The US grid network is over-stressed. Changes in climate increasing severe weather events, the aging of the system, and the need to incorporate innovative technologies that it was not designed for, such as DERs, means that there are many challenges to be faced.
The frequency and duration of power outages in the United States is increasing. Between 2011 and 2021, for example, there were 64% more power outages which were designated as major, meaning the event impacted more than 50,000 customers, than in the previous decade. Several power supply events recently, such as the Texas winter storms in 2021, have outlined the need for improved resilience.
Grid reliability will only become more important in the future. This is because society is increasingly relying on electricity to power everything from heating and cooling in buildings to industrial manufacturing processes as well as electric transportation. Many countries are attempting to phase out fossil fuels in favor of electric power.
This means that the grid will need to be upgraded, but there are also steps that can be taken that are cost effective and can be implemented in a short time-frame. Increasing the ability of utilities to detect and repair faults quickly would improve reliability.
Â
Limitations of Current Fault Location Methods
Fault passage indicators (FPIs) — FPIs are one method for providing an estimate of a fault location. They can provide an approximation of where the fault occurred, but not the exact point which needs repair, so repair teams will still have to search around to pinpoint the exact fault.
Impedance Models — Using impedance models is another common methodology for fault location. This method compares fault data recorded by protective relays, meters or other monitoring devices to an electrical system impedance model, and the result is an estimation of the electrical distance to a fault along the electrical lines. As the grid, is large and complex this method typically provides the utility with a short list of potential locations rather than an exact location of the fault.
Predictive maintenance — Detecting and addressing emerging the incipient and emerging faults that lead to outages is not always possible; some outages are caused when well-functioning grid infrastructure is damaged by wind, floods, falling vegetation or intruding wildlife. Surveillance by helicopters and drones is good but not the complete solution, even when combined with AI processing of the images.
All these systems need to be integrated and accessible from all locations, which means cloud-based data storage tends to be the most practical solution.
Â
Better Sensors, AI and Predictive Technology are the Future
What improved sensor systems should provide:
- Predictive maintenance by identifying weak spots
- Tracking of real- time power flow monitoring and power quality
- Grid insights into advanced distribution management systems (ADMS)
- Faster searches of permanent faults, reducing outage time (SAIDI)
- Data analysis of grid performance with increased power grid insights
By inserting these advanced sensors around various areas of the grid, including over- and under-ground lines, substations, and power plants, then uploading the data into the cloud so it is easily-accessible from anywhere, AI and machine learning systems can operate on the data. This should mean that resiliency and reliability are improved, and outages can be kept to manageable levels. This will also assist in the transition to a Net Zero emissions grid, and also ensure that new demands on the network, for example many more EVs connecting to it and microgrids both producing and consuming power will be serviced reliably.