All over the globe, gas turbines (GTs) play tremendous role in energy and power generation. Condition monitoring is also being used to obtain early warning of impending equipment failure to prevent costly downtime and damage to process equipment. Several scheduled visits were thus made to AFAM IV, GT 18, TYPE 13D plant located near Port Harcourt, in Rivers State of Nigeria. Continuous and periodic monitoring of the thermodynamics/performance parameters such as temperature, pressure, air pumping capability and fuel flow were carried out. These activities lasted for over a period of one year on hourly basis to examine the state of health of the engine compared with the data taken. The diagnostic method of trend performance monitoring was jointly used with multiple variable mathematical models (MVMMs), because they relate deterioration to consequences. A software code-named “THAPCOM” written in C++ programming language was used proactively monitor the engine based on this MVMMs. The values observed on the third month revealed that ηO was 27.0% and AL was 48MW. A significant variation in the results obtained shows that there is a deviation between the monitored data taken from the console and the reference data in the manufacturer’s manual. These are indications of impending failure or health uncertainty of the engine. This allowed maintenance to be scheduled, or other actions taken to avoid catastrophy.

Inefficiencies of compressors and turbines resulting from insufficient pressure ratio at successive blade stages in gas turbines often lead to flow reversals. A step to statistically correlate boost pressure and vibration velocity amplitude to avoid such flow reversals therefore led to the execution of this research. A computer simulation technique was used to actualize this purpose with the operational data obtained from Delta IV unit of Ughelli power station GT 18 using VC++ programming language. Results obtained show that maximum vibration manifests on each of the bearings at a pressure ratio of 9.47 for all cases considered. These results further show a linear relationship between the data using statistical z-test.

The ability to identify faulty equipment under severe conditions allows the manufacturers and operators to take appropriate proactive measures to rectify the faults and provide an assessment for early detection of the engine defects that could lead to catastrophic failure. This work therefore shows that such counter measures can be carried out by optimization, modeling, simulation or using a reliable analysis of gas turbine engine, rotor shaft and bearing vibration data through 2-Dimensional D’Alembert’s equation. Gas turbine plants on industrial duty for electricity generation were thus used to actualize the research. The data for vibration amplitude of rotor bearings from the reference engine which varied between 0.19 and 3.81 mm/s were compared with those obtained from a failed engine of similar characteristics just before failure and used for the simulation and modeling. The given engine speed and active load were also determined as falling between 7264 rpm to 7436 rpm and 10 MW to 20 MW respectively. A well packaged computer program code-named “MELBF” written in C++ programming language was developed. The results show that the machine should not be run beyond 3.81 mm/s vibration amplitude in order to avoid resonance and downtime of the engine.