Skip to main content
main-content

Über dieses Buch

This book constitutes the second part of the refereed proceedings of the International Conference on Life System Modeling and Simulation, LSMS 2014, and of the International Conference on Intelligent Computing for Sustainable Energy and Environment, ICSEE 2014, held in Shanghai, China, in September 2014. The 159 revised full papers presented in the three volumes of CCIS 461-463 were carefully reviewed and selected from 572 submissions. The papers of this volume are organized in topical sections on advanced neural network theory and algorithms; advanced evolutionary computing theory and algorithms, such as particle swarm optimization, differential evolution, ant colonies, artificial life, artificial immune systems and genetic algorithm; fuzzy, neural, and fuzzy-neuro hybrids; intelligent modeling, monitoring, and control of complex nonlinear systems; intelligent modeling and simulation of climate change; communication and control for distributed networked systems.

Inhaltsverzeichnis

Frontmatter

The First Section: Advanced Fuzzy, Neural Network Theory and Algorithms

Research of Metro Illumination Control Based on BP Neural Network PID Algorithm

This paper presents the metro constant illumination controller, which is designed to solve the current problem that the interior illumination in a moving metro fluctuates drastically with the changes of exterior environment. By detecting the real-time interior illumination values and adjusting the lights in a metro with PID controller based on BP neural network, the controller can keep the metro interior illumination values around the preset value. Simulations and actual test results show that the PID controller based on BP neural network has strong adaptability and robustness in a nonlinear system. It can both save energy and solve the problem of drastically fluctuating illumination in a moving metro, which cannot be achieved in conventional PID controllers.

Yong Shao, Fengbo Wang, Yuting Zhang, Peng Zan

A Study of Adaptive Neural Network Control System

In a hydraulic support system of heavy equipment, the oil pressure is required to be a constant value. Due to the disturbances come from the external environment and the running process, the hydraulic support system would not be stable; therefore, we here present a closed-loop feedback system, which has an adaptive neuronal network control system to make the oil pressure stable by controlling the rotate speed of the hydraulic pump motor. In this hydraulic system, the response and stability are the key factors to judge the control method is good or not. In our simulation, it shows that this adaptive neural network control system can meet the design requirements. It has good response and stability characteristics.

Heng Zhong, Dingyuan Li, Kun Tu

Design of Fuzzy Logic Controller Based on Differential Evolution Algorithm

In order to overcome the deficiency of fuzzy control algorithm, an adaptive fuzzy logic controller is proposed. In this method, the differential evolution algorithm (DE) was employed to optimize parameters of fuzzy controller: quantitative factor and proportional factor , they were designed as individuals of DE population, and evaluated using the fitness function provided until the termination condition was fulfilled. Then the selected parameter values were sent back to fuzzy logic controller. Simulation results concerning two-tank system show that the DE optimized fuzzy controller has good adaptability, as well as it‘s effectiveness, which provides a new approach to improve fuzzy control system.

Li Shuai, Sun Wei

Batch-Wise Updating Neuro-Fuzzy Model Based Predictive Control for Batch Processes

In order to guarantee the control performance of the batch processes when uncertainties and disturbances exist, a neuro-fuzzy model (NFM) predictive controller based on batch-wise model modification is developed. Model modification along batch-axis is used to improve the accuracy of neuro-fuzzy model, and predictive control along time-axis can guarantee the optimal control consequence, which lead to superior tracking performance and better robustness compared with the conventional quadratic criterion based iterative learning control (Q-ILC) approach. An illustrative example is presented to verify the effectiveness of the investigated approach.

Qinsheng Li, Li Jia, Tian Yang

The Second Section: Advanced Evolutionary Computation

A Novel Learning Algorithm for Pallet Grouping Technology

The Pallet Grouping Problem (PGP) is defined as minimizing the number of pallets for placing all materials in a collection and distribution center. A key to solving the PGP is to tackle the Pallet Loading Problem (PLP). The Pallet Loading Problem aims to maximize the number of identical rectangular boxes placed within a rectangular pallet. All boxes have identical rectangular dimensions and, when placed, must be located completely within the pallet. In this paper, a novel pallet grouping technology is proposed and a new learning algorithm, namely Learning Only from Excellence, (LOE) is presented for solving the pallet loading problem. Simulation results show that compared with the conventional Genetic Algorithm (AG) for two pallet loading problems with different scales, the new learning algorithm is proved to be more efficiently.

Weitian Lin, Zhigang Lian, Bin Jiao, Xingsheng Gu, Wei Xu

Optimization of Data Query and Display Algorithm for a Wireless Monitoring and Visualization System

As wireless sensor networks are widely used in industry, the wireless monitoring system for sniffing data packets from wireless sensor networks has been realized. Many novel methods has been utilized to efficiently store data and provide reliable query and display functions in real-time listening, such as adjustment of storage structure and introduction of memory module. This paper designs and realizes a wireless monitoring and visualization system, which can accurately collect data from the communication among a variety of wireless terminal devices in real-time through changing original data storage structure, improving data query algorithms and optimizing data display algorithms.

Jiqiu Chen, Jingqi Fu, Weihua Bao

A Simple Human Learning Optimization Algorithm

This paper presents a novel Simple Human Learning Optimization (SHLO) algorithm, which is inspired by human learning mechanisms. Three learning operators are developed to generate new solutions and search for the optima by mimicking the learning behaviors of human. The 0-1 knapsack problems are adopted as benchmark problems to validate the performance of SHLO, and the results are compared with those of binary particle swarm optimization (BPSO), modified binary differential evolution (MBDE), binary fruit fly optimization algorithm (bFOA) and adaptive binary harmony search algorithm (ABHS). The experimental results demonstrate that SHLO significantly outperforms BPSO, MBDE, bFOA and ABHS. Considering the ease of implementation and the excellence of global search ability, SHLO is a promising optimization tool.

Ling Wang, Haoqi Ni, Ruixin Yang, Minrui Fei, Wei Ye

An Improved Non-dominated Sorting Genetic Algorithm for Multi-objective Optimization Based on Crowding Distance

An improved non-dominated sorting genetic algorithm (INSGA) is introduced for multi-objective optimization. In order to keep the diversity of the population, a modified elite preservation strategy is adopted and the evaluation of solutions’ crowding degree is integrated in crossover operations during the evolution. The INSGA is compared with the NSGA-II and other algorithms by applications to five classical test functions and an environmental/economic dispatch (EED) problem in power systems. It is shown that the Pareto solution obtained by INSGA has a good convergence and diversity.

Tian-liang Xia, Shao-hua Zhang

Comprehensive Analysis of Cooperative Particle Swarm Optimization with Adaptive Mixed Swarm

Inspired by the collective intelligence of natural mixed flocking, the paper develops a mixed swarm cooperative search model for particle swarm optimization(MCPSO). Firstly, makes some analysis about the hinting principles and search mechanism behind the natural mixed flocking, and proposes the construction of mixed swarm for optimization. Secondly, introduces the mixed swarm into PSO and researches the main search behaviors of MCPSO, including coarse search and fine search, cooperative search and learning. Finally, the proposed MCPSO was applied to some well-known benchmarks. The experimental results and relative analysis show mixed swarm cooperative search mechanism can greatly benefit the global optimization performance of PSO.

Jing Jie, Beiping Hou, Hui Zheng, Xiaoli Wu

Imperialist Competitive Algorithm with Trading Mechanism for Optimization

International trade is the exchange of capital, goods, and services across different countries. Trading has been explored by economists to be an important mechanism for maintaining development. In an imperialistic country, trading makes imperialists capture resources from colonies, meanwhile providing technologies or cultures for colonies to develop themselves. Inspired by this economic phenomenon, this paper transplants the trading mechanism to imperialist competitive algorithm (ICA) and proposes an improved ICA with import and export mechanisms (IICA). IICA is designed to alleviate the problem of slow convergence without significantly impairing the parallel competitive feature of ICA. It is characterized by allowing the imperialist to capture useful aspects of colonies to enhance itself, and meanwhile making colonies learn advanced components from their imperialist. In this way, the trading mechanism enables imperialists and colonies to strengthen interactions during them. The performance of IICA is validated on 23 benchmark functions. Its high performance is confirmed by comparing with other ICA variants.

Shuaiqun Wang, Aorigele, Jingyi Luo, Shangce Gao

An Improved Artificial Fish Swarm Algorithm and Application

An improved Artificial Fish Swarm Algorithm (AFSA) based on Hooke-Jeeves (HJ) algorithm is proposed and improved AFSA is applied to design lamps of changeable color temperature and high luminous efficacy in this paper. The disadvantage of AFSA stochastic moving without a definite purpose is improved by HJ algorithm, owing to HJ’s great ability of local searching. Accuracy of solution is improved by the adaptive weight. The improved AFSA is verified through an example of how to search for the most luminous efficacy of LED mixing color. The white, red, green and blue LEDs are chosen to design LED lamp samples. LED proportions of 5000K color temperature among those LEDs are optimized by AFSA and new AFSA in the Matlab. The obtained results indicate that improved AFSA is faster and higher accuracy. After LED lamps are tested by integrating sphere, the results show that the difference between the actual value and simulation calculation value is tiny, the new AFSA is effective. The improved AFSA provides a new efficient calculation method of LED proportions. Compared with the traditional manual calculation LED proportions, new method not only saves a significant amount of time, but also achieves higher luminous efficacy for lamps. All this shows that the new method is effective and has high practical value.

Xinyuan Luan, Biyao Jin, Tingzhang Liu, Yingqi Zhang

A Novel Deterministic Quantum Swarm Evolutionary Algorithm

This paper presents a novel deterministic quantum swarm evolutionary (DQSE) algorithm based on the discovery of the drawback of the standard quantum swarm evolutionary (QSE) algorithm, in which a deterministic search strategy, inspired by the nature of qubit-based evolutionary algorithms and the characteristics of qubits, is proposed to avoid the misleading of search and strengthen the global search ability. The experimental results show that the developed DQSE outperforms the quantum-inspired evolutionary algorithm, the quantum-inspired evolutionary algorithm with NOT gate and QSE in terms of the search accuracy and the convergence speed, which demonstrates that DQSE is an effective and efficient optimization algorithm.

Xikun Wang, Lin Qian, Ling Wang, Muhammad Ilyas Menhas, Haoqi Ni, Xin Du

Application of the Improved Quantum Genetic Algorithm

The paper put forward an improved QGA in order to solve the shortcomings of traditional QGA in optimizing the multimodal function, like lower convergence speed, easier to local optimum. The improved QGA will adjust the rotation angle of quantum gate according to the dynamic evolutionary process, so it can speed up the convergence of the multimodal function. In order to avoiding the individual evolved to local optimum, so mutation was introduced. Lastly, through the typical complex multimodal function test to prove the validity of the improved QGA.

Yufa Xu, Xiaojuan Mei, Zhijun Dai, Qiangqiang Su

Strategy Analysis of an Evolutionary Spectrum Sensing Game

Evolutionary game has been shown to greatly improve the spectrum sensing performance in cognitive radio. However, as selfish users are shortsighted for the long-term profits, they are not willing to collaborate to sense. In this paper, we propose an evolutionary spectrum sensing game to improve the long-term spectrum utilization. The new spectrum sensing model takes advantage of the long-term effect of the future actions on the current actions by using the concept of present value (PV) in repeated game. The collaboration conditions of two strategies, i.e., tit-for-tat and grim strategy are discussed. It is proved that the grim strategy can enhance secondary users’ sensing positivity greatly, and so is the overall spectrum efficiency. Finally these new developments are illustrated in our experiments.

Dongsheng Ding, Guoyue Zhang, Donglian Qi, Huhu Zhang

Particle Swarm Optimization Based on Shannon’s Entropy for Odor Source Localization

This paper proposes the particle swarm optimization based on Shannon’s entropy to deal with the problem of odor source localization. First, a measurement model by which the robots can always observe a position is briefly described. When the detection events occur, the position of the odor source lies in the vicinity of the observed position with a higher probability. When the non-detection events occur, the position of the odor source does not lie in the vicinity of the observed position with a higher probability. Second, on the basis of the measurement model, the posteriori probability distribution on the position of the odor source is established where the detection events and non-detection events are taken into account. Third, each robot can understand the search environment by using Shannon’s entropy which can be calculated in terms of the posteriori probability distribution on the position of the odor source. Moreover, each robot should move toward the direction of the entropy reduction. By means of this principle, the particle swarm optimization algorithm is introduced to plan the movement of the robot group. Finally, the effectiveness of the proposed approach is investigated for the problem of odor source localization.

Nanqi Li, Qiang Lu, Yang He, Jian Wang

The Third Section: Pattern Recognition and Machine Intelligence

Face Detection and Tracking Based on Adaboost CamShift and Kalman Filter Algorithm

Face detection is an important component of the intelligent video surveillance system. Based on the MeanShift algorithm, we have developed into the CamShift algorithm. Although the traditional Camshift algorithm can track the moving object well, it has to set the tracking object by manually. Meanwhile it fails to track the object easily while the object is occluded and interfered by the same color obstructions. In order to solve the problem, according to the CamShift algorithm features, in this article, I will combine Adaboost, CamShift and Kalman filtering algorithm, which can be relied on to realize face detection and tracking automatically and accurately.

Kun Chen, ChunLei Liu, Yongjin Xu

A Quick Method for Matching Object Subspaces Based on Visual Inspection

In visual inspection, the object image subspace should be segmented and matched, then the affine relationship is built between the template image and the sample image. But sometime the illumination is uneven on the surface of object image, it is difficult to obtain accurate position of the object subspace quickly. In this paper, a novel strategy is proposed to adopt discrete radial search paths instead of searching all points in the image. Therefore, the searching time can be reduced. In order to reduce the influence coming from the industrial environment, the paper proposes another method that is local energy level set segmentation, which can locate the object subspace quickly and accurately. The detection upon crown caps is as an example in the paper, then the detection effects and computing time are compared between several detection methods, and the mechanism of inspecting has been analyzed. The industrial applications are also given in the paper.

Wenju Zhou, Zixiang Fei, Huosheng Hu, Li Liu, Jingna Li

Research on Visual Environment Evaluation System of Subway Station Space

Based on the energy crisis, LED with its energy-saving and environmental friendly is gradually used to the subway station space lighting. But now, there are little materials about the visual environment evaluation for semiconductor lighting, so that the use of LED lighting lacks theoretical basis and data support. So, in order to promote the LED lighting in subway station space, it’s very important to evaluate the visual environment. Therefore, the core of this paper was to build a theoretical model to evaluate the visual environment of subway station space using Particle Swarm Optimization. Firstly, chose 16 evaluation indexes which were fit for the subway station visual environment evaluation and got the initial judgment matrix through pair wise comparison, after that, established the non-linear consistency correction model. Finally, used Particle Swarm Optimization to calculate the judgment matrix with better consistency and the corresponding index weight, and constructed the theoretical model.

Fengqun Guo, Hui Xiao

Study on Pattern Recognition of Hand Motion Modes Based on Wavelet Packet and SVM

For pattern recognition-based myoelectric prosthetic hand control, high accuracy of multiple discriminated hand motions is presented in related literature. But in practical applications of myoelectric control, considering cost and simple installation, fewer sensors are expected to be used. A method of pattern recognition based on the wavelet packet decomposition and support vector machine (SVM) is proposed in this paper. Firstly, energy spectrum as feature vectors of the surface electromyography (sEMG) signal is acquired by wavelet packet transform. Then, SVM is used for pattern recognition of hand motion modes. Four channels of sEMG signals obtained from sensors placed on different positions of forearm are used to experiment of hand motion recognition. And different combinations of 2 or 3 signals are tried to recognize hand motion modes. The results show that recognition rate of proposed method can get 92.5% while using 4 sEMG signals to recognize 8 different hand motions, which is 2.5% higher than using traditional method. And when using 3 sEMG signals from specific positions, it can reaches as high as 90%. When using 2sEMG signals only 6 motions can be discriminated with more than 90% recognition rate. Thus, the proposed method can meet the demands of sEMG prosthetic hand control and has high practical value.

Fuxin Liang, Chuanjiang Li, Yunling Gao, Chongming Zhang, Jiajia Chen

Image Segmentation Using Multiphase Curve Evolution Based on Level Set

A novel multiphase curve evolution based on level set (MCELS) is presented, which is used for image segmentation. The MCELS method introduces

N

level set functions partition 2

N

sub-regions, which reduces the computational complexity. The double curve function is developed on the modified penalty function during the evolution. The experimental objects employ tablet packaging images. From the simulation results, the MCELS method can be used to partition multiple gray regions images for the noise, uneven gray scale, and intensity inhomogeneities. Comparing with recent researches based on level set methods, the characteristics of MCELS for image segmentation are superior robustness for noise, less run time and preferable computational efficiency.

Li Liu, Xiaowei Tu, Wenju Zhou, Minrui Fei, Aolei Yang, Jun Yue

Experimental Platform Design and Implementation for Plate Structure Shape Perception and Reconstruction Algorithm

For the problem of experimental verification for plate structure shape perception and reconstruction algorithm, an experimental verification platform was designed and constructed consisting of experiment base station, excitation system, measurement system and relative software, to the fitting algorithm based on plane curve as reference algorithm and the static error analysis and dynamic error analysis for the effect of reconstruction was conducted precisely. The results showed that the experimental platform was with good real-time capacity and high accuracy to meet the needs for the verification and data analysis of algorithms.

Mingdong Li, Hesheng Zhang, Kaining Liu, Xiaojin Zhu

Improved Mean-Value Coordinates Algorithm for Image Fusion

Image fusion is an advanced image processing, in which mean-value coordinates (MVC) algorithm based on Poisson image is a fast and effective algorithm. However, the algorithm may have unsatisfactory results if the source image and target image have many variations of color on the image boundary and image details. To solve the problem, this paper proposes two optimization methods, preserving color based on geodesic distance and matching details with modified detail layer. To verify the feasibility of the methods, the improved MVC results are compared with the original MVC results by experiments. The comparison results show that the improved approach can achieve better performance in image fusion.

Cai Fu, Yuying Shao, Li Deng, Gen Lu

Study of Face Recognition Technology Based on STASM and Its Application in Video Retrieval

The gradual perfection of video retrieval technology has a positive effect in maintaining public order. However, with the improving complexity of monitoring environment, the increase of related video data requires further improvement to the efficiency of video retrieval technology. Video retrieval technology aiming at processing massive video data is needed urgently and it has become hot research subject in multimedia retrieval area. In this paper, the application of face recognition technology in video retrieval is discussed. To improve the retrieval efficiency, STASM algorithm based on OpenCV software platform is designed. The research involves the acquisition of video image frame data, face recognition and detection. Experimental results demonstrate the effectiveness and efficiency of the algorithms.

Chunlei Liu, Kun Chen, Yongjin Xu

Robust Blurred Palmprint Recognition via the Fast Vese-Osher Model

In this paper, we propose a new palmprint recognition system by using the fast Vese-Osher decomposition model to process the blurred palmprint images. First, a Gaussian defocus degradation model (GDDM) is proposed to extract the structure layer and texture layer of blurred palmprint images by using the fast Vese-Osher decomposition model, and the structure layer is proved to be more stable and robust than texture layer for palmprint recognition. Second, a novel algorithm based on weighted robustness with histogram of oriented gradient (WRHOG) is proposed to extract robust features from the structure layer of blurred palmprint images, which can address the problem of translation and rotation to a large extent. Finally, the normalized correlation coefficient (NCC) is used to measure the similarity of palmprint features for the new recognition system. Extensive experiments on the PolyU palmprint database and the blurred PolyU palmprint database validate the effectiveness of the proposed recognition system.

Danfeng Hong, Wanquan Liu, Jian Su, Zhenkuan Pan, Xin Wu

Affinity Propagation Clustering with Incomplete Data

Incomplete data are often encountered in data sets for clustering problems, and inappropriate treatment of incomplete data will significantly degrade the clustering performances. The Affinity Propagation (AP) algorithm is an effective algorithm for clustering analysis, but it is not directly applicable to the case of incomplete data. In view of the prevalence of missing data and the uncertainty of missing attributes, we put forward improved AP clustering for solving incomplete data problems. Three strategies(WDS, PDS and IPDS) are given, which involve modified versions of the AP algorithm. Clustering performances at different missing rates are discussed, and all approaches are tested on several UCI data sets with randomly missing data.

Cheng Lu, Shiji Song, Cheng Wu

An Improved Algorithm for Camera Calibration Technology Research

Computer vision technology has wide application value in daily life and industrial production. Camera calibration is the base of computer vision technology, which is the key and necessary step to get three-dimensional spatial information from a two-dimensional image. In the paper, the geometric parameter of camera is considered as the research object. Firstly, the relationship model of camera calibration is established and used to unify the world coordinate system, the camera coordinate system and the image coordinate system. It takes the image pixel point as the optimization goal, then a differential evolution combined with particle swarm optimization algorithm is proposed to calibrate camera. Experimental simulation results show that the improved algorithm has good optimization ability and used for camera calibration has validity and reliability.

Gen Lu, Yuying Shao, Li Deng, Minrui Fei

Gearbox Fault Diagnosis Method Based on SVM Trained by Improved SFLA

A method of fault diagnosis based on support vector machine trained by the improved shuffled frog leaping algorithm (ISFLA-SVM) is proposed to promote the classification accuracy of the wind turbine gearbox fault diagnosis. Because the parameter selection for penalty factor and kernel function in support vector machine (SVM) have a great impact on the classification accuracy, we may use the improved shuffled frog leaping algorithm to select excellent SVM parameters, use the optimized parameters to train machine. Then three groups of data in UCI are used for performance evaluation. Finally ISFLA-SVM model will be applied to the wind turbine gearbox fault diagnosis. The result of the diagnosis indicates that the common fault of wind turbine gearbox can be exactly identified by this method.

Lu Ma, Guochu Chen, Haiqun Wang

Discrete Chaotic Synchronization and Its Application in Image Encryption

Based on the synchronization principle of three-dimensional discrete Henon chaos, this paper presents a new image encryption algorithm. A set of keys are generated by chaotic iteration from the sending part. Then the image can be encrypted. Via the same keys generated by chaotic synchronization, the receiving part can get decrypted image by inverse transformation. From the view of cryptology, high dimensional chaos in this paper is more complex than the general chaos, it is difficult to predict; and this algorithm can not only substitute the pixel values, but also scramble the pixel locations at the same time, but in most cases some algorithms can only encrypt the image in a way. As a supplement to the encryption process, this algorithm introduces the technology of chaotic synchronization and bits scrambling, increases the difficulty of cracking, and enhances the algorithm security. Simulation results illustrate the effectiveness of the proposed method.

Hua Wang, Jiu-Peng Wu, Xiao-Shu Sheng, Peng Zan

Structural Shape Reconstruction through Modal Approach Using Strain Gages

It is significant of the strain detection and shape reconstruction of flexible structures for guaranteeing the safe and reliable operation of large-scale and precision equipment, such as spacecraft, space station, satellite, et al. This paper presents a structural shape reconstruction method based on the modal approach using the real-time sensing strain data. Firstly, the displacement-strain transformation relationship is derived using a modal approach and the finite element method is used as a numerical analysis method. From the implementation point of view, the united simulation is employed by the software of MATLAB and ANSYS. Moreover, a plate flexible structure is regarded as the research object and the simulation of shape reconstruction is investigated. The simulation results show that the structural shape reconstruction method based on the modal approach is a novel strategy. Furthermore, the experiments are performed with strain gages and the reconstructed displacements are compared to the measured displacement data from laser displacement sensor. The error analysis is demonstrated and it is indicated that the modal approach based structural shape reconstruction method is an effective approach for structural displacement monitoring of flexible structures.

Li Li, Wuqian Li, Pingan Ding, Xiaojin Zhu, Wei Sun

A Fast Colorization Algorithm for Infrared Video

Color night-vision technology increases the representation ability of monochrome night-vision imagery by adding color to it, making observers’ understanding easier. Usually the color night-vision methods require the infrared and the low-light-level images at the same time, which hinders their application in the environment where totally without light or covered by heavy rain and thick fogs. To expand the application area of color night-vision technology, we propose a quickly colorization method based only on single band infrared video, which can provide all weather condition working. This method only requires a few pixels to be manually set with chrome values, and then the entire frame as well as the following frame sequence is automatically colorized. Experiments show that the colorization results are satisfactory and the algorithm is running fast.

Mengchi He, Xiaojing Gu, Xingsheng Gu

The Forth Section: Intelligent Modeling, Monitoring, and Control of Complex Nonlinear Systems

A New Networked Identification Approach for a Class of Hammerstein Systems

This paper investigates the problem of identification of a class of Hammerstein systems over a wireless network. An iterative identification method is implemented over a physical IEEE 802.11b wireless channel. Every time the identified model is used into next identification process to produce the estimated values of the plant outputs for compensating the influence of network delays. Finally the identified model can be optimized through the multiple iterations. The effectiveness of the proposed approach is demonstrated by numerical examples.

Weihua Deng, Kang Li, Jing Deng, Enyu Jiang, Qiang Zhu

Network-Based Stabilization of Linear Systems via Static Output Feedback

This paper is concerned with the problem of static output feedback control for networked systems with a logic zero-order-hold (ZOH). First, the networked closed-loop system is modeled as a discrete-time linear system with a time-varying delay, whose upper bound can be regarded as not only the admissible maximum delays induced by the network but also the admissible maximum number of packet dropouts between any two consecutive updating instants of ZOH. Then, in order to obtain a larger upper bound of the time-varying delay, a generalized finite-sum inequality is introduced, based on which, a less conservative stability condition is derived by incorporating with convex combination technique. By using the cone complementary linearization approach, the desired output feedback controllers can be designed by solving a nonlinear minimization problem subject to a set of linear matrix inequalities. Finally, some examples are given to show the effectiveness of the proposed method.

Xian-Ming Zhang, Qing-Long Han

Passivity-Based Control for Fractional Order Unified Chaotic System

This paper concerns with the fractional order unified chaotic control based on passivity. A hybrid control strategy combined with fractional state feedback and passive control is proposed, derived from the properties of fractional calculus and the concept of passivity. The fractional chaotic system with the hybrid controller proposed can be stabilized at its equilibrium under different initial conditions. Numerical simulation results present the verification on the effectiveness of the proposed control method.

Qiao Wang, Donglian Qi

Characteristics Analysis of Cloud Services Based on Complex Network

This paper researched the theory of cloud services based on complex network theory and graph theory. By virtue of the generalized network model of cooperation, a method of building cloud services network model was put forward, with regarding cloud services and their cooperative relationship as vertexs and edges in abstraction respectively. The network statistical properties are analyzed, including degree, degree distribution, shortest path length, vertex betweenness, similar matching coefficient and clustering coefficient. Using an example about fulfilling an order of LED spotlight, characteristics analysis of cloud services was suggested, to illustrate how to choose the appropriate cloud services.

Lilan Liu, Cheng Chen, Tao Yu

Room Cooling Load Calculation Based on Soft Sensing

The calculation of real-time dynamic room cooling load can be solved effectively by soft sensing technology based on auxiliary variable. This paper studies on the relationships between room cooling loads at different reference temperatures, and presents the system equations for cooling load calculation based on soft sensing. The undetermined coefficients in the system equations were identified via the least squares method, which reflect the magnitude relationship between the non-measurable primary variables and the auxiliary variables which can be measured accurately. Finally, commentary was presented based on the comparison between the results of simulation in Dwelling Environment Simulation Tools (DeST) and results of calculation via the equations in this paper.

Zhanpei Li, Xinyuan Luan, Tingzhang Liu, Biyao Jin, Yingqi Zhang

Trajectory Tracking of Nonholonomic Mobile Robots via Discrete-Time Sliding Mode Controller Based on Uncalibrated Visual Servoing

This paper considers the problem of trajectory tracking of nonholonomic mobile robots based on uncalibrated visual servoing. A prerecorded image sequence or a video taken by the pin-hole camera is used to define a desired trajectory for the mobile robot. First, a novel discrete-time model is present based on visual servoing. And then the discrete-sliding mode controller is designed for the model associated with uncertain parameter. The asymptotic convergence of the tracking errors is proved rigorously. Finally, simulation results confirm the effectiveness of the proposed methods.

Gang Wang, Chaoli Wang, Xiaoming Song, Qinghui Du

A Systematic Fire Detection Approach Based on Sparse Least-Squares SVMs

In this paper, a systematic approach adopting sparse least-squares SVMs (LS-SVMs) is proposed to automatically detect fire using vision-based systems with fast speed and good performance. Within this framework, the features are first extracted from input images using wavelet analysis. The LS-SVM is then trained on the obtained dataset with global support vectors (GSVs) selected by a fast subset selection method, in the end of which the classifier parameters can be directly calculated rather than updated during the training process, leading to a significant saving of computing time. This sparse classifier only depends on the GSVs rather than all the patterns, which helps to reduce the complexity of the classifier and improve the generalization performance. Detection results on real fire images show the effectiveness and efficiency of the proposed approach.

Jingjing Zhang, Kang Li, Wanqing Zhao, Minrui Fei, Yigang Wang

Sensorless Vector Control of PMSM in Wide Speed Range

In order to achieve the sensorless vector control of PMSM in wide speed range, a hybrid control mode strategy is presented, which included sliding mode observer (SMO) and high frequency injection (HFI). At medium or high speed, sliding mode observer method that was based on fundamental wave model is applied to estimate of speed and position of PMSM. While at low speed, for avoiding the shortcomings of SMO, it had to switch to the HFI method. Firstly, the application of SMO method was achieved and the speed limit of SMO method is calculated, and it is as basis for switching region of the speed of the hybrid mode. The simulation results show that the hybrid mode can reduce the buffet in the procedure of switching of algorithm effectively. And it achieves the control of PMSM in wide speed range.

Tao Yan, Jun Liu, Haiyan Zhang

Pitch Angle Control for Improving the Low Voltage Ride-Through Based on DFIG

Pitch angle control for wind generation is one of the most important segments in wind turbine. Wind turbine can capture the max wind energy which is called Maximum Power Point Tracking through regulating the pitch angle and then the output power of wind turbine is the best. On the base of it, one new method of pitch angle is put forward in this paper. When the grid was failure, we can reduce the output power through regulating the pitch angle in order to improve the performance of LVRT of the wind turbine. When the grid voltage is under the sudden dip, the rotor speed can be limited to reduce the engine and the sudden disconnection of wind turbine and grid can be confined. The pitch angle control is described in this paper in a short time when the emergency power supply is discussed so that we can keep the coordination between the wind turbine and the grid.

Ruming Li, Tianyu Liu, Qinghua Zhu, Li Zhang

A New Method for the Control of the Conditioning Temperature of Hoop Standard Granulator

The conditioning temperature is one of the important parameters of the hoop standard granulator production system. It can be influenced by temperature system with nonlinear, time-varying and hysteresis characteristics and feed quantity. This paper creates a control algorithm which combines disturbance observer and fuzzy PID to control the temperature modulation. In this way, feed quantity of the hoop standard granulator is also seen as a part of the interference. By constructing disturbance observer, this paper predicts the disturbance on temperature system and variations of parameters, so as to suppress the effect of interference on the system. Meanwhile this paper introduces fuzzy PID for adaptive PID tuning parameters to achieve optimal parameters. Finally, the numerical simulation on temperature system has strong adaptability and robustness.

Yuan Xue, Jianguo Wu, Lei Qin, Jin Liu, Kun Zhang

Integrated IMC-ILC Control System Design for Batch Processes

Considering conventional iterative learning control (ILC) is actually an open-loop control approach within each batch, which cannot guarantee the control performance of batch process when uncertainties and disturbances exist, an integrated iterative learning control scheme is presented in this paper. The proposed approach systematically integrates continuous-time information along with time-axis and discrete-time information along with batch-axis into one uniform frame, namely an internal model control (IMC)based PID control along time-axis, while the optimal ILC along batch-axis. As a result, the operation policy of batch process leads to superior tracking performance and better robustness compared with conventional ILC strategy. An illustrative example is exploited to verify the effectiveness of the investigated approach.

Qinsheng Li, Li Jia, Tian Yang

Decentralized Control for Power Systems Components Based-on Nonlinear Differential-Algebraic Equations Subsystem Model

Components of power systems essentially belong to a special class of nonlinear differential-algebraic equations subsystem, whose index is one and interconnection is locally measurable. In this paper, the decentralized control problem is discussed using inverse systems method for such class of power systems components. Firstly, the definition of

α

-order integral right inverse system is presented. Secondly, a recursive algorithm is proposed to judge whether the controlled component is invertible. Then a physically feasible

α

-order integral right inverse system is constructed with which the controlled component is made linearization and decoupled, so that linear control theory and methods can be applied. Finally, a decentralized excitation controller is designed for one synchronous generator within multi-machine power systems based on the proposed method in this paper.The simulation results demonstrate the effectiveness of the proposed control scheme.

Qiang Zang, Ying Zhou, Ping Mei, Baichao Zheng, Kaifeng Zhang

Event-Triggered State Estimation for Complex Systems with Randomly Nonlinearities and Time-Varying Delay

The event-triggered state estimation is investigated for a class of complex networks system with randomly nonlinearities. A novel event-triggered scheme is proposed, which can reduce the information communication burden in the network. Considering the effect of transmission delay, a time-varying delayed system model is constructed. Attention of this paper is focused on the analysis and design of a reliable estimator for the complex networks through the available output measurements under event-triggered scheme. In order to design the state estimator, a Lyapunov functional approach and the linear matrix inequality technique are employed. A sufficient condition is obtained in which the estimator error dynamics is exponential asymptotically stable, and a state estimator of considered complex networks can be achieved by solving some linear matrix inequalities. Finally, a numerical example is provided to illustrate the effectiveness of the proposed method.

Yushun Tan, Jinliang Liu, Yuanyuan Zhang

Integrated Iterative Learning Control Strategy for Batch Processes

An integrated iterative learning control strategy based on time-varying perturbation models for batch processes is proposed in this paper. A linear perturbation model is firstly obtained in order to control the perturbation variables rather than the actual process variables themselves. Next, an integrated control strategy which combines ILC with real-time feedback control is used to control the perturbation model. It leads to superior tracking performance and better robustness against disturbance and uncertainty. Lastly, the effectiveness of the proposed method is verified by examples.

Li Jia, Tian Yang, Min-Sen Chiu

The Fifth Section: Communication and Control for Distributed Networked Systems

Almost Sure Consensus of Multi-agent Systems over Deterministic Switching Agent Dynamics and Stochastic Topology Jumps

In this paper, consensus of multi-agent systems under both determinstic switching agent dynamics and stochastic topology jumps is investigated. The analysis relies on the fact that these two switching signals are independent. First, a necessary and sufficient condition for achieving a consensus under time-invariant agent dynamics and fixed communication topology is presented. Based on this result, the almost sure consensus condition under deterministic switching agent dynamics and stochastic topology jumps is investigated. It is shown that almost sure consensus can be reached if determinstic switching signal satisfies minimum average dwell time constraint.

Min Zheng, Zheng Mao, Kang Li

An Effective Integral Quadratic Constraint Construction Approach for Stability Analysis of a Class of Systems with Time Delays

This paper proposes an effective method for the construction of Integral Quadratic Constraints(IQC) based on Quadratic Separation(QS) framework for the stability analysis of a wide range of systems with time delays. An unified framework for both the QS framework and the Lyapunov theory is proposed, an effective IQCs construction method for Lyapunov candidates is thus obtained. This method is then applied to the stability analysis of linear systems with time delays. Stability criteria are established based on the proposed approach. Numerical examples are finally provided to show its effectiveness.

Min Zheng, Zheng Mao, Kang Li

An Online Recursive Identification Method over Networks with Random Packet Losses

Under the network environment, data packet losses bring an undesirable effect on parameter identification. To solve this problem, an online recursive identification method over networks with random packet losses is proposed. In this algorithm, the Bernoulli processes is firstly employed to describe the character of data packet losses. Considering random packet losses, an improved recursive least squares (RLS) algorithm is then presented by using Givens transformation, where the intermediate matrix can be updated recursively. Simulation results show that the proposed algorithm is able to significantly enhance the accuracy of parameter estimation, and the estimated variance becomes smaller as the number of iterations increases.

Dajun Du, Lili Shang, Wanqing Zhao

Output Feedback Reliable H  ∞  Control for Networked Control System

This paper investigates the static output feedback control for networked control systems with actuator failure. The failure is composed of two parts, the linear deficiency of the control gain and the nonlinearity varying with the control input. Based on the more general actuator fault model, linear matrix inequality (LMI) optimization approach is used to design the reliable

H

 ∞ 

output feedback control. Finally, an example is provided to demonstrate the design method.

Zhou Gu, Shumin Fei, Baochun Zhuang, Guangtao Shi

A Novel Adaptive Event-Triggered Communication Scheme for Networked Control Systems with Nonlinearities

This paper presents a novel adaptive event-triggered communication scheme for networked control systems (NCSs) with nonlinearities. Firstly, a novel adaptive event-triggered communication scheme for NCSs with nonlinearities is proposed, which can adaptively adjust the event-triggered communication threshold with respect to dynamic error to save the limited communication resource while ensuring the desired control performance. Secondly, a model of the considered system is built under consideration of the network-induced delay, adaptive event-triggered communication scheme and nonlinearities in a unified framework. Then, sufficient stability and stabilization criteria are obtained to judge the mean-square sense asymptotically stable for the studied system. Finally, two examples illustrate the effectiveness of the developed method.

Jin Zhang, Chen Peng, Dacheng Peng

Analytical Model for Epidemic Information Dissemination in Mobile Social Networks with a Novel Selfishness Division

With the prosperity of smartphone markets, mobile social networks (MSNs) attract much attention. In MSNs, information can be shared among users through their opportunistic contacts. The information forwarding mode needs mobile users to work in a cooperative and altruistic way. However, in the real world, most of users are not willing to forward information because of their selfishness. In this paper, we firstly divide selfishness into two new types: weak selfishness and extreme selfishness. Then we develop an analytical model to evaluate the influence of selfish behavior on the information dissemination process in the MSNs. Numerical results demonstrate weak selfishness and extreme selfishness hinder information dissemination.

Qichao Xu, Zhou Su, Bo Han, Dongfeng Fang, Zejun Xu

Distributed Collaborative Control Based on Adaptive Chaos Mutation PSO with WSAN

In large scale wireless sensor and actuator network (WSAN), unreliable wireless and multihop communications make challenges in designing centralized control due to severe packet dropout and latency. To this end, distributed collaborative control (DCC) scheme based on adaptive chaos mutation particle swarm optimization (ACMPSO) is proposed. In this scheme, control task is carried out collaboratively based on hierarchical clustering using only local information. Moreover, DCC is formulated as an optimization problem, and chaos mutation is introduced to improve the optimal performance of PSO. Simulation shows that the improved PSO has better convergence property, and the proposed scheme works well to obtain control objective.

Fuqiang Li, Jingqi Fu, Dajun Du, Weihua Bao

Distributed Fault Detection and Isolation for Discrete Time Multi-agent Systems

In this paper a distributed fault detection and isolation (FDI) scheme for discrete time multi-agent systems based on a filter is proposed. An FDI filter is designed such that the effects of external input disturbances on the residual signals are minimized and those of fault signals on the residual signals are maximized. These two optimal specifications are represented by

H

 ∞ 

and

H

 −

. Although sufficient conditions for the existence of such filters are obtained in terms of matrix inequality feasibility conditions, an algorithm to solve them is provided to find a feasible solution such that the filters can be established.

Dong Wang, Wei Wang

The Design Of Serial Communication Module between Android and Muti-lower Machines

Due to the low price of serial communication, it has been widely used among the industrial field. Human machine interface based on Android platform has not been widely used in the industrial field. By the article, Human-machine interface has been designed to achieve the communication between HMI and multiple lower machine.

Yun-xia XI, Xin Sun

Design for MIMO Networked Control Systems Based on Multi-threshold Dead Band Scheduling

The control and scheduling co-design of MIMO networked control system are studied in this paper. Considering the MIMO networked control system with multiple sensor nodes, the deadband scheduling strategy with multi-threshold is employed, then the transmit character of the sensor node’s data packets is analyzed. To compensate the effect of the deadband and network-induced delay, an observer is employed, and the closed loop system with the observer and dynamic feed back controller is modeled as a discrete time linear system with uncertainty. Then by the Lyapunov function and LMIs method, the sufficient condition for asymptotic stability is presented, together with the design method for the dynamic feedback controller and observer. Finally, a numerical example is given to validate the method.

Chenyu Zhang, Weihua Fan, Ronghua Xie, Qingwei Chen

Adaptive Distributed Event-Triggered Collaborative Control with WSAN

In large scale wireless sensor and actuator network (WSAN), severe packet dropout and latency make challenges in designing centralized control scheme. Moreover, if system runs stably with little fluctuation of states, period sampling always leads to waste of energy and communication resources. To this end, adaptive distributed event-triggered collaborative control scheme is proposed. In this scheme, distributed event-triggered (DET) mechanism is replaced by adaptive DET mechanism with variable triggering threshold in order to reduce updates of controller law further. Based on local information, distributed collaborative control (DCC) is carried out and formulated as optimization problem. And an improved particle swarm optimization (PSO) with linearly decreasing inertia weight is used to make a good tradeoff between global and local search ability. Simulation shows that, comparing with DET scheme, while guaranteeing required closed-loop performance, ADET scheme needs fewer executions of control task.

Fuqiang Li, Jingqi Fu, Dajun Du, Weihua Bao

Synthesis of Multi-robot Formation Manoeuvre and Collision Avoidance

This paper presents the synthesis of a multi-robot formation manoeuvre and collision avoidance. Turning-compliant waypoints are first achieved to support the multi-robot formation manoeuvre. The formation-based collision avoidance is then presented to translate the collision avoidance problem into the formation stability problem. The extension-decomposition-aggregation scheme is next employed to solve both the formation control problem and the collision avoidance problem during the multi-robot formation manoeuvre. Simulation study finally shows that the formation control and the collision avoidance can be simultaneously solved if the stability of the expanded formation including unidentified objects can be satisfied.

Aolei Yang, Wasif Naeem, Minrui Fei, Li Liu, Xiaowei Tu

Data Communications for Intelligent Electric Vehicle Charging Stations

This paper describes the communication requirements for an intelligent electric vehicle charge station which can provide inertial support to the electricity grid. The application is described. Telecoms delivery technologies are experimentally assessed, and an open source measurement system is discussed.

David Laverty, Kang Li, Jing Deng

Human-UAV Coordinated Flight Path Planning of UAV Low-Altitude Penetration on Pop-Up Threats

Human-UAV coordinated flight path planning of UAV low-altitude penetration on pop-up threats is a key technology achieving manned and unmanned aerial vehicles cooperative combat and is proposed in this paper. In the most dangerous environment, human’s wisdom, experience and synthetic judgments can make up for the lack of intelligence algorithm. By using variable length gene encoding based on angle for the flight paths planning, and combining artificial auxiliary decision with novel intelligence algorithm, it makes the best possible use of the human brain to guide solution procedures of the flight path planning on pop-up threats. A lot of simulation studies show that the on-line three-dimensional flight paths by this technology can meet the requirements of UAV low-altitude penetration, efficient implementation of threat avoidance, terrain avoidance and terrain following. This method has a certain practicality.

Peng Ren, Xiao-guang Gao, Jun Chen

Switched Topology Control and Negotiation of Distributed Self-healing for Mobile Robot Formation

In this paper, we investigate robot failure problem in mobile robot formation. A recursive and distributed self-healing algorithm is proposed to restore network topology when one or more robots fail. Firstly, a switched topology control method is introduced to restore the synchronization and connectivity of mobile robot network recursively. Then, a negotiation mechanism is further presented which achieves individual control in switched topology process. This mechanism only needs local information interactions between neighbors. Finally, the effectiveness of the proposed algorithm is validated by results of both simulations and real experiments.

Jianjun Ju, Zhe Liu, Weidong Chen, Jingchuan Wang

Design of Embedded Redundant Gateway Based on Improved TCP Congestion Algorithm

In wireless sensor networks, gateway is a hub in data transmission between wireless nodes and Personal Computer (PC). Once the gateway fails, the entire network will paralyze. To ensure the stability of the network, a methodology for the design of redundant gateway based on improved Transmission Control Protocol (TCP) congestion algorithm is proposed in this paper. In this methodology, two gateways physically connected by Ethernet are logically divided into a primary one and a secondary one. Secondary gateway is the backup of the primary gateway. The paper gives an improve method of congestion algorithm, which is applied in this scheme. It can better satisfy the high demands of real-time communication in the embedded system when compares with the traditional TCP congestion algorithm. Experiment shows the effectiveness of improved TCP congestion algorithm, when the primary gateway fails. Secondary gateway can rapidly replace the primary gateway and it ensures a better reliability of the network.

Qin Lai, Jingqi Fu, Weihua Bao

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise