A refrigeration equipment fault diagnosis device and system

Abstract

The invention discloses a refrigeration equipment fault diagnosis device and system. The refrigeration equipment fault diagnosis device comprises a management storage part, a node and layer number setting part, a deep neural network training part, a deep neural network model construction part, a training step number setting part, a fault diagnosis model building part, a learning rate optimizationpart and a fault diagnosis result generation part; the management storage part at least stores training group data and test group data; the deep neural network training part adopts an SGDM to train the deep neural network; the deep neural network model construction department establishes a deep neural network model according to the structure, activation function, loss function, etc. of the neuralnetwork; the fault diagnosis model construction part is used for constructing a fault diagnosis model after training by applying training group data in a deep neural network model, the learning rate optimization part is used for optimizing the learning rate by adopting a simulated annealing algorithm, and the fault diagnosis result generation part is used for carrying out fault diagnosis on test group data to obtain a fault diagnosis result.

Description

technical field [0001] The invention belongs to the refrigeration field, and relates to a refrigeration equipment fault diagnosis device and system, in particular to a chiller fault diagnosis device and system. Background technique [0002] Modern industrial and civil buildings are inseparable from refrigeration systems. The components of refrigeration systems are complex, resulting in frequent failures and various types during operation. The occurrence of faults affects the cooling effect, increases energy consumption, shortens the service life of the equipment, and brings potential safety hazards. Therefore, it is particularly important to ensure the operation quality of the refrigeration system. Real-time status detection and fault diagnosis of the refrigeration system can not only ensure the normal operation of the refrigeration system, but also detect and repair problems in time. In the past few decades, the fault diagnosis and detection technology of the refrigeration ...

AI Technical Summary

Problems solved by technology

In addition, neural networks usually have problems such as difficult to adjust the learning rate and difficult to determine the initial weight threshold.

Images