With the development of science and technology, the application of flight manipulators has received extensive attention. The flying manipulator has broad application prospects, such as the maintenance of high-voltage towers, the storage and retrieval of elevated goods in warehouses, and the delivery of express and takeout goods. Before the actual application of the flight manipulator, due to the complex task requirements and nonlinear environment, it is necessary to continuously optimize the Trajectory Planning and Control (hereinafter referred to as TPC) of the flight manipulator. In order to improve the recognition and positioning accuracy of the robotic arm on the surface of the aircraft, and achieve precise control of the autonomous motion and operation of the robotic arm on the surface of the aircraft, this paper studies the TPC of the flight robotic arm based on deep learning, image moment and vector product methods, establishes a bearing return function model based on deep learning, and a Jacobian matrix of the flight robotic arm based on image moment and vector product methods. Through the experimental research on TPC of the flight manipulator, it was proved that the DL trajectory planning method could reduce the collision risk of the flight manipulator by 4.79% compared with the traditional trajectory planning method, and could improve the task completion speed of the flight manipulator by 4.66%. The application of DL to the TPC of the flight manipulator could improve the trajectory planning effect of the flight manipulator.

Financial analysis is a method of analyzing the overall operating status of an enterprise based on financial information, which can help managers judge the company’ s operating risks and adjust the company’ s operating conditions in a timely manner, so as to better achieve business management. This paper aims to study the design of financial indicator analysis system through big data. This paper proposes to find the best clustering center by means of fuzzy identification algorithm, determine the quality of the company’ s operating status, calculate the company’ s overall operating indicators, and determine the company’ s risk level and improvement direction. The experimental results of this paper show that the fuzzy identification algorithm can help the enterprise to determine the overall state of the enterprise’ s operation, improve the financial risk identification ability by 20%, and better realize the enterprise’ s financial analysis and processing.

Fresh items have become an essential necessity for modern people, and the daily diet structure is growing more and more rich as people’s attention to health increases. One of the characteristics of fresh products is that they are hard to retain at room temperature. As a result, IoT logistics technology assistance is frequently needed in logistics linkages including distribution, transportation, and warehousing. Through the scientific and logical planning of the route of fresh food logistics distribution vehicles, this paper aims to effectively lower the overall economic cost of logistics distribution, guarantee the freshness of the fresh food distribution process, satisfy the various individualized needs of customers for delivery time, and enhance logistics distribution. security. This study suggests an enhanced ant colony algorithm in artificial intelligence that can efficiently determine the shortest path. This algorithm can be used to find the best route for new logistics distribution and lower transportation losses. It is based on 5G Internet of Things technology. The ant colony method prior to the enhancement had the longest optimization time of 25. 06 seconds in the 8 search process, according to the experimental data presented in this study. The enhanced ant colony algorithm had the longest optimization time of 17. 89 seconds. In finding the optimal path, after the improvement, the ant colony algorithm takes less time. In the comparison of transportation costs, the cost of the improved ant colony algorithm is reduced by about 1, 100 yuan, the vehicles required are less than those of the ant colony algorithm before the improvement, and the decay rate is also reduced a lot. It can be seen that the improved ant colony algorithm is more suitable for the analysis of the optimal path of fresh logistics distribution.

Students in adolescence are not mature in mind, thought, ability and other aspects, which are easily affected by various emotional behaviors. Positive emotional behavior contributes to students’ mental health and academic progress. Negative emotional behavior would lead to psychological problems and academic frustration. If it is not paid attention to, students may act out of control under the control of negative emotions, thus resulting in serious mental illness, which is not conducive to the education and management of students. Due to the rapid development of social information network and science and technology, the analysis of students’ emotional behavior and educational management by pure human intervention has fallen behind, and it is impossible to timely feedback, track and predict students’ status. This paper introduced the general direction and achievements of human-computer interaction research, and discussed the combination of big data and human-computer interaction. The method of applying human-computer interaction technology to students’ emotional behavior analysis and education management was studied. The pure human intervention method was compared with facial emotion recognition, voice emotion recognition, human-computer body feeling interaction and virtual scene education methods under human-computer interaction technology. Five experimental groups were designed to conduct research in three aspects of emotional behavior analysis, education and learning, and supervision and management. It was found that the average accuracy of facial emotion recognition for emotional behavior analysis was 88.0%; the average course learning efficiency of virtual scene education used for students’ educational learning was 82.8%, and the total progress was up to 99.81%; the average success rate of human-computer somatosensory interaction for supervision and management was the highest, which was 68.1%.

With the rapid development of China’s economy and the continuous improvement of its international status, “Chinese fever” is quietly emerging all over the world, and the teaching of Chinese as a second language has become an independent discipline in China. There is a lot of room for exploration in research. In the past few decades, with the rapid development of science and technology, “intelligence” has become the development trend of the whole society. The potential of smart devices has become more and more widely used, which has also inspired users’ love and ultra-fast adaptability to devices. This paper uses wireless network communication to study language recognition and its type induction in second language teaching, and proposes a resource allocation mechanism based on bilateral induction. Different expressions are designed with different priorities, and after transformation, the follow-up type induction research can be carried out smoothly, and finally the optimal induction allocation scheme is obtained. The research results show that in the teaching type induction, there are 684 new words in volume I, 778 new words in volume II, and 1462 total words in volumes I and II, and they are all summarized. Compared with traditional methods, the search and extraction speed is increased by 45%. Teaching type induction is more effective in the comprehensive use of multiple teaching methods in primary teaching, but it is still inseparable from traditional teaching methods. Therefore, in the specific teaching practice, we should choose a more suitable teaching method according to the individual factors of the teaching content and teaching objects.

As AI technology matures, computational intelligence has also been more widely used. Computational intelligence is an important branch of AI. Because of its global search, efficient parallel and other characteristics, it has become a new method to solve complex optimization problems and has received more and more attention. In the meantime, driven by other technologies such as big data, education has gradually broken away from the traditional teaching methods, broken the traditional time and space constraints, and opened a new chapter. With the reform of the national curriculum instructional patterns, the exploration and practice of educational informatization in the educational circle is rising, and the instructional patterns of intelligent service aided design curriculum is also coming. In this paper, a teaching mode of intelligent service aided design with human-computer interaction (HCI) as the core was proposed. Based on the available results, it is an important reference value for its application and promotion in practice. Through the empirical analysis of the course instructional patterns of intelligent service aided design of HCI under computational intelligence, the classroom instructional patterns method of intelligent service aided design of HCI in the intelligent era proposed in this paper has improved 12.7% in promoting students’ understanding of teaching content compared with traditional methods and has increased by 19.7% in improving students’ full satisfaction with course teaching. Besides, in terms of overall teaching effectiveness, it has improved 22.9% compared to the traditional. It illustrates that the teaching methods presented in this paper can better serve students’ development and improve their overall quality, and also meet the teaching needs of teachers, so that teachers can better carry out teaching activities. At the same time, the curriculum teaching mode of human-computer interaction intelligent service aided design in the intelligent era was discussed, which was conducive to promoting the gradual maturity of the development of computational intelligence and making its application in teaching more complete.

In order to provide various disturbance voltage waveforms for the test of power quality event detection and compensation device, it is necessary to develop a device that can simulate power grid faults. VSC power disturbances generator (VSC-IG, Interruption Generator based on Voltage Source Converter) is the importance of theoretical study and compensation device of power quality test tools, the flexibility to produce all kinds of disturbance voltage waveform, And reduce the harmonic pollution to the power grid. In this paper, the power injected by AC power supply into VSC-IG is controlled to stabilize the DC voltage, and a PI controller parameter tuning method considering the change of resistance parameters is proposed to optimize the dynamic performance of the controller. The SPWM rule sampling method is adopted as the underlying control strategy to realize the generation of disturbance voltage waveform. PSCAD/EMTDC platform is used to build the VSC-IG simulation model and carry out the simulation research. The VSC-IG device is designed and implemented by using the physical prototype hardware platform of 30kVA back-to-back converter in dynamic modeling laboratory. The experiment verifies the main circuit structure and control strategy of VSC-IG in this paper, and also verifies the function of digital controller, which lays a foundation for further research on power quality.

With the continuous development of the internet age, more and more art images are taking on digital forms, resulting in a new way of survival for art image digitization. However, the digitization process of art images is affected by various factors, resulting in poor results and low digital quality of art images. Therefore, this article conducted research on the digitization of art images based on metadata, and utilized BP (Back Propagation) neural network for metadata processing and analysis to achieve metadata visualization and interactive design. Animation production software was then utilized for image compression, transparent display, and modeling, and finally interactive display technology was used to display the dynamic design of art images. 4000 user feedback data and art image metadata from four age groups were collected and named A art image set. Starting from the visual communication effect, accuracy, and fidelity of art images, the differences in dynamic design of A art image digitization were compared. The experimental results showed that 2820 people were satisfied with the visual communication effect of dynamically designed art images, with a satisfaction rate of 70.5%. Only 1070 people in the control group were satisfied. The metadata accuracy of dynamically designed art images was greater than 80%, and the average accuracy was close to the median line, with small overall fluctuations. The deviation value between dynamically designed art image data and standard images is small, and the overall fidelity is relatively high. In short, the evaluation effect of digital dynamic design of art images is very good.

Short-term traffic speed prediction in Intelligent Transportation System (ITS) provides an important idea for solving traffic problems. To capture the spatio-temporal properties of traffic speed prediction, we proposes a Graph Convolutional Network-Gated Recurrent Units with Attention (GCN-GRUA) mode for expressway. The Graph Convolutional Network (GCN) and Gated Recurrent Unit (GRU) were used to extract the spatial and temporal features of traffic speed, and the attention mechanism was introduced to improve the prediction performance of the model. Experimental results from the real traffic data set of Qingyin Expressway show that the proposed model has a significant improvement in prediction accuracy compared with GCN, GRU and GCN-GRU models. In addition, the importance of speed characteristic variables and exogenous variables on the traffic speed prediction accuracy show that the speed data with the closest time interval has the greatest influence on the traffic speed prediction, followed by the daily cycle characteristics of traffic speed. As the prediction time increases, the relative importance of the velocity characteristic variable remains above 0.6, while the relative importance of the exogenous variable keeps rising.

As a new product of artificial intelligence, big data is widely used in daily life. Due to its appearance, people’s lives are more convenient and efficient, but at the same time, there are certain security risks, namely the leakage of private information, especially the financial information problem brought about by financial informatization has a more serious leakage problem. In order to effectively reduce the problems caused by the leakage of financial information privacy, this paper attempted to establish a model of related protection measures for financial big data information security by establishing a three-dimensional encrypted information model of big data or by using differential privacy method and using their own. The three-dimensional encrypted information model of big data overcame the defect that financial information is easy to be broken, while the differential privacy model overcame the defect of inaccurate protection of financial information, both of which can play a better protective role in different applications. The experimental results showed that in the process of accessing financial data information, with the increase of access frequency, the number of sensitive locations changes from 40 to 46. This also meat that a non-sensitive position becomes a sensitive position, which blurs the original sensitive position and achieves the effect of protecting the real sensitive position.