Battery testing for electric vehicles (EVs) has evolved from simple charging and discharging tests to advanced systems that simulate how EVs drive on roads. Recent testing equipment monitors important electrical values, such as state-of-charge (SOC), state-of-health (SOH), voltage, current, and temperature. It also uses artificial intelligence (AI) to predict when batteries may start to fail.
This first part of this article focuses on the electrical parameters of EV battery testing and highlights a few recent developments. The next one will cover the mechanical aspects of EV battery testing.
Where does electrical performance testing begin?
EV manufacturers need testing systems that copy real-world driving conditions and stress scenarios. Modern integrated battery testing platforms solve this problem by combining multiple testing functions into one system.
The testing workflow shown in Figure 1 shows how these platforms perform fault injection testing, timing control measurements, and load reduction commands while watching battery performance. The integrated testbed operates in conjunction with auxiliary battery cycling equipment to simulate real-world vehicle conditions.
Figure 1. An EV battery testing system showing the different integrated components for electrical performance validation. (Image: Quantel)
The testbed encompasses everything from key-on detection to complete charge and discharge cycles. Unlike traditional testing equipment that operates independently, these systems integrate with vehicle models and standard driving cycles. The integrated approach helps engineers identify potential failures before costly vehicle-level testing begins. This reduces development time and improves battery reliability in production EVs.
What are the key electrical parameters in EV battery testing?
Battery testing systems monitor multiple electrical parameters to check performance and safety. SOC shows how much battery capacity is left, while SOH measures how well the battery can store energy compared to when it was new. Voltage measurements track both individual cell voltages and overall pack voltage during charging and discharging cycles.
Current monitoring captures charging rates, maximum charging current, and discharge patterns. Temperature sensors measure the temperature ranges of the battery pack and detect overheating conditions that could lead to thermal runaway. The comparison table in Figure 2, between different manufacturers illustrates how various testing systems prioritize different parameters.
Some manufacturers focus on battery key indicators and power display SOC deviation, while others provide more detailed test analysis results covering basic parameters through performance and safety metrics.
Figure 2. A comparison of battery testing parameters implemented by different EV manufacturers for safety and performance validation. (Image: Energy Reports, Elsevier)
Additional parameters include insulation resistance testing, BMS communication accuracy, and power display deviation measurements. These parameters work together to provide a comprehensive picture of the battery’s condition. Testing systems compare measured values against manufacturer specifications and find batteries that need maintenance or replacement before failure occurs.
How are depth-of-discharge (DOD) and temperature important in EV battery testing?
Battery capacity testing shows important electrical parameters across different usage scenarios. Electrical testing parameters include SOH thresholds, functional capacity limits, and maximum DOD measurements.
Figure 3 consists of several capacity usage charts that show how electrical energy demands change significantly between different driving patterns and battery sizes. For smaller batteries, such as 16 to 24 kWh, the average DOD ranges from 30 to 60%, while larger 90-kWh batteries exhibit a significant amount of unused capacity.
Figure 3. Battery utilization patterns showing average, maximum, and unused capacity across six battery sizes. (Image: Elsevier)
The analysis reveals that most battery setups maintain electrical performance well below the traditional 80% state of health threshold. Temperature effects increase electrical energy consumption by about 29%, which means bigger batteries are needed for cold climate operations.
These electrical measurements help with proper battery sizing, stopping both the underuse of large batteries and insufficient capacity in smaller ones. The data helps improve electrical system design for different vehicle applications and environmental conditions.
Temperature alone has a significant impact on how batteries perform during tests. The resistance and capacity retention in Figure 4 clearly demonstrate this. When temperatures drop, the direct current resistance increases significantly. The measurements show that resistance values almost double when the temperature drops from 55° to 25° C across different SOC levels. This means that temperature changes affect the accuracy of the tests and our understanding of the battery.
Figure 4. Lithium-ion battery performance degradation under different temperature conditions. (Image: Energies, MDPI)
The capacity retention chart in Figure 4 indicates that batteries degrade more rapidly at high temperatures. Batteries tested at 45° C lose their capacity quickly after 10,000 cycles. However, batteries tested at 25° C continue to function well for over 40,000 cycles. These results enable us to determine the optimal testing conditions and demonstrate that battery thermal specifications are effective in real-world EV applications.
How is AI relevant in EV battery testing?
Modern testing platforms are starting to use AI to predict battery health and remaining useful life. Machine learning models analyze electrical data from charge and discharge cycles to identify patterns that indicate battery degradation. The architectural model depicted in the neural network diagram in Figure 5 illustrates how AI systems simultaneously analyze multiple electrical parameters.
Figure 5. A model architecture for battery state of health and remaining useful life prediction. (Image: Springer Nature)
These models use voltage, current, temperature, and capacity measurements as input data. Convolutional neural networks extract features from the electrical signals, while bidirectional LSTM networks handle temporal relationships within the data sequences. The attention mechanism weighs the most important electrical characteristics for accurate predictions.
Different AI approaches exhibit varying levels of success in electrical performance analysis. The performance comparison chart in Figure 6(a) shows that a hybrid AI-based BMS (HAI-BMS) outperforms other conventional methods. These AI systems handle electrical measurements differently during testing cycles. HAI-BMS combines multiple AI approaches to achieve better electrical parameter analysis compared to single-method systems.
Figure 6. The performance comparison of AI-based BMS across varying dataset sizes. (Image: International Journal of Low-Carbon Technologies)
The chart in Figure 6(b) shows that steady performance gets better with larger datasets. HAI-BMS keeps stable accuracy across all testing situations, making it work well for different types of batteries and operating conditions. This steady performance is necessary to verify electrical safety thresholds and predict battery degradation patterns in real-world applications.
Summary
Recent progress in EV battery testing systems has identified many more electrical parameters beyond just charging and discharging cycles. It is not surprising that the integration of AI will gain momentum, even though AI has already been utilized in EV battery testing for some time.
A few of the key parameters to look out for during testing include SOC, SOH, DOD, and temperature. It will be interesting to see any new parameters emerge with the use of AI.
References
- Implementation of artificial intelligence techniques in electric vehicles for battery management system, International Journal of Low-Carbon Technologies
- Estimation of electric vehicle battery capacity requirements based on synthetic cycles, Transportation Research Part D: Transport and Environment, Elsevier
- Application of state of health estimation and remaining useful life prediction for lithium-ion batteries based on AT-CNN-BiLSTM, Springer Nature
- Overview of EV battery testing and evaluation of EES systems located in EV charging station with PV, Energy Reports, Elsevier
- Lithium Battery Degradation and Failure Mechanisms: A State-of-the-Art Review, Energies, MDPI
- New Chroma Battery Pack Integrated Testbed for EV Composite Operation Conditions, Quantel
Related EE World Online Content
- What data acquisition and test setups are needed for EV battery testing?
- Ensuring EV battery safety with advanced temperature monitoring
- How can simulation improve EV battery thermal design?
- Battery testing: critical to the rise of electric vehicles
- The fundamentals of testing cycling battery cells
- What is battery cell formation and testing?
Filed Under: Batteries, FAQs, Testing and Safety