The Generation Gap


It is known that there are a number of different types of power generating stations in use. This piece will discuss the effects of their behaviour on planning of power generation.

Demand Curve

To begin with, let us consider the demand curve. ‘Demand’ is the amount of power instantaneously required by a piece of equipment. The combined demand of all the equipment running at any given time, plus the losses involved in transporting electricity from the generating station to the end user, are to be matched by the power generation in the power plants. This matching is essential as power cannot be stored on such a large scale, and because it is necessary to operate the power plants and maintain the grid in a stable condition. The demand curve usually follows a pattern over a 24-hour period. The demand is never zero, it does not drop below some minimum non zero value. Think of the midnight-early morning period, when electricity usage is minimum – things such as fans and refrigerators in houses, some essential services such as streetlights, water treatment plants, mobile phone towers, and other cases such as heaters in winter, still remain in operation during these hours, whereas the other equipment whose usage is typically associated with human activity, remains inactive during this period. This minimum load over an entire cycle is known as the ‘base load’. The peak load occurs during periods when electricity requirement is at its highest. The load keeps on varying throughout the day.



So what happens during a sudden significant increase or decrease in demand?

Consider the electricity being generated by a rotating generator. It has specific voltage and frequency levels which have to be maintained within narrow ranges. By the law of conservation of energy, the power output plus the losses will equal the power put into the generator (by the steam turbine, water turbine, gas turbine etc.). If the demand increases suddenly, the power input is momentarily insufficient, which causes the rotation of the generator to slow down and decelerate continuously, and causes a dip in frequency. It requires a bit of time to increase the power input into the generator and stabilize the system. On the other hand, a sudden dip in demand means that the input power is higher, which causes the generator to accelerate and increase the system frequency, and the input power has to be reduced within a specified time period to stabilize the system. If not controlled within time, both a sudden increase or a sudden decrease in demand can cause parts of the grid to shut down and cause blackouts over large areas.

Generation and Demand Matching

Now, there is a cost factor associated with the load variations. During peak demand requirement, the losses in transmission are also high and these also have to be matched by the power generated in the power plants. Generation of power, which is still largely dependent on fuels, comes with expenditure for the fuel consumed. Different types of power plants are operated on different types of fuel and have different costs associated with them. Different types of plants need to follow certain procedures to prepare for power generation.

Some types of power plants, such as coal-based and nuclear-based plants, require a long time to start up and are typically operated for long periods once started. In a coal-based power plant, for example, the steam needs to be heated to a certain temperature while maintaining a certain flow rate to achieve the required power output, and heating up the boiler to the required temperature takes a long time. Such plants are used to handle the ‘base load’, as the load is always there and the plant operation does not need to be cut off or adjusted significantly as the base load demand will remain fairly consistent. These are called as ‘base load’ plants.

Intermediate/peaking plants are able to start up and get prepared for power generation in a shorter time compared to coal-fired power plants. This makes it possible to control their operation as per the variations in demand. They can be switched on and switched off relatively quickly compared to base load plants. There are some generators which are operated below their rated capacity under normal conditions, so that in case of sudden increase in demand their output can be increased to the rated output, without requiring the complete start-up of an inactive power generating unit.

To maintain reasonable costs for power generation and match the generation to the requirement, the different types of plants have to be operated in an optimized pattern. This operation optimization is performed by the ‘load dispatch’ center at various levels, and also factors in the historical observed demand.

Renewable resources like solar and wind are more challenging to adjust because their output is significantly affected by weather conditions. Solar power plants usually show a sharp increase in generation over a small time period as the sun rises, then maintain a reasonably steady output throughout the day depending on the position of the sun, and then show a similarly sharp fall in production around evening time. This pattern can occur at various times of the day depending on season, and the output is highly susceptible to weather conditions and clouds can cause a significant dip in production.

Case Studies

An interesting case in power plant scheduling is the ‘duck curve’ issue reported from California. It looks like this:


This graph was first released in 2013. The curves show the 2013 observed power generation trend and predicted trends for later years for fuel powered plants over a 24-hour period. When a large proportion of solar power is available, it is included in the total power generation and a lower level of power generation is required from the fuel-powered plants. With high usage of equipment, the demand is highest during the daytime. This often includes the air conditioning load during the summer. Due to bright sunlight available at this time of the day, the generation by solar plants is also high during these periods and thus they can be preferred over operation of larger number of fuel-consuming power plants.

The problem comes up in the late evening hours when there is a large and sudden dip in the power generation from the solar plants due to significant reduction in ambient light levels around sunset, but the demand does not decrease at the same rate. This dip occurs within a very small time window, hence it is required to prepare a sufficient number of quick-acting plants to come online quickly as the solar generation dips but the demand increases. 

Another case is from India in early April 2020. Readers from India might be aware that it was requested by the Prime  Minister to switch off all lights in houses for a 9-minute duration for one day at 9 PM as a tribute to those working on essential services during the Covid-19 lockdown. As it was expected that people from all over the country would participate, it would be required to first swiftly ramp down the generation all over the country around the  9 PM mark, then maintain it at a reduced level for the duration of the lights-out, then again swiftly ramp it up around the 9:09 PM mark. This report shows some of the things which went on behind the scenes to prepare for and handle the event.

Comments

Popular posts from this blog

Water From Air

Demand Side Management and Energy Efficiency

Taking Charge of the Iron Horse