Energy Efficient Mines - Regulating energy consumption with data insights Business Landscape
Mining is an energy intensive industry with large amounts of energy required to extract the ore, process and ship it to the end customer. Artificial Intelligence, Data Science and Machine Learning can provide additional insights into each of these processes to help mining companies be more energy efficient in their operations.
Decisions made by the mining companies usually have an energy impact whether it be fuel consumption, explosives use, electricity consumption, or energy sourcing. Mining processes are complex, and often decisions are made with incomplete information such as uncertainty in the ore grade, new workforce operating old equipment, people with different levels of training etc.
How energy is used across the value chain may not be consistent and leads to energy inefficiency. Some causes of energy variance are:
- People factors - not everyone will make the same decision about how to operate equipment efficiently. Training and skill levels could be different between operators.
- Process Optimization - The process parameters change to adapt to the changing conditions of the process. Temperature and humidity changes may impact on the efficiency of the process.
- Material - The feed material may require different processing conditions. Ore characteristics such as hardness or contaminants may require different processing.
- Maintenance - Poorly maintained equipment would often have higher energy use.
- Age of Equipment - As equipment age, their energy use may change, or more efficient equipment may be available.
These points highlight some of the issues where the energy consequences of decisions are not always available or apparent to those making them.
Extracting insights from Data
The increase in available data about the equipment, process and the systems provide data which often has energy insights hidden within it. Some of the Data Science and Machine Learning toolsets and approaches will provide insights into the data to extract the impact on energy and the costs of what could happen.
Some examples of using data science to provide value include:
- Energy Prediction - When the planning team can predict their energy requirements and make informed decisions. These include decisions to defer the use of energy to a time when the cost is low or selecting the optimum energy source.
- Process State Classification - Using the data around the process will allow for the process to be better understood. Are there particular conditions, process parameters or modes of operation that cause inefficient energy use?
- People Analysis - Do different teams or operators have different energy profiles?
- Equipment Analysis - Higher Energy use may be an indication that equipment is requiring maintenance.
These insights and approaches show how data science tools can uncover insights from numerous data sources captured by the equipment and process. When process and energy data is studied, and analysis is undertaken, insights into the energy use is a key outcome that benefits the business.
However, getting the right data, at the right time at the right resolution for energy efficiency projects can be a challenge. An assessment of the data quality, completeness and availability of energy data is critical to ensure the efficacy of data science & its tools.