Method and device for battery life-optimized deployment planning of electric vehicles

Abstract

The invention relates to a method for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices, comprising: - Selecting (11) the vehicles (21, 22, 23) to be planned from the fleet, - Selecting (12) the routes to be driven by the selected vehicles, - Defining (13) an optimization target for an optimization parameter of the electrical energy storage (21a) of the selected vehicles (21, 22, 23), - Define (14) an optimization time point at which the optimization goal should be achieved, - Determining (15) a predicted optimization parameter for each selected vehicle (21) for the selected routes at the optimization time, - Assigning (17) the selected vehicles (21, 22, 23) to the selected routes according to the predicted optimization parameters.

Description

Technical field

[0001] The invention relates generally to the operation or battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices, such as electric trucks. State of the art

[0002] Vehicle data is increasingly stored in the cloud and serves as input for algorithms to determine operational or model parameters, such as the precise determination of battery state of charge (SoC), battery state of health (SoH), or predicted SoH values. This data can also benefit its owners, such as a freight forwarder who operates a fleet of trucks. Currently, freight forwarders lack access to this data and the software / programs to process it effectively for their business. Furthermore, their focus is on vehicle utilization, availability, and costs, not on vehicle or battery data.

[0003] WO 2023 / 096791 A1 discloses a method in which the load on a truck is reduced to a greater or lesser extent on different routes, thereby changing the energy consumption through weight reduction and thus also the SoC.

[0004] Route planning is based on the SoC value, energy efficiency, or cost of the truck.

[0005] US 9,079,505 B1 discloses a method in which the vehicle battery is configured differently depending on the route, so that battery packs with only high-energy cells, only high-performance cells or various configurations of both cell types are possible. Disclosure of the invention

[0006] According to a first aspect, a procedure is planned for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage systems, comprising: - Selecting the vehicles to be planned from the fleet, - Selecting the routes to be driven by the selected vehicles, - Defining an optimization target for an optimization parameter of the electrical energy storage of the selected vehicles, - Defining an optimization point in time at which the optimization goal should be achieved, - Determining a predicted optimization parameter for each selected vehicle for the selected routes at the optimization time, - Assigning the selected vehicles to the selected routes according to the predicted optimization parameters.

[0007] One idea is to determine the routes of several electric vehicles, such as electric trucks, with a fixed battery configuration, with the aim of achieving the same or as similar agility of the battery at a desired time.

[0008] A key aspect of the invention is the description of a method for creating a vehicle schedule optimized for the lifetime of an electric truck fleet, for use on different routes. This allows the freight forwarder to deploy their vehicles in such a way that, at the optimization point (e.g., time of inspection, sale, depreciation, or purchase of new vehicles), all vehicles have a specific battery aging state, ideally at the end of their service life (EoL) with a state of health (SoH) of 80%. This has the advantage of making the costs for new battery / vehicle purchases more predictable. The necessary data is all readily available or provided by battery service providers and can be combined with the method for predicting the SoH curve and determining the remaining useful life (RUL).Planning can be carried out for all electric trucks, regardless of whether the vehicles are used for urban, local or long-distance transport and on which routes they travel.

[0009] A general advantage is that an electric truck fleet can be planned for different routes in such a way that it meets specific technical and economic requirements at certain times. Battery data, for example from the cloud, is available for this purpose.

[0010] If this data is processed and visualized in an understandable way, a freight forwarder can also use this data for himself when it comes to planning routes for his vehicles and the remaining useful life (RUL) of the batteries in the electric trucks in his fleet.

[0011] Furthermore, the procedure can stipulate that the optimization parameter is the battery's state of health, its remaining service life, and / or a maximum or minimum battery aging. Ultimately, all technical and economic parameters of the battery can be used. These parameters can be taken directly from the vehicle's or battery's battery management system (BMS) or derived from it, possibly with additional data such as vehicle and / or route data.

[0012] According to one embodiment, the vehicles with electrical energy storage systems are electric trucks. The method is generally suitable for planned or repeatedly traveled routes, such as those driven by trucks, vans, or buses.

[0013] It may be stipulated that the fleet of vehicles with electrical energy storage systems be assigned to an electronic management unit. In addition to an organizational assignment, this could be a technical assignment to a control unit, a central computer, or the like.

[0014] Furthermore, the method can provide for the use of a current value of the optimization parameter as a starting value for determining the predicted optimization parameter. This value can then be used as a starting parameter for the iteration or simulation and can improve accuracy.

[0015] According to one embodiment, the vehicles with electrical energy storage devices can have a communication interface to a cloud network, wherein an electronic management unit for the fleet has a communication interface to the cloud network, and wherein the determination of a predicted optimization parameter is performed in the cloud network. Simple communication and distribution of computing power can be achieved via the cloud or a similar communication network.

[0016] It may be possible to determine a predicted optimization parameter by simulating the energy load profiles, charging frequency, and / or charging method using route profiles of the selected driving routes, optionally taking into account required rest periods. In this way, changes to the optimization parameter can be easily and reliably replicated for the optimization period under consideration.

[0017] Furthermore, the procedure can stipulate that the selected vehicles are distributed across the selected routes in such a way that as many predicted optimization parameters as possible for each selected vehicle fully or at least partially correspond to the optimization goal. This is in line with fleet optimization.

[0018] It can be configured that each selected vehicle is assigned one or more selected routes. Accordingly, a route can be driven multiple times or only once. The fundamental goal of fleet route planning—that all routes are served according to their respective frequency—is never neglected.

[0019] According to a second aspect, a device is provided for battery-life-optimized deployment planning of a fleet of vehicles with electrical energy storage systems, configured to execute a method as described above. The previously described embodiments of the method can also be embodiments of the device. Brief description of the drawings

[0020] Preferred embodiments of the present invention are explained in more detail below with reference to the accompanying drawings. These show: Fig.1 a flowchart of a procedure for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices; Fig. 2. A representation of a device for battery-life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices; and Fig. 3 Another representation of a device for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage. Description of embodiments

[0021] Fig. Figure 1 shows an example of a method 10 for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices. The vehicles with electrical energy storage devices can be, for example, electric trucks. The electrical energy storage devices are typically the vehicle batteries for the electric drive.

[0022] Since battery life decreases due to (frequent) fast charging, and consequently the operating life (RUL) is reduced, a trend further exacerbated by megawatt charging. This means the battery will reach its state of health (SoH) below 80% more quickly and should be replaced. Depending on whether an electric truck is used for urban, local, or long-distance transport, the number of fast charging cycles along the routes will vary. On long routes, megawatt charging during short breaks will be essential, while urban trucks will almost exclusively charge their batteries gently overnight at the depot using moderate currents.

[0023] It is proposed to provide the freight forwarder with a device in the form of a program (frontend) with which he can access the vehicle data in a cloud and process the data in such a way that an optimization goal specified according to his individual wishes is achieved, at the end of which is the allocation of his e-trucks for the different routes.

[0024] For connected cars, various data and services are offered as part of the "Battery as a Service" model. Currently, however, this data is only accessible to the manufacturer and authorized service centers. In the proposed concept, the transport company selects the vehicles it wishes to subject to the same optimization goal. For example, it can select vehicles intended for long-haul transport, perhaps based on the license plate and VIN, in order to schedule them so that their high-voltage batteries all exhibit nearly identical battery aging status after, say, six months.

[0025] A first step 11 involves selecting the vehicles to be planned from the fleet; this can be all vehicles, a part of the vehicles, or even just a single vehicle.

[0026] A second step (12) involves selecting the routes to be driven by the selected vehicles. This can include not only the routes themselves, but also the frequency with which the routes are driven until an optimization point.

[0027] A third step (13) involves defining an optimization target for an optimization parameter of the electrical energy storage system of the selected vehicles. The optimization parameter can be the battery's state of health, the remaining battery life, and / or a maximum or minimum battery aging.

[0028] A fourth step (14) involves defining an optimization point in time at which the optimization goal should be achieved. This point in time, or the end of an optimization period, could be, for example, 1 month, 6 months, 1 year, the time of an inspection, a sale, the depreciation date, the time of acquiring new vehicles, or similar.

[0029] Another optimization goal could be the decommissioning of an electric truck. The strategy here is to deploy the truck on routes where the battery experiences significant aging. An optimization point could also be the point at which a new electric truck is purchased, for which the state of health (SoH) can be optimized. This can be beneficial for resale but detrimental for scrapping an older vehicle.

[0030] A fifth step 15 involves determining a predicted optimization parameter for each selected vehicle for the selected routes at the optimization time.

[0031] In this example, the algorithm uses vehicle-specific data, such as the current state of health (SoH) value, and is aware of the load profiles as well as the charging behavior (frequency and charging method) on the routes. In long-haul transport, these profiles are almost identical for the same routes. The load profiles and operating volume profiles for the different routes, which are covered daily or weekly by the electric trucks, are available and form the basis for the optimization tool, which runs in the cloud. The current number of fast-charging cycles for each vehicle, compared to the maximum permitted number of fast-charging cycles, is also available in the cloud and is taken into account in future route planning.

[0032] The tool can also show the freight forwarder whether they should continue using the electric truck for long-distance transport or whether it would be better suited for local transport in the future. The optimization tool also takes into account potentially different battery capacities of the vehicles or different powertrain configurations (electric motorization) and simulates energy consumption, charging frequency, and charging method based on a predefined route profile, while also considering required rest periods.

[0033] A Lifetime Optimization (LO) service determines the respective state of health (SoH) values ​​of the batteries of the different vehicles on the routes, and the Aging Prediction (AP) service determines the predicted SoH value. The trucks usually travel the same routes, so a route change can lead to a more aged battery aging more slowly on a different route, while a less aged battery ages more quickly, resulting in the overall longer operating life of the electric trucks before a battery replacement is necessary.

[0034] In a sixth step (16), it is checked whether the optimization goal is achieved, that is, whether the predicted optimization parameter(s) reach the optimization goal(s). If this is not the case, the process branches back to step 15 to start the next iteration. If the optimization goal is achieved, the procedure continues to step 17.

[0035] A seventh step 17 involves assigning the selected vehicles to the selected routes according to the predicted optimization parameters.

[0036] From the pool of electric trucks selected for optimization, the vehicles are assigned to the corresponding routes, resulting in a nearly identical state of health (SoH) value at the desired time. Further route swaps within the selected period are also possible to achieve the optimization goal.

[0037] Even small freight forwarding companies can benefit from this process. Instead of direct optimization, the freight forwarder can manually select different routes for various electric trucks, assigning a specific route to each truck. This data set is then transferred to the cloud, where the state of health (SoH), predefined SoH, and reserve load (RUL) are determined. The result is then transmitted to the front end, allowing the freight forwarder to modify individual routes until the desired outcome is achieved.

[0038] This process benefits both freight forwarders and service providers, as changes to the routes of other fleet vehicles are updated in the cloud, similar to swarm intelligence. This provides users with an easy-to-use interface offering clear and actionable information.

[0039] Fig.Figure 2 shows a representation of a device for battery-life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices. The device 20 can be implemented in hardware and / or software and is configured to carry out the previously described method 10.

[0040] The device 20 or the system can include electric vehicles 21, 22, 23, such as electric trucks, or these can be arranged externally to the device 20. In this case, relevant data is transmitted to the device 20 via interfaces. Here, three vehicles 21, 22, 23 are shown as an example fleet.

[0041] Each vehicle 21 has an electrical energy storage device 21a and a battery management system 21b. Battery data is transmitted from the battery management system 21b to a cloud 24 and processed there. This may include the execution of the above procedure 10. Results of the procedure 10 and / or further processing are sent to an electronic management unit 25, such as a frontend. Input can also be made from there to the algorithm in the cloud 24.

[0042] The method or algorithm can also be distributed in the frontend 25 or in the components of the device 20.

[0043] Fig. Figure 3 shows another representation of a device 30 for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage systems. Part of the device 30 or the system is located in the cloud 24, which is connected to the frontend 25.

[0044] Speed ​​and altitude profile data 31 are entered into a vehicle model 32, which calculates user data 33. Instead of these two steps, the user data 33 can originate from a vehicle.

[0045] The user data is entered into an algorithm 34 for a predicted optimization parameter such as SoH, which calculates this predicted optimization parameter (as explained, for example, using method 10) and enters it into an optimization algorithm 35.

[0046] This optimization algorithm 35 continues to receive input from the frontend 25, such as selection criteria concerning the selected vehicles, the selected routes, the optimization goal and / or the optimization time.

[0047] Based on these inputs, the optimization algorithm 35 assigns the selected vehicles to the selected routes (as explained, for example, using procedure 10).

[0048] A specific example is described below.

[0049] First, the trucks from the vehicle pool or fleet that are to undergo optimization are selected. Based on the latest inspection data, especially the battery data and specifically the state of health (SoH) value, and optionally the specified predicted SoH value, the transport company knows which truck battery has aged the most, or which battery will need to be replaced next, assuming the same driving style / route.

[0050] For example, if his optimization goal is to replace three trucks from his fleet in 6 or 12 months or to replace the batteries, he will try to use the vehicles in such a way that all three trucks have the same (poor) state of health (SoH) value at that time, so that the replacement is economically optimal.

[0051] The optimization can be performed for trucks of the same weight class, i.e., trucks in the category up to 40t (heavy trucks, long-haul), up to 18t (medium trucks, intercity), and up to 7.5t (light trucks, urban traffic), so that logistics can continue to be operated with the same truck sizes as before. Otherwise, several smaller trucks would have to be used to transport the same amount of cargo, or a large truck would not be fully loaded.

[0052] The process begins with the freight forwarder entering various selection criteria on a mobile device, such as a PC, on which only the Frontend 25 of the optimization tool is installed. The following selection options are available: selection of the trucks to be optimized, selection of the routes on which the trucks to be optimized should travel, and selection of the optimization goal and the target time.

[0053] This data is transmitted to Cloud 24, where the actual optimization is carried out using the described algorithm. Since all trucks are connected to Cloud 24, the values ​​determined by the vehicle's BMS and transmitted to Cloud 24 (user data 33) for all routes are available there for each truck.

[0054] If the selected trucks all have the same powertrain (battery and electric motor), the user data 33 can be accessed directly and used for predicting the state of health (SoH) value. Otherwise, if the powertrains are different, the current profile is first generated using the corresponding vehicle model 32 based on the speed and altitude profile of the routes. This current profile serves as the input curve for predicting the SoH value.

[0055] Starting with the current state of health (SoH) value of the selected trucks, the predicted SoH values ​​of the batteries of the selected trucks are determined based on the individual, different routes up to the optimization target time; this is done by concatenating identical routes. A matrix structure (truck 1...n for route A...k) is thus simulated, and the corresponding determined SoH values ​​are compared with the target values, for example, SoH = 80%. Table 1: Example SoH results table % SoH_Start Route A Route B Route C Route D Route E Route F Truck 1 87 83 82 80 81 83 84 Truck 2 83 77 76 74 75 77 78 Truck 3 85 80 79 77 78 80 81

[0056] Table 1 shows that if truck 1 is deployed on route C, for example, for the next 6 months (optimization point), it will exactly meet the specified target value. Truck 2 will fall short of the target value on all routes; route F comes closest to the target value. Truck 3 reaches the target value of SoH = 80% on either route A or E. In this case, the freight forwarder can decide which route to deploy truck 3 on. Similarly, the freight forwarder can decide which route to send a truck on if the specified target value is achieved by several trucks on a single route. In this optimization, each truck is assigned only one fixed route. This represents the simplest version of the procedure.

[0057] In reality, however, each truck travels multiple routes. The frequency with which a truck travels its individual routes also varies, as does the route length / driving time. This information can be gleaned from user data 33. This results in a certain number of possible route combinations for the selected trucks over a specific period (for example, a week, a month, or the optimization target period). These combinations exclude the same routes for multiple trucks at the same time. From the remaining combinations, the SoH values ​​are determined, and the corresponding routes are assigned to the trucks whose SoH values ​​are closest to the target value. The order of the routes is irrelevant for determining the SoH value, but it is important for route planning. Table 2: Example of combining routes (routes A and B are traveled twice) Truck 1 Route B Route A Route B Truck 2 Route C Route F Truck 3 Route A Route D Route E

[0058] The primary goal of route planning—ensuring all routes are served and their respective frequencies are considered—is never overlooked. The procedure was described using the optimization goal of "same state of health" as an example; however, it can be applied analogously to other optimization criteria. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] WO 2023 / 096791 A1

[0003] US 9,079,505 B1

[0005]

Claims

Method, in particular a computer-implemented method for battery lifetime-optimized deployment planning of a fleet of vehicles (21, 22, 23) with electrical energy storage devices, comprising: - Selecting (11) the vehicles (21, 22, 23) to be planned from the fleet, - Selecting (12) the routes to be driven by the selected vehicles, - Defining (13) an optimization goal for an optimization parameter of the electrical energy storage devices (21a) of the selected vehicles (21, 22, 23), - Defining (14) an optimization time point at which the optimization goal is to be achieved, - Determining (15) a predicted optimization parameter for each selected vehicle (21) for the selected routes at the optimization time point, - Assigning (17) the selected vehicles (21, 22, 23) to the selected routes according to the predicted optimization parameters. Method according to claim 1, wherein the optimization parameter is the health status of the battery, the remaining service life of the battery and / or a maximum or minimum aging of the battery. Method according to claim 1 or 2, wherein the vehicles (21, 22, 23) with electrical energy storage devices are electric trucks. Method according to one of claims 1 to 3, wherein the fleet of vehicles (21, 22, 23) with electrical energy storage devices is assigned to an electronic management unit (25). Method according to one of claims 1 to 4, wherein a current value of the optimization parameter is used as a starting value for determining the predicted optimization parameter. Method according to one of claims 1 to 5, wherein the vehicles (21, 22, 23) with electrical energy storage devices have a communication interface to a cloud (24) network, wherein an electronic management unit (25) for the fleet has a communication interface to the cloud (24) network and wherein the determination of a predicted optimization parameter is performed in the cloud (24) network. Method according to one of claims 1 to 6, wherein determining a predicted optimization parameter comprises simulating the energy load profiles, the charging frequency and / or the charging type using route profiles of the selected driving routes, optionally taking into account required break times. Method according to one of claims 1 to 7, wherein the distribution of the selected vehicles (21, 22, 23) onto the selected routes is carried out in such a way that as many predicted optimization parameters as possible of the respective selected vehicles (21, 22, 23) fully or at least partially correspond to the optimization goal. Method according to any one of claims 1 to 8, wherein each selected vehicle (21) is assigned one or more selected routes. Device (20; 30) for battery life-optimized deployment planning of a fleet of vehicles with electrical energy storage devices, designed to carry out a method (10) according to one of claims 1 to 9 . Computer program product comprising instructions which, when the program is executed by at least one data processing device, cause it to execute the steps of the method according to any one of claims 1 to 9. Machine-readable storage medium comprising instructions which, when executed by at least one data processing device, cause it to execute the steps of the method according to any one of claims 1 to 9.

Images