Ensuring long-term energy savings in South African mines by improving energy data

Abstract

South African mines experience higher-than-inflation electricity tariff increases, additional expenses such as carbon taxes and mounting environmental obligations. The mining sector is, therefore, forced to maintain long-term energy savings. Over the past 20 years, many mine savings initiatives have been implemented. Unfortunately, studies have shown that most savings deteriorate over time due to limited monitoring and evaluation. Energy data on South African mines is usually of low-quality and not reliable for decision-making purposes. With the advent of more affordable measurement instrumentation, the amount of data is also increasing. Users are thus often overwhelmed to the point of ignoring data. To gain information from large amounts of data, the data must be of acceptable quality and be processed for presentation in an easily understandable manner. A method was developed to address the uncertain quality and reliability of mine data. Mine incomers are frequently calibrated and can be verified monthly through Eskom accounts. The data for sub-systems is, however, less reliable. Each sub-system total is a combination of several smaller power meters, resulting in a compounding error. Due to the large number of meters on a mine, calibration or the replacement of inaccurate meters is not financially feasible. To mitigate this, the total power consumption of the sub-systems is compared to the verifiable consumption measured by the incomer meter. Through normalisation, the difference is then spread over the sub-systems by weight. A validation is also performed on each system to confirm that the resulting values are within the expected performance parameters for the system. A method was developed to present data effectively for decision making. This involves a process of visualisation and evaluation in order to produce high quality figures. A selection method was developed to identify the best-suited means of visualisation. This includes the identification of relevant data and the preparation of that data for visualisation. The data is checked for outliers and common data quality issues before being visualised. A visualisation evaluation criterion was developed in parallel with this. The visualisation produced by this method will enable a decision-maker to evaluate the performance of energy intensive mine equipment. This method was implemented on four mines to demonstrate the value thereof. The quality improvement method allowed more accurate energy budgets, saving one mining group R55 million per year. The developed visualisation method enabled the user to evaluate project performance in order to keep energy savings projects on track. With this method implemented, mines could mitigate inefficiencies and achieve energy savings. This study was successful in developing a method to improve both data quality and presentation. This in turn can be used to achieve and maintain long-term energy savings on mines.