Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00072
Lei Zhang, Zhengying Wei, Weibing Jia, H. Wei
Multi-user random water distribution technology solves the shortage of available water and unreasonable water distribution in multi-user irrigation zones. With the goal of maximizing irrigation revenue, a multi-user random opportunity constrained programming water allocation model is established. Monte Carlo is used to simulate random variables. A hybrid intelligent algorithm is applied to solve the water allocation model to obtain the best allocation of different crops in different growth periods. Compared with traditional irrigation methods, the multi-users water optimal allocation model can increase the income by 21.32%.
{"title":"Multi-user Random Water Distribution in Greenhouse Based on Hybrid Intelligent Algorithm","authors":"Lei Zhang, Zhengying Wei, Weibing Jia, H. Wei","doi":"10.1109/ICCEAI52939.2021.00072","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00072","url":null,"abstract":"Multi-user random water distribution technology solves the shortage of available water and unreasonable water distribution in multi-user irrigation zones. With the goal of maximizing irrigation revenue, a multi-user random opportunity constrained programming water allocation model is established. Monte Carlo is used to simulate random variables. A hybrid intelligent algorithm is applied to solve the water allocation model to obtain the best allocation of different crops in different growth periods. Compared with traditional irrigation methods, the multi-users water optimal allocation model can increase the income by 21.32%.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130678064","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00039
Cheng Chen, L. Fan
Deep learning is the current advanced solution for remote sensing segmentation. Massive high-quality training datasets are the basic inputs to deep learning networks for solving the segmentation problems. Most of the existing remotely sensed image datasets have low segmentation accuracy due to their coarse spatial resolution and the susceptibility to image noise. Image augmentation is a technical means of effectively solving deep learning trainings in small and/or low-quality training datasets, which has continuously accompanied the development of deep learning and machine vision. Many augmentation techniques and methods have been proposed to enrich and augment the training datasets and to improve the generalization ability of neural networks. Common image augmentation methods are based mainly on image transformations, such as photometric changes, flips, rotations, dithering and blurring. In this paper, the segmentation task of multispectral remote sensing data is validated by augmentation methods. The segmentation accuracy was found to be 96.10%, which is higher than that (92.36%) of the corresponding un-augmented data.
{"title":"Scene segmentation of remotely sensed images with data augmentation using U-net++","authors":"Cheng Chen, L. Fan","doi":"10.1109/ICCEAI52939.2021.00039","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00039","url":null,"abstract":"Deep learning is the current advanced solution for remote sensing segmentation. Massive high-quality training datasets are the basic inputs to deep learning networks for solving the segmentation problems. Most of the existing remotely sensed image datasets have low segmentation accuracy due to their coarse spatial resolution and the susceptibility to image noise. Image augmentation is a technical means of effectively solving deep learning trainings in small and/or low-quality training datasets, which has continuously accompanied the development of deep learning and machine vision. Many augmentation techniques and methods have been proposed to enrich and augment the training datasets and to improve the generalization ability of neural networks. Common image augmentation methods are based mainly on image transformations, such as photometric changes, flips, rotations, dithering and blurring. In this paper, the segmentation task of multispectral remote sensing data is validated by augmentation methods. The segmentation accuracy was found to be 96.10%, which is higher than that (92.36%) of the corresponding un-augmented data.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"238 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116051747","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00073
Yongsheng Xie, C. Feng, Chenhao Gai, Changgang Li
Emergency load shedding (ELS) is an essential measure to prevent power system accidents from expanding. Economy and security need to be optimized comprehensively for ELS. In this paper, an ELS optimization model is established, which takes the minimum load shedding amount as the objective function and the transient angle security, transient voltage deviation acceptability, transient frequency deviation acceptability, maximum controllable load as constraints. The social learning-based particle swarm optimization (SL-PSO) algorithm is proposed to solve the ELS optimization problem, which adopts adaptive parameters. The portable and open-source power system dynamic simulation toolkit (STEPS) is used for numerical simulation to check the feasibility of the solution. Finally, the efficiency of the solution is improved by parallel computing. The proposed model and algorithm are validated with the IEEE 39 bus test system.
{"title":"Optimization of Emergency Load Shedding Employing Social Learning-Based PSO","authors":"Yongsheng Xie, C. Feng, Chenhao Gai, Changgang Li","doi":"10.1109/ICCEAI52939.2021.00073","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00073","url":null,"abstract":"Emergency load shedding (ELS) is an essential measure to prevent power system accidents from expanding. Economy and security need to be optimized comprehensively for ELS. In this paper, an ELS optimization model is established, which takes the minimum load shedding amount as the objective function and the transient angle security, transient voltage deviation acceptability, transient frequency deviation acceptability, maximum controllable load as constraints. The social learning-based particle swarm optimization (SL-PSO) algorithm is proposed to solve the ELS optimization problem, which adopts adaptive parameters. The portable and open-source power system dynamic simulation toolkit (STEPS) is used for numerical simulation to check the feasibility of the solution. Finally, the efficiency of the solution is improved by parallel computing. The proposed model and algorithm are validated with the IEEE 39 bus test system.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124172984","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00076
Hao He, Yong Yang
Intrinsically disordered proteins (IDPs) possess flexible 3-D structures, which make them play an important role in a variety of biological functions. We develop a method to predict intrinsically disordered proteins based on feature selection and convolutional neural networks (CNN). The combination of structural, physicochemical and evolutionary properties is used to describe the differences between disordered and ordered regions. Especially, to highlight the correlation between the target residue and adjacent residues, multiple windows are selected to preprocess the selected properties. After that, these calculated properties are combined into the feature matrix to predict IDPs through the constructed CNN. Our method is training as well as testing based on the DisProt database. The simulation results show that the proposed method can predict intrinsically disordered proteins effectively, and the performance is competitive in comparison with IsUnstruct and ESpritz.
{"title":"Prediction of Intrinsically Disordered Proteins with Convolutional Neural Networks based on Feature Selection","authors":"Hao He, Yong Yang","doi":"10.1109/ICCEAI52939.2021.00076","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00076","url":null,"abstract":"Intrinsically disordered proteins (IDPs) possess flexible 3-D structures, which make them play an important role in a variety of biological functions. We develop a method to predict intrinsically disordered proteins based on feature selection and convolutional neural networks (CNN). The combination of structural, physicochemical and evolutionary properties is used to describe the differences between disordered and ordered regions. Especially, to highlight the correlation between the target residue and adjacent residues, multiple windows are selected to preprocess the selected properties. After that, these calculated properties are combined into the feature matrix to predict IDPs through the constructed CNN. Our method is training as well as testing based on the DisProt database. The simulation results show that the proposed method can predict intrinsically disordered proteins effectively, and the performance is competitive in comparison with IsUnstruct and ESpritz.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122170621","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00040
Yu Chen, Jun Zhang, Zhicheng Wang, Gejian Liao, Shu Liu, Hai Tan, Guowei Yang, Ying Fang, Shuai Wang, Zhaoqun Sun
Massive small files access is the main challenge for the Hadoop Distributed File System. To solve these problems, we present a new Algorithm of archive file, A Faster Read and Less Storage Algorithm for Small Files on Hadoop. A new logical file name is used to identify the file which generated by the pair in the name node. Our experiments show that the algorithm is around 76.6% faster than original HDFS in the time of file storing, and around 31.9.6% faster than original HDFS in the time of file reading, around 73.9% less than original HDFS in the memory consumption of namenode.
{"title":"A Faster Read and Less Storage Algorithm for Small Files on Hadoop","authors":"Yu Chen, Jun Zhang, Zhicheng Wang, Gejian Liao, Shu Liu, Hai Tan, Guowei Yang, Ying Fang, Shuai Wang, Zhaoqun Sun","doi":"10.1109/ICCEAI52939.2021.00040","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00040","url":null,"abstract":"Massive small files access is the main challenge for the Hadoop Distributed File System. To solve these problems, we present a new Algorithm of archive file, A Faster Read and Less Storage Algorithm for Small Files on Hadoop. A new logical file name is used to identify the file which generated by the pair in the name node. Our experiments show that the algorithm is around 76.6% faster than original HDFS in the time of file storing, and around 31.9.6% faster than original HDFS in the time of file reading, around 73.9% less than original HDFS in the memory consumption of namenode.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115730555","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00038
Yan Zhimei, Pan Ping
The risk is uncertain. It is impossible to determine the influence of a risk by default risk, but only to cognize the influence degree of its existence. In an open behavior system, according to the quantum theory, the risk is reflected in the wave function of the behavioral system, and at risk, the dynamic evolution of the system is determined by the environmental Hamiltonian in system Hamiltonian. When the environment is perfectly correlated to the eigenvalue of the behavior subject, it does not affect its evolution, and the risk is controllable. Otherwise, the risk will harm the evolution of the behavior subject. Meanwhile, the corresponding control strategy is proposed through the dynamic analysis of the evolution of quantum wave function.
{"title":"Risk Analysis Based on Quantum Theory","authors":"Yan Zhimei, Pan Ping","doi":"10.1109/ICCEAI52939.2021.00038","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00038","url":null,"abstract":"The risk is uncertain. It is impossible to determine the influence of a risk by default risk, but only to cognize the influence degree of its existence. In an open behavior system, according to the quantum theory, the risk is reflected in the wave function of the behavioral system, and at risk, the dynamic evolution of the system is determined by the environmental Hamiltonian in system Hamiltonian. When the environment is perfectly correlated to the eigenvalue of the behavior subject, it does not affect its evolution, and the risk is controllable. Otherwise, the risk will harm the evolution of the behavior subject. Meanwhile, the corresponding control strategy is proposed through the dynamic analysis of the evolution of quantum wave function.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129597767","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00020
Zhong Guoyun, Liu Jun, Hong Yang, Liu Meifeng, Sun Hongyang
A fine-grained image classification method based on generating adversarial networks with SIFT (Scale Invariant Feature Transform) texture input is proposed to improve the recognition ratio of fine-grained image classification by deep learning. For the phenomenon of data sets that require a large amount of labeled information for strong supervised learning, active learning capabilities of generative and adversarial networks and excellent image modeling capabilities for target classification images are used to achieve active learning of image features. Then the difficulty of data set construction and the computational complexity are reduced, and the disturbance to the network that may be caused by manually set labeled boxes is lessened. The input method of generating the adversarial network to is fixed to balance the authenticity and diversity of the generated samples. The idea of image restoration is considered. The random input method of the generative adversarial network that combines image feature points and random noise to is used to reduce the training difficulty of the generative and adversarial network. Experiments results show that our method outperformances the current deep learning methods in fine-grained image classification.
{"title":"Fine-grained image classification method based on generating adversarial networks with SIFT texture input","authors":"Zhong Guoyun, Liu Jun, Hong Yang, Liu Meifeng, Sun Hongyang","doi":"10.1109/ICCEAI52939.2021.00020","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00020","url":null,"abstract":"A fine-grained image classification method based on generating adversarial networks with SIFT (Scale Invariant Feature Transform) texture input is proposed to improve the recognition ratio of fine-grained image classification by deep learning. For the phenomenon of data sets that require a large amount of labeled information for strong supervised learning, active learning capabilities of generative and adversarial networks and excellent image modeling capabilities for target classification images are used to achieve active learning of image features. Then the difficulty of data set construction and the computational complexity are reduced, and the disturbance to the network that may be caused by manually set labeled boxes is lessened. The input method of generating the adversarial network to is fixed to balance the authenticity and diversity of the generated samples. The idea of image restoration is considered. The random input method of the generative adversarial network that combines image feature points and random noise to is used to reduce the training difficulty of the generative and adversarial network. Experiments results show that our method outperformances the current deep learning methods in fine-grained image classification.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129895098","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00062
Zhuo Wang, Huan Li, Bin Nie, Jianqiang Du, Yuwen Du, Yufeng Chen
Aiming at the dimensional disaster and over-fitting problems in data analysis, this paper proposes a feature selection method using hybrid integration of difference models and random forests (Integrate-RF), firstly, Integrate-RF use CART, CHAID, SVM, BN, NN, K-Means, Kohonen to evaluate the importance of features, and then, for the above seven sorts, Integrate-RF use the arithmetic average method to calculate the importance of the features; secondly, Integrate-RF select the most important features from the remaining features into features subset, and use random forest classification to get the corresponding out-of-bag(OOB) data classification error rate; finally, the optimal features subset can be selected based on the OOB data classification error rate. Experiments show that feature selection methods proposed in this paper effectively reduces the data dimension, selects features better and more adaptable.
{"title":"Feature selection using different evaluate strategy and random forests","authors":"Zhuo Wang, Huan Li, Bin Nie, Jianqiang Du, Yuwen Du, Yufeng Chen","doi":"10.1109/ICCEAI52939.2021.00062","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00062","url":null,"abstract":"Aiming at the dimensional disaster and over-fitting problems in data analysis, this paper proposes a feature selection method using hybrid integration of difference models and random forests (Integrate-RF), firstly, Integrate-RF use CART, CHAID, SVM, BN, NN, K-Means, Kohonen to evaluate the importance of features, and then, for the above seven sorts, Integrate-RF use the arithmetic average method to calculate the importance of the features; secondly, Integrate-RF select the most important features from the remaining features into features subset, and use random forest classification to get the corresponding out-of-bag(OOB) data classification error rate; finally, the optimal features subset can be selected based on the OOB data classification error rate. Experiments show that feature selection methods proposed in this paper effectively reduces the data dimension, selects features better and more adaptable.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116141887","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00100
Chaolin Peng, Lixiang Zhong
China is promoting the action of “using numbers to enrich wisdom” and cultivating new economy. As a typical small and medium-sized manufacturing enterprise, Dongguan Taiwei Electronics constructs intelligent lean system through intelligent factory construction, including intelligent lean central control room, station intelligent management center, automatic line intelligent management center, team intelligent management center, application equipment management system, combining lean production with intelligent manufacturing concept, using intelligent lean tools with deep knowledge precipitation, directly solve pain point problem. China's small and medium-sized manufacturing enterprises can promote intelligent lean through the path of “standardization, lean, digital, intelligent”, follow the basic strategy of “overall planning, step by step implementation, pull forward”, expand from the field of production to the field of research and development, marketing, and gradually achieve the goal of intelligence.
{"title":"Intelligent Transformation of Small and Medium-sized Manufacturing Enterprise in China - Case Study of Dongguan Taiwei Electronics","authors":"Chaolin Peng, Lixiang Zhong","doi":"10.1109/ICCEAI52939.2021.00100","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00100","url":null,"abstract":"China is promoting the action of “using numbers to enrich wisdom” and cultivating new economy. As a typical small and medium-sized manufacturing enterprise, Dongguan Taiwei Electronics constructs intelligent lean system through intelligent factory construction, including intelligent lean central control room, station intelligent management center, automatic line intelligent management center, team intelligent management center, application equipment management system, combining lean production with intelligent manufacturing concept, using intelligent lean tools with deep knowledge precipitation, directly solve pain point problem. China's small and medium-sized manufacturing enterprises can promote intelligent lean through the path of “standardization, lean, digital, intelligent”, follow the basic strategy of “overall planning, step by step implementation, pull forward”, expand from the field of production to the field of research and development, marketing, and gradually achieve the goal of intelligence.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"30 10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133059754","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}
Pub Date : 2021-08-01DOI: 10.1109/ICCEAI52939.2021.00041
Liting Zhang, Wensi Wang, Qiang Gao, Mei Yang, Yanping Ji, Shuqin Geng
Water quality is a basic work in environmental governance, which has vital significance in promoting the sustainable utilization of water resources and instant pollution prevention and precise control. Water quality data is dynamic and frequently fluctuating with different temporal and spatial dimensions, therefore it can be challenging to predict. A hybrid AM-ConvLSTM deep learning algorithm is proposed in this paper to rapidly predict the trend of water quality which can run faster and require low computing power rather than the traditional MIKE 21 hydrological method. The ConvLSTM method and the attention mechanism are assembled to build AM-ConvLSTM model to better capture spatial correlation. Moreover, the statistic methods are used to evaluate the effectiveness of the model and then compared with varieties of deep learning baseline methods. The results reveal that the hybrid AM-ConvLSTM model can effectively replace MIKE 21 model to predict the future trend of water quality, and then the local environmental protection agencies will respond quickly to emergency water pollution.
{"title":"Study on the Rapid Prediction Model of Water Quality for Emergency Water Pollution","authors":"Liting Zhang, Wensi Wang, Qiang Gao, Mei Yang, Yanping Ji, Shuqin Geng","doi":"10.1109/ICCEAI52939.2021.00041","DOIUrl":"https://doi.org/10.1109/ICCEAI52939.2021.00041","url":null,"abstract":"Water quality is a basic work in environmental governance, which has vital significance in promoting the sustainable utilization of water resources and instant pollution prevention and precise control. Water quality data is dynamic and frequently fluctuating with different temporal and spatial dimensions, therefore it can be challenging to predict. A hybrid AM-ConvLSTM deep learning algorithm is proposed in this paper to rapidly predict the trend of water quality which can run faster and require low computing power rather than the traditional MIKE 21 hydrological method. The ConvLSTM method and the attention mechanism are assembled to build AM-ConvLSTM model to better capture spatial correlation. Moreover, the statistic methods are used to evaluate the effectiveness of the model and then compared with varieties of deep learning baseline methods. The results reveal that the hybrid AM-ConvLSTM model can effectively replace MIKE 21 model to predict the future trend of water quality, and then the local environmental protection agencies will respond quickly to emergency water pollution.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":"16 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116349540","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：