Grid responsive control device

Abstract

A load control device which is responsive to a physical variable representing the balance between load and generation on an electricity grid. The control device varies the energy consumption of the load based on the current value of the physical variable of the grid relative to a central value of that physical variable, which is derived from past readings of the physical variable of the grid. The grid responsive control device also takes into account the time since the load last varied its energy consumption in determining whether or not the grid variable load control should be provided.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]This application is a continuation of copending U.S. patent application Ser. No. 11 / 921,362 filed Nov. 30, 2007, which is the national stage of International Application No. PCT / EP2006 / 005252 filed Jun. 1, 2006.FIELD OF THE INVENTION[0002]The present invention relates to a means and method for controlling the balance between supply and generation on an electricity grid.BACKGROUND OF THE INVENTION[0003]A reliable source of electricity is essential for almost all aspects of modern life.[0004]Providing reliable electricity is, at present, an enormously complex technical challenge. It involves real-time assessment and control of an electricity system consisting of generation, of all types (nuclear, coal, oil, natural gas, hydro power, geothermal, photovoltaic, etc.), and load e.g. the appliances, instruments etc. using the electricity.[0005]The electricity is supplied over a distribution network consisting of transmission lines interconnected...

AI Technical Summary

Benefits of technology

[0062]Conventionally, a nominal frequency of the grid and a current value of the physical variable is used for controlling the energy consumption of the load. The present invention, however, uses some function of the past readings of the physical variable. This gives a long-term past value for the central value and it is this central value that is taken into account for controlling the energy consumption of the load. Modeling of the present invention has shown the use of a historically based central value for controlling the energy consumption of the load removes the strange effects on the grid frequency found to exist with prior art grid responsive control devices.

[0123]In this way, the limits are moved at a rate less than the physical variable could theoretically move. The lower rate of the limit movement means that there is still some provision for the load to be grid responsive even while variable limits are being changed.

Problems solved by technology

Providing reliable electricity is, at present, an enormously complex technical challenge.

Reliable operation of a power grid is complex as, at present, electricity must be produced the instant it is used, meaning power generation and demand must be balanced continuously.

Failure to match generation to demand causes the frequency of an AC power system to increase when generation exceeds demand and decrease when generation is less than demand.

However, large deviations in frequency can cause the rotational speed of generators to move beyond tolerance limits, which can damage generator turbines and other equipment.

The variation in frequency can also damage loads.

There are problems with the present supply-side style architecture of matching generation to demand.

This may have the effect of stabilizing the system, but is extremely inconvenient and even hazardous to the user.

After a blackout the grid is at a particularly sensitive stage and recovery is slow.

Every step when load or generation is added is a shock to the system and the grid can take seconds or minutes to stabilize after it happens.

This inevitably slows down the overall process, prolonging the blackout for those who are still to be reconnected.

This, however, means that the generation is not operating as efficiently as possible, with a resulting increase in electricity supply costs.

High air conditioning and other cooling loads in the summer and high space heating loads in the winter are a normal cause of peak-loads.

Utilities can use additional peaking generators, which have high running cost, to provide additional electricity when needed or, alternatively do not operate main generators at capacity so as to leave some potential for extra generation to satisfy excess loads.

Both of these methods result in a higher unit cost of electricity than if the system was operating nearer to capacity.

Such devices would not be able to (fully) fulfill their function if the power was not supplied at a designated time and rate.

Otherwise, the pump is idle.

This means that the excess in generation is being taken up.

One problem with the FAPER device is that, without any randomization, the smallest movement of the frequency could result in all loads with FAPERs applied responding in the same way and doing so simultaneously.

This could result in a destabilizing influence on the grid.

For example, it would not be appropriate if a particular device was constantly sensitive to the slightest change in frequency whereas another device had such broad trigger frequencies that it only provided frequency response in extreme grid stress situations.

One problem with this system is that the controller is not tamperproof.

Thus, if the air conditioner is generating in a lower temperature range, that is the air conditioner is working harder and is on more frequently, because of an increase in grid frequency, and a user notices this and turns the air conditioning down, before the frequency returns to an acceptable level, then the response has been lost.

In this way, manipulation of a set point controller, such as a thermostat, is made ineffective for the duration of the period of high stress.

Further, the potential energy of a population of water pumps controlled in this way will have increased their average depth of water.

In the case that a stressed state is determined, the control limits of the device are frozen at pre-stress settings so that manipulation of a control panel to adjust a set point for the sensed physical variable (e.g. temperature) is ineffective.

Thus, the user of the controlled load cannot adjust the loads settings, for example by using a thermostat control.

These extreme measures are only taken in the most serious of grid conditions, when the alternative is a blackout.

Modeling of the prior art frequency and responsive control devices has uncovered previously unknown problems with the above described prior art grid responsive loads.

It has been found that after response has been affected for a period of time, a population of the devices will tend to approach the physical variable control limits, and start switching at an excessive rate.

This results in very frequent switching as the device is attempting to provide frequency response to a unit close to its physical variable limits.

This is not desired behavior as it could damage the controlled loads.

Excessive oscillating on and off switching of the load will reduce the lifespan of the device.

This did not, however, entirely bear out during modeling, and some previously unknown strange behavior of the grid frequency was observed as a result of the responsive loads.

The prior art grid responsive control devices do not provide any special assistance to a grid recovering from a blackout, but the stabilizing effect of responsive loads are needed more than ever at this time.

The loads and generators can be reconnected more quickly, so speeding recovery.

Images