In order to improve the treatment efficiency of the constrained layer damping (CLD), it is meaningful to have a study on its layout optimization. This paper based on current research of Cellular Automata (CA) algorithm for CLD optimization, points out that the current application of CA algorithm is less efficient, considering that adding or deleting a single CLD patch almost does no change to the entire structure, then makes a simplification and modification of CA algorithm. With the aim at improving the structural modal loss factor, this paper uses the current CA algorithm and its modified algorithm to optimize a partial CLD plate. The results have shown that the modified CA algorithm for CLD optimization can not only substantially increase computing efficiency, but to some extent, improve the optimization effectiveness.

A longitudinal vibration coupling with a flexural vibration in a curved fluid-filled periodic pipe is studied. The pipe is fabricated by a periodic composite material structure based on the Bragg scattering mechanism of Phononic Crystals. Using the transfer matrix method, the band structure of an infinite periodic straight pipe is calculated, and the elastic wave propagation characteristics of the periodic pipe are discussed. Furthermore, the vibrational frequency response functions of a finite curved pipe are performed and the coupled vibration is studied. Finally, the transmission properties of longitudinal vibration, flexural vibration and their coupling vibration in the curved periodic fluid-filled pipe are investigated.

Considering the uncertainty of maintainability design factors and the fuzzy preference information in evaluation values, an evaluation and decision-making method of maintainability was proposed. The linguistic assessment values of maintainability design factors were converted to trapezoid fuzzy numbers by fuzzy set theory, and decision weights under fuzzy preference were given based on maximizing deviation method. Then, evaluation model of maintainability design was established, and confidence index of evaluating decision was given. At last, a case study was given and the result showed that the method is reasonable and effective for maintainability evaluation.

In conventional federated learning, each device is restricted to train a network model of the same structure. This greatly hinders the application of federated learning where the data and devices are quite heterogeneous because of their different hardware equipment and communication networks. At the same time, existing studies have shown that transmitting all of the model parameters not only has heavy communication costs, but also increases risk of privacy leakage. We propose a general framework for personalized federated learning (PerHeFed), which enables the devices to design their local model structures autonomously and share sub-models without structural restrictions. In PerHeFed, a simple-but-effective mapping relation and a novel personalized sub-model aggregation method are proposed for heterogeneous sub-models to be aggregated. By dividing the aggregations into two primitive types (i.e., inter-layer and intra-layer), PerHeFed is applicable to any combination of heterogeneous convolutional neural networks, and we believe that this can satisfy the personalized requirements of heterogeneous models. Experiments show that, compared to the state-of-the-art method (e.g., FLOP), in non-IID data sets our method compress ≈ 50% of the shared sub-model parameters with only a 4.38% drop in accuracy on SVHN dataset and on CIFAR-10, PerHeFed even achieves a 0.3% improvement in accuracy. To the best of our knowledge, our work is the first general personalized federated learning framework for heterogeneous convolutional networks, even cross different networks, addressing model structure unity in conventional federated learning.

No abstract