Energetic and Economic Assessment of a Solar Thermally Driven Innovative Tri-Generation Unit for Different Use Cases and Climates

Abstract

The energy sector is currently under enormous transition, moving from fossil fuels to renewable energies and integrating energy efficiency measures. This transition can hold opportunities for new and innovative energy systems. This study presents an energetic and economic assessment of an innovative tri-generation unit working with a two-phase thermodynamic cycle. The tri-generation unit is driven by heat and is capable of providing heat at lower level, cold, and electricity to end users. The use cases—residential, day-use offices, commercial retail, and manufacturing industry—are integrated in a dynamic simulation model, indicating the operation mode of the unit. The results show that the tri-generation unit is able to provide heat and cold with an Energy Utilization Factor of 35% to 68%, depending on the use case. Solar thermal has a limited to potential to supply the unit with heat, due to the high temperature of 180 °C and the required unit operation at nighttime. The economic comparison indicates that the driving heat must be as low as possible and that savings through self-consumption is most relevant.

1. Introduction

The climate target of not exceeding 1.5 °C compared to the pre-industrial climate was already exceeded in 2024 [1], resulting in more frequent weather extremes and calling for a rapid transition in our energy production and consumption. The EU has made this transition legally binding: climate neutrality by 2050 and at least −55% net greenhouse gas cuts by 2030 under the European Climate Law and Green Deal [2]. The main challenge is the more efficient use of energy and upscaling renewables. The third version of the Renewable Energy Directive (RED III) raises the EU’s renewables share to a minimum of 42.5% in 2030. Additionally, the recast Energy Efficiency Directive (EED) sets a binding 11.7% reduction in EU energy consumption vs. the 2020 reference scenario by 2030 [3].

Variable renewable energy production brings challenges to the electricity and energy systems. The EU is therefore prioritizing “efficiency-first”, high-efficiency cogeneration, and efficient district heating and cooling into policy, and is setting milestones for integration of renewable energies and waste heat through 2050. These frameworks reward solutions that deliver multiple energy vectors with primary energy savings compared to separate generation [4]. Innovative multi-generation concepts, like tri-generation, also referred to as Combined Cooling, Heating, and Power (CCHP), that simultaneously supplies electricity, heat, and cooling from a single energy driver, match this need by coupling sectors and balance renewable variability at the point of use.

The concept of CCHP units or systems is already presented and discussed in different forms. The typical CCHP concept is based on the combination of a CHP unit coupled with a thermally driven absorption chiller [5]. The CHP unit predominantly generates heat, with an estimated electrical efficiency between 20% and 40% [6]. The surplus heat then can be used to drive a thermally driven absorption chiller, characterized by an estimated thermal COP of 0.6 to 0.8. The electricity generation can be increased if the generated heat is further used in an Organic Rankine Cycle (ORC) unit. However, the electrical efficiency of an ORC unit driven with temperature below 200 °C is relatively low, with ca. 10% up to 25% [7]. Thus, the potential of additional power generation, especially for smaller units is very limited. Additionally, those CHP-based concepts of CCHP need a driving fuel, such as natural gas, biogas, or synthetic fuels. Heat-driven solutions have the benefit that they require fewer steps and energy transfers to provide the driving energy for CCHP compared to fuel-driven technologies.

Dupuy et al. identify three types of heat-driven systems [5]:

• Single-generation: 1 Output.
• Cogeneration: 2 Outputs.
• Tri-generation: 3 Outputs.

Any CCHP unit relies on individual components, such as generators for electricity supply and heat exchangers for heat and cold extraction. Based on the cold generation typology, components such as an absorber and a generator for the sorption process, or a compressor for the vapour compression cycle, are required.

Unlike a CCHP system with multiple units involved, a tri-generation unit is a single unit that is able to provide both heating, cooling, and electricity. The tri-generation cycle developed and patented by Stefano Briola follows such an approach [8]. The patented set-up of the tri-generation unit is presented in Figure 1a. No thermally driven sorption cooling is involved in the tri-generation concept; instead, a vapour compression cycle is used. Additionally, the tri-generation cycle uses only one refrigerant that is capable to absorb thermal energy at high temperature (driving energy), extract heat for the user, drive generators through expansion work, and extract cold at low temperature. A fundamental concept of the process design is the isothermal heat transfer in the two-phase area. In this area, the refrigerant is neither fully liquid nor fully gaseous. The thermodynamic cycle is presented in Figure 1b in a Temperature-Entropy (T-S) diagram. The cycle can be divided into an upper sub cycle (steps 1–8) and a lower sub cycle (steps 1–12). However, operating in the two-phase area poses particular challenges to the performance of both compressors and expanders. First investigations on how to tackle these challenges already have been made and presented to the scientific community [9,10].

This study will refer to a total of 13 and 15 states of the tri-generation unit, respectively, as presented in Figure 1. The change in refrigerant characteristics between each state is referred to as step.

Table 1. Description of the single steps of the tri-generation process.

Step	Description
1–2	Compression; rise in temperature, pressure, and entropy
2–2*	Heat transfer, heat absorption; isothermal, rise in entropy
2*–3	Compression; rise in temperature, pressure, and entropy
3–4	Heat absorption from ext. heat source; superheating, rise in temperature, and entropy
4–5	Expansion (power generation via generator); reduction in temperature and pressure, rise in entropy
5–5*	Heat transfer, heat disposal; reduction in temperature and entropy
5*–6	Expansion (power generation via generator); reduction in temperature and pressure, rise in entropy
6–7	Condenser (heat supply to user); isothermal heat rejection, entropy loss
7–8	Expansion (power generation via generator); reduction in temperature and pressure, rise in entropy
8–13	Mix with state 12 to state 13
1–9	Heat rejection to ambient; isothermal loss of heat, entropy reduction
9–10	Expansion; reduction in temperature and pressure, rise in entropy
10–11	Evaporation (cold supply to user); isothermal heat absorption, rise in entropy
11–12	Compression; rise in temperature, pressure, and entropy
12–13	Mix with state 8 to state 13
13–1	Heat rejection to ambient; isothermal loss of heat, entropy reduction

briefly describes each step of the tri-generation thermodynamic cycle. The mentioned steps will further be referred to throughout this article. More details on the structure and concept of the thermodynamic cycle can be found in the publications of Briola [8,12].

Another challenge in the set-up of this tri-generation unit is the selection of a suitable refrigerant, being in line with the latest legal developments regarding the Global Warming Potential (GWP) of refrigerants. A detailed screening of about 60,000 structures revealed that the refrigerant R1233zd(E) is most promising, considering thermodynamic, process-related, constructional, safety, and environment constraints [13]. In addition, the refrigerant must also work with the lubricating oil, which directly has an influence of the performance of the compressors and expanders. The actual thermodynamic and fluid dynamic properties of refrigerant–lubricating oil mixture are very difficult to estimate and hold a wide range of uncertainties. An investigation on the performance of the selected refrigerant R1233zd(E) mixed with a Polyolester oil (POE) lubricating oil showed that enthalpy and entropy of the working fluid are influenced, given a modified characteristic of density and viscosity [14].

This study evaluates the dynamic performance of a novel tri-generation unit based on a two-phase thermodynamic cycle. The tri-generation unit is assessed including heat supply from concentrating solar thermal and analyzed in different climates. Building on the recent EU research project (Regen-By-2), two-phase expanders/compressors enable efficient conversion of renewable or waste heat into electricity, useful low-temperature heat, and chilled water in one plant [11]. The study tests its economic and energetic performance across representative use cases, within residential, day-used offices, commercial retail, and industrial manufacturing in different climates in Europe. The target is to identify where multi-vector supply can outcompete conventional single-generation systems and lower the use of GHG-emitting fossil fuels. Although tri-generation technologies are conceptually well known, their techno-economic feasibility under broad market economy operating conditions and policy constrains remains poor documented. This research aims to fill this gap by providing case study-based investigations.

2. Materials and Methods

This study uses software-based simulations to model the tri-generation unit in a simplified way. Software-based simulations are the preferred method of investigation, as it allows analysis of the unit at low stage of the technology readiness level (TRL).

2.1. Modelling of Tri-Generation Unit

The study presents a dynamic assessment of the unit, including the impact of environment related to the heat dissipation component (step 13–1 and step 1–9 in Figure 2). The dynamic calculation model is based on the unit modelling by a previous study [15]. The dynamic model is deviated from a stationary model created in EBSILON Professional 15.0 software, which is suitable and favoured for modelling of stationary thermodynamic cycles. A performance map for dynamic behaviour of efficiency rates was developed from 1125 individual stationary calculations to cover, first of all, different boundary conditions in the ambient air temperature and its impact on the heat dissipation in the cycle, as well as different heat source and heat supply temperatures. The model applies the refrigerant Rz1233zd(E) with an isentropic efficiency in expansion components of 0.75, within a range from 0.7 to 0.88 [8,16]. The isentropic efficiency of compressors is 0.65, with the references to the literature stating efficiencies between 0.35 and 0.75 [8], and up to 0.89 [17]. Since both the expander as well as the compressor are working in the two-phase area, their performance and behaviour cannot yet be fully pictured. As this study targets primarily the dynamic representation and modelling of the tri-generation unit, these assumptions serve as a baseline. Generators and motors are pictured with an efficiency of 0.98 and 0.95, respectively.

The tri-generation unit is capable of providing heat, cold, and electric power simultaneously. It is estimated that by modifying the mass flow in the heat supply section to the user (step 6–7) and in the cold supply section to the user (step 10–11), the level of heat and cold supply can be adjusted and controlled. The dynamic model allows to alter the heat and cold supply in 25% steps (0%, 25%, 50%, 75%, 100%). The control of heat and cold supply has a direct impact on the overall performance of the thermodynamic cycle. A reduced cold supply to the user, meaning a reduced flow rate in the lower cycle, results in a lower power consumption of the compressor and thus to a higher net electricity generation of the unit. Reduced mass flow in the condenser (step 6–7) would lead to a lower heat supply to the user, but also to an increased power generation in step 7–8. Thus, the mass flow control of the cycle’s evaporator and condenser is relevant not only for heat and cold supply, but also for the overall performance of the cycle. The 0.25 steps in mass flow control allow a fundamental management and freedom of the unit’s operation.

The dynamic model reacts to the ambient temperature, which directly affects the cycle’s performance in step 13–1 and step 1–9. Since heat must be released in these two steps to the ambient, the refrigerant temperature at step 13 is set to be 5K above the ambient temperature. Thus, subsequently an increase in ambient temperature results in a higher refrigerant cooling temperature (T_cool). As this further affects the refrigerants enthalpy and the corresponding vaporization pressure, the total cycle is affected. The simulation in Figure 3 shows the impact of increased T_cool on the upper sub cycle (a) and on the lower sub cycle (b) separately. The illustrations show T_cool temperatures varying between 21 °C and 51 °C.

At a higher cooling temperature, the pressure drop in step 7–8 turns out smaller, which results in a lower power generation in the installed generator. For the lower sub cycle (Figure 3b), an increased T_cool results not only in a higher pressure lift in step 11–12, but also in a reduced potential cold supply, as step 10–11 has a decreased potential in heat absorption. This behaviour is not only modelled for the given refrigerant cooling temperatures but also calculated in 2 K steps for the given range. The results were transferred into a set of equations for dynamic modelling.

The previously developed dynamic model marks the backbone of the investigation of this study, as it is the first and only known approach of dynamically modelling the cycle developed by Briola 2017 [8]. The model allows a control of heat and cold supply, managed through a 0.25 stepwise alteration of the mass flow. It also pictures the performance of the thermodynamic cycle of the tri-generation unit dynamically for changing refrigeration cooling temperatures. In the following, this control of heat and cold supply via alteration of the mass flow is referred to as heat/cold supply mode.

The described dynamic model calculates the energy efficiency for each energy vector based on the driving heat energy input. Thus, the delivered energy output is always directly related to the energy input. The size of the tri-generation unit is adjusted linearly by changing the heat supply. The efficiency of electric power net output ranges between 4.4% and 14.8%. The net electric power yield strongly depends on the heat and cold supply. The heat supply efficiency (COP) for a heat supply mode of 1 is at about 38%, whereas the cold supply efficiency (EER) for a cold supply mode of 1 is at about 36%. Both COP and EER are linearly linked to the 0.25 stepwise control. Figure 4 shows the electric efficiency, COP, and EER of the tri-generation unit for different T_cool temperatures and for a heat and cold supply mode both at 0.5 and at 0.75, respectively. The COP increases with rising T_cool as the heat consumption from external source (step 3–4) reduces, while even more heat is available to exploit in the condenser (step 6–7). This behaviour is presented in Figure 3a. The combination of electric energy efficiency, COP, and EER indicates the total conversion of heat input to output of useful energy and is further referred to as the Energy Utilization Factor (EUF).

Each conversion efficiency is represented in individual equations. This set of equations is included in the dynamic simulation software TRNSYS18, which allows modelling the tri-generation in a large environment linked with other components and units. Figure 5 presents the methodology on how the heat supply mode is defined based on the given heat demand and the maximum capacity of the tri-generation unit.

The heat and cold supply modes are calculated in the dynamic simulation for each time step based on the given heat and cold demand. This means during hot season and no heat is demanded, the heat supply mode is set at 0. If heat is required, but less than the tri-generation unit can supply at maximum, the heat supply mode is calculated to the closest factor mode (0/0.25/0.5/0.75/1). Accordingly, this mode is the factor to multiply the maximum heat supply capacity of the tri-generation as the mass flow of the condenser (step 6–7) is factorized accordingly. The same control is applied to both heating and cooling. As a consequence of this control, the final heat or cold supplied from the tri-generation unit may not fit exactly to the actual demand, causing either a shortage or an excess of heat or cold. However, this study follows an approach in thermal energy balance, assuming a lossless thermal energy storage. Any energy supply is calculated by multiplying the conversion factor by the heat source input. Thus, the tri-generation unit size is based on the limitation of the heat source capacity. This heat source capacity is influenced by an initial sizing for a balanced heat supply to minimize over production or a lack of energy. The calculation of the heat/cold supply modes was then adapted to the reduced unit size.

2.2. Types of End Users

The performance of the tri-generation unit depends on the climate and the degree of heat and cold supplied to the user. Therefore, the characteristic of the user and its thermal requirements is a relevant aspect for the tri-generation unit. Four main user types have been identified and included in the dynamic simulation model. The heat and cold demand characteristics were estimated for each of the four user types.

• Industrial site: Manufacturing industry that requires mainly heat, but also cold. It is characterized by daily operation, and the heat/cold demand is independent from the ambient temperature. A batch process is applied for both heat and cold demand.
• Residential user: This use case represents typical residential users, applicable for both single family and multifamily houses. Heating and cooling demand depends on the outdoor temperature and is mainly operated during daytime. During nighttime, both demands are reduced.
• Office user: The office use case represents the typical day-use office. Thermal energy demand depends on the outdoor temperature. Heating is required at cold temperatures, whereas cooling is required for hot outdoor temperature.
• Retail and commercial user: This case represents shops and supermarkets, with heating demand based on the outdoor climate. Cooling demand is always on for the storing of fresh products, with peaks in the morning and in the afternoon for filling fresh products.

These four described user types can also be combined to represent further use case. The following four combined use cases have been identified and characterized:

• Industrial area: The industrial area combines the manufacturing industry and the office of industrial companies. The thermal energy demand of this use case is mainly driven by the constant demand of the industrial manufacturing. In addition to that, climate dependent peaks from the company’s office buildings influence the total thermal demand.
• Central urban district: This use case can be found in various city centres. It is a combination of the office and retail user types. The thermal energy demand of this use case is mainly driven by the climate, with cooling demand in hot season and heating demand in cold season.
• “Berlin Mix”: This use case combines the industrial, office, and residential user types. “Berlin Mix” is a type of urban structure named after the sector combination that can be historically found in Berlin. The energy demand in this combined use case is dependent on the climate on the one side, but also strongly affected by the demand in the industrial sector. The structure of the “Berlin mix” becomes more significant in modern city planning typologies where mixed used structures with urban manufacturing and the target of gaining new residential space becomes a key solution [18].
• Rural municipality: The rural municipality combines industrial retail user types with residential user types. It represents the combination that can be found in villages, where supermarkets and manufacturing SMEs are located close to residential areas.

The four described combined use cases have been modelled in detail and integrated in the dynamic simulation, with the individual demands of each user type separately. The energy demand for both heating and cooling depends on the daily schedule and partly on the outdoor climate. The schedule and the heat/cold supply modes are multiplied with the maximum heating and cooling load, respectively.

Table 2. Description and maximum heating and cooling load for the four investigated use cases.

	1: Industrial Site	2: Urban Central District	3: “Berlin Mix”	4: Rural Municipality
Case description	Industry:	Office:	Industry:	Industry:
	Heating: 800 kW	Heating: 500 kW	Heating: 150 kW	Heating: 600 kW
	Cooling: 800 kW	Cooling: 700 kW	Cooling: 150 kW	Cooling: 600 kW
	Office:	Retail:	Office:	Retail:
	Heating: 200 kW	Heating: 200 kW	Heating: 300 kW	Heating: 150 kW
	Cooling: 200 kW	Cooling: 400 kW	Cooling: 300 kW	Cooling: 300 kW
			Residential:	Residential:
			Heating: 300 kW	Heating: 350 kW
			Cooling: 300 kW	Cooling: 70 kW
Heating demand	4352 MWh/a	1896 MWh/a	2582 MWh/a	4435 MWh/a
Cooling demand	5143 MWh/a	1277 MWh/a	1117 MWh/a	4379 MWh/a
Dimensioned heat source capacity of the CCHP plant	2898 kW	2060 kW	2183 kW	3290 kW

presents the maximum heating and cooling loads considered for each of the four investigated use cases.

The tri-generation unit is not sized to answer the maximum heating or cooling load of a use case. Instead, it is optimized on meeting the heat demand by prevent massive over dimensioning. The tri-generation is sized for the lowest difference between heat surplus and not covered demand for an initial operation. The heat/cold supply mode calculation is then adapted, see Figure 5. As heat source, an ideal waste heat was taken into consideration, with no specification of the industrial sector.

2.3. Economic Data

To identify suitable applications for the commercial roll-out of tri-generation systems, an economic assessment is necessary. Techno-economic assessments based on simulation results are a fruitful pathway to demonstrate the financial effectiveness of innovative technologies, such as for future hydrogen pipelines [19]. For the tri-generation unit, the economic assessment is performed for all four investigated use cases. The analysis is based on a 20-year calculation model that considers cost of investment, operation, and energy costs as well as potential revenues. The specific investment cost for the Regen-By-2 plant is estimated account 3250 EUR/kW without the integration of the heat source. Annual cash flows include revenues from selling all the generated electricity and surplus of heat and cooling. Cost savings from the self-supplied heat and cooling are taken into account for the amount of heat/cold demanded. The energy prices are backed and based on recent EU statistics and regulatory benchmarks. For sold electricity and self-consumed, heat savings of 0.20 EUR/kWh are assumed, consistent with the EU non-household average electricity tariff of 0.1899 EUR/kWh in the second half of 2024 [20], and within the range of reported district heating prices [21]. Given the limited harmonization of district-cooling tariffs, the sale price for cooling is estimated at 0.12 EUR/kWh. An example for cooling sales prices is provided by the Paris cooling network at 0.10–0.13 EUR/kWh [22]. Any surplus heat and cold is sold at 0.08 EUR/kWh, representing wholesale or large-consumer procurement conditions below regulated end-user tariffs. Maintenance and repair costs are calculated with 3% of CAPEX per year. An economy-wide 3% inflation is applied as a modelling assumption.

These economic indications allow the calculation of payback period and profitability indicators. Through the annual calculation, it is possible to identify the year of amortization for an investment in the tri-generation plant in different use cases. The motivation of this economic comparison is to demonstrate not only the energetic but also the financial viability of the proposed concept, which is necessary for a large-scale deployment.

2.4. Solar Thermal Collector

A separate analysis was performed to assess the potential of solar thermal to be the driving energy of the tri-generation unit. The heat source temperature of the tri-generation is estimated to be 180 °C, see Figure 1b. When there is no industrial waste heat source at the given temperature available, solar thermal energy can be considered as a potential driving renewable energy source. For this study, Parabolic Trough Collectors (PTCs) were considered, as they are the most commercially proven CST option [23]. The study on solar thermal energy integration in the system targets two objectives. The first is to determine the collector field size required to cover the heat source demand of the tri-generation unit in a yearly balance or for a monthly balance in winter or summer in use case “Berlin mix”. The second is to quantify the amount of backup energy still required at each site for a fixed collector field size of 3000 m². The comparison between sites in Madrid, Stuttgart, and Uppsala span high, medium, and low solar resource conditions. For the fixed-field case, backup heat is assumed from burning natural gas, and its CO₂ emission is calculated with an emission factor of 0.201 tCO_2,eq/MWh [24].

The PTC system was implemented in TRNSYS18 in the dynamic simulation model. The parabolic trough collector (PTC) is modelled via TESS Type 1288. The solar loop is a closed loop that transfers heat to the tri-generation unit/plant via a dedicated heat exchanger. A pump and iterative feedback controller regulate the loop mass flow to supply 180 °C when sufficient solar radiation is available. Whenever the collector outlet exceeds the setpoint, a tempering valve controller blends part of the return flow to cap the supply temperature at 180 °C. Mixers/dividers handle the recirculation.

2.5. Climate and Weather Data

The heat and cold demand for the user types residential and office depends on the climatic conditions, as presented in Section 2.2. Accordingly, the thermal energy demands of the “Berlin Mix” use case and the performance of the tri-generation unit are analyzed in different climates. Additionally, the potential of the PTC as promising energy source is assessed accordingly in different climates.

Therefore, three different types of climates in Europe were considered. A colder climate in Northern Europe, a moderate climate in Central Europe, and a warm climate in the South of Europe. This analysis provides answers to the usefulness of using solar thermal heat sources for the unit.

Weather files with information on an hourly base for a full year including temperature, radiation, humidity, wind, and other parameters, typically based on measurements from a nearby weather station were used in the simulation. Uppsala in Sweden was chosen for the representation of Northern Europe, with an annual global horizontal radiation of 960 kWh/m². Stuttgart in Germany was chosen for the climate in Central Europe, with 1182 kWh/m² annually global horizontal radiation, and Madrid in Spain was chosen to represent the climate in Southern Europe, with the highest annually radiation of 1721 kWh/m².

3. Results

The investigation performed through the dynamic simulation includes in an energetic assessment with a comparison of the identified four mixed use cases, an analysis on the impact of the climate, the potential for solar thermal as heat source of the tri-generation unit, as well as an economic assessment.

3.1. Dynamic Performance of Tri-Generation Unit

This section presents the behaviour of the tri-generation unit in the dynamic simulation set-up and how it reacts to the user demand and climate. The control of the unit is set to provide electricity at any time and to manage the heat or cold supply based on the actual demand for heat or cold. Thus, a constant operation of the unit is considered for this analysis to identify and distinguish differences in the operation for different user groups. The total thermal demand of a use case consists of the individual heat or cold from each user type. The heat supply from the tri-generation unit is influenced by the actual heat demand and calculated using the heat supply mode (see Figure 5). Figure 6a shows the composition of heat and cold demand for each month of a year for the use case “Berlin mix” for the climate of Stuttgart. It shows that heating demand rises in wintertime, whereas cooling demand peaks in summer. Due to the influence of industrial user types in this mix, heat and cold are required year-round, regardless of climate. Figure 6b,c show the ordered load for heating and cooling for one year and the resulting heating and cooling mode, respectively.

Figure 7 illustrates typical daily profiles on an hourly basis for heating during winter and cooling during summer in the “Berlin mix”. The black dotted line (right axis) represents the discrete supply mode (0, 0.25, 0.5, 0.75, 1), resulting from the stepwise mass flow control at the condenser or evaporator, as described in Section 2.1.

On a winter day (Figure 7a), the mode increases during the day as heating demand from residential and office buildings grows, often operating at full load during peak hours, and decreases stepwise in the evening, while the base heating demand of the buildings maintains a residual mode above zero. On a summer day (Figure 7b), the mode rises rapidly in the morning due to office cooling demand, remains at high level throughout the day following the load plateau, and decreases stepwise in the evening. The industrial demand keeps the mode continuously above zero, even during nighttime.

3.2. Energetic Analysis

The energetic analysis presents the performance and the total generation of heat, cold, and electricity of the tri-generation unit. The unit is considered to operate nonstop all year, but the heat and cold supply modes are managed based on the actual thermal demand at a time, as presented in Section 3.1. Figure 8 shows the annual course of cumulative energy generation per month. It shows that electricity is constantly provided throughout all four use cases. It shows that the monthly energy generation is different for each use case. During the summertime, for the central urban district use case, no heat is supplied to the user, but the cooling demand increase. The industrial site, however, shows constant supply of heat, cold, and electric power, regardless of the seasonal weather conditions.

The unit’s capacity is based on a preliminary and initial sizing, not to cover the total peak load of heating/cooling demand of the use case.

Table 3. Annual energy demand, generation, and efficiency of the tri-generation unit for the different use cases.

		1: Industrial Site	2: Urban Central District	3: “Berlin Mix”	4: Rural Municipality
Heat	Demand	4352 MWh/a	1896 MWh/a	2582 MWh/a	4435 MWh/a
	Generation	5640 MWh/a	2256 MWh/a	2753 MWh/a	5953 MWh/a
	Balance	+1288 MWh/a +29.6%	+360 MWh/a +18.99%	+171 MWh/a +6.62%	+1518 MWh/a +34.23%
Cold	Demand	5143 MWh/a	1277 MWh/a	1117 MWh/a	4379 MWh/a
	Generation	8723 MWh/a	1497 MWh/a	2155 MWh/a	8160 MWh/a
	Balance	+3580 MWh/a +69.61%	+220 MWh/a +17.23%	+1038 MWh/a +92.93%	+3781 MWh/a +86.34%
Electricity	Generation	2842 MWh/a	2630 MWh/a	2701 MWh/a	3433 MWh/a
Dimensioned heat source capacity of the CCHP plant		2898 kW	2898 kW	2060 kW	2183 kW
Heat source demand		25,392 MWh/a	18,046 MWh/a	19,123 MWh/a	28,821 MWh/a
Total energy generation		17,205 MWh/a	6383 MWh/a	7609 MWh/a	17,545 MWh/a
EUF		67.8%	35.4%	39.8%	60.9%

presents an overview of the energy demands and generations through the tri-generation unit. It shows, that for each use case, the tri-generation was still oversized, with surplus heat generation of 6.6% to 34.2%. Also, the generated cold exceeds the demanded cold by 17.2% to 92.9%. This indicates that the control of the heat and cold supply modes need further modification as well as the sizing of the tri-generation unit.

Across the four mixed use cases, the unit delivers concurrent heat, cold, and electricity, with significantly different energy utilization factors (EUFs). Annual demands span between 1.9 and 4.4 GWh for heat and 1.1–5.1 GWh for cold, with the industrial site and rural municipality showing the highest simultaneous loads and the highest EUF of 68% and 61%, respectively. The Berlin Mix sits in the lower mid-range (EUF 40%), while the urban central district is lowest (35%). In both cases, heat demand is reduced or even non-existent in summertime. This, in turn, leads to a reduced heat supply and consequently a reduced EUF. However, as pictured in Figure 8, an increase in power generation is given for both cases during summertime. Based on the initial sizing of the four use cases, as presented in Table 2 and Table 3, the required heat demand for the constant operation of the tri-generation unit ranges between 18.0 and 28.8 GWh/a across the four cases.

Heat and cold supply dominate the EUF. Thus, a constant heat and cold demand, independent from seasonal weather conditions by commercial and industrial users, leads to an increase in EUF. If no heat or cold is provided to the user, the EUF is characterized by the electricity generation efficiency only. Figure 9 shows the total heat, cold, and electricity generated by the tri-generation in each use case. It shows that, for both the industrial site and the common rural municipality, cold is generated the most. An operation for both the urban central district and the Berlin Mix leads a relatively equal generation of heat, cold, and electricity.

3.3. Climate Analysis and Solar Thermal Heat Supply

The thermal energy of use case 3, the Berlin Mix, is modelled in the three climates of Uppsala, Stuttgart, and Madrid. Figure 10 presents the heating and cooling energy demand for all the three investigated locations. Heating demand is highest in Uppsala, which characterized by cold climate. In the warm climate of Madrid, on the other hand, cooling demand is highest, and heating demand lowest.

The climate also has an effect on the dimension of the tri-generation unit, with the smallest scale in warm climate and the largest scale in colder climate.

Table 4. Heat source demand, efficiency, and dimensioning of “Berlin mix” in different climates.

	Uppsala (Sweden)	Stuttgart (Germany)	Madrid (Spain)
Dimensioned heat source capacity of the tri-generation unit	2232 kW	2183 kW	1986 kW
EUF	41.2%	39.8%	42.5%
Annual heat source demand	19,554 MWh/a	19,123 MWh/a	17,399 MWh/a

shows the resulting required heat source capacity for each location and the resulting EUF. For the Berlin Mix case, moving from Uppsala (cold) to Stuttgart (temperate) to Madrid (warm) changes the driver-energy need only modestly by about 11% and keeps EUF in a tight 40–43% band, with slightly smaller design capacity in Madrid. This shows an internal flexibility of the cycle against ambient variations.

Aside from a constantly available heat source, the potential of solar thermal heat as primary energy source is analyzed. For this investigation, the unit is operating only to cover the heat and cold demand, not constantly.

Table 5. Overview and comparison of the climate conditions in the three locations and the required collector field size for balanced heat source supply for the year, January, and July, respectively.

	Balance Period	Uppsala (Sweden)	Stuttgart (Germany)	Madrid (Spain)
Mean ambient temperature		7.51 °C	9.55 °C	14.71 °C
Global horizontal radiation (GHI)		960 kWh/m2a	1182 kWh/m2a	1722 kWh/m2a
Energy demand Heat Source		11,034 MWh/a	9921 MWh/a	9071 MWh/a
Collector Field size (no. of collectors in array × number of arrays) aperture area	Year	367 × 10 37,700 m2	320 × 10 32,900 m2	114 × 10 11,700 m2
	January	-	8215 × 10 842,900 m2	513 × 10 52,700 m2
	July	86 × 10 8900 m2	100 × 10 10,300 m2	57 × 10 5900 m2
Hours of solar only operation	Year	1504 h (17%)	1548 h (18%)	2393 h (27%)
Solar energy generation	Year	11.05 GWh/a	9.91 GWh/a	9.06 GWh/a

shows the results of this investigation. For the location of Madrid, the GHI is highest, while at the same time, the heat source demand of the tri-generation unit is lowest. This already indicates the highest potential for a solar thermal operation of the tri-generation unit. The parabolic trough collector (PTC) is the selected solar thermal collector type, capable of providing heat at the required temperature of 180 °C. Table 5 shows the theoretically required size of the collector field if the collector field should generate enough heat to cover the demand on a balance for different periods, (i) the total year, (ii) the month January, and (iii) the month July. The collector field consists of multiple arrays, each containing 10 collectors with 10.25 m² each. To generate as much solar heat as required to operate the tri-generation throughout all the year, more than 30,000 m² of collector area are required for Uppsala and Stuttgart, whereas in Madrid, 11,700 m² would be sufficient. Thermals storages and losses are not considered here. An operation of the tri-generation unit in the wintertime via solar thermal only is not feasible.

The second investigation on the solar thermal energy supply considers a fixed-field size of 3000 m² aperture area with no consideration of a thermal storage. Thus, any solar heat surpassing the demand at a given time is not used. The demand of the Berlin Mix is the same as in Table 5.

Table 6. Overview and comparison of the impact of a 3000 m² solar thermal collector field and the requirement of a back-up system.

	Uppsala (Sweden)	Stuttgart (Germany)	Madrid (Spain)
Collector field size	3000 m2	3000 m2	3000 m2
Total solar energy yield	564 MWh/a	603 MWh/a	1384 MWh/a
Solar surplus heat (not used)	50 MWh/a	45 MWh/a	86 MWh/a
Solar energy directly used	514 MWh/a (5%)	558 MWh/a (6%)	1298 MWh/a (14%)
Energy from back-up supply	10,519 MWh/a (95%)	9362 MWh/a (94%)	7773 MWh/a (86%)
Hours when back-up boiler needed	8585 h (98%)	8578 h (98%)	8465 h (97%)
GHG emission savings	114.8 tCO2-eq	124.6 tCO2-eq	289.9 tCO2-eq

shows the results of this second investigation. With such a system, the solar share in primary energy supply remains limited, about 5% (Uppsala), 6% (Stuttgart), and 14% (Madrid). Again, the system shows the best performance for the location of Madrid. The rest of the required heat must be supplied from a back-up system, considered to be a natural gas boiler with an efficiency of 0.9. Thus, a solar thermal collector field of 3000 m² would lead to a GHG emission savings of 115 tCO₂-eq to 290 tCO₂-eq. This analysis demonstrates that a back-up system is necessary even if a 3000 m² field of CST collectors is installed.

The study shows that the tri-generation circle is robust against different climates, and the annual EUF shows no great deviation. CST can technically supply the required heat input, but annual balancing becomes nearly impossible in Northern/Central Europe due to low radiation in the winter period. In the southern climate, driving the unit by heat from CST could be an alternative to missing industrial waste heat. At large field sizes (3000 m²), CST acts as a partial decarbonisation measure.

3.4. Economic Analysis

The economic analysis includes an initial investment of the tri-generation unit for each case individually as the size of the unit is adapted accordingly. Figure 11a shows that the earliest amortization is given for the use case 4, rural municipality, after 5.4 years. When residential and office are there in a use case only, the longest amortization rate after eight years is given. It shows that a large energy consumption from industry and retail is economically advantageous, as they demand heat and cold simultaneously. Figure 11b shows the economic curves for the Berlin Mix application in the three different climates. Economic-wise, the location is less relevant than the selection of use case. However, it shows that the operation in Stuttgart with temperate climate achieves best performance.

Table 7. Economic data of the use case “Berlin mix” in different cities with separate climate.

	Uppsala (Sweden)	Stuttgart (Germany)	Madrid (Spain)
Investment	7415 k€	7095 k€	6455 k€
Ø 20 yrs sale of energy earnings	822 k€/a	845 k€/a	720 k€/a
Ø 20 yrs energy savings	903 k€/a	827 k€/a	693 k€/a
Ø 20 yrs missing energy costs	−89 k€/a	−47 k€/a	−50 k€/a
Ø 20 yrs maintenance and replacement	−292 k€/a	−286 k€/a	−260 k€/a
Ø 20 yrs annual balance	1273 k€	1293 k€	1053 k€
Amortization time	7.0 yrs	6.8 yrs	7.5 yrs

shows the investment and the annual average of savings, costs, and income for the Berlin Mix use case for the three investigated locations of Uppsala, Stuttgart, and Madrid.

The economic analysis does not take energy costs for the heat source into consideration, assuming that industrial waste heat is available for free, even though the practical implementation is not feasible. This assumption works only if the industrial company also operates the tri-generation unit. Rather than focusing on the heat source cost, this first economic assessment focuses on the energy outputs of the tri-generation machine. However, if the heat source cost account for 6 ct/kWh, the tri-generation may not reach an attractive payback period. A cost-effective operation of the unit implies either lower costs or higher energy supply costs to users.

Due to lightly different demands in the different climates, the investment costs vary from EUR 7.4 million (Uppsala) and EUR 7.1 million (Stuttgart) to EUR 6.46 million (Madrid). Cumulative 20-year operating cash-in (energy sales and savings) lands near EUR 1.4–1.8 million, with Stuttgart and Uppsala showing higher cash-in. When considering missing energy costs and maintenance costs, Stuttgart has the highest annual balance income with EUR 1.29 million, while Madrid has the lowest at EUR 1.1 million. The payback times fall within a narrow range of 6.8–7.5 years, as presented in Figure 11, indicating that economic performance is not primarily based on the investment cost.

For the location of Stuttgart, the tri-generation unit presents the lowest period of amortization. This results, on the one hand, from slightly lower investment costs than in northern climate, but also from the balance between the high heating demand and the still existing cooling demand—which is higher than in northern climate—leading to a surplus production of cooling that is higher than in the warm and cold climates. This case profits from selling cooling energy the most. The tri-generation unit for the location of Madrid is smallest among the three sites, and its operation leads to a reduced generation of surplus energy and thus to less income from selling surplus energy.

A sensitivity analysis was performed for the Berlin Mix use case in Stuttgart to illustrate the impact of the specific energy prices, as presented in Section 2.3. For the energy prices of sold surplus energy, an alteration of +25% was considered, causing prices of 10 ct/kWh instead of 8 ct/kWh of sold thermal energy and 25 ct/kWh instead of 20 ct/kWh of electric energy. The same accounts for a price reduction by −25%. Also, the energy prices taken into consideration in savings through self-consumption were included in the sensitivity analysis, also showing an alteration of +/− 25%. The results, presented in the form of the amortization line in Figure 12, show that in this case, a variation in the prices of sold energy is very close a variation in self-consumed energy. Both vectors change the annual performance by +/− 14–16%. The time when amortization is reached changes by about +/− one year. Thus, even if either the economic income of selling surplus energy or the economic savings through self-consumption is reduced by 25%, amortization is still achieved within 7.8 years at the latest.

The biggest impact on the energy-specific economic performance comes from savings through self-consumption and reduced energy purchase from external sources. Second, it is relevant that further customers outside the initial customer group are in spatial range for energy transport. Otherwise, the distribution of surplus energy is limited or even blocked. Surplus energy comes from overproduction of energy, which is directly aligned to the selection of the heat/cold supply modes, as well as due to the dimensioning of the unit. In the southern climate, the plant dimension has the lowest capacity and the lowest surplus generation. In conclusion, payback around 6–7 years is feasible across the different climates.

4. Discussion

The results of this study are derived from a first-of-its-kind dynamic simulation study of a tri-generation unit based on a single thermodynamic cycle. Therefore, the validity of the results is limited on the previous results in modelling the dynamic performance of the tri-generation derived from a performance map generated through static simulations. This affects the calculation of electric efficiency, COP, and EER. The dynamic simulation model of the tri-generation unit is significantly influenced by earlier assumption and input values on component level, such as the isentropic efficiency of expanders and compressors. Furthermore, the model represents the tri-generation applying the refrigerant R1233zd(e) only, which does not allow a direct validity of the performance of the unit with other refrigerants.

Heat and cold generation of the tri-generation unit is controlled via a selection of heating/cooling supply mode. Each mode represents the level of mass flow to run through the unit’s evaporator and condenser, respectively. The mode allows an alteration in 25% steps, and the selection is based on the actual heat/cold demand. The given mode arrangement requires further modification to meet the actual heat/cold demand more precisely. A Thermal Energy Storages (TES) is relevant and very important for the unit’s operation. The characteristics and role of TES, namely, size and losses, are not investigated in this study, as the unit is not modelled in a full-system approach but for itself instead. The relevance of TES can be reduced through controlling the heat/cold supply modes in short steps, such as 5% or 10%.

The energetic analysis highlights the different needs of the four investigated use cases that include a combination of individual user types. The thermal energy demand of office and residential buildings depends on the climate and varies throughout the year.

As presented in Table 2, the heat source capacity may outstrip the combined thermal energy demand load for both heating and cooling by factor 2–3. This is both a technical and an economic constraint. The unit dimension in this study is sized form the demand-side, assuming a heat source is available at any capacity and any time. This simplified approach was selected as the study targets the demand first before assessing the integration to any heat source. There is room for improvement in sizing the unit, as even with an initial sizing, surplus of heat and cold is generated.

The economic analysis presents the economic performance of the tri-generation for four different use cases and for the Berlin Mix use case in three different locations in separate climates. It includes investment costs and annual costs in contrast to savings and earnings from supplied energy. The operation of the unit needs further investigations and a detailed economic analysis assessing suitable business models. The economic analysis of this study presents a first pilot analysis that highlights the most relevant aspects but does not cover the full picture.

The application of the tri-generation unit is most promising when industry and retail is part of the user types in a use case, as they provide a constant demand for thermal energy. Office and residential users rarely have a simultaneous demand for heat and cold, thus the tri-generation cannot exploit its full potential in simultaneous supply of heat, cold and electric power. Additionally, industry is a promising user type, as industrial waste heat is a promising heat source.

For the three investigated locations, the best economic performance is identified in the temperate climate of Stuttgart. Even though any excess heat and cold, as well as all electricity, is sold to external users, the most relevant are the financial savings through self-consumption of heat and cold. The tri-generation unit generates more heat and cold than required in all three locations. An adjusted sizing of the unit will directly affect not only the investment costs, but also the savings and earnings from energy sales.

5. Conclusions

This study presents a first approach in dynamically modelling a tri-generation unit that is based on a single refrigerant thermodynamic cycle operating in the two-phase area, able to deliver heat, cold, and electric power to the user. When driven by renewable heat, this unit can provide emission-free energy to users and play a supporting role in the clean energy transition. The research contributes to a better understanding of the critical points in the operation of the presented tri-generation unit, such as the influence of T_cool, its performance to supply different user types, as well as presenting the performance in different climates. The presented tri-generation unit is a promising and innovative solution for the effective use of excess heat or renewable heat. As the unit is still under development at the moment of writing, simulation studies are the preferred way to investigate the unit’s performance and potential in future applications. The study’s highlights are as follows:

• Initial assumptions were used to develop a dynamic calculation model of the heat, cold, and electricity supply from the tri-generation unit and applied for annual simulations.
• The developed control strategy allows to adapt the unit’s heat and cold supply in 25% steps based on the users’ demand.
• Four use cases were identified, and the performance of the tri-generation unit in each case was analyzed.
• The Energy Utilization Factor (EUF) is highest for industrial sites and rural municipalities, due to the thermal demand of industry and retail throughout the year.
• Climate influences both user demand and the performance of the unit.
• Selling electricity is a significant aspect in the economic performance of the tri-generation unit.

Lab-scale testing of a physical prototype is necessary to validate the performance in simulation and to analyze the actual flexibility in start–stop operation. The impact of TES integration on the unit’s behaviour and the set-up in a system are topics of interest and worthy of future investigations to deliver a comprehensive assessment of the technology. The roll-out of the technology, aiming at a higher technology readiness level (TRL), benefits from this study by highlighting relevant aspects for further optimization. Next steps for the roll-out of products to commercialize this tri-generation technology include not only scientific lab tests but also demonstration in real use cases to further assess its potential.