In order to protect our country's territorial waters from infringement and maintain maritime rights and interests, it is necessary to establish a marine defense system that combines both offshore defense and far sea defense. Coast-to-shore/GEO satellite-ship-to-ship bistatic inverse synthetic aperture radar (inverse synthetic aperture radar, ISAR) has all-day and all-weather characteristics, and can realize long-distance continuous tracking, imaging and identification of ship targets in offshore/far sea areas. In addition, these two bistatic ISAR imaging systems also have the advantages of high concealment, strong anti-strike capability, high flexibility, multiple imaging modes, and rich target information, thus improving the reliability of the marine defense system.

With the increasing scale of the Internet medical market, a large number of platforms have started to provide free medical diagnosis services to users. The number of questions on medical Q&A platforms is growing rapidly, and there are a lot of repetitive and similar questions, which can no longer meet the demand of users to get fast and accurate answers only by the manual answers of medical experts. To solve the above problems, this paper proposes a Chinese medical short text matching model based on fine-tuning BERT-Attention-BiLSTM (FBAB), use fine-tuning BERT model for representation of text, the attention mechanism is used to obtain the interaction information between sentences, use the Bi-directional Long Short-Term Memory (BiLSTM) to fuse the contextual information of the 2 sentences., and the feature information is further extracted by maximum pooling and mean pooling to obtain the sentence-level semantic representation, which is input to the prediction layer after stitching, and the softmax classification function is used to calculate the probability that the two sentences are similar. Experiments show that compared with the classical deep short text matching model, the FBAB model proposed in this paper shows better results on the medical short text corpus, which proves the effectiveness and feasibility of the model.

The Traveling Salesman Problem is a typical combinatorial optimization problem, which has not been well solved until now. In this paper, The Genetic Algorithm is used to solve this problem, and the gene segment is regarded as an urban sequence. The introduction of the crossover rate and the mutation rate not only ensures the current good genes, but also produces better genes. In this way, the optimization of the problem solution is continuously realized. We experimentally validate The Genetic Algorithm with 100 cities, and for this problem, The Ant Colony Algorithm and The Particle Swarm Algorithm are used for experimental comparison. The experimental results show that The Genetic Algorithm has a strong global optimization ability and better overall performance under the premise of ensuring a faster operation speed.

According to the characteristics of ship maneuvering motion control, this paper constructs the membership function and control rules of the input and output language variables of the ship course fuzzy neural network controller, and designs a fuzzy controller. Aiming at the deficiencies of the basic fuzzy controller, a two-stage fuzzy controller with fuzzy PI control and self-adjusting parameters is proposed. When it is applied to the ship heading control system, the system has satisfactory results in terms of immunity and response speed. By establishing the description function of the fuzzy controller, the analysis concludes that the ship course fuzzy control system is stable.

This paper utilizes the BERTbased-BiLSTM-CRF models to complete Chinese named entity recognition tasks, including finetuned and unfinetuned BERT models. Use the pre-training model BERT(Bidirectional Encoder Representations from Transformers), a BiLSTM(Bi-directional Long Short-Term Memory) network and CRF(Conditional Random Field) to perform NER(Named Entity Recognition) on Chinese. Tested on the people-daily-ner-pretreatment corpus, compared with other models, the BERTbased models can effectively identify entity information, and the evaluation metrics on the dataset have been improved.

Named entity recognition (NER) is an important research direction in natural language processing (NLP). Traditional machine learning algorithms in NER have problems such as low accuracy, highly dependent feature design, poor domain adaptability, and inability to handle the different contexts of multiple meanings of the term in recognizing Chinese entities. Based on these problems, this paper adopts a method based on the BERT-CRF model in Chinese NER. The BERT preprocessing language model generates word vectors that represent contextual semantic information, automatically extract numerous word-level features and semantic features in text, and decodes through the CRF layer generates entity tag sequences. In this paper, the BERT model has been fine-tuned to make the model perform better on NER tasks, and the experimental verification is carried out on the People’s Daily dataset, and the F1 value reaches 94.5%.

With the increasingly severe global energy crisis and environmental pollution, the development of renewable energy has become the key to protecting the environment, saving energy and reducing emissions. As a kind of clean and low-cost renewable energy, wind power has been paid more and more attention by more and more countries. However, the time-varying and intermittent wind speed often leads to instability of wind power and hinders the integration of wind power into the grid. Therefore, effectively predicting the wind spee of wind farms can effectively reduce the cost of wind power grid connection and improve the security and stability of the power grid. In this paper, we propose a wind speed prediction model LSTM-FWA based on long short-term memory (LSTM) network and fireworks algorithm (FWA). The experimental results show that the prediction performance based on LSTM-FWA model is the most obvious. The LSTM-FWA was compared with other prediction models, and the performance of the model under various optimization strategies was comprehensively analyzed. Experimental results show that the proposed spatio-temporal optimized LSTM-FWA model has the best performance for wind speed prediction.

In view of the increasingly complex navigation situation, in order to ensure the safety of ship navigation, a neural network is proposed to establish an environmental safety risk assessment model for ocean-going passenger ships. In the case of analyzing the factors affecting the safety of the ship's navigation environment, each factor is quantified and assigned, and each group of data is evaluated and scored according to experience and quantitative rules. Input the data to the neural network and train the neural network model. The test data is input into the neural network model, and the data predicted by the neural network model is compared with the expert's score. The evaluation results reflect the feasibility of the model.