1.1 Electric power industry
It is crucial to understand the electric power industry before tackling the electric load forecasting problem. What’s the fundamental characteristic that differentiates this industry from the others? What about its history? How does it work today? This section provides a brief, non-technical, and conceptual level overview with emphasis on how electric load forecasting interacts with the industry.
1.1.1 Electricity and end use
The fundamental characteristic that makes the electric power industry unique is the product, electricity. More specifically, it is the limited storage capability of electricity that makes it distinctive. Since there is no “inventory” or “buffer” from generation to end users (customers), ideally, power systems have to be built to meet the maximum demand, the so called peak load, to insure that sufficient power can be delivered to the customers whenever they need it. In transmission and distribution systems, the capacity of any equipment has to exceed the peak demand of all the downstream customers it serves. It is due to this nature that many utilities are concerned with the forecasting quality of peak loads more than that of the off-peak ones.
When Pearl Street Station, the first central power plant in the US, opened and started generating electricity on September 4th, 1882, lighting was the sole end use for electricity. In fact, power systems were invented by Thomas Edison to promote his invention of the incandescent light bulb. Since homes and businesses were not electrically ready, in addition to building the distribution systems, the lighting companies had to set up the internal wiring. Therefore, it was straightforward to determine the system capacity by counting the light bulbs installed on the system. Later on, the end uses of electricity quickly expanded to include electric fans and electric irons, both of which were invented in the early 1880s, together with many other electric inventions. Electric motors started to replace steam engines in factories. Starting from 1888, many cities introduced electric streetcars to take the place of horse-drawn streetcars. Because of the transmission limit of direct current (DC), which was the standard of US power systems at that time, the central stations were rather small, supplying only small areas. Although the end uses for electricity were getting more and more diverse, system capacity was not a concern due to the small scale of each station’s service territory.
The alternating current (AC) power system devised by Nikola Tesla, initiated a major change to the industry. On November 16, 1896, electrical power was generated and transmitted as alternating currents to industries in Buffalo from the hydroelectric generators at the Edward Dean Adams Station at Niagara Falls. The Niagara Falls generation project was the first large-scale system to successfully supply electricity from one circuit for multiple end uses, and opened the era of large-scale AC power systems. Inventions in the 20th century, of which several representative ones are listed in Table 1-1, further enriched the end uses and stimulated the growth of electricity demand. For example, Figure 1-1 shows the US residential electricity consumption by end use in 2008.
Table 1-1 Several representative inventions in the 20th century
|Electric washing machine||1906|
|Radio stations in the US||1920s|
|Personal computer (Sphere 1)||1975|
Figure 1-1 US residential electricity consumption in 2008
Eventually, electricity demand began to be tied in with human activities through end uses – the equipment or appliance had to be installed and turned on/off according to someone’s will. In the early 20th century, economic factors, which indicated and impacted human activities nationally or even worldwide, started to be used to forecast the long-term trend (i.e. 10 years ahead) of the demand. Moreover, the increase of load due to special events, such as radio broadcasts “of a night ball game, of a prize fight, and of the President’s speech to the Congress”, which affected local human activities, was also discovered by pioneering power engineers. On the other hand, the electricity demand of certain appliances was observed to be closely tied to meteorological conditions. For instance, it was discovered in the 1940s that “the increase of load during the summer as the temperature increases is the result of increasing operations of fans, refrigerators, air conditioning and other cooling devices”. While the sales of room air conditioners escalated from 74,000 in 1948 to 1,045,000 in 1953, a major effort in the field of electric load forecasting was to investigate the impact of weather on electricity demand.
As the electricity end uses expand from lighting only to multiple end uses (i.e., Figure 1-1), electric load forecasting has gradually become more complicated. For instance, with hundreds or even thousands of messages, talk shows, and comedies broadcasted through radio stations, TV stations, and the internet, it is no longer as easy as it was in the 1940s to predict which ones could significantly affect the electricity demand and the magnitude of the affect.
1.1.2 Regulation and deregulation
The electric power industry consisted of many municipally owned utilities and privately owned multiservice utilities competing with each other locally, until Samuel Insull first set up a successful example of the business model for the industry. He consolidated his company with others, so that he could sell electricity to a broader market using bigger and more efficient generators. Moreover, he also sought new customers to diversify the pattern of electricity consumption and fill in more off-peak loads. By 1907, his company Commonwealth Edison was already known as one of the most progressive and lowest cost utilities in the world. As a result, Insull's strategies became the strategies of the electric power industry in the United States. Small utilities gradually merged with or purchased electricity from large, vertically integrated, private multiservice utilities. To further exploit the benefits of large and efficient equipment, private electric utility ownership also consolidated into large utility holding companies.
As the service territories of the large utilities started to grow beyond the city boundaries, and even across state lines, state regulation and Federal involvement were brought to the electric power industry. Federal and state regulation recognized electric utilities as natural monopolies, which allowed them to grow with little competition to sell electricity to broader segments of the market, and to charge rates that were reasonable and yet sufficiently high to keep the utilities financially healthy. On the other hand, the utilities were obligated to provide service, with as few interruptions as possible and without discrimination, to all customers who sought to buy power.
Abuses by holding companies appeared in the 1920s, which eventually led to Federal intervention. In the 1930s, the early years of the Great Depression, the Federal Government became a regulator of private utilities and a major producer of electricity, and oversaw the holding companies in order to avoid financial abuses. The electric utility industry was still a host of natural monopolies. The benefits and regulation, together with technological development, helped keep the rapid and healthy growth of the industry through the 1960s.
Several challenges such as environmental concerns, high inflation, and increased fossil-fuel prices which resulted in the energy crisis of the 1970s appeared, and led to the introduction of the electric power industry deregulation. In 1978, the US Congress passed The Public Utility Regulatory Policies Act (PURPA), a law meant to promote greater use of renewable energy by opening the wholesale markets to nonutility electricity producers, primarily private businesses and industrial manufacturers that generated their own electricity. Figure 1-2 shows how the share of electricity generation varied over time. In 1970, the electric utilities produced 93% of the total generation. The share steadily increased to 97% in 1979. Due to the growth of nonutilities in the 1980s, this number declined to 91% in 1991. In 1992, US congress passed the Energy Policy Act (EPACT92) that further promoted competition in the bulk power market.
Figure 1-2 Share of electricity generation in the US
With the hope of economic benefits and price reductions for consumers, the government regulation of the electric power industry started to be replaced by competition. Passages of PURPA and EPACT92 initiated the deregulation of the electric power industry from a regulated monopoly with regulated rates to a competitive market of independent electricity producers and distributors. Deregulation not only changed the structure of the electric power industry, but also had huge impacts in many other aspects, which have led to the reasons of “why” and “how” we should perform electric load forecasting nowadays.
1.1.3 Implications of deregulation on electric load forecasting
One role of regulation was to ensure that the utilities could raise enough capital for building infrastructure so that the projected load could be met. Sometimes, the customers had to pay higher rates than in previous periods to maintain the financial health of the utilities. The legitimate status of a natural monopoly allowed the utilities to grow, to use larger and more efficient equipment, and to invest in technological development, which further reduced the unit cost of electricity. As a result, the retail price of electricity was decreasing prior to the energy crisis of the 1970s, as shown in Figure 1-3. Hearings to approve rate increases, so called “rate cases,” were initially not an issue for utilities. This changed with deregulation. Although the industry is deregulated, to avoid financial abuses, the utilities are still being watched by the government. It has the authority to approve or deny requests in rate cases. To favor the voters, of which most are the customers of utilities, politicians have the tendency to reduce the rates rather than the counterpart. This makes the defense of rate cases a more and more challenging for utilities. Since it is the “projected load” that reflects the future income and potential costs of a utility and thus determines the rate, a formal load forecasting process becomes a necessity for a utility to succeed in a rate case.
Figure 1-3 Average retail prices of electricity
Uncertainty and the difficulty of defending rate cases, together with many other factors derived by deregulation, changed the operating strategy of the industry from “build and grow” to “wait and hold.” Beginning in the late 1980s, utilities slowed down infrastructure development and only performed system expansion or equipment maintenance and upgrade when they had to. In the short run, (i.e., one to three years), this new strategy helped them achieve better financial benefits than the “build and grow” strategy would have done. Meanwhile, the load/capacity ratio was being pushed higher and higher to accommodate the increasing electricity demand. After a decade of consistently executing the new strategy, the equipment was operating near its limit, which further required the utilities to deploy a formal forecasting process to help justify decisions about operations, maintenance, and planning.
Deregulation creates a competitive market for parties to buy and sell electricity. In order to make proper decisions for energy purchases, utilities have to have a good idea of their future electricity demand and price. Failing to do so would result in excess costs in the market. While the market price is not transparent to the end users in real-time, utilities would have to be responsible for the costs, which causes the potential for them to be inefficient due to bad forecasts.
Counter-intuitively, regulatory commissions become more active in the deregulated industry than before. Energy efficiency is encouraged by the regulators through demand response programs, which are mandated in many states. In the competitive market, electricity prices rise as the demand goes up. On the other hand, as the system approaches its limits, outages are more likely to occur during the high demand periods. Demand response programs are set up to help utilities manage system peaks as well as to avoid outages. Since power systems are generally sized corresponding to the peak demand, reducing peak demand could also result in savings in plant and capital costs. Nevertheless, none of these benefits could be realized without fairly accurate load forecasts indicating when the peak load would occur and how much the peak load would be.
Deregulation promotes the use of alternative energy resources, such as solar, wind, etc. With more and more solar panels and wind turbines interconnected to the grid and being treated as negative loads, the load forecasting problem becomes even harder, because the production of these energy resources is highly dependent upon weather, such as cloud cover, wind speed and direction, etc., which cannot be easily predicted.