Metering device fault tracing method based on deep belief network
A deep belief network and metering device technology, applied in the field of power system distribution and transformation data processing, can solve the problems of slow processing speed, difficulty in processing high-dimensional features, and large memory footprint.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0062] The present invention will be further described below in conjunction with the accompanying drawings.
[0063] 1. Basic principles of deep belief network model:
[0064] Deep belief network (deep belief network, DBN) is a neural network model composed of multiple restricted Boltzmann machine (RBM) stacks, and its core is the RBM unit. DBN is an improvement on the basis of the traditional BP neural network, using a neural network model composed of multiple RBM stacks. The first step is to use the unsupervised greedy algorithm to train each layer of RBM separately in the pre-training process. A more reasonable pre-training process can provide a good initial weight value for the entire DBN network, and also provides convenience for subsequent training. . In the second step, the traditional feedforward BP backpropagation is used to change the value, so that the optimal position of local convergence can be obtained.
[0065] figure 1 Represents a simple DBN network struct...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


