ETD Collection

Permanent URI for this collectionhttps://wiredspace.wits.ac.za/handle/10539/104


Please note: Digitised content is made available at the best possible quality range, taking into consideration file size and the condition of the original item. These restrictions may sometimes affect the quality of the final published item. For queries regarding content of ETD collection please contact IR specialists by email : IR specialists or Tel : 011 717 4652 / 1954

Follow the link below for important information about Electronic Theses and Dissertations (ETD)

Library Guide about ETD

Browse

Filters

2018 - 201912020 - 20211

Settings

Author: search.filters.author.Dheda, Dhruti

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Data-driven sensitivity mitigation techniques for genetic algorithm - long short term memory water quality prediction model
    (2021) Dheda, Dhruti
    A long short-term memory (LSTM) model developed for the prediction of water quality, based on the historical data of a particular water body, and as such a particular water quality dataset, will only be applicable to that dataset. Thus if a specific LSTM prediction model is applied to another dataset, then it is quite possible that the prediction model will fail to make an accurate prediction. These models tend to be case study specific. This research focuses on improving the tolerance (mitigating the discrepancies in model prediction capability that arise from differences in datasets) of LSTM prediction models. The two different LSTM models developed from two different water quality datasets, the Burnett and Baffle models, are optimised using the metaheuristic genetic algorithm (GA). The two hybrid GA-optimised LSTM base models, the GA-Burnett and GA-Baffle models, are fused together using a weight-based approach to forma final robust and tolerant predictive ensemble model. Both the models contribute equally to the average ensemble model. In the weighted ensemble model, the GA-Burnett model only has a 10% greater contribution than the GA-Baffle model. Generally, the ensemble models outperform the GA-optimised hybrid LSTM models. The four models are tested on unseen and unrelated datasets and the performance of all the models are consistently similar to one another on each dataset. The consistency of performance exhibited by the different models on any particular dataset is evidence of the successful mitigation of the discrepancies of the individual LSTM models through the implementation of the linear weight based fusion of two hybrid GA-optimised LSTM models. The models are not only applicable for the prediction of water quality, but also for domains outside of the water sector; thus asserting the relevance of the models, especially the weighted ensemble model in the wider field of LSTM and ensemble prediction. This research involves the water quality of rivers. Water is a critical natural resource that is currently under threat, especially rivers. The models are able to successfully predict the quality of river water ahead of time, in terms of dissolved oxygen concentration. Water quality prediction aids in increasing the efficiency of water quality monitoring. Efficient water quality monitoring enables effective water management. Effective water management is necessary for the preservation of rivers
  • No Thumbnail Available
    Item
    The quantification of water usage in a South African Platinum refinery using various water accounting methods
    (2018) Dheda, Dhruti
    South Africa is the darling of the platinum world with majority of the global platinum reserves being located in its backyard. Despite boasting extensive platinum mining activity, South Africa contrastingly has limited water resources. Additional pressure is placed on existing water resources due to climate change, poor water infrastructure and greater urbanisation. Hence water management in the mining sector, particularly the platinum mining sector is of great significance. Platinum precious metal refineries are often neglected in terms of water related studies as they are comparatively smaller than other components involved in platinum production, such as platinum mines, hence the significance of this study as a means to increase awareness about platinum PMRs. Accurate accounting of water usage in mining operations is necessary if water is to be effectively managed and minimised. Two water accounting methods were employed to evaluate water usage in a South African platinum precious metal refinery, namely the Water Accounting Framework and Water Footprint Network method. Flowrates and rainfall data were provided by the refinery, whilst evaporation data was obtained from the South African Department of Water and Sanitation. This information along with the appropriate assumptions was used to generate a comprehensive water account for the refinery. The Water Accounting Framework found the volume of the total water inputs into the refinery to be 48.51 ML/year and the total volume of water outputs from the refinery is about 0.99 of the volume of the total inputs. The Water Footprint Network method found the total water footprint to be 49086.07 m3 /year or 49.09 ML /year, comparable to the results of the Water Accounting Framework. The total water footprint was equivalent to the blue water footprint. The total product water footprint of the refinery being valued at 1.20 m3/kg PGM was found to be greater than that of base metal refineries. After viable recommendations were taken into consideration the total product water footprint was reduced by 25%.