Developing sustainable public transport: the electric bus charging project
24 April 2017
Consider a system that uses as little as 20 per cent of the energy of traditional vehicle systems – and which could have an even smaller carbon footprint with the right primary energy source. Consider the removal of health-damaging pollutants from the streets of a city, as well as noise reduction. Add a much smoother journey experience for passengers, with no smell of exhaust fumes, and you have what some leading manufacturers are offering by way of the electric bus. Are benefits like these worth changing the way we operate our bus fleets?
Electric buses already exist, with many trolley systems worldwide. However, cities want to avoid the major investment in inflexible and unsightly overhead wire systems. The latest battery developments offer us the opportunity of gaining the benefits without this.
Just as with the eCar, ‘range anxiety’ is likely to be the first objection: if we cannot get a decent range from a wee mini-car, how are we going to do it with a heavy vehicle? But buses, with their set routes, have a very predictable energy usage, making them particularly suitable for electric operation. In late 2012, Volvo Bus Corporation chose to work with Siemens to deliver an automatic charging system for its planned range of plug-in hybrid and all electric vehicles.
The first question then is around the charging strategy: should the bus be able to operate all day without a charge, or should the opportunity be taken to charge en route? Carrying a charge for maximum duty over the day still means adding a very significant weight to a bus, almost to the state where the energy usage is determined by the unloaded weight of the vehicle. The additional weight also means additional structural weight to carry the batteries and extra axles or wheels, unless we sacrifice passenger capacity to maintain legal axle loads.
The other extreme is charging at almost every stop, sometimes with flash charging. Experience on initial projects indicates that bus passengers are very dissatisfied when buses need to stop even when no passengers board or alight. It also means that we have a very significant additional electrical infrastructure throughout the city.
In order to maintain a reasonably low utility connection charge, this infrastructure will also need the additional cost of storage in order to avoid very high maximum import capacity with very low utilisation. While studies indicate that repeated small charges give a better battery life than medium or overnight charging, this does not appear to outweigh the significant additional infrastructure costs.
Volvo Bus Corporation already has a significant fleet of hybrid electric buses in operation worldwide and has implemented several early plug-in pilot projects. Based on this experience and experience of other electric bus projects, Volvo chose to use opportunity charging at route termini. While the charging time may initially appear to add to the operating cycle time, much of this can be covered by driver break times and part of the buffer time that is usually provided to allow for traffic hold-ups.
We do not want drivers to have to get out of the bus and manhandle a cable capable of carrying 400kW through people waiting at terminus and plug such a large device in safely to the vehicle. We do not want to waste the time involved and it would be a major HR issue in any city. So we need to connect automatically.
Of course, one option is not to ‘connect’ at all, but to use inductive charging, just like your toothbrush. Electrical engineers are generally horrified at this option, since such a ‘loose coupled’ transformer is against everything they have been trying to do for years to reduce losses. This is nominally overcome by using higher frequency systems. There are still significant losses and we require very precise parking of the bus to keep these to reasonable levels. In one system, a camera is located under the vehicle so that the driver can check the accuracy of the parking. This obviously has extra cost and slows down the parking process. However, the major drawback is the weight of equipment needed on the vehicle along with additional screening from the cabin.
With a contact system and DC charging, we can minimise the equipment on the vehicle to essentially the contact system and the communications hardware.
The most common option for connecting vehicles with power supply is the pantograph connection to overhead contacts, familiar from electric trains and trolley buses. In order to minimise weight and complexity on the vehicle, we decided to put the pantograph on the overhead system and to leave just the ‘overhead lines’ on the vehicle. The roof contacts are only connected to the battery system during charging and, being on the roof, are separated by a safe distance from people. While this means additional maintenance on the infrastructure side, the technology is relatively simple and familiar to transport operators worldwide.
However, in order to ensure safe connection to a battery-powered electric road vehicle, we need more than the one or two contacts on the existing systems. In this project, we used the requirements of the (then draft) IEC/EN 61851 standard as a guide. Initially many felt that three contacts (isolated plus and minus together with a protective earth) would be sufficient for an automatically connected system but we felt that the pilot signal to confirm earth connection was necessary for safety under all fault conditions.
The pilot itself, together with the geometry of the connection system should then be sufficient for proximity detection. This means that we required four contacts for a safe power connection.
Using a simple pantograph, we use two rails on the bus roof, each with two contacts and a similar arrangement at 90° on the inverted pantograph. Allowing for roof construction and isolation distances, this gives us a lateral parking tolerance of up to 25% of the bus width and a tolerance in the driver direction depending on the length available on the bus roof.
Allowing also for maintaining contact in a sideways kneeling position and for construction and system tolerances, we end up with a parking zone of about 600mm in the driver direction and up to about 400mm from the kerb. Experience over the last year has confirmed that drivers can comfortably achieve this accuracy of parking.
Where the connection is made mechanically from the vehicle, a similar communication system to that for manually connected charging can be used. However, when the mechanical connection needs to be initiated from the infrastructure, we need an initiating request from the vehicle to the charger to make the connection. This requires a wireless connection. This is similar to the situation for non-contact charging.
We also need to be sure that the vehicle at the charging station is the vehicle requesting connection. This projects deals with this simply by using directional WLAN antennae with limited range on the vehicle and infrastructure, so that the bus can only connect to the wireless point when parked at the charging station. We also wanted to reduce the possibility of communication with a vehicle on the roadway beside the parked bus. On this basis, it was decided to mount the bus antenna on the boarding side of the vehicle with the mast antenna preferably directly above it when in the centre of the permitted parking area.
The V2X standard ISO15118 is used as the basis of the communications protocol. Standard message stacks already developed for this standard were used in order to speed up development. A message structure was developed jointly by Siemens and Volvo to achieve charging and to exchange information between the vehicle and the infrastructure. This has delivered a very speedy and reliable charging communication systems but some additional development of the standard message stacks for this use case would be of benefit.
On the street – charger
A 300kW DC charger is a significant load for the electricity network and a noticeable piece of equipment to place on the street. Early discussions with the distribution network operator (DNO) and with relevant city authorities are important.
In the pilot projects in Stockholm and Gothenburg, the local DNO was a partner in the project and provided low-voltage connections for the single charging stations. This was also the case in the Volvo sites. In Hamburg, the charging stations on Line 109, the Innovation Line, were duplicated at each terminus, so we had a load of over 600KVA. In that case, a decision was made to invest in a medium voltage connection, connecting from the metro network in the city centre and from the public network in the suburbs.
The medium voltage connection required its own concrete MV substation building, even though we used modern vacuum switchgear, as used in ESB unit substations, and dry transformers, to avoid any requirement for oil bunds or blast walls. This was supplied as a pre-fabricated, pre-wired unit delivered to site. The chargers were fitted on site in this case as these were some of the first production units.
While the LV connected chargers could be accommodated in new developments, provision had to be made for stations where there was no suitable building. The housing had to be able to handle the ventilation for compact high-power converters at ambient over 40°C with minimum temperature requirements of -25 °C, as well as dampening noise to requirements of residential areas.
A containerised unit was supplied for installation at the Volvo test track outside Gothenburg and for a temporary mobile charger for summer testing of the bus in the South of Spain. However, a conventional shipping container was never going to be acceptable as part of an urban streetscape, so some effort was spent in delivering a housing acceptable for a modern urban environment.
A standard unit was developed with Loughryan Engineering Services of Clonmel. This was a small prefabricated structure with doors at one end for the utility isolation and metering cabinet, at the other end for power and signal terminals to the contact system and its supporting mast, as well as a door in one side wall for access to the charger cabinet. All equipment including the isolation transformer were delivered to Loughryan where they were installed and wired in the housing structure before walls of noise-dampening Kingspan architectural panels and secure doors were assembled to complete the structure. The structure was then dispatched complete to site where it could just be dropped on a couple of prepared slabs, with just the external power and signal connections to be made on site.
On the street – contact system
The inverted pantograph had to be positioned over the bus parking position, with the centre of the contact arrangement approximately over the centre of the bus. The centre of the support structure was about 550mm offset from this position. With the communications antennas optimally over the boarding side of the bus, the supporting structure offset would be to the driver or normal roadside of the vehicle. In the nature of things, requirements of the sites or of the cities meant that the mast could be on the boarding or the driver side and appearance requirements could determine the positioning of the equipment, whether this was technically optimal or not.
A structure to hold the pantograph in this position is a very obvious addition to the streetscape. At the very least, it has to hold the raised pantograph at a minimum required clearance, typically 4.75m, above the roadway. Depending on local requirements, the mast may be close to the kerb or may be required to be some distance back. This obviously has an effect on the size of the supporting structure and necessary foundation.
- A very simple mast structure with external cable ladders was used in the Volvo locations. Different solutions were needed for public spaces and we worked with some variations in each city:
Stockholm Ropsten represented one extreme case with the mast base some 3m in from the kerb and the outer limit of the pantograph just under 3m out from the kerb. The location, at a busy traffic junction and directly beside the elevated T-Bana station, lent itself to a large, bold mast design, including a generous protective cover for the pantograph.
- In Hamburg, there was an interest to provide a less intrusive mast design that might be more acceptable in a quiet residential area but reflected at the same time the leading edge nature of the system. The mast arm is sized to be the minimum length, requiring the moving part of the pantograph to cover the optimum position of the antenna. This results in a somewhat larger association area in the suburban location but with minimal possibilities of mis-association in the actual road layout used. A leading industrial designer proposed a more rounded mast with GRP cladding. The mast structure was produced by Loughryan Engineering Services and the cladding by a local specialist in the Hamburg region. Hamburg was also interesting as a location where most official approvals were required for each stage, from planning, through mast and foundation design and manufacturing details.
- The outdoor station in Gothenburg chose a structure similar to the Hamburg mast but with a simpler aluminium cladding, manufactured by Tara Cladding in Limerick.
Other cities will have other priorities and the variety in the initial projects gives a basis for a process for delivering technical and aesthetic requirements of future projects.
The flexibility and professionalism of Con O’Regan Associates, in the structural design and input to planning documentation for Germany and Sweden, was of great assistance in responding to these differing demands, sometimes at very short notice.
Parking – precision and speed
A significant issue was reliability of parking as precisely as necessary to enable a good connection without taking excessive time. Siemens and other companies provide various automatic guidance systems for buses and other vehicles but current models tend to require a very slow approach in public spaces, especially in crowded locations, as well as involving significant extra complexity and cost in the vehicle.
A decision was made early on to rely mainly on the driver. We have a reasonable contact zone and experience from bus rapid transit systems with gated platforms has shown that the human eye is quite capable of very precise parking while sacrificing only a minor reduction in speed of approach. It took a little time working with drivers to find a reference view that was reasonably consistent for all drivers and seat positions. Once this had been identified, there were very few charging failures due to incorrect position.
However, there was a demand to provide driver-independent confirmation of position. We do provide a check of the distance from the kerb using an ultrasonic sensor on the supporting mast but measurement in the driver direction on the open street without any additional structures has proved much more difficult. An accuracy of +/- 5cm would be desirable in order to maintain maximum contact zone size, though +/-10cm would probably be adequate. We considered at least 5 different systems but none could meet the requirements of the open street. The ability to adjust WLAN power to limit the association zone more closely to the contact zone was made available but this does not help much where urban design requirements moved the antenna from the optimal position.
While experience has shown the drivers to be more than capable of parking consistently and accurately, there is always an interest in supporting the human input and options are still being investigated.
Operational experience with the development systems and with on-street service over the last few years indicates that we have developed a technically reliable and usable system. It does not represent an ideal for every individual stakeholder but it does represent a near optimal engineering solution that can enable us to move forward with the electrification of our transport systems.
Over the last two years, more bus and equipment manufacturers have started to use the system, which is being promoted under the name Oppcharge. One small city, Namur in Belgium, has placed orders to make 90% of its bus fleet electric. In addition, in North America, an initial pilot is under way in Montreal and plans are well advanced for systems in several other US and Canadian cities. The success of the system is confirmed by the fact that other companies are now developing compatible chargers and contact systems, which is a necessary step to providing adequate supply and a competitive market to achieve a faster rollout of the solution.
We should also remember that at least 25% of heavy trucks operate almost exclusively in a short-range urban environment. Given this type of charging infrastructure at depots and in strategic city locations, there is no good reason why we should not now move to electrification of all urban operating commercial vehicles and remove diesel pollution and inefficiency from our streets over the next decade.
Author: Liam Mulligan has extensive experience of issues of automation, energy and transport in his long career with Siemens in Ireland. He is a former chair of the Electrical and Electronic Division and is a member of the technical committee of ETCI – the Electro-Technical Council of Ireland. Liam worked on the international project with Siemens headquarters in Germany for eBus automated charging. Volvo Bus, part of Volvo AB, is one of the largest worldwide manufacturers of buses. He is now an independent consultant.http://www.engineersjournal.ie/2017/04/24/ebus-electric-bus-transport/http://www.engineersjournal.ie/wp-content/uploads/2016/04/eBus-1024x768.jpghttp://www.engineersjournal.ie/wp-content/uploads/2016/04/eBus-300x300.jpgElecelectric vehicles,transport