DERMS: “Distributed Energy Resources Management Systems”.
“Beware acronyms bearing high word counts” (unknown skeptic). And KISS, the engineering mantra: ”Keep it simple, stupid”.
As an old hardware/software/network/security engineer, with some experience in CPUC hearings (even during power failures), let’s start with a dose of reality…
a) Electric power systems already are, and provide, distributed energy (DE).
b) Power systems are already “managed” by both software & live humans.
c) All DERs must provide utility-grade power as/when demanded by the established regulator.
What a) reflects is that an electric-generating site is placed wisely, in relation to loads, customers and generation technology realities (hydro sources at reservoirs, steam-driven sources near cooling (rivers, oceans, or artificial evaporative structures). System trunks (backbones) distribute power from major sources (e.g., Diablo Canyon) across our state and allow interstate sourcing (e.g., the 500-1MV DC Pacific Intertie). That assures some geographical and jurisdictional robustness.
Distribution continues from such backbones to trunk lines feeding substations that service neighborhoods and industries -- even our BART, new CalTrain and everanticipated high-speed rail.
As b) indicates, power systems already rely on considerable software management, even for the moment-to-moment contracting with sources that fills CAISO’s day. ENRON constructed a particularly evil, collusive example of human+software management of DERs. Responsible engineers keep it in mind when going gung ho on computerization, especially of the inter-networked kind, which can make ‘juicy’ targets for hackers across the world.
In c) we must face the reality/responsibility all engineers/scientists assume when stepping up to receive their degrees. At those moments, we pledge, much as do doctors, to “do no harm”. In other words, all us engineering professionals pledge to be honest brokers of fact to society. We don’t mislead and simply help to ‘market’ personally profitable projects/actions to society.
With that in mind, consider “utility” and “utility-grade” DERs. Society depends on certain things without which “society” falls apart. In the OECD** world, water, sewer, fire, police, healthcare and energy are prime “utilities” that society depends on 24/7. We sufer what happens when any of those services fails. And “failure” can include poor maintenance and untethered pricing (e.g., ERCOT).
What energy resources (ERs) should we consider placing under existing or better/more software management without increasing failure points/failures?
We know that wind/solar sources are extremely unreliable and wasteful of materiel (DoE Quadrennial Review) and installed Capacity to boot -- DoE comparisons show wind energy consumes over 10x the materiel of building an equal nuclear (e.g., LWR) wattage, while PV consumes over 16x.
So, even without belief that truly massive battery-storage additions might hide their undependability, we know, in reality, that wind/solar are not utility-grade ERs, even if we could afford massive overbuilds, environmental intrusions and massive storage installations that include predictive systems looking ahead days/weeks to determine, with utility-grade reliability, when and how much storage to drain, charge, or add.
We can also simply listen to moneyed interests we blithely subsidize:
Warren Buffet 2014: "For example, on wind energy, we get a tax credit if we build a lot of wind farms. That's the only reason to build them. They don't make sense without the tax credit."
US News & World Report (2014): "Big Wind's Bogus Subsidies - Giving tax credits to the wind energy industry is a waste of time and money." Charlie Archambault, Nancy Pfotenhauer, https://tinyurl.com/Buffett-Wind-Scam
For wind power, there’s particular danger to all around of what occurs in strong storms, especially tornados. What happens can spread over miles. Here’s an Iowa storm chaser sidling up to a large tornado, at speed, that’s making expensive trash of a wind ‘farm’ or two, or more… https://tinyurl.com/3yn6k3uv Those unreliable ERs were already “Distributed” and may end up in Nebraska, awaiting clean up, loss claims and perhaps metals recycling.
Apart from wind hardware’s vulnerability to Nature’s reality, we have the sad reality of worker injuries/deaths, working >>100 feet up, servicing several- hundred-ton prop-generators, or ‘just’ replacing blades that so often experience damage and wear…
.
A question for wind ‘farmers’ is: are workers employees, whose safety is covered by OSHA rules? Or, are they purposely chosen by wind owners from a migrating pool of gig workers, working at their own risk? This used to be the danger celltower workers often faced. Do ‘green’ energy jobs provide proper safety?
As for old blades – they aren’t actually old in the common sense (they do suffer hard lives). Their production/replacement has decimated balsa forests in SE Asia, and the suggestion to chip them for burning in coal plants just illustrates how far some are from grasping environmental reality.
Even the newest, optimally located UK wind machines in the North Sea only deliver about 42% (Capacity Factor) of their built generation capability. This means their materiel/construction investments start with a yearly 58% assetstranding (non-performance) loss. Only vast subsidy forces this to look sensible to owners. But to society, it’s expensive and far from utility-grade service.
Why all this trashing of actual wind/solar? Because one myth is that they represent clean “distributed generation” and are worth including in any overarching management system, software or not. Management software becomes an enabler (in the minds of those in charge) to make fashionable but poor decisions on what energy producers (ERs) to deploy and maintain at utility customer expense.
For example, ERCOT (Texas) doesn’t regulate reliability of power sources, so during 2021’s Storm Uri, natural gas pipelines were not all weatherized, so some generation was lost when gas plants discovered pipelines clogged by methane clathrates, killing plant power outputs. Wind/solar power demonstrated vapid dependability by simply not generating much during the cold, stagnant, overcast days from 16 February on…
People died (>1000) for lack of gas heat, gas generation, or simply ineffectual wind/solar generation. Texas’ 4 nuclear reactors, however, plugged reliably and cleanly along, preventing a grand catastrophe (Texas intentionally avoids electrical links to outside utilities). The state’s electrical system was within minutes of total collapse on 2/17. Such would only be recovered via “black start” – a few sufficiently powerful plants coordinating AC voltage and phasing as they gradually restored power to key loads. Wind/solar (‘renewables’) cannot engage in such large, critical, coordinated power-system restarts.
In Calif., our fanciful “Duck Curve” hides daily PV waste – “curtailment” of overbuilt PV generation at citizen expense***. Our ‘duck’ also has sneaky backside**** – ~80% waste heat/IR emitted by commercial PV (military/space PV’s waste is ‘only’ ~60%). So the ‘duck’ has a hot, tall waste-energy hump on its back, lots higher than it’s daily spinal dip is deep. Funny that no PV marketers/promoters point this out and our legislators appear not to know of its simple threat to climate. As we learned in NJ kindergarten: “Follow the money”.
Another sizeable energy/pollution waste that PV induces is forcing off system our most efficient (combined-cycle) gas plants (Colusa…) when the sun is visible. This forces inefficient gas-turbine “peakers” to idle (still burn gas) while we all await the sun’s daily ‘bye-bye’ – returning us to dependency on reliable power sources (a gas-turbine owner must keep it hot and ready in case of clouds, etc. and certainly warm enough to meet our ~60 MW/minute demand ramp as the sun sets. This extra pollution & costs aren’t lain at solar owners’ feet.
Solar PV is illusory in benefit. In 2005, Cal Tech’s Lewis Group estimated that the US’s then ~1TW power demand could be met by the solar energy (1kW/sqm) falling on a 200mi square. Even for today’s world’s 20TW 24/7 demand this sounds benign.
Our unregulated, inefficient PV, however, would need 5 such ‘farms’, or about 270x1000mi. Worse, the worldwide waste heat/infrared generated would be about 60-100TW –truly climate changing. And then, the diurnal ‘duck’ raises its head to demand we somehow simulate 24-hour sunlight. Batteries? How many? Where? Resources, materiel, pollution, transmission, cost… 20TW ready all day every day from some source(s) just to always charge a 12-hour societal ‘battery’ that kept things running all last night. Local (rooftop) PV, substation storage, reversible distribution etc. solve nothing. They fail the KISS test, but sure add product/service revenue opportunities (e.g., Capital Cost Recovery in CA).
Of course, we want less fuel burning, but solar PV has never been regulated for its massive energy waste, directly adding to local/global warming. This irony is a good example of failure to “Do no harm”. We’ve known how to make PV at least twice as efficient and how to transfer waste heat (blackbody radiation) safely into space**** (e.g., via radiative cooling), but no regulations have been forthcoming.
Commercial PV is thus an imperfect, unregulated product that, like wind power, has acquired subsidized status for illusory societal reasons, despite massive environmental detriments. Just like subsidizing prop-generators on stormy lands, placing flat glass panels in the open, weather-prone world often leads to messy realities…
https://tinyurl.com/y83g6htx (Puerto Rico’s hurricane Maria) The video above includes wind machine damage.
We could learn from glider pilots who look for large PV arrays and use the hot daytime updrafts created by their ~800 Watts per square meter (~600W/sqm for military/space PV) of waste heat. Migrating birds think they’re lakes to land on, cool off in and rest a bit. Our ignorance is their regret.
A little California energy history…
California has The California Council on Science and Technology to call on any time we want good answers to science/engineering/environmental issues. We did that in 2011 – we asked what our energy future should look like – they answered….
Answer: ~1/3 nuclear power, ~1/3 wind/solar, ~1/3 geothermal/biofuel, etc. And, in 2011, we were on track to such a goal, having The Geysers geothermal, Oroville plus other hydro, Diablo Canyon and San Onofre nuclear – all yielding utility-grade power to meet ~30% of total CA demand. Our various wind/solar/biomass installations brought us to ~50%. The remainder of our usage was served by gas and out-of-state coal combustion (LA has long had its own coal plant in Delta Utah, which is planned to cut output and switch to gas – a baby step to cleanliness).
CA’s energy pie (approaching CCST’s advice) with San Onofre would have been a good start, but we faltered, when long festering national politics brought about 2005’s repeal of the 1936 Public Utilities Holding Company Act. San Onofre was shut for no nuclear problem, rather for its holders Sempra Energy (gas) and Edison International (gas + electric) to access its ~$3B trust fund (the PUHCA would have forced them to get SEC approval).
The CEO of Edison International also stated (personal communication) that held utility SoCal Edison would be charging its customers an extra $1.5B.
Thus, CA emissions went up and power reliability went down (Oroville Dam too). But, CAISO can always depend on Diablo Canyon…
9 Sep. 2015: “Planned Maintenance at Diablo Canyon Unit 2 Delayed to Meet State Energy NeedsDuring Heat Wave” CAISO: “Requests Both Units Operate at Full Power”.
Unlike sun/wind/fuel-dependent generation, nuclear plants already have fissile energy in-reactor to last several years and never run to ‘empty’ (the US Navy runs nuclear ships for ~30 years per fissile ‘fillup’).
Fortunately, PG&E has maintained its relatively clean power mix, but “unspecified” (below) includes considerable coal power, as from BerkshireHathaway’s BHE-owned Pacificorp. Warren Buffet wants to keep that business going and is trying out an interesting ploy called WRAP (Western Resource Adequacy Plan) which would rush his coal power into any breach brought on by any CA wind/solar, etc. power evaporation. Buffet also expects CA to run a ~1000mi HV transmission line up to Pacificorp furnaces/generators. Interestingly, WRAP would be meaningless if San Onofre were repaired and restarted – reliability would rise and emissions would drop. Gotta watch those billionaires.
Because PG&E has extended Diablo Canyon’s nuclear operation (per NRC licensing), its generation suite remains ~50% clean (pie chart). Peninsula Clean Energy (a CCA) bought 420GWH from PG&E in 2017, perhaps not aware it was cleaner power simply because of its large nuclear/hydro proportion.
A most valuable analysis has been completed that shows just how profitable was PG&E’s decision to extend Diablo’s licensing under NRC approval…
A $billion or so helps offset those sneaky CA wind/solar curtailment fees***.
Social value
A management decision to invest in a clean-power life extension nets CA electric customers ~1.3$ billion. In fairness, the Governor and legislators had a hand in this decision. An example of wise MS action applied to a clean DER? The CA CPUC has a charter whose first commitment is: “All decisions shall be made in the public interest” (our CPUC was founded partly because Leland Stanford overcharged for cargo transport on his railroad).
A corollary question for DERMS enthusiasts is: what benefit of DERMS if a DER installation’s outputs are undependable/polluting/costly and effective control functions or corrective actions are unavailable?
Arun Majumdar’s group at Stanford expended great effort to analyze electricutility loads from a full range of customers, from homes through industries. They compiled diurnal and weekly load profiles falling within several defined categories. The intent was to expose how we might aggregate certain customer loads in ways to improve overall power service delivery and manageability, and to establish how best to incorporate local household generation, such as rooftop solar PV. Could local electric loads (appliance, EV…) and generation (PV) work together to improve utility reliability and customer service/cost?
A simple outcome is: “not really”.
One intriguing example has existed for managing EV charge/discharge cycles – Betterplace.com provided battery management, including charging, foe EV owners in Israel and Denmark. The batteries were owned and mostly charged by Betterplace. Each held its own cellular network interface for remote management by Betterplace itself. The car’s owner could contact Betterplace about a planned trip or weekly commute and Betterplace would be sure the EV owner would leave the car plugged in long enough to charge for leg 1 of the planned trip or commute week. The owner’s power reliability is key here.
Going on a trip meant notifying Betterplace of enough details so that a stop along the trip could be made at a Betterplace battery station, whereupon, the EV’s battery would be swapped for an equivalent charged one. A long trip might incur several swap stops. But the EV owner wouldn’t care – the batteries are free and charged, awaiting a few minute swap (Tesla once demonstrated their version).
This EV management system provides social benefits: 1) battery charging times evaporate 2) batteries can be charged at optimal hours (e.g., at lowest power pricing), 3) battery maintenance/failure is almost eliminated from customer’s experience and 4) Betterplace could offer local utilities both load and ‘generation’ services. That last can directly help smooth daily utility loads and help eliminate the perversities of, say, the CA ‘duck’. Because Betterplace batteries (even in EVs at homes) are cell-network connected, they can participate in ‘smoothing’, which may result in credits to the EV owner’s utility bill. Majumdar’s work includes dynamic, utility-controlled EV battery charging.
Management systems
That is a good example of one absolute need, if DERMS is considered at all by any entity – remote, secure access. Computer and Internet based command & control functions are remarkably weak against intrusions. A ‘fun’ fact of the Internet (TCP/IP protocols) is that it was never intended to be secure, unlike IBM’s vintage SNA. And, it has no concept of physical location/address of Internet-connected devices – a boon to hackers. As DARPA’s past head of Internet development said a few years back, when interviewed about Internet data breaches: “I’d do it differently”. Thanks.
So, for years, Internet users have had to encapsulate important Internet traffic in encrypted, virtual networks (VPNs) riding (like gibberish) atop TCP/IP. Only the user’s end equipment/software knows how to decrypt and reveal the user’s private content (text/data). Unless we treat management systems/software for utilitygrade services with deserved respect, we guarantee society-threatening failures, violating our oaths as engineers to “do no harm”. Should a PUC allow fragile, penetrable management systems, it will violate any pledge to place public interests first.
Subsidy has long been a bad influence on systems development. As we learned in NJ 1st grade: “the best money is government money”. There’s lots of it around now, a lot subsidizing what we already know are misdirections like wind/solar power. An odd consequence of our CA emissions-credits program was related by the head of a CA water agency whose clean hydro power garners them a state carbon-credit certificate. They sell it to a GHG emitter. The environmental effect?
It’s not like we haven’t been warned, as by The World bank of all places…
The graphic simply shows we’ve no hope of meeting IPCC targets with wind/solar power builds – something engineers & scientists have been warning of for years. Apparently, politicos and some environmentalists prefer to validate Churchill…
“Americans always do the right thing, after first trying everything else.” (also attributed to Abba Eban re our policy for creating Israel after WWII)
A sad example is trying to make wind generators a bit less wasteful by using improved polymers to allow crushed worn-blade material recycling (AAAS Science 7/23/24, p854). Yet, even a perfect wind-generator prop wastes 1 - the Betz Limit (or ~41%) of oncoming wind energy, leaving at most 355 Watts per swept square meter delivered to the generator at typical 40km/h wind speeds. Storm-downed prop generators leave much more afield to be recycled.
Any proposed DERMS design must comprehend infrequent but large, valid power demands. A northern CA example is NASA Ames – its large wind tunnel consumes ~200MW. Ames staff can’t just wheel in an aircraft and throw the tunnel fan motor’s switch. They must call CAISO and request a mutually workable load time & duration. Otherwise, the power system will likely fail for many other loads and users. Human action and oversight will likely not be replaced by any MS.
A less stringent example is Amazon’s new nuclear-powered datacenter… https://tinyurl.com/vz47cpb7
If this is the beginning of a trend for corporate large clean-energy capture, then PUC rules and MS functions must reflect societal clean-power needs. In other words, this type of DER is environmentally and economically desirable as long as its owner agrees to operate it under rules set by the local PUC and the public gets fair use of power throughout any day/month/year. And, as in the Ames example above, some sporadic loads may have special access for social benefit. Because of AI and other power-hungry applications, there will be more of these, so an MS here will be carefully thought out to avoid Amazon monopolizing its clean, reliable power.
A similar situation exists in TN, where a steel-recycling company has two 100MW furnaces. Economics keeps then hot 24/7. There power comes from a mix of coal and nuclear. An MS in this case should work to maintain plant operation, maximizing its access to regional nuclear/hydro power and minimizing coalsourced power, looking forward to its eventual nuclear replacement.
A challenge for DERMS implementations is evidenced by aggressive attempts to profit from undependable sources. For instance, despite Warren Buffet saying (2014) wind ‘farms’: “don't make sense without the tax credit" his BHEis hot on the subsidy trail – “BHE Montana will add 397.5 megawatts of clean energy generation from three windfarms and 75 megawatts of a battery energy storage system. In June 2024, BHE Montana also announced plans to develop Glacier Solar Park, a new 100-megawatt solar” (per CAISO).
Note that: “75 megawatts of a battery energy…” isn’t “energy” but power. How long will that power continue to flow from that storage (4 hours is often implied)? This illustrates why MS designers have work cut out for them. Also, at this writing (3pm PST), CAISO reports that of all our amassed CA ‘renewables’ just 60% are “serving demand”. That makes 40% of all CAISO-accessible ‘renewaables’ stranded/nonperforming assets, perhaps invisible at times to any MS. BHE’s Montana PV is also non-performing by the time of day CA needs ramp up for evening loads.
We should build or connect with more?
And, CAISO created the Extended Day-Ahead Market (EDAM) following entry into the Western Energy Imbalance Market (WEIM) created by Buffet’s BHE and others. Using these markets may add complication to an MS. It certainly adds hidden subsidy costs to end-customer power bills. The danger is that reliability can fall below utility-grade despite MS design.
What should a management system (MS) see and do?
Many utilities already have automated customer meter reading that can report real time power use. Similar information may already be aggregated at or for neighborhood substations. These paths can be made bi-directional so that an MS can influence substation and individual load behaviors. That can include customer battery charge/discharge with fee/credit functions.
Customers with local PV installations are usually required to have inverters that automatically disconnect (island) when utility power fails, allowing safe utility staff repair work. This function can also be triggered by an MS with proper inverter interfacing. The utility can decide to allow daytime power from PV to pass back into the local substation, perhaps on a per-address basis. The consideration here is that neighborhood power wiring and pole transformers may be overloaded by aggregated customer power at high PV hours (e.g., solar noon). An MS must be designed to report and handle this dynamically by shedding PV input(s) appropriately. The same concepts apply as PV-sourced power is aggregated back into the utility itself. The MS must have appropriate access and real-time control.
Because of the significant MW totals of undependable western wind/solar installations, an overall MS function must be predictive. In other words, if WEIM, for example, has wind/solar/storage Watts for sale, an MS wanting to use the implied Capacity must be sure it’ll be available when needed by the entity employing DERMS. Not only is there a risk winds won’t blow or clouds may appear, but the accompanying storage may lack full charge due to the WEIM member’s actions. Obviously, both the client MS and the WEIM vendor must have means to estimate probability of success/failure to fulfill a given need.
Storage in any undependably-sourced power system only corrects for limited generation shortfalls, even with the best predictive management systems. Ideally, for instance, storage with accurately-predictive charging could eliminate gas peaking and backdown-mode pollution.
We should understand that as long as government subsidies are provided to investors to build undependable (wind/solar…) rather than utility-grade generation (hydro/nuclear…), we’ll see inherently wasteful deployments atop vast tracts whose developers’ intent is other than utility-grade power provisioning to society. DERMS designs must take account of this.
Finally, DERMS should expose all relevant control points and functions, whether physical or economic, in whatever power system is so managed. Abstractly, this is a security nightmare. Every exposed MS component must be physically and logically secured. This means that communication interconnects among MS components must be both duplicated and secured. Any utility implementing DERMS must provide redundant yet separate communication among all MS operators and control entities. Communication must be privately implemented – no public network use, no unauthorized access to software, data or control functions.
The most important step in secure systems design is physical security -- separate MS installs/sites/paths, etc.
Training on system use should be compartmentalized, so that only a few individuals know each management function and each MS instance. Overall MS operation/reporting/control is known by just a few key personnel. The system also implements auditing to help with training, fault and security detection/reporting.
Some Background…
The critical need for reliable electric power was clear long ago, when President Kennedy established the Seaborg commission: https://tinyurl.com/ye27te2z The conclusion was simple – nuclear power based on Thorium and Uranium provides unlimited, reliable heat and electric power to sustain world economies, environments and national security. Environmental benefits weren’t much known in 1962. Now they are. Some important environmentalists…
Appendices
* Dr. Alexander Cannara Menlo Park, CA 94025 [email protected] 650-400-3071
** OECD is a global policy forum that promotes policies to improve the economic and social well-being of our societies - helping to shape better policies for better lives.
*** California’s 2019-20 increasing monthly & annual wind/solar “curtailment” Cost (>1.5 trillion watthours) ensconced in our electric bills…
Owners of curtailed generation still get paid and CA pays out-of-state utilities wishing some ‘green cred’ to take the curtailment, all at CA utility customer expense – interesting how wordsmithing (“curtailment“) can cover up foibles or worse. As we learned in NJ kindergarten: “Follow the money”.
**** The CA PV ‘Duck’…
The backside of the ‘duck’ is a huge thermal waste ‘hump’ (lefthand red dashed boxes) ~4 times taller (commercial PV) than the height of the green ‘duck’ body of actual PV electrical energy delivery. CA energy folks don’t show us the hump, but Ma Nature and we all pay for its (blackbody radiation) waste. Proper, wasteminimizing PV would appear much as the righthand rooftop picture shows for radiative-cooling panels that redirect PV’s waste blackbody spectrum to spectral ‘holes’ that radiate into space without atmospheric warming (absorption).
Anyone making decisions or advising on energy sourcing is obligated to study alternatives accurately and thoroughly. For example, solving PV’s thermal waste problem doesn’t solve its diurnal or weather-sensitive generation problem.