Ants may teach us sustainable mining
This article is the first of a series of short articles specifically for professional engineers and mining executives who are asked to reduce greenhouse gas emissions but may be of interest to others. The point I wish to make, that we do not need to change the feasibility study process to adopt renewable energy. However, we will have to change operations to achieve success, and one useful method mimics animal behavior. We must challenge ourselves to learn these ways and then design better technologies that are both of optimal economic value and environmentally sustainable.
The article describes opportunities offered by emerging technologies like swarm intelligence for reducing emissions. Solutions that still will deliver optimal economic value for shareholders. In other words, I believe that as a professional engineer you can deliver these new requirements without compromising. You can deliver both the shareholders environmental governance and financial requirements. You can do so by retaining the design philosophy engineers have always used. However, you must include “without emissions” in the scope of your feasibility study, it is now a nonnegotiable functional requirement. You then should investigate innovation offered by behavior of animals who have learned how to operate in the most energy efficient way. Doing so favors survival.
Surprisingly, studying ants will help you. Artificial intelligence offers ant optimization as an algorithm to control an energy efficient process. An ant colony optimization algorithm will choose the ore of correct hardness from the stockpile. The right one to use which expends the least energy and over the operating cycle optimizes energy use depending on the irradiance of the sun. The algorithm decides how to use the available energy effectively.
A growing number of artificial intelligence algorithms use swarm intelligence, they are inspired by nature. Mimic how animals learn by a trial-and-error process that is repeated over and over. The longer the trial-and-error learning is repeated the more likely it is to produce the most efficient way to operate. You could say ants learn success from repeated failure.
The most surprising fact about the evolutionary history of ants is that they have been on Earth for a much longer period than humans. Evidence suggests that ants originated between 140 and 160 million years ago. This means ants existed during the Jurassic period, the same age when dinosaurs roamed the Earth. That is a long time to learn the best way to live.
In their long history, ants have learned efficient ways. They have learned the best way to acquire the material they need to survive. For example, when Leaf Cutter Ants forage for leaves they cut them and then transport them to their processing plant. Ants secrete an evaporating pheromone that marks the path that tells their fellow worker ants the shortest distance to the best leaves. Other workers deep in tunnels, which they dug with fit for purpose highly evolved mandibles, process food with the aid of a special fungus. They process food from the leaves they cut to sustain the colony. The queen ant is fed to lay more eggs to keep the worker ant population thriving, workers serve the queen. Does this sound a bit like mining? Like finding and processing metal ore to sustain a growing human society.
When a mining company produces metal, the ore is crushed at a mine site using a mill that consumes approximately twenty megawatts of power. However, energy consumption is variable, it depends on how hard the ore is. If the site is remote, and many are, power for the crushing of ore is generated by burning diesel and that is no longer acceptable. Metal production is causing unwanted emissions that you are now being tasked with eliminating.
Imagine you are in the engineering team that is asked to design a crushing circuit that both reduces emissions, to the lowest level possible, and then reduces the ore to transportable dimension cheaper than the former diesel-powered circuit. You first simulate in your model a hybrid power system to reduce the consumption of diesel, fill your model with attributes of arrays of photovoltaic panels, inverters, and lithium battery power walls. You retain the diesel generator sets as back up for mill power on low sun days and early mornings when batteries flatten. However, you find the capital cost of the new circuit is too high and even with the saving in operating costs from burning less diesel you can’t justify it. The capital is not returned before, the end of mine — life. You report the situation to your management, still under pressure from shareholders to reduce emissions. What do you do?
Engineers are not asked to design uneconomic systems. However, today the public person may say otherwise. Imagine after work you are at a coffee shop discussing climate policy with a friend, he might say by following the politics you are not doing your job, the rules of good engineering are being broken. He might say that common sense is not being followed by engineers because of political reasons and when designing and commissioning renewable energy powered systems the friend might say the engineers are looking for government handouts. However, you respond and tell him such behavior is not sustainable. It is not the way to engineer, that no engineer will receive a leave pass from what is the foundation of engineering science for long, and repeated uneconomic design is not sustainable. Engineers cannot afford to change practices because of political considerations, nor should they, you say, lower standards and rush the energy transition. That will end badly and ruin the economy. But doing nothing is not acceptable. Inventing a new economic evaluation process is not an option. One must instead conduct a reevaluation of the physical process and improve the design to comply with the functional requirements. If the design powered by renewable energy is too expensive one must not accept that and ask for government subsidies but instead innovate. Innovate until a less expensive solution is found.