In order to improve the efficiency of fault diagnosis in low-voltage distribution network, a fault data classification algorithm based on rough set and SVM is proposed in this paper. Firstly, the redundant and incomplete data stored in the fault symptom data can be reduced by using rough set, and a multi class support vector machine classifier is established and trained with the reduced new sample data to realize the accurate fault classification based on the optimal decision attributes. The test results of the software simulation model show that the proposed method can not only identify the common faults of typical medium and low voltage distribution networks with high accuracy, but also has high accuracy and good stability, which is helpful to improve the industrial production efficiency.
In order to improve the voltage control accuracy of AVC system and reduce time-consuming, a new voltage control strategy of AVC system based on data mining is proposed in this paper. The fuzzy c-means clustering mining method is used to mine the voltage data of AVC system to obtain the key voltage parameters. On this basis, taking the minimum active power loss of AVC system as the goal, the voltage control objective function of AVC system is constructed. Under the constraints, the objective function is solved to complete the voltage control. The experimental results show that compared with the traditional control method, the control accuracy of this method is higher and the control time is shorter.
Power IoT dumb terminals such as web cameras and network printers based on IP protocol and no user interaction interface are usually based on embedded system development, hard to update program firmware, limited computing resources, and simple security authentication mechanisms. The upgrade is easily controlled by the attacker to initiate a network attack. Aiming at the above problems, this paper designs and implements a dumb terminal security management and control system based on traffic feature recognition. The system extracts the traffic characteristics of the terminal and realizes the identity authentication and behavior supervision of the terminal. When the device is accessed, the static characteristics of the traffic of the terminal are extracted to implement identity authentication. After the device is accessed, the abnormal behavior of the network access terminal is analyzed and analyzed, and the session connection is blocked. The performance of the system is better in the experimental environment and the measured environment. The accuracy of equipment identification is 96.6%, and the accuracy of abnormal detection is 97.7%. It can effectively detect DOS, port scanning and other network attacks.
In recent years, the installed capacity of new energy has increased year by year. The characteristics of randomness and volatility of new energy make the safe and stable operation of power system suffer severe challenges. The participation of high energy-consuming enterprises such as cement in power grid dispatching can effectively relieve the pressure of safe and stable operation of power grid brought by new energy generation. This paper takes the new dry cement production process as the research object, mainly studies the electricity-using characteristics of cement production process. At the same time, starting from the three aspects of the cement production process, the labor cost of load adjustment and the time-of-use electricity price strategy considering the actual load change, we take the sum of power cost and labor cost of cement plant as the objective function, the optimization scheduling modeling of adjustable load of cement plant is carried out, and the great potential of the new dry cement industry to participate in power grid peak cutting and valley filling has been proved by an example.
With the rapid development and growth of the communication industry, the commutation system has higher and higher requirements for the performance of antennas. The radiation of omni-antennas can cover the angular range of all directions. Its simple structure, low cost, and miniaturization make it more and more popular. More and more scholars have paid attention to the wide-band omni-directional microstrip antenna research, which has a wide range of application prospects and practical significance. This article takes the mobile communication system as the background, designs and realizes a broadband omni-antenna. The antenna is based on an array of doublet antenna, with triangular balun balanced feed, which effectively broadens the bandwidth of the antenna through electromagnetic coupling feed, all indicators meet the requirements, and the simulation results match the actual test results well.
Intelligent vehicle-infrastructure system (IVIS) uses advanced wireless communications and the new generation of Internet technology to enable comprehensive real-time dynamic information exchange between different vehicles and between vehicles and roads. It also provides active vehicle safety control and collaborative road management based on omnidirectional acquisition and fusion of traffic information, so as to fully realize effective coordination among people, vehicles and roads. This ensures traffic safety and improves traffic efficiency. In light of the urgent need to evaluate IVIS and the technology bottleneck in current testing, a flexible, reliable method for evaluating IVIS is implemented in this paper. First, an industrial camera with quick response and low delay is used to acquire video warning signals from on-board IVIS equipment in real time. The collected images are then accurately time-aligned through a synchronous acquisition scheme based on BeiDou Pulse Per Second (PPS). After that, different grayscale template matching algorithms are used to identify video warning signals. Then these algorithms are quantitatively evaluated and compared based on the accuracy and timeliness of recognition so as to select the optimal template matching algorithm.
Convolutional neural networks (CNNs) have been widely used in the field of target recognition, and it has become a challenge to deploy convolutional neural networks with high accuracy and the ability to recognize multiple targets on a single chip. In order to save FPGA resources and improve target recognition accuracy with limited resources, this paper proposes an idea of constantly changing the weights (w) and biases (b) of the convolutional and fully connected layers in order to recognize different kinds of targets, and dynamically configuring w and b into the neural network to achieve the effect of improving target recognition accuracy[1]. The improved network is deployed on a Xilinx ZYNQ xc7z020 FPGA, which effectively improves the target recognition accuracy with fast recognition speed and low power consumption.
In order to improve the accuracy of athlete's wrong movement detection and reduce the detection time, an athlete's wrong movement detection method based on multi feature map fusion is proposed. Firstly, the acceleration sensor is used as the core to collect athletes' action data, and the optimal classification function is used to classify and recognize the collected image data. Secondly, histogram equalization is used to measure the similarity of athletes' action images, and the correlation factor of athletes' action images multi feature image fusion is used to complete the multi feature image fusion. Finally, the edge contour information of multi feature map is clustered to complete the athlete's wrong action detection. The experimental results show that compared with the traditional error action detection method, the maximum detection accuracy is 98%, and the maximum detection time is no more than 0.4 min.
In order to improve the stability and reliability of missile during flight, this paper designs an active microstrip antenna for missile based on GNSS. It covers BD B1, GPS L1 and Glnoass L1. By combining microstrip antenna with circuit, broadband, miniaturization and high gain are realized. The metallized through hole in the antenna center is used to make DC grounding and eliminate the electrostatic interference at the RF channel input, increase the reliability of antenna installation and reduce the resonance of antenna in the flight environment. The radome is made of F4BM to improve the heat insulation of the antenna. In the 1584±25MHz operation bandwidth, the range of antenna is greater than 130°when main lobe gain is greater than 7dBic, the axial ratio is less than 3dB, the VSWR is less than 1.8, the low noise amplifier gain is 36dB ± 0.5dB, and the noise factor is less than 1.79dB. The design can meet the requirements of missile satellite navigation application.
In view of the problems of excessive load rate of trunk lines and low voltage qualification rate in the traditional active distribution network scheduling method, an active distribution network optimization scheduling strategy of multi-source cooperation is proposed. Through the analysis of multi-source cooperation of active distribution network, the optimization objective function of active distribution network is determined, and the scheduling analysis is carried out according to the objective. The experimental results show that the improved scheduling strategy has some advantages, such as low load rate and high voltage qualification rate.
The rapid development of the Internet has posed a more serious challenge to communication. The widely used authentication scheme is now a centralized authentication scheme, which obviously cannot meet the demand in the era of explosive information growth, and the development of blockchain has provided the impetus for the development of decentralization, so the design of decentralized authentication scheme has become a hot topic, and the use of blockchain and asymmetric encryption technology can solve the authentication problem. However, due to the limitations of blockchain, it cannot meet the conditions of non-Internet. A two-way authentication needs to be designed to solve the trust problem between the client and the service node. In this paper, we propose a two-way authentication scheme based on digital signature using elliptic encryption algorithm to achieve reliable authentication of identity.
In order to study the impact of autonomous emergency braking (AEB) intervention on vehicle safety, design a simplified trolley to simulate AEB braking and proved the difference between the dummy and the real human body in kinematic response. Through the out-of-position test of volunteers on the real vehicle, the factors affecting the out-of-position displacement were analyzed. The most important factor affecting out-of-position displacement was the peak braking deceleration, with the increase of the peak braking deceleration, the out-of-position displacement will increase, while increasing the initial speed under the same braking deceleration has little effect on out-of-position displacement. Total Human Model for Safety (THUMS) was used to simulated the occupant kinematics change during AEB braking and occupant injuries in frontal collision. Compared the occupant injuries with and without AEB braking, the performance of the current restraint system can not meet the requirements of occupant protection with AEB braking and the risk of occupant injuries was greatly increased.
With the development and popularization of autonomous driving technology, its safety issues have attracted more and more attention from the public. The performance limitations and insufficient functions of the perception system and execution system of autonomous driving, the algorithm defects of perception algorithms, the decision-making algorithms, the control algorithms, special weather and traffic, and the safety risks caused by reasonably foreseeable misuse of personnel, are the biggest challenges in the development and industrialization of autonomous vehicles. Aiming at solving these challenges, the concept of Safety of The Intended Functionality (SOTIF) of autonomous driving is proposed in this work. The conceptual definition and research status of the SOTIF of the automatic driving system were introduced, the main challenges and the system analysis methods of the SOTIF of the automatic driving system were analyzed, and research outlooks were put forward.
In smart factories, the wide application of a variety of intelligent manufacturing devices and advanced sensors leads to the increasing demand for data exchange between different devices. In order to realize the flexible forwarding of multi-source manufacturing data in heterogeneous networks and the optimal utilization of network resources, in this article, we combined smartphone manufacturing scenarios, referred to the data flow reconstruction algorithm in related documents, analyzed the high dynamics of the data flow and proposed a data flow reconstruction algorithm based on routing-protocol, according to the particularity of different data forwarding and processing procedures, redesigned and defined the data flow in the scene. Finally, the results of Mininet simulation experiments showed that our proposed algorithm reduces the average transmission delay.
Aiming at the problem of inaccurate monitoring caused by too little inclination analysis in traditional monitoring methods, an intelligent monitoring method of tunnel section deformation based on inclined photogrammetry is proposed. Firstly, the tilt photogrammetry technology is analyzed; Secondly, based on the inclined photogrammetry technology, the tunnel section information is extracted, the invalid section information is eliminated, and the section thickness is calculated; then; Based on the back removal of the section, the influence of the section inclination on the tunnel section deformation is determined to realize the final section deformation monitoring; Finally, the tunnel section deformation monitoring method is verified by experiments. The results show that this method can intuitively and comprehensively monitor the overall deformation of the tunnel.
In order to realize the miniaturization of the antenna used in the navigation system, the slot array is slotted in the microstrip antenna substrate, and the coupling cavity is loaded at both ends of the slot to optimize the radiation characteristics of the slot array; The compact PBG structure is loaded at the three corners of the ground plate to reduce the surface radiation, so as to realize the miniaturization and circular polarization of the microstrip antenna. The antenna adopts an equilateral triangle structure, the side length is only 18.7% of the wavelength (the wavelength is the wavelength corresponding to the central working frequency of the antenna), and the working frequency band is 1.57Ghz ~ 1.58Ghz. The experimental results show that the antenna covers the working frequency band of GPS navigation system and Beidou navigation system.
The robustness of a virtual enterprise (VE) system is an essential prerequisite for its reliable and efficient performance. This paper performs a study on robust simulation and analysis of VE network. Firstly, the VE network is modeled with an expression as (N, E, W) taking advantage of complex theory. Secondly, two defined global metrics, namely NE and NA, are defined to evaluate the robustness properties of VE network from the perspective of both the nodes and the edges, and static as well as dynamic robustness analysis on the two NE networks is launched. Finally, several experiments are undertaken to analyze the dynamic robustness of the two VE networks reacting to random or target attacks from the point of nodes and edges. Through the experiment result, it is found that the VE1 network is more robust than VE2 facing random attacks, while VE2 network is relatively more immune to target attacks comparing to VE1. Methods utilized in this article can provide an overall indication of the characteristics and robustness analysis of the VE network in static status as well as dynamical ones and also can provide an early caution of failures facing random or target attacks.
Aiming at the problems of image encryption algorithm in terms of security and encryption effect, an image encryption algorithm based on multiple chaotic maps and DNA encoding is proposed. First, using Logistic map to randomly scramble the rows and columns of the plaintext image, and convert the result into a binary matrix; second, according to the DNA coding rules, convert the binary matrix into a DNA matrix, and use 2D-LSMM map and Logistic-Sine system to perform in DNA encoding operation, XOR operation is performed on the obtained encoding matrix to obtain an encrypted DNA matrix; finally, according to the DNA decoding rules, the encrypted DNA matrix is converted into a digital image to obtain a ciphertext image. Experimental simulation and performance analysis show that the algorithm has better encryption and decryption effects, larger key space, and higher security and robustness.
In order to overcome new business changes that bring new security threats and challenges to many Industrial Internet of Things (IIoT) fields such as smart grids, smart factories, and smart transportation, the paper proposed the architecture of the industrial Internet of Things system, and analyzed the security threats of the industrial Internet of Things system. Combining various attack methods, targeted security protection strategies for the perception layer, network layer, platform layer and application layer are designed. The results show that the security protection strategy can effectively meet the security protection requirements of IIoT systems.
In order to solve the problem of limit sample in the detection of surface coating defects of plastic parts, an improved image dataset enhancement algorithm of Deep Convolutional Generative Adversarial Networks (DCGAN) is proposed, which is based on the DCGAN.The algorithm improves the existing activation function without changing the computational effort, and enhances the richness of the generated features, alleviating the gradient disappearance or explosion during the training process of the network. Next, introduces the deep convolutional residual blocks into the existing generator structure to obtain relatively high-resolution generated images, compares the results with Generative Adversarial Networks(GAN) and DCGAN. The results show that the improved DCGAN has a more significant improvement in the effect of image sample generation for surface painting of plastic parts.
