Search

In this paper, we propose a novel hybrid algorithm to construct an improved recurrent neuron network (RNN) for modeling tunnel microwave thermal process. The new design involves a hierarchical learning process, in which the recurrent neurons of RNN are optimized by the cuckoo search (CS) algorithm, while effectiveness and efficiency of the model are guaranteed by using the orthogonal least squares (OLS) method, which is a fast approach for construction of neural networks in a stepwise forward procedure. The major contribution is to integrate seamlessly the OLS model selection and CS neuron optimization in an innovative way so that it can well track the underlying dynamic of this complicated thermal process with a very sparse model. By conducting a microwave rice drying experiment, a set of real-world datasets is used to drive the RNN model. Simulation results demonstrate the effectiveness of the proposed model compared with existing well-known approaches in terms of modeling accuracy and model compactness. [ABSTRACT FROM AUTHOR]

• We propose a fresh distributed algorithm to solve the EDP, which not only allows the utilization of constant step-size, but also provides a wider range of its selection. • If the cost functions meet the strongly convex and smooth condition, theoretical analysis indicates that the algorithm can achieve the optimal dispatch at a linear convergence rate. • The linear convergence rate of the algorithm is tunable, and a relationship among the linear rate, cost functions, network topology, weight matrix and constant step-size is established. The increasingly complex modern energy network arouses the need of flexible and dependable approaches to solve the economic dispatch problem (EDP) in the smart grids. Toward this end, this paper develops a fresh distributed algorithm with constant step-size, which aims to schedule the power generation among generators by complying with individual generation capacity limits to satisfy the total load demand at the minimized cost. The convergence of the proposed algorithm is analyzed through utilizing the Lyapunov method and the spectral decomposition technique. When the selected constant step-size is smaller than a specifically provided upper bound, the theoretical analysis demonstrates that the proposed algorithm can linearly achieve the optimal solution of the EDP under the smooth and strongly convex assumption on generation cost functions. In particular, the linear convergence rate of the proposed algorithm is tunable, and a relationship among the linear convergence rate, generation cost functions, network topology, weight matrix and constant step-size is established. The availability of the proposed algorithm is verified through simulation experiments. [ABSTRACT FROM AUTHOR]

• We propose an asynchronous distributed primal-dual algorithm to solve the EDP, which utilizes asynchronous communication strategy for dealing with the link failures over stochastic networks. • If the expectation of stochastic networks is connected, theoretical analysis manifests that the algorithm can exactly seek the optimal solution with probability one. • The algorithm allows uncoordinated constant step-sizes for the local optimization of each node, which guarantee the algorithm achieves a linear convergence rate to the optimal solution in contrast to most existing literature. The economic dispatch problem (EDP) in power systems is usually formulated as a category of convex optimization problems in which the collective cost function is expressed as the sum of all individual objectives of generators. It is inclined to be solved through a distributed way, which is to minimize the total generation cost while meet the demands under generator capability restrictions. In this paper, we consider this class of convex optimization problems that involve coupling linear constraint and individual box constraints. In order to make information communication of network decentralized, reliable and computationally inexpensive, an asynchronous distributed primal–dual optimization algorithm is proposed over stochastic networks. This algorithm allows nodes to use asynchronous communication strategy to update their state, and the step-sizes for seeking the optimum solution are uncoordinated constants. The asynchronous operation for generators takes link failures of networks into account. Under strongly convex assumption on objective functions, it is proved that the proposed algorithm can seek the exact optimal solution with probability one if the expectation of communication network is undirected connected. The effectiveness of the proposed optimization algorithm over stochastic networks is illustrated by provided simulation results. [ABSTRACT FROM AUTHOR]

Since most real-world processes exhibit both nonlinear and time-varying characteristics, there exists a need for accurate and efficient models that can adapt in nonstationary environments. Also for adaptive control purpose, it is vital that an adaptive model has a fixed small model size. In this paper, we propose an adaptive tunable gradient radial basis function (GRBF) network for online modeling of nonlinear dynamic processes, which meets these practical requirements. Specifically, a compact GRBF model is constructed by the orthogonal least squares algorithm in training, which is capable of modeling variations of local mean and trend in the data well. During online operation, the adaptive GRBF model tacks the time-varying process's dynamics by replacing a worst performing node with a new node which encodes the current new data. By exploiting the local predictor property of the GRBF node, the new node optimization can be done extremely efficiently. The proposed approach combining the advantages of both the GRBF network structure and fast tunable node mechanism is capable of tracking the time-varying nonlinear dynamics accurately and effectively. Extensive simulation results demonstrate that the proposed fast tunable GRBF network significantly outperforms the existing state-of-the-art methods, in terms of both adaptive modeling accuracy and online computational complexity. • Fast adaptive gradient radial basis function (GRBF) network for online modeling and prediction of nonlinear and nonstationary dynamic processes. • Small fixed number of nodes encode the learnt process's states, and each node is an accurate local predictor of the corresponding process state. • During online adaptation, the worst performing node is replaced by the newest process state which automatically becomes the perfect local predictor for the process output. • This tunable GRBF model is suitable for adaptive control system applications. [ABSTRACT FROM AUTHOR]

We calculated the crustal stress field using the composite focal mechanism method based on the P-wave initial motion polarity data of the Tengchong volcanic area from January 2011 to April 2019 obtained from the Bulletin of Seismological Observations of Chinese stations. The magnitude range of earthquakes used in this study is 0–4, and their magnitudes are mainly approximately 1.0. To investigate the influence of the source location on the stress field and obtain reliable stress fields of the study area, we applied the double-difference algorithm to relocate the seismic events, obtaining more accurate and reliable relative positions of seismic events with a clearer seismic belt. On the basis of relocation results, the study on the stress field along the fault zone was conducted, and the influence of seismic event position on the stress field was analyzed. Results show that, first, the current stress regime in the shallow crust of the Tengchong volcanic area is strike-slip faulting, the orientation of the principal compressive stress axis is NE-SW, the orientation of the principal extension stress axis is SE-NW, the principal compressive and extension stress axes are nearly horizontal, and the dip angle of intermediate principal stress axis is relatively large. This reflects that the volcanic and seismic activities in the Tengchong volcanic area are mainly controlled by the collision and squeezing effect of the Indian-Eurasian plate. It also reflects that the current tensile action caused by deep magma activity has little influence on the shallow crustal stress field. Second, the stress field along fault zones reveals that there exist local stress fields, such as the thrust stress regime at the strike-slip fault terminal area, which is consistent with the compressional area at the intersection of conjugate strike-slip faults indicated by previous study. Third, the stress field results are consistent, regardless of using the original location in the bulletin or the relocated location, indicating that the influence of the event location error can be neglected when there are sufficient data and reflecting the stability of the composite focal mechanism method. The findings can serve as a reference for investigating geological structure movement, seismic activities, and volcanic activities in the Tengchong volcanic area. [ABSTRACT FROM AUTHOR]

Background: Insomnia as one of the dominant diseases of traditional Chinese medicine (TCM) has been extensively studied in recent years. To explore the novel approaches of research on TCM diagnosis and treatment, this paper presents a strategy for the research of insomnia based on machine learning. Methods: First of all, 654 insomnia cases have been collected from an experienced doctor of TCM as sample data. Secondly, in the light of the characteristics of TCM diagnosis and treatment, the contents of research samples have been divided into four parts: the basic information, the four diagnostic methods, the treatment based on syndrome differentiation and the main prescription. And then, these four parts have been analyzed by three analysis methods, including frequency analysis, association rules and hierarchical cluster analysis. Finally, a comprehensive study of the whole four parts has been conducted by random forest. Results: Researches of the above four parts revealed some essential connections. Simultaneously, based on the algorithm model established by the random forest, the accuracy of predicting the main prescription by the combinations of the four diagnostic methods and the treatment based on syndrome differentiation was 0.85. Furthermore, having been extracted features through applying the random forest, the syndrome differentiation of five zang-organs was proven to be the most significant parameter of the TCM diagnosis and treatment. Conclusions: The results indicate that the machine learning methods are worthy of being adopted to study the dominant diseases of TCM for exploring the crucial rules of the diagnosis and treatment. [ABSTRACT FROM AUTHOR]

Temperature, especially temperature distribution, is one of the most fundamental and vital parameters for theoretical study and control of various industrial applications. In this paper, ultrasonic thermometry to reconstruct temperature distribution is investigated, referring to the dependence of ultrasound velocity on temperature. In practical applications of this ultrasonic technique, reconstruction algorithm based on least square method is commonly used. However, it has a limitation that the amount of divided blocks of measure area cannot exceed the amount of effective travel paths, which eventually leads to its inability to offer sufficient temperature information. To make up for this defect, an improved reconstruction algorithm based on least square method and multiquadric interpolation is presented. And then, its reconstruction performance is validated via numerical studies using four temperature distribution models with different complexity and is compared with that of algorithm based on least square method. Comparison and analysis indicate that the algorithm presented in this paper has more excellent reconstruction performance, as the reconstructed temperature distributions will not lose information near the edge of area while with small errors, and its mean reconstruction time is short enough that can meet the real-time demand. [ABSTRACT FROM AUTHOR]

In Yangtze River traffic management, to manage ship sequencing and scheduling effectively and efficiently in restricted waterways has long been a challenging issue. This paper proposes a Sliding Window based Online ship Sequencing and Scheduling algorithm (OSS-SW) to tackle this problem. The OSS algorithm is capable of generating a more efficient ship sequence by introducing the 'position shift' concept which takes advantages of ships' sailing time differences in the restricted waterway. The sliding window mechanism on the other hand is introduced to handle the traffic uncertainties and to reduce the computational complexity. The impact of the restricted waterway congestion on the performance of OSS-SW is also investigated. Further, the parameter settings of the proposed OSS-SW are investigated in detail. Both simulation studies and the real data applications confirm that the proposed OSS-SW algorithm outperforms the existing widely adopted Traffic Signal Revealing System (TSRS) and the Expert System based algorithm (FAHP-ES) in solving the ship sequencing and scheduling problem. • An online sequencing and scheduling algorithm for inland waterway traffic management. • The ships' sailing time in the restricted waterway is considered. • The sliding window is to decrease the computational cost and to handle uncertainties. • The parameters are discussed in detail to find the optimal settings. [ABSTRACT FROM AUTHOR]