Fault tracing method of metering device based on deep belief network
A technology of deep belief network and metering device, which is applied in the field of power system distribution transformer data processing, can solve the problems of slow processing speed, difficulty in processing high-dimensional features, and large memory usage
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0062] The invention will be further described below in combination with the accompanying drawings.
[0063] 1. Basic principle of deep belief network model:
[0064] Deep belief network (DBN) is a neural network model composed of multiple restricted Boltzmann machines (RBM), and its core is RBM unit. DBN is an improved neural network model based on traditional BP neural network, which is composed of multiple RBM stacks. In the first step, unsupervised greedy algorithm is used to train each layer of RBM separately in the pre training process. A more reasonable pre training process can provide a good initial value of weight for the whole DBN network and facilitate subsequent training. In the second step, the traditional feedforward BP back propagation is used to change the value, so as to obtain the optimal position of local convergence.
[0065] Figure 1 Represents a simple DBN network structure composed of three-layer RBM. Figure 1 Medium V 1 Is the visible layer connecting the ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


