At a time when the world is dynamically moving toward the development of sustainable solutions while focusing on innovation, various industries, including automotive, are implementing technologies that will help shape their future. With electric vehicles (EVs) playing a critical role in decarbonization and reducing CO2 emissions, several automakers are investing heavily in the adoption of technologies such as plug-in hybrid electric vehicles (PHEVs), battery electric vehicles (BEVs), fuel cell electric vehicles (FCEVs), and solar electric vehicles (SEVs). The introduction of additional hardware and software components in a vehicle/machine increases the complexity manifold. This underlines the need for improved operation, reliability, safety, and availability of EVs with real-time monitoring, fault diagnosis, fault prediction, and health management.

Prognostics and Health Management (PHM) has become an important tool for monitoring and maintaining the health of electric vehicles. By shifting the focus from pure diagnostics to prognostics, PHM enables early detection and prediction of potential faults or failures, thereby increasing safety and minimizing downtime. Let us explore the importance of prognostics in electric vehicles and how predictions are made to ensure safe operation.

From diagnostics to prognostics

Data in conventional vehicles with internal combustion engines are usually extracted according to industry standards from an OBD (onboard diagnostics)port. However, since most electric vehicles do not have OBD ports (except for hybrid powertrains) and rely primarily on the battery, the incidence of unknown variables is greatest. Electric vehicles include a motor system, inverter, integrated charger, DC-DC converter, electronic control systems, etc., which are also safety-critical systems and are monitored for performance, efficiency, safety, and longevity.

Traditional diagnostic techniques in electric vehicles involve identifying and analyzing existing faults or failures in real time. While diagnostic techniques are important to identify immediate problems, they are not sufficient to prevent unforeseen failures. This is where prognostics comes into play. Prognostics in electric vehicles is critical for monitoring battery health, predictive maintenance, and safe operation of electrical systems. Due to the complexity of EV systems, the importance of battery health, predictive maintenance, data availability, and safety aspects, prognostics is of particular importance in the context of electric vehicles.

Why prognostics?

Prognostics offers several advantages over conventional diagnostics in terms of EV safety:

1. Early fault detection: Prognostics enable the identification of early warning signs of potential failures, allowing for timely intervention before catastrophic events occur. This proactive approach enhances safety by minimizing the risk of accidents or breakdowns on the road.
2. Predictive maintenance: By predicting failures, prognostics facilitates proactive maintenance scheduling. This ensures that necessary repairs or replacements are carried out before the failure actually occurs, reducing the likelihood of unexpected breakdowns and maximizing vehicle availability.
3. Cost savings: Early fault detection and predictive maintenance not only improve safety but also result in cost savings. By addressing issues before they escalate, EV owners can avoid expensive repairs and replacements, ultimately reducing operational costs.

How is the prediction made?

The prediction process in prognostics involves analyzing data collected from various sensors and systems in an EV. This data includes parameters such as temperature, voltage, current, vibration, and other relevant variables. The following steps outline how predictions are made using prognostics:

1. Data acquisition: Sensor data from different EV components, such as battery, motor, and power electronics, is continuously collected and monitored in realtime. Advanced sensing technologies and onboard monitoring systems play a vital role in capturing this data accurately.
2. Data pre-processing: The collected data is processed and filtered to remove noise, anomalies, and irrelevant information. Data cleaning techniques and algorithms are applied to ensure high-quality data for accurate predictions.
3. Feature extraction: Relevant features are extracted from the pre-processed data to characterize the health condition of the EV. These features capture the behavior and performance patterns of the components under normal and abnormal operating conditions.
4. Model development: Predictive models, such as machine learning algorithms, statistical techniques, or physics-based models, are built using the extracted features. These models learn from historical data to establish relationships between the features and potential failures.
5. Prognostic analytics: The developed models are used to analyze real-time data and make predictions about the remaining useful life (RUL) of EV components. RUL estimation provides valuable insights into the future health and performance of the vehicle, enabling proactive maintenance planning.
6. Decision support: Based on the predictions, maintenance actions can be prioritized and scheduled. Alerts or notifications can be generated to inform vehicle owners, fleet managers, or service centers about impending failures and recommended actions.

Prognostics and health management play a critical role in improving electric vehicle safety and reliability. By shifting the focus from diagnostics to prognostics, early fault detection and predictive maintenance become possible. By analyzing real-time data, prognostics enables accurate prediction of potential failures, facilitating proactive maintenance and minimizing risks on the road. Further development of PHM holds immense potential to ensure safer operation of electric vehicles in the future.

About the author

Name: Shyamlendu Panda

Designation: Senior Director & Industry Offer Head -Automotive & Mobility (Technology Group)

Shyamalendu Panda is Industry Offer Head – Automotive & Mobility at Cyient. He has 19 years of experience in consultative technology selling of Engineering solutions for customers across multiple industries. In his current role, he is responsible for building Automotive industry specific Go-to Market offerings, leveraging Embedded and Digital technologies. Prior to Cyient, he worked with Wipro, Capgemini, Huawei &Webex-Cisco.

If you have an interesting Article / Report/case study to share, please get in touch with us at editors@roymediative.com/ roy@roymediative.com, 9811346846/9625243429.