RBF神经网络在\"薄管板\"结构强度分析中的应用

叙述了RBF神经网络的结构和\"薄管板\"结构强度分析过程以及应用RBF神经网络对\"薄管板\"结构应力的非线性系统进行了系统辩识,网络输出值与计算值符合较好.说明基于神经网络的系统辩识方法在结构强度分析过程中具有一定的实用性.     

Bifurcation of a three-unit neural network

Considered is a system of delay differential equations modeling a time-delayed connecting network of three neurons without self-feedback. Discussing the change of the number of eigenvalues with zero real part, we locate the boundary of the stability region and finally determine the largest stability region of trivial solution. We investigate the existence of bifurcation phenomena of codimension one/two of the trivial equilibrium by considering the intersections of some parameter curves, which, in the aτ-half parameter plane, correspond to zero root or pure imaginary roots. In particular, the equivariant bifurcation is studied because of the equivariance of the system. We also present numerical simulations to demonstrate the rich dynamical behavior near the equivariant Pitchfork-Hopf bifurcation points, Hopf-Hopf bifurcation points, and some higher codimension bifurcation points.     

Exact solutions for network rewiring models

Evolving networks with a constant number of edges may be modelled using a rewiring process. These models are used to describe many real-world processes including the evolution of cultural artifacts such as family names, the evolution of gene variations, and the popularity of strategies in simple econophysics models such as the minority game. The model is closely related to Urn models used for glasses, quantum gravity and wealth distributions. The full mean field equation for the degree distribution is found and its exact solution and generating solution are given.     

A hybrid genetic algorithm–neural network strategy for simulation optimization

Simulation optimization aims at determining the best values of input parameters, while the analytical objective function and constraints are not explicitly known in terms of design variables and their values only can be estimated by complicated analysis or time-consuming simulation. In this paper, a hybrid genetic algorithm–neural network strategy (GA–NN) is proposed for such kind of optimization problems. The good approximation performance of neural network (NN) and the effective and robust evolutionary searching ability of genetic algorithm (GA) are applied in hybrid sense, where NNs are employed in predicting the objective value, and GA is adopted in searching optimal designs based on the predicted fitness values. Numerical simulation results and comparisons based on a well-known pressure vessel design problem demonstrate the feasibility and effectiveness of the framework, and much better results are achieved than some existed literature results.     

Process prediction of Ni–SiC coatings based on RBF-BP model

The preparation of Ni–SiC coatings using magnetic field-assisted jet electrodeposition under various plating settings is described in this study. A RBF-BP composite neural network with 4 × 4 × 4 × 7 × 10 × 1 was used to predict the corrosion resistance of Ni–SiC coatings prepared by employing different plating parameters. The results show that the fitting degree between the expected value and the actual value of the RBF-BP composite neural network is 0.97497. Moreover, the hybrid neural network can accurately predict the corrosion resistance of Ni–SiC coatings prepared under different process parameters. The corrosion weight loss of the coating is the lowest at the current density of 4 A/dm², a jet rate of 3 m/s, a SiC particle concentration of 8 g/L, and at a magnetic field intensity of 0.8 T, demonstrating its corrosion resistance under these conditions. According to the coating characterization analysis, the coating's grain size was significantly refined, and the surface was smoother with a high amount of uniformly sized SiC nanoparticles.     

A novel algorithm for chromatogram matching in qualitative analysis

The matching of the pattern of peaks produced during gas chromatography is of importance to many applications. At present, this task is generally performed manually, but this generates the usual problems associated with human inspection, such as a lack of objectivity and reproducibility, proneness to errors, and practical restriction of the volume of data which can reasonably be processed. This paper explores the use of a novel algorithm for automation of this task. The performance of the method on well controlled simulated data sets and real chromatograms is used to show not only how problems of manual inspection can be circumvented, but also how the existence of such a powerful method should open up the possibility of many new analyses for quality control, discrimination of varieties of sample, and the identification of specific components within a sample.     

Coordination Polymers. 9. Chelate Polymers Derived from Bisphenolic Complexes and Propylenediamine

Chelate polymers derived from bis(2,4-dihydroxybenzaldehyde)propyl-enediimine M and bis(2,4-dihydroxyacetophenone)propylenediimine M (M = Fe²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺) with aromatic acid chlorides were prepared by interfacial polycondensation. Also, chelate polysiloxanes were obtained from the same monomers and α,ω-dichloropolydimethyl-siloxane. The spectral, thermal, magnetic, and electrical properties of the polychelates were studied.     

Neural Network Inference of Molar Mass Distributions of Peptides during Tailor-Made Enzymatic Hydrolysis of Cheese Whey: Effects of pH and Temperature

The fine-tuning of the enzymatic hydrolysis of proteins may provide a pool of peptides with predefined molar mass distributions. However, the complex mixture of molecules (peptides and amino acids) that results after the proteolysis of cheese whey turns unfeasible the assessment of individual species. In this work, a hybrid kinetic model for the proteolysis of whey by alcalase, multipoint-immobilized on agarose, is presented, which takes into account the influence of pH (8.0-10.4) and temperature (40-55 degrees C) on the activity of the enzyme. Five ranges of peptides' molar mass have their reaction rates predicted by neural networks (NNs). The output of NNs trained for constant pH and temperatures was interpolated, instead of including these variables in the input vector of a larger NN. Thus, the model complexity was reduced. Coupled to differential mass balances, this hybrid model can be employed for the online inference of peptides' molar mass distributions. Experimental kinetic assays were carried out using a pH-stat, in a laboratory-scale (0.03 L) batch reactor. The neural-kinetic model was integrated to a supervisory system of a bench-scale continually stirred tank reactor (0.5 L), providing accurate predictions during validation tests.     

Hopfield neural networks in large-scale linear optimization problems

Hopfield neural networks and affine scaling interior point methods are combined in a hybrid approach for solving linear optimization problems. The Hopfield networks perform the early stages of the optimization procedures, providing enhanced feasible starting points for both primal and dual affine scaling interior point methods, thus facilitating the steps towards optimality. The hybrid approach is applied to a set of real world linear programming problems. The results show the potential of the integrated approach, indicating that the combination of neural networks and affine scaling interior point methods can be a good alternative to obtain solutions for large-scale optimization problems.