Community Solar Microgrids

Michael Casey, Andy Lippman and Harvey Michaels
October, 2017

Abstract:

We propose an experimental deployment of a community-owned and operated off-grid solar power system that uses the blockchain to manage intra-community electricity management.  The point of the experiment is twofold: i) to provide instant power to regions that have no centralized option and, ii) to demonstrate the efficiencies and social gains that can be extracted from integrating smart meters and Internet-of-Things-connected devices into a fluid, multiparty electricity marketplace that forms the foundation for a future “Energy Democracy.”  The deployment will be in the US Virgin Islands where the existing electrical grid has been obliterated by Hurricanes Irma and Maria.  Centralized power will not be returned to some areas for six months from the disaster but electricity is needed to make the place habitable.  There is a clear need and few other options.  The plan is, in effect, to create a condominium-style distributed electrical plant that gangs all of the roofs and battery plants in a neighborhood into one, smart local grid, owned and operated by the members of the community.  We will develop and package all components so that this can be safely and effectively replicated and scaled by anyone.  The smart grid optimizes electrical use among all members of the condominium and uses the blockchain as a low-friction accounting system for all members.  When a legitimate public grid becomes available, it can be tied to our system for net metering and extensibility.


Introduction:  The need

We all know that solar power is a viable option for household power in many parts of the world, both where there is a functioning electric company and where one is absent.  Solar power is now affordable and understood.  It is developed and successfully deployed in on-grid and off-grid cases; there are many financing options for people who cannot afford the up-front capital investment and some tax credits for those who can.  It is green, effective, efficient and increasingly popular.

To date, there is no such “instant electrical plant” for places where electricity is urgently needed, such as disaster areas and places where individuals cannot bear the cost by themselves.  An obvious systematic approach might be the equivalent of communications company’s “cells on wheels” such as were used at the NYC World Trade Center to restore connectivity immediately after 9-11.  An electric company could install local mini-plants and build out power from there.  However, this neither seems to exist, nor is it within the mandate of providers to supply.  When power is gone, it is gone until the utility manages to rebuild; individuals are left to their own resources to generate power, generally in the face of other urgent needs.
Power is necessary for habitation and humanitarian assistance.  People in Puerto Rico are suffering because medicines cannot be kept cool for lack of refrigeration.  Cistern water requires pumps to provide pressure, raising legitimate concerns about the risk of disease. Communications has all but disappeared from these places.  Power for construction tools is ad hoc.  Reliable electricity is a prerequisite for any other action.

Most important, electricity use has high variance both across time and across residences.  And in the case of  solar, not all properties are equally amenable to a solar installation.  Some are in the shade, lack the appropriate roof area, or don’t have space for an on-ground panel farm.  We argue that there is much to be gained by aggregating a community into a de facto electric company that they manage in common.  A subtle corollary of this is that in the case of emergencies, people generally act together to face the issues; a microgram of their own can be a significant source of community action, pride, and support. 

The Opportunity in the USVI and the world

Island economies like those of the U.S. Virgin Islands make a compelling case for the social and economic value of distributed, renewables-based energy systems. The prevalent centralized model currently in place is inefficient, expensive, environmental harmful and insecure. With limited hydrocarbon sources of their own, the legacy, fossil fuel-based infrastructure on most islands relies on imports of diesel. That dependency has resulted in significantly higher costs of energy for island residents than for their mainland counterparts. It has also created a systemic vulnerability to natural disasters such those that have just occurred, which can disrupt fuel delivery and electricity generation and distribution.

These problems — and the benefits of an alternative, decentralized structure — were brought into stark relief during the hurricanes. On the British Virgin Islands, for example, public grid generation is entirely located on the island of Tortola and then transmitted via submarine cables to Virgin Gorda and all other islands. When the Tortola-based generation plants and transmission lines were taken out of operation, the entire country lost power. By contrast, the core structure of the solar microgrid on Richard Branson’s Necker Island remained intact. Although about half the panels sustained some damage, it was relatively easy to restore power generation without having to wait on the public utility to fix the centralized generation and transmission system. Stronger, hurricane-resistant PV arrays and protective infrastructure are being developed to enhance resiliency of such systems.

The same situation is true in the USVI islands of St Thomas and St John. Power is shipped by undersea cable to St John from St Thomas. St John currently has no power at all; both the connection and the local distribution system need to be completely rebuilt.

We believe the change can and should go further than solely shifting to distributed solar resources, however. The combination of blockchain and IoT technologies gives us an opportunity to build more “democratic” grid systems that could be used as a model, not only for the island economies but for the rest of the world. Our vision is to show how communities, when armed with smart devices, open data, and market pricing signals, can use energy resources far more efficiently than under the current, centralized paradigm.

In that vision, “prosumer” households, responding to price signals, can independently and automatically make sophisticated decisions on what to do with the power they generate, as well as for that which they purchase.  Giving households the right and ability to sell home-generated solar energy to whomever will pay them the highest price, and offering them smart-contract- and digital currency-based financing tools to rapidly add new capacity when market signals show a money-making opportunity, should allow market forces to help modulate and optimize supply across the grid. And on the demand side, smart devices and that can turn on or off based on pre-determined price thresholds can given different usage priorities to optimize consumption patterns. Together, these technologies point to the prospect of a more efficient, flexible system, a system that adjusts to the variation of demand and supply over time in an organic manner, with much less waste.

Automatic and conscious decision also make an off-grid solar installation far more cost effective and efficient for all. Modern solar planning generally builds in a reserve for cloudy days and to preserve battery life by sparingly using them. By optimizing electrical usage across groups of houses and local businesses we can save the capital costs of the batteries and extend their life. We note that their expense is a major reason that most installations in the USVI are on-grid and rely on the utility for backup and high load periods.

Four Aspects of community microgrids

The first and primary goal of our plan is an instant community grid.  This is designed to be installed and operated by amateurs, not electricians.  We will construct it as a plug-and-play system:  instant electricity, just add hot sunlight.

The second aspect is community management.  The idea is that each person contributes what they can to an electric bank and withdraws as needed.  We will net meter across all members of the association. 


The third aspect is automatic, frictionless operation.  We will use a blockchain as the accounting mechanism to allow free and easy establishment of community credit and to facilitate acting in the community interest.


The fourth aspect is smart power control.  We will use this to test automated management of appliances to minimize the demand variance.  For example, to insure that pumps and refrigerators operate sequentially and that everyone’s food doesn’t spoil.

To the greatest extent possible, we will rely on available equipment.  Solar needs precious little invention, it needs deployment.  The goal of our work is to get a disaster-afflicted area operating quickly and simply.

The community

We are inspired by the hurricanes in the Caribbean because this is a case where rapid response is needed and because there is no infrastructure that might stand in our way.  No one will stop a grassroots installation of power: there is no one there to do so and the imperative trumps any opposition.


A second reason is that there is a strong community there.  Their isolation has amplified an existing commonality of interest and lifestyle.  We think they will be receptive to a local power system because it is their style as well as their need.


The vagaries of solar power

Residential solar power is installed with either on-grid or off-grid configurations.  An on-grid system is generally preferred where utility connections are available:  the grid saves the need for an expensive battery system for night and bad weather operation, and the panels have been optimized.  It reduces home energy costs and pays for itself in a few years.  Often generators are used as plant backup for (short) periods when the grid is unavailable. 

Off-grid system are more complicated.  Sophisticated ones can drive a community with a mix of solar, diesel-generation, and battery storage.  For example, the Isles of Shoals, off Portsmouth, New Hampshire, has the largest off-grid solar installation in New England and supports a population of approximately 300 during the summer months.  There are solar farms distributed opportunistically throughout the property, and a set of controllers that occupies a small barn.  It is the only option for that area.

Appledore Island controller barn
Appledore Island controller barn

Appledore Island complete control system

Personal off-grid systems can be as simple as two panels, some batteries and an inverter to provide 100AC.   Costs for the panels are on the order of $1/watt and the associated equipment is reasonably affordable.  A complication is the fact that such system may not have appropriate roof area or ground space to mount a set of panels, and the batteries remain costly and short-lived (on the order of five years).  A general design parameter is that one plans for 4 cloudy days in winter.  This is the worst case and it results in a surplus when the weather and season improves.  Design is thus a function of attitude, with more over-design at higher latitudes where there is more variance in incident, harvestable, solar energy.

Individual panel controllers and inverters
Individual panel controllers and inverters

Recent demonstrations by Tesla (citation) include panels and a purpose-build battery.  Mass replication may make it cost-effective.  Indeed, we have this factored in our plan as will be explained below.  


The Installation

We estimate that the average St John household uses approximately 21KWh per day. With conscious attention to conservation, it could be reduced to less the 13KWh. A 16 panel installation using modern 250watt panels can provide 38KWh/day (with normal conditions) which mates to a 7.2KWh battery bank. The battery can just drive the house through a night at a 50% discharge limit. This is detailed in the appendix.

Each house or business will get/install a 4KW panel system and a 7.2KWh battery system, with 4KW inverters and a specially built “net” meter. This is sufficient to operate a small house with fans, refrigeration, washing machines and normal appliances. This rig is scalable and be made larger.

The local net meter is designed to measure flow to and from the community grid. It is similar to that used by a utility but simpler and less ironclad.

Optionally, we will add to fuseboxes circuit-level measuring and control devices. This way we can evolve to sequence large, automatic appliances such as hot water heaters and refrigerators. The intent is to reduce the usage variance and to insure that there is sufficient energy retained in the batteries to prolong their life and run the community at night.

Each installation will have a small computer that runs the blockchain client software and manages the accounting. Software will be specified later.


Why the blockchain?

In the world we envisage, supply and demand are matched inside peer-to-peer micro-markets. But to achieve that, we need an enabling technology to intermediate trust. Without the centralized public utility sitting in the middle, peer-to-peer trading households must trust that each others’ metering information is correct. They will also need a sufficiently decentralized digital payments system. That’s where the blockchain technology comes in -- providing a structure of trust and a digitally programmable form of money that can communicate and manage IoT devices according to pre-defined smart contracts.

Various companies are already developing blockchain-based systems, including LO3 Energy in Brooklyn and Powerledger in Australia as models. Meanwhile, the Rocky Mountain Institute has joined forces with Grid Singularity to form the Energy Web Foundation, with backing from a variety of energy companies and other stakeholders, to map out an energy future based on blockchain-enabled smart grids. We have an opportunity to both tap into and inform this growing movement for change by demonstrating the value of these ideas in a real-live use case.

Conclusion: The Imperative for On-the-Ground Research


Strategic reflection and discovery is needed to “stretch the thinking” of practitioners and researchers beyond the conventional paradigms that serve as defaults.  Especially when practitioners are distracted by urgent near-term demands, there is a need to have a discovery team that reflects on the new paths forward.  Such strategy projects often inspire new research and development directions, and also may influence near-term decisions to build in contingencies for new paradigms.

We see an urgent need to encourage research/discovery projects that address the hypotheses attached to our proposed model. Research questions and areas of focus include: i) evaluating the relative merit of blockchain for democratized grid transactions; ii) defining energy-democracy business and financing models that can overcome up-front costs and create potential benefits (cost, climate, reliability, resilience, equity); iii) analyzing and composing a blueprint that establishes enabling standards and regulatory policies, and iv) providing high-value case analyses for climate change-impacted and energy-deprived communities, compelling cases that can sharply demonstrate the advantages of this model and expose the challenges in achieving it.

We believe that a rapid-deployment research project in the U.S. Virgin Islands, such as the one described above, would meet these requirements. But our window for action is limited.

The situation in the Caribbean is acute and urgent. The humanitarian priority is to restore power to communities that desperately need it. The risk is that this urgency leads to a default solution in which the old, fossil-fuel dependent, centrally vulnerable model is rebuilt, resulting in the very same breakdowns when the next hurricane comes through. On the other hand, the absence of an effective public utility in much of the U.S. Virgin Islands, coupled with the dire need for power, creates a political and economic vacuum into which we can enter. This environment is, we believe, an ideal one for building a bottoms-up model for advancing our vision of energy democracy.

 Appendix: Toward Energy Democracy


At the core of these ideas lies the notion that decentralization — of both energy distribution and markets — results in individual and community empowerment and that this will ultimately produce the most efficient resource-usage outcomes for society. It’s about “Energy Democracy,” which we define as equitable access for smaller, disadvantaged energy consumers to the high-value energy options currently realized only by larger consumers. That equitable access is enabled by a democratized grid infrastructure composed of microgrids, IoT, and blockchain technologies. It is supported by innovative business models and pro-active energy regulation policies.

Energy Democracy, we believe, will widen energy access well-being, lower costs, grow businesses and jobs, and support an effective worldwide climate solution. It’s also consistent with self-reliance and personal empowerment, values consistent with entrepreneurship and resiliency against the failure of political and economic institutions.

It is especially relevant when we consider the urgent challenges of climate change and the central role that the future of the electric grid plays in that. The bulk of the Paris climate solution comes down to emphasizing three components: i) Ubiquitous renewable electricity production (central or distributed), coupled with ii) electrification of transportation (BEVs/PHEVs) and space/water heating (heat pump), and iii) batteries and efficient IoT-controllable end uses to make the transition work. 

All of these technologies are economic now in most parts of the world . Solar and wind have the largest share of new grid scale power resources. But we’re not moving fast enough in achieving these goals, not by a long stretch. The barriers are no longer in the advancement of these base technologies, whose efficacy continues to improve and become increasingly cost-efficient, but in the design of the economic system that will use them. With renewables increasing the variance of supply, a more effective market mechanism is needed to match it with demand. Otherwise, the default is to back up the system with a surfeit of inefficient, carbon-emitting fossil fuel power.

We must accelerate the evolution of a smart grid that can better integrate intermittent renewable supply, as well as provide net load to consumers with distributed energy resources, or DERs. The problem is that that transition to a more diverse power system runs into the core political-economic barriers of a legacy grid model that’s defined by a hodgepodge of monopoly and market structures and an ad-hoc regulatory framework.

Generation deployment largely follows markets but distribution is mostly managed by monopoly structures. Meanwhile, consumer pricing is heavily regulated, with a hedging strategy that sends inaccurate price signals to households, resulting in inaction and inefficiency, while wholesale markets are far more open. Blockchain-enabled, transactive microgrids potentially change both these dynamics, bringing more competition to the distribution market and spreading the informational power of price signaling to all sectors of the grid. These must be foundational elements of the smart grid of the future.

The path forward lies in democratized grid systems, with multidimensional metering, blockchain-enabled transaction systems and IoT-enabled device measurement and control, bringing efficient markets down to the consumer and device level.  By reducing cost and market barriers to energy/climate business and technology innovations, these systems have the potential to induce: greater electricity access where networks are currently inadequate, damaged, or unreliable; greater consumer access to energy democracy, including potential for sharing-economy-type business models; greater economic value for climate-beneficial consumer choices, by reducing friction to positive-NPV solar, battery, electric vehicles, efficiency/control solutions, all resulting in greater market penetration for these climate-friendly solutions; potential for business and job creation, leading to economic growth.

Appendix 2: Initial Calculations


The Team


Andy Lippman, is Associate Director of the Media Lab and a Senior Scientist at MIT. His research group is called Viral Communications. It addresses grassroots, scalable system. His undergraduate, Jeremy Rubin brought bitcoin to MIT. In addition, his group developed a solar mining facility to show the feasiblility of turning excess power into a fungible good. He has a BSEE and MS from MIT and a PhD from EPFL, Lausanne. He also has a small residence in St John and knows some of the community there.

Michael Casey, TK

Harvey Michaels, TK

Aditya Pittie, TK

Masakazu Kimura, TK

Erick Pinos, TK

Katherine Anabel-Curiel, TK