Machine Learning for Energy Generation

Original article can be found here (source): Artificial Intelligence on Medium

Machine Learning for Energy Generation

The 4 use cases of ML for electric power plants

Source: author

If you are managing wind parks or any other power plant, then you are well aware: Maintenance, human errors, downtime and planning inefficiencies can cost millions per year.

Where simpler statistics has made little progress, machine learning is proving to be a very effective tool to:

  • Predict malfunctions sooner and more accurately
  • Detect human errors, before they become a big problem
  • Optimize power plant schedules, to increase your profitability
  • Predicting the merit order, to optimize the scheduling of different power sources

Predict Turbine Malfunction

Wind is a great renewable energy source, but wind turbine maintenance is notoriously expensive. It accounts for up to 25% of the cost per kWh. And fixing problems after they occur can be even more expensive.

Machine learning can help you get ahead of this problem, reducing maintenance costs by catching problems before the turbine malfunctions. This is particularly important when wind farms are located in hard-to-access places, such as the middle of the ocean, which makes repair costs even higher.

Real-time data gathered with Supervisory Control and Data Acquisition (SCADA) can help identify possible malfunctions in the system far enough in advance to prevent failure.

For example, we can use data from sensors found within the turbines — such as oil, grease, and vibration sensors — to train machine learning models to identify precursors to failure, such as low levels of lubricant. This method can predict failures up to 60 days in advance.

Reduce the Potential for Human Error

Source: author

Each year, human error accounts for as much as 25% of power plant failures. Along with the loss of up to 30 million megawatt-hours of energy generation annually, this causes service interruptions for customers, or worse — just think of Chernobyl and Three Mile Island. It also means unnecessary costs associated with fixing the error and getting the system back online.

To combat this, we can use machine learning to support decisions made by control room operators.

Machine learning provides constant systems monitoring that helps you detect anomalies. We also automatically suggest an action plan to prevent the situation from getting worse. It can even deal with a problem before human intervention becomes necessary.

This reduces the risk of human error due to distraction, lack of knowledge, or reaction speed — sometimes control room operators simply can’t move fast enough to stop the problem.

Increase Power Plant Profitability with Optimized Scheduling and Pricing

The sheer volume of data we have at our disposal these days is helping people make better decisions about how they operate businesses. Utilities are no different. The volatile nature of energy prices means that running a generation plant can be more or less profitable depending on something as simple as the time of day.

But because the utilities market is so fast-paced, it can be hard to manually track all the data required to make these decisions.

We can use machine learning to help. By feeding historical data on prices and usage into a machine learning algorithm, you can predict the best times to run their plant — and make more money. Machine learning can find times when usage is high but prices for the raw materials used to produce energy are low. These extremely accurate predictions creates an optimized generation schedule that maximizes profitability.

Predict Merit Order of Energy Prices

Source: author

Utility providers have a lot of options when it comes to sourcing where they get their energy — everything from renewables like wind and solar to fossil fuel and nuclear. When it comes time to sell power, these different sources are organized into a merit order based on price. This determines the order in which power from these various sources is sold.

Because we have access to data from a wide variety of sources, we can use machine learning to analyze both real-time data and historical data. Machine learning algorithms are also better at taking into account all the different factors that influence the price -things like weather, demand, how much energy is available from the various sources, historical usage, etc.- to create an optimized merit order.

This helps you make more informed decisions about where you’re getting your power. This is especially helpful in markets where there is a lot of renewable energy, such as wind, because it’s hard to guarantee energy availability from these sources.