Predictive analytics and artificial intelligence in the wind energy industry

Is it true that predictive analytics may increase O&M costs?


For some wind energy experts, it may be early to openly talk about predictive analytics and artificial intelligence in the wind energy industry, the truth is companies have been providing digital solutions for wind asset management for the last 10 years, the matter is that companies are buying more sophisticated software solutions without taking 100% advantage of it.

To get familiar with predictive analytics it’s important to understand how it works and why it is so trendy in the management of renewables. It combines big data with data science to get hidden insights from a windfarm's operation. Every wind turbine type contains tons of data, difficult to extract and analyze, but with artificial intelligence and machine learning algorithms help, a detailed scanning and diagnostic can be done to unveil potential failures or problems that a wind turbine may present in the short or long term. Acknowledging in advance when things may go wrong gives managers the power to better control the assets lifecycle.

Some energy experts may think predictive analytics will increase O&M (Operations and Maintenance) costs due to more frequent interventions in WTG’s, nevertheless, if we take a look into the evolution of O&M costs, these are already changing due to two main trends. First, global onshore O&M costs increased by 20.9% between 2010 and 2020, due to higher-than-expected costs for servicing older assets and more robust service scopes (Richard C, 2020). Secondly, as shown in Figure 1. younger and more technological wind turbines have demonstrated to reduce maintenance costs significantly vs turbines manufactured in the past century (MilborrowD, 2020). So how is predictiveness involved in these situations?

Figure 1. O&M costs evolution between old and young WTGs.

Wind Technologies Market report

In both scenarios, though older assets have older technologies making difficult the scanning and processing of data, predictive analytics can early detect outages allowing the operations team to take precautions avoiding highly expensive parts to be replaced. On the other hand, this derives into revenue maximization, improving asset management KPI’s. At the end of the day, technology is making huge progress to be as accurate as possible in its predictions to reduce O&M costs.

Renewable energy providers have found in SCADA software a solution to ease assets analysis but deep diving into these tools to make the most of wind energy is necessary. Asset managers are still relying on what the O&M team suggests without paying much attention to all possible analytics provided by digital solutions. It is necessary to find balance by trusting artificial intelligence plus the O&M human touch. This is why at Proxima Solutions we have been developing a suite of modules that will accompany asset managers into more accurate decision making. Wind-Log is a digital tool that captures portfolio and individual asset information, through 10 different modules decision-makers can deep dive into assets health, revenues, energy production and losses, reporting, budget, outages, and so on.

It is time for wind farms' decision-makers to start trusting artificial intelligence and technologies that other industries have been taking advantage of for years, not only to decrease O&M costs, but to get insights that will improve the assets lifecycle and management.

References: