Random vibration analysis with radial basis function neural networks

Random vibrations occur in many engineering systems including buildings subject to earthquake excitation, vehicles traveling on a rough road and off-shore platform in random waves. Analysis of random vibrations for linear systems has been well studied. For nonlinear systems, particularly for multi-degree-of-freedom systems, however, there are still many challenges including analyzing the probability distribution of transient responses of the system. Monte Carlo simulation was considered the only viable method for this task. In this paper, We propose a method to construct semi-analytical transient solutions of the probability distribution of transient responses of nonlinear systems by using the radial basis function neural networks. The activation functions consist of normalized Gaussian probability density functions. Two examples are presented to show the effectiveness of the proposed solution method. The transient probability distributions and response moments of these examples are presented, which have not been reported in the literature before.

Study On the Bistable Vibration Behaviour of a Rod Fastened Rotor-Bearing System

Based on the Lagrange equation, the motion equation of a rod fastened rotor-bearing system considering the damping of the contact interface is established. The numerical method is employed for numerical analysis. The bifurcation diagram, time series, frequency waveform, phase spectrum and Poincare map are used to illustrate the nonlinear dynamic behaviour. The transient responses during acceleration and deceleration are calculated to reveal the dynamic behaviour of the system. The numerical results hold that since the oil film is nonlinear, the system presents obvious bistable behaviour and a jumping phenomenon. In addition, a test bench of the rod fastened rotor-bearing system is built. The bistable behaviour and jumping phenomenon are experimentally proven, and the effect of the eccentric distance of the rotor on the bistable behaviour is experimentally explored. The results of this paper can be used for the basic design and fault diagnosis of rod fastened rotors.

A Nonlinear Model and Parameter Identification Method for Rubber Isolators under Shock Excitation in Underwater Vehicles

Rubber isolators are usually used to protect high-precision equipment of autonomous underwater vehicles (AUVs), avoiding damage from overlarge dynamic excitation. Considering the nonlinear properties of the rubber material, the nonlinear behavior of rubber isolators under shock exaltation is hard to be predict accurately without the available modal and accurate parameters. In view of this, the present study proposes a nonlinear model and parameter identification method of rubber isolators to present their transient responses under shock excitation. First, a nonlinear model of rubber isolators is introduced for simulating their amplitude and frequency-dependent deformation under shock excitation. A corresponding dynamic equation of the isolation system is proposed and analytically solved by the Newmark method and the Newton-arithmetic mean method. Secondly, a multilayer feed-forward neural network (MFFNN) is constructed with the current model to search the parameters, in which the differences between the estimated and tested responses are minimized. The sine-sweep and drop test are planned with MFFNN to build the parameter identification process of rubber isolators. Then, a T-shaped isolator composed of high-damping silicon rubber is selected as a sample, and its parameters were determined by the current identification process. The transient responses of the isolation system are reconstructed by the current mode with the identified parameter, which show good agreement with measured responses. The accuracy of the proposed model and parameter identification method is proved. Finally, the errors between the reconstructed responses and tested responses are analyzed, and the main mode of energy attenuation in the rubber isolator is discussed in order to provide an inside view of the current model.

Engineering digitizer circuits for chemical and genetic screens in human cells

Cell-based transcriptional reporters are invaluable in high-throughput compound and CRISPR screens for identifying compounds or genes that can impact a pathway of interest. However, many transcriptional reporters have weak activities and transient responses. This can result in overlooking therapeutic targets and compounds that are difficult to detect, necessitating the resource-consuming process of running multiple screens at various timepoints. Here, we present RADAR, a digitizer circuit for amplifying reporter activity and retaining memory of pathway activation. Reporting on the AP-1 pathway, our circuit identifies compounds with known activity against PKC-related pathways and shows an enhanced dynamic range with improved sensitivity compared to a classical reporter in compound screens. In the first genome-wide pooled CRISPR screen for the AP-1 pathway, RADAR identifies canonical genes from the MAPK and PKC pathways, as well as non-canonical regulators. Thus, our scalable system highlights the benefit and versatility of using genetic circuits in large-scale cell-based screening.