Which Factors Determine the Fuel Savings From Electricity Generation Using Waste Heat?
Author: Jeroen Berger • Publication date:
Fuel savings from electricity production using waste heat are determined by the net electricity that a waste heat recovery system actually supplies and the resulting reduction in fuel use by the equipment it displaces or unloads. A waste heat recovery system (WHRS) can convert heat from exhaust gases or engine cooling water into electricity for the shipboard electrical system using an Organic Rankine Cycle (ORC).
However, this electricity production does not automatically lead to proportional fuel savings. The result depends on the available waste heat, engine load, the WHRS's own energy consumption and how the load on existing generating sets changes. The specific fuel consumption of these engines and the possibility of shutting down an entire generating set in particular may affect the savings. It also matters whether ORC electricity displaces generation from auxiliary-engine-driven generators, a shaft generator or another energy source. Additional losses in the exhaust gas and cooling systems may also reduce the net savings.
Fuel savings are therefore established by comparing the ship's total fuel consumption with and without a WHRS under sufficiently comparable operating conditions and at equivalent ship performance. Net electricity generation provides an important technical basis for this, but the amount of fuel actually avoided is ultimately decisive.
Why Is WHRS Electricity Production Not the Same as Fuel Savings?
An ORC installation uses heat released during engine operation to generate electricity through a separate thermodynamic cycle. Some of the electrical energy demand can thus be met without burning fuel directly in a separate engine to generate that electricity. However, the amount of electricity generated does not yet show how much fuel is actually saved. For this purpose, it must be known which existing energy source needs to supply less power because of the ORC and how that energy source's fuel consumption consequently changes.
When a diesel generating set is partially unloaded, fuel consumption may decrease. The size of this reduction depends on the original and new engine operating points. A generating set that continues to run at a lower load still consumes fuel. Moreover, specific fuel consumption does not necessarily change in the same proportion as the electrical power delivered.
With a different generator configuration, ORC output may make it possible to shut down an entire generating set. Consequently, not only does the load on the individual engines change, but so does the number of engines that remain in service. Fuel savings may then differ from those achieved by partially unloading the same generating sets.
The recovered electricity must also displace generation that would otherwise require fuel. If the ship is supplied with shore power, for example, additional ORC electricity need not automatically reduce fuel consumption on board. With battery storage, the effect also depends on the energy source that would otherwise be used to charge the battery. Electricity that replaces another energy source does not necessarily reduce fuel consumption at the time it is generated.
Fuel savings therefore do not arise solely because the ORC produces electricity. They arise because this production changes the deployment and fuel consumption of other energy sources. The assessment must therefore cover the ship's entire electrical energy supply. Gross or net ORC electrical output provides a technical starting point, but the difference in fuel consumption of the engines concerned determines how much fuel is actually saved.
How Do the Heat Source, Engine Load and Operating Hours Determine the Available Electricity?
How much electricity a WHRS can produce depends first on the waste heat available during engine operation. Exhaust gases and engine cooling water are important heat sources, but their temperature and thermal power change with engine load. For exhaust gas heat, gas temperature, mass flow rate and the limit on exhaust gas cooling are key factors. Together with heat exchanger performance, these determine how much thermal power can be transferred to the ORC.
A high exhaust gas temperature does not automatically mean that a large amount of heat is available if the gas flow rate remains limited. Heat extraction may also be limited by the permissible exhaust back pressure, material temperature limits or other necessary functions of the exhaust gas path. For engine cooling water, the available temperature, flow rate and engine heat release are relevant. The original engine cooling must continue to function within the specified limits. Not all heat rejected through the cooling water can therefore be extracted for the ORC without additional conditions.
The amount of heat actually absorbed then determines which process conditions are possible in the ORC. If heat input decreases, the working-fluid flow rate, evaporating conditions and available expansion work may change. Electrical power need not decrease in proportion to thermal input power. Heat transfer and expander performance may also change at part load. The conversion of available heat into electricity must therefore be assessed for the relevant operating conditions.
The condenser also affects how much electrical power the ORC can deliver. After expansion, the remaining process heat must be rejected. If the cooling water becomes warmer, condensing pressure may rise and the available expansion range may narrow. At the same engine load, the ORC may consequently produce less electrical power, even if the available heat supply on the hot side remains almost unchanged. Heat source and cooling conditions must therefore be assessed at the same operating point.
For fuel savings, it is then important how often these conditions actually occur. An ORC can produce substantial electrical power at high engine load, while the ship operates under those conditions only for limited periods. Annual electricity production may therefore be much lower than a calculation based on rated generator power and all engine hours would suggest. Conversely, an installation with lower maximum power may receive enough usable heat over many operating hours to make a useful contribution.
Heat source conditions must therefore be linked to the actual operating and load profile. This also includes periods in which insufficient heat is available, the ORC must be started or electricity production is temporarily limited. For ships with several engines, it must also be clear which heat sources are available during the same operating periods. An engine that is out of service supplies no continuous heat from combustion. The combined heat supply therefore depends on the engines actually running and their individual loads.
Possible ORC electricity production must therefore be based on the heat that can be used within acceptable technical limits under these conditions. Only when that production has been determined sufficiently reliably can it be established which existing generation the ORC actually displaces.
Why Is Net Electricity Generation Decisive for the Savings?
The ORC generator's gross electrical output exceeds the net output of the complete WHRS by the power used or lost within the system boundary. The installation itself needs energy to operate the thermodynamic process and associated equipment. The working-fluid pump pressurizes the liquid organic working fluid and circulates it through the cycle. Depending on the configuration, intermediate circuit pumps, additional cooling-water pumps and other supporting equipment may also use electricity.
Electrical losses may also occur between the generator terminals and the connection point at which WHRS electricity actually becomes available. This may be relevant when power electronics, transformers or additional distribution equipment are used. The system boundary determines which auxiliary loads and losses a stated power figure already accounts for. For a reliable fuel savings calculation, it must therefore be clear how much net electricity the complete installation supplies to the ship.
For existing shared ship systems, this boundary definition requires particular attention. A cooling-water pump may, for example, already have served other equipment before the WHRS was installed. The pump's entire original power demand is then not automatically a new electrical load caused by waste heat recovery. However, it must be established whether the WHRS increases the pump's electricity consumption. That additional load must be included if it has not yet been accounted for in the net power value used.
The same energy consumption must not be deducted several times. If the working-fluid pump's consumption is already included in net module power, it must not be counted again as a separate loss. Conversely, necessary equipment outside the module must be included if it consumes additional electricity.
The distinction between electrical power and electrical energy also remains important. Net power in kilowatts is the rate at which the WHRS supplies electrical energy at a particular moment. Net energy in kilowatt-hours measures how much electricity was supplied over a particular period. For fuel savings, that energy must be linked to the existing generation it actually displaces. Positive net ORC production does not mean that every kilowatt-hour produced automatically replaces the same amount of fuel.
The shipboard electrical system must be able to accept the electricity, and the controls must be able to reduce output from other sources accordingly. If these capabilities are limited, some of the available WHRS output may go unused. With energy storage too, fuel savings may arise at a different time from when the ORC electricity is produced. It must therefore be established which energy production is ultimately avoided, instead of assessing only the electrical exchange at a single moment.
The relevant starting point is therefore not only the net electricity measured at the module connection, but the electrical contribution that actually replaces other energy production. Any additional load on shared pumps or other existing equipment must be included in the ship's overall energy balance. Only when the net electrical contribution has been established within a consistent system boundary can it be reliably linked to how much less fuel the ship needs.
Which Existing Energy Source Is Replaced by WHRS Electricity?
Fuel savings depend strongly on which existing source supplies less electricity as a result of the WHRS. The same amount of net ORC electricity may, after all, have different effects on fuel consumption, depending on the original energy supply. In a conventional electrical system, electricity is usually generated by one or more generating sets. When the ORC supplies power to the shipboard electrical system, these generating sets may need to supply less electrical power. Fuel savings then follow from the difference between their original and new fuel consumption.
For a ship with a shaft generator, the situation is different. A shaft generator converts mechanical power from the propulsion engine into electricity. When the WHRS replaces some of this production, the mechanical power that the propulsion engine must supply to the shaft generator may decrease. The associated fuel savings depend on the changed engine operating point and losses in mechanical and electrical power transmission. A kilowatt of electrical ORC power therefore does not automatically equal a reduction of one kilowatt in mechanical engine power or a fixed amount of avoided fuel.
For a correct comparison, it must be established how much power the propulsion engine must actually deliver with and without the WHRS. The same propulsion performance and electrical energy demand must be used as the basis.
On ships with battery storage, the relationship between ORC electricity and fuel consumption may change further. If the WHRS supplies electricity that would otherwise be drawn from a battery, fuel savings do not necessarily arise at that same moment. The ultimate effect depends on how the battery would have been charged without the WHRS. If that energy is normally supplied by fuel-fired generating sets, the ORC may reduce the amount of charging energy required.
Fuel savings may then arise at a later time. The relevant charging, storage and conversion losses must be included. The battery must also have a sufficiently comparable state of charge at the beginning and end of the period examined. Drawing less energy from the battery during a single operating period must not be regarded as definitive fuel savings without further assessment if the energy must still be replenished later by a fuel-fired generator.
If the battery is charged mainly using shore power or another source that is not fuel-fired, the effect on fuel consumption on board may be different. The same applies if the ship receives electricity directly through a shore connection during a port stay. If ORC electricity then reduces shore power consumption, this need not reduce the fuel consumed by the ship's engines. However, less external electricity may be needed as a result.
The actual availability of waste heat during port operation must also be included. An ORC that relies on a propulsion engine for heat cannot continue generating when that engine is shut down without another suitable heat source. The same amount of net ORC electricity may therefore lead to different fuel savings on different ships. For each relevant operating mode, it must be clear which energy source the WHRS actually replaces and how the fuel consumption of the complete energy supply consequently changes.
How Does the Specific Fuel Consumption of Generating Sets Affect the Savings?
The fuel consumption of a generating set depends on the power delivered and the efficiency of the engine and generator. Specific fuel consumption is often expressed in grams of fuel per kilowatt-hour. The stated basis must be clear, whether mechanical engine output or the electrical output of the complete generating set. These values are not automatically equal.
If specific fuel consumption relates to mechanical shaft power, generator efficiency and other relevant losses must also be taken into account for electrical production. Moreover, a generating set does not operate with the same specific fuel consumption at every load. At part load, combustion conditions, air supply and internal losses, among other factors, may affect the amount of fuel per kilowatt-hour produced.
A generating set that delivers less electrical power need not use proportionally less fuel. The difference in fuel consumption must be determined for the original and new loads. For WHRS savings, it is therefore established how much fuel the generating set used per unit of time before unloading and how much it uses per unit of time at the new operating point. The difference between the two fuel flow rates represents the savings under those conditions, provided that the other relevant factors are sufficiently comparable.
It is therefore insufficient simply to multiply the specific fuel consumption at the original engine load by the ORC's electrical power. After unloading, the generating set runs at a different operating point, at which specific fuel consumption may differ. Nor must a single general part-load efficiency be assumed for all generating sets without technical substantiation. Part-load fuel consumption depends on engine type, design, controls and the relevant load range. Even engines with comparable rated power may have different fuel consumption characteristics.
When several generating sets operate in parallel, their combined fuel consumption must be assessed. The load on each engine may change when the power management system redistributes the remaining electrical load among the existing generating sets. The same net WHRS output may consequently deliver different fuel savings with different generator combinations. Not only the total electrical load, but also its distribution among the individual engines is therefore important.
For a reliable calculation, engine-specific or generator-specific data recording fuel consumption across the relevant load range are preferably available. A fixed conversion factor between net ORC kilowatt-hours and fuel saved can be used as a simplification only if it has been sufficiently demonstrated to produce a representative result for the engine loads examined.
Why Does It Matter Whether a Generating Set Is Unloaded or Shut Down?
When a WHRS supplies electricity, the power management system can adjust the load on the existing generating sets. It matters whether the engines concerned continue running or whether the number of active generating sets can be reduced. With partial unloading, the engines remain in service but produce less electrical power. Their fuel consumption changes according to the new load and the corresponding fuel consumption curve.
When several generating sets operate in parallel, ORC generation can reduce the load shared by these generating sets. Combined fuel consumption then depends on the original load distribution, the new load on each engine and the selected generator control. Partial unloading need not be the most favourable operating strategy under all conditions. If an engine continues to run at very low load for a prolonged period, this may also affect combustion, maintenance requirements and permitted operation. Which limits apply depends on the engine design and the manufacturer's requirements.
Under suitable conditions, the available WHRS electricity may make it possible to reduce the number of running generating sets. When an entire generating set is shut down, that engine's fuel consumption ceases during the period concerned. However, the remaining generating sets must take over the remaining electrical load. They may consequently run at higher loads and consume more fuel than before the other engine was shut down.
Total fuel savings therefore do not automatically equal the original consumption of the generating set that has been shut down. The change in fuel consumption of the remaining engines must also be included. The necessary electrical power reserve, any redundancy requirements and safe operation of the shipboard electrical system also remain constraints on generator dispatch.
A WHRS depends on the available engine heat. If the heat supply suddenly decreases, electricity production may also decrease. Other energy sources must then retain sufficient capacity to continue supplying the necessary electrical load. A generating set that must first start and synchronize does not immediately deliver electrical power to the shipboard electrical system. During certain ship activities, it may therefore be necessary to keep additional generating capacity online or another suitable source of reserve power available.
Energy storage may also influence the permitted generator strategy, provided that the electrical installation is designed for it and enough stored energy and reserve power are available. Shutting down a generating set may therefore make favourable fuel savings possible, but is not automatically the best or technically permitted solution. The correct comparison concerns the total fuel consumption of all engines involved within a generator configuration that actually suits the ship's operational and safety limits.
What Happens When the ORC Unloads Its Own Heat Source?
A particular interaction arises when an ORC uses waste heat from a generating set whose load is reduced by the ORC's own electrical output. The initial estimate of ORC output is based on the heat that the engine concerned releases at a particular load. When the ORC subsequently supplies electricity to the shipboard electrical system, that same generating set may need to produce less electrical power.
Exhaust gas temperature, exhaust gas mass flow rate and heat release through the cooling-water circuits may consequently change. Under these new conditions, the ORC may receive less thermal power than at the original, higher generator load. Electrical production may consequently decrease relative to the value possible before the generating set was unloaded. In this configuration, heat availability and electrical output therefore affect one another.
The final fuel savings must therefore be determined at an operating point with consistent values for generator load, available waste heat and net ORC output. It is not technically sound to combine full ORC production at the original generator load with the fuel consumption that the same engine would have at a substantially lower load without further assessment.
Such a calculation may assume heat that is no longer available in the same quantity or at the same temperature level after the engine is unloaded. The estimate may therefore overstate both electricity generation and fuel savings. This interaction becomes even more important when shutting down an entire generating set is considered. If the ORC receives its heat exclusively from that engine, the continuous heat source is also lost on shutdown. The original electricity production cannot then be maintained without another suitable heat supply.
For a WHRS that receives heat from several generating sets, the situation may be different. Shutting down one engine removes a heat source, but may simultaneously increase the load on the remaining engines. Their heat release may consequently change. The combined heat supply must then be recalculated for the resulting combination of running engines and their individual loads.
The thermal inertia of heat exchangers and intermediate circuits also merits attention. After generator load has been reduced, heat previously stored in the installation may still be present temporarily. The ORC output may therefore take time to settle at the level associated with the new operating condition. However, this temporarily stored heat is not a lasting additional source for electricity production.
A reliable savings calculation must therefore reflect steady operating conditions or, during highly variable operation, take account of the actual transition periods and dynamic behaviour of the installation. Conversely, if the ORC receives heat exclusively from an independently running propulsion engine, unloading individual electrical generating sets need not directly affect that heat source.
This distinction between the different heat source configurations is important. For an ORC that unloads its own heat source, generator load changes because of electricity production, while that changed generator load in turn determines the available heat supply. Only when this mutual influence has been sufficiently accounted for can it be established how much less fuel the complete engine and generator installation actually consumes.
Which Additional System Losses Reduce Net Fuel Savings?
A WHRS can produce electricity while its connection to existing ship systems causes additional energy losses. These effects must be included to determine net fuel savings for the whole ship. For recovery from exhaust gases, the heat exchanger is an additional component of the exhaust gas path. It creates flow resistance and may consequently increase total exhaust back pressure.
If that additional back pressure affects engine fuel consumption, this effect must be included in the fuel savings calculation. The magnitude of the effect depends on the engine design, original exhaust gas installation and pressure drop across the heat exchanger. Fouling may also change flow resistance during operation. It is therefore insufficient merely to establish how much electricity can be produced from exhaust gas heat.
Any additional fuel consumption resulting from the modified exhaust gas installation may offset part of the savings. The effects must be assessed for the actual engine conditions and permitted operating limits.
Ship cooling may also require additional energy. The ORC rejects most of its unused heat through the condenser. Existing cooling circuits can be used for this purpose, but additional pumps or other cooling equipment may also be needed. If an existing cooling-water pump consumes more electricity because of the WHRS connection, this additional load must be accounted for. However, the pump's existing demand cannot automatically be attributed to the WHRS if the pump would continue to serve other ship systems without recovery.
It must therefore be clear which auxiliary loads the net output figure already accounts for. If additional pump power has already been deducted from gross ORC production, the same consumption must not be counted again as a separate loss. Conversely, additional electrical consumption still outside the net ORC value must be included in the overall ship energy balance. Otherwise, the additional electrical contribution may be presented too favourably.
Existing uses of recovered heat may also affect the final savings. A WHRS can use heat that, without the new installation, was already used for heating, fuel treatment or another useful application. If this heat is subsequently used for electricity production, additional energy may be needed to continue meeting the original heat demand. Any extra fuel used because of the change in heat distribution must be included.
An ORC may consequently generate a net electrical output, while the additional fuel savings for the whole ship are smaller than would be expected from that electricity production alone. Under unfavourable conditions, the additional losses and required replacement energy production may significantly limit the possible benefit.
Fuel savings must therefore not be assessed solely at ORC module level. The effects on engines, heat exchangers, cooling, electrical distribution and existing equipment that uses heat together determine how much fuel is actually avoided.
How Do Operating Hours, Availability and Electrical Offtake Affect Annual Savings?
Fuel savings established at one favourable engine operating point do not automatically describe the savings over a full voyage or calendar year. Heat availability and energy demand vary as the ship's operating conditions change. Engine load, waste heat, electrical demand and cooling-water temperature may vary substantially across the ship's sailing and working modes. The net electricity generation that the ORC can deliver in the different operating modes consequently also changes.
For annual fuel savings, it must therefore be established for how many hours the WHRS can actually operate under each relevant condition. This includes both high-load periods and part-load operation when less heat is available. During certain operating periods, insufficient suitable heat may be present to provide a positive net electrical contribution. Starting up, reducing output and stopping may also affect the amount of useful electricity.
Technical availability is another important factor. During maintenance, faults or necessary downtime, less or no energy can be recovered. An installation that performs well under suitable heat conditions may consequently still deliver limited fuel savings over the full year. The annual calculation must therefore take account of actual availability or a well-supported availability estimate.
The electricity generated must also actually be usable. If electrical demand on board is limited, ORC production may be restricted. The existing generator strategy may also require certain engines to remain in service. The savings then depend on which generating equipment can actually use less fuel under these conditions. Not every technically available kilowatt-hour automatically leads to an equivalent reduction in generator output.
With energy storage, it must also be established when electricity is supplied and which later battery charging the supply displaces. Lower fuel consumption during a single operating period must not be regarded as definitive savings without further assessment if the energy demand is merely shifted to another time.
Annual fuel savings therefore follow from the changes in fuel consumption during the individual operating periods, taking account of their duration and the generator configuration actually used. If the operational profile changes, the savings calculation must also be reassessed. A different average ship speed, a change in engine dispatch or additional energy-saving measure may affect both the available heat supply and the original electrical energy demand.
Annual savings may consequently change without the ORC installation having deteriorated technically. Fewer suitable operating hours or lower electrical demand may limit energy output, while the installation continues to function as expected under comparable heat and cooling conditions. For a reliable annual result, the heat actually available, electricity actually used on board and fuel-fired energy production avoided must therefore be assessed together across the full relevant operating profile.
How Are Actual Fuel Savings Reliably Established?
A reliable determination of fuel savings compares the ship's total fuel consumption with and without waste heat recovery at comparable ship performance and under comparable operating conditions. For this purpose, a suitable baseline without the WHRS must first be established. This describes which engines and other energy sources would be needed to provide the same propulsion performance, electrical load and any heat demand when the ORC produces no electricity.
The permitted generator strategy and necessary electrical reserve must also be included in this reference case. The comparison must not assume a generator configuration that is unsuitable for the ship's actual operating conditions. The situation with waste heat recovery is then determined. The heat sources actually available, net ORC electricity production, changed engine loads and any changes in generator dispatch are included.
The relevant additional losses and existing uses of heat must also remain part of the overall energy balance. Only then does it become clear which change in fuel consumption can actually be attributed to the WHRS. If the ORC produces electricity from the heat of generating sets that are unloaded by that same electricity, the heat supply and generator load must correspond to the same final operating condition.
For each relevant engine, fuel consumption without and with the WHRS is then established. Suitable fuel consumption curves from the engine or generator manufacturer can be used for this purpose. The data must relate to the correct power basis and the load range in which the engine actually operates. If consumption data are available only for certain load points, it must also be clear how fuel consumption for intermediate loads is determined.
The difference between the total fuel flow rates without and with the WHRS then gives the fuel saving rate under the conditions examined. For an existing installation, operational measurements can support this assessment. Net electrical WHRS production, engine loads, fuel consumption and other relevant operating data are recorded over the same periods.
The measurement method must be sensitive enough to detect the difference under investigation. If the expected savings are small relative to the ship's total fuel consumption, uncertainty in readings from fuel meters and other instruments may play an important role. The reliability of the reference case without a WHRS also remains decisive. If the ship operates under substantially different conditions before and after the ORC is installed, the difference in fuel consumption cannot automatically be attributed entirely to waste heat recovery.
Changes in ship speed, propulsion power, weather conditions, electrical demand and the use of other equipment may also affect the result. The operating periods examined must therefore be sufficiently comparable, or the relevant differences must be accounted for using a substantiated method. A comparison between two arbitrary voyages does not automatically constitute suitable evidence for this.
The assessment period must also be clear. Fuel savings established at one stable engine operating point do not automatically describe the result over a full voyage. For an annual assessment, savings in the relevant operating modes must be linked to the hours in which they actually occur. Periods without WHRS production, any start-up and shutdown losses, maintenance downtime and electrical output limits are also included.
Another important consideration is the definition of percentage fuel savings. Savings expressed as a percentage of the original fuel consumption of one generating set may differ substantially from the same savings expressed as a percentage of the ship's total fuel consumption. For every stated fuel saving percentage, it must therefore be clear which original consumption is used as the reference. Without that baseline, technically correct percentages may still give a misleading picture of the effect on the ship's complete energy supply.
Forecast savings and savings established in service must also be distinguished. A calculation based on engine data, expected generator loads and the operational profile may be suitable for assessing potential savings in advance. Savings established afterwards can be supported by recorded energy and fuel data. Even then, however, a sufficiently reliable reference case without a WHRS remains necessary, because the net electricity generation actually measured is not automatically equal to the amount of fuel consequently not consumed.
If the reference case is calculated rather than measured directly, the starting conditions and assumptions used must remain identifiable. Measurement uncertainty and the sensitivity of the result to changing operating conditions also merit attention. The final assessment must therefore show which fuel-fired energy production has actually been replaced, how much less fuel the engines concerned consequently use and which additional effects influence the outcome.
Fuel savings from electricity production using waste heat are thus determined by the ship's complete energy supply. The heat supply and ORC performance determine how much net electricity becomes available, while load distribution, specific fuel consumption and the use of other energy sources determine how much fuel is actually avoided as a result.
Higher WHRS electricity production therefore does not automatically lead to proportionally greater fuel savings. The ship must demonstrably consume less fuel under the new operating conditions. Only when the net electrical contribution, energy production actually replaced and total fuel consumption without and with the WHRS have been assessed under sufficiently similar conditions can fuel savings be reliably established.
This Article in the Series
Net electricity generation from a WHRS leads to fuel savings only when existing energy sources actually consume less fuel. How those savings contribute financially is discussed in the economic viability of waste heat recovery. Here, the additional costs of a WHRS retrofit must also be included. The balance between avoided fuel costs, total project costs and financial risks is central to the business case and investment decision for WHRS installations.