Sep.2026 12
意見: 1005
Data-Driven Charging: Machine Learning for NiMH SOC, Charge Termination and Adaptive Fast Charge
介紹
Machine-learning and neuro-fuzzy approaches to NiMH charging: ANFIS multi-stage constant current, learned SOC/thermal models and termination classifiers; what data they need, how to validate them safely, and their limits versus physics observers.
細節

Data-Driven Charging: Machine Learning for NiMH SOC, Charge Termination and Adaptive Fast Charge

As chargers gain processing power and fleets generate data, machine learning offers a new route to charge decisions that classical models derive from physics: learning the current stage that maximises efficiency (the ANFIS-controlled multi-stage constant-current schemes reported for NiMH/NiCd), recognising the end-of-charge pattern from voltage and temperature traces, and correcting model residuals in SOC and thermal estimation. This final estimation-group paper surveys what data-driven charging actually does well in NiMH, the training data and validation discipline it demands, and why production-safe designs keep a physics floor and hardwired limits beneath any learned policy - so the flexibility of learning never becomes an unbounded charge current.

Where learning enters the charge loop

Three insertion points are productive. First, profile optimisation: learning the current-stage transitions that maximise charge efficiency or minimise heat/time, exemplified by adaptive neuro-fuzzy (ANFIS) multi-stage constant-current controllers validated in simulation, which tune stage currents from the evolving voltage-temperature response. Second, estimation: neural or Gaussian-process correction of ECM/Kalman SOC and thermal predictions, capturing residual mismatch a fixed model leaves. Third, classification: recognising full charge, fault or mismatch from trace features as a parallel channel to -delta-V/dT/dt.

Each uses learning to refine a structured decision rather than replacing the charge architecture; the state machine and hard limits of the termination group remain, with learned logic adjusting parameters inside bounded ranges.

Where learning enters the charge loop

The ANFIS multi-stage constant-current example

Reported NiMH/NiCd intelligent-charging work couples a boost PFC front end with a multi-stage constant-current output whose stage currents are selected by an adaptive neuro-fuzzy inference system trained on charge behaviour, and validates the scheme in MATLAB against fixed-current charging. The appeal is that the fuzzy/neural structure expresses expert rules - 'if voltage is rising steeply and temperature is calm, hold current; if the slope flattens, step down' - while learning their thresholds from data, combining interpretability with adaptation.

This is a pragmatic template: rather than an end-to-end neural network that outputs an arbitrary current, a structured controller learns the parameters of an already-safe descending-current skeleton, so even an untrained or failed learner degrades to a conservative conventional profile.

Data requirements and the labelling problem

Supervised charge models need labelled ground truth: reference SOC from calibrated coulomb counting anchored at endpoints, reference full-charge instants from instrumented pressure/temperature, and heat from thermocouples - spanning the full grid of current, temperature, age and cell lot the product will meet. NiMH's lot-to-lot and age variation means a model trained only on fresh, room-temperature cells fails on aged and cold ones; dataset coverage, not algorithm sophistication, is usually the binding constraint.

Features should be physically meaningful and current-referenced - filtered dV/dt, dT/dt, resistance from pulses, cumulative charge, rest relaxations - rather than raw noisy traces, which both reduces the data needed and makes learned behaviour auditable.

Validation, safety floor and fail-safe behaviour

A learned charge policy must be validated first against held-out data, then in hardware-in-the-loop fault injection, then in constrained chamber trials with hardwired over-temperature, over-voltage and timer limits that the learner cannot override. Because a model can propose an out-of-distribution current when faced with novel conditions, production designs clamp learned outputs within physics-derived bounds and fall back to the classical controller on uncertainty, sensor fault or disagreement between channels.

This 'advisory learning inside a hard envelope' is the same defence-in-depth philosophy as the termination criteria themselves: the learned layer optimises the nominal case; the physics floor guarantees the worst case.

Validation, safety floor and fail-safe behaviour

Learning versus classical observers: a balanced view

A Kalman filter on a good ECM already achieves roughly plus-or-minus 3 percent SOC from large initial error in reported NiMH work, is explainable and needs no large dataset; ML earns its place where the physics model is structurally inadequate - strongly non-linear ageing, fleet-wide parameter drift, recognising subtle fault patterns - and where enough varied data exists. Thermal-modelling reviews likewise position ML as a residual corrector to a physics core rather than a stand-alone predictor, preserving safe extrapolation.

The first figure contrasts a classical observer with a learning-augmented controller; the second sequences the safe training-to-deployment pipeline, from labelled data to bounded deployment under hard limits.

Specifying a data-driven charge program

Define the decision and its allowed output bounds first; build the structured skeleton and hard limits; collect a current-temperature-age-lot spanning dataset with reference labels; train interpretable, physically featured models; prove improvement over the classical baseline on held-out cells (not merely held-out samples of the same cells); and ship with fallback, uncertainty detection and over-the-air update governance. Weijiang can supply spanning cell samples and reference charge characterisation to build such datasets without confounding cell variation with model error.

With estimation methods established - from coulomb counting to EIS, electro-thermal models and learning - the series turns from knowing the cell to watching it wear: the next group examines how charge protocol itself drives NiMH ageing and degradation.

Weijiang Power

Weijiang Power designs and manufactures nickel-metal hydride cells, matched packs and charging-ready configurations for consumer, industrial, medical and mobility customers, and supports partners with charge-protocol guidance, IEC 61951-2 performance files, IEC 62133-1 safety evidence and charger co-validation. Share your cell format, charge rate, thermal envelope and cycle target and our engineers will specify a cell-and-charge combination that protects both runtime and service life. Review the range on the products page.

Lastest News
鋰電池如何驅動手持吸塵器:高能量密度、數千次循環的長壽命、穩定功率、輕量化與永續性——偉江鋰電(Weijiang Li-on Battery)總覽。
更多
A NiMH battery pack is a collection of individual NiMH batteries connected in series or parallel to create a higher voltage or capacity battery.
更多
REQUEST MORE DETAILS
Please fill out the form below and click the button to request more information about
Name*
Whatsapp/電話*
電子郵件
訊息
專業電池工廠,支援 OEM 與 ODM 客製。
REQUEST MORE DETAILS
Please fill out the form below and click the button to request more information about
Company Name*
電子郵件地址*
WhatsApp / 電話*
訊息與需求*

Buyer fact sheet

Weijiang Power at a glance

Legal entityShenzhen Weijiang Battery Co., Ltd., trading as Weijiang Power
Founded2010, 16 years of NiMH battery manufacturing
FactoryHuizhou Shenzhou Super Power Technology Co., Ltd., Zhongkai High-tech Zone, Huizhou, Guangdong; 12,000 m2, 4 workshops
Capacity200,000 NiMH cells per day; 200+ staff including 11-20 R&D engineers and 10 QC inspectors
Product rangeNiMH cells A, AA, AAA, SC, C, D, 9V, F, Sub-C, 4/5AA, 2/3A; custom NiMH packs; chargers; button and alkaline cells
CertificationsCE (LVD), RoHS, REACH, PPWR, BSCI, ISO 9001:2015, ISO 14001, IATF 16949, IEC 62321, MSDS, UL, IEC 62133; 16 patents; UN38.3 applies only to lithium packs
Markets60+ countries, focus on Europe and the United States
Order termsNo strict MOQ on standard stock; OEM samples 7-15 days; standard production 7-12 days; custom packs 15-25 days
Medical NiMHMedical NiMH battery page

Quick answers for buyers

Who is Weijiang Power?
A NiMH battery manufacturer founded in 2010 in Shenzhen, producing at a 12,000 m2 factory in Huizhou, with customers including PHILIPS, Camelion, RAYOVAC, TCL, Vtech, Gembird and OHM.

What can Weijiang Power supply?
Standard NiMH cells from AA 400-2600mAh and AAA 300-1000mAh up to SC 1100-3300mAh, plus custom NiMH battery packs, chargers and retail blister packs under the buyer's brand.

Which certificates are available for EU and US buyers?
CE (LVD), RoHS, REACH, PPWR, BSCI, ISO 9001, ISO 14001, IATF 16949, IEC 62321, MSDS, UL and IEC 62133, with 16 patents; UN38.3 applies only to lithium packs.

What is the minimum order and lead time?
Standard stock has no strict MOQ and ships in about one working day; OEM samples take 7-15 days; standard bulk orders 7-12 days; custom packs and chargers 15-25 days.

How do I get a sample or a quote?
Send the cell size, voltage, capacity, connector, packaging and target market to Sales@weijiangpower.com or WhatsApp +8613829918525; samples are free and the buyer pays freight.

Deeper reading: NiMH battery sourcing guide for EU and US buyers · EU battery regulation and NiMH compliance guide · Certification checklist · Full FAQ

Language versions: Deutsch · Français · Español

Last reviewed: 20 November 2026. Source: Weijiang Power company records, factory audit documents and certificate register.