Current Issue

15 December 2021, Volume 3 Issue 4

Previous Issue    Next Issue

Review Intelligence

MetaGrid: a parallel grids based approach for next generation smart power systems

Xiaoshuang LI, Xiao WANG, Linyao YANG, Yonglin TIAN, Yutong WANG, Jun ZHANG, Fei-Yue WANG

2021, 3(4):  387-398.  doi:10.11959/j.issn.2096-6652.202139

Asbtract ( 525 )   HTML ( 67)   PDF (2460KB) ( 392 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the large-scale integration of new energy devices, the fragility, openness and uncertainty of the bulk power grid system have increased significantly, and the system management and control are facing significant challenges.Constructing a parallel grid based on the idea of parallel systems and metaverse to form an intelligent bulk power grid management and control theoretical approach of virtual-real integration is one way to solve the above problems.The basic framework of the parallel power grid system MetaGrid was introduced, the key technologies of the parallel power grid system were analyzed, and the potential application scenarios of the parallel power grid were prospected.Based on the 300-nodes thermal stability case, a prototype of a parallel grid application case was provided.It is expected that by constructing a parallel grid system, the real grid system will be accurately modeled, and with the help of large-scales computational experiments and parallel execution, the virtual-real interaction between the real grid system and the artificial grid system will be realized, and the power system will be promoted from simulation-dependent control to parallel grid-based intelligent management and control.

Surveys and Prospectives

Review of pedestrian trajectory prediction methods

Linhui LI, Bin ZHOU, Weiwei REN, Jing LIAN

2021, 3(4):  399-411.  doi:10.11959/j.issn.2096-6652.202140

Asbtract ( 1601 )   HTML ( 349)   PDF (2826KB) ( 1328 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

With the breakthrough of deep learning technology and the proposal of large data sets, the accuracy of pedestrian trajectory prediction has become one of the research hotspots in the field of artificial intelligence.The technical classification and research status of pedestrian trajectory prediction were mainly reviewed.According to the different modeling methods, the existing methods were divided into shallow learning and deep learning based trajectory prediction algorithms, the advantages and disadvantages of representative algorithms in each type of method were analyzed and introduced.Then, the current mainstream public data sets were summarized, and the performance of mainstream trajectory prediction methods based on the data sets was compared.Finally, the challenges faced by the trajectory prediction technology and the development direction of future work were prospected.

Theoretical framework of brain modelling and highlighted problems

Dongwei HU, Xiaolu FENG

2021, 3(4):  412-434.  doi:10.11959/j.issn.2096-6652.202141

Asbtract ( 335 )   HTML ( 33)   PDF (2928KB) ( 459 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Understanding how the brain works is an important way to constructing the final artificial general intelligence.Building models, and testing the correctness of these models play a pivotal role on understanding how the brain works.Firstly, the experimental techniques and mathematical modelling methods for the study of brain were reviewed, and then the block diagram model and theoretical framework were shown, with an emphasis on reinforcement learning.Finally, some highlighted problems in this area, and the relation with closely related disciplines were addressed.

Special Column: Data Based Learning and Optimization

Parallel control-based event-driven approximate optimal control of discrete-time nonlinear systems

Zehua LIAO, Ziyu LIANG, Tianmin ZHOU, Jingwei LU, Qinglai WEI

2021, 3(4):  435-443.  doi:10.11959/j.issn.2096-6652.202142

Asbtract ( 245 )   HTML ( 37)   PDF (1278KB) ( 406 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

A parallel control-based event-driven approximate optimal control method was proposed for discrete-time nonlinear systems.Firstly, based on the time-triggered optimal value function and optimal control law, a novel triggering condition was developed and the asymptotic stability of the closed-loop system was proved based on the Lyapunov method.Secondly, a parallel control method using neural networks and adaptive dynamic programming techniques was proposed to predict the next state of the system and to obtain the optimal value function and control law.Finally, the effectiveness of the developed method was validated by numerical examples.

Synchronization control of unknown heterogeneous multi-agent system via model-free adaptive dynamic programming

Lina XIA, Qing LI, Ruizhuo SONG, Zihan WANG, Zhen XU

2021, 3(4):  444-448.  doi:10.11959/j.issn.2096-6652.202143

Asbtract ( 211 )   HTML ( 30)   PDF (1921KB) ( 285 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Synchronization of multi-agent system has been gradually applied in the most fields, but there are still many unsolved problems.The synchronization control of unknown heterogeneous multi-agent system based on model-free adaptive dynamic programming (MFADP) algorithm was studied.Firstly, an observer was designed for each follower to estimate the information of the leader, including the state and the system dynamic matrix of the leader.Then, the optimal controller was obtained by exploiting the Bellman optimality principle.Under the condition that the dynamics of the follower was unknown, a MFADP algorithm was proposed.Finally, two-mass-spring systems were used to verify the effectiveness of the algorithm.

Action dependent heuristic online tracking control for a class of nonaffine systems

Huiling ZHAO, Ding WANG, Jin REN

2021, 3(4):  449-455.  doi:10.11959/j.issn.2096-6652.202144

Asbtract ( 142 )   HTML ( 16)   PDF (1573KB) ( 183 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the tracking control problem for nonaffine systems, an online design method was developed by using the action dependent heuristic dynamic programming (ADHDP) structure.Firstly, the tracking control problem for the unknown nonaffine system was transformed into the error regulation problem.Then, the ADHDP tracking controller was designed and the online learning method was adopted to synchronize the system control with the training of action networks and critic networks, so that the desired trajectory could be tracked by the system state.Finally, a simulation example was given to verify the effectiveness of the proposed method.

Papers and Reports

Dynamic water quality warning with seasonal decomposition and long short-term memory network

Bowen XU, Jing BI, Haitao YUAN, Gongming WANG, Junfei QIAO

2021, 3(4):  456-465.  doi:10.11959/j.issn.2096-6652.202145

Asbtract ( 231 )   HTML ( 20)   PDF (6186KB) ( 158 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Surface water quality is increasingly deteriorated in recent years, and therefore, high-quality early warning and prediction of water quality are essential for sustainability of water resources and emergency response mechanisms.Long short-term memory (LSTM) network is widely applied in the existing literature on the prediction of water quality time series.However, only applying LSTM for the prediction of water quality time series cannot well address irregular fluctuations in the water quality series caused by multiple complex factors.To solve this problem, a data-driven prediction model for the water quality time series was proposed, named STL-LSTM-ED, which was composed of seasonal-trend decomposition using locally weighted scatterplot smoothing (STL) and LSTM based on encoder-decoder (LSTM-ED).Compared with several typical models of LSTM, LSTM-ED, and a sequence decomposition method based on LSTM, the proposed STL-LSTM-ED can significantly improve the prediction accuracy and reliability of the water quality time series, and also provide the effective data support for dynamic warning of water quality.

Tibetan entity relation extraction based on multi-level attention fusion mechanism

Like WANG, Yuan SUN, Sisi LIU

2021, 3(4):  466-473.  doi:10.11959/j.issn.2096-6652.202146

Asbtract ( 187 )   HTML ( 12)   PDF (2120KB) ( 131 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Compared with Chinese and English, the training corpus of Tibetan entity relation is smaller, so it is difficult to obtain higher accuracy based on traditional supervised learning methods.And there exists the problem of wrong labels in distant supervision for relation extraction.To solve these problems, the distant supervision method was used to construct the data set of Tibetan entity relation extraction through aligning the knowledge base with texts, which could alleviate the problem of lacking of large-scale corpus in Tibetan.And a Tibetan entity relation extraction model based on multi-level attention fusion mechanism was proposed.The self-attention was added to extract internal features of words in word level.The selective attention mechanism could assign weights of each instance, so as to make full use of informative sentences and reduce weights of noisy instances.Meanwhile, a joint score function was introduced to correct wrong labels, and neural network was combined with support vector machine to extract relations.Experimental results show that the proposed model can effectively improve the accuracy of Tibetan entity relation extraction, and is better than the baseline.

Classification method of dermoscopic image based on hierarchical convolution neural network

Hong SHAO, Mingkun ZHANG, Wencheng CUI

2021, 3(4):  474-481.  doi:10.11959/j.issn.2096-6652.202147

Asbtract ( 162 )   HTML ( 21)   PDF (1079KB) ( 439 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to solve the problem of insufficient number of dermoscopic image and the imbalance of image data among various diseases, a classification method of dermoscopic image based on class weighted cross entropy loss function and hierarchical convolution neural network was proposed.Firstly, the dermoscopic image was processed by color constancy to eliminate the ambient light noise.Then, the hierarchical convolution neural network based on ResNet 50 was constructed, and the two classification and multi classification convolution neural network models were constructed respectively, and the class weighted cross entropy loss function was set according to the quantitative characteristics of the dermatoscopic image.The experimental results show that the method achieves good classification effect, and the classification accuracy reaches 85.94%.Compared with the improved classification model ResNet 50, the test accuracy is improved by 5.752%.

Improved M-ORB based direct-loop closure detection algorithm for visual SLAM

Wei LI, Menghan REN, Weihao HUANG, Xiaoyu DU, Yi ZHOU

2021, 3(4):  482-491.  doi:10.11959/j.issn.2096-6652.202148

Asbtract ( 231 )   HTML ( 28)   PDF (5478KB) ( 162 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Most kinds of direct methods do not extract image feature points in the front end of SLAM system, resulting in that they cannot use loop closure detection with bag-of-words models to eliminate the cumulative error of the system.To resolve this problem, an improved mature-oriented fast and rotated BRIEF (M-ORB) based direct-loop closure detection algorithm for visual SLAM was proposed, which designed an improved M-ORB, generated the bag of words model required for loop closure detection, and then used the term frequency-inverse document frequency (TF-IDF) algorithm to adaptively assign weights to the visual words in each sub-node of the dictionary tree.Finally, an accurate representation of the scene information was obtained.In the end, the proposed algorithm and conducted comparative experiments were verified though two public data sets TUM and KITTI.The experimental results show that the algorithm proposed in this paper can effectively detect the loop closure, and has better real-time and robustness performance without reducing the accuracy.

Implicit knowledge learning:taking clinical simulation for example

Jun DONG

2021, 3(4):  492-498.  doi:10.11959/j.issn.2096-6652.202149

Asbtract ( 175 )   HTML ( 16)   PDF (853KB) ( 340 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In the course of computer-aided cardiovascular disease diagnosis, field experience modeling usually being ignored.Hereafter the strategy to focus on the discovey and study of implicit knowledge was presented.The importance of implicit knowledge in knowledge engineering and its transformation process to explicit knowledge was discussed.Technical roadmap simulating physicians’ thinking by infusing rules inference and machine learning was introduced.And the computer-aided traditional Chinese medicine new prescription generating was analysed, that is, the machine gave a new prescription for a certain disease by simulating the thinking process of traditional Chinese medicine.It is hoped that this approach can provide a reference for the modeling of applications in other fields of artificial intelligence.

Windblown sand control decision-making support system based on parallel intelligence: Taklimakan desert highway and its sand-breaking system

Fangle CHANG, Mengzhen KANG, Xiujuan WANG, Jiaqiang LEI, Fei-Yue WANG

2021, 3(4):  499-506.  doi:10.11959/j.issn.2096-6652.202150

Asbtract ( 242 )   HTML ( 23)   PDF (1101KB) ( 198 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Using Taklimakan desert highway and its sand-breaking system as the research object for complex system management and control, the ACP-based parallel intelligence theory to deal with the problems of the difficulty in modeling, analyzing, and predicting of sand-breaking system was applied, to realize the intelligent decision support for sand-breaking system management and control, support the sustainable development of aeolian environment.The expert experience knowledge to construct the artificial desert highway sand-breaking system was extracted by simulating physical process.The sand-control efficiency index of the artificial system was calculated using equations, then evaluated and modified by comparing and learning with the actual system.Using parallel intelligent theory, the artificial system and the actual system can learn from each other to provide decision support for sand control.

Game 5.0: social cognitionparallel game based on the parallel systems and machine game

Yaling LI, Linyao YANG, Jun GE, Yuanqi QIN, Xiao WANG

2021, 3(4):  507-520.  doi:10.11959/j.issn.2096-6652.202151

Asbtract ( 357 )   HTML ( 24)   PDF (6969KB) ( 287 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Machine game is a kind of artificial intelligence in which machines imitate the way people think and make decisions.It is currently the most challenging research direction in the field of artificial intelligence, whose research status largely represents the development level of artificial intelligence.The basic concepts, research status and typical applications of machine game were summarized, its challenges and development trends were discussed, and the possible applications of machine game on the social governance were analyzed.Then, a social cognitive parallel game method based on the parallel systems and the machine game were proposed, aiming at providing new ideas for the research of social governance.

Parallel battery: the framework and process for an intelligent and ecological battery system and related services

Fei-Yue WANG, Huaiguang JIANG

2021, 3(4):  521-531.  doi:10.11959/j.issn.2096-6652.202152

Asbtract ( 262 )   HTML ( 17)   PDF (3822KB) ( 441 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The concept, framework, process methodology and applications of parallel battery were proposed from both virtual and real aspects.The parallel battery was an application of ACP-based parallel intelligence in battery and related energy system areas.The real battery system was running with its equivalent, and the artificial battery system was in a virtual space, in a parallel and interactive manner.The artificial battery system contained the descriptive, predictive, and prescriptive functions on the real battery and related energy systems.There was a closed-loop workflow between the real battery system and the artificial battery system, which iteratively optimizes the battery and related energy systems, leading to a new paradigm of intelligent and ecological parallel battery system management.