In Part One, we discussed how Data-as-a-Service platforms enable utilities to operate competitively in a sector driven by flexible demand, distributed generation and affordable storage. We also introduced how DaaS solutions support predictive and pre-emptive analytics, which provide utility personnel with the context needed to minimize the impact of power outages.
Now we'll dive deeper into predictive analytics use cases. Could this technology affect the way utilities and consumers buy and sell energy? How will DaaS impact asset performance?
Identifying the True Value of Each Asset
The main advantage to using a Data-as-a-Service platform is that it aggregates datasets located across disparate systems.
For example, suppose a regional manager wants to identify the fiscal impact of substation outages. He or she also needs to estimate the likelihood of specific substations experiencing downtime whether due to inclement whether or asset depreciation. In order to do that, the regional manager needs the following information:
- The amount of power each substation transfers over a 24-hour period.
- Seasonal demand across regional subsections.
- The age of every substation's equipment.
- Which assets technicians have had to repair over the past five years.
- The purpose and nature of each asset service request.
- How much power customers generate on a seasonal basis, if any.
- Historical outage reports.
While a field service management application may contain technician notes and other equipment repair information, a customer relationship management system possesses household electricity consumption and generation rates. This and other data are necessary to understand which customers each substation services.
Through a DaaS platform, the regional manager can pull the requisite data together and run various analyses. For instance, he or she may discover that Substation A's step-down transformer has required four service requests in the past year, and is 11 years old. He or she can then calculate the cost of a failure occurring during certain conditions, and determine whether it's worth it to purchase new equipment.
Applying Predictive Analytics to Utility Asset Tracking
The purpose of applying predictive analytics to utility equipment monitoring systems is to register symptoms indicative of impending asset failures. If a lineman receives an alert that a transformer is getting overheated, he or she could address the problem before harmonic distortion occurs.
When analysts apply preemptive analytics to utility asset tracking data, utilities acquire the foundational capabilities needed to establish preventative maintenance programs. The Institute of Electrical and Electronic Engineers discovered transformer outages are twice as likely to occur when preventative maintenance strategies aren't in place.
It is possible to execute preventative maintenance programs without using predictive analytics. The problem is, this can result in a misallocation of scarce resources which have alternative uses.
For example, suppose a technician has to maintain 10 assets, but only two of those assets are at risk of failing. Instead of dedicating most of his time toward refurbishing those pieces of equipment, he needs to distribute his time to maintain the other assets, which don't require such attention. Pre-emptive analytics reveal high-risk assets, allowing technicians to allocate their time to more valuable uses.
Developing New Business Models
Data-as-a-Service platforms could empower utilities and their customers to participate in open power exchanges. Energy Sage noted that the cost of installing solar panels decreased 9 percent between 2016 and 2017. If this trend continues, it could become more feasible for the average homeowner to install photovoltaic systems.
The challenge of selling energy to utilities is that there is no fixed price for power. Demand fluctuates with the changing seasons, customer behaviors and other such variables. DaaS platforms could rectify this problem by centralizing information from multiple sources. That way, utility analysts could run weekly or daily formulas that calculate how much the power company should bid on electricity from households across a certain region.
What we're really talking about here is a more informed business. An analyst could determine how much power a household will generate, the costs of distributing that energy through available equipment and where demand for that power will be greatest. That's what the utility of the future looks like.