ASK A QUESTION
ASK A QUESTION
×
All news
искусственный интеллект

Using neural networks, anomaly detection algorithms, AI

SmartEAM ,
11.03.2020

Artificial intelligence is the property of computer systems to perform creative, intellectual tasks.

штучный интеллект

Machine learning is a set of algorithms and statistical models with which computers can solve the problem without using detailed instructions, using templates and logical conclusions.

Anomaly detection – identification of rare events of elements, values that cause suspicion due to differences from most data.

Machine learning. Local outlier factor – detection of anomalies in the operation of the equipment

Attention!!! Anomaly!

For complex equipment (turbines, furnaces in metallurgy) it is often necessary
control tens, hundreds and thousands of different indicators.
The method of “local emission level” can warn us about violations during the operation of equipment in the early stages.

Machine learning. LSTM Neural Networks – detection of anomalies in the parameters of operational statistics of equipment.

Attention!!! Anomaly!

Methods for detecting anomalies without human intervention can determine deviations from the normal course of the production process.
For signals with a complex structure (vibration, pressure in the valves), pattern analysis is necessary. To analyze such data, LSTM neural networks are used.

нейронные сети

Kyiv


+38097 97 05 113
+38095 69 49 957

Ave Bazhana 14a, 4th floor

Minsk


+375 29 312 29 93
+375 29 622 24 69

Ltd "Softeq Lab"

© SmartEAM 2019.