This study evaluates designing and simulating wireless sensor network as a solution to enhance the operations of natural gas plant in a metering station in Niger Delta region. There are several problems in natural gas plants ranging from hydrate formation, condensate carryover to customer's facility, low equipment availability and so on. All the problems enumerated were due to inability of some units of the plant not linked online nor on Realtime basis. To solve these problems requires the use of transmitter to route process parameters in the respective plant units to the control room and display such parameters for online monitoring. As such, there are two basic options. Either the use of wired transmitters or wireless transmitters. Both technologies are great but considering the prevailing circumstances of an existing old gas plant in Niger Delta region, the cost of excavation to lay fresh signal cable will be relatively high. Hence, the choice of wireless transmitters was chosen. The methodology involved using literature review and case study of existing metering station gas plant in Niger Delta. The literature reviewed involved wired sensors network, wireless sensor network, and application of wireless sensor network. There was also comparison between wired fieldbuses, wireless network technologies and wired versus wireless fieldbuses. Matlab software was used to design wireless sensor network. The Matlab code was written such that it can read data from Microsoft Excel sheet into the program for analysis and simulation. K-means algorithm was used to cluster the sensor nodes data and determine the centroids of the clusters. The program when run will create sink node matrix, concatenate the cluster centroid, and sink matrices. The results of the software simulation and discussion show that the proposed wireless sensor network if implemented will solve the identified problems and therefore, enhanced the operations of the gas plant in the metering station.

You can access this article if you purchase or spend a download.