Home Energy Physics Nuclear Power Electricity Climate Change Lighting Control Contacts Links



By Charles Rhodes, P. Eng., Ph.D.

The following description of Ontario and California electricity market problems includes contributions from Paul Acchione, P. Eng., FCAE in an April 2017 email to the Future of Energy Initiative (incorporated with permission from the author).

Be careful what utilities you invest in!

California has just discovered that when you add clean generation (it is adding solar generation aggressively) the design of the current wholesale energy market will drive wholesale prices negative as producers compete to stay on line.  Some producers have very high shutdown costs and will bid negative prices to stay on-line to avoid a temporary shutdown during low demand or high renewable production periods.

This problem is not new for Ontario because we were one of the first mixed generation power systems to transition to an almost non-fossil power system.  In a mixed power system the number of hours when prices go negative increases as the CO2 emissions decrease.  To keep producers whole financially, there must be mechanism to pass the fixed capacity costs through to the retail rates. Otherwise the producers will go bankrupt as the power system emissions decrease.

Ontario added a capacity cost pass-through mechanism to retail rates with its Global Adjustment.  Ontario also introduced minimum wholesale market bid prices on dispatchable non-fossil generation to avoid excessively large negative market prices. PJM in the USA is using a separate capacity market.

Unfortunately Ontario made two fundamental errors in designing its retail rate structure.  The first error was recovering it's global adjustment via energy use rather than via capacity use at the retail level.  The second error was moving to a flat charge at the retail level for distribution costs rather than a capacity based charge.  Both of those decisions created incentives to use electricity in sub-optimal ways from a power system cost perspective.  Those two errors will drive electricity rates higher over time as asset utilization deteriorates.  For example Ontario is currently exporting large amounts of surplus non-fossil electricity at very low prices and is curtailing (wasting) what it cannot export.  That non-fossil electricity should instead be used in Ontario to displace fossil fuels. However, the present Ontario retail electricity price plans do not allow economic use of surplus non-fossil electricity for displacement of fossil fuels.

Clean power systems (under 50 grams carbon dioxide per kWh) require a different design for retail electricity rates.  Clean power systems like Ontario's incur more than 90% of their costs based on capacity (peak kW) and less than 10% of their costs based on energy production (kWh). High CO2 emission power systems in most other jurisdictions incur more than 50% of their costs based on energy production.  Most wholesale and retail markets in deregulated power system jurisdictions were designed for a high CO2 emission power system.

California is transitioning, like Ontario, from a high CO2 emission power system to a low CO2 emission power system.  As it does so its wholesale energy rates will collapse like Ontario's did.  Ontario's weighted average wholesale energy price in 2016 was 1.6 cents/kWh.  That's not enough money to pay for any known electric power production technology.  So unless a capacity mechanism is added to the wholesale market like Ontario and PJM have done, producers in other markets will likely go bankrupt as the electricity system is forced to become cleaner by well intentioned politicians and their green energy mandates.  Europe has also identified this wholesale electricity energy price as an existential crisis for its electricity producers and is now looking for solutions.

Here is the link to the California news article:


The fundamental concept that politicians and electricity system regulatory bodies must grasp is that in a non-fossil electricity system the retail electricity price must be primarily based on consumer peak power (kW) demand at any time when interruptible power is not enabled. Similarly non-dispatched generator compensation must be primarily based on minimum power supplied at any time when interruptible power is not enabled. That methodology fairly compensates parties for their fixed costs, which in a non-fossil electricity system are typically 90% of total costs.

Energy (kWh) metering should be used for allocation of variable costs, which in a non-fossil energy system are typically only 10% of total costs.

Energy metering can be used for total cost allocation in major buildings where the building management uses shared central equipment to do energy storage and peak demand control for the benefit of all of the building occupants.

In Ontario the electricity market is further skewed by faulty Ontario government policies that encourage reduction of electrical energy consumption (kWh) instead of encouraging reduction of electrical peak demand (kW) and encouraging use of surplus non-fossil electricity for displacement of fossil fuels in other sectors using fuel switching equipment and automation.

An increasing fraction of electrical power is being sourced from power inverters that generate significant harmonics. A typical power inverter approximates a sine wave by six succesive voltage levels. In so doing it introduces significant harmonic power content. When this power passes through transformers the harmonics are attenuated more than the fundamental. Most of the harmonic energy is converted into heat. From an end user's perspective, if the purpose of the electricity is to smoothly drive an AC electric motor, the harmonic energy has little value. Hence if a generator is paid in accordance with total energy delivered to the grid, and that energy has significant harmonic content, then from the end user's perspective the generator is over paid. Further, dissipation of harmonic energy in transformers requires derating the transformer, so from a transmission/distribution cost perspective the harmonic energy adds to the system cost. Hence the compensation paid to the generator should be further reduced to reflect the increased transmission cost as well as the reduced energy value delivered to the end user.

A fundamental question that needs resolution is should an electricity customer pay for total energy absorbed from the grid as if that customer was a purely resistive load or should the customer only pay for energy provided at the fundamental frequency as if that customer had only an induction motor load? Similarly, should generators be paid for total energy delivered to the grid or just for energy delivered at the fundamental frequency? Who is responsible for the extra costs of managing harmonic energy?

An issue worthy of thought is that in the future small (20 MWe) distributed nuclear generators may use inverters or cycloverters to transfer power from steam turbine driven direct coupled generators that typically operate at 14,000 RPM to 60 Hz. The alternative gear drives for frequency reduction seldom reliably withstand the rigors of continuous operation. However, the solid state inverters will have some of the harmonic issues and moment of inertia surge current constraints that presently beset wind and solar generators.

Moment of inertia is an important aspect of electric power systems that is seldom understood by the public and is generally only appreciated by a limited subset of the engineering community. Moment of inertia in an electricity system is discussed on the web page: Generation Valuation. It is a subject that is briefly discussed in a 4th year power machines course but then is usually forgotten about by most engineers.

Up until about 1990 almost every significant electricity generator incorporated a mechanical rotating device that inherently provided moment of inertia which contributed to electricity system stability. Then came the availability of large power semiconductors and power inverters with ratings and costs suitable for use in distributed electricity generation systems.

There are two classes of power inverters, current source inverters that rely on the external electricity system for voltage control and voltage source inverters that provide their own voltage control.

A voltage source inverter can be programmed to simulate the moment of inertia of a synchronous generator. However, when a synchronous generator experiences a load transient an energy spike is transferred to or from the synchronous generator rotating moment of inertia. Accommodating this energy spike in a power inverter requires switching elements and energy storge that must accommodate peak currents much higher than normal. Those higher peak current rated switching and energy storage elements significantly increase the cost of a power inverter. Consequently the power inverters used with most solar and wind generation are current source rather than voltage source inverters. Thus in a real power system with a high penetration of wind and solar electricity generation a large fraction of the generation provides no moment of inertia.

As the ratio of AC power to moment of inertia increases the power system becomes more unstable. In a centrally managed power system unloaded synchronous motors with large flywheels can be added to the system to provide the required stability. However, in a pure uncontrolled energy market no party is compensated for providing the moment of inertia required for electricity system stability.

From a legal perspective the fix is simple. All generators must spend whatever money is necessary to provide their fair share of the required electricity system moment of inertia. C. Rhodes identified this problem to the Ontario Power Authority about 13 years ago. However, no one listened. The lawyers refused to incorporate this technical complexity into wind and solar generation contracts. Today in Ontario nuclear and hydroelectric generation provide the required electricity system moment of inertia even while nuclear and hydroelectric power are being discarded. Today in ERCOT there is a high penetration of wind generation, fossil fuel generation has been taken out of service and parts of the ERCOT system are on the threshold of instability. As PJM introduces increasing amounts of wind generation this instability issue will become more pronounced. In CASIO the projected decommissioning of nuclear generation may have such severe instability consequences that the existing generators may be left running as unloaded synchronous motors simply to provide electricity system stability.

A significant issue for FOEI to address is how, in current circumstances, to force generators to either provide the required amount of stabilizing moment of inertia themselves or to fund others to provide it. Clearly a kWh supplied by a generator that provides its own moment of inertia should be worth more than a kWh supplied by a generator that provides no moment of inertia. However, this issue is not reflected in current electricity market rules.

This web page last updated April 28, 2020.

Home Energy Physics Nuclear Power Electricity Climate Change Lighting Control Contacts Links