Repozytorium publikacji - Politechnika Gdańska

Machine foundations is a critical topic in the gas and oil industry, which design and exploitation require extensive technical knowledge. Machine foundations are the constructions which are intended for mounting on it a specific type of machine. The foundation has to transfer dynamic and static load from machine to the ground. The primary difference between machine foundations and building foundations is that the machine foundations are a separate structure, even if they are inside the building. Failures of machine foundations can be very dangerous due to its carry loads from machines in operation. There is also an economic aspect because every break in the operation of industrial machines is expensive, especially in the gas and oil industry, where technological processes are complex and multi-stage. Repairs to concrete machine foundations are problematic, so the capability to predict what exactly affects failures seems extremely necessary. The failure of concrete machine foundations depends on many factors that are not fully understood. Modern achievements of science and technology, especially machine learning techniques may allow determining what affects the failure rate. This paper presents an analysis with the use of machine-learning techniques to predict in which way loads can affect the failure of foundations. This study examines whether and what relations exist between variables describing loads about the machine concrete failures occurrence. The analysis concerned some variables such as cross-section reinforcement amount, the grate load, measured concrete strength, motor short circuit moment load, the engine unit and rotor with shaft load, the pump unit and rotor with shaft load, the weight of the foundation, total load with foundation self-weight. The primary parameter of concern is the failure occurrence rate.

Autorzy

• dr inż. Patryk Ziółkowski link otwiera się w nowej karcie ,
• dr hab. inż. Maciej Niedostatkiewicz link otwiera się w nowej karcie ,
• mgr inż. Sebastian Demczyński