Many applications for Restricted Boltzmann Machines (RBM) have been developed for a large variety of learning problems. Recent developments have demonstrated the capacity of RBM to be powerful generative models, able to extract useful features from input data or construct deep artificial neural networks. In this work, we propose a learning algorithm to find the optimal model complexity for the RBM by improving the hidden layer. We compare the classification performance of regular RBM use RBM() function, classification RBM use stackRBM() function and Deep Belief Network (DBN) use DBN() function with different hidden layer. As a result, Stacking RBM and DBN could improve our classification performance compare to regular RBM.

Comparative Analysis of Restricted Boltzmann Machine Models for Image Classification

At present,there is not a unified formula to analyze the output power and efficiency of the class D audio amplifier.According to the power amplifier equivalent circuit,the paper analyzes the waveform and the average value of the output current.The mathematical model is built for computing the output power and efficiency.It is proved by the concrete examples that the model is feasible.

Output power and efficiency of class D power amplifier

The high efficiency and small size of the switching power amplifier (SPA) make it more ideal than amplifiers of other types, and it has been utilized widely. Effective fault prognosis of the SPA is extremely necessary for improving system reliability. This paper proposes a way to use these state variables containing a large number of component fault information to predict system faults. This method relies on a Restricted Boltzmann Machine (RBM) and its derived algorithm, which has excellent information filtering, component analysis and feature extraction capabilities. The classification-restricted Boltzmann model was constructed, and its classification performance was tested using data sets. It has good performance in avoiding over-fitting and local optimal solutions, and reducing machine learning process complexity.

Fault Prognosis Based on Restricted Boltzmann Machine and Data Label for Switching Power Amplifiers

The invention discloses an accelerated storage and life test system for a flat spiral spring, which comprises an initial stress applying device, a measuring device and a temperature stress applying device, wherein the initial stress applying device is composed of a base, a center shaft, a sensor baffle, a lug piece force transmission structure and an auxiliary rod; the measuring device comprises aforce sensor and a strain gauge and is used for measuring the decay change of the spiral spring; and the temperature stress applying device is a high-temperature test box and is used for realizing temperature control between room temperature and 200 DEG C During a test, the spiral spring is arranged in the initial stress applying device and is placed in the high-temperature environment test box;and the strength and high-temperature-resistant design of the initial stress applying device meet the loading requirement under the highest stress condition According to the invention, the continuous with-temperature measurement of the stress relaxation of the flat spiral spring at a certain temperature can be realized; direct measurement of with-temperature stress relaxation at different temperatures from room temperature to 200 DEG C can be realized, which is closer to the actual situation; the system has the advantages of convenient assembly and disassembly, timely measurement and high measurement precision

Sheng Tao

Papers

Fault Prognosis Based on Restricted Boltzmann Machine and Data Label for Switching Power Amplifiers

Accelerated storage and life test system for flat spiral spring