Royal Academy of Sciences New Zealand Open Science
Open Science

Emergency power supply enabling solar PV integration with battery storage and wireless interface

Published:

ABSTRACT

This paper presents a detailed investigation of an emergency power supply that enables solar photovoltaic (PV) power integration with a battery energy storage system (BESS) and a wireless interface. Through the utilisation of solar PV-based generation and BESS with wireless/contactless power transmission, the proposed method offers an easy-to-setup and flexible alternative solution for the emergency power supply (EPS) for household appliances and wireless electric vehicle (EV) charging for all weather conditions. During bad weather conditions, the battery acts as the main power supply and can be charged from the solar PV panel and during rainy days, it can be charged from the grid by the proposed wireless interface for emergency use. The proposed system is analysed by mathematical modelling, focusing on the interface of the solar PV, BESS, and load, with inductive power transfer (IPT) as a wireless power transfer (WPT) method. The simulation and a scaled-down experimental prototype are built to demonstrate that the proposed system enables wireless power transfer with PV and BESS, and easy installation can be achieved by just placing the primary charging coil of the proposed power supply close to the wireless charging pad that is available in the existing system for e.g. EV charging. The results show that the load voltage is kept constant at 48 V against the varying input characteristic of the solar panel which is the solar irradiance ranging from 20% to 100% and output characteristics of the load of the proposed integrated system with variation of 20–50%, respectively. Moreover, the wireless power transfer efficiency across the inductive coils is maintained at about 97% under these variations.

1. Introduction

In the past decade, the global market for producing electricity from renewable energy sources (RESs) has been rapidly expanding (Anderson 2022). Solar photovoltaic (PV) generation, in particular, is the rapidly expanding sector for standalone household and electric vehicle (EV) charging applications. The efficiency of stand-alone PV generation can be further enhanced by implementing energy management systems to optimise energy use and reduce fossil waste. Because of the stochastic nature of solar irradiance (Sr) and dependency on weather conditions in solar PV generation, there can be irregular fluctuations in the PV output characteristics, leading to uncertainty in the power systems operations (Zhang X et al. 2021). Solar power generation and household energy consumption have completely opposite characteristics, with solar output peaking in the afternoon and household electricity demand peaking in the evening. Integrating battery energy storage systems (BESS) with solar PV (Tikkiwal et al. 2021; Williams et al. 2023) can compensate for such shortcomings. The battery can store the excess energy by charging from the solar array and, during the off-peak time, can discharge to the local loads (Manandhar et al. 2017; Cho and Valenzuela 2020, 2022; Ouédraogo et al. 2021).

New Zealand has an abundance of renewable energy resources, accounting for around 85% of its present electricity generation. In New Zealand, hydropower and geothermal power projects are often the most prominent, with hydropower consistently serving as the nation's main source of electricity (Zhang Z et al. 2023). A typical New Zealand can require up to 5kW rating system depending on a typical household operation, but if there are other amenities such as an electric vehicle (EV) or spa pool, it might require a high power rating system (Energy 2021). Solar Zero, a New Zealand-based company, have commercially produced battery for 5.4 kWh with a standard charge rate of 2.5 kWh (SolarZero 2023). Investing in solar PV by households may potentially promote the adoption of electric vehicles. Nearly one in four newly registered vehicles in New Zealand for each month in 2023 are plug-in hybrids or battery-electric vehicles as given (Evdb 2023) in . PV generation has also been explored in an innovative way by using floating PV in hydro reservoirs in New Zealand (Brent et al. 2024).

Figure 1. EV Market Stats: Monthly number of new battery electric cars and market share.

Figure 1. EV Market Stats: Monthly number of new battery electric cars and market share.

There have also been several pieces of literature on integrating solar PV and BESS in a stand-alone system used as the emergency power supply in weather disaster management during a power blackout or black start. In this paper Li et al. (2019), PV-BESS in a microgrid is combined to be used in a black start strategy. To improve the resilience of the power system, an optimum sizing of PV generation and battery is proposed in Zhang B et al. (2017).

In solar PV generation, converters would be connected between the PV array, battery and loads. Generally, the DC-DC boost converter is added for the PV panel to extract the maximum power from the panel. It is done so by the maximum power point tracking algorithm (MPPT), which uses the perturbation in the output characteristics of the PV panel to reach the maximum power point (MPP) for maximum efficiency (Paz and Ordonez 2016). There are several studies done on the MPPT algorithm (Esram and Chapman 2007). In these methods, hill-climbing, Perturb and observe (P&O), and incremental conductance (InC) are the MPPT methods mainly used in commercial PV systems. The hill-climbing and P&O methods are based on the duty ratio or voltage perturbation, respectively. Incremental conductance is based on the change in the slope of the PV voltage or current (Yan et al. 2018). The step size of the algorithm is a significant factor when designing the MPPT. As if the step is high, the tracking will be faster but inaccurate, and if it's too small, the algorithm will be very accurate, but the computational time will increase and hence, the tracking will be slow. Therefore, the methods need to be adjusted between computational speed and algorithm accuracy (Paz and Ordonez 2016).

Traditionally, a bidirectional DC-DC converter integrates BESS and PV with loads. In addition, several converter topologies have been analysed for integrating the RESs (Jha and Kumar 2019) and solar PV system in a DC microgrid based on voltage levels, power levels, switching and topologies (Uno and Sugiyama 2018; Singh et al. 2021; Vettuparambil et al. 2021; Zhang T et al. 2021). For Zhang T et al. (2021), a distributed MPPT based on a multiport DC converter is proposed for reducing the power losses in the PV distributed system. In Uno and Sugiyama (2018), a novel multi-port converter is proposed for stand-alone PV and battery systems combined with bidirectional pulse width modulation (PWM) and series resonant converter. In Vettuparambil et al. (2021), another novel converter topology has been proposed where multiple converter module has independent PV port system for extracting the maximum power. An automated transition for power management from grid-connected to islanded mode and maintaining constant power with enhanced stability to the utility is proposed in Singh et al. (2021).

Among these traditional wired converter designs, an innovative and new-edge solution can be a contactless/wireless power transfer interface. It is easily installed, a highly reliable way of transmission, and very safe to use (Zhang Z et al. 2022). Wireless power transfer is based on transferring power at a high frequency between galvanically isolated coils or plates with very high efficiency in an environment independent of dirt and other weather conditions. It uses mainly near-field electromagnetic induction and is divided into two methods: inductive power transfer (IPT) or capacitive power transfer (CPT) (Luo et al. 2020). For high voltage applications, IPT has been mainly used in EV charging (Vu et al. 2017). Very little literature has been on where IPT has been used in emergency power supply applications.

A typical IPT system is given in . It mainly consists of two isolated sides: primary coil and pick-up or secondary coil. The primary side will have a DC or an AC source with a primary converter and resonating circuit. Similarly, the pick-up side will have the secondary coil, pick-up resonating circuit, rectifier and DC load. Resonating circuits are mainly used to improve power transfer efficiency by reducing the reactive power and the inductance across the coils. The simplest way is by compensating the inductive coils with capacitors (C) and inductors (L) in series-series (SS) (Baros et al. 2018), series-parallel (SP) (Sohn et al. 2015), parallel-parallel (PP) (Su et al. 2018) topologies. One of the significant disadvantages of these topologies is that power transfer gets highly altered due to the misalignment (Mai et al. 2019). Therefore, there can also be more than one resonating or compensating component, such as LCL or LCC topologies (Li et al. 2015). Another factor for efficient wireless power transfer is the axial alignment and optimum distance between the inductive coils.

Figure 2. Typical IPT system.

Figure 2. Typical IPT system.

Therefore, considering all the factors above, this research showcases a novel configuration of a portable and compact PV and BESS system with an IPT interface with experimental validation. This proposed system can be used as an emergency power supply for household applications and EV charging. The proposed system is expected to maintain the load voltage to be constant against the output load and solar irradiance input variations of the proposed integrated system.

The remainder of the paper is organised as follows: Section 2 discusses this research's main objectives and aim. Section 3 presents the mathematical modelling of the proposed integrated system, mainly focusing on PV-MPPT and LCL-IPT systems. This model has been simulated in MATLAB/Simulink Software. A scaled-down experimental setup of a 24 V battery and 50 W PV module interfaced with IPT coils and 48 V load voltage has also been built to evaluate the performance of the integrated system. The description of this integrated system is given in Section 4. All the simulation and experimental results are presented and analysed in Sections 5 and 6. Section 7 presents the discussion from this study. Finally, the conclusions are given in Section 8.

2. Proposed system using WPT for emergency power supply

In this proposed study, the solar PV module-enabled BESS is the primary source for charging the EV battery and supplying the household load when there is a loss of power during an emergency. The proposed model and its applications are illustrated in and , respectively. illustrates the compact PV-BESS modular box in detail. (a) shows the normal operation of wireless power transfer, which supplies the power for the household loads and the EV battery. This setup also charges the proposed PV-BESS modular box as an emergency backup. The wireless charging coil is already available as the EV wireless charging pad and is connected to the AC mains as presented in (a). (b) shows the proposed system application as an emergency power supply. The PV-BESS coil acts as the primary side of this proposed modular PV-BESS box integrated model.

Figure 3. Proposed PV-BESS modular box.

Figure 3. Proposed PV-BESS modular box.

Figure 4. (a) Normal operation and (b) emergency power supply.

Figure 4. (a) Normal operation and (b) emergency power supply.

When there is a loss of power or a blackout from the main grid due to bad weather or a grid outage during an emergency, the household load will become the pick-up side or the receiving end through the wireless charging coil and power can be transferred contactlessly from the proposed solar PV-BESS modular power box as given in (b). Mobile PV module enables the flexible use of solar energy and BESS making the system resilient and offering electricity in critical situations. Similarly, the proposed PV-BESS modular box can also be used as the auxiliary power source for charging the EV battery in case of dire weather emergencies or when no charging ports are available, as presented in (b).

Therefore, for the scaled-down experimental prototype, a flexible and movable 50 W solar module with 24 V batteries modular box is incorporated in the primary side of the contactless power transfer. For this proposed model, on the pick-up side, a DC load is used for the experimental setup. It is a simplified representation of an EV battery acting as a voltage source and various load types in a household application for emergency power supply (EPS). shows the main difference between the traditional wired and IPT connections.

 

Table 1. Difference between IPT and Traditional wired methods.

The aim of this work is not to compare the efficiency with a standard wired system but rather to prove the feasibility of such wireless emergency power supply with solar PV. In (b), the EV battery is used with the pick-up coil and converter. The operating frequency is 85 kHz, the standard for EV charging (Bosshard and Kolar 2016), for this proposed setup. This setup can be very beneficial, especially in countries such as New Zealand, where the infrastructure for EV charging is yet to be developed beyond major cities. When the EV battery runs out of charge, the compact PV and BESS modular box can be used to charge the EV battery. This could be particularly useful for long-distance drives, where EV charging station availability might be a big issue. The proposed system could be placed wirelessly near the EV pick-up system to recharge.

The pick-up side can be available in a household that needs to be powered for the emergency power supply. The modular PV and BESS primary side box can be used to transfer the power to the loads depending upon the demand by contactless power transfer. This type of setup might be of great use for restoring power during a blackout or emergency situation. This proposed system would be quick and easily installable due to the lack of wiring.

The main objective of this proposed work is as follows :

  • Feasibility: The main focus of this paper is to show that it is achievable to integrate solar PV and BESS with load demand with contactless power transfer. The DC load on the secondary side is a simplified representation of various loads used in a household application for EPS or EV charging when no EV battery charger is available.

  • Easy installation: The integrated model has the setup in modular boxes. The primary side modular box contains the flexible PV module and battery backup with the primary resonating circuit and inductive coil. Similarly, the secondary side has the pick-up coil with the resonating circuit, rectifier and DC load. These modular boxes can be easily installed and maintained as a reliable way for EPS.

  • Load demand: This proposed model can also maintain constant load voltage when there is a load demand change and is discussed more comprehensively in the results sections.

3. System modelling and analysis

In this section, the calculation for parameter selection for the design has been analysed. shows the block diagram of the proposed model used in both the simulation and experimental setup. The DC input side comprises of a battery and bi-directional buck-boost converter for charging and discharging. It also consists of the PV module, MPPT algorithm, DC-DC converter, and DC link capacitor. The IPT system consists of a high-frequency (HF) inverter at 85 kHz and an LCL compensating network with the primary inductive coil. The pick-up side consists of the pick-up coil with a diode-rectifier and DC load to maintain the 48 V for the load voltage.

Figure 5. Block diagram of integrated solar PV and battery-IPT system.

Figure 5. Block diagram of integrated solar PV and battery-IPT system.

3.1. MPPT design for PV system

For this proposed integration of solar PV and battery with an IPT system, the P&O method is used for extracting the maximum power from the PV module as given in . For this proposed research, the P&O method is used as it is very simple and fast in computation.

Figure 6. P&O MPPT algorithm flowchart.

Figure 6. P&O MPPT algorithm flowchart.

The equivalent circuit of a single-diode of a PV cell can be described as equation (1), Ipv=IphID[exp(Vpv+IRsnaVT)1]Vpv+IRsRsh,(1) where Vpv and Ipv are output PV voltage and current, q is the charge of the electron, k is the Boltzmann's constant and ID is the diode current. Iph can be defined as the equivalent circuit of the PV cell, and it can be described as equation (1), Iph=Isc+ki(TTr)Ir100.(2) Isc is the short circuit current, T is the cell temperature, Ki is the coefficient, and Ir is the irradiance. Therefore, the PV output characteristics depend on the irradiance change.

P&O MPPT algorithm used in this proposed method is based on the PV voltage and power perturbation. The algorithm generates the duty cycle depending on the direction of the maximum power point (MPP) change. This duty cycle is then fed to the IGBT of the DC-DC boost converter for the solar PV module, and the maximum power is maintained.

3.2. LCL-IPT model

For IPT system modelling, the equivalent circuit is given in where Vprim is the voltage for the primary inductive coil and Vpick is the pick-up voltage. The primary and the pick-up currents are Ip and Is, respectively. The compensating components on the the primary side network are L1, Lp and Cp and the secondary side are L2, Ls and Cs respectively.

Figure 7. Equivalent circuit of LCL-IPT model.

Figure 7. Equivalent circuit of LCL-IPT model.

M is the mutual inductance between the coils, jωMIp and -jωMIs are the mutual voltage across the coils.

The resonant frequency is given by equation (3), ωr=2πfr=1(Lp.Cp)=1(Ls.Cs).(3) Then impedance can be defined as the induced voltage divided by the secondary current Is with RL as the load resistance and is given as equations (4) & (5), Zs=jωMIpIs=Ls+1Cs+1RL+L2,(4) Zp=jωMIsIp=ω2M2Zs.(5) The total impedance ZT of the IPT system can be calculated as equation (6), ZT=L1+1Cp+1Lp+Zp+Zs.(6) And the coupling coefficient kbetween the primary and the pick-up coil is defined by equation (7), M=k(Lp.Ls).(7) The voltage equation for both the coils under resonance can be written in equations (8) & (9), Vprim=RpIp+jωMIs(8) and, Vpick=RsIs+jωMIp.(9) Vprim and Vpick are the fundamental RMS values of the high-frequency VSI and rectifier AC voltages, respectively, and are across the coil. The equivalent resistances Rp and Rs are neglected here as they are very small and can be ignored. The primary coil and pick-up coil current under resonance is deduced by equations (10) & (11), Ip=VpickjωM,(10) Is=VprimjωM.(11) Therefore, the RMS values of the transmitter and receiver coil with respect to battery and load can be given by, Ip=22.VLπ.ωM(12) and, Is=22.VBatπ.ωM.(13) ILis the load current and is given by, IL=22.Isπ=8.VBatπ2.ωM.(14) The output power Pout is given as, Pout=VL.IL=8.VL.VBatπ2.ωM.(15) Therefore, from equations (8)–(15), the resonating components, respective voltages, current and output power can be calculated.

5. Simulation results

Based on the parameters given in the above section in and , the integrated PV and battery model simulations are conducted in MATLAB/Simulink software. All the simulation results and discussions are shown in this section. Both setups are based on the schematic diagram in .

5.2. Battery characteristics

Shows the simulation of a Li-ion battery's charging and discharging characteristics. (b) is the State of Charge (SOC) of the battery and is at around 85%.

Figure 11. (a) Solar irradiance (b) state of charge (SOC) of battery (c) PV and reference battery current (c) battery voltage.

Figure 11. (a) Solar irradiance (b) state of charge (SOC) of battery (c) PV and reference battery current (c) battery voltage.

From t=03 s, the SOC increases as the irradiance is more than 60%. From t=47 s, As the irradiance is low, the battery discharges to the load as the solar power is insufficient to supply the load. Once the irradiance is around 80%, the PV is sufficient again to charge the battery and supply the load from t=78 s.

Once the irradiance becomes less than half, the battery discharges, replicating the sundown time or bad weather conditions. The battery voltage is at around 26 V, as given in (c), and is the nominal discharge voltage of the 24 V battery. The small changes are due to the irradiance and load change, respectively as shown in (b–d). (d) shows the PV and the battery reference current. The load PI controller generates the battery reference current IBref. It is the error computed between the reference voltage and the load voltage.

From t=04 s, the battery current is negative, resulting in the charging profile of the battery. Here, the PV current is more than IBat, and the battery charges to t = 4 s. Furthermore, the discharge profile continues again from t=47 s.

6. Experimental results

6.1. Experimental setup

gives the experimental overview of the proposed model to demonstrate the feasibility of the proposed system. All the parameters are given in and .

Figure 14. Experimental setup of (a) battery with load setup (b) 50 W PV module (c) IPT with converters.

Figure 14. Experimental setup of (a) battery with load setup (b) 50 W PV module (c) IPT with converters.

Circular spiral IPT coils of 70 mm diameter are constructed in parallel, and both coils are completely aligned with each other and are connected with their respective LCL compensating circuits as given in (c). The secondary coil setup can move to measure the optimum distance for maximum power transfer. The output from the Solar PV system at (b) and battery-load setup at (a) is connected to the primary side for the IPT coils through the HF frequency inverter at 85 kHz. The pickup coil is connected to the diode rectifier and the electronic load on the secondary side.

6.2. Case 1: Optimum distance between the IPT coils

compares the various distance between the coils for the maximum efficiency of wireless power transfer for Solar-IPT and Battery-IPT setup, respectively.

Figure 15. Efficiency of contactless power transfer based on distance between inductive coils (a) 30 mm (b) 40 mm (c) 60 mm (d) 70 mm.

Figure 15. Efficiency of contactless power transfer based on distance between inductive coils (a) 30 mm (b) 40 mm (c) 60 mm (d) 70 mm.

This setup operates at an open loop with a DC input with the same input characteristics as PV-load at Vin is 24 V and battery-load system at Vin is 48 V at 85 kHz. This setup is built for EPS applications and can maintain constant load voltage with load demand change. Both sides of the coil will be stationary, but it is essential to find the optimum position so that the maximum power with constant voltage can be delivered to a DC load or EV. The distance differs by 10 mm from (a–d). At a distance of 30 mm between the coils, the k is lowest, and the efficiency is at 90%, but the coils have to be very close to each other, which might not be the best wireless power transfer setup as the coils are very close to each other. At 40 mm, the efficiency is the highest at 97% and has been chosen as the optimum distance for the closed-loop experimentation of two case studies. The plot also depicts that as the distance is increased beyond 50–70 mm, the k increases and the efficiency decreases, which is the reasonable outcome as the leakage inductance increases. Therefore, for 85 kHz resonating frequency (fr), 40 mm is the distance to get the maximum power in this proposed system.

6.3. Case 2: Hardware PV setup with IPT coils

The PV module is connected to the load with IPT coils for this section. shows the PV current when the load decreases in steps at t=11 s from 144-105 Watts and t = 24 s from 105-86 Watts. In this case, the solar PV module is at full irradiance at 1000 W/m2.

Figure 16. Hardware results of PV current with Load change.

Figure 16. Hardware results of PV current with Load change.

shows the primary and pick-up voltage of this study. The PV voltage is around 17 V, and the primary voltage is 25.8 V after the inverter and resonating circuit. The secondary pick-up voltage is 24.9 V, and the efficiency of the power transferred is around 97% which is around the same efficiency as given for the optimum distance between the IPT coils.

Figure 17. Hardware results of PV with IPT coils (a) primary voltage (b) pick-up voltage.

Figure 17. Hardware results of PV with IPT coils (a) primary voltage (b) pick-up voltage.

6.4. Case 3: Hardware battery setup with IPT coils

shows the primary and secondary voltage of the closed loop battery-IPT setup.

Figure 18. Hardware results of Battery with IPT coils and load (a) primary voltage (b) pick-up voltage.

Figure 18. Hardware results of Battery with IPT coils and load (a) primary voltage (b) pick-up voltage.

These voltage characteristics are the discharging profile of the battery. The figure shows that the primary voltage is around 64 V and the pick-up voltage is around 63 V at 85 kHz resonating frequency. The efficiency for contactless power transfer is maintained at around 97%.

Finally, this integrated PV-Battery is interfaced with the IPT coils in to investigate its performance when loads change. This experiment focuses on the transient response of load demand change with the WPT system. It is built in MATLAB software and then compiled in dSpace for real-time operation. The main function of this control is to regulate constant load voltage at 48 V when there is a load change. The controller generates the duty cycle for the DC-DC converter of the battery.

Figure 19. Load voltage against output Load variation.

Figure 19. Load voltage against output Load variation.

When the step load change occurs at 4.8 s, the load doubles from 50Ω to 100Ω, the transient settling time is around 50 ms, and the overshoot is 1 V. For the following step change at t = 15.2 s, there is a slight change in load from 100Ω to 80Ω, the transient response is again around 50ms, and the undershoot is −0.5 V. In the third step, the load decreases from 80Ω to 50Ω. The undershoot is −0.5 V. The load voltage is maintained at 48V±1V when the load is changed in large and small steps. Hence, from this figure, it can be proven that the proposed system can maintain constant load voltage with contactless power transfer at 97% efficiency across the inductive coils.

7. Discussion

Discussion on this research are provided as follows:

  • (Electricity Authority 2023) shows the planned outage from this year 2023 for all types of generation, such as hydro, geothermal and wind, among other sources. In these situations with planned or unplanned outages, this research of PV and battery with WPT system can be quite beneficial, as the household won't have to depend on the grid to supply the load. Instead, with the help of PV and battery, the fast and efficient wireless power transfer method can meet the load demand.

  • This study shows a proof-of-concept for a fully integrated system that uses solar PV as the renewable energy source and a battery as the energy storage, with power transferred via a wireless/contactless interface. This system is simple to install and provides a reliable power source for stand-alone residential applications in normal or emergency conditions. This study focuses on the system's practicality and feasibility, as well as the usage of RESs and battery backup, and an innovative method of power transmission by WPT.

  • Furthermore, the main objective of this study is to maintain load voltage by wireless power transfer from the battery and solar PV system in a new way by utilising WPT with solar PV and battery, no financial and economic analysis has been performed for this study. However, this can be investigated in future studies.

Figure 20. Planned outage for New Zealand (Electricity Authority 2023).

Figure 20. Planned outage for New Zealand (Electricity Authority Citation2023).

8. Conclusion

In this study, PV generation and battery storage are integrated for contactless emergency power delivery that can be put in a compact portable power box for an easy setup. The proposed system can serve as an emergency power box that can be used for wireless EV charging with a pickup coil already on board or for powering household appliances by using the primary charging pad of the EV as a power pickup coil. Complete theoretical analysis has been conducted to calculate the parameters for the system design, especially for the solar PV, battery and IPT coils. P&O MPPT method is applied to extract the maximum power from the solar module. For wireless power transfer, maximum power transfer is demonstrated by testing the optimum distance between the inductive coils. An LCL-IPT system with a 50 W PV module and 24 V battery storage is built, and the power transfer efficiency across the coils is found to be around 97%. A constant bus voltage at 48 V is achieved by designing a PI controller in a voltage and current loop. All the transient settling time and fluctuations in load voltage changes are within 2% against load change. The simulation and experimental results show that the contactless power transfer can be interfaced in the solar PV-BESS system to maintain a constant load voltage when the input solar irradiance of the PV panel changes from 20% to 100% in various steps and the load demand changes by 20–50%.

Disclosure statement

No potential conflict of interest was reported by the author(s).