Pith. sign in

REVIEW 4 major objections 5 minor 29 references

Multi-Objective Nonlinear Power Split Control For BESS With Real-Time Simulation Feedback

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A mixed-integer nonlinear MPC controller that splits power between parallel battery strings, with feedback from an electro-thermal simulation, can raise battery, inverter, and derating efficiency by 1–2 percentage points while shaving 5…

desk verdict Useful MINLP-MPC formulation for BESS power split, but the headline efficiency and temperature numbers have no defined baseline and cannot be audited from the manuscript. read the letter →

arxiv 2507.04800 v1 pith:3OJZ54KA submitted 2025-07-07 eess.SY cs.SY

Vivek Teja Tanjavooru , Prashant Pant , Thomas Hamacher , Holger Hesse This is my paper · ORCID
classification eess.SYcs.SY MSC 90C1190C29
keywords BatteryEnergyStorageSystemsmulti-objectiveoptimizationMINLPmodelpredictivecontrolelectro-thermalsimulationpowersplitthermalderatinginverterloss
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that a model predictive controller built on mixed-integer nonlinear programming can decide how to split power among parallel battery strings in a battery energy storage system (BESS) while balancing inverter efficiency, battery efficiency, and thermal safety. The controller is coupled to a high-fidelity electro-thermal simulation that feeds back state of charge and temperature every control horizon, so the optimizer sees realistic battery states. In a two-string case, the balanced weighting improves battery efficiency by 1%, inverter efficiency by 1.5%, and derating efficiency by 2%, while cutting peak BESS temperature by 5 °C and keeping availability high. The central message is that these conflicting objectives can be navigated with a Pareto sweep rather than a single fixed rule.

What carries the argument

The carrying object is the MINLP-MPC optimizer with lexicographic priorities (availability first, then derating, then inverter loss, then battery loss) and blended weights for the lower-priority objectives. Loss models enter as piecewise-linear lookup tables and Big-M binary constraints: resistance as a function of SOC and temperature, inverter loss as a function of power, and derating as a piecewise-linear approximation of a PI-controlled derating factor. Current is derived from the quadratic root of the equivalent-circuit relation between open-circuit voltage, resistance, and power, and heat is the I²R product. Each optimization horizon is re-initialized with temperatures from a finite-difference spatial thermal simulation, giving the controller feedback it would not otherwise have.

What would settle it

Run the same controller on a physical two-string BESS with measured inverter loss and battery resistance data, and compare predicted versus measured temperature and efficiency traces; if the 5 °C peak-temperature reduction and the 1–2% efficiency gains do not appear, the reported improvements are artifacts of the model.

Watch

Extended reading notes

Core claim

The central claim is that a MINLP-based MPC can simultaneously improve efficiency and thermal behavior in a multi-string BESS by treating availability as a hard priority and then exploring Pareto trade-offs between inverter and battery losses. The optimizer uses a piecewise-linear lookup-table approximation of the inverter loss curve, an SOC- and temperature-dependent internal resistance model with an equivalent circuit for current, and a temperature-based derating factor, all updated with states from a 1D spatial electro-thermal simulation. Applied to a two-string homogeneous BESS, the balanced scenario yields the highest system efficiency, with gains of 1% battery efficiency, 1.5% inverter efficiency, and 2% derating efficiency, a 5 °C lower peak temperature, and no loss of availability.

Load-bearing premise

The loss models and thermal coefficients inside both the optimizer and the co-simulation are assumed to represent a real BESS; the inverter lookup table comes from a commercial inverter and the battery data from one manufacturer, and the thermal coefficients and LUT slopes are not shown or validated against measurements in the paper.

Editorial extensions

If this is right

  • BESS operators could improve round-trip efficiency and reduce thermal stress without sacrificing availability by choosing balanced weights between inverter and battery losses.
  • The Pareto sweep provides a principled way to choose operating points that favor either inverter efficiency or battery longevity, depending on market or degradation priorities.
  • Real-time feedback from a high-fidelity thermal simulation makes the MPC state-aware, enabling predictive derating rather than reactive temperature limits.
  • The framework can be extended to include aging and economic objectives, turning the power split decision into a lifetime-cost optimization.
  • The PI-controlled derating curve can be tuned to shift the trade-off between thermal degradation and system uptime.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The value of the method depends heavily on how well the lookup tables and thermal coefficients represent a real system; validating the unshown k1 and k2 coefficients and the inverter LUT against measurements is the natural next step.
  • The same architecture should transfer to heterogeneous strings with different SOC, temperature, or aging states, where the Pareto trade-offs are likely to be larger.
  • A finer Pareto sweep could generate a dynamic weight lookup table for online decisions, an extension the authors list as future work.
  • Because only a two-string homogeneous case is shown, scaling to many strings may hit MINLP computational limits, so cheaper surrogate models may be needed for larger installations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes an MINLP-based model predictive control strategy for power split among parallel battery strings in a BESS, co-simulated with a high-fidelity electro-thermal model that returns SOC and temperature feedback. The objective is formulated as a lexicographic plus blended multi-objective optimization over availability, thermal derating losses, inverter losses, and battery losses, with a Pareto weight sweep between inverter and battery efficiency in the second case study. The central claims are that the balanced operating point improves battery efficiency by 1%, inverter efficiency by 1.5%, derating efficiency by 2%, and reduces peak temperature by 5 °C while maintaining availability. The manuscript also presents the modeling equations, two case studies, a radar plot, and temperature trajectories.

Significance. If the quantitative claims were properly supported, the paper would address a practically important problem in multi-string BESS dispatch: explicitly trading off inverter efficiency, battery losses, thermal derating, and availability while using electro-thermal feedback. The architecture — MINLP-MPC with co-simulation feedback and a lexicographic objective hierarchy — is a sensible and nontrivial contribution, and the attempt to model temperature-dependent internal resistance and derating inside the optimizer is commendable. However, the current evidence is insufficient: the claimed improvements have no defined baseline, the KPI values are not tabulated, several load-bearing model parameters are undisclosed, and the KPIs are computed from the same variables being minimized. These issues must be resolved before the central claims can be accepted.

major comments (4)
  1. [Abstract; §IV-A] The quantitative claims in the abstract and conclusion (1% battery efficiency, 1.5% inverter efficiency, 2% derating efficiency, and a 5 °C peak-temperature reduction) are not supported by a defined baseline or by reported numerical values. Section IV-A presents only a radar plot (Fig. 6) and a mean-temperature plot (Fig. 7); no table gives the KPI values from Eqs. (31)–(35) for scenarios S1–S3, and no comparator controller (e.g., equal-power split, rule-based dispatch, or one of S1/S3) is specified or simulated. Because S2 is a compromise between S1 and S3, it cannot improve over both on both inverter and battery efficiency simultaneously; the claimed deltas must be relative to an unspecified baseline. Please define the baseline explicitly, tabulate the KPI values for all scenarios, and report solver tolerances or other numerical uncertainties.
  2. [§II-D, Eqs. (8)–(14)] The availability-loss model is under-constrained. Equations (12)–(13) are one-sided upper bounds of the form p_avail,high ≤ M_avail·b_high + ε; when b_high = 1 the loss variable can still be zero without violating any constraint. The text states that the loss 'will be non-zero' when the binary is active, but no lower-bound constraint of the form p_avail ≥ m·b is present. As a result, the availability KPI in Eq. (31) may be identically 100% regardless of SOC saturation, and the lexicographic priority on availability is vacuous as written. Please add the missing lower bounds or revise the formulation and the KPI definition accordingly.
  3. [§II-E, §II-F, §II-G; Eqs. (15), (20), (21), (25), (30)] The parameterization that determines the claimed efficiencies is not disclosed. The lumped-thermal coefficients k1 and k2 in Eq. (15), the SOC- and temperature-dependent resistance LUT breakpoints and slopes in Eqs. (20)–(21), the derating LUT slopes and intercepts in Eq. (25), and the coefficients of the commercial-inverter loss LUT in Eq. (30) are all absent from the manuscript; the PI derating gains are selected by trial and error (Fig. 2). Because the claimed improvements are 1–2% efficiency and 5 °C in temperature, they are the same order as plausible parameter uncertainty, so the results cannot be reproduced or falsified from the manuscript. Please provide the parameter values and breakpoints (or a repository with the full data) and include a sensitivity analysis over them.
  4. [§II-B vs. §II-H; Eqs. (24), (26), (30), (31)–(35)] The KPIs are computed from the same loss variables that appear in the objective function: p_heat from Eq. (24) is minimized as battery loss and then used in Eq. (34), p_inv from Eq. (30) is minimized and used in Eq. (33), and p_derate,loss from Eq. (26) is minimized and used in Eq. (32). The Pareto rankings therefore partly encode the model's own loss definitions rather than an external measure of performance. To support the central claim, compute the KPIs from the independent high-fidelity electro-thermal simulator (or from measured power flows) and report the discrepancy, if any, with the optimizer-internal values.
minor comments (5)
  1. [Fig. 2] The figure shows three gain settings for the PI derating controller, but the text only states that the highlighted gains of Kp = 0.035 and Ki = 0.0035 were sufficient; please explain why the other settings were rejected and how the trial-and-error selection was evaluated.
  2. [Eq. (30)] The zero-power branch is written as '0, p^B,ch(/dch) = 0', which is unclear; furthermore, the charge and discharge branches appear to use the same index i with different segment bounds. Please rewrite the LUT definition cleanly with separate indices or explicit breakpoints.
  3. [References] References [12] and [14] have identical titles ('Evaluating the impact of model accuracy for optimizing battery energy storage systems'); please verify the citations and clarify whether they are companion papers or one is mis-cited.
  4. [§IV] The term 'Pareto analysis' is used for a sweep with only three weight combinations (S1–S3); a three-point sweep is not a Pareto frontier. Please either add more points along the weight sweep or temper the wording to 'weight-sweep analysis'.
  5. [Abstract; §VII] The abstract and title claim real-time operation and real-time simulation feedback, but no computation times, solver settings, hardware, or horizon length are reported, and Section VII lists computational benchmarking as future work. Please add runtime measurements or temper the real-time claim.

Circularity Check

3 steps flagged · score 6.0 of 10

Efficiency and temperature gains are read from the same loss terms the controller minimizes, and the thermal claim rests on the authors' own unvalidated simulation.

  1. self definitional [Section II-B Table II; Section II-H Eqs. (31)-(35)]
    "The objective function accounts for losses arising from limited available capacity for charge/discharge, inverter switching, battery heat generation, and power derating required for safe operation. ... Inverter Efficiency (%) = 1 − Σ_t Σ_m pinv[m]_t / Σ_t Σ_m pB[m]_t ... Battery Efficiency (%) = 1 − Σ_t Σ_m pheat[m]_t / Σ_t Σ_m pB[m]_t ... Derating Efficiency (%) = 1 − Σ_t Σ_m pderate,loss[m]_t / Σ_t Σ_m pB[m]_t"

    The KPIs are normalized sums of the very terms the MINLP minimizes (inverter loss pinv, heat pheat, derating loss pderate,loss). Scenario S1 with W3=1 minimizes inverter loss, so its highest inverter efficiency is a restatement of the objective weights, not an empirical prediction; S3 likewise maximizes battery efficiency by construction. The 1% battery / 1.5% inverter / 2% derating 'improvements' cited for balanced operation are differences of these objective-defined KPIs, so the qualitative ranking is forced by the weight sweep even though the magnitudes depend on the model.

  2. fitted input called prediction [Section II-F, Eqs. (25)-(26), Fig. 2]
    "The controller gains Kp and Ki, as shown in Fig. 2, were tuned using a trial-and-error approach to ensure that the battery temperature remained within the safe operational limits under various load conditions. ... In the optimizer, this nonlinear derating factor (kderate_t) as shown in Fig. 2 is approximated as a piecewise linear LUT as defined in (25) and the derating loss is computed as per (26)."

    The derating-efficiency KPI is computed from pderate_loss (Eq. 32), and pderate_loss = kderate_t · pB. The kderate_t LUT is itself an approximation of a PI-derating curve whose gains were hand-tuned by the authors. Thus the reported 2% derating-efficiency improvement is an outcome of the chosen tuning and the same optimization variable, not a measured or externally calibrated derating behavior.

1 more flagged steps
  1. self citation load bearing [Section II-A; Section II-E, Eq. (15), Fig. 1]
    "To support thermal-aware control, a high-fidelity Electro-Thermal Simulation is integrated, comprising an Equivalent Circuit Model (ECM) and a Finite Difference Method (FDM) based 1D spatial thermal simulation that evaluates internal temperature distributions based on battery power and ambient air cooling [5]. ... After each horizon of optimization, the temperature estimations of the string are corrected using a high fidelity spatial thermal simulation as shown in Fig. 1 [5]."

    The 5°C peak-temperature reduction is generated by this simulation, which is cited to the authors' own prior work [5]. The optimizer's lumped model (Eq. 15) uses coefficients k1 and k2 that are never reported, and the paper shows no measured validation of the spatial thermal model. The temperature claim therefore rests on an unverified self-citation chain rather than on an independent, machine-checked, or externally falsifiable benchmark.

full rationale

The paper's own KPI definitions use the exact loss variables that appear in the optimization objective (inverter loss, pheat, pderate_loss), so the scenario comparisons S1-S3 restate which weights were chosen rather than providing an external benchmark. The derating-efficiency improvement additionally depends on Kp/Ki gains that were hand-tuned. The temperature reduction is generated by an electro-thermal simulation cited only to the authors' prior work [5], with coefficients k1/k2 in Eq. (15) undisclosed and no measured validation in this manuscript. These are genuine circularity concerns: the claimed 1%/1.5%/2% and 5°C numbers are read from the same model that the controller optimizes. However, the MINLP-MPC architecture, the Pareto trade-off exploration, and the numerical magnitudes of the KPI differences are not literally equal to the objective by construction; S2's 'highest system efficiency' is a nontrivial model outcome. Hence partial circularity (score 6), not complete reduction.

Assumptions & free parameters 8 free parameters · 8 assumptions · 0 invented entities

The central results rest on many undisclosed or fitted components: trial-and-error PI gains, unstated thermal coefficients, hidden LUT slopes for resistance, inverter loss and derating, Big-M constants, and a static forecast. The efficiency KPIs are computed from these same models, so any error in the LUTs propagates directly into the claimed 1 to 2 percent improvements and 5 degrees C reduction.

free parameters (8)
  • PI derating gains Kp, Ki = Kp=0.035, Ki=0.0035
    Tuned by trial-and-error in Section II-F; define the derating factor curve and therefore the derating efficiency KPI.
  • Lumped thermal model coefficients k1, k2 = not reported
    Eq (15) updates string temperature inside the optimizer; coefficients are never specified.
  • Internal resistance LUT coefficients and breakpoints = 40 mOhm at SOC=0, 2.5 mOhm at SOC=0.1 and below 25C, 0.75 mOhm at 80C; slopes not reported
    Section II-E; battery heat loss and battery efficiency KPI are computed from this LUT, parameterized by FENECON data not disclosed.
  • Inverter loss LUT coefficients = 0.0193*pN charge, 0.0177*pN discharge at nominal power; slopes not reported
    Section II-G; empirical LUT from an unnamed commercial inverter; defines the inverter efficiency KPI.
  • Derating LUT slopes and intercepts x_i, y_i = not reported
    Section II-F, Eq (25); piecewise linear approximation of the PI-based derating curve used to compute derating loss.
  • SOC-OCV LUT for charge and discharge = not reported, cited to [25], [26]
    Section II-E; used in the ECM quadratic current calculation, Eq (23).
  • Thermal sensitivity of BESS to power input = 0.012 degrees C per kW over 15 minutes
    Section II-F; used to justify the slow derating strategy, derived from simulation rather than experiment.
  • Big-M constants and tolerances = M_avail, M_inv, epsilon, epsilon_soc, epsilon_inv not reported
    Required by disjunction constraints (12)-(13), (27), and lexicographic relaxation (1d); choices affect feasibility and solution quality.
assumptions (8)
  • standard math Lexicographic optimization with tolerance epsilon preserves the exact priority ordering
    Eq (1b)-(1d); relies on Gurobi's lexicographic MOO [21]; no proof that epsilon avoids suboptimal high-priority solutions.
  • domain assumption Lumped thermal model (15) adequately represents string mean temperature for control
    Section II-E; k1, k2 not identified; simulation correction assumed to fix bias.
  • domain assumption Equivalent circuit model with LUT-based internal resistance and OCV accurately represents battery electrical behavior
    Section II-E; parameters from one battery system without validation against independent measurements.
  • domain assumption Empirical inverter LUT from a commercial inverter system generalizes to both strings
    Section II-G; inverter model identity and loss data are undisclosed.
  • ad hoc to paper PI-based derating curve with trial-and-error gains represents true derating behavior
    Section II-F; not an industry-standard curve, and Kp/Ki were tuned to the simulation.
  • domain assumption Static demand forecast is perfect over the horizon
    Fig 1 labels forecast management as static; no forecast error model.
  • domain assumption 1D FDM electro-thermal simulation gives accurate real-time battery states for MPC feedback
    Section II-A; builds on authors' prior work [5] without independent validation in this paper.
  • domain assumption Two homogeneous strings are representative of multi-string BESS behavior
    Section III-A tests only a two-string homogeneous system; inhomogeneities are deferred to future work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-Objective Nonlinear Power Split Control For BESS With Real-Time Simulation Feedback." pith.science (2026). https://pith.science/paper/3OJZ54KA

@misc{pith2026250704800,
  author       = {Pith},
  title        = {Pith review of: Multi-Objective Nonlinear Power Split Control For BESS With Real-Time Simulation Feedback},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3OJZ54KA}},
  note         = {Machine review of arXiv:2507.04800}
}
read the original abstract

This paper presents a mixed-integer, nonlinear, multi-objective optimization strategy for optimal power allocation among parallel strings in Battery Energy Storage Systems (BESS). High-fidelity control is achieved by co-simulating the optimizer with a BESS electro-thermal simulation that models spatial thermal dynamics of the battery, providing real-time State of Charge (SOC) and temperature feedback. The optimizer prioritizes reliability by enforcing power availability as a hard constraint and penalizing battery thermal derating. Within these bounds, the controller performs a Pareto sweep on the relative weights of inverter and battery losses to balance the trade-off between inverter efficiency and battery efficiency. The inverter loss model is based on an empirical lookup table (LUT) derived from a commercial inverter system, while the battery thermal loss model uses SOC and temperature-dependent internal resistance, with electric current computed from the battery Equivalent Circuit Model (ECM). When the optimization was applied to a two-string BESS, the competing effects of inverter and battery losses on system availability and thermal derating were observed. The balanced operation yielded improvements of 1% in battery efficiency, 1.5% in inverter efficiency, and 2% in derating efficiency, while maintaining higher availability. Additionally, a 5 degrees C reduction in BESS peak temperature also suggests reduced thermal stress without compromising availability.

Figures

Figures reproduced from arXiv: 2507.04800 by the authors.

Figure 1
Figure 1. Framework showcasing the interaction of Power Split Control with [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Temperature-based derating factor vs. battery temperature. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Performance values of KPIs in Case 1: Availability as the primary [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: SOC evolution of two string system for Case 1. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Temperature evolution of two string system for Case 1. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 5
Figure 5. Figure 5: These results suggest that prioritizing inverter effi [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Pareto Analysis for inverter (W3) and battery (W4) losses in Case 2. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Plot of mean BESS temperature of S1–S3 scenarios in Case 2. [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

  1. [1]

    Prashant Pant, Thomas Hamacher, and Vedran S. Peri ´c. An- cillary services in germany: Present, future and the role of battery storage systems. TechRxiv, March 30 2024. DOI: 10.36227/techrxiv.171177336.66870344/v2. 10

    work page arXiv 2024

  2. [2]

    Lithium-ion battery lifetime extension: A review of derating methods

    Haijun Ruan, Jorge Varela Barreras, Timothy Engstrom, Yu Merla, Robert Millar, and Billy Wu. Lithium-ion battery lifetime extension: A review of derating methods. Journal of Power Sources , 563:232805, 2023

    work page 2023

  3. [3]

    Lifetime estimation of grid-connected battery storage and power electronics inverter providing primary frequency regulation

    Marco Stecca, Thiago Batista Soeiro, Laura Ramirez Elizondo, Pavol Bauer, and Peter Palensky. Lifetime estimation of grid-connected battery storage and power electronics inverter providing primary frequency regulation. IEEE Open Journal of the Industrial Electronics Society , 2:240–251, 2021

    work page 2021

  4. [4]

    Barreras, Billy Wu, and Gregory J

    Michael Schimpe, Jorge V . Barreras, Billy Wu, and Gregory J. Of- fer. Battery degradation-aware current derating: An effective method to prolong lifetime and ease thermal management. Journal of The Electrochemical Society, 168(6):060506, jun 2021

    work page 2021

  5. [5]

    Optimal power split control for state of charge balancing in battery systems with integrated spatial thermal analysis and aging estimation

    Vivek Teja Tanjavooru, Melina Graner, Prashant Pant, Thomas Hamacher, and Holger Hesse. Optimal power split control for state of charge balancing in battery systems with integrated spatial thermal analysis and aging estimation. Energy Storage, 7(4), 2025

    work page 2025

  6. [6]

    Critical review of optimal control methods for li–ion batteries in electric vehicles

    Yeonsoo Kim. Critical review of optimal control methods for li–ion batteries in electric vehicles. ChemElectroChem, 11(3), 2024

    work page 2024

  7. [7]

    Model predictive control for lithium-ion battery optimal charging

    Changfu Zou, Chris Manzie, and Dragan Nesic. Model predictive control for lithium-ion battery optimal charging. IEEE/ASME Transactions on Mechatronics, 23(2):947–957, 2018

    work page 2018

  8. [8]

    Multi- objective energy management strategy based on pso optimization for power-split hybrid electric vehicles

    Aimin Du, Yaoyi Chen, Dongxu Zhang, and Yeyang Han. Multi- objective energy management strategy based on pso optimization for power-split hybrid electric vehicles. Energies, 14(9):2438, 2021

    work page 2021

Show all 29 references
  1. [9]

    Multi- objective predictive balancing control of battery packs based on predic- tive current

    Wenbiao Li, Longyun Kang, Xiangwei Guo, and Yuan Yao. Multi- objective predictive balancing control of battery packs based on predic- tive current. Energies, 9(4):298, 2016

    2016

  2. [10]

    Assessing battery degradation as a key performance indicator for multi-objective optimization of multi-carrier energy systems

    Lingkang Jin, Milad Kazemi, Gabriele Comodi, and Christina Papadim- itriou. Assessing battery degradation as a key performance indicator for multi-objective optimization of multi-carrier energy systems. Applied Energy, 361:122925, 2024

    2024

  3. [11]

    Temperature effect and thermal impact in lithium-ion batteries: A review

    Shuai Ma, Modi Jiang, Peng Tao, Chengyi Song, Jianbo Wu, Jun Wang, Tao Deng, and Wen Shang. Temperature effect and thermal impact in lithium-ion batteries: A review. Progress in Natural Science: Materials International, 28(6):653–666, 2018

    2018

  4. [12]

    Evaluating the impact of model accuracy for optimiz- ing battery energy storage systems

    Sheng Yin, Vivek Teja Tanjavooru, Thomas Hamacher, Christoph G¨obel, and Holger Hesse. Evaluating the impact of model accuracy for optimiz- ing battery energy storage systems. arXiv preprint arXiv:2507.00628 , 2025

    2025 arXiv

  5. [13]

    Design and analysis of an aging–aware energy management system for islanded grids using mixed–integer quadratic programming

    V olkan Kumtepeli, Yulong Zhao, Maik Naumann, Anshuman Tripathi, Youyi Wang, Andreas Jossen, and Holger Hesse. Design and analysis of an aging–aware energy management system for islanded grids using mixed–integer quadratic programming. International Journal of Energy Research,...

    2019

  6. [14]

    Evaluating the impact of model accuracy for optimizing battery energy storage systems

    Martin Cornejo, Melina Graner, Holger Hesse, and Andreas Jossen. Evaluating the impact of model accuracy for optimizing battery energy storage systems. arXiv preprint arXiv:2506.17059 , 2025

    2025 arXiv

  7. [15]

    Hierarchical distributed coordinated control for battery energy storage systems participating in frequency regulation

    Bingqing Yu, Qingquan Lv, Zhenzhen Zhang, and Haiying Dong. Hierarchical distributed coordinated control for battery energy storage systems participating in frequency regulation. Energies, 15(19):7283, 2022

    2022

  8. [16]

    Mahamudul Hasan, Boris Berseneff, Tim Meulenbroeks, Igor Cantero, Sajib Chakraborty, Thomas Geury, and Omar Hegazy

    Md. Mahamudul Hasan, Boris Berseneff, Tim Meulenbroeks, Igor Cantero, Sajib Chakraborty, Thomas Geury, and Omar Hegazy. A multi-objective co-design optimization framework for grid-connected hybrid battery energy storage systems: Optimal sizing and selection of technology. Ener...

    2022

  9. [17]

    Optimal allocation of bess in distribution network based on improved equilibrium optimizer

    Yang Li, Chengshan Wang, Hongzhan Li, and Zhipeng Wu. Optimal allocation of bess in distribution network based on improved equilibrium optimizer. Frontiers in Energy Research, 10:936592, 2022

    2022

  10. [18]

    Multi- objective battery energy storage optimization for virtual power plant applications

    Hui Song, Mingchen Gu, Chen Liu, Ali Moradi Amani, Mahdi Jalili, Lasantha Meegahapola, Xinghuo Yu, and George Dickeson. Multi- objective battery energy storage optimization for virtual power plant applications. Applied Energy, 352:121860, 2023

    2023

  11. [19]

    Prashant Pant, Holger Hesse, Vivek Teja Tanjavooru, Martin Cornejo, Thomas Hamacher, and Vedran S. Peri ´c. Soc-temperature based power sharing algorithm for battery racks in grid ancillary services. In 2023 4th International Conference on High Voltage Engineering and Power Sy...

    2023

  12. [20]

    A hierarchical and self-evolving digital twin (hse-dt) method for multi- faceted battery situation awareness realisation

    Kai Zhao, Ying Liu, Yue Zhou, Wenlong Ming, and Jianzhong Wu. A hierarchical and self-evolving digital twin (hse-dt) method for multi- faceted battery situation awareness realisation. Machines, 13(3):175, 2025

    2025

  13. [21]

    L. L.C. Gurobi Optimization. Gurobi optimizer reference manual: Multi- objective optimization, 2024,

    2024

  14. [22]

    Lithium-ion battery state-of-charge estimation based on an improved coulomb-counting algorithm and uncertainty evaluation

    Fazel Mohammadi. Lithium-ion battery state-of-charge estimation based on an improved coulomb-counting algorithm and uncertainty evaluation. Journal of Energy Storage , 48:104061, 2022

    2022

  15. [23]

    Entropy measurement of a large format lithium ion battery and its application to calculate heat generation

    Chil-Hoon Doh, Yoon-Cheol Ha, and Seung-wook Eom. Entropy measurement of a large format lithium ion battery and its application to calculate heat generation. Electrochimica Acta, 309:382–391, 2019

    2019

  16. [24]

    Evaluation of lithium-ion battery equivalent circuit models for state of charge estimation by an experimental approach

    Hongwen He, Rui Xiong, and Jinxin Fan. Evaluation of lithium-ion battery equivalent circuit models for state of charge estimation by an experimental approach. Energies, 4(4):582–598, 2011

    2011

  17. [25]

    Improved ocv model of a li-ion nmc battery for online soc estimation using the extended kalman filter

    Ines Baccouche, Sabeur Jemmali, Bilal Manai, Noshin Omar, and Najoua Amara. Improved ocv model of a li-ion nmc battery for online soc estimation using the extended kalman filter. Energies, 10(6):764, 2017

    2017

  18. [26]

    Standard battery energy storage system profiles: Analysis of various applications for stationary energy storage systems using a holistic simulation framework

    Daniel Kucevic, Benedikt Tepe, Stefan Englberger, Anupam Parlikar, Markus M ¨uhlbauer, Oliver Bohlen, Andreas Jossen, and Holger Hesse. Standard battery energy storage system profiles: Analysis of various applications for stationary energy storage systems using a holistic simu...

    2020

  19. [27]

    Temperature-driven path dependence in li-ion battery cyclic aging

    Max Feinauer, Margret Wohlfahrt-Mehrens, Markus H ¨olzle, and Thomas Waldmann. Temperature-driven path dependence in li-ion battery cyclic aging. Journal of Power Sources , 594:233948, 2024

    2024

  20. [28]

    Jorge Varela Barreras, Trishna Raj, and David A. Howey. Derating strategies for lithium-ion batteries in electric vehicles. 44th Annual Conference of the IEEE Industrial Electronics Society , 2018

    2018

  21. [29]

    Power flow distribution strategy for im- proved power electronics energy efficiency in battery storage systems: Development and implementation in a utility-scale system

    Michael Schimpe, Christian Piesch, Holger Hesse, Julian Paß, Stefan Ritter, and Andreas Jossen. Power flow distribution strategy for im- proved power electronics energy efficiency in battery storage systems: Development and implementation in a utility-scale system. Energies, 1...

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.