13.1 Context: When the Best Storage Solution Is Not an Electrochemical Battery
Battery storage often receives the most attention in power access discussions, but it is not always the best storage solution. Many facilities do not need stored electricity for every load. They need to reduce the electrical demand created by cooling, refrigeration, process temperature control, short-duration power interruptions, or equipment transitions. In those cases, thermal storage or other non-battery storage may solve the problem with lower cost, longer life, simpler operation, or better fit with the facility’s load profile.
Thermal storage does not store electricity directly. It stores heating or cooling capacity so electric equipment can run at a different time from when the thermal service is needed. A chilled-water tank can produce cold water at night, when utility capacity may be available, and use that stored cooling during the afternoon peak. Ice storage can make ice during off-peak hours and melt it later to offset chiller load. A cold-storage facility can use product, air, building mass, or dedicated storage to shift part of refrigeration demand, provided temperature limits and product quality are protected.
This distinction matters in a power access acceleration project. If the facility’s peak is driven by chillers, refrigeration compressors, HVAC recovery, or process cooling, an electrical battery may be an expensive way to solve a thermal problem. A thermal solution may reduce the peak more directly. It may also avoid some battery-specific issues such as degradation, fire safety concerns, replacement cycles, and inverter limitations.
Non-battery storage can also include flywheels, compressed air, gravity systems, and other technologies. In facility projects, flywheels are most often relevant for very short-duration ride-through, power quality, or UPS support. They are not a substitute for long-duration energy supply, but they can protect sensitive equipment during brief disturbances or support transfer between power sources.
The executive question is not, “Should we install a battery?” The better question is, “What form of storage best matches the load we are trying to shift or protect?” If the constraint is thermal, thermal storage deserves early review. If the constraint is a short power interruption, a flywheel may be useful. If the constraint is multi-hour facility-wide operation, batteries, generation, or hybrid systems may still be required.
Thermal and non-battery storage also fit the broader principle of matching the solution to the constraint. If the power gap lasts several years, the facility may need assets that remain useful after full grid service arrives. A chilled-water tank may continue to lower demand charges and improve resilience. Improved refrigeration control may remain valuable for product protection. A flywheel may continue to support critical controls or UPS systems. This residual value can make non-battery options attractive even when they do not solve the entire power access problem alone.
13.2 Methodology: Evaluating Chilled Water, Ice Storage, Thermal Mass, Flywheels, and Other Non-Battery Options
The methodology begins by identifying the type of demand the storage system must address. Storage should not be evaluated as a generic technology choice. It should be tied to the specific load, operating window, power constraint, and business requirement.
Step 1: Identify the target load. The team should determine whether the peak is driven by HVAC, chilled water, refrigeration, process cooling, heating, compressed air, motor starts, UPS events, or another load. Thermal storage is most relevant when a large portion of the facility’s electrical demand is used to create or maintain a temperature condition.
Step 2: Quantify the thermal load profile. The analysis should estimate cooling or heating demand by hour, day, season, and operating phase. For existing facilities, use utility data, building automation data, chiller logs, refrigeration data, temperature records, and production schedules. For new facilities, use design loads, weather assumptions, equipment specifications, and operating scenarios. The team should identify how much of the peak can realistically be shifted without compromising the business.
Step 3: Define the storage objective. The objective may be to reduce a utility demand cap, avoid operating chillers during peak hours, reduce generator size, support commissioning, improve resilience, stabilize refrigeration, or lower demand charges. A two-hour peak-shaving objective leads to a different design than an eight-hour cooling support objective.
Step 4: Evaluate chilled-water storage. Chilled-water tanks store cold water produced by chillers during off-peak hours. They can be attractive for campuses, data centers, hospitals, laboratories, and industrial sites with central chilled-water systems. The evaluation should consider tank size, site space, piping, pumping energy, temperature differential, controls, water treatment, integration with existing chillers, and whether the stored cooling aligns with peak demand.
Step 5: Evaluate ice storage. Ice storage generally stores more cooling in less volume than chilled water because it uses the phase change from ice to water. It can be useful where space is constrained or where cooling demand is concentrated during specific windows. The trade-off is that making ice may require lower evaporator temperatures, which can reduce chiller efficiency. The economics depend on demand reduction value, electricity rates, equipment cost, and operating schedule.
Step 6: Evaluate thermal mass and pre-cooling. Some facilities can use the building, product, water loops, or process materials as temporary storage. Pre-cooling a warehouse, refrigerated zone, or process stream before peak hours may reduce compressor or chiller demand later. This must be tested carefully. Temperature, humidity, product integrity, worker comfort, condensation, and regulatory limits may constrain how much flexibility exists.
Step 7: Evaluate flywheels and short-duration storage. Flywheels store energy mechanically in a rotating mass. They can respond quickly and are useful for short ride-through, voltage events, or UPS applications. They are generally not designed for multi-hour peak shifting. The team should evaluate duration, power rating, maintenance, safety enclosure, integration with UPS systems, and whether the problem is truly short-duration.
Step 8: Model charge and discharge windows. Thermal storage must be charged before it can be used. The team should model when chillers, refrigeration equipment, or other systems will create the stored cooling, how much electricity is required during charging, and whether charging creates a new peak. The model should include weather, production schedules, utility limits, bridge power, and operating restrictions.
Step 9: Design controls and operating procedures. Storage value depends on control discipline. The system must know when to charge, when to discharge, what temperature limits apply, which equipment is locked out, and how operators override controls during abnormal conditions. Controls should integrate with the building management system, refrigeration controls, energy management system, or microgrid controller where relevant.
Step 10: Evaluate site, construction, and maintenance requirements. Thermal storage may require tanks, foundations, piping, insulation, pumps, valves, heat exchangers, water treatment, structural support, and access for maintenance. Ice systems require specialized equipment and controls. Flywheels require appropriate siting, safety protection, and service procedures. The construction path should be tested against the facility launch schedule.
Step 11: Compare economics and power access value. The business case should include capital cost, installation, controls, maintenance, pumping energy, efficiency impact, utility bill savings, avoided generator or battery capacity, avoided delay, incentives, and residual value. The ranking should reflect both lifecycle economics and the immediate value of getting the facility operating under a constrained power supply.
The team should also decide whether the storage system is a temporary bridge asset, a permanent operating asset, or a hybrid investment. Temporary assets should be judged on speed, rental or resale options, and demobilization complexity. Permanent assets should be judged on lifecycle value, integration with the final utility service, maintenance staffing, and future expansion. Hybrid assets should be designed so they provide immediate demand relief now and continue to support cost, resilience, or sustainability goals later.
The final deliverable is a non-battery storage options memo. It should state the target load, storage technology, peak reduction potential, duration, required equipment, site impact, controls approach, cost range, schedule, operating risk, and whether the option changes the utility request or bridge power requirement.
13.3 Skills Required: Mechanical Engineering, HVAC Design, Refrigeration, Thermal Modeling, Controls Engineering
Mechanical engineering is the core skill for thermal storage. The engineer must understand heat transfer, pumping, piping, equipment sizing, thermal losses, temperature differentials, and constructability. This role determines whether the storage concept can physically deliver useful heating or cooling when the facility needs it.
HVAC design expertise is required for facilities where cooling and ventilation drive demand. HVAC specialists evaluate chillers, air handlers, chilled-water loops, economizers, humidity control, setpoints, and comfort requirements. They also understand how thermal storage changes system operation across a full day rather than only at peak hour.
Refrigeration expertise is essential for cold-storage, food, pharmaceutical, and process facilities. Refrigeration specialists understand compressors, evaporators, condensers, defrost cycles, suction pressure, door openings, product temperature, and regulatory limits. They can distinguish safe flexibility from changes that would put product quality at risk.
Thermal modeling expertise converts design assumptions into an hourly operating view. The modeler estimates storage capacity, charge and discharge rates, weather sensitivity, temperature drift, pumping energy, and peak reduction. This work should be linked to the facility’s 8760 load model and ramp-up plan.
Control engineering is required because storage only works if equipment operates in the right sequence. Control specialists design charge logic, discharge logic, demand caps, temperature safeguards, alarms, operator overrides, and integration with building or energy management systems. The team may also need structural, civil, commissioning, maintenance, EHS, and financial expertise.
13.4 Terminology: Thermal Storage, Chilled-Water Tank, Ice Storage, Thermal Load, Flywheel, Discharge Window, and Pre-Cooling
Thermal storage: A system that stores heating or cooling capacity for later use. It shifts when thermal energy is produced, reducing electric demand during constrained periods.
Chilled-water tank: A tank that stores cold water produced by chillers. The stored water can later provide cooling while chillers operate less or not at all.
Ice storage: A system that makes ice during one period and melts it later to provide cooling. It can store significant cooling in a smaller volume than chilled water.
Thermal load: The heating or cooling demand required to maintain a space, product, process, or equipment condition.
Flywheel: A mechanical storage device that stores energy in a spinning mass. It is typically used for very short-duration power support or ride-through.
Discharge window: The period when stored energy or stored cooling is released to support facility operation or reduce grid demand.
Pre-cooling: Cooling a space, product, or thermal system before a peak period so cooling equipment can run less during constrained hours.
Charge window: The period when storage is filled, such as when chillers make chilled water or ice before the peak period.
Temperature drift: The change in temperature that occurs when active cooling or heating is reduced. Acceptable drift depends on product, process, comfort, and compliance requirements.
Phase change: The transition between physical states, such as water freezing into ice or ice melting into water. Phase change can store or release significant thermal energy.
