BMS designs for EVs reach cloud, adopt digital twins

Article By : Majeed Ahmad

Semiconductor suppliers are joining hands with automotive software specialists to create innovative new features for BMS architectures.

The battery is the costliest element in an electric vehicle (EV), making battery management system (BMS) a key design recipe in vehicle electrification. Not surprisingly, therefore, BMS requirements are growing as EVs go mainstream. So, in a bid to improve battery performance and enable new applications, semiconductor suppliers are joining hands with automotive software specialists to create innovative new features for BMS designs.

NXP Semiconductors, for instance, is joining forces with Electra Vehicles, a specialist in artificial intelligence (AI)-based battery controls, to utilize digital twin models in the cloud to better predict and control the physical BMS in real-time. NXP will utilize Electra’s EVE-Ai 360 adaptive control technology to connect its high-voltage BMS (HVBMS) through its S32G GoldBox vehicle networking reference design to the cloud to leverage AI-powered battery digital twins.

Figure 1 Digital twins in cloud can improve battery state of health by up to 12% and enable multiple new applications. Source: NXP

Electra’s AI-enabled battery-adaptive digital twin technology powers EVE-Ai architecture to process the data to identify cycles and extract features both at the battery and vehicle levels. Then, the adaptive cell modeling system dynamically selects the most appropriate model for a specific usage profile. In short, it connects digital twins seamlessly to the cloud to ensure access to accurate sensor data, real-time closed-loop control of the BMS, high-performance in-vehicle processing, and secure connectivity for services and over-the-air (OTA) updates.

BMS in the cloud

Andreas Schlapka, director and segment manager for Battery Management Systems at NXP Semiconductors, says that the company has addressed the two main challenges associated with the digital-twin approach by integrating Electra’s EVE-Ai architecture. “These are coping with the abundance of data from our electrification solutions, which requires cleansing and appropriate feature selection, and the variance of use cases, which requires model selection and adaptive training.”

NXP’s hardware integrating Electra’s AI-powered battery-adaptive digital twin solution encompasses the S32K3 HVBMS reference design and S32G GoldBox vehicle networking reference design.

The S32K3 HVBMS reference design facilitates precise measurements of the battery’s state of health (SoH) and state of charge (SoC) to leverage the full potential of the battery and thus maximize driving range with accurate diagnostics. Next, the S32G GoldBox, a vehicle networking processing solution, provides high-performance computing capacity and real-time network performance with secure cloud connectivity for data-driven cloud-based automotive services.

As a result, the digital-twins-enabled BMS solution helps adapt to ongoing changes in battery health due to operating conditions and provides updated figures back to the BMS for continuously improving control decisions. So, fleet operators can employ usage insights such as vehicle charging times for battery predictive diagnostics. Likewise, battery care centers can use this information to reduce downtime with rapid diagnostics and EV charging station operators can optimize charging service and energy efficiency.

NXP will demonstrate this cloud-connected BMS solution at the electronica fair in Munich to be held from 15 to 18 November 2022.

Figure 2 BMS architectures are increasingly demanding a software-centric approach for abstracting communication and control interactions between the BMS microcontroller and the battery cell controllers. Source: NXP

Acknowledging the shift toward software, NXP is also joining hands with embedded software supplier Elektrobit to reduce the entry-level cost of BMS development and enhance ease-of-use of its HVBMS reference designs. The Eindhoven, Netherlands-based chip company will co-develop a software platform that supports NXP’s HVBMS RD reference design using Elektrobit’s Classic AUTOSAR tooling and software.

NXP’s HVBMS RD reference design comprises three modules: battery management unit (BMU), cell monitoring unit (CMU), and battery junction box (BJB). The reference design will integrate application software and complex device drivers (CDDs) by utilizing Elektrobit’s EB tresos software platform.

BMS system solutions

Meanwhile, BMS ICs continue to hone EV battery designs, especially for high voltages like 400 V or 800 V. That leads to more sophisticated BMS architectures that enable unique application features. Take, for example, Infineon’s new battery management ICs that aim to create system-level solutions for cell-to-pack and cell-to-car battery topologies.

Figure 3 The BMS chips facilitate a system-level solution for battery modules. Source: Infineon

Chinese automaker NETA Auto is adopting Infineon’s BMS solution in collaboration with B&Z, which handles system design and development, testing and verification, and mass production and delivery. Infineon’s BMS solution comprises the Li-ion battery monitoring and balancing chip TLE9012DQU, the iso UART transceiver chip TLE9015DQU, and AURIX microcontrollers for voltage and temperature sensing as well as balancing and communication in BMS designs.

The Li-ion battery monitoring and balancing chip TLE9012DQU performs highly accurate voltage measurements for SoC and SoH calculations: key requirements for BMS designs. It facilitates noise rejection through internal oversampling as well as advanced sync-on-sample at less than 10 µs across an entire battery pack.

The battery monitoring and balancing system IC also enables high accuracy through 16-bit analog-to-digital converter (ADC) that allows simultaneous accurate cell voltage sampling. Here, each ADC is equipped with a digital low-pass filter down to 35 Hz to minimize peripheral circuit design and reduce the cost of the BMS.

The independent ADC channel of each cell significantly reduces the measurement time of the entire battery string and optimizes the synchronization of voltage and current sampling. That, in turn, facilitates highly accurate SoC and SoH estimations as well as other condition predictions for EV battery pack safety.

The ASIL-D compliant monitoring and balancing IC designed for lithium-ion battery packs can be controlled directly by a microcontroller via UART or an isolated UART interface with the new TLE9015DQU iso-UART transceiver IC (Figure 3). TLE9015DQU—internally equipped with a bidirectional daisy-chain signal transceiver function—sends acquired data in both directions to ensure that each analog front-end (AFE) draws the same amount of power during operation. That equalizes power consumption and reduces the risk of unbalanced cell voltage.

BMS beyond EVs

It’s important to note that secondhand batteries that have reached the end of their automotive life still have a significant residual capacity of up to 80%. So, they can be tapped in energy storage systems (ESS) for homes to reduce homeowners’ energy bills. That’s why BMS designs are enabling new features such as continuous battery monitoring to support battery recycling and reuse of EV batteries in energy storage systems to support the circular economy.

The software-first approach in BMS designs and the availability of features like digital twins will go a long way in preparing EV batteries for new applications and markets.


This article was originally published on EDN.

Majeed Ahmad, Editor-in-Chief of EDN and Planet Analog, has covered the electronics design industry for more than two decades.


Leave a comment