By Jim Foerster, Schneider Electric
Utilities can waste millions of dollars due to inaccurate weather forecasts, and they might not even know those forecasts are steering them wrong.
Inaccurate weather data can make it hard for utilities to accurately predict outages across their service areas, leading to ineffective restoration efforts, frustrated customers and potential health risks to crews or customers. With the right information, however, outage prediction can be highly accurate and localized so utilities can call up the right number of crews and position them correctly, as well as smartly invest in infrastructure to better prevent storm damage.
Topography, geography and climate can differ significantly across a utility’s service area. It is common, therefore, for weather conditions at specific localities to differ significantly from the broader forecast for the region. This makes it difficult for utilities to predict how weather will affect a specific point of their service area on a given day. Certain points within a service area may be battered by high winds and rain while a different point experiences ice and snow. These forecasts can cause utilities to over- or under-prepare for storm damage, as well as overestimate or underestimate the demand load on the grid. Any of those scenarios put a utility’s bottom line at risk.
Commonly available weather forecasts for a specific location are part of generalized forecasts for a larger geographical area. These weather forecasts are gathered by national weather stations commonly found at airports, which are often miles outside the city. These forecasts usually combine the weather each city or town might experience throughout an entire day, without any specificity around the time and location of each weather event.
Historically, weather-dependent businesses have relied on generalized weather forecasts to make critical decisions that don’t necessarily provide accurate outlooks for their specific needs. These businesses can benefit dramatically from local weather station technology.
Hyper-local Weather Data AT CENTRAL HUDSON
Utilities need weather stations to provide the most accurate temperature forecast across their service area, not just one for the whole territory. This allows them to more precisely model and predict where outages will occur.
One example of this is Central Hudson Gas & Electric (Central Hudson), a regulated transmission and distribution utility that delivers natural gas and electricity in a service area that extends from the suburbs of metropolitan New York City north to the Capitol District at Albany. The utility wanted to develop an outage prediction model, but found itself relying on historical weather data from National Weather Service weather stations-most of which were outside of the utility’s service territory. In addition, historic weather data was not available for many areas within the service territory, including some of the most topographically diverse locations.
Because of this, the National Weather Service showed that the service area would experience certain weather conditions that were vastly different from those occurring at specific localities. Consequently, Central Hudson was unable to accurately predict outages, and creating an effective outage prediction model was especially difficult. Because it couldn’t accurately predict outages before they happened, Central Hudson was risking its crews’ safety and frustrating its customers.
To address this, Central Hudson built its own forecasting network with local weather stations from Schneider Electric, which helped it make informed, business-critical decisions. These hyper-local forecasts generated forecasts for specific locations and times. Accurate, localized weather conditions and forecasts across the territory are reported with real-time notifications, allowing better awareness of weather influences on service.
“The addition of 24 weather stations in our service territory has allowed us to measure significant weather data in remote areas on a real-time basis, as well as perform detailed investigations immediately after an event,” said Tim Hayes, T&D operations services and emergency response at Central Hudson. “The stations have provided us more focused information than we have been able to gather through any weather service outlet, thus allowing us to determine the resiliency of our system to specific weather occurrences.”
As Central Hudson continues to develop its outage prediction model using data from its local weather stations, it will help the utility better prepare for outages by assisting in the pre-positioning of crews. Crews are safer as a result of the weather stations, because they have a better sense of the weather conditions with which they are dealing when solving outages. In the future, Central Hudson hopes to integrate this weather data into its outage history to develop an outage management forecasting system.
Hyper-local weather data aids energy management
In addition to outage management, localized weather forecasting also plays an important role in effective energy management.
In the U.S., residential and commercial buildings account for 39 percent of energy demand. Up to one third of that usage derives from heating and cooling alone. This number is highly dependent on weather, and could be lowered with accurate, precise forecasts.
Sudden, sometimes even miniscule changes in the temperature can significantly impact the demand on the grid. In fact, up to 90 percent of errors in load forecasts are the result of a poor weather forecast. Because utilities buy and sell energy days before they actually need it, the price of purchasing energy the day it’s needed comes with a hefty penalty, not to mention the risk of outages or damaged equipment. Large power utilities can save more than half a million dollars a day when the temperature forecast accuracy is improved by a mere half a degree.
Similar to local factors significantly impacting storm profiles, these factors can also create diversions from the broader temperature forecasts, as well as change renewable energy generation profiles.
Hyper-local forecasts help utilities plan a more effective distribution strategy, so they can provide the right amount of power to the right place at the right time. For example, if one location in a service area typically experiences temperatures several degrees lower than the general forecast, utilities can plan for less power delivery. Using hyper-local forecasts will allow a utility to adjust its power delivery based on accurate forecasts without adding to the grid.
In the case of renewable energy, hyper-local forecasts allow utilities to predict how much energy will be created by renewable sources, such as solar or wind. This is crucial to ensuring that the load and supply of power stays balanced.
Solar is the most abundant renewable energy source available, but the ability to harness and distribute it alongside existing sources of energy is new territory for many utilities. It isn’t easy for grid operators-who need to balance electricity supply and demand at all times-to control solar energy because not only does weather affect the amount of power that is generated but also when that generation will be available on the grid. With solar energy, there is no “on” or “off” button, and grid operators must accept the solar energy generated. With proper weather information, solar power can be more predictable and therefore less disruptive to the grid.
Hyper-local forecasts will help weather-dependent businesses and the U.S. reduce money spent on electricity generation and ultimately decrease overall energy consumption.
Hyper-local weather supports building management
Because weather varies all the time, building management systems (BMS) can use weather forecasts to predict the energy needs required to keep buildings at a roughly consistent temperature. The colder the outside air temperature, the more energy it takes to heat a building, and cooling vice versa.
Sophisticated BMS use forecasted temperature data and other factors, such as human occupants and equipment that give off their own heat, to predict the building’s energy needs. BMS then can make informed decisions to construct a strategy to meet those needs-how much energy needs to be created or purchased to meet desired temperatures, using the cheapest energy available.
An example of this is a hospital that chills water or makes ice at night when energy is cheap, rather than during the day when energy is most expensive. The hospital then uses the cold water or ice to keep the buildings cool during the next day. Accurate forecasts will show a BMS how much cold water or ice to create, then enable the BMS to create it when energy is cheapest. If the hospital runs out of chilled water or ice during the heat of the day, the BMS must consume energy to create more ice or chilled water at a time when electricity is most expensive.
Accurate weather data can positively impact the efficiency and performance of BMS by more than 25 percent. By adding a weather station to the top of a building or parking lot, or attaching it to an object, such as an electrical pole, BMS can get a more sophisticated, accurate view of the weather affecting the building and further improve building efficiency.
Weather stations enable businesses to make smarter decisions when it comes to energy consumption and generation using hyper-local forecasts. Utilities that use hyper-local forecasts won’t have to purchase extra generation the day it’s needed because their temperature data will be specific to their service area, leaving little room for error. Buildings won’t face a guessing game of temperatures, because weather stations will enable BMS to understand future weather patterns. Hyper-local weather forecasting is the right choice for utility, operations and building managers, because such precise forecasts will reduce expenses, increase efficiency and improve safety.
Jim Foerster is the director of the product management team for the weather division of Schneider Electric. He has responsibility for all of the weather software products offered to the market, as well as the meteorological content in those products. Jim is one of four Certified Consulting Meteorologists with Schneider Electric, one of the most prestigious peer-awarded certifications available in the weather industry. He has a bachelor’s degree in meteorology from the University of Wisconsin, Madison. In his spare time, Jim is a professional soccer coach in Bloomington, Minnesota, where he resides.