- Webinars
Consumer-Centric Electricity System Planning
While policymakers may drive the pace of decarbonization overall, the depth, timing, and location of consumer adoption of Behind the Meter (BTM) resources, i.e. rooftop solar PV systems, battery storage, electric vehicles, and appliances, are the key drivers of the future energy system in Australia.
While policymakers may drive the pace of decarbonization overall, the depth, timing, and location of consumer adoption of Behind the Meter (BTM) resources, i.e. rooftop solar PV systems, battery storage, electric vehicles, and appliances, are the key drivers of the future energy system in Australia.
While policymakers may drive the pace of decarbonization overall, the depth, timing, and location of consumer adoption of Behind the Meter (BTM) resources, i.e. rooftop solar PV systems, battery storage, electric vehicles, and appliances, are the key drivers of the future energy system in Australia.
Figure 3 illustrates the same customer following grid integration to maximize their load factor, i.e. to maximize the utilization of the grid capacity and minimize the amount of grid capacity. While this is an extreme example, it is technical feasible, and with the right rules and regulations, will be economic.
Once consumers transition to full electrification of appliances and transportation and have installed rooftop PV (including over-parking and other structures) and storage to minimize their energy costs, their grid needs will be completely different:
- Wheeling services to buy/sell from neighbors, including coordination and system security services.
- Less large scale, centralized resources, as higher cost due to amount if infrastructure required, and the economies of manufacturing scale afforded to consumer resources.
- Less higher voltage conductors are most energy is traded close to its source to minimize costs.
While this may take 20 years or more to realize due to the rate of turnover of appliances and vehicle stock and declines in the cost of battery storage, grid investments such as conductors and transformers typically last over 50 years.
Therefore, to avoid the risk of significant stranding of today’s grid investments, the electricity industry needs to anticipate future consumer needs today with as much accuracy and precision as possible.
The State of the Art in Consumer Behavior and Load Impact Forecasting
The electricity forecasting community has been slow to change in the face of significant change in the things they are forecasting. Below shows the Australian Energy Market Operator’s forecast of peak demand over the 2011 and 2012 timeframe, when consumers starting using more solar PV.
Listen or click through at your own pace
The industry’s track record in forecasting electric vehicle adoption has not been much better, as shown in Figure 5 below.
A key reason for the forecasting errors seen above is due, in Energeia’s view, to the use of forecasting methods that were no longer fit for purpose. Regression based methods, which largely project the future using the past, were the tools used in the above examples. As new technologies emerged, and new behaviors emerged as a result, the past was a poor indicator of the future.
Luckily, the state of the art in consumer behavior forecasting has come a long way, and it is possible to not only estimate what the future state looks like down to the LV transformer level, but also how key strategies can optimise it over time.
Figure 6 below compares a range of forecasting methods, with the older but easier to use methods like regression, etc. shown at the top. These methods are less robust, and therefore less well suited to forecast an evolving and emergent factor like consumer behavior.
Figure 6 below compares a range of forecasting methods, with the older but easier to use methods like regression, etc. shown at the top. These methods are less robust, and therefore less well suited to forecast an evolving and emergent factor like consumer behavior.
The more complex, less easy to use methods can all be seen to be more robust to typical forecasting risks, but come at the cost of more effort, more complexity and computational complexity. Of these, Agent Based Modeling (ABM) stands out for its relative ease.
Agent Based Simulation 101
ABM is a category of computational models that invoke dynamic action, reaction, and intercommunication protocols amongst the agents in their shared environment to derive insights about their behaviour and emergent properties, as illustrated by the diagram in Figure 7.
ABM has grown in popularity with leading edge forecasting practitioners due to it s ability to address key complexities seen in modeling and forecasting consumer behavior:
- Range of choices, which become complex and cumbersome to manage individually
- Ability to integrate with other methods, e.g. regression or probit, for uptake parameterization
- Path dependency, future behavior is a function of historical behavior, e.g. PV adoption
- Factor and consumer inter-dependency, consumers can sell to other consumers
- Traceability / auditability of outcomes from inputs and functions
ABM is originally based on the work of Jon Von Newman, a Hungarian polymath, who first invented the method. It’s first application by someone other than Von Neumann was in the social sciences in the 1970s, which used it to model the behavior of migrating populations based on their level of tolerance. It expanded into other areas from the 1990s, and its first identified application in the electricity industry was in 1999. From the 2010s, it has become more widely applied in the electricity industry.
With a more robust and accurate consumer behavior forecasting method, which can not only address decisions around new technology adoption, but also address participation in electricity industry programs, and therefore the development and operation of distributed resources, it is now possible to integrate this behavior into electricity system planning.
Truly Integrated, Consumer-Centric System Planning
The State of the Art in electricity planning is truly integrated system planning, which includes consumer resources and the distribution system as an endogenously modeled element. This stands in contrast to typical electricity planning, sometimes referred to as Integrated Resource Planning, which endogenously models and co-optimizes transmission and generation, with consumer and distribution factors treated as exogenous, potentially varying according to a range of scenarios, without any feedback loops.
The figure illustrates conceptually the difference between an IRP and a true[1] ISP in terms of how different elements of the electricity interact dynamically. For an IRP, you can see the generation and transmission elements interacting. A true ISP integrates each of the systems, modeling their interactions, and feedback loops.
The differences in the level of optimisation between the two are growing in significance due to the potential role of consumer resources to reduce generation and grid costs. Figure 10 shows the difference between a business as usual (BaU) level of system planning and a true ISP level of system optimization, using consumer centric, agent-based simulation methods.
Interestingly, true ISP not only identifies lower cost pathways where consumer resources are used to offset utility investments, but it can also identify opportunities for reducing the level of uneconomic consumer investment in behind-the-meter resources, as shown in this example.
Despite the potential benefits of employing a true ISP approach, the significant complexity involved in setting up more sophisticated methods like ABM have limited its popularity to date. The table below summarises the results of Energeia’s recent review of international system planning methods. Our research found that California utilities are one of the few examples of true ISP at the moment.
For more detailed information regarding the key challenges of analysing and optimising a truly integrated system plan, best practice methods, and insight into their implementation and implications, please see Energeia’s webinar and associated materials.
For more information or to discuss your specific needs, please request a meeting with our team.
[1] A number of plans refer to themselves as integrated, but do not endogenously model consumer behavior.
You may also like
The Future of Data Centre Electrical Grid Impacts
The rapid expansion of data centres to support AI, cloud computing, and digitization is reshaping electricity demand and challenging grid planning. Energeia’s research highlights the
Natural Gas Decarbonisation Strategies and Impacts
Energeia’s research outlines key pathways, including renewable gas blending, thermal energy networks, and end-use electrification, all vital for achieving carbon targets while minimizing economic disruption.
The Balance of Ancillary Services Pricing
FCAS exist to provide AEMO with operational reserves in case of unplanned variations in demand and supply to keep demand and supply of electricity in
