HDL Code Generation from SIMULINK Environment for Li-Ion Cells State of Charge and Parameter Estimation

Stighezza, Mattia; Bianchi, Valentina; Munari, Ilaria De

doi:10.1007/978-3-030-66729-0_16

Cited by 1 publication

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In some implementations, SoC evaluation combines ECM and different methods, such as Kalman Filters (KF) or Proportional-Integral (PI) Observer [14], to improve accuracy despite the increase in computational demand [11]. Different implementations were presented in research based on microcontroller systems [15][16][17], but complex models and algorithms also led to solutions exploiting the intrinsic parallelism of FPGA devices [18][19][20][21]. For instance, in [21], the model-based design of a Mixed Algorithm (MA) combining ECM and Coulomb Counting (CC) for SoC estimation was proposed.…”

Section: Introductionmentioning

confidence: 99%

“…Different implementations were presented in research based on microcontroller systems [15][16][17], but complex models and algorithms also led to solutions exploiting the intrinsic parallelism of FPGA devices [18][19][20][21]. For instance, in [21], the model-based design of a Mixed Algorithm (MA) combining ECM and Coulomb Counting (CC) for SoC estimation was proposed. The Simulink model was automatically converted into VHDL code through the HDL Coder Tool.…”

Section: Introductionmentioning

confidence: 99%

“…For example, the aforementioned case of 60 cells in series could be easily managed by this solution. By comparison, the ECM model presented in [21] occupied 23% of the whole FPGA, preventing more than four instances from being programmed on the same platform. Moreover, unlike in [43], no DSP slices were used, allowing the use of smaller and low-cost devices as well.…”

mentioning

confidence: 99%

“…The accuracy estimation of the proposed ACO-optimized SVR model was compared with the cell model-based approach [21] in Table 5. In this case, two different implementations were considered.…”

mentioning

confidence: 99%

“…In the first implementation, a constant parameters solution was analyzed, i.e., mean values of extracted ECM parameters were assumed. Next, a variable parameters approach was used, i.e., each ECM parameter value was stored in a Look-Up-Table indexed by the current estimated SoC [21].…”

mentioning

confidence: 99%

See 4 more Smart Citations

FPGA Implementation of an Ant Colony Optimization Based SVM Algorithm for State of Charge Estimation in Li-Ion Batteries

Stighezza¹,

Bianchi²,

Munari³

2021

Energies

Self Cite

View full text Add to dashboard Cite

Monitoring the State of Charge (SoC) in battery cells is necessary to avoid damage and to extend battery life. Support Vector Machine (SVM) algorithms and Machine Learning techniques in general can provide real-time SoC estimation without the need to design a cell model. In this work, an SVM was trained by applying an Ant Colony Optimization method. The obtained trained model was 10-fold cross-validated and then designed in Hardware Description Language to be run on FPGA devices, enabling the design of low-cost and compact hardware. Thanks to the choice of a linear SVM kernel, the implemented architecture resulted in low resource usage (about 1.4% of Xilinx Artix7 XC7A100TFPGAG324C FPGA), allowing multiple instances of the SVM SoC estimator model to monitor multiple battery cells or modules, if needed. The ability of the model to maintain its good performance was further verified when applied to a dataset acquired from different driving cycles to the cycle used in the training phase, achieving a Root Mean Square Error of about 1.4%.

show abstract