CN-121995222-A - Battery SOH estimation method based on fractional equivalent and memristor degradation modeling

CN121995222ACN 121995222 ACN121995222 ACN 121995222ACN-121995222-A

Abstract

The application belongs to the technical field of battery SOH estimation, and particularly discloses a battery SOH estimation method based on fractional equivalent and memristor degradation modeling, which comprises the steps of constructing a joint state space model according to a fractional equivalent circuit model and a memristor degradation model based on operation data of a battery; the method comprises the steps of establishing a joint state space model by combining a fractional equivalent circuit model and a memristor degradation model, solving the joint state space model by adopting an unscented Kalman filtering algorithm to synchronously estimate the health state of a battery and parameters of the fractional equivalent circuit model on line, correcting and adjusting an estimation process based on a double time scale, wherein the double time scale comprises short time scale on-line correction based on the intrinsic consistency of the model and long time scale off-line correction based on the review of historical data, and outputting a corrected and adjusted battery health state estimation result. The application can realize the self-adaptive estimation of the SOH of the battery with high precision, strong mechanism and full life cycle.

Inventors

CHEN JIE
LIU FENG
LIU MENGYANG
Peng Zhuangzhuang

Assignees

中国地质大学（武汉）

Dates

Publication Date: 20260508
Application Date: 20251215

Claims (10)

1. The battery SOH estimation method based on fractional equivalent and memristor degradation modeling is characterized by comprising the following steps: S10, acquiring operation data of the battery in a current charge-discharge cycle, wherein the operation data comprise working current, port voltage, temperature and state of charge; s20, constructing a joint state space model according to the fractional equivalent circuit model and the memristor degradation model based on the operation data; The memristor degradation model is used for calculating comprehensive stress factors based on a plurality of stress factors in the battery charging and discharging cycle process, and updating memristor state values according to a nonlinear updating rule, wherein the state values are used for representing accumulated damage of a battery; s30, combining the fractional equivalent circuit model with the memristor degradation model to establish a joint state space model, and solving the joint state space model by adopting an unscented Kalman filtering algorithm to synchronously estimate the health state of the battery and the parameters of the fractional equivalent circuit model on line; S40, correcting and adjusting the estimation process based on a double time scale, wherein the double time scale comprises short time scale online correction based on the internal consistency of the model and long time scale offline correction based on the historical data review; s50, outputting the corrected and adjusted battery state of health estimation result.
2. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S20, the order of the fractional element comprises an order characterizing a charge transfer process And characterizing the order of the solid-phase diffusion process ; The order of the charge transfer process Is modeled as: the order of the characterization of the solid phase diffusion process Is modeled as: in the formula, Indicating that the new battery is at the reference temperature The initial order below; is the attenuation coefficient of memristive damage to the order; is a temperature influence coefficient, T represents the battery temperature, and M represents the memristor state value.
3. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S20, the integrated stress factor Calculated according to the following formula: Wherein, the And The current multiplying factor influence coefficient and the average absolute multiplying factor are respectively; And Depth of discharge and its power law index, respectively; Is equivalent activation energy; Is a gas constant; Is the average temperature; indicating the reference temperature.
4. The method for estimating the SOH of the battery based on fractional equivalent and memristor degradation modeling as claimed in claim 1, wherein in step S20, the memristor state value is updated according to a nonlinear update rule, and a calculation formula is as follows: in the formula, Representing the memristor state value after the current cycle; representing the memristor state value after the last cycle, and the initial value ; Is the basal aging rate associated with the average SOC; A composite stress factor representing the current cycle; Is provided with a power exponent Is used for the saturation inhibition term of (c), Is the theoretical maximum charge loss.
5. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S20, the state vector of the joint state space model includes a state of charge of the battery, an overpotential indicative of a charge transfer process, an overpotential indicative of a solid phase diffusion process, a charge transfer resistance, a solid phase diffusion resistance, and a memristor state value.
6. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S40, the short-time scale online correction includes dynamically adjusting a process noise covariance matrix of the unscented Kalman filter based on a consistency residual between a charge transfer resistance estimated from the unscented Kalman filter and a theoretical internal resistance calculated from the memristor degradation model.
7. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S40, the long-time-scale offline correction includes periodically utilizing historical charge-discharge cycle data and corresponding health status truth values, constructing a global optimization problem targeting at minimizing health status estimation errors, and solving by an optimization algorithm to update the super-parameters in the fractional equivalent circuit model and the memristor degradation model.
8. The method for estimating the SOH of the battery based on fractional equivalent and memristor degradation modeling of claim 7, wherein the optimization algorithm is a particle swarm optimization algorithm.
9. The method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of claim 1, wherein in step S50, the output battery state of health estimation result includes a state of health value updated in real time, and a multidimensional diagnostic index derived from the fractional equivalent circuit model parameters and memristor state values.
10. A battery management system, comprising a processor and a memory, the memory storing a computer program which, when executed by the processor, implements the steps of the method for estimating SOH of a battery based on fractional equivalent and memristor degradation modeling of any one of claims 1 to 9.

Description

Battery SOH estimation method based on fractional equivalent and memristor degradation modeling Technical Field The application belongs to the technical field of battery SOH estimation, and particularly relates to a battery SOH estimation method based on fractional equivalent and memristor degradation modeling. Background With the rapid development of industries such as electric automobiles, energy storage power stations and the like, the lithium ion battery is used as a core energy carrier, and the safety, the reliability and the economy of the whole life cycle of the lithium ion battery are paid unprecedented attention. A Battery Management System (BMS) is a brain that ensures efficient and safe operation of a battery pack, and accurate estimation of a State of Health (SOH) of the battery is a fundamental basis for the BMS to make key decisions such as energy management, thermal management, equalization control, and life prediction. Inaccurate SOH estimation may cause "cliff" drop of battery overcharge and overdischarge, and endurance mileage, and even cause serious safety accidents such as thermal runaway. Currently, the dominant SOH estimation technique relies mainly on the Equivalent Circuit Model (ECM) of the battery. Conventional integer-order ECMs, such as the widely used second order RC model, are physically limited in interpretation capability, although simple in structure and easy to implement. The capacitive, resistive elements in these models cannot exactly correspond to complex electrochemical processes inside the cell, such as the double layer effect during charge transfer and the diffusion process of lithium ions in the electrode material (i.e. Warburg impedance). This model-level approximation makes it difficult to maintain high accuracy over a wide range of operating temperatures and dynamic conditions, and model mismatch problems become more serious especially at the late battery aging stage. To improve model accuracy, researchers began to introduce fractional calculus theory. Fractional order models, with their inherent "memory properties", are able to more naturally and accurately describe the impedance spectrum characteristics of an electrochemical system through a constant phase angle element (CPE). However, most of the existing studies stay "better fitting" experimental data with fractional order models, without deep mining the physical connotation of the core parameter of fractional order. In these studies, the order is often regarded as a fixed mathematical fit constant, neglecting the key fact that as the cell ages, the internal microstructure (e.g. electrode surface morphology, active particle integrity) changes irreversibly, which changes necessarily lead to changes in its electrochemical kinetics and ultimately are reflected in the dynamic evolution of the fractional order. Thus, establishing a model that directly relates the order parameters to the internal physical state of the battery is critical in achieving the transition of the model from "mathematical fit" to "physical characterization". On the other hand, the essence of SOH is the decay of the maximum available capacity of the battery relative to its initial capacity. The attenuation is the accumulated result of internal irreversible chemical side reactions (such as SEI film thickening and active lithium loss) under the combined action of various external stresses such as cyclic charge and discharge, temperature change, high-rate working conditions and the like. The prior attenuation models describing the process are mostly empirical formulas or data-driven black box models, and although the attenuation models can fit a capacity attenuation curve under specific conditions, the attenuation models are poor in generalization capability and cannot reveal the dependency of aging paths on different stress histories. How to construct a mechanism model capable of "memorizing" and "quantifying" the historical accumulated damage and directly correlating it with capacity fade is a bottleneck to be broken through in the current SOH estimation field. Memristors are used as a nonlinear element with a memory function, and a theoretical framework provides a potential mathematical tool for describing the cumulative nonlinear degradation process, but the application exploration in the field of battery health modeling is still in a starting stage. In summary, the existing methods have significant shortcomings in terms of model mechanization, parameter physics, and adaptive characterization of the aging process. Therefore, how to realize high-precision, strong mechanism and full life cycle self-adaptive estimation of the SOH of the battery is a problem to be solved currently. Disclosure of Invention Aiming at the defects of the prior art, the application aims to provide a battery SOH estimation method based on fractional equivalent and memristor degradation modeling, which can realize high-precision, strong mechanism and full life cycle se