New Energy Vehicle Software Functional Testing: Standards, Scope, and Methods

New Energy Vehicle (NEV) software functional testing is a systematic verification process that confirms whether the embedded software governing battery management, drive control, charging, and connectivity behaves exactly as specified under all intended operating conditions. As modern electric vehicles depend on more than 100 electronic control units and millions of lines of code, professional functional testing at accredited third-party laboratories helps manufacturers guarantee safety, reliability, and regulatory compliance before mass production. This article explains the technical scope, applicable standards, sample types, test items, and methods involved in NEV software functional testing.

Technical Overview

Modern new energy vehicles are increasingly described as "software-defined vehicles," in which the majority of vehicle functions—from torque allocation and battery protection to over-the-air (OTA) updates and intelligent cockpit experiences—are implemented and continuously evolved through software. A typical electric vehicle contains dozens of electronic control units (ECUs) interconnected via CAN, CAN FD, LIN, and automotive Ethernet networks, collectively running tens of millions of lines of code. Any defect in this software can compromise driving safety, energy safety, or user experience, which makes rigorous functional testing indispensable.

Software functional testing verifies that each software function behaves according to its specified requirements. It is a requirements-driven, predominantly black-box activity that complements structural code testing and is embedded within the broader automotive V-model development process. For safety-related software, testing rigor is scaled to the Automotive Safety Integrity Level (ASIL) defined in ISO 26262 and its Chinese counterpart GB/T 34590. Related frameworks such as ISO/SAE 21434 for cybersecurity engineering, ISO 24089 for software update engineering, ISO 21448 for safety of the intended functionality (SOTIF), and ISO 14229 for unified diagnostic services (UDS) further define the functional behaviors that must be verified.

Accredited third-party laboratories provide independent functional testing services using model-in-the-loop (MIL), software-in-the-loop (SIL), hardware-in-the-loop (HIL), and real-vehicle environments. Independent verification is particularly valuable for regulatory submissions, supplier acceptance, and launch risk reduction, because it eliminates developer bias and provides objective evidence of conformity.

Testing Samples

Battery Management System (BMS) Software

BMS software governs state-of-charge and state-of-health estimation, cell balancing, charge and discharge power limits, insulation monitoring, and high-voltage interlock protection. Functional testing verifies estimation accuracy under dynamic load profiles, correct execution of protection strategies, and reliable fault detection for overvoltage, undervoltage, overcurrent, overtemperature, and communication loss conditions.

Vehicle Control Unit (VCU) and Motor Control Unit (MCU) Software

VCU software coordinates energy management, torque arbitration, gear and mode selection, and regenerative braking blended with friction braking. MCU software executes torque commands, field-oriented motor control, inverter protection, and sensor fault handling. Testing confirms correct torque paths, smooth transitions between driving modes, and fail-safe responses to sensor and actuator faults.

Charging Control and On-Board Charger (OBC) Software

Charging software manages AC charging, DC fast charging, guidance circuit diagnostics, charging pile handshake, and communication protocols. Functional testing covers protocol conformance, charge curve behavior, interruption recovery, and protection logic during abnormal charging events.

Intelligent Cockpit and Infotainment Software

Cockpit software includes instrument cluster rendering, center-stack applications, voice interaction, Bluetooth and smartphone connectivity, and camera-based surround view. Testing verifies display logic, HMI response times, audio and video behavior, connectivity stability, and correct interaction between cockpit and vehicle control domains.

ADAS and Driving Assistance Software

Advanced driver assistance functions such as adaptive cruise control, lane keeping, and automatic emergency braking depend on perception, fusion, planning, and control software. In addition to functional checks, these systems are evaluated against the intended-functionality principles of ISO 21448 to address performance limitations and foreseeable misuse.

OTA Update and Diagnostic Software

Software update modules must correctly download, verify, install, and roll back updates without bricking ECUs or violating safety constraints, following the software update engineering requirements of ISO 24089. Diagnostic software implementing UDS services must correctly execute session control, fault memory read and clear, routine control, and ECU flashing as defined in ISO 14229.

Thermal Management and Body Control Software

Thermal management software coordinates battery, motor, and cabin cooling and heating loops for energy efficiency and cold-weather performance, while body control software manages door locks, lighting, wipers, and alarm logic. Testing confirms control strategy correctness, mode transitions, and fault degradation behavior.

Testing Items

Typical functional testing items for NEV software include:

• Functional requirement verification: systematic testing of every functional requirement with full requirements-to-test-case traceability.
• Interface and signal testing: verification of CAN, CAN FD, LIN, and Ethernet signal timing, byte order, scaling, and invalid-value handling consistent with ISO 11898 and interface specifications.
• State machine testing: validation of power modes, sleep and wake-up behavior, charging states, and driving mode transitions, including illegal transition rejection.
• Diagnostic function testing: verification of UDS services, diagnostic trouble code management, and ECU reprogramming behavior per ISO 14229.
• Fault injection and degradation testing: simulation of sensor failures, actuator faults, network errors, and power interruptions to confirm detection, reaction, and safe-degradation strategies.
• Boundary and extreme condition testing: evaluation of software behavior at the limits of input ranges, temperature ranges, battery charge levels, and communication loads.
• Charging interoperability testing: conformance and compatibility testing of charging communication, including protocol behavior aligned with ISO 15118 and applicable national charging standards.
• HMI and usability verification: checking of display content, alarm priorities, language resources, and user interaction flows.
• OTA update verification: testing of update packaging integrity, installation sequencing, preconditions, power-loss recovery, and rollback in line with ISO 24089.
• Cybersecurity function verification: confirmation that secure boot, access control, secure diagnostics, and intrusion detection functions operate as designed under ISO/SAE 21434.
• Regression testing: re-execution of test suites after software changes to detect unintended functional side effects.
• Resource and stability testing: assessment of CPU load, memory usage, and long-duration endurance behavior to detect leaks and performance drift.

Testing Methods

NEV software functional testing is executed across a staged environment strategy aligned with the V-model. Model-in-the-loop (MIL) testing validates control algorithms against plant models in the simulation environment, allowing early detection of logic errors. Software-in-the-loop (SIL) testing executes compiled production code on a development platform to confirm behavior consistency between algorithm and code. Hardware-in-the-loop (HIL) testing runs the real ECU hardware against real-time simulated vehicle physics, enabling exhaustive and hazardous scenario testing—such as battery thermal events or sensor failures—safely in the laboratory. Processor-in-the-loop (PIL) testing verifies code on the target processor, and vehicle-in-the-loop or real-vehicle testing provides final system-level confirmation under actual operating conditions.

Test design applies recognized techniques including requirements-based analysis, equivalence partitioning, boundary value analysis, state transition coverage, and cause-effect analysis, ensuring systematic coverage of both normal and abnormal scenarios. Fault injection is performed by manipulating signals, bus messages, and sensor outputs to trigger the software's diagnostic and degradation logic. Communication testing uses bus simulation and analysis tools to send, monitor, and disturb network traffic, while diagnostic testing employs protocol testers to exercise UDS services. Automated test frameworks enable high-volume regression execution, and results are documented with requirement coverage matrices, pass/fail statistics, and defect reports that support ISO 26262 verification evidence and safety case documentation.

Applications

NEV software functional testing serves multiple stakeholders throughout the vehicle lifecycle:

• OEM research and development: early functional validation of new platforms, software integration milestones, and release candidate acceptance.
• Tier-1 supplier delivery: independent verification that delivered ECUs and software releases meet customer functional specifications.
• Regulatory and homologation support: objective test evidence demonstrating conformity with applicable national and international functional safety, cybersecurity, and software update requirements.
• Mass production launch: final regression and stability confirmation before start of production, reducing recall and field-campaign risk.
• OTA release management: verification of every software update campaign prior to deployment across vehicle fleets.
• Field issue analysis: reproduction, root cause identification, and fix validation for customer-reported software defects.

FAQ

Why is software functional testing especially important for new energy vehicles?

NEVs concentrate high-voltage energy storage, high-power drive electronics, and deep connectivity in one platform, so a single software defect can simultaneously affect safety, range, charging, and user trust. Functional testing provides objective evidence that protective and control functions work correctly in all specified conditions, which is essential for safety certification and brand reputation.

What is the difference between MIL, SIL, and HIL testing?

MIL tests the control algorithm within the simulation model, SIL tests the implemented software code on a development environment, and HIL tests the real ECU hardware running production software against a real-time simulated vehicle. The three levels progressively increase realism, with HIL offering the best balance of fidelity, safety, and test repeatability before real-vehicle trials.

Which standards govern NEV software functional testing?

The primary frameworks include ISO 26262 and GB/T 34590 for functional safety, ISO/SAE 21434 for cybersecurity engineering, ISO 24089 for software update engineering, ISO 21448 for safety of the intended functionality, ISO 14229 for diagnostics, ISO 11898 for CAN communication, and ISO 15118 for charging communication, together with applicable national vehicle and charging standards.

How long does a typical software functional test campaign take?

Duration depends on scope: a single ECU function test campaign may take several weeks, while a full vehicle-domain program covering BMS, VCU, charging, cockpit, and OTA functions can span several months of iterative testing aligned with software release milestones.

Can testing be performed by independent laboratories?

Yes. Accredited third-party laboratories provide MIL, SIL, HIL, and vehicle-level functional testing with standardized environments, calibrated instrumentation, and traceable reporting, giving manufacturers unbiased evidence for supplier acceptance, regulatory submission, and production release decisions.