Compressed Air Energy Storage

Compressed air energy storage (CAES), in two different forms, underground and underwater, is a technology that has been used for decades. In CAES plants, energy is stored in the form of potential energy of compressed air that is accumulated in

TABLE 6.5

Hydrodynamic Losses for Considered Floats

Source: Bassett, K., et al.../. Energy Storage, 14,256-263, 2017.

either rigid, underground tanks or flexible, underwater balloons. CAES units operate like turbines except that the compression and expansion cycles occur at different times [II]. CAES has critical advantages compared to other storage technologies. They are relatively low-cost systems that can store large amounts of energy for medium to large scale power [11,18,19]. When compared to other forms of energy storage, CAES is generally cheaper than most, size for size [20]. CAES can be used in a large number of places. It is scalable and environmentally friendly. It can be used both inland and offshore. The biggest issue that is associated with CAES is the overall low efficiency that comes from the losses that occur during the conversion processes (roundtrip efficiency of approximately 60%-70% [21]). It is generally known that losses would occur within any type of conversion, storage, and heat transfer cycle, so it should come as no surprise that the efficiency is relatively low. Like all other storage systems discussed in this chapter, CAES technology can be mixed with alternative energy harvesting technologies. So far, two concepts have been proposed to mix CAES into (i) solar energy and (ii) vortex hydrokinetic energy harvesters to store excess energy during low demand times to regulate the power output.

CAES-Integrated Photovoltaic Plant

The concept of floating storage-integrated photovoltaic (PV) plants using underwater CAES is proposed by Cazzaniga et al. [19]. As shown by Cazzaniga et al. [19], efficiency of CAES increases to approximately 80% when it is mixed into floating PV panels. The floating aspect of this concept alleviates some negative impacts of the PV plants, including the need for large land areas, which are driven by the low power density of solar energy. In the literature, there are two different systems that are proposed for floating PV arrays: (i) a raft structure that will support the PV panels with a pontoon system on the surface of the water and (ii) a submerged system that uses floats to hold the PV panel at a certain depth of water while it is anchored onto the bottom of the body of water [19]. For the storage module, underwater compressed air storage is preferred over well-developed batteries because the batteries are costly with a shorter lifetime and cause environmental and disposal difficulties. Underwater CAES, however, is environmentally friendly, inexpensive, and scalable. The proposed design was taken a step further by making the compressed air reservoir an “integral part” of the platform [19]. Pontoons of the support structure were used as the accumulators of high-pressure air to reduce costs and required space. In this technology, air is compressed through an isothermal process so the heat created through the compression phase can be efficiently transferred to an external thermal bath. The isothermal process is facilitated by reducing the speed of the process and providing large heat transfer interfaces. Another version of this technology is also presented where the compression occurs through an adiabatic process, using large insulated, high-pressure reservoirs. The adiabatic version of this technology does not require a thermal energy storage [19]. Mixed approaches that combine features from both adiabatic and isothermal compression have been used, too.

Working principles of the isothermal version of this technology are: (i) the floating PV platform will supply electric energy through modules and inverters to the grid or air compressor, (ii) the air compressor will increase the pressure in the

FIGURE 6.3 Block flow scheme for compressed air energy storage (CAES)-integrated floating photovoltaic (PV) plants. (From Cazzaniga, R., J. Energy Storage, 13,48-57, 2017.)

pontoons from 0.1 to 20 MPa, and (iii) the compressed air can be used whenever necessary to produce energy through a turbine. The heat stored in the thermal bath will be added to the compressed air before it enters the turbine during the discharge process. The flow scheme for the system is given in Figure 6.3. This scheme shows how the system takes in energy from the PV array while having access to the energy from the compressed air storage module. The direction of power flow is all controlled by a power controller. The amount of energy stored in the pontoons is calculated as:

where R is the gas constant, P_pj and P_IK, are the initial and final air pressure at the pontoons, and T_ca is the temperature of the compressed air. Equation (6.2) was used to identify the optimal range of pressure for the pontoons. Since the description is assumed for a perfectly isothermal process, adjustments were needed because it is not possible to have a perfectly isothermal process in practice. To ensure the compression process in practice is as close as possible to an isothermal process, all the work done by mechanical forces was transferred to a heat reservoir (thermal bath). This heat will be injected back into the compressed air flows before it expands within the turbine to release the stored energy.

Storage-Integrated Vortex Hydrokinetic Energy Converter (SAVER)

Storage integrAted Vortex hydrokinetic Energy ConverteR (SAVER), a modified version of the device described by Vasel-Be-Hagh et al. [22], uses the best features of two existing technologies: (i) vortex hydrokinetic energy harvester [23,24] and (ii) underwater compressed air energy storage [21]. The energy harvesting module of the proposed device is a device that converts kinetic energy of vortex-induced vibrations (VIV) into electrical power. VIV is a naturally occurring phenomena that happens to solid bodies in cross-flow. As water flows over a bluff body, vortices shed off of the surface in what is known simply as vortex shedding. Vortex shedding causes the bluff body to oscillate.

How SAVER Works

The process flow diagram of SAVER is shown in Figure 6.4. The system consists of five main subsystems including the air compression unit, transmission pipeline unit, expansion unit, heat recovery unit, and most importantly, the submerged

FIGURE 6.4 Overview of the proposed power-generating storage plant: (#1) compressor unit, (#2) starter supply, (#3) rigid compressed air pipes, (#4) heat exchanger no. 1, (#5) heat tank, (#6) flexible compressed air lines, (#7) a charged accumulator-converter (A-C), (#8) crankshaft, (#9) gearbox, (#10) generator driven by oscillating A-Cs, (#11) AC-DC converter, regulator, inverter, and filter, (#12) heat exchanger no. 2, (#13) turboexpander driven by compressed air released to quickly respond to high demands, and (#14) generator driven by turboexpander. To evade the effects of the upstream units on the vortex-induced vibrations (VIV) of the downstream ones, each A-C should be placed in a staggered pattern off the wake of the upstream.

accumulator-converter (A-C) unit. In the proposed system, energy is stored in the form of potential energy of compressed air underwater, where a charged compressed air accumulator also plays a second role as an active energy converter. Highly durable flexible cylindrical lift bags are suitable choices to perform as the accumulator-converters (A-Cs) for this hybrid system. See Figure 6.4 for a plant of N SAVER units. Four sample units, that is, 1st, 2nd, (N-l)th and Nth, are shown in the figure. A single SAVER unit with more details is displayed in Figure 6.5. The surplus electrical energy generated in the off-peak hours in addition to the power harvested by charged A-Cs is used to run the compression unit (Figure 6.4: #1). If there is no surplus power from the external grid, and all A-Cs are discharged, the system can be initiated using a starter battery (Figure 6.4: #2). This compresses atmospheric air and stores it in sealed underwater flexible accumulator-converters (A-Cs) at much higher pressures. This is a major modification in comparison with conventional underwater CAES where the accumulator balloons are all open from the bottom and the compressed air would be discharged into the water if the air pressure exceeds hydrostatic pressure at the installation depth. This broadens the application of underwater CAES beyond very deep waters. The terminal pressure depends on the strength of material used for manufacturing the A-C balloons. The temperature of air increases during the compression process, hence, using a plate heat exchanger (Figure 6.4: #4); the heat added to the air will be taken and stored in an insulated heat tank (Figure 6.4: #5) before transferring the compressed air into underwater A-Cs where compressed air will reach a thermal equilibrium with cold water. Storing heat of compressed air in a heat tank pushes this process toward an adiabatic condition and significantly reduces the thermal losses, which

FIGURE 6.5 A charged balloon performing as the accumulator-converter of the proposed Storage-integrAted Vortex hydrokinetic Energy converteR (SAVER). Rigid rods are strapped to the flexible pressurized balloon to prevent deflection under hydrostatic pressure and dynamic loads. The perforated plate installed at the top would ensure that in case of bursting of the balloon, which is very unlikely, any generated vortex ring is dispersed into smaller spherical bubbles before it gains any momentum. (See Bernitsas, M., et al., J. Offshore Mech. Arctic Eng., 131, 1-13, 2009 for further details.) leads to higher efficiencies. Cold compressed air is stored in the A-Cs (Figure 6.4: #7) through a rigid (Figure 6.4: #3) and flexible (Figure 6.4: #6) piping until all A-Cs are charged. A-Cs are always locked during the charging process to avoid technical complications; hence, they do not contribute to the plant power generation until they are fully charged w'ith compressed air at a predefined terminal pressure. Once an A-C is fully charged, it is unlocked to start its linear oscillations in response to the VIV from passing water currents. This linear motion is converted to rotary motion using a crankshaft mechanism (Figure 6.4: #8), the speed of which is regulated using a gearbox (Figure 6.4: #9). The crankshaft allows the use of electric machines that do not use rare Earth magnets and can be modularly installed with relative ease. It also offers the capability of placing the generators above the water for easier maintenance. The rotary motion drives a generator (Figure 6.4: #10). The generator output is passed through an AC-DC converter consisting of an electronic rectifier, regulator, inverter, and filter (Figure 6.4: #11) to produce a 60 Hz power signal that could interface with the grid. Electrical power produced by this generator is transmitted either to the compressor unit (if there are still uncharged A-Cs) or to the external grid (if all available A-Cs are charged). If more power is urgently required during peak hours, A-Cs can be locked one by one, and the stored compressed air can be quickly discharged through a turbine (Figure 6.4: #13) with the help of hydrostatic pressure of water that continuously acts on compressed air stored in flexible A-Cs. The electrical power generated using released compressed air is then transmitted to the grid to meet the demand. Note that the delivered cold compressed air from A-Cs is first heated by passing through the plate heat exchangers (Figure 6.4: #12) before entering the turbine. Heat exchanger liquid supplied from the heat tank (Figure 6.4: #5) flows in the reverse direction to heat this cold air (60% ethylene glycol solution can be used as the heat exchanger liquid).

Advantages of SAVER

The proposed device mixes two existing cost-effective, simple, environmentally friendly, and scalable technologies (underwater compressed air energy storage [UWCAES] and vortex hydro energy harvester [VHEH]) to develop a highly efficient device, that:

• 1. Costs in the same order of traditional hydro energy harvester plants (in utility scale), while, unlike traditional harvesters, SAVER is both a power generator and an energy storage device
• 2. Does not suffer from intermittency and unpredictability due to its integrated storage module
• 3. Operates at currents as slow as 1 m/s
• 4. Operates at waters as shallow as 5 m, making it a more feasible option than conventional UWCAES (this will expand the renewable resource capacity of the market, as there are more potential areas that can host this device)
• 5. Requires much shorter piping and wiring as it can be installed close to the shoreline, making the installation and maintenance operations much easier and cheaper.

Efficiency Calculations

Energy calculations for a typical SAVER unit are presented. The numerical calculations presented in this section are based off deploying the unit at the Bourne Tidal Test Site, located in the Cape Cod Canal. Characteristic values for this typical unit are given in Table 6.6. Ultimately, the roundtrip efficiency for a single unit is obtained. While in practice, the SAVER unit would be used in an array system and one needs to take into account the negative effects of hydro-mechanic wakes on the efficiency of the downstream units. The following calculations should give an idea of what should be reasonably expected.

Step 1: Calculating the storage capacity of a single SAVER unit E_s,„_rrj [MJ] The storage capacity of a single SAVER unit is calculated via the following equation:

By substituting the data provided in Table 6.6 into Equation (6.3), the amount of energy that can be stored in the storage module of each unit is calculated as E_s,_ored = 7.72 MJ.

Step 2: Calculating the amount of energy that can be delivered from storage module [MJ]

The amount of energy that can be delivered from the storage module is calculated as E_m„ = r)_round_,_rip E_Slored. Roundtrip efficiency of a conventional UWCAES unit is reported to be approximately 65% [26]. Although

TABLE 6.6

Characteristic Values for a Typical Unit

this value changes with isentropic efficiencies of used turbines and compressors. Using heat exchangers that capture more thermal energy would also increase system performance, although to a lesser extent compared to the impact of isentropic efficiencies of turbine and compressor units. The authors, however, calculated the target level of performance for a single SAVER unit assuming that the roundtrip efficiency of the storage module of this device is i)_romd4rip = 55%, approximately 10% less than what is reported in the literature to account for uncertainties and other potential inefficiencies that might cause a lower roundtrip efficiency. Therefore, = 3.68 MJ.

Step 3: Calculating the charge/discharge period(s)

Charge: Mass of compressed air stored in a charged SAVER unit is calculated as m = p₂V/ RT₂. Intake mass flow rate at the compressor unit is calculated as m = pQ = p / RTQ. Therefore, charge period can be calculated as:

In other words, charge period depends on the pressure ratio (pressure of stored air divided by atmospheric pressure), temperature ratio (atmospheric air temperature divided by water temperature, assuming thermal equilibrium between water and stored air), storage volume of each SAVER unit, and compressor flow rate in CFM. Therefore, using data given in Table 6.6:

Discharge: Discharge period depends on the discharge rate, which is controlled by the operator who sets it based on the instantaneous demand. The discharge period is usually longer than the charge period.

Step 4: Calculating energy that can be harvested by a single charged SAVER unit

Total hydrokinetic power available in water current is calculated as:

where p, A, and U are density of water, swept area of the oscillating A-Cs, and current speed, respectively. Swept area is calculated as A = WL, where

L represents length of each A-C unit and W is defined as W = H + D, in which H is the amplitude of the oscillations and D is the diameter of each cylindrical balloon.

According to [27] energy conversion efficiency of conventional vortex hydro energy harvester is approximately tj_conv = 22%. However, the authors assumed an energy conversion efficiency of r_com, = 17%, which is 5% less than values reported in the literature to account for any potential additional losses. Therefore, power harvested by the single SAVER unit described in Table 6.6 at a current speed of 1 m/s is:

Due to complications caused by the variation of mass during charge/dis- charge periods, which causes a continuous change in the optimal required set of spring/damper system for the device, the cylinder is locked during the charging/discharging process. Power production starts when the cylinder is fully charged and continues until there is a need for quick power release by quickly discharging compressed air stored in the cylinder (beginning of discharge process).

Step 5: Calculating roundtrip efficiency

To calculate the roundtrip efficiency of the integrated system, it is critical to know how much energy (J) was produced by power harvester module of the SAVER. The amount of energy produced by the power harvester module of SAVER directly depends on the period between the end of charge process and the beginning of discharge process. The longer this period of power production lasts, the more energy can be harvested for making up for losses. Here the roundtrip efficiency of a single SAVER unit is calculated for two different cases: (i) assume the power production period is in the same order of the charge period, and (ii) the power production period is one order of magnitude larger than charging period.

Sample 1: Assume the power production period (i.e., the period between the end of charging and the beginning of discharging) is in the same order of the charge period. Then, using the charge period calculated at step 3, and power production rate calculated at step 4, the harvested energy is:

Therefore, the roundtrip efficiency of the unit is:

Sample 2: Normally, the power production period is one order of magnitude larger than charging period, approximately equal to the off-peak duration, which varies between 7 and 10 hours depending on the season. Assuming power production period equal to approximately 8 hours:

Leading to:

Therefore, not only are all losses compensated, but also an additional amount of electrical energy equal to 2.67 E_slored is produced during an 8-hour period of operation of the proposed SAVER.