To achieve early fault warning for feedwater pumps in power plants， a feedwater pump fault warning model， MHA-NGU， based on Multi-Head Attention （MHA） mechanism and New Gated Unit （NGU） was proposed. MHA mechanism was employed to capture long-range dependencies in the time series data， thereby enhancing key dimensions through multi-perspective feature interaction. NGU is an improved Recurrent Neural Network （RNN） proposed in this paper， in which the output gate and reset gate were designed with two levels， thereby avoiding the degradation in global dependency capture capability caused by single?level gates’ over?reliance on local information. Additionally， a T-transformation module was embedded within NGU to further alleviate the gradient vanishing problem of the model. Experimental validation was conducted using data from a 1 030 MW thermal power unit in Zhejiang Province as an example. The results demonstrate that the proposed model increases the Coefficient of Determination （R2） by 4.38% and 1.90% compared with GRU and the second-best model， MHA-GRU， respectively. Furthermore， it triggers a warning 5.39 hours earlier than the GRU model， realizing an early fault warning for feedwater pumps effectively.

A large number of self-made videos on the Internet lack user ratings and the recommendation accuracies of them are not high. In order to solve the problems, a Video Recommendation algorithm based on Danmaku Sentiment Analysis and topic model (VRDSA) was proposed. Firstly, sentiment analysis was performed to video' danmaku comments to obtain the sentiment vectors of the videos, which were used to calculate the emotional similarities between the videos. At the same time, based on the tags of videos, a topic model was built to obtain the topic distribution of the video tags which was used to calculate the topic similarities between the videos. Secondly, the emotional similarities and topic similarities were merged to calculate synthesis similarities between the videos. Thirdly, combined with the comprehensive similarities between the videos and the user's history records, the user preference for videos was obtained. At the same time, the video public recognitions were quantified by user interaction metrics such as the number of likes, danmakus and collections, and the comprehensive recognitions of the videos were calculated by combining the user's history records. Finally, based on the user preference and video comprehensive recognitions, the user's recognitions of videos were predicted, and a personalized recommendation list was generated to complete the video recommendation. Experimental results show that, compared with Danmaku video Recommendation algorithm combing Collaborative Filtering and Topic model (DRCFT) and Unifying LDA (Latent Dirichlet Allocation) and Ratings Collaborative Filtering (ULR-itemCF), the proposed algorithm has the precision increased by 17.1% on average, the recall increased by 22.9% on average, and the F1 increased by 22.2% on average. The proposed algorithm completes the recommendation of videos by analyzing the sentiments of danmakus and integrating the topic model, and fully exploits the emotionality of damaku data to make the recommendation results more accurate.

Hierarchical routing algorithm is now a hotspot in the field of wireless sensor network. Aiming at the problem that the energy of sensor nodes are limited, a hierarchical routing algorithm for wireless sensor networks based on small world model (HASWNM) was proposed. The Wireless Sensor Network (WSN) could reflect characteristics of the small world by adding nodes with high performance as well as shortcuts among cluster heads. As the energy consumption was mainly concentrated in the data transmission phase, the energy of the cluster head was taken into account while choosing the relay node between clusters. Besides, the different adaptive search area was determined according to the distance between the cluster head and the base station. The experimental results showed that the network can show the characteristics of small world when the number of high-performance nodes reached 100, and compared to the algorithms of CSWN (topology Control based on Small World Network), TSWN (Tree-based Small World Network), DASM (Directed Angulation towards the Sink node Model), the number of death rounds of the first node was delayed by 6%,6%,29% separately, and the average energy consumption of the network per round was reduced by 5%,12%,17% respectively. Thus the wireless sensor network constructed by the proposed algorithm has the characteristics of small world and low energy consumption.

In order to solve the problems of high outage probability and unbalanced load in dense small cellular networks, an energy-efficient scheme based on load balancing was proposed. The network energy-efficiency was maximized through joint optimization of load balancing and base station on/off control strategy under the constraints of guaranteed user outage probability and minimum rate. The optimization problem is a non-convex Non-deterministic Polynomial-hard (NP-hard) problem, and it is quite complex to obtain the optimal solution. Therefore, the original optimal problem was decomposed into two suboptimal subproblems. Firstly, the optimal load balancing strategy with the given base station on/off control strategy was given by the proposed load balancing scheme. Secondly, the optimal base station on/off control strategy was designed under the constraint of satisfying the minimum user rate. The experimental results show that, when the number of users is less than 180, the outage probability of the proposed scheme is zero while the outage probability of the traditional Maximum Signal to Interference plus Noise Ratio (Max-SINR) algorithm reaches 11%. The network energy-efficiency of the proposed scheme is higher than those of the base station Randomly-off (Ran-off) algorithm and the base station Not-off (No-off) algorithm. The proposed scheme can improve network energy-efficiency and ensure load balance.

Focusing on the issue that most research results on correlation between advertising phrases stay in the literal level, and can not exploit deep semantic information of the phrases, which limits the performance of the task, a novel method was proposed to calculate the correlation between the phrases by using deep learning technique. Recursive AutoEncoder (RAE) was developed to make full use of semantic information in the word order and phrase, which made the phrase vector contain more deep semantic information, and built the calculating method of correlation under the advertising situation. Specifically, for a given list of a few phrases, reconstruction error was produced by merging the adjacent two elements. Phrase tree, which similar to the Huffman tree, was produced by merging two elements with smallest reconstruction error in turn. Gradient descent and Cosine distance were used to minimize the reconstruction error of phrase tree and measure the correlation between the phrases respectively. The experimental results show that the contribution of the important phrases is increased in the representation of the final phrase vector by introducing weight information, and RAE is more suitable for phrase calculation. The proposed method increases the accuracy by 4.59% and 3.21% respectively compared with LDA (Latent Dirichlet Allocation) and BM25 algorithm under the same condition of 50% recall rate, which proves its effectiveness.

In order to further improve the accuracy and efficiency of product recommendation in E-commerce recommender system, by analyzing the problems of traditional recommender system and prior optimization technique, a multi-Agent recommender system model was proposed. By adopting multi-Agent technique in Artificial Intelligence (AI) and terminal self-adaption, this model improved the efficiency of traditional recommender system in the condition of using multi-terminal, and dynamically returned recommendation to users according to different terminals. The experimental results show that multi-Agent cooperative E-commerce recommender system improves the accuracy and efficiency to a certain extent.

An affine invariant local feature detector has been put forward for 3D object image matching. In order to cope with view angle and scale changes, this algorithm changed the image structure to fit the circular Gaussian filter according to the principle that Gaussian filter and image structure should be compatible. Local image structures were measured by covariance matrixes of Maximally Stable Extremal Regions (MSER) having been detected in the image. Anisotropic image structures must be rotated and squeezed into isotropic image structures to guarantee the correctness of image transformation. Finally, Scale Invariant Feature Transform (SIFT) features were extracted on isotropic image structures. Coordinates of SIFT features should be changed into the original image coordinates after being extracted. The experimental results indicate that the local features extracted by this algorithm are fully affine invariant. They are suitable to be used in wide baseline image matching for 3D objects.

The forecasting model about aviation ammunition storage-transportation system with automatic regulatory system in aircraft carrier to replenish the required aviation ammunition timely was established. This paper studied the storage and transportation system about aviation ammunition, built the causal loop diagrams and stock and flow diagrams about every element impacting the storage-transportation efficiency of aviation ammunition with the method of system dynamics and simulated by the system dynamics software Vensim PLE 5.9. The simulation results show that this model can predict the storage and transportation trend about aviation ammunition under the condition of perturbation, which correspond with the data and conclusion of exercises with aircraft ammunition storage system.

Concerning the common multi-path transmission and reflection factors in the radio communications, broadband coherent sources must be considered. A new algorithm for the Direction Of Arrival (DOA) estimation of correlation wideband Non-Linear Frequency Modulation (NLFM) signals based on Discrete Polynomial-phase Transform (DPT) and virtual array was introduced. And comparison, analysis and improvement were done on the new algorithm. It can settle the problem of not finding coherent signals in routine MUSIC and ESPRIT algorithm. This approach can more accurately estimate and does not decrease effective array aperture, and it also can improve the utilization of array. Then the angle estimation algorithm of two closely-spaced emitters was proposed. The simulation results verify the correctness and efficiency of the new algorithm.

In order to find connections between network structure and network action, a new dynamic evolving model based on Dynamic Agent (DA) was proposed. Every node in network was modeled as an agent that had life and evolved according to its own strategy. The experimental results show that this model possesses good properties such as small-world property, scale-free property and large clustering coefficient, which reflect the real networks very well. Furthermore, this model improves the robustness of traditional models when attacked.