Inverters: From "DC–AC Converters" to "Active Supporters" of New-Type Power Systems
The functional definition of an inverter is simple and clear: it converts direct current (DC) into alternating current (AC). PV modules output DC, and storage batteries charge and discharge in DC, while the grid and the vast majority of loads operate on AC-the inverter stands precisely at the boundary between these two forms of electricity. But it is this "boundary position" that determines that its role in the energy system is far more than that of a converter. When the share of PV and wind power in the power system crosses a certain critical point, the behavior of inverters begins to affect grid stability itself, and their technical routes, topology choices, and control strategies rise from equipment-level engineering issues to system-level strategic issues.
To understand this shift, one must first return to the basic action of how an inverter "turns DC into AC."
Pulse-Width Modulation: "Cutting" a Sine Wave with Switches
The voltage and direction of DC are constant, while the voltage and direction of AC change periodically. The core means by which an inverter achieves this conversion is pulse-width modulation (PWM): power switching devices are repeatedly turned on and off at a speed far higher than the output frequency, "cutting" the DC voltage into a series of pulses of varying widths, which are then smoothed by a filter circuit into a waveform close to a sine wave. Sinusoidal pulse-width modulation (SPWM) is the most classic method-a low-frequency sinusoidal modulating wave is compared with a high-frequency triangular carrier to generate a pulse sequence that drives the switching devices, so that the width of the output pulses varies according to a sinusoidal law. In a three-phase system, three modulating waves are 120° apart in electrical angle, and a three-phase symmetrical AC voltage is obtained at the output.
This principle appears simple, but "what device to use for switching," "how many levels to use to approximate a sine wave," and "at what frequency to switch"-each choice involves efficiency, cost, reliability, and grid adaptability, and also constitutes the underlying logic of inverter topology evolution.
Topology: Finding the Optimal Solution Between Two-Level and Multi-Level
The most basic two-level three-phase inverter consists of six switching devices forming three phase legs, with each leg having only two states-"connected to the DC positive bus" and "connected to the DC negative bus." It has a simple structure and low cost and is the mainstream choice for small- and medium-power scenarios. However, the voltage waveform jumps of a two-level output are large and the harmonic content is high, requiring a considerable filter circuit to "clean up" the waveform.
The emergence of three-level topologies-especially the neutral-point-clamped (NPC) type-largely responds to this contradiction. Each phase leg has additional clamping diodes and switching devices, and the number of output levels increases from two to three, making the waveform closer to a sine wave, reducing harmonics, and easing the filtering burden; at the same time, the voltage stress borne by each switching device is only half of the DC bus voltage, giving it an advantage in high-voltage, high-power scenarios. The cost is increased circuit and control complexity, and DC bus neutral-point potential balancing becomes an issue that must be additionally handled. Further expansion of the multi-level family-flying-capacitor type and cascaded H-bridge type-each has its own advantages at higher voltage levels and under modularization requirements.
The choice of topology is never "the more advanced, the better"; rather, it must match the power level, voltage platform, cost constraints, and reliability requirements of the application scenario. Ground-mounted PV plants tend toward high-power centralized or string-type solutions, commercial and industrial and residential scenarios are more sensitive to volume and flexibility, and energy storage systems require inverters to have bidirectional power flow capability-the same core function of "DC–AC conversion" has grown into significantly different product forms in different scenarios.
Beyond PV: An Underestimated Application Landscape
PV grid-connected inverters are undoubtedly the most closely watched segment of the inverter industry today, but inverter applications go far beyond this. In industry, variable-frequency inverters drive motor speed regulation and are a basic means of energy saving in manufacturing; the inverter link in uninterruptible power supplies (UPS) for data centers ensures continuous operation of critical loads during grid fluctuations; in electric vehicles, on-board inverters convert the DC power of the traction battery into the AC power required to drive the motor, while V2L (vehicle-to-load) functionality turns the vehicle into a mobile AC power source-related inverter module shipments exceeded 20 million units in 2025, and the penetration rate of new models equipped with bidirectional charging and discharging functions is expected to reach 40% in 2026. The power conversion system (PCS) is the "heart" of an energy storage system, undertaking battery charge/discharge control and bidirectional AC–DC conversion and accounting for about 15% to 20% of the total cost of an energy storage system.
These scenarios share the underlying technology of power electronic conversion, but the weighting of efficiency, power density, response speed, and environmental adaptability differs from case to case. It is precisely this pattern of "common technology, differentiated demand" that gives the inverter industry both a foundation for large-scale manufacturing and sustained vitality for scenario-based innovation.
Silicon Carbide and Grid-Forming: Two Technical Variables Rewriting the Rules
If topology determines the "skeleton" of an inverter, power devices and control strategies determine its "muscles" and "nerves." The two most noteworthy changes currently taking place happen to occur at these two levels.
At the device level, silicon carbide (SiC) is moving from a "high-end option" to a "standard configuration." The switching loss of SiC devices can be reduced to about 30% of that of silicon-based IGBTs, and there is almost no reverse-recovery tail current, enabling switching frequencies to leap from the 8–20 kHz of the IGBT era to 50–100 kHz or even higher. This means filter inductors can be significantly reduced in size, power density increases, and system efficiency rises accordingly. In energy storage PCS, the total loss of 125 kW-class products based on SiC modules can be reduced by 20%–30%; in centralized utility-scale storage scenarios, the turn-off loss of SiC PCS is about 47% lower than that of IGBT-based PCS, and the maximum system efficiency exceeds 99.3%. Companies such as Huawei, Haier, and CRRC Zhuzhou Institute have concentratedly launched SiC-based PV-storage products in 2026, and the industry expects the penetration rate of SiC PCS in the energy storage sector to reach 10% that year.
At the control level, the more systemically significant change is the shift from "grid-following" to "grid-forming." Traditional grid-following inverters use grid voltage and frequency as references to inject current and are essentially "followers"; when the penetration of new energy rises and synchronous generators are largely replaced, system inertia and voltage support capability decline, and mere "following" is no longer sufficient to maintain stability. Grid-forming inverters actively provide voltage and frequency support to the grid by simulating the inertia and damping characteristics of synchronous generators. This shift is moving from a technical initiative to a regulatory requirement: the national standard "General Technical Specification for Grid-Forming Converters" was released in 2026 and will be implemented in November, specifying 19 key indicators including inertia and damping, phase-angle jump response, and wideband impedance characteristics. Grid-forming capability currently still carries a premium of 10%–30%, but as grid connection guidelines tighten, it is becoming an entry threshold for new utility-scale projects.
Market: Duopoly, String-Type Dominance, and the Rise of Energy Storage PCS
The global PV inverter market shows a clear oligopolistic structure. In 2025, Huawei ranked first with a share of more than 20.5%, followed closely by Sungrow, with the two together accounting for about 55% of the global share; the top five companies (Huawei, Sungrow, Ginlong Solis, GoodWe, and Growatt) together accounted for about 60%. In terms of shipment structure, string inverters have established a dominant position, accounting for more than 68% of global shipments, the centralized share has contracted to around 20%, and the penetration rate of microinverters in residential distributed scenarios continues to rise. The driving force behind this structural change comes from the expansion of distributed PV and the inherent advantages of string solutions in MPPT granularity and O&M flexibility.
Energy storage PCS constitutes a segment of the inverter industry with a steeper growth slope. The commercial and industrial bidirectional energy storage inverter market is expected to grow from US$685 million in 2025 to US$1.746 billion in 2032, with a compound annual growth rate of 14.5%. In terms of technical routes, the shipment share of string-type PCS is moving from 30% in 2025 toward 35%–40% in 2026, and 400 kW+ high-power string-type products are becoming a new focus of competition. The competitive landscape is also loosening: companies such as Sungrow, BYD, and Envision, which previously focused mainly on self-supply or system integration, have begun to supply PCS externally; battery cell companies and system integrators are extending upstream to core components, while third-party PCS manufacturers face pressure from a narrowing "ceiling." The gross margin structure of PCS also confirms its value position-gross margins for residential storage PCS/microinverters are 30%–33%, and those for utility-scale storage PCS are 20%–22%, significantly higher than those of battery cells and domestic system integration links.
Conclusion
The development trajectory of inverters reflects the evolving role of power electronics technology in the energy transition. It was initially defined as a "conversion device," and its core metric was conversion efficiency; later it became a "control node" of PV and energy storage systems, where MPPT tracking, battery management, and grid connection compliance all converge; now, with the implementation of grid-forming requirements and the spread of SiC devices, it is becoming a "grid equipment" that undertakes active support functions in new-type power systems. The direct impact of this role leap is that competition in inverters is no longer merely a contest of efficiency figures and cost per watt, but rather a test of depth of understanding of grid needs, adaptability to application scenarios, and collaborative capability across the entire chain of devices–topology–control–system integration. For this component that once "hid behind modules and batteries," this is both pressure and an opportunity to move toward center stage.