Mehdi Jafari | Massachusetts Institute of Technology (MIT) (original) (raw)

Papers by Mehdi Jafari

Renewable Energy, 2020

Decarbonization of the electricity sector is one of the major measures in slowing down the pace o... more Decarbonization of the electricity sector is one of the major measures in slowing down the pace of
climate change. In this paper, we analyze the impacts of energy storage systems (ESS) and year-to-year
variability and uncertainty in the hourly profiles of variable renewable energy (VRE) on power system
decarbonization in 2050. We perform this analysis through capacity expansion optimization based on
technology cost projections and CO2 emission restrictions based on 11 years of wind, solar, and load data
variations in Italy’s power system, with a particular focus on how ESS changes the optimal generation
portfolio and system performance. We also explore the impact of ESS duration on the renewables’
adoption and system costs. To quantify the impact of VRE variability in different years, we present a
comparative analysis of capacity expansion optimization based on multiple-year and single-year data.
Our results indicate a high RES penetration even in the absence of decarbonization policies, due to expected
declines in future technology costs. In the transition to a zero-carbon electricity system, carbon
capture and storage (CCS) with less than 100% carbon capture efficiency will play a role only in intermediate
low-emission scenarios unless future technology advancements improve its capture efficiency.
ESS investments contribute to lower total system costs by replacing more expensive flexibility resources.
However, the value of ESS changes by its duration, with longer ESS durations having lower marginal
value per added kWh storage capacity. Variability in VRE profiles leads to substantial variation in the
system’s configuration and energy cost, depending on what year is used in the optimization. Decision
making based on single-year data can therefore lead to substantial increases in the systems’ operational
costs in other years due to increased probability of capacity shortages and load curtailments. In contrast,
optimizing over multiple VRE and load years provides a more robust and cost-effective generation
expansion strategy.

Applied Energy, 2020

We investigate six different lithium-ion battery modeling approaches to highlight the importance ... more We investigate six different lithium-ion battery modeling approaches to highlight the importance of accurately
representing batteries in decision tools. Advanced mixed-integer-linear battery models account for efficiencies as
a function of the discharge power, power-limits as a function of the state-of-charge, along with degradation,
which are usually not accounted for in power systems models. The revenue potential from offshore wind paired
with battery systems is then examined using the more advanced representation where degradation is the sum of
the capacity fades resulting from calendar- and cycle-aging. The impacts of variability of offshore wind output
along with energy- and capacity-market prices are evaluated using publicly available data from 2010 to 2013
using NYISO as a test case. For 2013, results highlight that without accurate battery representations, models can
overestimate battery revenues by up to 35%, resulting primarily from degradation-tied costs. Advanced dispatch
algorithms that account for calendar- and cycle-aging of the battery can help operate the battery more efficiently.
Locating the battery onshore yields higher revenues and with wider useable SOC windows, it is possible
to monetize higher arbitrage opportunities, which can compensate for any additional degradation-tied costs. The
added value of a MWh of energy storage varies from 2to2 to 2to4.5 per MWh of wind energy, which leads to a
breakeven cost range of $50–115 per kWh for the battery systems. As such, energy- and capacity-market revenues
were found to be insufficient in recovering the investment costs of current battery systems for the applications
considered in this analysis.

IEEE Transactions on Transportation Electrification, 2019

This paper proposes a hierarchical Bayesian model for probabilistic estimation of the electric ve... more This paper proposes a hierarchical Bayesian model for probabilistic estimation of the electric vehicle (EV) battery capacity fade. Deterministic estimation of the EV battery degradation is not reliable, as its aging factors are mostly external and varies by time, location and the driver's usage pattern. Whilst, probabilistic models can address the observations uncertainties and provide more accurate estimations. We have developed a comprehensive Bayesian Network model which incorporates two main aspects of the EV battery aging: hierarchy and variety of external effectual factors. Multiple levels of hierarchical relationship with intermediate hidden variables connect the external factors such as driving behavior and habits, charging options and grid services to the battery capacity fade. The mathematical expression of the model is extracted based on Bayes' theorem, the probability distributions for all variables are carefully chosen and Metropolis-Hastings Markov Chain Monte Carlo sampling method is applied to generate the posterior distributions. The model is trained with a subset of experimental data (85%) and tested with other 15% of data to prove its accuracy. Also, three case studies for different drivers, different grid services' repetitions, and different climates are explored to show model's flexibility with different input data.

IEEE iTEC , 2018

In this study, we present a Bayesian Networks (BNs) approach for the electric vehicle (EV) batter... more In this study, we present a Bayesian Networks (BNs) approach for the electric vehicle (EV) battery degradation modeling. Battery aging is caused by factors that carry heavy uncertainty, such as battery usage depending on driver behavior, temperature profile depending on location and thermal management system, etc. with all these variations complicating the battery aging modeling with traditional frameworks. That is why we propose that the modeling should be carried out in a Bayesian Network framework that is capable of incorporating uncertainty and causality. The battery aging model is developed in the Bayesian framework and set of training and test data are used to validate the model. Results show that the BN model has a promising performance in the battery aging modeling. The model is also used to estimate the battery capacity loss in real driving cycles.

Journal of Energy Storage , 2018

Due to the increasing interest on the Li-ion battery aging studies among researchers, there are n... more Due to the increasing interest on the Li-ion battery aging studies among researchers, there are numerous battery degradation models presented in literature. However, they are either focused on a single technology, form factor or scale. This can be challenging for researchers that typically have to bridge multiple technologies and are interested in crossing multiple scales, from material-level, to cell and module/pack level. This paper explores the models presented for Li-ion battery degradation in different scales from the material level to the application level. In each scale, the main aging variables are summarized, the mathematical presentation of models are analyzed and the merits and disadvantages of each scale is discussed. This review aims at bringing together methods and results for multiple technologies, form factors for the most common Li-ion battery technologies.

-In this paper, transient stability improvement using superconducting magnetic energy storage bas... more -In this paper, transient stability improvement using superconducting magnetic energy storage based fault current limiter (SMES-based FCL) is presented. The proposed SMES-based FCL is installed at the beginning of faulty line. The proposed SMES-based FCL inserts a superconducting inductance into the fault current pass. The insertion of high value of inductance not only limits the fault current level in an acceptable level but also improves transient stability of power system by storing excessive energy of synchronous ...

In this paper, transient stability improvement of power system using fault current limiter (FCL) ... more In this paper, transient stability improvement of power system using fault current limiter (FCL) based on superconducting magnetic energy storage (SMES) is proposed. SMES technology is capable to make dc reactors with large values because of its superconducting property. Using such dc reactors in FCL structure can effectively limit the fault current during short circuits in power lines. To show the additional merits of such FCL structure, it is applied to transient stability of single machine infinite bus (SMIB) system ...

An energy storage system simulation requires a battery model capable of precisely predicting the ... more An energy storage system simulation requires a battery model capable of precisely predicting the dynamic behavior and I-V characteristics. An equivalent circuit model (ECM) of a battery generates an electric circuit able to replicate complicated battery characteristics. Combination of time domain and frequency domain tests can be utilized to parameterize ECM components properly. The goal of this paper is to discuss two systematic approaches to decide the equivalent circuit model parameters, which can be replicated on any battery technology. Time and frequency domain testing are carried out on a commercial electric bicycle Li-ion battery composed of Samsung ICR18650-22P. Then, Impedance Analyzer approach and Bayesian Network (BN) method are applied to estimate the ECM circuit elements with different circuit topologies. The models were developed in MATLAB/ Simulink environment. A highly dynamic drive cycle based on the E-Bike was utilized to validate the ECMs through hardware-in-loop testing.

IEEE Transactions on Transportation Electrification

n this paper, battery lifetime estimation of an electric vehicle (EV) using different driving sty... more n this paper, battery lifetime estimation of an electric vehicle (EV) using different driving styles on arterial roads integrating recharging scenarios in the neighborhood of the vehicle-to-grid (V2G) integration is studied. The real-world driving cycles from a fleet of connected vehicles are evaluated in an EV model with different charging options. Daily utility services are added to the simulations to explore the whole day performance of the battery and its daily degradation. Fifty driving cycles from different drivers on arterial roads are classified into aggressive, mild and gentle drivers based on their driving acceleration behavior. The standard level 1 and 2 chargers are considered for recharging and the frequency regulation, peak shaving and solar energy storage are assumed for the daily ancillary services. The results indicate that the aggressive driving and recharging behavior have significant effect on the battery life reduction. In addition, the daily utility services impose extra degradation of the battery. Also, the effect of temperature change on the battery degradation is explored. Simulation of active vs. passive thermal management systems in three different climates shows the significant impact of the battery temperature on its capacity fade.

Electric vehicles are increasingly being adopted due to environmental awareness and competitive t... more Electric vehicles are increasingly being adopted due to environmental awareness and competitive technical performance and reducing prices. Their research and development has sometimes relied on the use of standard driving cycles. However, these cycles cannot reproduce the variations of traffic flow in real world. That is why in this paper we develop a software tool able to analyze real-life driving cycles for electric vehicles. To do so, a driving trajectory process tool is used to obtain large data for vehicles driving in the same stretch of highway. To show the performanc of the developed tool, sample cycles are analyzed and simulated for electric vehicles automatically in the REV-Cycle (Real Electric Vehicle Cycle analyzer) software presented.

—This paper presents the large data analysis of the real-world driving cycles and studies the eff... more —This paper presents the large data analysis of the real-world driving cycles and studies the effect of different driving styles on the electric vehicle (EV) battery performance and aging. For this study, a MATLAB-based software tool, real EV cycle (REV-cycle) analyzer, is developed. The real-world driving data are recorded from the I-80 highway, CA, USA. In this study, driving cycles are classified into three styles as aggressive, mild, and gentle driving based on their average acceleration. Also, two standard driving cycles (EUDC and HWFET) are simulated by the software and the results are compared. The results show that the real driving cycles are very different from the standard cycles. Also, it is observed that the traffic flow affects the driving style, as the drivers tend to drive more aggressive during the light traffic hours of the highway, while the heavy traffic limits the driver's aggressive behavior. On the other hand, the driving style has considerable effect on the energy consumption and the battery aging. The aggressive driving style demands higher average power from the battery compared to the mild and gentle driving style. From the aging point of view, the aggressive driving style leads to higher C rate demand from the battery and it expedited the capacity fade process compared to the mild and gentle driving styles.

Power and Energy (PECon), …, Jan 1, 2011

This paper proposes a new topology for fault current limiter (FCL) that is based on series LC res... more This paper proposes a new topology for fault current limiter (FCL) that is based on series LC resonance circuit. The proposed FCL uses a high speed switch to put an inductor in series with a capacitor and make resonance condition between them. This structure can hold the peak of fault current in the constant value that is not possible in previously introduced series resonance type FCLs. In such conditions, if fault continues, power system will not be experience large short circuit currents. Using non-superconducting inductor instead of superconducting one, leads to low construction and maintenance costs in this structure. In addition, in normal operation, capacitor of this FCL can be used as a series compensator. Analytical analysis for the proposed series resonance type FCL is performed in detail. Simulations are presented using PSCAD/EMTDC software to show the operation of the proposed FCL. Also, comparison between the proposed FCL and previous series resonance type FCLs is included in simulations.

In this paper, transient stability of Single Machine Infinite Bus (SMIB) system with a Non-superc... more In this paper, transient stability of Single Machine Infinite Bus (SMIB) system with a Non-superconducting Fault Current Limiter (NSFCL) is proposed. Analytic analysis of system stability is discussed in detail and as result, optimal value of NSFCL resistance during fault that leads to effective improvement of stability is calculated. Equal area criterion is used for this calculation. Study system is simulated by EMTDC/PSCAD to verify the effectiveness of the optimal resistor value of the proposed NSFCL.

In this paper a variable and controllable resistor type fault current limiter (FCL) is introduced... more In this paper a variable and controllable resistor type fault current limiter (FCL) is introduced for enhancement of transient stability of single machine infinite bus (SMIB) system with a double circuit transmission line. The optimal value of resistor during fault to reach the maximum stability of power system is computed. It is shown that, this optimal value depends on the fault location and the pre-fault active power of synchronous generator considering power demand changing. To show the effectiveness of the proposed FCL, analytical analysis including transient stability study and optimum resistor value calculation are presented. In addition, simulation results using PSCAD /EMTDC software are included to confirm the analytical analysis accuracy.

ELECO 2011, Dec 1, 2011

In this paper, inrush current limitation of transformer using ‎a bridge type fault current limite... more In this paper, inrush current limitation of transformer using ‎a bridge type fault current limiter (FCL) as inrush current ‎limiter is proposed. The proposed ICL consists of three ‎single phase sets of diode bridge, small non-superconductor ‎and semiconductor switch parallel with a resistor. Because ‎of quick damping of inrush current by resistance, this ‎topology inserts a resistance in power system. By simple ‎control circuit and fast operation of the proposed ICL, the ‎maximum peak value of inrush current decreases in an ‎acceptable level. Using small value of non-superconducting ‎dc reactor reduces voltage drop on the FCL as inrush ‎current limiter (ICL) and construction cost. ‎PSCAD/EMTDC software is used for getting simulation ‎results. These results show good capability of the proposed ‎ICL to limit the inrush current of transformers.‎

ELECO 2011, Dec 1, 2011

This paper proposes using a fault current limiter (FCL) for ‎soft starting of induction motors (I... more This paper proposes using a fault current limiter (FCL) for ‎soft starting of induction motors (IMs). The base of this ‎approach is on the primary resistance starting method. The ‎proposed FCL which is installed at the beginning of an ‎industrial feeder helps to soft starting of induction motors ‎by controlling the current during the starting process. The ‎current of motors starting is controlled by inserting a ‎resistance to the current path. Using this method not only ‎reduces the stresses on both motor’s windings and power ‎system equipments by controlling the magnitude of starting ‎current, but also suppresses motor torque pulsations. In ‎addition, it restores voltage of point of common coupling ‎‎(PCC) and helps to other loads of power system. Control ‎strategy of the proposed structure is presented and analytic ‎analyses are performed in detail. Simulation results using ‎PSCAD/EMTDC software are involved to validate the ‎effectiveness of this structure for soft start up of IMs.‎

ELECO 2011, Dec 1, 2011

In this paper, transient stability improvement using bridge ‎type fault current limiter (FCL) is ... more In this paper, transient stability improvement using bridge ‎type fault current limiter (FCL) is presented in single ‎machine infinite bus (SMIB) system with a double circuit ‎transmission line. Three single-phase sets of the proposed ‎FCL are installed at the beginning of feeder. The proposed ‎FCL inserts an inductance and a resistance in the fault ‎current pass. The insertion inductance and resistance not ‎only limits the fault current level in an acceptable value but ‎also improves transient stability of power system by ‎consuming excessive energy of synchronous generator ‎during fault. To reach maximum transient stability, the ‎optimal resistor value of the proposed FCL is calculated. ‎Analytical analysis and simulation results using ‎PSCAD/EMTDC software are presented to show the ‎current limiting future and transient stability enhancement ‎using the proposed FCL in SMIB.‎

In this paper a controllable resistive type fault current limiter (CR-FCL) is introduced. The CR-... more In this paper a controllable resistive type fault current limiter (CR-FCL) is introduced. The CR-FCL inserts a pre-specified value of resistance based on a pre-defined function, by using a simple switching method, in series with the fault current path. When a fault occurs, a self turn off switch starts switching with a pre-specified frequency and duty cycle. By this switching pattern, the controlled value of resistance enters to the fault current path. So, the CR-FCL limits the fault current to the desired values. In addition, from transient stability point of view, by inserting the optimal resistance value, the CR-FCL is capable to enhance power system transient stability in a good manner. In fact, generation of the controllable resistance that depends on the duty cycle of the self turn off switch is the main idea of the CR-FCL. The variable duty cycle results the variable resistance and the fixed duty cycle results the fixed resistance. Analytical analyses of the proposed FCL are presented in details. Simulation results by power system computer-aided design/electromagnetic transients, including dc (PSCAD/EMTDC) software and corresponding experimental results are studied to validate the effectiveness of the CR-FCL. Considering error analyses, there is the good agreement between the simulation results and the experimental results.

Power and Energy (PECon), …, Jan 1, 2010

In this paper, enhancement of transient stability of Single Machine Infinite Bus (SMIB) system wi... more In this paper, enhancement of transient stability of Single Machine Infinite Bus (SMIB) system with a double circuit transmission line using a Non-superconducting Fault Current Limiter (NSFCL) is proposed. Stability analysis for such system is discussed in detail. It is shown that, the stability depends on the resistance of NSFCL in fault condition. To effective improvement of stability, the optimum value of NSFCL resistance is calculated. Simulation results by PSCAD/EMTDC software are presented to confirm the analytic analysis accuracy.

Renewable Energy, 2020

Applied Energy, 2020

IEEE Transactions on Transportation Electrification, 2019

IEEE iTEC , 2018

Journal of Energy Storage , 2018

IEEE Transactions on Transportation Electrification

Power and Energy (PECon), …, Jan 1, 2011

ELECO 2011, Dec 1, 2011

Power and Energy (PECon), …, Jan 1, 2010

Applied Energy Symposium: MIT A+B, 2020

Due to the environmental impacts brought by current energy schemes, the energy transition, a new ... more Due to the environmental impacts brought by current energy schemes, the energy transition, a new paradigm-shift from fossil fuels to renewable energy, has been widely accepted and is being realized through collective international and local efforts. Electricity, as the most direct and effective use of renewable energy sources (RES), plays a key role in the energy transition. In this paper, we first discuss a viable pathway to energy transition through the electricity triangle, highlighting the role of RES in electricity generation. Further, we propose methodologies for the planning of wind and solar PV, as well as how to address their uncertainty in generation expansion problems. Finally, by using a web-based tool, "RES-PLAT" 1 , we demonstrate the scheme in a case study of the North Africa, which evaluates the impacts and benefits of a large-scale RES expansion.

INFORMS Annual Meeting, 2019

The power systems across the world have been changing rapidly over the past decade as the deploym... more The power systems across the world have been changing rapidly over the past decade as the deployment of renewables-based generation technologies, particularly wind and solar photovoltaics (PV) has grown. With the rise of intermittent renewables to the point where they are now delivering a meaningful proportion of total generation in many power systems, concerns have begun to emerge regarding the challenges of grid integration and system reliability. One mechanism to mitigate any issues related to intermittency is using energy storage systems. Advanced energy storage technologies such as batteries, can provide the grid with the added flexibility needed to reliably accommodate much higher levels of variable and intermittent renewable generation.
The traditional approach in assessing the economic performance of energy storage systems such as lithium-ion batteries for grid-applications has been to adopt simple formulations to represent the storage capability limiting the representation of the non-linear nature of the physical behavior of the storage technologies. Simplistic representations of energy storage technologies in power systems models typically include fixed roundtrip efficiencies or fixed rated power while these parameters are dependent on the storage device’s power output/input as well as the state of charge. Additionally, the batteries degrade over time and these result from both calendar and cycle aging mechanisms. We have developed a model for combined offshore wind-battery energy storage system while modeling the power-limit of the battery and its efficiency as a function of the output power and the state of charge and accounting for its degradation. The resulting mathematical model is in the form of mixed-integer linear programming (MILP), while accounting for nonlinear behavior of the battery through piecewise linearization method. With developed model, we study different configurations of the system (onshore, offshore and hybrid design of the battery) on multiple li-ion chemistries in different power system application such as energy arbitrage and evaluate each scenario’s profitability.

IEEE PES General Meeting, 2019

This paper elaborates on the characterization of vanadium redox flow battery (VRFB) performance f... more This paper elaborates on the characterization of vanadium redox flow battery (VRFB) performance for energy arbitrage optimization based on the experimental data obtained in-house. Typical figures-of-merit used for evaluating VRFBs include coulombic, voltaic, and energy efficiencies. However, these metrics along with the deliverable power vary as a function of discharge/charge current during cycling. Thus, using a basic energy storage model with constant efficiency and fixed maximum power is not a rigorous approach for predicting the performance of VRFBs in applications with variable supply/demand of electricity. Moreover, optimization based on such an oversimplified treatment may result in inaccurate battery dispatch signal and may overestimate arbitrage profit. Here, we propose a more detailed VRFB model with dynamic efficiency and maximum power limits as a function of state of charge (SOC). These data were obtained using lab-scale VRFB cells over a range of operating conditions. The dynamic model's performance is compared to the basic model for various day-ahead electricity price profiles. Substantial difference between the predictions of two modeling approaches on the battery dispatch and profits was observed. The results indicate that the dynamic model provides more accurate predictions on the battery performance for applications with intermittent energy profile. Index Terms-Vanadium redox flow battery; energy arbitrage; day-ahead price; mixed integer linear programming.

Applied Energy A+B Symposium , 2019

In this paper, we evaluate the current status and future outlook (i.e. for 2022, 2030, 2050) for ... more In this paper, we evaluate the current status and future outlook (i.e. for 2022, 2030, 2050) for renewable electricity sources (RES) in Italy, considering the present challenges and solutions for relevant dimensions such as technology advancement and environmental impacts. We provide quantitative projections for future RES penetration levels in Italy, based on a cost-based analysis of future generation expansion. The economic expansion analysis indicates that by 2050, more than 80% of the electricity will be provided by RES even in the absence of a CO2 price. This share can even go above 90% with CO2 emission reduction measures such as a CO2 price. These high penetrations of RES will lead to a substantial reduction of the CO2 emissions from the electricity sector in Italy.

BATTERIES: BOTTLENECK FOR ELECTRIFICATION • Hybridized/Electrified transportation system require ... more BATTERIES: BOTTLENECK FOR ELECTRIFICATION
• Hybridized/Electrified transportation system require an energy storage system on-board.
• To provide the maximum range with minimal weight and volume, Li-ion is preferred over other technologies.
• Li-ion technologies require advanced battery management systems (BMS) to operate safely at maximum performance.

BATTERY MULTI-TASKING
• Batteries provide different tasks during driving, recharging and with the possibility of vehicle-to-grid integration
• Aging cause to be factored when scheduling a specific task

GOALS OF THIS RESEARCH
• Quantify the accuracy variations in a Li-ion battery model
when considering higher and lower order ECM models.
• Time domain and frequency domain testing and modeling.
• Develop a simulation tool (REV-Cycle) including EV powertrain and battery model for daily pattern aging study.

To reduce greenhouse gas emissions, fuel efficiency driver interfaces (FEDIs) have been proposed ... more To reduce greenhouse gas emissions, fuel efficiency driver interfaces (FEDIs) have been proposed and shown positive results. While FEDIs can
be a stopgap, it does not seem to solve the problem in a fundamental way. Electric vehicles can radically decrease greenhouse gas emissions, but they have also brought about another issue, “battery” performance and aging. Research shows that driving style can affect battery
performance and aging of electric vehicles. In the present paper, we introduce our multiple approaches to develop persuasive technology for driver behavior modification in electric vehicles. To investigate the effects of driver characteristics and driving style on battery performance and maintenance, we propose a research study using driving simulation data, naturalistic driving data, and driving style questionnaires.