Liao Ruiquan

Generalized accumulated GM(1,1) cosine model and its application

Based on the generalized accumulated generating operation, in this paper GAGM(1,1) cosine model is proposed to improve the classical GM(1,1) cosine model. The GAGM(1,1) cosine model can fit in many kind of periodic sequences, such as non-equidistant periodic sequences and non-equidistant periodic sequences with jump point or with multi-stage. Also GAGM(1,1) cosine models parameter space and solving process are discussed in details. At last the proposed model is used to predict oil production. The results showed that the precision of GAGM(1,1) cosine model is higher than GM(1,1) model and confirmed that the GAGM(1,1) cosine model has important theoretic and practical significance.Purpose – In practical engineering, there are various sequences. The purpose of this paper is to improve and put forward a new model.Design/methodology/approach – Based on the generalized accumulated generating operation, the paper improved the GM(1,1) cosine model and proposed the GAGM(1,1) cosine model.Findings – It is found that the GAGM(1,1) cosine model can satisfy many kinds of periodic sequences and the precision of GAGM(1,1) cosine model is higher than the GM(1,1) model.Originality/value – The paper shows that the GAGM(1,1) cosine model has important theoretic and practical significance.

A Modified Particle Swarm Optimization Algorithm with Dynamic Adaptive

To overcome premature searching by standard particle swarm optimization (PSO) algorithm for the large lost in population diversity, the measure of population diversity and its calculation are given, and an adaptive PSO with dynamically changing inertia weight is proposed. Simulation results show that the adaptive PSO not only effectively alleviates the problem of premature convergence, but also has fast convergence speed for balancing the trade-off between exploration and exploitation.

Dynamic thickening investigation of the gelation process of PAM/PEI system at high temperature and high pressure

ABSTRACT The gelation process of organically polymer gel was investigated by dynamic thickening measurements. Rheological measurements were used to evaluate the viscosity of the gel. During the gelation process, high temperature resulted in higher rate of crosslinking. Rigid and stable gel was formed in neutral and alkaline media, and the higher of the pH value, the faster of the gelation process. However, gel could not be formed in acid condition. Moreover, the rate of crosslinking increased with the increase of concentration of polymer and crosslinker. The addition of NH4Cl elongated the gelation time significantly, but played a negative role in the gel strength, while a rigid gel was formed in the presence of Sodium acetate or trisodium citrate dehydrate. This paper summarizes the results and discusses how various parameters affect the gelation process of the gel. GRAPHICAL ABSTRACT

Optimal Capacity Allocation on Heterogeneous Complex Transport Networks

We study how the node delivering capacity is allocated so that the traffic transport efficiency can be enhanced maximally. Network heterogeneity of degree distribution and processing delay of the traffic are considered. An explicit analytical solution is provided, which is based on the M/M/1 queueing theory and optimization principle, provided that the network structure and routing strategy are given. In particular, we extend the relevant conclusions in the literature [Europhys. Lett. 83 (2008) 28001]. Finally, an order parameter simulation example by comparing results with those obtained via simple capacity allocation in large Barabasi—Albert (BA) scale-free network is provided to illustrate the effectiveness of the theoretical results.

A Modified Particle Swarm Optimization and Simulation

To overcome premature searching by standard particle swarm optimization (PSO) algorithm, a new modified PSO with information of the closest particle is proposed. In the algorithm, the particle is updated not only by the best previous position and the best position among all the particles in the swarm, but also by the best previous position of the closest particle. To balance the trade-off between exploration and exploitation and convergence to the global optimum solution, a linearly varying acceleration coefficient over the generations was introduced. The simulation results show that the algorithm has better probability of finding global optimum and mean best value than others algorithm, especially for multimodal function.

Analysis and dynamical changing inertia weight strategy of particle swarm optimization

Convergence of particle velocity and effect on optimization performance were analyzed in particle swarm optimization, and a new algorithm with dynamical changing inertia weight was proposed. The information defined as the average absolute value of velocity of all particles was used in the algorithm, which can avoid premature convergence for the velocity is closed to 0 in the early search part. The simulation results show that the algorithm has better probability of finding global optimum and mean best value than other algorithms for maintaining the population diversity.