1 Introduction

The recent advancements in component size minimization and the availability of commercial off-the-shelf (COTS) components have enabled the development of CubeSat missions with tremendous capabilities in all major areas of science and technology, which were only possible with bigger spacecraft. CubeSats are classified according to their form factor and mass, and they are one of the dominant subclassifications of nanosatellites [1, 2]. One of the core purposes of developing nanosatellites was to educate students on how to design and deploy CubeSats in orbit. Small satellites, like CubeSats, have made it possible to perform science experiments and try out new space technologies at a minimal cost [3]. The very first six CubeSats were launched into space in 2003 on June 30; after that, over 1300 more CubeSats have been launched [4]. Academic institutions launched the most CubeSats in the early years, whereas commercial groups came into CubeSat business in subsequent years. Table 1 below shows the CubeSats that were launched by academic institutes in recent years [5]. CubeSats are frequently launched in low earth orbit (LEO), a low-earth orbit with an altitude of less than 2000 km. The orbital period of LEO generally lies between 90 and 120 min [6]. For the satellites present in low-earth orbit with reference to the equatorial plane, a broad variety of inclination angles (0 to 98 degrees) are accessible. During each orbit, they are viewable from an Earth-based station for a very short period of time. During its orbit, the CubeSat is subjected to both sunlight and eclipsed phases. The CubeSat’s photovoltaic (PV) cells produce power only during the sunlit period. During the sunlit period, PV cells have an angle of incidence ranging from 66.5 degrees to 90 degrees of sunlight illuminating them. The Earth blocks the sun’s light from reaching the CubeSat during an eclipse, resulting in temperature drops. The length of the eclipse is governed by height, inclination, or the angle at which the radiation from the sun impacts the plane of the orbit (β angle) [7, 8]. There is a maximum eclipse duration at = 0. An increase in the value of results in a decrease in the duration of an eclipse. So, there is no eclipse in polar orbits. It is critical for CubeSat EPS, particularly the storage system [9], that the eclipse time be accurately calculated. The successful launch of a CubeSat to Mars orbit has increased interest in using this technology beyond low-Earth orbit (LEO) missions [10, 11]. The CubeSat’s major subsystems are as follows:

  • Attitude determination and control system (ADCS)

  • Communication transmitter (COM TX)

  • On-board computer (OBC)

  • Communication receiver (COM RX)

  • Command and data handling system

  • Electrical power system (EPS)

Table 1 List of some University affiliated Cubesats

In a CubeSat form factor, one of the major challenges is to best fit all the platform subsystems and meaningful payloads in a small and affordable volume. Additional requirements include being able to withstand high radiation levels, vast temperature ranges, and unexpected events [12]. Solar cells orbiting in low Earth orbit typically perform their operations in a temperature range of − 99 to 99 °C, while electronic components operate in a range of − 40 to 80 °C [13]. The CubeSat’s heat regulation is accomplished using both active and passive approaches [14, 15]. This work provides a detailed review of EPS for generalized nanosatellites, including CubeSats. The CubeSat Electrical Power System (EPS) provides uninterrupted power to different subsystems and payloads of the CubeSat. Figure 1 illustrates a basic CubeSat’s EPS configuration with a fixed solar panel array. The typical EPS configuration of CubeSat includes solar arrays connected in parallel, converting light energy from the sun into electrical energy. Power storage is implemented in the form of different types of batteries depending upon the mission requirements for providing power to the CubeSat’s subsystems and payload when the main power generation from the solar panel is unavailable (in case the CubeSat is in an eclipse). Power is regulated and controlled in the form of DC–DC converters and regulators, including the additional functionality of CubeSat’s protective measures in the event of a fault situation.

Fig. 1
figure 1

Essential elements of CubeSat’s EPS

Similarly, the power distribution unit of EPS can include switches to provide regulated power to any of the subsystems and payloads when required.

These switches can either be controlled by the on-board computer (OBC) of the CubeSat or by the EPS board local microcontroller. The maximum power point (MPP) of the CubeSat’s PV panels is measured using a DC–DC converter under various lighting conditions and battery voltages. The battery voltage and voltage at MPP of the PV panel dictate the kind of DC–DC converter that we may use, and that can be a Buck converter, Boost converter, or Buck Boost converter in the power regulation and control unit.

Since each PV panel includes a diode in series, non-illuminated PV panels will not drain the power generated from illuminated PV panels. Generally, a bus containing regulated DC voltage, also called a “main power bus,” distributes the regulated voltage lines from the output of MPPT converters to all payloads and subsystems in CubeSats. The CubeSat PV panels’ arrangement and sizing are normally decided based on the specific requirements of the mission [14].

Similarly, there are volume and weight limitations that limit the PV panel’s capacity. Solar panels on CubeSats are generally placed on their facets. If the power requirements are high and cannot be met with PV panels installed on facets alone, then deployable solar panels are also used along with PV panels installed on facets of CubeSat [16].

The overall solar power production of an orbit is governed by the angle at which sunlight hits the PV panels as well as the layout of the panels altogether (fixed or deployable) [17, 18]. Operating a CubeSat in nadir, partial sun tracking mode, and full sun tracking modes influences the CubeSat’s total power output, which is decided by the ADCS [19]. CubeSats may function in both partial and full sun-tracking modes with the help of deployable panels [16]. Solar power generation is highest in full sun tracking mode because, in this mode, the orientation of PV panels is completely towards the sun [20]. This mode of operation, however, should not be employed indefinitely due to its impact on payload performance and data gathering. As a result, the CubeSat is oriented within 10 degrees off nadir using the partial sun-tracking mode [21]. Power production in this mode is greater than that produced in nadir pointing mode, but still less than that produced in sun tracking mode. CubeSat is directed towards the earth in nadir-pointing mode. The energy storage system is critical during high power demands and during eclipse times. The energy storage system in CubeSats is implemented by utilizing lithiumion cells because of their lower self-discharge rate, more charge/discharge cycles, and greater energy density [22]. But they also have the following drawbacks:

  • Increasing the charge/discharge rate shortens the life of a Li-ion battery [23].

  • At temperatures below 20 degrees Celsius, the energy density of Li-ion batteries is restricted.

  • Temperatures below 32 degrees Fahrenheit may reduce the life expectancy of a lithium-ion battery by as much as 50% [24].

Therefore, low-duty cycle high-power applications like synthetic aperture radar in low Earth orbit (LEO) have performance issues. For higher values of power density, reliability, and enhanced efficiency, especially at low temperatures, CubeSats have turned to hybrid energy storage systems, or ultra-capacitors (UCs) [25], which combine UCs with batteries [26]. To keep the temperature of the batteries within the allowed working range, EPS can also employ a battery warming circuit. The purpose of this article is to give a comprehensive evaluation of all the available EPS designs and organize them into distinct categories while also describing how they work and making qualitative comparisons. Based on the information that is currently available, this comprehensive review provides a high-level overview of different CubeSat EPS designs. These studies will aid researchers in identifying the gaps and best potential power supply designs for CubeSats so that mission-specific comparative studies may be conducted. Furthermore, this article suggests potential study topics related to CubeSat’s EPS for additional research and invention [2, 27].

1.1 Literature reviews

In recent years, there has been significant advancement in power systems for small satellites, including optimization models and protection schemes that enhance system resilience and efficiency. For instance, optimization strategies developed for renewable energy microgrids can be adapted to CubeSat EPS design, where limited power resources must be managed dynamically to meet fluctuating demands.

An example of this study [27] presents a comprehensive optimization framework that jointly considers the design and operational aspects of microgrids coupled with water supply systems, highlighting the synergies between energy and water resources.

Anudeep et al. [28] introduces a protection scheme utilizing differential power measurements to enable selective phase tripping, thereby enhancing the fault resilience of microgrids.

Also, this paper [29] proposes an advanced protection strategy tailored for Doubly-Fed Induction Generator (DFIG)-based wind farm collector lines, addressing specific challenges associated with wind energy integration.

By incorporating the methodologies and findings from these studies, we have strengthened the theoretical foundation of our work and provided a broader perspective on the current advancements in microgrid optimization and protection strategies.

In this paper, we extend these principles to CubeSat EPS, exploring how optimization, fault tolerance, and emerging technologies can be applied to CubeSat power systems. By building on these recent advancements in microgrid optimization and power protection, our work provides a practical and updated perspective on improving CubeSat EPS reliability and efficiency in space missions.

1.2 Contributions and novelty

The primary contributions are summarized as follows:

  1. 1)

    Comprehensive review with a practical perspective:

    • This paper offers a practical, mission-oriented review of CubeSat EPS, addressing challenges in power generation, energy storage, and reliability, in contrast to theoretical studies.

    • We provide design recommendations specifically tailored for CubeSat developers, enabling system designers to utilize these insights for the construction of more resilient CubeSat EPS.

  1. 2)

    Introduction of a new hybrid energy storage system:

    • This study proposes a hybrid storage system integrating ultracapacitors (UCs) with lithium-ion (Li-ion) batteries.

    • This hybrid system enhances power density, energy availability, and reliability, allowing CubeSats to manage power surges with greater efficacy.

    • This section presents case studies and use scenarios that demonstrate the performance improvements realized with this system.

  1. 3)

    Investigation of novel technologies:

    • This study examines and advocates for the implementation of Gallium Nitride (GaN) transistors and multi-junction solar cells as advanced approaches to enhance the performance of CubeSat EPS.

    • This paper provides a practical framework for the application of these technologies to minimize power loss, enhance thermal performance, and increase overall system efficiency.

  1. 4)

    New design guidelines for CubeSat EPS:

    • Recommendations for modular, scalable, and fault-tolerant designs of CubeSat EPS are presented.

    • This framework for future mission design is presented to CubeSat developers by incorporating redundancy, error detection, and hybrid energy storage.

  1. 5)

    Practical insights from actual Cubesat missions:

    • Analyzing MarCO and other interplanetary CubeSat missions reveals essential insights for enhancing the robustness and scalability of future CubeSat EPS.

    • These insights provide mission designers with a framework for interplanetary exploration, emphasizing the necessity for power systems to function effectively in extreme, prolonged space environments.

  1. 6)

    Comparative analysis of MPPT controllers:

    • This analysis of MPPT controllers elucidates the advantages and trade-offs associated with digital and analog MPPT techniques.

    • We provide targeted recommendations regarding the most suitable MPPT system in relation to the CubeSat mission profile.

  1. 7)

    New metrics and models:

This study presents the Mathematical Framework for EPS Design as an innovative technical contribution. It formalizes essential elements of CubeSat power systems, encompassing solar power generation, battery state of charge modeling, and converter efficiency assessment. This methodology offers a mathematical foundation for evaluating and optimizing EPS performance, including both design and operational issues.

This paper transcends theoretical categorization to offer a mission-focused, pragmatic viewpoint on CubeSat EPS. Our work provides a distinctive contribution to the area through a comprehensive review of advanced energy storage, power optimization techniques, and future technologies, aiding developers in the design of EPS systems that are resilient, efficient, and adaptive to the changing requirements of CubeSat missions.

Unlike previous works, such as “A Comprehensive Review on CubeSat Electrical Power System Architectures” [26], this document focuses on mission-specific solutions, practical design principles, and real-world lessons from recent CubeSat missions, such as MarCO. Furthermore, it suggests innovative hybrid energy storage systems and investigates sophisticated technologies such as GaN-based converters, multi-junction solar cells, and redundancy methods to enhance fault tolerance, which were not covered in prior works.

The rest of this paper is structured as follows:

Section 2 delineates the classification of CubeSat EPS architecture, encompassing DC Bus Regulation, PV Panel Interface, Converter Locations, and Conversion Stages. Section 3 presents the mathematical framework for EPS design. Section 4 offers an extensive analysis of CubeSat EPS architectures, classified into many categories. Section 5 delineates the EPS architectures of CubeSats, offering essential comparative and design insights. Section 6 analyzes the EPS topologies, focusing on component quantity, efficiency, and trade-offs. Section 7 examines current advancements in CubeSat technologies, encompassing GaN transistors, multi-junction solar cells, and interplanetary missions such as MarCO. Section 8 delineates the prospective research avenues, emphasizing enhancements in reliability, sophisticated topologies, and hybrid energy storage systems. Section 9 ultimately finishes the paper by encapsulating the principal results and contributions.

2 CubeSats’s EPS architecture classification

It is important to note that the terms ‘solar panels’ and ‘photovoltaic (PV) panels’ are used interchangeably throughout this paper. Both refer to devices that convert sunlight into electrical energy using photovoltaic cells through the photovoltaic effect. While ‘solar panels’ is a broader term that may include thermal panels, in the context of CubeSat EPS, we specifically refer to photovoltaic panels for power generation [72, 73]. For that, Table 2 has been added to highlight the main differences between solar panel and PV.

Table 2 Comparison between solar panels and photovoltaic (PV) panels

The selection of EPS architecture is a critical step in the CubeSat’s EPS design process, and it is critical for the success of a CubeSat mission ADDIN ZOTEROITEM CSLCITATION. Based on DC-bus voltage regulation, PV panel interface, power converter location, and number of conversion stages, existing CubeSat’s EPS topologies can be categorized (see Fig. 2 for a visual representation).

Fig. 2
figure 2

CubeSat EPS architecture classification

2.1 DC bus regulation of Voltage

With the help of a DC bus, you may connect your solar panels to batteries (the storage system) and to various subsystems of the CubeSats acting as a load. There are three categories of EPS based on the regulation of the voltage of DC buses: regulated DC-Bus, unregulated DC-Bus, and sun-regulated DC-Bus. EPS having unregulated DC-Bus is the most popular kind of COTS EPS and contains battery connections linked to the DC Bus [39]. To keep the voltage of the bus very close to the typical values that are possible in the DC Bus (regulated), a dedicated converter (DC–DC) is commonly utilized. Sun regulated DC bus design regulates the voltage of the DC bus only during times of sunlight and eclipse, and the battery is linked to this bus through a diode. Semi-regulated and quasi- regulated DC buses are other names for this kind of topology [7].

2.2 Interface of PV panels

Based on the interfacing of the PV panels, CubeSat’s EPS may be divided into Direct Energy Transfer (DET) and Peak Power Tracking (PPT) architectures [74].

1) DET Architecture: The storage system and loads are directly linked to the PV panels using series diodes in this arrangement [7, 75]. Shunt regulators are often installed in parallel with PV arrays to divert extra PV power when the system is fully charged or connected to low values of load, as shown in Fig. 3. Extra power is normally wasted as heat within the CubeSat and discharged onto the PV panel with the help of a resistor in the shunt regulator. The mission might fail if the shunt regulator fails. It is the simplest and most cost-effective option for radiation environments, with a lower number of components and hence greater reliability. Due to the low power generation and storage capabilities of CubeSats, this architecture has a major flaw: it underutilizes the PV panel generation capacity.

Fig. 3
figure 3

Direct energy transfer (DET)

2) PPT Architecture: In PPT architecture, a DC–DC converter-based design uses PV panels to perform MPPT throughout a broad range of operating circumstances, including solar irradiance and the temperature of the PV panels. A series regulator is normally placed between the load and solar array in PPT architecture, as shown in Fig. 4. Because of the restricted power-generating capabilities of CubeSats due to their smaller size and lack of room for bigger photovoltaic panels, this technology is commonly utilized in their development. Digital microcontrollers (MCUs) or analogue controllers may both be used to implement MPPT [39, 76, 77]. The simplicity and tuning flexibility are two main advantages of using MCUs, but in parallel, they are also vulnerable to radiation damage [37]. A discrete analogue controller, though less efficient than an MCU, is considered more reliable [38, 41]. Analogue controllers may be used to operate the CubeSats either as the primary control or as a backup in case of MCU failure. Due to the wide range of PV array sizes, EPS may use both DET and PPT architectures [41]. PPT architecture is used on a panel that is not the same size as other panels to operate at the same voltage as larger PV panels. Some of the PV panels of CubeSat utilize DET to power Li-ion batteries, whereas the remaining PV panels of the same CubeSats use PPT to power their microcontrollers [30].

Fig. 4
figure 4

Peak power tracking (PPT)

2.3 Location of power converters

Architectures of EPS may be classified as centralized/concentrated or decentralized/distributed depending on where their power converters are located (Figs. 3 and 4).

1) Centralized architecture: In this topology, a single EPS PCB board houses all the controllers and power converters, which are connected to the payloads, PV arrays, batteries, and subsystems through specific voltage buses. Centralized architecture distributes all or most of the voltage rails used by the CubeSat from one central location. Since one regulator may supply the same regulated voltage to several sub-systems or components, the main benefit of the centralized architecture is that fewer regulators are needed. Figure 5 shows the block diagram of the centralized architecture.

Fig. 5
figure 5

Centralized EPS architecture

CubeSats mostly utilize centralized architecture due to their simple design, physical space efficiency, and accessibility of many COTS EPS designs. Using a single voltage rail for several payloads and subsystems reduces the number of voltage regulators in this architecture. As a result, the converter operates at a reduced efficiency for most of the time since the voltage regulators were built for maximum load demands while designing the centralized architecture. This type of architecture has one more major flaw: if a single regulator fails, then the payloads connected with this regulator are also affected [39, 40, 42,43,44, 49, 69, 78,79,80,81].

2) Distributed architecture: A single bus voltage is often spread to the various subsystems using a distributed EPS, as seen in Fig. 6. A separate bus exists for each subsystem. In a distributed EPS architecture, it is possible to turn on and off a single component without having an impact on the other subsystems or components. In distributed architecture, power converters or regulators are installed close to various subsystems of the CubeSats, including MPPT power converters near PV panels [35, 36, 56, 69, 71]. Because of the increased number of dedicated power converters and regulators, this type of architecture is more typical on larger satellites. CubeSats are also adopting distributed designs be- cause of the current advancement of tiny monolithic DC-DC converters with high efficiency [55, 82]. Because of the creation of smaller footprints for charge pumps and switching capacitor converters, distributed designs in CubeSats may be more readily adaptable. Distributed designs provide feature modularity, redundancy, and the advantage of reusability for several missions. Distributed architectures also have reduced radiation noise. Distributed architectures have point-of-load (POL) converters and regulators installed near the load, minimizing radiation noise by reducing DC-bus coupling and current loops [71]. This type of architecture also has a good distribution of heat as power converters and regulators are distributed.

Fig. 6
figure 6

Distributed EPS architecture

2.4 Conversion stages

We can also categorize the EPS of CubeSats based on the number of conversation stages into two types: single-stage architecture and multi-stage architecture [83].

1) Single-stage architecture: Fig. 7 depicts a CubeSat EPS design with a single-stage architecture. In this architecture, MPPT, battery charge/discharge management, and voltage regulation are performed utilizing multiple in- put/multiple output (MIMO) converters as a single conversion stage. It features a smaller size, fewer components, and a more efficient conversion. This type of architecture has more complexity of control and is more difficult to regulate; hence, it performs worse across a broad range of operating situations. This type of architecture is generally implemented in small satellites and exploration rovers [83,84,85].

Fig. 7
figure 7

Single stage EPS topology

2) Multistage architecture: Multistage architecture is mostly used in CubeSats, with single input, single output (SISO) converters having multiple conversion stages involving dedicated functionality. In multistage architecture, load-side converters and regulators manage output voltages, whereas PV-side converters and regulators check MPPT. Each converter or regulator is designed according to its unique functionality, resulting in a simpler control scheme and better system performance over a wider variety of operating conditions. The EPS architectures shown in Fig. 5 (centralized architecture) and Fig. 6 (distributed architecture) are both multistage architectures. But it comes at a cost in terms of the number of components in the system and the overall system’s efficiency. Various CubeSat COTS EPS designs use multistage architectures [40, 44].

3 Mathematical framework for EPS design

This section delineates the mathematical models employed to characterize and evaluate the fundamental elements of CubeSat Electrical Power Systems (EPS). The quantitative framework encompasses models for solar power generation, battery energy storage, DC–DC converter efficiency, and energy balance. These models establish a mathematical basis for the comparative analysis discussed in the next sections.

3.1 Solar power generation model

The equation for quantifying the electricity provided by PV panels has been incorporated below [86].

$$ P_{PV} = \mu_{PV} \cdot A_{PV} \cdot G \cdot {\text{cos}}\left( \theta \right) $$
(1)

where

  • PP V: is the power generated by the PV panel (W).

  • AP V: is Surface area of the PV panel (m2)

  • G: is Solar irradiance (W/m2) at the CubeSat’s orbit.

  • θ: is the angle of incidence of sunlight on the PV panel. This equation delineates the relationship between power generation and the efficiency, orientation, and area of the photovoltaic panels, in addition to the available solar irradiation.

We have elucidated how these characteristics affect CubeSat power generation across various orbital situations.

3.2 Battery state of charge (SoC) calculation

To track the energy available in the CubeSat battery, we have included the following equation for the state of charge (SoC) [87]:

$$ {\text{SoC}}\left( t \right) = {\text{SoC}}(t_{0} ) + \frac{1}{{C_{{{\text{bat}}}} }}\int\limits_{{t_{0} }}^{t} {(P_{{{\text{in}}}} - P_{{{\text{out}}}} )} dt $$
(2)

where:

  • SoC(t): State of charge at time t (as a percentage).

  • SoC(t0): Initial state of charge at time t0.

  • Cbat: Capacity of the battery (Ah).

  • Pin: Power input to the battery (W).

  • Pout: Power output from the battery (W).

This model offers a dynamic framework for monitoring battery state of charge, essential for maintaining power continuity during eclipse phases and times of elevated subsystem demand. We offer illustrations and background regarding the management of battery charging and discharging in CubeSats.

3.3 DC–DC converter efficiency

To compute the power conversion efficiency of DC–DC converters, the following equation has been introduced [88]:

where

$$ \mu_{conv} = \frac{{P_{out} }}{{P_{in} }} \times {1}00 $$
(3)
  • µconv: is the efficiency of the converter (as a percentage).

  • Pout: is the power output from the converter (W).

  • Pout: is the power input from the converter (W).

This equation demonstrates the efficacy of DC–DC converters, which is crucial for the overall power supply of CubeSat subsystems. This discussion addresses the impact of energy loss in converters on system performance and elucidates the necessity of developing technologies, such as GaN-based converters, for enhancing efficiency.

3.4 Maximum power point tracking (MPPT) efficiency

The following formula has been added to characterize the maximum power point tracking (MPPT) system’s efficiency [89]:

$$ PMPPT = PPV \cdot \mu MPPt $$
(4)

where

  • PMP P T is the power at the maximum power point (W).

  • PP V is the total power generated by the PV panels without MPPT (W).

  • µMP P t is the efficiency of the MPPT controller (between 0 and 1).

This model illustrates how the MPPT controller optimizes power generation from photovoltaic panels under fluctuating variables, including orbital position, sun incidence, and temperature variations. We examine the importance of the MPPT controller in enhancing power output from the photovoltaic system.

3.5 Solar panel energy calculation over an orbit

To calculate the total energy produced during one orbital period, the subsequent equation has been proposed [90]:

$$ Eorbit = PPV \cdot tsunlit $$
(5)

where

  • Eorbit is the total energy generated during the sunlit phase (Wh).

  • PP V is the power generated by the PV panel (W).

  • tsunlit is the duration of the sunlit period (hours).

The total energy produced by solar panels throughout each orbit is modeled by this equation. It establishes a fundamental framework for assessing available energy in CubeSat EPS, guiding decisions on power distribution, battery charging, and load balancing.

3.6 Battery charge/decharge process

It can calculate the energy stored in or supplied by the battery over a certain time period as follows [91]:

$$ E_{battery} = V \cdot I \cdot t $$
(6)

where

  • Ebattery: is the energy stored in or discharged from the battery (in watt-hours or joules).

  • V: is the Voltage of the battery (in volts).

  • I: is the Current flowing in or out of the battery (in amperes).

  • t: is the Time during which the battery is charging or discharging (in hours if energy is in watt-hours or seconds if energy is in joules).

This equation computes the total energy transfer in the battery, where energy is contingent upon the voltage (V), the current (I), and the duration (t) of operation. It is essential for assessing the energy input while charging or the output during discharge.

4 Review of Cubesat’s EPS architecture

Based on the above-mentioned classification in Section II, here is a comprehensive review of some commercial and university-affiliated CubeSat’s EPS architectures.

4.1 With unregulated DC bus and DET

For the OUFT1-1, its EPS design with a DC bus voltage in the range of 2.7–4.2 V is described in detail in [31]. Regulated voltages for loads are available, having values of 3 V, 5 V, and 7.2 V, respectively, and two Li-ion batteries are used for backup. For the CP1, an EPS design has been developed using a 3.6 V nominal DC bus voltage [32]. Two series- connected cells with a typical voltage of 4.2 V make up the solar panels. It has a lithium-ion main battery and a secondary battery made up of three Li-ion cells connected in series. Two buses of regulated voltages having values of 5 V and 3.6 V are provided by load-side converters. The EPS design for the Pilsen II CubeSat [30] includes three independent supply channels. Two of the sources of supply to increase the reliability of the EPS—a supercapacitor-based EPS design—were evaluated in [33]. For Ky-sat-3, there exists a DC bus voltage of up to 2.7 V, whereas there are two unregulated voltages (5 V and 3.3 V) that are supplied to the load. Battery, payload module, and solar panel module form a distributed EPS architecture that is developed using rectifiers that are synchronously functioning as DET interfaces [35]. Load-side converters and shunt regulators are utilized to link PV panels’ output directly to the DC bus in [38, 92], whereas a battery charge regulator (BCR) is utilized to connect batteries to the DC bus. Batteries’ voltage and charging current are controlled by the BCR, which has two control loops. When the PV panels provide the load demand, one DC–DC converter will lose power, but two converters will lose power when the battery supplies the load requirement. The EPS design is implemented in [93, 94] with two 1000 mAH Li-ion batteries linked in series and powered by a special BQ2405 charger. The EPS design is implemented by utilizing two BCRs linked in series with two lithium-ion cells for the Quake Sat 3U CubeSat [37]. Four solar arrays out of the total twelve solar arrays are body-mounted, and the remaining eight are deployable. The payloads are powered by controlled voltage buses of − 5 V and + 5 V, while part of the equipment is coupled to an unregulated DC-bus on the load side.

1) With unregulated DC bus and PPT: All the battery connections presented in [33, 47, 55, 56, 78] are directly linked to the output of an MPPT converter, forming an EPS architecture with an unregulated DC bus and PPT. To keep the battery voltage at its maximum, the MPPT converter switches to voltage control mode if its top limit is reached. Here, the PV panels provide the same amount of current as the load demands, and there are also conversion losses. Two converters handle the load requirement at the time when PV arrays are providing current, and electric supply from the battery to the load is handled by one DC to DC converter. Nanoavionics [33], Clydespace [40], Crystalspace [95], and GOMspace [96] all employ this form of EPS architecture. CubeSats containing power of PV equal to or equal to 30W have already been realised by the GOMspace, while modular EPS design has been designed for bigger than 3U CubeSats having 300 Watts of PV power [96]. A technology is built by Crystalspace, having capacities of approximately uninterrupted 10 watts as solar input and 12 V, 3.3 V, 5 V, and unregulated 3.7 − V supply for CubeSats. Input from three parallel-connected PV panels feeds two MPPT converters. Based on this EPS architecture, AAC Clydespace offers CubeSats with integrated batteries, body-mounted panels in 3U CubeSats, and deployable panels in 3 to 12U CubeSats [33, 40]. A single-ended primary inductor converter (SEPIC) is used in [41] to design and implement EPS (with unregulated DC-Bus and PPT) for CubeSats. Because of the benefits of noninverted input and output voltages, short-circuit safety, a low-side driver, and continuity of current flow, SEPIC converters are the preferred choice. There are six battery packs separated into two groups in PiCPoT CubeSat’s MPPT converter, which is manufactured using a hysteretic switching converter and an MPPT controller. In the AoxiangSat 12U CubeSat, solar panels have 57.7 watts of output power and 6 cells connected in a 2S3P configuration rated at 57 − watt hours [43, 44]. To improve power management and the efficiency of energy distribution across components, the ERPSat-1 CubeSat uses a fuzzy logic technique to implement its EPS [46]. Libertad 2 is a 3U CubeSat that uses this design after assessing various operating situations in terms of power converter efficiency [97]. EPS is achieved in the Ncube with the parallel configuration of PC arrays and by utilizing POL converters on the load side for supplying various loads with just one MPPT converter with redundant components to prevent the mission from being jeopardized by component failures [56]. The ESTCube-1, which performed the very first experiment of an electric solar wind sail in orbit, used this configuration of EPS with redundant components to guarantee that the mission was not jeopardized by component failures [48]. There were no serious issues with component failures, and the parallel-functioning converters boosted system efficiency. This CubeSat design, on the other hand, encountered difficulties because of the high level of complexities related to the software required for monitoring and handling the redundant systems caused by the hot redundant arrangement of multiple components.

Some CubeSats make use of hybrid EPS systems, which integrate DET with peak power tracking. The University of Toronto has created a nanosatellite bus [98]. At full charge, the voltage of the DC bus is controlled by the battery voltage, which is at its maximum during the day and at its MPP voltage at night. EPS systems that use PV panels only process the load once, while EPS architectures that use batteries only process the load twice. By utilizing the MPPT converter, MAX5717 IC, and two prismatic polyester lithium-ion batteries with nominal voltage values, the MEROPE CubeSat has been developed with a nominal voltage of 3.7 V [99]. Tel-USat’s EPS uses lithium-ion batteries with a nominal voltage value equal to 7.4 V and regulated voltage outputs of 5 V and 3.3 V to execute this design.

2) With regulated DC bus and PPT: An extra DC–DC converter is used in this sort of design for adjusting the DC voltage of the bus close to the reference value. In this type of EPS architecture, there is a direct linkage of the battery terminals with the outputs of the MPPT converters. The load is connected serially to the voltage regulator of the DC bus. A three-converter system is needed when load is provided by the PV panels, whereas only two converters are needed when using a battery for demand. One channel of power supply in Pilsen Cube II CubeSat has an energy storage system implemented as UC and DC bus voltages within the range of 3.34 V–4.9 V are also controlled, which is linked to battery terminals [30]. With DC-bus voltages as high as possible and currents as low as possible, fewer conduction losses may be achieved by selecting a battery voltage that is within the limitations of the duty cycle of the converter. Keeping the DC bus voltage equal to some reference value is an additional function of the MPPT converter. The BCR controls the battery discharging and charging currents. As an MPPT converter, it uses the SPV1040 IC, the LTC4066 IC, and three lithium-ion batteries connected in parallel with a nominal voltage of 3.7 V. In addition, regulated voltage lines of 5 V, 4.2 V, and 3.3 V are provided to the load [65]. As an additional option, there may be a DC–DC converter that is connected in parallel that keeps the DC voltage on the bus constant by serving as a voltage regulator [66,67,68]. Batteries in this system can be charged or discharged throughout the day, depending on the amount of demand placed on them during eclipses. As PV panels are only providing power to meet load demand and to compensate for converter losses, MPPT converters restrict DC bus voltage to their maximum value of battery voltage. There must be two power converters in this form of EPS architecture, whether the batteries or the PV panels are supplying the load requirement. 5 V, 3.3, and an unregulated voltage bus are provided to the RVSAT-1 CubeSat in its load equipment, which utilizes two parallel and two series lithium-ion batteries. Energy production, storage, and sizing recommendations for EPS are examined in [68]. PV panels may be linked in series or parallel, with each panel’s output sent into a voltage regulator. The load-side converters and the MPPT battery charger are both linked to the regulator’s output. Two converters handle all the electricity, whether it comes from solar panels or batteries. The EPS of Tel-USat uses a regulated DC of 12-V to do this. Loads were powered by a 7.4-V Li-ion battery and provided by voltage regulators that included an LM2577-12 integrated circuit (IC) and an MPPT integrated circuit (IC) [100]. If a voltage regulator and MPPT converter are provided with the outputs of all the se-ries-connected PV panels for DC buses, then another EPS arrangement may be created with PPT and controlled DC buses [70]. Another DC–DC converter controls the DC-bus voltage, while the output voltage of PV arrays is regulated by the MPPT converters to MPP voltage. Two power converters handle the power whenever the PV arrays provide the load requirements, while a total of three power converters control the power whenever the batteries provide the load requirements [56]. Disconnecting loads from PV and/or batteries due to an open circuit failure on a semiconductor device is a common problem in this sort of design. 3.4. With Sun-regulated DC bus and PPT To keep the voltage of the DC bus around a certain reference value, EPS on certain CubeSats with Sun-Regulated DC buses uses a voltage regulator to charge or discharge the battery depending on whether the load demand is greater or less than the production of power from solar panels. In the case of a fully charged battery, PV panels fulfil conversion losses and load demands as MPPT converters keep the voltage level at optimal battery voltage. A forward-biassed diode and a disabled voltage regulator are the only power sources available during the eclipse; thus, the battery is used to power the loads. Two serially linked lithium-ion cells with distributed POL converters are used in [69] to increase the efficiency of EPS. A sun-regulated DC bus design based on additive stacking is possible [71]. The MPPT converters and voltage regulators are connected in series with the PV panel in this EPS configuration. The DC bus voltage is created by connecting voltage regulator outputs and the MPPT converters in series. If the DC bus voltage is greater than the voltage of PV panels, the functionality of the voltage regulator is achieved by the voltage boost converter. If you have a variety of PV panels and batteries, a buck-boost converter should be used as an MPPT converter. Batteries may be charged more efficiently using the MPPT algorithm. Higher DC-Bus voltage reduces conduction losses, which is the primary benefit. The voltage regulator is deactivated during the eclipse so that we have a straight supply from the battery to the loads, making the diode forward biassed [101, 102].

5 Summary of EPS architecture of Cubesats

Table 3 presents a summary of the EPS topologies utilized in diverse CubeSat missions. It methodically classifies the essential design criteria and decisions implemented for the EPS in each mission, emphasizing notable themes and discrepancies.

Table 3 Summary of EPS architecture

6 Comparison of EPS architecture

COTS components are often used in CubeSat missions; however, some developers choose to create their own EPS systems. Because of the weight and volume limitations involved in CubeSats, most architectures are centralized, and multiple conversion stages are used in all present EPS topologies due to their extensive flight history and simplicity. There are two EPS designs that leverage unregulated DC bus topologies to provide improvements in efficiency and a reduction in component count. It’s possible to show that most EPS systems with aviation history use MPPT for their PV panels due to their restricted power-generating capabilities.

Table 3 compares the efficiency and component count of the four EPS topologies that were examined. It relies on power generation and load profiles to determine the number of stages of power conversion in EPS design. With no other means of supplying loads with power during an eclipse, battery-to-load power conversion losses are a key determinant of overall system performance. The total number of steps of power conversion during sunlit time is dependent on whether the load demand exceeds the production of power. Power conversion losses are defined by the factor of how many power converters there are spanning load and PV panels, as well as panels to batteries, if somehow the demand for the load is very low as compared to the power produced. If demand for power is greater than what is available, the total conversion losses are determined by the quantity of power converters connecting PV arrays, batteries, and loads. Because of CubeSats’ severe volume and weight restrictions, the EPS architecture’s total number of components is critical. The smallest number of components is found in DET systems, but the PV panels’ ability to collect energy is sacrificed. There are several variables to consider while choosing the appropriate EPS design for a CubeSat mission. These include factors like efficiency, quantity of components, and, of course, reliability.

7 Overview with recent updates in CubeSat

  1. A)

    Integration of gallium nitride (GaN) transistors for high- efficiency DC–DC converters [103]:

    • A discussion on the transformational role that GaN transistors play in CubeSat EPS has been included in this article. When it comes to CubeSat conditions that are power-sensitive and tiny, these transistors are appropriate because they offer higher efficiency, faster switching rates, and decreased power losses.

    • As an illustration, we referred to recent developments made by Texas Instruments, which highlighted the potential of GaN technology for use in space applications through the development of multi-MHz GaN power stage reference designs.

    • This new update sheds light on the utilization of GaN-based DC–DC converters in order to increase efficiency, decrease the size of the system, and meet the power requirements of CubeSats.

  1. B)

    Advancements in solar cell technology (multi-junction solar cells) [104]:

    • We have incorporated a discourse on the application of multi-junction solar cells in CubeSat systems, which offer markedly superior efficiency (exceeding 32%) relative to traditional silicon-based cells.

    • Our research is based on the findings from NASA’s State-of-the-Art of Small Spacecraft Technology Report, which highlights the increasing utilization of multi-junction cells to enhance power generation for small satellites.

    • This feature highlights how multi-junction cells allow CubeSats to produce greater power from restricted surface areas, rendering them vital for missions with stringent power limitations.

  1. C)

    CubeSat missions beyond low earth orbit (LEO) [104]:

    • We have included a summary of the Mars Cube One (MarCO) mission, which effectively showcased the viability of CubeSats in interplanetary endeavors.

    • The MarCO mission offered real-time communication assistance for NASA’s InSight Mars lander, signifying the inaugural interplanetary mission for CubeSats.

    • This case study demonstrates that CubeSats are no longer confined to Low Earth Orbit missions and can now facilitate deep-space exploration, so enabling future missions to the Moon, Mars and beyond.

8 Research scope in future

Improved CubeSat EPS reliability is a major issue for future studies. According to the CubeSat database, EPS failure has been a prominent cause of the failure of CubeSats. If we want to shield the CubeSat from cosmic radiation, substantial shielding materials cannot be used because of the CubeSat’s weight constraint. EPS reliability may be improved by providing redundancy for crucial components in case of a single-event failure. However, due to the limited amount of space on the EPS PCB, it is not possible to include redundant parts [105]. Passive components, particularly inductors with large footprints, may be made smaller, or the number of passive components can be reduced to overcome this problem. The use of semiconductor devices with a wide band gap, like GaN FETs [106, 107], to operate converters at high switching frequencies is one way to minimize the footprint of passive components. High-performance, low-loss, small-footprint devices that can withstand extreme radiation environments are available [108]. For CubeSat EPS, new multiport converter topologies may be developed that have various benefits, including smaller footprints, improved conversion efficiency, and a more affordable solution. In terms of CubeSats EPS multiport converters, there have been no published or patented papers or patents concentrating on this problem till now. A further area for investigation is the development of innovative system topologies that reduce the overall system complexity and footprints by reducing the amount of energy stored in passive components. Additionally, the development of advanced EPS architectures that reduce the requirement of energy storage for various passive components, for example, in inductors, is another research area that should be studied.

9 Conclusion

This article provides an in-depth look at the various CubeSat EPS designs. All other CubeSat subsystems are powered by the EPS, so it is a key subsystem of any CubeSat. It is essential to start by choosing an EPS architecture that is optimized for overall efficiency, control simplicity, component count, and battery configuration flexibility, as well as dependability and fault tolerance. According to the set of stages of power conversion, current EPS layouts are characterized based on either a DC bus having any voltage rating, which is the output of the EPS board providing regulated voltage, or unregulated voltages to various other subsystems of the CubeSats acting as a load. Similarly, where power converters are positioned and how PV panels are linked are also the parameters based on which this article has provided a classification of various EPS designs of the CubeSats. EPS architectures are organized into 4 major categories, each with a description of how they work and the benefits and drawbacks that come with them. Similarly, references to various recently launched CubeSats are provided in each relevant category of EPS topologies. Various CubeSat developers and academic institutes who are researching customized EPS designs are anticipated to find this article to be a helpful resource. In addition, a significant contribution of this work is the introduction of a Mathematical Framework for EPS Design, which offers quantitative instruments for assessing the processes of energy generation, power conversion, and storage. This methodology tackles practical mission difficulties and facilitates the optimization of CubeSat EPS systems.