How Fleet Electrification is next in EV Evolution

Source: mouser.com

Electrification of medium- and heavy-duty vehicles (MHDVs) is on the rise. Growing fleets of buses, delivery trucks, logistics trucks, and other MHDVs will present new opportunities for technology advancement compared with the electrification of individually owned light-duty vehicles (LDVs). These advancements will benefit makers and users of all types of electric vehicles, from LDVs to off-road agricultural and mining vehicles.

MHDVs require much larger chargers and can support different charging infrastructures compared with smaller LDVs. These new fleets will require a complex mix of charging technologies, including conventional plug-in chargers, in addition to wireless chargers at fixed locations and wireless charging built into some highways for continuous charging while the vehicle is in motion.

Fleet operators are already well-versed in telematics to improve the operational efficiencies of existing fleets of internal combustion-powered vehicles. Software and real-time controls will become even more crucial to managing the energy consumption and charging demands of MHDV fleets.

This article considers various MHDV use cases for buses, delivery and refuse trucks, and logistics trucks. We then look at how the emergence of MHDV fleets will result in the development of new charging technologies and discuss the importance of software, including the sensors needed to supply the data, as well as advanced analytics and the use of artificial intelligence and machine learning to optimize the operation MHDV fleet operation.

MHDV Use Cases

MHDVs differ from today’s LDVs in several important dimensions (Figure 1). Heavy-duty vehicles are expected to last for 1 million kilometers or more, about three times longer than LDVs. MHDVs consume much greater amounts of power than LDVs: MHDVs consume from 300Wh/km to 2000Wh/km, while LDV energy usage is typically under 250Wh/km. The larger power consumption per kilometer combined with the longer daily driving distances for MHDVs requires larger battery packs. In some cases, that can be mitigated somewhat by employing opportunity charging, as we’ll review. LDVs typically have a battery pack under 100kWh, while current MHDV battery packs can be 660kWh. Future MHDV designs anticipate battery packs of 1MWh.

Figure 1: Classification standard used for commercial MHDVs. (MED = Medium; COE = Cab Over Engine) (Source: WTWH Media, LLC)

The MHDV itself, driver compensation, and fuel are the biggest operational costs when operating a conventional internal combustion engine (ICE). The adoption of electric MHDVs is expected to reduce fuel costs significantly. To justify the economic use of MHDVs, however, the lower fuel expenditures need to be large enough to offset the batteries’ high cost. Battery costs represent a significant cost in electric MHDVs that has no corresponding element in ICE-powered MHDVs. For example, suppose MHDV battery packs need replacement at much less than 1 million miles of use. In that case, it will significantly increase the cost of electric MHDVs and negatively affect the potential for large-scale deployment.

In addition, the differing driving patterns of various MHDVs is expected to impact the architecture of the corresponding charging infrastructure:

• Buses travel a dedicated route and can use opportunity charging at stops, and can benefit from the use of embedded highway charging
• Delivery trucks, refuse collection trucks, and similar services need depot charging because routes vary from day-to-day
• Logistics trucks, depending on routing and travel distance, can use depot charging, strategically placed charging stations along highways, or embedded highway charging for traveling on fixed point-to-point routes, such as back and forth from an airport or port to a warehouse district.

Charging Options for MHDVs

A growing number of MHDV charging systems are compatible with standard J1772-CCS Type 1 plug-in charging connections, and industry-standard SAE J3105 pantograph systems. Although plug-in chargers are found almost universally in depot charging scenarios, pantograph charging can be found in depot charging and in opportunity charging, primarily for buses during stops.

A typical use pattern for opportunity charging of buses is the use of a pantograph system and a charge time of three to six minutes during stops (Figure 2). The system also includes remote diagnostics of the charging system and batteries and fleet management software. Operating voltages of pantograph systems range from 150V to 850V, and power ratings typically range from 150kW up to 600kW, allowing the systems to support various bus sizes and charging needs.

Figure 2: Pantograph for electric bus ultra-quick charging. (Source: Scharfsinn/Shutterstock.com)

Wireless opportunity charging of MHDVs is an emerging technology expected to grow significantly in the next several years. For example, while pantograph systems are used to opportunity charge fleets of municipal buses, wireless opportunity charging of buses is still primarily under development and in field trials (Figure 3).

Figure 3: Electric bus at a stop being opportunity charged using wireless induction charging. (Source: Scharfsinn /Shutterstock.com)

Opportunity charging for buses and other MHDVs can provide benefits beyond the ability to support a given driving range with a smaller and lower-cost battery pack. Depth of discharge has a significant impact on the cycle life of batteries. The greater the depth of discharge, the shorter the cycle lifetime. For example, discharging a battery pack to near zero charge, instead of a 50 percent level, can reduce cycle lifetime by half. By supporting a lower depth of discharge rates, opportunity charging also supports longer battery cycle lifetimes.

Figure 4: Fleet of autonomous hybrid trucks driving on wireless charging lane. (Source: Chesky/Shutterstock.com)

MHDV charging will require higher power levels, and higher-input voltage chargers will support those higher power levels. Battery bus voltages of MHDVs are also expected to increase. Today, it is common to find 800V to 900V main power buses feeding the drivetrain. Research is underway to develop devices and circuits to support 1200V battery voltages and drivetrain power buses. For the largest applications (on the order of 1MW), even higher voltages are under consideration. The higher the voltage, the lower current for a given power level. Lower currents mean smaller and lighter-weight power buses can be used.

For high-power MHDV chargers, 480V mains are common. In the future, MHDV chargers are expected to operate from 1200V mains voltages. Future chargers are also expected to include digital controls that will enable a single charger to be used with various vehicles. These chargers will support multiple input voltages and identify the charging need of individual vehicles based on central controls that know the anticipated use of each vehicle and the state of charge and condition of each battery pack and modify the charging voltage and charging rate accordingly. For example, vehicles scheduled for later use can be charged more slowly compared with vehicles that are needed in the immediate term.

Figure 5: Delivery vans and trucks and refuse trucks are depot charged, and the charging stations can include integrated solar panels. (Source: Chesky/Shutterstock.com)

Software and Sensors

The electrification of MHDVs will result in both software-defined vehicles and software-defined vehicle fleets. In both cases, large quantities of data are required. Today’s vehicles have between 60 and 100 onboard sensors, and the data is almost exclusively analyzed on board the vehicle itself. Next-generation electric vehicles, including MHDVs, are expected to include double that number of sensors as vehicles become increasingly smart and connected using a combination of cloud computing and onboard computing to optimize vehicle performance.

The various mechanical and energy storage systems in an MHDV are relatively fixed once the vehicle has been built. Software is different; it can be updated regularly, supporting continuous learning and improvements to the various control systems to make MHDVs more efficient. Software also enables MHDVs to respond to changing operating conditions. Software has been called the new aerodynamics and controls everything from the drivetrain to the battery and energy systems. The drivetrain software that manages the energy flow between the battery and motor in today’s EVs includes over 1 million lines of code. It is expected to get even more complex for next-generation MHDVs.

Artificial Intelligence and Fleets of MHDVs

A significant difference between MHDVs and conventional electric vehicles is that most MHDVs will be used by fleet operators instead of individual owners. That central control over fleets of vehicles will provide increased operational optimization incentives using advanced analytics and artificial intelligence.

With the continued advancement of smart girds, the wide adoption of electric MHDV fleets is expected to bring extended environmental benefits. For example, the MHDVs will augment the grid and act as distributed energy storage nodes. Fleet operators and utility companies are expected to develop energy management capabilities and corresponding business models that help support optimal energy use from intermittent renewable energy sources. Several challenges need to be addressed if such benefits are to be achieved. Fleets and even multiple fleets of MHDVs need to be organized in such a way as to avoid peaks on the grid that can result in high electricity prices and overload local distribution grids.

Big data and the application of artificial intelligence will be important aspects of this expanded optimization. Machine-learning techniques are being developed to analyze energy-usage patterns of MHDV fleets and then apply AI algorithms to optimize charging parameters and scheduling. Machine learning and AI will be applied within the MHDVs themselves for quick and high-level performance adjustments. Data will also be uploaded to the cloud for more in-depth analysis and fine-tuning. As a result, it will be possible to deliver more energy without costly and time-consuming charging infrastructure upgrades.

Finally, EEMBC is planning an extension of the ADASMark benchmark suite to include machine-learning capabilities. ADASMark is a performance measurement and optimization tool for automotive companies building next-generation advanced driver-assistance systems (ADAS) and could find application in next-generation MHDVs. Intended to analyze system-on-chips (SoC) performance used in autonomous driving, ADASMark utilizes real-world workloads that represent highly parallel applications such as surround-view stitching, segmentation, and convolutional neural-net (CNN) traffic sign classification. The ADASMark benchmarks emphasize various forms of compute resources, such as the CPU, GPU, and hardware accelerators, allowing the user to determine the optimal utilization of available computing resources.

Conclusion

Growing fleets of buses, delivery trucks, logistics trucks, and other MHDVs will present new for technological advancement opportunities. These new fleets will require a complex mix of charging technologies, including conventional plug-in chargers in addition to wireless chargers at fixed locations, and wireless charging built into some highways for continuous charging while the vehicle is in motion.

Fleet operators are already well-versed in the use of telematics to improve the operational efficiencies of existing fleets of internal combustion-powered vehicles. Software and real-time controls will become even more crucial to managing the energy consumption and charging demands of MHDV fleets. The numbers of sensors on MHDVs will continue to grow and drive the need for advanced analytics, including cloud computing, AI, and machine learning, to optimize MHDV fleet operations and maximize the economic and environmental benefits