Although there is enthusiasm from a wide range of stakeholders for the take-up of low carbon vehicles there is an ongoing need to provide an evidence base for wider public acceptance of new technologies through vehicle deployment trials.

The approach adopted by Cenex and detailed here follows technology from the laboratory to the test track through to real-world fleet deployment to yield a uniquely comprehensive evidence base for low carbon vehicle performance.  As part of these activities Cenex held a seminar in conjunction with the Millbrook Proving Ground on Electric Vehicle Studies.

The event addressed four aspects of testing and operation that influence electric vehicle performance:

  • Driver style and duty
  • EV range testing
  • Traffic flow
  • Drive cycle standards and issues

The event presentations are summarised below, with full presentations available for download by clicking here.

1.    Driver style and duty

One of the advantages of hybrid and battery electric vehicles compared to conventional vehicles is the ability to capture braking energy through ‘regenerative braking’ whereby the electric motor operates as a generator, turning the rotational motion into electrical energy in order to charge the battery.  Few studies have yet been published as to the effect of this behaviour on vehicle energy consumption in test or real-world conditions.

In order to help address this issue Steve Carroll of Cenex presented a study of six drivers using the first generation smart ED (as used in the Smart Move trial) on a specially designed EV test cycle at the Millbrook training ground.

The 11.8 km cycle includes:

  • A high speed circuit
  • City course
  • Hill route
  • And handling circuit (designed to simulate a UK B road with free flowing traffic)


The results presented showed that:

  • The most inter-driver variation occurs where numerous deceleration events dominate the drive cycle
  • For different driving styles the range in motoring efficiency (measured as the transfer of energy from the vehicle battery to the vehicle motor) was low, at 80-84%, compared to a 15-93% range in deceleration efficiency.  Lower efficiency decelerations were due to high blends of friction braking being used
  • When comparing a diesel smart CDi and the smart ED, both vehicles were least efficient over the City track circuit.  The most efficient application for the smart CDi was on the high speed circuit, whereas the smart ed was most efficient on the lower steady speeds representative of UK B roads.  During positive energy transfers the EV efficiency ranged from 70-90% (excluding energy storage and electrical conversion losses) compared with a lower and wider range of 5 to 33% for the smart CDi
  • Average regeneration rates of 16% were seen during test track studies compared to 11.3% in real world trials, reflecting the less predictable deceleration scenarios of public roads
  • The study highlighted the advantages of driver training for regular EV users as an eco driving instructor at Millbrook achieved an average of 87 % more energy regeneration over the City circuit compared to the least efficient driver

2.    EV range testing

In order to further elaborate on the effects of the transition from the laboratory to the road on vehicle performance, Chris Walsh of Cenex presented a comparison of two EVs – the second generation smart ed and the Mitsubishi i MiEV – based on dynamometer, test track and real world testing.

Extrapolated range over real world drive cycles

A charts showing extrapolated range over real world drive cycles.

Laboratory testing shows:

  • A step performance increase from the first to the second generation smart ed
  • Considerably more range from the smart over the i MiEV on all cycles
  • Higher regeneration from the i MiEV, plus the effect of different gearbox modes
  • High speed cycles severely limit the range of both vehicles, but speed limiting on the smart to 60 mph alleviates this effect

Track cycles show:

  • Good correlation between lab and track results for real world cycles
  • i MiEV work shows that increasing the vehicle regeneration rate is more effective in extending the range of an EV than limiting its power consumption

For the real world drive events:

  • Data backs up the conclusions of the lab and track work
  • Demonstrates typical driver behaviour in EVs with acceleration and deceleration rates that far exceed the NEDC cycle

Heating/cooling effects:

  • Hotel load levels demonstrate typical use range reduction of 10% and a possible worst case of 33%

3.    Traffic flow

The effect of congestion on the efficiency of an EV was discussed by Dr Graeme Hill of the University of Newcastle.  His presentation of the preliminary results of the performance of smart ED vehicles travelling around the Metro Centre in Gateshead revealed no observable difference in vehicle efficiency and power usage with increased levels of congestion.  This was explained by the improved efficiency of the EV owing to the greater use of regenerative braking in increased congestion.

4.    Drive cycle standards and issues

The applicability of the New European Drive Cycle (NEDC) to the real-world driving conditions that many of us face, particularly congestion and higher speeds needed on motorways, is under debate.  As revealed in the previous presentations, adaption of driver behaviour of the characteristics of electric vehicles, such as regenerative braking, is crucial to optimising their efficiency, and therefore to maximising their environmental benefit.  This is hard to capture by measuring vehicles over the NEDC.

Andy Eastlake, Head of Laboratories at Millbrook, presented a thought-provoking discussion of the issues raised by the electrification of vehicles with regards to vehicle testing and standards.  Amongst his conclusions on the testing and comparison of electric and conventionally-fuelled vehicles were:

  • The public needs better and consistent information on the comparative performance of low carbon and conventional vehicles
  • In looking at real-world performance of electric vehicles energy pricing assumptions will cause an influence and COand energy generation factors are important
  • Overall, comparisons to conventional fuels are complex and the NEDC Drive Cycle may be inappropriate as fundamental energy efficiency benefits may be hidden