Mobile-Menu

Discover the PCIM Europe

CHARACTERIZATION & SIMULATION Energy storage system with high durability and power density

From Nigel Charig |

Related Vendors

Würth Elektronik ICS GmbH & Co. KG
Finepower GmbH
Rogers Corporation
MESAGO Messe Frankfurt GmbH

As 14 % of the world’s 2010 greenhouse gas emissions came from cars, trucks and other vehicles, plans exist to reach 125 m electric vehicles by 2030. However, by 2018, the total was only 5.1 m. Accordingly, developing an economically-viable fast-charging infrastructure has become of interest, to accelerate the changeover to EVs.

Today, there is a worldwide impetus to reduce our carbon footprint by electrifying society. Therefore 125m electric vehicles are planned until 2030.
Today, there is a worldwide impetus to reduce our carbon footprint by electrifying society. Therefore 125m electric vehicles are planned until 2030.
(Source: gemeinfrei / Unsplash)

However, 33 % of 2018 US CO2 emissions were related to energy production, mainly from coal-fired power stations. To overcome this, massive deployment of renewable energy sources such as solar panels and wind turbines has become mandatory – but this requires a grid-tied energy storage system to balance energy production and consumption demands and to help grid stabilization.

Past energy storage systems composed of supercapacitors were not competitive with lithium ion batteries for energy and power density, making such systems economically unviable. Accordingly, current research efforts are looking at new types of carbon based supercapacitors to increase the performance metrics while simultaneously decreasing production costs.

This topic has been addressed in a paper presented at the PCIM Europe Digital Days conference in July 2020 by Daniel Evans, Nicolas Sockeel, Marco Verlohner, Jim Gafford, Somasundaram Essakiappan, Madhav Manjrekar, and Mike Mazzola. Key points from the paper, which was titled ‘Characterization, modeling, and simulation of a new high longevity and high power density energy storage system’ are given below.

The paper is concerned with the methodology used for extracting the main parameters (capacitance and equivalent series resistance) of Carbon-ion (C-ion) cells, as well as the development of a MATLAB Simulink model of C-ion packs and module with its balancing circuit. The parameters of the cell model are derived from a stochastic model of individual cells defined by the data set comprised of a large number of cells empirically characterized. An active voltage balancing circuit provides a simple, open loop, non-dissipative method to maximize accessible energy at the system level.

Carbon-ion (C-ion) is a new category of energy storage that uses nano-carbons and ionic electrolytes. This allows devices to operate at voltages exceeding 3V, which in turn offers the potential to increase the theoretical energy density up to 50 Wh/kg. C-ion cells have a very high power density capability (1.5 kW/kg) allowing safe charge and discharge at high rate. The lifetime expectancy and cycle capability of this technology is extremely high (> 100,000 full cycles) and with no fire risk. The cells can be safely discharged and stored at a zero state of charge with no performance degradation.

The paper focuses on a system level model of a C-ion based energy storage system. It describes how model simulation shows the need for voltage balancing of series connected cells, which is resolved by the introduction of an active voltage balancing circuit.

The paper begins by describing the empirical parameter extraction method used to characterize the C-ion cells, from which the model simulation is developed. It then introduces the Simulink models for varying C-ion cell arrangements, simulation setup, and performance results. Finally, it presents the performance of an active voltage balancing circuit and the simulation of an energy storage system composed of parallel and series connected groups of carbon-ion cells.

Carbon-Ion cell characterisation

Fig.1: Electrical equivalent model of a Super capacitor.
Fig.1: Electrical equivalent model of a Super capacitor.
(Source: Karlsruhe Institute of Technology)

The characterization method used carbon-ion cells produced by ZapGo, Inc.; these are modeled in Simulink and then used for further simulations examining the performance of a 25 VDC energy storage system as well as an active voltage balancing circuit. As shown in Fig. 1, the model chosen was a classical type for electrochemical double layer capacitor. All three elements of the classical model - the capacitance, the equivalent leakage resistance, and the equivalent series resistance (ESR) - are extracted and utilized for performance metrics.

Table 1: Cell ESR and capacitance characteristics.
Table 1: Cell ESR and capacitance characteristics.
(Source: Karlsruhe Institute of Technology)

To ensure accuracy of the measurements to industry standards, the energy conversion method procedure described in IEC international standard 62391 was adhered to during the capacitance extraction process. This test has been repeated on 90 different C-ion cells, and the distribution of the cell capacitance and ESR estimation is shown in Table I.

Simulink C-ion cell and system model

Time domain simulation of an energy storage system based on C-ion technology is performed with MATLAB using the Simulink software package. The C-ion cells are first arranged in a parallel configuration of 20 cells, which is referred to as a pack. Building on the C-ion cell distributions of capacitance and ESR, new distributions are calculated for the pack level.

Simulink modules of a pack and a module – which comprises eight packs – are then built, using capacitance and ESR values from these distributions. A state machine then performs multiple charge and discharge cycles in Simulink. The pack level configuration consisted of a constant current charge to 3.125 V at 120 A, and then held at constant voltage until the current falls below 5 A, at which point the source is turned off. The system is then left at the charged point for some time, to allow a comparison of the voltage deviation, and future system level simulations with a voltage balancing circuit to see the voltage convergence.

Subscribe to the newsletter now

Don't Miss out on Our Best Content

By clicking on „Subscribe to Newsletter“ I agree to the processing and use of my data according to the consent form (please expand for details) and accept the Terms of Use. For more information, please see our Privacy Policy.

Unfold for details of your consent

Following the hold, the pack is then discharged at 170 A to a voltage of 1.6 V. The state machine then loops back, and continues looping until the simulation finishes.

The system level simulation of the C-ion energy storage module utilizes the same state machine as the pack simulations. Therefore, it follows the same cycling pattern except that the voltage charge and discharge set points are changed appropriately.

Computational limits existed on the computer system used for simulations in the form of time and memory limitations, so simplifications were introduced to reduce both computational complexity and time duration for each simulation.

Charge Management and System Simulation Performance

Fig. 2: Active charge balancing circuit.
Fig. 2: Active charge balancing circuit.
(Source: Karlsruhe Institute of Technology)

Manufacturing variations cause the C-ion cells to have varying parameters in the model, and imbalances to occur in the pack level voltages when charged and discharged in the energy storage system. More specifically, the series connections of the packs result in the same current to flow through the cells of varying capacitance, ESR, and leakage resistance. When charging under constant current, the integration of the current causes the voltage across the capacitor to rise in proportional to its capacitance.

When the cells are left charged for extended periods of time, the varying leakage current internal to the cell results in a voltage imbalance between series connected cells and packs. Both imbalance mechanisms can result in cell overvoltage conditions and reduce the accessible energy of the system under standard use.

To ensure the voltage remains constant in the series connected C-ion system, an active charge balancing circuit is implemented as a series of interleave connected cascading half-bridge circuits as shown in Fig.2.

Fig. 3: Simulation results of a 160 cell, 25 V carbon-ion system undergoing a cycling test while managed to examine the voltage and ensure convergence with the active charge management circuit.
Fig. 3: Simulation results of a 160 cell, 25 V carbon-ion system undergoing a cycling test while managed to examine the voltage and ensure convergence with the active charge management circuit.
(Source: Karlsruhe Institute of Technology)

Empirical results validated the scaled system simulation, which shows that the active balancing circuit with open loop control causes pack voltages to converge as shown in Fig. 3.

Conclusion

The paper presents a realizable method for large format energy storage using C-ion supercapacitors. C-ion energy storage is capable of providing high power density, high longevity, and the potential to reach an energy density of 50 Wh/kg.

The paper’s main contribution is the methodology used for extracting the main parameters (capacitance and equivalent series resistance) of C-ions cells, as well as the development of a MATLAB Simulink model of C-ion packs and module with its balancing circuit. An active voltage balancing circuit provides a simple open loop non-dissipative method to maximize accessible energy at the system level.

Future work will move beyond simulation of the energy storage module and show the empirical results as measured.

(ID:46918437)

Related content