This paper presents a tolerance analog circuit hard fault and soft fault diagnosis method based on the BP neural network and particle swarm optimization algorithm. First, select the mean square error function of BP neural network as the fitness function of the PSO algorithm. Second, change the guidance of neural network algorithms rely on gradient information to adjust the network weights and threshold methods, through the use of the characteristics of the particle swarm algorithm groups parallel search to find more appropriate network weights and threshold. Then using the adaptive learning rate and momentum BP algorithm to train the BP neural network. Finally, the network is applied to fault diagnosis of analog circuit, can quickly and effectively to the circuit fault diagnosis.

Base on improved particle swarm algorithm, this paper proposes a linear decreasing inertia weight particle swarm algorithm and error back propagation algorithm based on hybrid algorithm combining. The linear decreasing inertia weight particle swarm algorithm and momentum-adaptive learning rate BP algorithm interchangeably adjust the network weights, so that the two algorithms are complementary. It gives full play to the PSO's global optimization ability and the BP algorithm local search advantage, to overcome the slow convergence speed and easily falling into local weight problems. Simulation results show that this diagnostic method can be used for tolerance analog circuit fault diagnosis, with a high convergence rate and diagnostic accuracy.

In order to improve the simulation accuracy for arc welded structures, the refined numerical simulation approach using the solid model with material mechanical properties has been presented. The arc welded joints are frequently utilized in the manufacture of neighborhood electric vehicle. Firstly, the mechanical properties of the base metal and three types of welded joints were determined by the uniaxial tensile experiments, and experimental results revealed that the mechanical property significantly reduced after welding. Then the numerical simulation approach using solid model and approach using weld constraint were conducted on the butt welded joints of 6061 aluminum alloy in LS-DYNA software, the simulation results were compared with experimental results. Moreover, the refined simulation approach using solid model was validated by different types of arc welded joints and their corresponding experiments. It is concluded that the proposal simulation approach using solid model shows higher accuracy than the approach using weld constraint on predicting deformation of butt welded and fillet welded joints.

When medical image transmitted and stored on the Internet, it is vulnerable to many kinds of attacks, thus, the security of the patients personal information is low. A robust watermarking algorithm has been proposed to increase the security of medical images. The scheme obtains the visual feature vectors of the medical image using DWT-DFT. At the same time, applying Logistic Map to encrypt the watermarking image, which can enhance its security. The experiment results show that the scheme has strong robustness against geometrical attacks.

This paper presented a new watermarking algorithm based on 3D-DCT and chaotic neural network in order to protect three-dimensional medical images. The algorithm adopts a wealth of information grayscale image as a watermark and three-dimensional medical image as the original carrier. It utilizes chaos neural network for scrambling and encryption of watermarking image. The embedded watermark has cryptographic security significance. Experimental results show that the algorithm is simple, which is a blind watermarking algorithm; the watermark extraction process does not require the original image. It is capable of resisting shear and filtering attacks.