Computerbaseret fjernovervågning af pumpesystemers behov for vedligehold
Oversat titel
Computer-based remote monitoring of pump systems need for maintenance
Forfatter
Arensman, Mareike
Semester
4. semester
Uddannelse
Udgivelsesår
2012
Afleveret
2012-06-07
Antal sider
105
Abstract
Dette projekt undersøger, om helbredet af pumpesystemer kan overvåges ved kontinuerte trykmålinger i pumpestationer. Hypotesen er, at tre tilstande kan identificeres i trykdata: pumpeslid (der reducerer ydelsen), ændret ruhed i trykledningen (som øger modstanden i røret), og luft i systemet (luftlommer). Vi bruger forløbet af trykket ved start og stop af pumperne til at bestemme starttryk og driftstryk, som siger noget om pumpens og ledningens tilstand. Tilstedeværelse af luft vurderes ud fra trykbølgens forplantningshastighed: en lavere hastighed tyder på luft i rørene. Trykket er målt på seks pumpestationer. I én station med to ens pumper havde den ene markant lavere ydelse. Der blev også observeret luft i et af pumpesystemerne. Til at analysere data er der udviklet to modeller: en analysemodel, som udtrækker trykbølgens hastighed samt start- og drifttryk, og en numerisk model, som kan genskabe de observerede trykforløb. Modellerne er ikke umiddelbart egnede til direkte integration i eksisterende overvågningsanlæg, men resultaterne viser, at de tre tilstande kan aflæses i trykmålingerne. Dermed bekræftes hypotesen.
This project investigates whether the health of pump systems can be monitored using continuous pressure measurements in pumping stations. The hypothesis is that three conditions can be identified from pressure data: pump wear (which reduces performance), changes in the roughness of the pressure pipeline (which increase pipe resistance), and air in the system (air pockets). We use the pressure behavior when pumps start and stop to determine the start pressure and operating pressure, which indicate the condition of the pump and the pipeline. The presence of air is assessed from the propagation speed of pressure waves: a lower speed suggests air in the pipes. Pressure was measured at six pumping stations. In one station with two identical pumps, one pump showed a marked reduction in performance. Air was also observed in one of the pump systems. To analyze the data, two models were developed: an analysis model that extracts the pressure-wave speed and the start and operating pressures, and a numerical model that reproduces the observed pressure profiles. The models are not suitable for direct integration into existing monitoring systems, but the results show that the three conditions can be read from pressure measurements. This confirms the hypothesis.
[Dette resumé er genereret ved hjælp af AI]
Emneord