Power generation from renewable energy plants is intermittent, that is, variable in nature as it depends significantly on weather conditions. The generation of electricity varies immensely due to changes in wind speeds and solar irradiation. This variability becomes a concern for grid operators, utilities and market participants, especially with the rapidly increasing share of renewable energy in the total energy mix of a country.
To this end, the prediction of electricity generation from renewable energy plants becomes pertinent. This is mostly done by making forecasts based on models that incorporate weather conditions. This exercise is cumbersome, but not futile. Wind and solar power forecasts provide valuable information and insights that assist in planning by various stakeholders in the energy industry. The information is used for unit despatch or sale of renewable energy to energy markets. Over the past two decades, several countries have gained significant experience in ways to integrate variable renewable energy into electrical grids and energy markets efficiently.
Need for forecasting
The need for forecasting arises when renewable energy plants have been set up and energy generated from these plants needs to be either integrated with the power grid or sold in the energy market. Generation forecasts are thus needed in order to know in advance the amount of power that wind or solar plants will feed into the grid over the next few hours or days. In a way, for different stakeholders, forecasts are an expectation of the generation from intermittent renewable energy plants. For different requirements, several time scales of forecasts are used. One, medium-term forecasts for the following 2 to 20 days; two, short-term forecasts for the next 6 to 48 hours; and three, shortest-term forecasts for the next minute to the next six hours.
By and large, forecasts are used by wind and solar plant developers to schedule maintenance. Also, owners of rooftop solar plants need forecasting to raise the share of internal consumption of the energy produced from the rooftop system.
Examples of best practices in Europe
A study, titled “Variable Renewable Energy Forecasting – Integration into Electricity Grids and Markets – A Best Practice Guide”, published by GIZ, gives an overview of the forecasting techniques for wind and solar generation and shares experiences of countries with different regulatory frameworks and market environments. Renewable Watch provides an extract from this study of the regulatory history in Germany, Denmark and Spain and how it impacted the need for renewable energy forecasting…
In 1996, the European Parliament and the Council of the European Union proposed to liberalise the electricity markets in member countries to remove market access barriers in the energy sector. Until then, customers had no option but to purchase electricity from the regional electricity supplier and electricity prices used to be controlled. After two years, in 1998, Germany liberalised its energy market, which significantly changed the dynamics of the energy sector in the country. At present, Germany has a vibrant competitive energy market that ultimately benefits end consumers. In 2020, the Renewable Energy Sources Act in Germany was introduced, which guaranteed feed-in tariffs (FiTs) for power procured from renewable energy plants. These developments helped attract high investments in the country’s renewable energy space.
The advent of energy markets and the substantial increase in the renewable energy penetration made renewable energy forecasting necessary in Germany. In the new energy market scenario, both transmission system operators (TSOs) and electricity traders sold renewable energy in the day-ahead and intra-day market and required accurate day-ahead forecasts to submit their trading schedules.
To this end, the German government introduced a bonus system for TSOs to make accurate forecasts, which used economic incentives in the form of a market premium to encourage transition to the energy market option. Moreover, the costs of balancing energy were reimbursed through a surcharge paid by power consumers. The traders’ incentive for using accurate forecasts in the energy market was mainly to minimise the payment for balancing power. In this way, a favourable market environment was created wherein a large portfolio, with more regionally distributed assets, led to better prediction of renewable energy. This, in turn, helped in better balancing of power and thus increasing profits. Traders now competed to increase the portfolio of renewable energy, which would further help in better forecasts. Therefore, the benefits of having accurate forecasts motivated traders in the energy market to improve the quality of generation forecasts from renewable energy plants.
The electricity market design of Spain is similar to the one prevalent in Germany. As in Germany, the Spanish government wanted to promote renewable energy sources. To this end, the Spanish government also implemented an economic incentive to make this transition from fixed FiTs to a market-dependent premium scheme, although in 2014, the FiT and the premium scheme were phased out.
Nevertheless, similar to energy market dynamics in Germany, participation in the market meant that the Spanish traders had to accurately forecast generation from renewable energy plants. Penalties were applied if the actual generation deviated from the forecast. Thus, penalties and market-based profits led to the need to calculate accurate renewable energy forecasts.
The electricity market design of Denmark is quite similar to that in Germany. In Denmark, renewable energy can be sold in the electricity market at the market price with a price supplement. Thus, accurate forecasts were required for this promotion scheme of renewable energy with price supplements on top of the market price. Moreover, in Denmark, the need for accurate forecasts was driven by the possibility of using wind power plants to assist in the balancing of the power market.
Application of renewable energy forecasts
In Europe, renewable energy forecasts have several applications due to high liberalisation of the electricity market. With liberalisation, it has become important for different stakeholders in the energy value chain to have renewable energy forecasts. For instance, in Germany, TSOs use forecasts not only for trading purposes but also for gaining information of the grid and for doing security analysis. For these reasons, forecasts are calculated not only at the regional level, but also at the federal and the entire country level.
Since variable renewable energy generation has a strong influence on the spot market price, national forecasts are also used as a spot market price indicator. Thus, countrywide aggregate renewable energy forecasts are used by electricity traders, even those that do not have a renewable energy portfolio. Energy forecasts are also used by speculators that need to determine the best time to buy and sell energy, in order to make short-term profits.
Renewable energy forecasts are also used for load flow calculations at the grid level, driven by the high integration of renewable energy into the distribution grid. The transition from a one-directional electricity flow in the grid to a bidirectional one in Germany has further increased the importance of accurate forecasts. This change has, to a certain extent, changed the way distribution system operators as well as TSOs look at their responsibilities. This is so because now huge loads of electricity from renewable energy plants are to be handled at transformer stations and grid capacity has to be monitored consistently. Moreover, forecasts are not only needed for single countries, but also to predict the import or export of electricity from one country to another.
The way forward
The cases of Germany, Denmark and Spain show that different stakeholders in the energy sector needed renewable energy forecasts after the liberalisation of electricity markets and much before large-scale penetration of renewables into the grid. In many countries of the world, this has not been the case. The share of renewable energy in the total energy mix is increasing, while forecasting and scheduling norms are still absent or inadequate. It has often been observed that policymakers realise the need for forecasting and scheduling of renewable energy only when a large capacity of it has been added to the grid and integration issues start cropping up.
Therefore, it is important to set clear and well-defined regulations and policies on forecasting, and collect accurate data on weather conditions and renewable energy generation before it becomes difficult to manage the increasing share of renewable energy capacity. In this respect, the Indian power market seems to be at a crossroads as it is simultaneously working to promote renewable energy markets with ambitious targets and mega auctions and building a strong regulatory environment of forecasting and scheduling of renewable energy.