Three Implementation Phases

The need for energy systems analysis tools depends on the share of renewable energy in the system. The following three phases of implementing renewable energy technologies can be defined.

The introduction phase: This phase represents a situation in which no or only a small share of renewable energy is present in the existing energy system. The phase is characterized by marginal proposals for the introduction of renewable energy—for example, wind turbines integrated into a system without or with only a small share of wind power. The system will respond in the same way during all hours of the year, and the technical influence of the integration on the system is easy to identify in terms of saved fuel on an annual basis.

The large-scale integration phase: This phase represents a situation in which a large share of renewable energy already exists in the system—for example, when more wind turbines are added to a system that already has a large share of wind power. In this phase, further increases in renewable energy will have an influence on the system, which will vary from one hour to another, depending, for example, on whether a heat storage is full or whether the electricity demand is high or low during the given hour. The influence of wind power integration on the system, and thereby the calculation of the fuel saved on an annual basis, becomes complex and requires a proper calculation with hourly simulation models.

The 100 percent renewable energy phase: This phase represents a situation in which the energy system has been or is being transformed into a system based 100 percent on renewable energy. The system is characterized by the fact that new investments in renewable energy must be compared not to nuclear or fossil fuels but to other sorts of renewable energy system technologies. These technologies include conservation, efficiency improvements, as well as storage and conversion technologies—for example, wind turbines introduced to replace the need for biomass resources. The influence on the system is complex not only in terms of differences from one hour to another but also with regard to the identification of a suitable combination of changes in conversion and storage technologies.

The definitions of these three implementation phases can be used in the selection and design of proper tools for the technical analyses. In the first phase, the technical calculations are rather simple and do not require complex models. Typically, annual fuel savings can be calculated without models or by using simple models based on duration curves or similar data. However, in the next phase, it becomes essential to make hour-by-hour calculations due to the fluctuations in most renewable energy sources. In the third phase, it also becomes essential to include proper analyses of advanced conversion and storage technologies in the system.

Different Types of Energy System Analysis Models

On a global scale, a large number of different computer models exist that can all be called energy system analysis models in the sense that they make calculations related to the analysis of energy systems. Table 4.1 lists some of the models. In general, all models address the implementation of renewable energy sources or other technologies related to renewable energy systems, such as, for example, CHP.

Table 4.1 Energy System Analysis Models

With regard to the idea of Choice Awareness and the introduction of the three implementation phases just described, important differences can be found among the models. One important difference is whether the model makes a detailed hour-by-hour simulation or is based on aggregated annual calculations, possibly made by using duration curves or similar data. Another important difference is whether the model addresses the national or regional system level or the project or single station level. In Table 4.2, the models are grouped according to these essential differences.

Table 4.2 Grouping of Energy System Analysis Models

Not all models are designed in such a way that they fit perfectly into such groupings. Consequently, one model is located in two groups, and other models may have aspects of more than one group as well. The MESAP model is not shown in Table 4.2 because it is really a database system that combines the use of other models. Moreover, some models are being further developed on a continuous basis. Typically, models based on aggregated annual calculations are added the possibilities of hourly simulations in different parts of the calculation process. Nevertheless, Table 4.2 shows two important differences.

As described with regard to the three implementation phases, models based on aggregated data are typically suitable for the introduction phase. They can provide the needed information in a proper way, and in this phase it is not necessary to run detailed hourly simulations, which is often a complicated and data-consuming process. Moreover, models based on aggregated data typically have the advantage of being easy to document and communicate compared to hourly simulation models. However, for the analysis of large-scale integration or 100 percent renewable energy systems, hour-by-hour simulation models are essential.

In the division between levels of operation, the models at the project/station level typically cannot evaluate the influence of, for example, fluctuating renewable energy sources on the overall regional and/or national system. On the other hand, they are typically better equipped for making detailed analyses of the business-economic operation and design of single stations. Consequently, in the groupings in Table 4.2, only detailed hour-by-hour simulation models at the regional/national level can fulfill the requirements for energy system analyses of renewable energy systems in the large-scale and 100 percent renewable energy implementation phases.

It should be emphasized, however, that the design of alternatives by use of such models in many cases may benefit from being combined with either models based on aggregated data or models at the project/station level. For example, analyses made with the EnergyPLAN model have, in some cases, been combined with analyses conducted by one of the “sister models”: energyPRO and EnergyBALANCE (Lund et al. 2004). The energyPRO model has been used to add analyses of how single stations will respond to changes in the energy system design defined by use of the EnergyPLAN model. The EnergyBALANCE model has been used to make fast and easy first estimates of different large-scale options, before involving the EnergyPLAN model.

Hourly Simulation Models at the National Level

Historically, energy system analysis models based on hourly simulations at the regional/national level have typically been developed with two purposes in mind. Some models have been developed to optimize the load dispatch between single power stations in utility systems, whereas others have been made for planning purposes—that is, identifying proper investment strategies.

The overall aim of the load dispatch models is to design suitable operation strategies on a day-to-day basis. Such models have, prior to the introduction of national and international electricity markets, been used and designed by electricity system operators to plan least-cost production strategies of electricity supply systems with several production units. Because the models must be able to calculate exact operational costs and emissions, they are typically very comprehensive and detailed in their description of each individual power station.

The overall aim of planning models is to identify suitable future investment strategies. Such models have been used and designed by public planning authorities, utility companies, and different NGOs, including universities and research institutions. Sometimes load dispatch models have been used for planning purposes, or planning models have been developed on the basis of load dispatch models. In practice, such models have sometimes proven to be very conservative in the sense that they have mainly been able to analyze small, short-term adjustments to the existing system rather than radical changes in the overall system design and regulation. Moreover, the use of such models is typically very time-consuming because of the need for detailed data.

Both load dispatch models and planning models have been influenced by the introduction of international electricity markets, which has taken place in many countries around the world since the late 1990s. Operation strategies for power production units are now determined by the market, and the models are used to identify optimal market behavior. Planning models have been facing the need to include the modeling of international electricity markets in the analyses. Some models have chosen to convert to solely making simulations of the results of market operations based on the present institutional setup of the international electricity market, while, for example, the EnergyPLAN model has been equipped with the ability to make both market analyses as well as simulations on the basis of pure technical optimizations.

Consequently, hour-by-hour simulation models at the regional/national level have important differences. One is whether the present electricity market structure forms the only institutional basis for the simulation and/or operational optimization procedure of the model or if this simulation is also based on technical and/or economic optimization. In the latter case, it is also important to define whether this optimization is made on the basis of a business-economic market strategy or some sort of socioeconomic least-cost strategy.

Another important difference is whether the model combines all sectors of a regional/national energy system or includes only parts of this system, such as, for example, the electricity supply. This difference is important to the analysis of large-scale integration and especially 100 percent renewable energy systems. These systems will benefit from efficiency measures such as CHP production, which makes an integrated analysis of the electricity and the heating sector relevant. They will also benefit from the use of electricity for transport purposes, which makes combined electricity and transport sector analyses essential.

In Table 4.3, the hour-by-hour simulation models applied to the regional/national level have been grouped according to their optimization and their operation levels. Again, such grouping is not 100 percent accurate for all models. The LEAP model is left out. It is characterized as a simulation model rather than an optimization model because other models optimize the operation of the system. In principle, the two models SIVAEL and Balmorel do not include all sectors, as individual heating and transportation are left out. However, they do comprise the aspect of electricity grid involvement in the case of micro-CHP or electric vehicles.

Table 4.3 National Level Hour-by-Hour Simulation Models

In Table 4.3, the EnergyPLAN model is placed in both the technoeconomic optimization and the electricity market optimization model groups. The model is designed for calculating the consequences of both types of optimization strategies. Moreover, the calculation of the market-economic strategy is based on a business-economic optimization of individual stations in which different taxes and subsidies can be specified. The aim is to be able to provide information for the discussion on proper public regulation measures.

It must be emphasized that the models have differences other than those presented in Tables 4.2 and 4.3. Models such as LEAP and RAMSES, for example, emphasize the analyses of scenarios and include the calculation of series of years.

Purpose and Application

The main purpose of the model is to assist the design of national or regional energy planning strategies on the basis of technical and economic analyses of the consequences of implementing different energy systems and investments. The model encompasses the whole national or regional energy system, including heat and electricity supplies as well as the transportation and industrial sectors. With regard to electricity supply, the model emphasizes the analysis of different regulation strategies with a focus on the interaction between CHP and fluctuating renewable energy sources.

The model is a deterministic input/output model. General inputs are demands, renewable energy sources, energy station capacities, costs, and a number of optional different regulation strategies emphasizing import/export and excess electricity production. Outputs are energy balances and resulting annual production, fuel consumption, import/exports of electricity, and total costs, including income from the exchange of electricity (see Figure 4.2).

Figure 4.2 Input-output structure of the EnergyPLAN model.

EnergyPLAN is an input/output energy systems analysis model. Compared to other similar models, the following characteristics of EnergyPLAN can be highlighted:

• • EnergyPLAN is a deterministic model as opposed to a stochastic model or models using Monte Carlo methods. With the same input, it will always come to the same results. However, as we shall see in Chapter 5, the model can perform a calculation on the basis of RES data of a stochastic and intermittent nature and still provide system results that are valid for future RES data inputs.
• • EnergyPLAN is an hour-simulation model as opposed to a model based on aggregated annual demands and production. Consequently, the model can analyze the influence of fluctuating RES on the system as well as weekly and seasonal differences in electricity and heat demands and water inputs to large hydropower systems.
• • EnergyPLAN is aggregated in its system description as opposed to models in which each individual station and component is described. For example, in EnergyPLAN, the district heating systems are aggregated and defined as three principal groups.
• • EnergyPLAN optimizes the operation of a given system as opposed to models that optimize investments in the system. However, by analyzing different systems (investments), the model can be used for identifying feasible investments, as we shall see in Chapters 5 and 6.
• • EnergyPLAN provides a choice between different regulations strategies for a given system as opposed to models into which a specific institutional framework (such as the Nord Pool electricity market) is incorporated.
• • EnergyPLAN analyzes 1 year in steps of 1 hour as opposed to scenario models analyzing a series of years. However, several analyses each covering 1 year may of course be combined into scenarios.
• • EnergyPLAN is based on analytical programming as opposed to iterations, dynamic programming, or advanced mathematical tools. This makes the calculations direct and the model very fast when performing calculations. In the programming, any procedures that would increase the calculation time have been avoided, and the computation of 1 year requires only a few seconds on a normal computer, even in the case of complicated national energy systems.

Energy Systems Analysis Structure

The model can be used to calculate the consequences of operating a given energy system in such a way that it meets the set of energy demands of a given year. Different operation strategies can be analyzed. Basically, the model distinguishes between technical regulation—that is, identifying the least-fuel-consuming solution—and market-economic regulation—that is, identifying the consequences of operating each station on the electricity market with the aim of optimizing the business-economic profit. In both situations, most technologies can be actively involved in the regulation. And in both situations, the total costs of the systems can be calculated. In the documentation of the model, a list of energy demands is presented as well as an overview of all components of the model. A short description of how they are operated in relation to the two different regulation strategies is also presented together with a list of the main inputs for each component.

The model includes a large number of traditional technologies, such as power stations, CHP, and boilers, as well as energy conversion and technologies used in renewable energy systems, such as heat pumps; electrolyzers; and heat, electricity, and hydrogen storage technologies, including Compressed Air Energy Storage (CAES). The model can also include a number of alternative vehicles—for instance, sophisticated technologies such as V2G (vehicle to grid), in which vehicles supply the electric grid. Moreover, the model includes various renewable energy sources, such as solar thermal and photovoltaic (PV), wind, wave, and hydropower.

Figure 4.3 shows the procedure of the energy system analysis. As a first step, calculations are based on a small computation, which is made simultaneously with the typing of input data in the input and cost tab sheets. The next step consists of a series of initial calculations that do not involve electricity balancing. Then the procedure is divided into either a technical or a market-economic optimization. The user chooses which one to apply. However, each calculation lasts only a few seconds, and, consequently, it is possible to make both calculations, one after another. The technical optimization minimizes the import/export of electricity and seeks to identify the least-fuel-consuming solution. On the other hand, the market-economic optimization identifies the least-cost solution on the basis of the business-economic costs of each production unit. In both situations, the model can calculate the socioeconomic consequences that provide important information for the design of different public regulation measures.

Figure 4.3 Overall structure of the energy system analysis procedures.

The principle of the energy system of the EnergyPLAN model is shown in Figure 4.4. In the EnergyPLAN model, an aggregated analysis is made of the many individual stations that together form a regional or national energy system. Thus, all boilers, CHP stations, and so forth that produce heat for district heating are grouped into three district heating systems. Moreover, the EnergyPLAN model includes a series of optional renewable energy sources as well as a large number of conversion and storage technologies. In this way, the model can make comprehensive analyses of rather complex 100 percent renewable energy systems without the need for an enormous quantity of detailed data.

Figure 4.4 Overall sketch of the energy system described in the EnergyPLAN model.

Energy System Analysis Methodology

A short outline of how to use the model in the design of energy alternatives at the national level (Lund, Andersen, and Antonoff 2007) is presented here. On the home page of the model, www.EnergyPLAN.eu, one can download a whole set of exercises and assignments, including detailed answers, that together constitute a comprehensive set of guidelines on how to use the model. As already shown in Table 4.3, the model can be used for different kinds of energy system analyses.

• Technical analysis: Design and analysis of complex energy systems at the national or regional level and according to different technical regulation strategies. In this analysis, input is a description of energy demands, production capacities and efficiencies, and energy sources. Output consists of annual energy balances, fuel consumptions, and CO₂ emissions.
• Market exchange analysis: Further analysis of trade and exchange on international electricity markets. In this case, the model needs further input to identify market prices and to determine the response of these prices to changes in import and export. Input is also needed to determine marginal production costs of the individual electricity-producing unit. The modeling is based on the fundamental assumption that each station optimizes according to business-economic profits, including any taxes and CO₂ emissions costs.
• Feasibility studies: Calculation of feasibility in terms of total annual costs of the system according to different designs and regulation strategies. In this case, inputs such as investment costs and fixed operation and maintenance costs must be added together with lifetime periods and an interest rate. The model determines the socioeconomic consequences (taxes and subsidies are not included) of the productions. The costs are divided into fuel costs, variable operation costs, investment costs, fixed operation costs, electricity exchange costs and benefits, and possible CO₂ payments.

A Step-by-Step Approach to National Energy Systems Analysis

The approach to energy systems analysis used in the EnergyPLAN model can be divided into four steps:

• Step 1: Defining reference energy demands
• Step 2: Defining a reference energy supply system
• Step 3: Defining the regulation of the energy supply system
• Step 4: Defining alternatives

Sister Models to EnergyPLAN

The EnergyPLAN computer model has three sister models (sister models in the sense that they originate from Aalborg University and have been made to supplement and support one another): EnergyBALANCE, energyPRO, and COMPOSE. The EnergyBALANCE model is a simple spreadsheet model based on aggregated annual calculations of energy balances. The model is designed for an easy integration of typical inputs from national statistics. The model adds different data of efficiencies, and so forth, to make overall analyses of changes in demand and supply technologies. The model is now made available by the Danish Organisation for Renewable Energy (OVE) on its home page: www.orgve.dk.

The energyPRO model excels in modeling and optimizing the operation characteristics of a single station. It has the ability to evaluate many different types of technologies and performance criteria for generating units, particularly CHP stations, and enables the user to add detailed definitions of parameters, such as heat production, electricity production, fuel costs, power curves, and control strategies. It can also conduct a sophisticated economic analysis that considers varying values for revenues (e.g., heat prices and spot market electricity prices) and costs (e.g., fuel costs, taxes, and other operational expenses). However, this level of detail may also require a significant amount of research and data input to properly initialize the model, and it implies a high level of understanding of the specific performance characteristics of the station.

The model focuses primarily on production aspects and, apart from a few exceptions such as heat distribution losses, it does not consider how the station fits into the broader energy system. The model is an advanced computer tool for the design and operation of CHP stations, and it has been used to design most of the existing small CHP stations in Denmark. The initial version of energyPRO was designed in the late 1980s. Shortly thereafter, the program was made commercially available by the software company Energy and environMental Data (EMD). Based on ongoing dialogue with users, EMD has refined and added new facilities and features to the model on a continuous basis. It has become a widely used software package for the analysis of local energy stations based on gas engines, gas turbines, and steam turbines burning both waste and wood chips, as well as stations based on boilers only. EnergyPRO has been used in Lund and Andersen (2005) and Andersen and Lund (2007).

COMPOSE (Compare Options for Sustainable Energy), designed by Morten Blarke, is a technoeconomic energy project assessment model. COMPOSE enables the evaluation of user-defined sustainable energy projects in user-defined energy systems and includes user-selected methodology options. The mission of COMPOSE is to combine the strength of energy project operational simulation models with the strength of energy system scenario models to create a modeling framework that supports an increasingly realistic and qualified comparative assessment of sustainable energy options.

The current functionality of COMPOSE focuses on the modeling of framework design. The model currently enables the user to calculate the relocation coefficient of an energy project and an energy system defined by the user. User-defined uncertainties may be specified to enable extensive risk analyses—for example, specifying uncertainty ranges for wind production. Special features currently include, among others, Monte Carlo risk assessments, import of projects from energyPRO, and import of hourly distributions from EnergyPLAN.

The vision is to establish COMPOSE as a cost-benefit and cost-effectiveness toolbox for private and public decision makers. COMPOSE focuses on assessing to which degree energy projects may support intermittency, while generally offering a realistic evaluation of the distribution of costs and benefits with uncertainty. In coming releases, COMPOSE will increasingly improve the evaluation of energy projects in a project-system hybrid perspective with respect to fossil energy consumption, emissions, economic costs, financial costs, fiscal costs, employment, balance of payment, and distributional aspects of costs and benefits.