With the exponential growth of various network resources, the use of search engine has become one of the most basic skills of everyone in today's society, and an efficient information retrieval model is also of more significance. The traditional text-based music information retrieval method can retrieve music data by inputting text information such as song name, composer, singer and album name. The content-based music information retrieval queries the target music through the input music melody information. In the actual music information retrieval scene, there is interaction between the user and the retrieval model. The user gives feedback on the retrieval results, and the retrieval model returns a new page of document according to this feedback. The existing ranking learning model regards ranking as a one-time process, ignores user feedback, and the ranking effect needs to be improved. With the increasing demand for digital music information and the continuous expansion of application fields based on massive music data sets, content-based music information retrieval method is attracting more and more researchers' attention.
Physical movements are an important part of people’s activities and communication. The movements of the hands and upper limbs are relatively fast, and the postures in the lower part of the ordinary camera will produce smears, and the images will be clearly obtained only when they are stopped, which seriously affects the recognition speed. This article takes the hands-guessing game in the communication robot as an example. Using a high-speed camera, the corresponding posture can be recognized when the posture is not completed during the hands-guessing game. It is expected that postures can be recognized in a more timely manner in human posture recognition, and the sense of delay in communication can be reduced. The corresponding data is collected and the model is generated after deep learning. In the fast-moving stage of postures, using a ordinary camera, the percentage of time that different postures are unrecognizable is between 50% and 60%. Compared with ordinary cameras, using high-speed cameras, the unrecognizable time percentage of different postures is reduced from 50%-60% to 0%, and the effect is obvious. In human-computer interaction, using ordinary cameras to infer postures, the people participating in the test have a significant sense of delay. With a high-speed camera, this feeling of delay is barely noticeable.
In recent years, economic and information globalization, e-commerce booming, and the application and popularization of the Internet and mobile payments have accelerated the development of the e-commerce industry. People are relying more and more on the Internet for shopping, and all kinds of shopping needs can basically be met online. For e-commerce companies, the construction of e-commerce service platforms can help promote the improvement of corporate economic benefits; for users, it is greatly satisfy their shopping needs. Based on this, the purpose of this paper is to study the construction of e-commerce service platform under hybrid genetic algorithm. This article first conducts a system analysis, expounds the design of an e-commerce service platform based on hybrid genetic algorithm, and clarifies the overall design framework and specific functional modules of the e-commerce service platform. Finally, this article has carried on the performance and the function test to the system. The test results show that the average response time of system login is less than 2 seconds when there are more than 10 to 1500 users concurrently; when more than 10 to 2000 users are concurrently, the business data declaration time is less than 4 seconds. It can be seen that the system has good performance.
In recent years, with the rapid development and wide application of modern information technology, the management environment and management concept of enterprises have been changed. The environment of enterprise accounting has changed greatly, and the accounting data system came into being. The application of accounting data system not only brings convenient and fast information services to enterprises, but also brings many security risks due to its own openness and the vulnerability of data storage media. With the expansion of the scope of use of computers, it is common to use computers for corruption, fraud and criminal activities. For example, the data stored on computer disks is easy to be tampered with; The data of the database is highly centralized, and unauthorized personnel may invade the enterprise's database through the computer network, browse all data files, copy, forge and destroy important data of the enterprise. Because computer crime has great concealment and harmfulness. Therefore, it is particularly important to strengthen the security of accounting data. This paper uses DES algorithm to encrypt accounting data.
Prefabricated buildings have been highly recognized for their advantages such as ease of construction and construction, and the scope of popularity is becoming wider and more numerous, and there are still a large number of building construction plans to use this type of building form. Under such circumstances, what we need to do most is the design of prefabricated buildings, especially through BIM technology application to improve the design quality of prefabricated buildings, to better guide the construction of prefabricated buildings. To ensure the efficiency and quality of building construction, and reasonably control costs. This article first analyzes the advantages of BIM technology in the design of prefabricated buildings, and then mainly introduces the design of prefabricated buildings based on BIM and conducts research and analysis on BIM technology application in building construction.
In this paper, image noise reduction research is carried out based on in-depth learning. In specific life, due to the lack of perfection of equipment and system, the image will often be polluted by more noise, resulting in unclear image details and reduced image clarity. Better image display ability can be obtained when BP neural network is used to denoise the image. Through the research on the activation function and optimization network function based on weighted neural network (CNN), combined with multi feature extraction technology and other in-depth learning models, we can learn and extract the important features of the input image. At the same time, we propose CNN back propagation optimization algorithm. At the same time, the training speed of the model is improved and the convergence speed of the algorithm is accelerated. Based on the deep residual learning of convolution network, the algorithm is used to remove the noise in the model. This is a better image denoising network model. Compared with other excellent denoising algorithms, the analysis and comparison show that the optimized denoising algorithm can not reduce the clarity of the image. At the same time, the image noise pollution is greatly improved and the image details are clearer.
We have developed walking assistance system considering the condition of the emotion and fatigue. This system consists of the emotion estimation, the fatigue evaluation, decision function of the walking condition, and walking assistive device. We already made the emotion estimation system, and obtained the method of muscle fatigue evaluation by NIRS, and the walking assistive device. To complete the decision function, it is necessary to recognize the current condition of the user’s emotion and fatigue. In this paper, we proposed the three dimensional human condition model to decide the suitable walking setting. Finally, the example of using this model and the importance of the recognition both emotion and fatigue were shown.
The traditional direct torque control is mainly based on the stator flux vector value to achieve the purpose of stator torque control. The most effective way is to change the double hysteresis structure of stator and torque flux. Although the structure of this control mode is relatively simple and the torque can respond quickly, it will lead to serious torque and speed fluctuations. In this paper, the SVPMSM technology is introduced to replace the original voltage vector selection module, and the MATLAB/Simulink simulation model is built. The results verify the effectiveness of the direct torque control algorithm.
With the continuous development of modern information technology, my country has gradually entered the era of big data. The salient features of the big data era are rich data resources, convenient data processing and information exchange, and smoother learning and communication between people. The impact of big data on education is also very significant. This paper studies the online education big data platform based on data mining and data collection technology, uses data mining technology and data collection technology to design the online education big data platform, and tests the designed platform. The test results show that this paper improves the algorithm the accuracy of clustering analysis is good, and the number of errors is controlled within 5, and then the query time of the platform is tested. The time for the platform from data query to acceptance is within 30 minutes, which meets the requirements of platform design.
Starting from the form, image and significance of environmental design, the path planning algorithm in navigation algorithm has always been one of the core contents that need to be solved and optimized in automatic navigation research. Industrial technology and Internet, people have higher and higher requirements for the performance of path planning search, such as truthfulness, timeliness, efficiency and accuracy. Nowadays, many algorithms implemented on CPU or GPU can not meet people's requirements for the performance of the algorithm, so more and more algorithms are integrated into hardware to improve the performance of the algorithm. Integrating the algorithm into hardware can solve the problem of insufficient efficiency of path planning algorithm on industrial computer. At present, FPGA (field programmable gate array) has the characteristics of strong portability, repeated configuration of computing resources, and internal logic can be changed according to user needs. Many embedded navigation algorithms are suitable to be developed through FPGA platform to realize the synthesis of hardware, so as to accelerate the algorithm. This paper puts forward the method of integrated design, the shaping of place spirit and cultural intervention, and makes a case analysis combined with many years of design practice.
With the large-scale construction of smart grids in my country, as well as the continuous improvement of power system dispatch automation and substation automation technology, the work of ensuring and coordinating the smooth operation of substations continues to deepen. With the development of science and technology, deep learning and machine learning technologies are becoming more and more intelligent. Nowadays, deep learning has made great achievements in the fields of target detection, image recognition, character recognition, etc., serving the work of substations. The purpose of this article is to study the intelligent analysis of substation images based on deep machine learning technology. Starting from the analysis of images, this paper uses deep machine learning technology as the technical support for intelligent analysis of substation images, combines deep machine learning technology with substation image detection, and focuses on the application of deep machine learning technology in substation image intelligent analysis, improve the intelligent level of substations and ensure the safe and stable operation of substations. Experimental data shows that the correct recognition rate of the BP neural network proposed in this paper for the normal operation of substation equipment and the heating fault of the three types of arresters are 98.06%, 98.25%, 99%, and 98.75%, respectively. It can be concluded that the BP neural network has a high image recognition accuracy rate and it is suitable for the actual work of infrared detection of lightning arresters in substations.
Solar greenhouse is an important agricultural facility in China's agricultural production. It has the functional advantages of cold proof and heat preservation, rain proof cultivation, disaster resistance, anti-seasonal regulation and so on. According to the wind sensitive characteristics of solar greenhouse, this study takes the typical solar greenhouse in Shandong as the research object, uses the CFD numerical wind tunnel method, takes the wind direction angle as the variable, carries out the simulation calculation of the surface average wind pressure of solar greenhouse, carries out the optimization research of the wind load shape coefficient, and makes a comparative analysis with the code for the design load of horticultural greenhouse structures. The practical recommended value of wind load shape coefficient is given, which provides a theoretical reference for the design, construction, production, operation and maintenance of solar greenhouse.
Green manufacturing is a modern manufacturing mode with significant social and economic benefits, so it is a frontier and hot topic at home and abroad. As the basis of operation technology optimization, management and technology development of green production system, workshop planning plays an important role in the research of green production theory. In recent years, N - P genetic algorithm has been gradually applied in solving workshop optimization problems. In this method, the system generates the initial overall solution of the problem, and then selects the best solution for further improvement. This process continues until all possible solutions are exhausted or through a specified algebra. Genetic algorithms are often used in engineering applications because they provide a decision-making method based on experience rather than relying only on rules and formulas. This technology provides an effective solution for analysis and decision-making by using genetic algorithm, which is helpful to solve the complex problems with limited time and resources.
In recent years, with the improvement of living standards, people's demand for electricity has gradually increased. With the rapid development of substation informatization and digitalization, intelligent analysis technology can conduct real-time status assessment of substation equipment and improve the efficiency of fault handling in smart substations. The purpose of this paper is to study the research and application of intelligent analysis technology in the design of substation condition monitoring system. This paper analyzes the requirements of the substation condition monitoring system based on intelligent analysis technology, and describes the application architecture of the system and the total volume architecture of the software in detail. This article tests the designed system, and the experimental results show that the system CPU utilization rate is maintained at 50%~60%, and the memory utilization rate is maintained at about 61%~64%. It can be seen that the various components of the system can coordinate and operate stably, and the resource utilization rate fluctuates little, indicating the feasibility of the system architecture.
Coal is an important basic energy source in our country, accounting for about 70% of the total energy, and its development prospects are very broad. Our country has abundant coal reserves, and 95% of coal mining is underground operations. Although major accidents have declined in the past two years, the safety production situation in our country's coal mines is still severe, with a large number of casualties every year. The BIM technology, which is currently in the research and promotion stage, provides an effective way to solve such problems. Promoting the application of BIM technology will be the trend of future development. However, the research on coal mining risk assessment is relatively lagging, and most of the original evaluation methods are not suitable for complex coal mine systems. Therefore, this article is mainly based on BIM-based smart coal mine safety mining RA (Risk Assessment) research. First of all, the analysis of the current situation of coal safety issues shows that the situation of CM (coal mine) safety production in our country is still very severe. Secondly, analyzed the impact of BIM on coal mining RA. The results show that 41% of people believe that BIM technology can improve the accuracy of RA. 37% also pointed out that the application of BIM technology to coal mining RA can help save companies' capital cost.
With the decrease of sensor cost and the increase of performance, the application of sensor is becoming more and more popular. Sensors in different locations and devices produce different data streams. We expect to analyze the data or events we are interested in in in real time from these data streams. With the increase of the number of sensors, a large number of sensors are organized into the form of networks. How to carry out data mining in sensor network data stream will bring a new challenge. As one of the most important data flow technologies, sliding window technology has been widely studied and applied. By accurately estimating the data distribution in the sliding window, we can carry out anomaly detection and other important applications.
With the continuous development of computer technology, the coverage of informatization solutions covers all walks of life and all fields of society. For colleges and universities, teaching and scientific research are the basic tasks of the school. The scientific research ability of the school will affect the level of teachers and the training of students. The establishment of a good scientific research environment has become a more important link in the development of universities. SR(Scientific research) data is a prerequisite for SR activities. High-quality SR management data services are conducive to ensuring the quality and safety of SRdata, and further assisting the smooth development of SR projects. Therefore, this article mainly conducts research and practice on cloud computing-based scientific research management data services in colleges and universities. First, analyze the current situation of SR data management in colleges and universities, and the results show that the popularity of SR data management in domestic universities is much lower than that of universities in Europe and the United States, and the data storage awareness of domestic researchers is relatively weak. Only 46% of schools have developed SR data management services, which is much lower than that of European and American schools. Second, analyze the effect of CC(cloud computing )on the management of SR data in colleges and universities. The results show that 47% of SR believe that CC is beneficial to the management of SR data in colleges and universities to reduce scientific research costs and improve efficiency, the rest believe that CC can speed up data storage and improve security by acting on SR data management in colleges and universities.
In recent years, with the development of computer science and technology, computer analysis methods have become an important scientific method for solving fluid engineering and other problems in today's society. Fluid dynamics has always been a popular research direction in the academic circle and the industry. In-depth, the radial fluid diffusion problem has gradually become the focus of this field. The purpose of this article is to explore the relationship between radial fluid diffusion and time scale based on computer analytical methods. This article first describes the solution process of computational fluid dynamics based on computer analytical methods, and analyzes the radial fluid dynamic axis, the movement direction of fluid particles, and the velocity of fluid particles. Then through experimental simulation, observe the fluid diffusion process, explore the relationship between radial fluid diffusion and time scale, and the influence of material structure on the diffusion results. Experimental data shows that when the contact angle is 1 degree, the absorption rate is 97.8%, when the contact angle is 45 degrees, the absorption rate is 84.7%; when the contact angle is 89 degrees, the absorption rate is 7.6%. This shows that the smaller the contact angle, the larger the contact area between the liquid and the solid fabric.
With the rapid development of mobile wireless Internet, the arrival of the information age has changed the way we think and communicate, and the pervasiveness of social media has prompted us to live in a culture of continuous learning and sharing. The influx and collection of information need to be controlled and applied, which has prompted changes and innovations in the means of information dissemination. The human visual system has amazing processing functions, and the visualization of information enables us to quickly and effectively gain insight and understanding of the communication of information. This article aims to study the application of Internet-based new media in information visualization. Based on the analysis of the shortcomings of traditional media information visualization design, the development characteristics of new media, and the superiority of information visualization design in new media, it is necessary to conduct new information on the residents of a certain city. Media information visualization questionnaire survey to understand the city’s residents’ understanding of new media information visualization. The survey results show that new media information visualization has a wide range of applications. What residents know best is new media charts, accounting for 24.5%; next are data news and information websites, accounting for 20% and 19.5% respectively; what residents don’t know is interaction design, the number of people who understand is only 10%.
Internet security incidents are frequent, the forms of attacks against Internet business systems are diversified, and the traditional defense means can hardly meet the growing security protection needs. This paper proposes a new system based on cyberspace asset discovery and identification method, which contains five layers: data collection layer, data aggregation layer, data storage layer, permission control layer and web display layer, to comprehensively and accurately sort out the network assets, clearly grasp the attack surface exposed assets, and effectively manage and protect the cyberspace for security.
