Prototype to production: accelerating electric vehicle development

With software defining the next generation of cars, engineers must master the exploding complexity of software and ensure that its free of bugs and runtime errors.

Prototype to production: accelerating electric vehicle development

By R Vijayalayan

The market for electric vehicles in India is expected to cross annual sales of 17 million by 2030, growing at a CAGR of 49% as per a report by the India Energy Storage Alliance (IESA). Several factors, such as a growing emphasis on sustainability and rising fuel prices, are driving the popularity of EVs. In addition, favourable government policies provide more tailwinds to further accelerate EVs’ popularity.

While this is an exciting time to be a part of the EV industry, there are two fundamental challenges that all ecosystem participants must address in order to effectively meet the demands of a rapidly growing market.

Given that the EV industry is not yet mature, it’s important for auto engineers to understand that vehicle designs are still evolving. Therefore, taking designs from the prototype to the production stage quickly is often challenging. For instance, the engineering team must evaluate design concepts for various riding and usage scenarios to make informed trade-off decisions that optimise their design parameters. For example, while increasing battery capacity can improve range, this decision would also increase the cost and size of the EV.

Then there is the time and resource constraints of many EV makers, building a large number of physical prototypes is not practically feasible and financially advisable. In such instances, simulation and model-based development is key to quickly delivering a viable product while de-risking the development process and reducing costs. Simulation can bring down the development cycles from months to weeks. Model-Based Design can also help accelerate testing and allow for the mitigation and resolution of field issues via software updates.  The systematic reuse of models is a basic principle of Model-Based Design, where models form a digital thread connecting development, design optimisation, code generation, and verification and validation

Here are some of the top areas to consider at each stage of the EV development cycle as you develop the prototype, test and get into production.

Is your EV system design optimised for range and thermal efficiency?

Range anxiety is one of the biggest deterrents to buying an electric vehicle. As per the JD Power 2022 US EV Owners’ Satisfaction Ranking, range satisfaction is a key purchase reason for electric vehicles. While this is a global survey, the sentiment holds true in India, too.

Given the increasingly competitive market, building vehicle prototypes or simulation models from scratch to optimise the vehicle range is simply not a viable approach.  This is where using pre-built, customisable, and open vehicle simulation models to optimise the vehicle range plays a critical role. Using pre-built models not only reduces the need to build vehicle prototypes, but also helps optimise the range on an existing drivetrain as Tesla has proven. The team at Tesla used MATLAB models to identify losses due to inefficiencies and were able to continually tweak the hardware to increase efficiency. Interestingly, this process improved motor efficiency by 8-10% and improved the range by 15-18%.

Also, given the multiple instances of EV battery fires, thermal management of batteries has emerged as a key focus area. The government of India’s Union Ministry of Road Transport and Highway recently announced additional safety requirements for battery safety that covered battery cells, battery management systems, the design of battery packs, and unabated thermal propagation, also known as thermal runaway.

A Model-based design approach can help study the fidelity of battery models at different stages of development, simulate battery management systems, and understand thermal behaviour in different environments. For instance, even a ready-made battery pack will behave differently in Indian weather conditions as compared to colder climates. Therefore, the ability to simulate local conditions is critical.

Simulation can also help analyse trade-offs, predict battery performance, study the impact of temperature on the range of the vehicle, and put in place design controls without hardware.

How do we design for safety and compliance?

Functional safety spans both system and software safety. It is about ensuring designs meet safety standards and workflow compliance. Using an ISO-26262-compliant product development process and qualified toolchain can help reduce development time by using an automated application and a specific verification workflow. Applying best practices such as developing back-to-back models and code testing is also important.

With software defining the next generation of cars, engineers must master the exploding complexity of software and ensure that its free of bugs and runtime errors. Integrating the system software and data by bringing together people, processes, and tools is necessary for the successful transition from prototyping to production. A model-based design approach lends itself perfectly to this process, bringing in operational aspects, such as predictive maintenance, to build a continuous cycle.

How do we use AI to enhance driver experience and vehicle performance?

EV companies are focused on making their vehicles and their systems more connected and smarter. We see organisations addressing these challenges through two main methods.

First, many businesses committed themselves to simply, ‘do things better’ Detecting anomalies quickly, scheduling maintenance before the system fails, or optimising the operation of a fleet’s systems are possible once the system is connected and improves customer experience. The second strategy is to ‘do new things.’ This means that EV makers transform their organizations with new business models and revenue opportunities. For many, these are strategic shifts towards digital transformation. 

The few innovation trends that we are observing with respect to EV technologies globally and locally are:

• AI-based sensor models:  As accurate SoC is critical to effective operation, AI is being explored as a method to generate accurate SoC prediction in those cases where traditional techniques have limitations.

• IoT for remote monitoring: With the need for remote performance monitoring for batteries, we would be seeing more technologies like IoT being leveraged to make them happen.

• Digital Twins: A digital twin provides access to verified simulation models across cloud platforms, hence supporting collaboration and rapid innovation across multiple engineering teams.

While data is available in abundance, using it optimally requires us to bridge the knowledge gap between domain engineers and data scientists to ensure data does not remain siloed. Model-Based Design at the system level spans the entire process right from data preparation, AI modelling, simulation and testing, and deployment. This encompasses the complete AI workflow to support use cases such as battery analytics, remaining useful life (RUL), digital twins, and predictive maintenance.

How do we make the workforce future-ready?

While there is a large pool of professionals trained in conventional automotive technologies, there is a pressing need to reskill and upskill the existing workforce and bring in system-level thinking as the industry moves from Conventional powertrain to Electrified powertrains. In order to achieve this, there’s a need for the workforce to develop and expand their EV development skills in motor control, battery management, fuel cells, electrical systems, and system simulation. 

Industry-academia collaboration is a proven method for building industry-specific expertise among entry-level employees.

In addition, the Indian government is working with organisations like SAE, ICAT, ARAI, and private companies to build the right skill sets within our region. India is home to a robust EV ecosystem that includes not just domestic EV OEMs, academia, and government, but also deep software and technology maturity through software companies and GICs. At the same time, government policies are incentivising the shift towards manufacturing of batteries with focus on large-scale battery packs batteries and in-house development of battery management systems. There is also a push on bringing motor control in-house, not only to benefit from subsidies but also to build competitive differentiation.

To summarise, electrified vehicle development requires one to address many challenges in parallel, such as total vehicle efficiency, new engineering capability – power electronics, high-voltage battery, motors, embedded software quality, and transition from prototyping work style to production mindset and processes. Model-Based Design involving the systematic use of models and data and getting additional efficiencies through automation enables engineers to prototype and productionise software with well-defined processes – reducing the time, cost and dramatically improving the quality of software and hence the electric vehicle.

The author is Manager, Automotive industry and control design vertical application engineering teams, MathWorks India.

Disclaimer: Views expressed are personal and do not reflect the official position or policy of Financial Express Online. Reproducing this content without permission is prohibited).

Get live Share Market updates and latest India News and business news on Financial Express. Download Financial Express App for latest business news.

First published on: 04-12-2022 at 12:49:44 pm