Process Control and Computer Systems

Permanent URI for this community

https://hdl.handle.net/10352/182

Browse

Browsing Process Control and Computer Systems by Author "Ejike, Christian Nonyelum"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Design and implementation of a real-time smart sensor network-based system for monitoring potable water quality
    (Vaal University of Technology, 2022-07-29) Ejike, Christian Nonyelum; Mathaba, T. N. D., Dr.; Ochieng, G. M., Prof.; Ohanga, M. O., Prof.
    Water is a necessity for living and having clean water is a priority. Lately, the water problem in South Africa has become a concerning subject matter and the risk it poses to livelihood of the people and agriculture cannot be overlooked. The Vaal region has recorded a considerable number of water contamination cases in the Vaal River coupled with a near-drought incident of the Vaal Dam. Water-supply centres are trying their best to uphold the quality of water delivery in the region, however, most of the water treatment facilities are noted to be in a perilous state. Moreover, the traditional methods of water quality assessment used within the region have been deemed untrustworthy and unreliable, given that they are slow, costly and do not necessarily rely on real-time data analysis. These challenges demonstrate that there is a need to rethink the overall water quality monitoring system approach and enhance the assessment methods by taking advantage of the evolution of wireless sensor networks (WSNs) and the Internet of things (IoT), which have led to many applications for environmental monitoring studies. This study demonstrates the viable application of LoRa and LoRaWAN for in-situ water monitoring. A smart autonomous system is proposed that explores the possibility of using IoT technology and the LoRa to monitor, in real-time, the drinking water supplied to communities, assess and log the data remotely and keep authorities up to date about the status of the water supply. The experiment was carried out at the Vanderbijlpark campus of the Vaal University of Technology in Gauteng, South Africa. The transmitting and receiving nodes were modelled after the TTGO LoRa32 board's features. The sensors used are the DS18B20 which monitors temperature, the SEN0161 used for pH and the SEN0189 for turbidity sensing. The tests ran for a duration of three months. The tests demonstrate how the structure of the terrain may impact the covering range, while models that account for building diffraction pinpoint higher attenuation. In comparison to lab simulations from other studies which considers possible path loss effect on LoRa communication in real-life environments, as seen with conventional wireless communication models in both indoor and outdoor settings, four test locations were selected. Multiple measurements of SNR, RSSI are observed as spread factors (SF7-12) are varied for different locations. The result show that the minimum SNR and RSSI value for good transmission were -9.4 dBm and -114 dBm respectively given the same experimental conditions. This can be improved by adding more gateways between paths as recommended as part of future improvement work.