JingJing Wang, Dawei Xie, Wei-Hsin Peng, Chao Ding, Huan Wang
{"title":"Identification of Distribution Network Parameters Based on Particle Swarm Optimization Method","authors":"JingJing Wang, Dawei Xie, Wei-Hsin Peng, Chao Ding, Huan Wang","doi":"10.1109/ISPDS58840.2023.10235740","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235740","url":null,"abstract":"Incomplete and inaccurate information of line parameters affects the condition monitoring, analysis and control of the distribution network. The accuracy of distribution network parameters is of great significance for its optimal operation. However, in recent years, the structure of distribution networks has become more and more complex, and there are large errors between the distribution network parameters and the actual values calculated by traditional theories. At the same time, there are errors in the actual measurement data, which makes it difficult to identify its parameter information accurately. Many current methods require the use of a large number of phasor measurement units to measure the phase angle of node voltages, which is currently not feasible for relatively conventional distribution networks. For this reason, this paper uses different parameter equations for the line for parameter identification by considering the presence or absence of phase measurement units $mutext{PMU}$. Firstly, the bad data are removed and then the parameter identification equation is derived and established by the formula, and the parameter identification is carried out by the particle swarm optimization algorithm, and the 13-node system is tested. The results verify the feasibility and good performance of the proposed parameter identification method, which is beneficial to the safe and stable operation of the distribution network.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121827940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on Mask Wearing Detection Algorithm in Complex Scenes","authors":"Dongzhao Zhang, Hongqiong Huang","doi":"10.1109/ISPDS58840.2023.10235428","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235428","url":null,"abstract":"In recent years, various viral outbreaks have had a huge impact on the world, and wearing masks has become an important public health measure. Incorrect wearing of masks can increase the risk of virus transmission, so how to detect and monitor the correctness of mask wearing has become an important research direction. However, complex factors such as occlusion, dense crowds, and small-scale targets often exist in natural scenes, which have an impact on the detection performance of wearing masks. To address this issue, a mask wearing detection algorithm incorporating attention mechanism is proposed based on the RetinaNet algorithm. Specifically, it is to establish a branch in the residual learning unit of the original target detection network RetinaNet, and incorporate an attention mechanism with feature recalibration characteristics into this branch. The idea of this attention mechanism originates from the human visual system. In deep learning, it can be understood as obtaining the importance of each feature map channel through feature learning, and then assigning weights to each channel based on this importance, thereby achieving the effect of enhancing useful information and suppressing useless information. Due to severe facial occlusion in densely populated scenes, incorporating attention mechanisms into the object detection network can better learn local features of the target by enhancing important information weights, thereby achieving better detection performance. The experimental results show that compared with YOLOv3 algorithm, the detection accuracy of this algorithm on facial targets and facial mask wearing targets is improved by 0.11% and 3.16%, with an average detection speed improvement of 2 FPS.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122768908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimization of Throughput and Fairness Performance in Multi-Rate WLAN Network","authors":"Guonian Jin, Shejie Lu, Siping Hu","doi":"10.1109/ISPDS58840.2023.10235398","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235398","url":null,"abstract":"In IEEE 802.11 wireless networks, as the number of network nodes increases, impact increase accordingly, channel utilization decreases, and aggregate throughput decreases, resulting in system function degradation. In response to this problem, this article proposes a solution: number nodes-based minimum contention window dynamic adjustment (NNMCWDA), which dynamically adjusts the minimum competition window to achieve channel resources' rational allocation. Simulation tests have shown that the NNMCWDA method can effectively work out the problem of channel utilization degradation, greatly improve fairness performance, and increase system throughput. On the side, what's particularly nice about this approach is that it requires only minor modifications over 802.11. In multi-rate networks, the minimum competition window is dynamically adjusted in line with the transfer rate. Its physical realization is very simple and handy, and can be achieved only by writing driver programs without any hardware modifications. Therefore, it is easy to execute and popularize in an actual network.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"415 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123002154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An Innovative Similar Complaint Recommendation Model Integrating Semantic and Graph Embeddings","authors":"Ruisi Li, Renwei Ou, Dejun Wang","doi":"10.1109/ISPDS58840.2023.10235356","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235356","url":null,"abstract":"Artificial intelligence plays a vital role in driving the intelligentization of the complaint hotline 12345. To assist hotline operators in accurately understanding public demands and allocating them to municipal departments, we propose BERT-Graph that recommends historical complaint to pending complaint through a dual-tower recommendation model integrating semantic and graph embeddings. Since the complaint records are sparse, colloquial and similar complaint texts may transcribed semantically different by operators, we construct an event interact graph for the complaint text to avoid implying sole reliance on semantic information. We consider each complaint text as a node, the co-occurrence times of key words calculated by TF-IDF between texts as weighted edges, and we use the Graph Attention Network (GAT) to obtain graph embeddings. After obtaining the semantic embedding through pre-trained language model BERT, we concat it with the graph embedding and feed it to the dual-tower model. We add a self-attention module as the neural network in the dual-tower model and we use a fully connected layer to output the similarity score of the input texts. Experimental results on complaint hotline 12345 datasets of the Chinese city of Wuhan demonstrate that compared to other baseline models, BERT-Graph shows a significant improvement on the Mean Squared Error (MSE) metric, from 0.86 to 0.24.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116612233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research and Verification of the Scheme to Ensure Voice Over Wi-Fi Call Quality","authors":"Xiaojing An","doi":"10.1109/ISPDS58840.2023.10235424","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235424","url":null,"abstract":"Operators' cellular networks have problems such as weak coverage, insufficient deep coverage, and high-level pilot pollution, which lead to poor voice call experience. The VoWiFi service can make up for the above problems caused by the cellular network. However, V oWiFi also has many factors that affect its call quality. When Wi-Fi data services and VoWiFi services occur at the same time, this paper proposes a method to suspend other Wi-Fi data services to ensure VoWiFi call quality according to service priority, and verifies the effectiveness of this method.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117303870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Robust and Efficient Measurement Method of Lumbar Lordosis Based on Three Dimensional Curvature Line","authors":"Yingbin Liu, Yanbin Guo, Shuai Yu, Yong Li, Jingran Kang, Yuanyang Zhao, Guoping Wang","doi":"10.1109/ISPDS58840.2023.10235615","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235615","url":null,"abstract":"Lumbar lordosis (LL) is an important physiological parameter used to evaluate the degree of curvature of human lumbar in the sagittal plane. The accurate and reliable measurement of lumbar lordosis is of great significance for diagnosis, clinical treatment and rehabilitation evaluation in the care of degenerative lumbar diseases such as lumbar disc herniation and lumbar spondylolisthesis. Cobb and Centroid are the standard methods for measuring LL in clinical practice but show poor reliablity and low efficiency since the accuracy of the methods depends on the subjective judgment and clinical experience of operators. Here, we present a robust and efficient method for measuring LL. The core of the method involves an algorithm based on the variable step Runge-Kutta (RK) integral, extracting the three-dimensional curvature line (3DCL) on the lumbar simplified model constructed from computed tomography (CT) images. Our experimental findings show that the standard error (STD) and root mean square error (RMSE) of the method proposed in this paper (0.43 $pm$ 0.506, 0.383 $pm$ 0.441) are superior to the Cobb method (1.915 $pm$ 1.237, 1.712 $pm$ 1.079) and the centroid method (1.789 $pm$ 1.297, 1.601 $pm$ 1.134)(p<0.05). Moreover, the correlation coefficient R between the result of the single measurement in our method and the average of multiple measurements in the standard method was relatively high (0.984, 0.978). The results of our comparative analysis showed that the proposed method is more robust and efficient, having the potential for improving the diagnosis and therapeutic monitoring of lumbar diseases.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129146218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Network Topology Analysis of NCW Based on SPDI Model and Modularity Optimization Algorithm","authors":"Xufan Wan, Guangjun Zeng, Yunjun Lu, Kebin Chen, Zixi Xu","doi":"10.1109/ISPDS58840.2023.10235502","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235502","url":null,"abstract":"Recent years, network-centric warfare (NCW) has become a hot topic in the global military theory circle. The NCW operation style, which shows the characteristics of precision strike, flexible maneuvering, platform integration and so on. In this article, we abstract the combat entities of NCW as nodes and construct the SPDI model. Based on the modularity optimization algorithm in Gephi, we make an in-depth study on the topology structure of the combat network and analyze the characteristics. The specific methods are as follows: First, we take NCW as a sample and abstract the combat network topology based on SPDI model. Then, relying on the modularity optimization algorithm in Gephi, we obtain the optimal community division. Last, according to the combat network topology and the optimal community division, we analyze the degree distribution and community structure, and find the key nodes in different node types, together with the local areas which have greater influence on the whole network.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126912956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jie Liu, Zhicong Li, Zhenyong Liu, Jiarong Ru, Minsheng Chen, Zelong Cai, Chengqiang Huang, Xianxin Lin
{"title":"The Design of Digital Inkjet Printing Control System Based on FPGA","authors":"Jie Liu, Zhicong Li, Zhenyong Liu, Jiarong Ru, Minsheng Chen, Zelong Cai, Chengqiang Huang, Xianxin Lin","doi":"10.1109/ISPDS58840.2023.10235632","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235632","url":null,"abstract":"Aiming at the high cost and low real-time performance of adjusting image grayscale in traditional inkjet printing systems, a digital inkjet printing Control system based on FPGA(Field Programmable Gate Array) has been designed to balance the speed differences across clock domains between the FPGA and SDRAM(Synchronous Dynamic Random Access Memory), achieving real-time adjusting the grayscale of images in a single print head devices. The system selects FPGA as the main control chip and SDRAM as the memory, simplifies the SDRAM interface using asynchronous FIFO(First Input First Output), and inputs images through the serial port and stores them in SDRAM. Before printing each column of pixel data, a DAC(Digital-to-Analog Conversion) circuit and a proportional amplification circuit are used to generate an adjustable amplitude drive voltage. By adjusting the amplitude of the drive voltage, the ink droplet amount for each column of pixels can be controlled during printing. The system experimental platform that has been constructed verifies that the system possesses characteristics of strong real-time grayscale adjustment, low hardware cost, and high reliability. It can satisfy the requirement of quick and small-batch printing with grayscale adjustment, and it has significant importance and value in advancing technology in the inkjet printing field.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"54 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116330612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Kinematics Parameter Analysis of Long Jump Based on Human Posture Vision","authors":"Shuai-yu Zhou, Bin Li, Jing Chen, Hao Yuan","doi":"10.1109/ISPDS58840.2023.10235733","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235733","url":null,"abstract":"In order to enable long jumpers or coaches to accurately analyze each training result and help athletes improve their performance, this paper designs a method of analyzing the kinematic parameters of long jump based on human posture vision. Firstly, the attitude recognition model is used to analyze the athlete's posture, and the key point position information of the athlete is given. At the same time, the trajectory of the movement is accurately tracked through perspective transformation and background difference, and the kinematic parameters of the long jump are obtained based on the trajectory.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"5 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113979533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Deep Learning-Based Text Processing Method for Power Grid Information System","authors":"Xuesong Qi, Li Tian, Jia Sun, Zhenxin Li, Yusong Li, Xiaozhuo Guan","doi":"10.1109/ISPDS58840.2023.10235409","DOIUrl":"https://doi.org/10.1109/ISPDS58840.2023.10235409","url":null,"abstract":"The existing power grid dispatch automation system lacks the ability to comprehend natural language information. This limitation hampers the efficiency of operating personnel when they read and process power grid operation event information, which is often presented in the form of natural language sentences. These personnel can only manually search for relevant event information from a vast array of unordered sentences, leading to inefficiencies and increased chances of errors. To address this issue, this article proposes a data synthesis processing model that combines Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) networks. This model effectively handles large batches of Chinese short textual sequences by considering their contextual correlations. To align with the format of the composite information system, the model converts Chinese short texts into English letters and numerical short sequences. Moreover, the proposed model analyzes the contextual relationships within the text to identify the types of equipment described in each short text entry. It then generates the corresponding event codes and device codes for each entry. By employing this model, power grid operation event information can be processed, enabling the identification and classification of power grid operation events based on equipment and behavior. Consequently, this enhances the understanding of power grid operators and enables further processing by other business systems, thereby elevating the intelligence level of the power system as a whole.","PeriodicalId":237615,"journal":{"name":"2023 4th International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128024821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}