What does a distribution system fear most? Power outages. But the words “power outage” mean completely different things to different people — for a data center, one second of outage can lose a batch of data; for a hospital operating room, one second can endanger a life; for a continuous production line, one minute can scrap an entire batch of products. For this reason, “backup power” becomes an unavoidable core topic in distribution system design. But the technical routes for backup power go beyond one — UPS, EPS, diesel generators, and energy storage each have their own “personality” and “temperament.” Choose correctly, and they can save the day in a critical moment; choose incorrectly, and they fail when you need them most. Let’s first clarify the concepts — what exactly does each of the four solutions “back up”? A UPS (Uninterruptible Power Supply) backs up “uninterruptibility” — zero transfer time between utility and backup, so the load never senses an outage. A UPS contains a rectifier, inverter, and battery; during normal operation, utility power is rectified and inverted to supply the load while charging the battery; when utility power fails, the battery continues supplying through the inverter with zero transfer time. UPS output is high‑quality, voltage‑ and frequency‑stabilised AC, suitable for protecting power‑quality‑sensitive loads (servers, precision instruments). An EPS (Emergency Power Supply) backs up “emergency” — after a utility outage, EPS switches to battery supply within tens of milliseconds to several seconds, mainly for fire equipment, emergency lighting, and lifts. The core differences from UPS: UPS pursues “zero transfer,” EPS allows “short interruption”; UPS outputs high‑quality AC, EPS typically outputs emergency power with lower power quality requirements. A diesel generator backs up “long duration” — after a utility outage, the diesel generator starts and takes load within seconds to tens of seconds, and can run continuously for hours or even days (as long as fuel is available). Diesel generators can be built with large capacities (from tens of kW to several MW), making them the mainstay for long‑duration backup. Their disadvantages: long starting time, high noise, fuel storage requirements, and periodic maintenance. An energy storage system (battery storage) backs up “flexibility” — it can provide short‑duration backup (replacing UPS), long‑duration backup (partially replacing diesel generators), and also perform peak‑valley arbitrage, demand management, and dynamic capacity expansion. Energy storage responds within milliseconds, capacity can be flexibly expanded, but initial investment is relatively high. A comparison of the four solutions: in transfer time, UPS is zero (0ms), EPS is tens of ms to seconds, diesel generators are seconds to tens of seconds, and energy storage is millisecond‑level. In backup duration, UPS provides 15‑30 minutes (battery), EPS provides 30‑90 minutes (battery), diesel generators provide hours to days, and energy storage provides 1‑4 hours (expandable). In output power quality, UPS is extremely high (stabilised voltage and frequency), EPS is standard, diesel generators are standard, and energy storage is high. In capacity range, UPS spans a few kVA to several MVA, EPS spans a few kW to hundreds of kW, diesel generators span tens of kW to several MW, and energy storage spans tens of kWh to tens of MWh. In initial investment, UPS is medium‑high, EPS is medium, diesel generators are high, and energy storage is high. In operating cost, UPS is low, EPS is low, diesel generators are high (fuel + maintenance), and energy storage is low. In noise, UPS is low, EPS is low, diesel generators are high, and energy storage is low. In environmental performance, UPS is good, EPS is good, diesel generators are poor (emissions), and energy storage is good. The “four‑question method” for selection decisions: ask yourself four questions. First, how much outage time can your load tolerate? If the answer is “not even one second” (data centers, operating rooms, precision manufacturing), UPS or energy storage is the only choice. If the answer is “a few seconds is acceptable” (fire pumps, emergency lighting, lifts), EPS or diesel generators will do. If the answer is “a few minutes is acceptable” (general production lines, office air conditioning), a diesel generator is sufficient. Second, how long do you need to back up? If you only need to “hold until the generator starts” (typically 15‑30 minutes), UPS or EPS is enough. If you need to “hold until utility power returns” (possibly hours), diesel generators or energy storage are better. If you need to “last through the entire day” (e.g., solar power at night), energy storage is the only choice. Third, how sensitive is your load to power quality? Servers, precision instruments, and medical equipment are sensitive to voltage fluctuations, frequency deviations, and harmonics — the high‑quality output of UPS or energy storage is essential. Fire pumps, fans, and lighting have low power quality requirements — EPS or diesel generator output is sufficient. Fourth, what is your budget? UPS and EPS have relatively low initial investment but limited backup duration; diesel generators have high initial investment and high operating costs; energy storage has the highest initial investment but can participate in peak‑valley arbitrage and demand management, recovering part of the investment. The optimal solution is often a combination — not choosing just one. In real projects, the most rational approach is often “combined use” — UPS or energy storage for “zero transfer” and short‑duration backup, and diesel generators for “long‑duration” backup. For example: a data center adopts a “UPS + diesel generator” solution — UPS provides 0‑15 minutes of supply after a utility outage, and the diesel generator starts within 15 seconds to take over. Energy storage can replace part of the UPS function while also participating in peak‑valley arbitrage. This combined approach addresses both “zero transfer” and “long duration” needs — it is the standard configuration for large critical facilities. Three common pitfalls in selection: first, UPS batteries “expired but not replaced.” UPS batteries have finite lifespans — lead‑acid typically 3‑5 years, lithium typically 8‑10 years. Many users never replace batteries after installing a UPS, only to discover during a real outage that “the UPS only lasts 3 minutes.” Batteries are the most vulnerable part of a UPS — they must be regularly tested and replaced when due. Second, diesel generators “never tested.” Diesel generators require periodic testing — at least monthly no‑load tests, and quarterly load tests. Many units have generators installed for years that have never been started — when a real outage occurs, they either won’t start or won’t take load. Backup power reliability is not “installed” — it is “proven through testing.” Third, ignoring the match between “transfer time” and load sensitivity. Some loads are extremely sensitive to transfer time — for example, contactors release when voltage drops for more than 100ms, and motors stop. If you choose EPS (transfer time of several hundred milliseconds), the motor will still stop. In such cases, the “zero transfer” of UPS or energy storage is the only choice. There is no “best solution” for backup power — only the “most suitable solution.” UPS, EPS, diesel generators, and energy storage — each has its own applicable scenarios and limitations. When making the selection decision, first clarify how much outage time your load can tolerate, how long you need to back up, how sensitive your load is to power quality, and what your budget is — the answers to these four questions are your selection criteria. Choose correctly, and backup power is the “lifeline” in a critical moment; choose incorrectly, and it is the most expensive “decoration” in your electrical room.