PhD Student, Department of Electronics & Electrical Engg., Indian Institute of Technology Guwahati
Battery Model Identification and its State of Charge Estimation
Lithium-ion (Li-ion) batteries are popularity due to high power rating, high energy density, longer life cycle, better electrochemical strength, rechargeability, low weight and low maintenance. The dynamics of Li-ion battery are non-linear, the exact modeling is a challenging task. Of late, different equivalent circuit models (ECMs) have been used to represent the electrochemical model of Li-ion batteries. Such models result in computational efficiency and structural simplicity leading to an easy real-time implementation. Some of the ECMs include Rint model, Thevenin model and RC model etc. In order to ensure a reliable performance of hybrid electric vehicle (HEV), a precise estimation of state of charge (SOC) in Li-ion battery is a crucial step, wherein the exact knowledge of battery model parameters is essential. Some of the commonly employed methods for SOC estimation include ampere-hour (Ah) counting approach, open circuit voltage (OCV) method and artificial neural networks.
We have proposed a novel online structure for an integrated parameter extraction and state of charge estimation for Li-ion batteries. A state space based relay feedback approach for parameter extraction of Li-ion battery is utilized for exact model identification. The identification method is capable of combating the ill effects of sensor noise yielding an accurate estimation of battery parameters. The identification procedure has the merits over a step/pulse testing as follows; 1) it identifies process information around the important frequency, the ultimate frequency (the frequency where the phase angle is −π), 2) it is a closed-loop test therefore, the process will not drift away from the nominal operating point, 3) for processes with a long time constant, it is a more time-efficient method than conventional step or pulse testing. Further, a smooth and robust estimation of SOC is achieved using a finite time extended state observer with no chattering. Key contributions are as follows: 1) a smart online structure is proposed, yielding accurate system parameters in addition to precise SOC estimation; 2) the proposed integrated scheme is robust towards large measurement noise leading to a bounded estimation error, without the requirement of the signal averaging technique; 3) this scheme is devoid of any current sensor usage, merely the voltage available at the output terminals of the battery is utilized in entire SOC estimation process. 4) no persistence of excitation required.
Abstract: In this paper, the problem of optimal sizing of a series PHEV is studied by formulating a convex program that minimizes the sum of operational and component costs. The solution gives the optimal sizes of the main powertrain components, simultaneously with the vehicle's optimal energy management. Investigations are performed on driving cycles generated stochastically from real data using Markov chains, with different driving distance distributions and charging patterns. The results show that the optimal component sizing is affected more from the driving distances between charging opportunities, than the speed profile of the driving. With anticipated future battery and petroleum prices, larger battery sizes are obtained.
Pub.: 28 Feb '16, Pinned: 25 Aug '17
Abstract: This paper investigates networked control of a collection of battery-powered systems with seriously limited communication capacity and power resources. We aim to stabilize the systems by effectively assigning the communication channels and appropriately allocating the transmission powers so that the energy consumption is within an energy budget. The role of channel assignment is to guarantee network access for all plants when needed; the mission of power allocation is to ensure a desired rate of successful packet transmission for each channel. These two aspects are achieved by a scheduling policy and a power allocation method, respectively, each of which is derived based on stability and schedulability requirements. An interesting co-design framework is derived for communication scheduling, transmission power allocation and stabilizing control. The presented methodology can guarantee a desired decay rate and a given energy consumption for each plant. The effectiveness of the results is demonstrated by numerical simulations. Copyright © 2017 John Wiley & Sons, Ltd.
Pub.: 05 Jan '17, Pinned: 25 Aug '17
Abstract: In this paper, we provide a general framework for robust optimal estimation over a lossy and delayed network. A threshold principle is introduced to integrate network‐induced uncertainties into packet losses, which are modeled with a Bernoulli process. Based on stability conditions derived from two Riccati equations, we show the existence of critical observation arrival probabilities below which the optimal estimator stochastically fails to converge. Moreover, the result is extended to a real system with variable process disturbance, which has an indicator for its admissible bound in terms of a given restriction of estimation accuracy. The proposed method is experimented on a specific automobile application, the battery state of charge estimation. Copyright © 2015 John Wiley & Sons, Ltd.
Pub.: 01 Oct '15, Pinned: 25 Aug '17