In order to solve the power supply problem of the oil field downhole water distributor, a wireless power transmission scheme with a coaxial solenoid type as an electromagnetic induction coupler was proposed, and the relationship between frequency and system output power under this condition was studied. The coupling coefficient of coaxial solenoid with uniform coupling medium and aligned two ends is related to the radius ratio of primary and secondary coils. In the tubing environment, the coupling coefficient is larger, and the influence of different natural resonance frequency and working frequency on the transmission power is greater in the case of series to series compensation. When the natural resonance frequency of the system is larger, the frequency splitting will occur. The working frequency can ensure that the system has high power output and transmission efficiency. Through simulation and experiment, the influence of natural resonance frequency and working frequency on power transmission is verified.

Magnetotelluric signal has the characteristics of weak signal and wide frequency band. The quality of electromagnetic data obtained in the field is extremely reduced by strong human interference. In order to improve the signal-to-noise ratio of the collected data, it is necessary to adopt a reasonable denoising method to reduce the noise of the data. Wavelet transformation has the function of joint analysis of “ time-frequency", which can conduct multi-scale detailed analysis of signal. So it is very suitable for processing magnetotelluric data with complex frequency components. The wavelet transform principle combines the signal and noise characteristics of the field electromagnetic data, reasonable wavelet transformation parameters of parent wavelet, decomposition layer number and threshold are determined. Different combinations of parameters are used to handle different human noise interference. The signal-to-noise ratio of field electromagnetic data is effectively improved after wavelet transformation, which provides a high-quality time series for subsequent visual resistivity calculations.

With the rapid development of the fifth-generation communication technology, the amount of communication data has grown rapidly. In order to solve the problem of low communication efficiency, a data receiving system with higher data transmission rate and higher frequency efficiency is proposed. By analyzing the application of AD9361 technology in wireless communication systems, a MIMO ( Multiple-Input Multiple-Output)、 OFDM (Orthogonal Frequency Division Multiplexing) synchronous timing receiving system based on AD9361 is designed. This system has the advantages of high spectrum utilization, flexible demodulation mode and fast communication speed having important significance in 5G construction. The test result shows that MIMO- OFDM synchronous timing receiving system based on AD9361 has high convergence and the FPGA ( Field- Programmable Gate Array) resource occupancy of the system is generally good, and the collected data is relatively stable. In addition, during the synchronization of data receiving the system can effectively avoid the influence of bit error rate, without ISI(Inter-Symbol Interference) and ICI(Inter-Channel Interference), and the data receiving system meets the transmission channel standard which has high wireless communication stability.

Power system disturbance data is of great significance for monitoring power system operation state and regulating its working mode. A new algorithm for power system disturbance data acquisition and classification is proposed based on compressed sensing. Firstly, an AtOMP ( Atom adaptive Matching Pursuit) algorithm is designed based on compressed sensing and K-SVD ( K-Singular Value Decomposition ) dictionary learning algorithm to compress and collect various disturbed data. Then, the sparse feature of the disturbed data, the standard deviation of the adaptive dictionary atom, kurtosis, margin factor and the number of principal frequencies are extracted as the training samples, and the BP neural network is used to realize the sample learning and classification. The experimental results show that the proposed algorithm can achieve highly compressed data collection of disturbed data, small data amount, high classification recognition, strong anti-interference and other advantages.

In order to avoid the influence of the fluctuation and uncertainty of wind power output on power grid operation, a combined algorithm of improved variational modal decomposition algorithm and integrated gated cyclic neural network is proposed for the short-term power prediction of wind farm. Firstly, aiming at the problems of low prediction accuracy and unstable prediction of single GRU(Gated Recurrent Unit) model, the stacking model is used for fusion and the SGRU(Stacking-GRU) prediction model is established. Secondly, for the common endpoint effect problem in VMD (Variational Modal Decomposition)decomposition algorithm, GRU extension method is used, and ALO ( Ant Lion Optimizer ) algorithm is introduced to optimize the super parameters in the model to establish GAVMD(GRU-ALO-VMD) decomposition model. Firstly, the wind power data is decomposed by GAVMD model, which reduces the instability of the data and improves the training difficulty of the prediction model. Then, the SGRU model is used to predict and reconstruct the decomposed IMF (Intrinsic Mode Function) component, and finally the predicted value of the original wind power data is obtained. The performance of the model is verified by the actual wind farm sample data. The simulation results show that the combined forecasting model based on GAVMD-SGRU maintains the average absolute percentage error index within 6% in different seasons compared with other models, which meets the requirements of the power system for wind power dispatch.

In order to intuitively and accurately eliminate the image blur phenomenon caused by the defocus error and improve the accuracy of digital image processing, a fuzzy image visual effect optimization algorithm based on stereo binocular vision is proposed. The fuzzy image optimization problem is transformed into an optimization estimation problem under the constraint of the fuzzy core path relationship. Based on the fuzzy core path correspondence, a stereo binocular image fuzzy core path corresponding relationship model is constructed, and the model is used as a convergence condition to obtain the coupling between the fuzzy image features relation and three-dimensional geometric information. Directional filter is used to remove fuzzy image noise, Radon transform is used to deal with the error caused by directional filter in fuzzy function estimation, convolution operation is used to remove blur phenomenon, through the gradient norm’s variational functional optimization algorithm. The visual effects of blurred images is optimized. The experimental results show that the proposed method can optimize the image contaminated by noise and blur, and retain the texture and detail information of the image completely. When the number of iterations is 7 times, the difference energy of the image reaches convergence, and the visual effect of the blurred image can be completed efficiently optimization.

In order to improve the efficiency and accuracy of the monitoring personal and vehicles on the substation construction site, a method of non-perceptible interactive based on multiple video analysis technologies is established to replace traditional monitor methods and to build a smart construction site. First, a basic information database of on-site personnel and vehicles in the monitor software system is built to provide a benchmark for video analysis and comparison. Then two-way personnel / vehicle identification drums, panoramic cameras and perimeter dome cameras are deployed at the main entrance and perimeter of the substation. Face recognition, intrusion detection and other video analysis technologies for real-time recognition and analysis of video streams are used to check the attendance of personnel and vehicles and to trigger abnormal behavior alarms. The average multi-target recognition accuracy rate of this research application can reach 95% , which greatly improves the efficiency and accuracy of the entry and exit control of personnel and vehicles on the substation construction site.

In order to meet multi-level experimental teaching needs in cultivating talents majoring in communication engineering, an experimental teaching system of optical communication is established and the problems faced by practical optical communication network training are summarized. The commercial passive optical network equipment is expensive, and single equipment does not support multi-person operation simultaneously. The internal communication signal processing can not been seen or touched, and communication service opening process is complicated. The size of teaching laboratory is limited, however the space span of broadband access application scenario is large. The virtual simulation models of passive optical network equipment with high fidelity are established; complex and comprehensive communication service opening process is streamlined; training scenario of broadband access application with large space span is constructed; virtual simulation experiment of passive optical network is thus developed. The whole training procedure is divided into three different stages, including cognitive basis, experimental practice and experimental evaluation. Teaching mode based on virtual-real combination is designed, which improves students’ comprehensive ability to solve problems in the application of optical communication network.

The current research progress of coal mine machinery intelligence always stays on a single function and a single device. The intelligence of complete equipment or complete systems is still in its infancy. Therefore, based on the optimal power algorithm of terminal-to-terminal technology, a complete set of equipment and mining technology combined with a complete set of systems for intelligent is studied. According to the commonly used continuous mining technology, the whole continuous mining face complete set of equipment is constructed, combined with the mining technology, the operation correlation between the intelligent continuous mining face equipment is explored. The terminal-to-terminal direct communication technology is adopted to construct the cellular network framework communication model. Through analysis of multiple frequency bands, the transmission capacity can be obtained, the upper limit of the total power of the frequency band is obtained. The optimal power algorithm is used to solve the uplink interference problem of the shuttle receiving end. In the simulation experiment stage, the system performance is evaluated from multiple perspectives such as Gaussian white noise and multipath interference, different numbers of subcarriers, different multipath fading amplitudes, and whether there is a guard interval. Experimental results show that the research results have an impact on interference and multipath. The effects of fading and implementation methods have ideal suppression effects.

In view of the current limitations of counting high-density population, a population counting method based on Yolov5 is proposed. The input layer is mainly used for Mosaic data enhancement, adaptive anchor frame and adaptive picture zooming technology. Backbone Yolov5 mainly adopts Focus and CSP( Cross Stage Partial) structure. The Neck layer mainly adopts SPP ( Spatial Pyramid Pooling) module and FPN ( Feature Pyramid Networks)+PAN(Pixel Aggregation Network) structure. The output end adopts CIOU_Loss as the loss function mainly for Bounding Box Loss function and DIOU_Loss as average indicator of the NMS(Non Maximum Suppression). Finally the training results are output. Experimental results show that this method can effectively improve the accuracy of crowd counting.

Aiming at the shortcomings of traditional pipeline welding robots such as long start-up time, slow speed response, insensitive speed change, and slow climbing speed, and the problem of variable parameters and multiple disturbances during the orbital motion of the welding robot, an improved FSMC ( Fuzzy Sliding Mode Control) is proposed. The discontinuous intelligent control method combined with fuzzy PI(Proportional Integral) realizes double closed loop control. The inner loop is the current loop control based on fuzzy PI, and the outer loop is the speed loop control based on the improved fuzzy sliding mode algorithm. This kind of control effectively reduces the external disturbance of the orbital motion system, and increases the response speed and robustness of the system. For the speed setting and excessive speed change under variable load, variable speed, and variable welding mode, a follow-up is proposed. The system adopts a tracking and setting system of subregional fine target speed, which effectively divides the range of different welding methods of the welding robot and the precise feedback mechanism. After simulation and experimental testing, the entire control system has the functions of parameter self-tuning and self-adjustment during the welding process. The walking robot has smooth speed transition, fast speed response, good robustness, and stable and reliable pipeline all-position welding.

In order to deepen students' understanding of semiconductor device performance, a comprehensive experimental project and performance tester of static parameters of MOS ( Matal-Oxide-Semiconductor ) is designed and developed according to the characteristics of undergraduate experimental teaching. This tester can measure a number of parameters either by item or by combination. It is helpful to deepen students' understanding of device performance, and is helpful to cultivate students' practical ability. Years of practice have shown that this comprehensive experimental project is helpful to improve students' learning interest and thinking of the basic course “Semiconductor Device Physics". In teaching practice, students change from passive learning to active experimental designers and participants. It has realized the interaction between teachers and students and the cooperation between students, and achieved good teaching results.

Aiming at the problem that patients' enthusiasm for active training is gradually weakened in the traditional pronunciation rehabilitation training of hearing-impaired children, an immersive pronunciation rehabilitation training system for hearing-impaired children based on Virtual Reality VR technology is designed. The system uses the Unity3D Engine to develop the rehabilitation scene and uses the virtual reality helmet as the output to display the virtual environment. It can reproduce the virtual environment with a high degree of restoration and has a high sense of immersion. Through the face capture technology, the standard pronunciation and mouth shape are trained. Through the high-definition visual virtual reality imaging system and hand motion capture system, the human-computer interaction is realized to trigger the training function, and the virtual reality technology is used to better help the hearing-impaired children with immersive pronunciation rehabilitation training. The results show that the pronunciation training method in the system can effectively improve the pronunciation ability of hearing-impaired children and play a positive role in the rehabilitation of patients' language ability.

Aiming at the low detection rate of real-time material procurement business in the traditional material management and control platform, an intranet material procurement information management and control platform based on multi-layer access control is proposed. The internal and external network of the management and control platform share the same database, and the internal and external network data interaction is realized through two- dimensional code scanning. The service index of real-time material procurement business, i. e. the maximum capacity of the platform is calculated. And the maximum length of CFP(Certified Financial Planner) phase, the minimum length of CP(Content Provider) phase are calculated to judge whether the transmission channel is in saturation state, and whether the user access platform is accepted according to the service index and access conditions, so as to realize the information management and control of Intranet material procurement. The experimental results show that the proposed method has good control effect in Intranet, high detection rate of real-time material purchase business and high system load capacity.

The spread of public opinion caused by the outbreak of public emergencies has caused major distress to all sectors of society. An effective method is to find public opinion points in advance and stifle public opinion in the cradle. A system including visual analysis results and attribution analysis is constructed. The system takes public opinion events as the theme, uses data collection technology, and combines sentiment analysis and text processing methods to monitor and analyze public opinion information. Experiment results show that this system is effective for the analysis of public emergencies. By monitoring the changes in mood fluctuations in the process of public opinion, the purpose of assisting the government in decision-making is finally achieved, laying a foundation for the analysis of future public emergencies.

In the face of a large number of enterprise files, it is time-consuming and laborious to label the encryption points simply by manual, and its division standard is affected by human subjective consciousness. It is an important issue for the automatic classification of enterprise documents, which needs to be solved urgently in enterprise confidentiality management is proposed. Therefore, a file dense point labeling system for power grid enterprises based on transformer. It includes file preprocessing, Chinese word segmentation, word vector construction and secret information annotation. The proposed model is trained and tested on the data set constructed by the internal core commercial secret files and ordinary commercial secret files of State Grid Jilin Electric Power Corporation. The accuracy is 97. 79% and the recall is 99. 08% , indicating that the model has achieved high recognition effect. The recognition of secret information is accurate. There are only a few secret information that have not been marked, which prevents the leakage of secret information effectively.

In order to visualize relative positions for multi-target network without base stations, an interactive multi-target map matching visualization scheme is proposed. Localization modeling for multi-target network, multidimensional scaling transformation for solving the relative position are adopted first, then with the help of interactive translation, rotation, scaling, and mirroring operations the target positions of nodes as a whole are matched with local map, thus the mapping between relative coordinates, absolute latitude and longitude is established, and the multi-target scene is realized in indoor environment with UWB(Ultra Wide Band ). The experimental results show that the interactive matching method is simple and feasible, and the interactive response time is less than 100 ms, which has a good user experience.

In recent years, benefiting from the mature application of communication, big data, cloud computing and other technologies, “ Internet + ” has been widely popularized in people’s livelihood, economy and government affairs. With the rapid increase of equipment and explosive growth of data, the network environment becomes increasingly complex and brings huge hidden dangers to network security. The need for the development of network security technology and how to apply artificial intelligence AI(Artificial Intelligence) to help solve some problems is introduced. And how to apply machine learning ML(Machine Learning) to improve network security performance is analyzed in detail for specific domain or specific network technology. First, we summarize existing research work on using AI to combat cyber attacks, including using traditional machine learning approaches and existing deep learning solutions. We then analyzed the counterattacks that the AI itself might be subjected to, dissected their characteristics, and classified the appropriate defenses. We also provide some advanced concepts of artificial intelligence network security technology i. e. how to better apply artificial intelligence to the field of network security in the future.

In view of the multi-element and heterogeneous demand of red resources information integration in Northeast China, we develop a red resources information system for northeast China by using WebGIS technology and combining with the map display, data analysis and database management of geographic information system (GIS: Geographic Information System). On the basis of GIS data acquisition, conversion, integration and database building technology, the northeast red resources database of visual management is established. And the functions of red resources visualization, user management, map service, data management and data display are realized. Through the establishment of northeast red resources information system, the overall planning and visual management of northeast red resources data are realized, which is conducive to promoting the influence of northeast red culture in the whole country, promoting the development of northeast red education and tourism industry.

In order to solve the problem of automatic identification and detection of violent speech on Weibo network, after analyzing the domestic and foreign research on violent text recognition, based on microblog corpus, a data set is established, and data cleaning work is carried out. An improved TFIDF text vectorization method is proposed. The vector of traditional method and the vector constructed by this method are used for the input of the logistic regression model, and the logistic regression violent text classification models of the traditional method and the improved method are created respectively. The above models are evaluated and compared. The experimental results show that the AUC and accuracy of the improved method are 0. 969 and 0. 970, respectively, which are 14. 4% and 15. 5% higher than those of the traditional method.

With the continuous development of wireless communication, the demand for wireless communication channel simulation is gradually increasing. In order to solve the problems of low limitation and poor accuracy of wireless communication channel simulation, a wireless communication channel simulation system is designed by combining hardware and software. The propagation loss of signal can be divided into large scale fading and small scale fading. Clear sky model is used for large-scale fading. Small-scale fading is simulated by multipath prediction model. The hardware part of the simulation system uses FPGAA(Field Programmable Gata Array) and DSP(Digital Signal Processing) to build the system so as to calculate the loss. For the clear-sky model, FPGA is used to calculate the line-of sight propagation loss, diffraction loss, tropospheric loss, atmospheric waveguide layer loss and additional basic loss. In multi-path simulation model FPGA+DSP is used to build the tap delay first simulation system. Qt is used to write visual interface of the upper computer, providing the user input entry and the result display, and using the serial communication way to transmit the model set parameters and output results. After testing and adjusting the simulation system, we finally can better simulate the wireless channel under different environments, so as to improve the convenience of channel simulation and reduce the cost.