Method for optimizing the service life of a drive battery of a vehicle

Abstract

In a method for optimizing the service life of a drive battery of a vehicle, an effective state of health is raised (71) with a first periodicity. Proceeding from an effective state of health raised last, an estimated state of health of the drive battery is ascertained (61, 62) by applying an ageing model for the drive battery with a second periodicity, taking into account use data of the vehicle since the estimated state of health was last ascertained. Finally, the vehicle is operated (21) on the basis of the estimated state of health and a current drive energy requirement, wherein at least one battery loading-related control parameter is newly determined with at least a third periodicity. The ageing model for the drive battery is calibrated (72) each time the effective state of health is raised.

Description

[0001]Method for optimising the service life of a drive battery of a vehicle Technical field The invention relates to a method for optimising the service life of a drive battery of a vehicle and to an arrangement for operating an electrically powered vehicle. State of the art The drive battery is an essential component of electrically powered vehicles. This is subject to an aging process which leads to a decrease in the usable capacity (i.e. the energy that can be stored and accessed) and an increase in the internal resistance of the drive battery. Aging can be divided into aging due to the loads during operation and calendar aging. Aging due to vehicle operation depends on various parameters, including the charging and discharging power levels occurring and the cell temperatures. Different operations therefore also lead to different aging.Accordingly, the service life of identical batteries can vary, with service life generally referring to the number of kilometers driven until a specified residual capacity (e.g., 80% of the initial capacity) is reached. Methods are known for increasing the service life of drive batteries by influencing vehicle operation. 28115 WO PR / zol Keller Schneider Patent- February 8, 2024 and Trademark Attorneys AG For example, US 2022 / 0126723 A1 (GM Cruise) concerns the fleet-based optimization of drive batteries in electric vehicles. For this purpose, the current state of health (SOH) is measured using impedance spectroscopy. The measurement results from many vehicles in the fleet are compared to determine parameters such as remaining service life or factors for increased battery aging; the data can also be used to train a machine learning model to estimate battery capacity, remaining service life, or other parameters.Measures to optimize SOH include, in particular, restrictions on the use or charging of a specific battery or an adjustment of route assignments. It is also possible to limit the energy consumption of the heating, ventilation, and air conditioning (HVAC) system. The current SOH can be determined from time to time by charge counting; based on the corresponding data, approaches can be developed that enable a faster and less stressful determination of SOH on the battery and that can be performed more frequently. Increasing battery life is often accompanied by restrictions on driving and / or a loss of comfort. Excessively battery-friendly operation to ensure a minimum service life is therefore undesirable.Description of the Invention The object of the invention is to create a method belonging to the technical field mentioned above for optimizing the service life of a drive battery of a vehicle, which ensures that a predetermined service life of the drive battery is achieved with minimized adverse effects on driving operation and / or user comfort. The solution to the problem is defined by the features of claim 1. According to the invention, a method for optimizing the service life of a drive battery of a vehicle comprises the following steps: 28115 WO PR / zol Keller Schneider Patent- 08.February 2024 and Markenanwälte AG a) Determining an effective state of health of the traction battery with a first periodicity; b) Determining an estimated state of health of the traction battery with a second periodicity, based on a last determined effective state of health, taking into account vehicle usage data since the last determination of the estimated state of health, by applying an aging model for the traction battery; c) Operating the vehicle based on the estimated state of health and a current drive energy requirement, whereby at least one battery load-relevant control parameter is re-determined with at least a third periodicity. The aging model for the traction battery is calibrated after determining the effective state of health. The state of health of a traction battery is regularly referred to as the State of Health (SOH).It is defined as the ratio between a remaining charge capacity ^. ^ ^^^ to an original charge capacity ^ ^ǡ^^^௧ : ^ ^^^ ൌ ^ୠ^^^ ^ Ǥ The usage data is Vehicle usage relevant to battery aging. These can be collected at the vehicle level (e.g., speeds, accelerations, drive power, operating times, etc.) and / or at the drive battery level (e.g., consumed power, charging power, charge level, cell temperatures, etc.). Various aging models for the drive battery can be used. Such models are available, for example, in the open source project OpenSesame (M. Beyeler, SS Bhoir, S. Broennimann, Y. Moullet, Open-SESAME-Battery, V1.1, https: / / gitlab.ti.bfh.ch / oss / esrec / open-sesame (2022)). 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG According to the invention, when operating the vehicle, in addition to the current drive energy requirement (and possible further input parameters, relating, for example, to the desired operation of a heating, cooling or air conditioning system as well as other components of the vehicle), the estimated health status of the drive battery is taken into account. In this context, at least one control parameter relevant to the battery load is repeatedly re-determined. By specifying this control parameter, battery aging can be influenced, in particular with regard to ensuring a predetermined battery service life with the least possible restrictions in driving operation and / or minimal energy consumption. The periodicities are selected in particular such that the re-determination of the battery load-relevant control parameter occurs frequently in order to take into account the changing requirements in driving operation.The estimated health status can be determined less frequently, e.g., after each driving mission or when parking the vehicle in a depot, etc. The actual health status is determined, in particular, by explicitly determining the remaining capacity of the drive battery using common methods. These are time-consuming, and accordingly, this process is performed much less frequently. The aforementioned steps can essentially take place regularly, according to a predetermined time frame and / or upon predetermined triggering criteria, such as when parking the vehicle, during maintenance work, etc. According to the invention, the aging model is calibrated depending on the behavior of the specific drive battery, i.e., the model is adapted to best describe the aging of this specific drive battery. An initial aging model can initially be universally specified, wherebyFindings from the operation of a vehicle fleet or a sub-fleet with identical or similar traction batteries can be taken into account when determining the initial aging model and / or during calibration. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG The method according to the invention thus makes it possible to ensure that the traction battery neither ages too quickly and therefore needs to be replaced too early, nor is it over-protected, which can lead to inefficient driving, increased energy consumption, or a loss of comfort. With the help of the method according to the invention, it can also be avoided that batteries with excessive capacity and thus unnecessarily high weight are used to ensure the service life and / or the requirements of driving operation. Likewise, the choice of the vehicle used on a specific route and / or its configuration can be optimized. Due to theRegular determination of the actual health status and its adjustment through interim assessments ensure that a realistic SOH value is continuously available as a basis for vehicle operation. At the same time, systematic deviations of the SOH value from the actual health status that accumulate over the vehicle's service life are avoided. The control parameter is advantageously determined based on a deviation of the estimated health status from a reference curve. The deviation is updated, in particular, according to the second periodicity, i.e., when the estimated health status is reassessed. The reference curve represents an expected development of the health status of the traction battery. If the estimated health status is better than the reference curve, measures to protect the traction battery, which could, for example, result in restrictions in driving operation or higher energy consumption, can be reduced.If the estimated health status is worse than that according to the reference profile, measures to protect the drive battery can be initiated or intensified. The relatively close determination or estimation of the health status based on concrete usage data enables short-term measures in this context and ultimately avoids both excessive degradation of the drive battery and unnecessary restrictions or excessive energy consumption due to excessive protection of the drive battery. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG Preferably, the reference profile is determined such that the drive battery has a predetermined health status for a given mileage. For example, it is specified that the drive battery still has an SOH or residual capacity of 80% for a mileage of 1.2 million km. The reference profile of the SOH is, in particular, aFunction of the distance driven and can have a linear or non-linear dependency. The latter can be useful in order to take into account age-dependent influences on battery aging. Depending on the drive battery and vehicle, for example, increased aging can occur at lower SOH. Thus, the specified service life is targeted in a closed control loop. The goal is the most efficient and comfortable operation possible while ensuring the required battery service life. It is advantageous to recalculate the reference curve after the actual health status has been recorded. Instead of the overall curve between the first use of the new drive battery and the specified mileage, the newly calculated reference curve represents the time period between the last recording and the specified mileage, with the SOH value of the last recording forming the starting value of the curve. If this is higher than according to the previous definition of theIf the reference curve is higher than the reference curve, a (slightly) faster aging is accepted; if it is lower, the battery is additionally protected to achieve a (slightly) slower aging. Advantageously, the usage data taken into account to determine the estimated state of health include a battery charge level curve since the last determination. It has been shown that a reliable estimate can be achieved based on the curve of this single value, whereby both the influences of consumers and charging and braking processes are indirectly taken into account, and influences of the respective charge level can be incorporated into the estimate. The estimate can be based on the curve itself and / or on the first and / or second derivative of the battery charge level. If necessary, even higher derivatives can be considered. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG In a preferred embodimentThe estimated health status is calculated from the most recently recorded effective health status less a sum of degradation estimates that were made based on the aging model, based on the usage data, since the last recording of the effective health status. Each degradation estimate covers a period since the previous estimate. To calibrate the aging model, the estimated health status of the traction battery at the time the effective health status was recorded is preferably compared with the recorded effective health status. The difference corresponds to an over- or underestimation of the effective aging. By adapting the aging model, the degree of over- or underestimation can be reduced, which generally leads to a better estimate for the subsequent interval. In particular, the calibration takes into account individual characteristics of the specific traction battery.automatically taken into account, i.e. the aging model automatically adapts to the specific traction battery. To determine the control parameter (at least temporarily or in addition to other target variables), an optimization is advantageously carried out with a view to maximizing a first target variable corresponding to the service life of the traction battery. This optimization thus enables an increase in the service life or an increase in the residual capacity (or the SOH) for the specified mileage. It is particularly necessary if the health of the battery is worse than necessary at a certain point in time, e.g. lower than according to a reference curve. Maximizing the service life is achieved in particular by minimizing damaging operating phases, e.g. by limiting the charging power, particularly during braking, or by limiting the reference power through reduced acceleration and / or a temporaryReducing the reference power of vehicle components (e.g., the HVAC). 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG With a view to maximizing the first target value, it is advantageous to determine the influence of operating parameters on the health status of the drive battery based on the vehicle's usage data and the recorded effective health status. This allows the vehicle- and drive battery-specific determination of which operating configurations, measures, and events lead to increased (or reduced) battery aging. Likewise, the influences of battery aging itself can be automatically taken into account: For example, it is automatically determined if certain processes represent an additional load with increasing battery age. This results in a self-learning control system that detects operating points that are harmful to the drive battery (including vehicle- and battery-specific dependencies and influences).The control system automatically detects the battery ageing and initiates measures within a control loop to ensure the required service life of the drive battery with the least possible impairment of driving operation and user comfort. The relationship between the operating parameters and the influence on the health of the drive battery can be determined, for example, using machine learning techniques, including regression methods. To determine the control parameter (at least temporarily or in addition to other target variables), an optimization is advantageously carried out with a view to minimizing a second target variable corresponding to a (primary) energy consumption of the vehicle. The method according to the invention has the particular aim of ensuring a specified battery service life without the drive battery having to be excessively protected, which would lead to restrictions in driving operation and, in particular, to increased energy consumption.For example, to protect the battery, the maximum power consumption of the drive battery during braking can be limited, which means that a certain portion of the braking energy is not recuperated but is converted into heat, for example, in braking resistors. If such a limitation is not necessary at all or only to a reduced extent without jeopardizing the service life, optimization with a view to reducing energy consumption is possible. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG Particularly preferably, the weighting of the first and second target variables is adjusted in the optimization depending on the deviation of the estimated health status from the reference curve. In particular, the first target variable is given a greater weighting if the estimated health status falls below the reference curve, while the second target variable is given a greater weighting ifif the estimated health status is better than according to the reference curve. Thus, the trade-off between (primary) energy consumption and battery aging can be optimally adjusted automatically and by varying a single parameter. It is advantageous if a first period of the first periodicity is longer than a second period of the second periodicity, and the second period is longer than the third period of the third periodicity. In particular, the following are: - the first period 14 days to 400 days, - the second period 1 hour to 24 hours, and / or - the third period 0.1 to 10 s. The actual health status therefore only needs to be recorded relatively rarely, e.g., during periodic maintenance work. The new determination of the estimated health status, e.g., after each mission or daily after the end of the mission, simultaneously enables sufficiently close consideration of battery aging during driving operation. The determination of theThe control parameter is adjusted more frequently in order to take account of the changing requirements of driving operation. If several control parameters are redetermined depending on the estimated state of health, the corresponding periodicities can be the same or different from one another. For example, the drive power can be redetermined more frequently than the control parameters for the HVAC. The method according to the invention is particularly applicable to the operation of battery-electric vehicles, particularly preferably in buses for passenger transport, e.g., in scheduled service. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG. In the present case, a "battery-electric vehicle" is understood to be a vehicle whose drive power is generated by means of one or more electric motors fed by one or more drive batteries, wherein the electrical energy for the drive battery(ies) is supplied from external power sources during charging processes.The charging processes can be carried out stationary and / or during travel, e.g. via overhead lines. The method according to the invention is particularly advantageous for operating vehicles that comprise a pantograph to collect electrical energy from an overhead line during operation. This particularly concerns trolleybuses that are able to cover sections of the journey without drawing electrical drive energy from the overhead line, with the drive energy on these sections being taken from the drive battery. This also makes it possible to travel on lines that are not continuously equipped with overhead lines. The drive battery is recharged on the sections with overhead lines. In trolleybuses, excessively battery-saving operation can lead to increased overall energy consumption. This can be avoided with the help of the method according to the invention. In particular, the method according to the invention also makes it possibleadjust the charging power as a control parameter depending on the estimated health status of the traction battery. Trolleybuses that can be operated according to the invention can comprise charging connections for receiving electrical energy in addition to the current collectors for the overhead line, for example, to pre-charge the vehicle's traction battery before an operating cycle in a depot. In principle, the method can also be used for hybrid vehicles that, in addition to the traction battery and one or more electric drive motors, comprise an internal combustion engine for generating electrical energy and / or direct mechanical drive power. In power-split hybrid vehicles, the 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Markenanwälte AG at least one battery load-relevant control parameter includes, in particular, the distribution of the reference power between the battery and the internal combustion engine. The method is for operating aIt is suitable for the control of a vehicle used in scheduled service, whereby forecast driving data of a route being traveled or of routes to be traveled are taken into account when determining the control parameter. Using such information, for example, the charge level of the traction battery can be regulated in such a way that, on the one hand, sufficient drive energy is available along the entire line and, on the other hand, as much braking energy as possible can be recuperated in corresponding sections with the lowest possible load on the traction battery. Similarly, in trolleybuses, the charge level of the traction battery can be conditioned with regard to an impending charging process in a section with overhead lines. The method can also be used to assign similar vehicles within a fleet to individual lines based on the estimated health status of their traction battery. For example, vehicles whose traction batteries have ahave a below-average state of health are assigned to lines with a low load on the traction battery, while vehicles whose traction batteries have an above-average state of health are assigned to lines with a higher load. Alternatively, a battery swap between two or more vehicles is recommended. The periodic reassessment of the health status avoids the excessive effects of such measures: based on the updated estimates for all vehicles in the fleet, a reassignment can take place at any time, so that the loads are balanced across the entire fleet. The control parameter preferably comprises one or more of the following parameters: a) a state of charge; b) a charging power; c) a reference power from the traction battery; 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG d) a maximum acceleration of the vehicle;e) a maximum recuperation current of the vehicle; f) a consumption power for a braking resistor; g) an operating parameter of a heating, ventilation and / or air conditioning system of the vehicle; h) an operating parameter of an interior lighting and / or an interior information system and / or other auxiliary consumers of the vehicle; i) an operating parameter of a thermal management system for the traction battery. For the state of charge and the charging power, the control parameter can be a specified value or a minimum or maximum value. One possible control parameter is therefore the maximum or optimal depth of discharge (DOD). The charging power can refer to either the stationary supply or the supply from an overhead line. The supply power from the traction battery can be specified as a control parameter if there is an immediate possibility of supplying primary energy during driving, in particular from aOverhead line, from a fuel cell or – in a hybrid vehicle – from an internal combustion engine, either directly as a drive or indirectly via a generator. If the specified maximum recuperation current is reached and additional braking power is required, a braking resistor can be used to convert excess energy into heat and / or an additional mechanical braking system. The operating parameter of the HVAC can be a specified temperature of the vehicle interior. However, the HVAC system is preferably operated at reduced power or switched off for short periods during operating phases of high load. If a heat storage or cold storage system is available, this can be filled during phases with low load on the traction battery or with excess recuperation energy and used to operate the HVAC system during phases of high load. Instead of operating parameters of the HVAC system, a maximum power of theThe entirety of the auxiliary consumers can be specified. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Trademark Attorneys AG The thermal management system for the traction battery can also be influenced within certain limits in order to avoid short-term load peaks, e.g., by temporarily reducing the power of a corresponding unit or switching it off, whereby higher temperature levels can also be accepted depending on the situation. The effects of a switch-off process or a power reduction can be compensated for by a corresponding power increase before or after such a measure. If the vehicle control system is already aware of an impending measure, e.g., when operating on a known route, the compensation can be carried out in full or in part in anticipation. Other control parameters relate to the triggering of instructions to the driver, e.g., regarding a target speed or the initiation of a braking process. If theIf the method according to the invention is used in connection with a trolleybus with an integrated traction battery, the power to be absorbed from the overhead line forms one of the control parameters, since generally, by increasing the proportion of the required power that is taken from the overhead line, the battery can be relieved and thus ageing can be delayed, while at the same time, due to the higher resistances in the overhead line network, the total energy required increases. By suitably varying the proportion of the power taken from the overhead line, according to the method according to the invention, the desired balance between energy efficiency and battery life can be achieved. Load peaks in the overhead power network can also be favorably influenced. An arrangement according to the invention for operating an electrically powered vehicle according to the method according to the invention comprises: a) the electrically powered vehicle with a traction battery forDriving at least one drive unit of the vehicle; and b) a measuring device for determining an effective state of health of the drive battery of the vehicle; wherein the vehicle comprises the following: 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG c) a processing device for determining an estimated state of health of the drive battery; d) a calibration device; e) a vehicle controller for operating the vehicle; wherein - the processing device is configured to receive the last measured effective state of health from the measuring device, to receive usage data of the vehicle from the vehicle controller and to determine the estimated state of health based on the effective state of health and taking into account the usage data since a last determination of the estimated state of health, by applying an aging model for the drive battery; - theThe calibration device is configured to calibrate the aging model based on the actual health status received from the device; and - the vehicle control system is configured to determine at least one battery load-relevant control parameter based on the estimated health status and a current drive energy requirement. The measuring device is, in particular, arranged stationary, e.g., in a maintenance center or in a workshop. The most recently measured actual health status can be received directly from the measuring device or indirectly, e.g., from the vehicle control system, to which this data was previously transmitted from the measuring device or from a corresponding server. Further advantageous embodiments and feature combinations of the invention arise from the following detailed description and the entirety of the patent claims. 28115 WO PR / zol Keller Schneider Patent- February 8, 2024 and Trademark AttorneysAG Brief Description of the Drawings The drawings used to explain the exemplary embodiment show: Fig. 1 a schematic block diagram of an embodiment of an arrangement according to the invention for operating an electrically powered vehicle; Fig. 2 a schematic representation of the method according to the invention for optimizing the service life of a vehicle's drive battery; Fig. 3 a schematic representation of the power balance of the sources and loads of the electrically powered vehicle; Fig. 4A the resulting power distribution for an operating strategy to maximize total energy consumption; Fig. 4B the resulting power distribution for an operating strategy to maximize battery service life; and Fig. 5 results of a lifetime simulation to compare the results of the method according to the invention with a method without regular recalibration. In principle, the same parts have the same reference numerals in the figures.Ways of carrying out the invention Figure 1 shows a schematic block diagram of an embodiment of an arrangement according to the invention for operating an electrically powered vehicle. Connections for transmitting electrical power are shown in solid lines, connections for transmitting control and regulation signals are shown in dot-dash lines, and connections for the (temporary) transmission of measurement signals are shown in dashed lines. The vehicle 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG is a trolleybus 10 that can receive power from an overhead line via a current collector 13 during operation, both while in motion and when stationary. The trolleybus 10 also has a charging connection 14 that enables a wired power supply, e.g., in a depot where the trolleybus 10 is parked outside of its operating phases. The electrical energy is stored in a traction battery 12 of the trolleybus 10. ItThese are, as is known per se, Li-ion batteries with a capacity of, for example, 72 kWh, which enables travel over longer stretches without overhead lines. In addition to the traction battery 12 and the charging connection 14, the operating components 10a of the trolleybus 10 include the drive unit 11, which comprises several electric motors powered by the traction battery 12, auxiliary consumers 15 (e.g., heating, ventilation and air conditioning, lighting, on-board entertainment, etc.), a braking resistor 17, and a power distributor 16, via which the traction battery 12 is connected, on the one hand, to the current collector 13 and a downstream unidirectional DC-DC converter 18 and the charging connection 14, and, on the other hand, to the consumers, i.e. the drive unit 11, the braking resistor 17, and the auxiliary consumers 15. It should be noted that the drive unit 11 acts as a generator in recuperation mode and accordingly feeds electrical energy into theThe trolleybus 10 further comprises a control system 20. This includes a vehicle control system 21 with various control modules, which, on the one hand, control and monitor the drive unit 11 and, on the other hand, the other vehicle components, including the auxiliary consumers 15. Furthermore, the control system 20 comprises a processing device 22 for estimating the health status of the drive battery 12 and a calibration device 23 for updating the aging model. Both are in turn connected to the vehicle control system 21 for receiving vehicle information and transmitting control parameters. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG When stationary, the trolleybus 10 can be connected to a measuring device 50 for determining the effective health status of the drive battery 12. This determination provides, in particular, the current (residual) capacity and is carried out using known methods.e.g., through a complete charging and discharging cycle with recording of the transferred energy. Figure 2 is a schematic representation of the inventive method for optimizing the service life of a traction battery of a vehicle, specifically the trolleybus 10. The method comprises several feedback loops at different time levels. In the lowest level 40.1, driving signals 45 and other control signals from the driver (and, if applicable, a control center of the operator) are recorded in a manner known per se and transmitted to the operating components 10a of the trolleybus 10. At the same time, the driving signals 45 and other control signals, or a selection of these signals, are transmitted to the vehicle control system 21. In a next higher level 40.2, a degradation estimate 61 is made based on signals from the operating components 10a (which may be provided via the vehicle control system 21) using a model. This represents aA measure of the degradation that the traction battery 12 has experienced due to driving operation and the operation of the auxiliary consumers 15 since a last update. The degradation is then subtracted from a last known value for the state of health (SOH) to obtain an updated SOH estimate 62 of the health status of the traction battery 12. Based on the updated SOH estimate 62, an SOH comparison 63 is now carried out with a target value. If the updated estimate falls below the target value, measures must be taken to reduce the degradation of the traction battery 12. If the updated estimate exceeds the target value, the operation of the trolleybus 10 can be optimized more with regard to other criteria (e.g., total energy consumption, comfort functions, etc.). Ultimately, the result of the SOH comparison 63 therefore has an impact on driving operation; it is therefore transmitted accordingly to the vehicle control system 21.28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Trademark Attorneys AG In the highest level 40.3, an effective SOH measurement 71 is carried out, with a stationary trolleybus 10, using the measuring device 50. The remaining battery capacity is determined and from this, by comparing it with the initial capacity, the current effective value of the SOH ^ ^ ^ The result of this SOH measurement 71 is used, on the one hand, for the model calibration 72 of the model used for the degradation assessment 61 and, on the other hand, for the update 73 of the reference curve, which is required in the SOH comparison 63. The reference curve is derived from the following variables: - a given health status ^ ாை^ at a given mileage ^ ாை^ ; - the current mileage ^; and - the last determined effective SOH ^ ^ ^ .at a mileage ^ ^ In the simplest case, a Decrease of the SOH is assumed, so that the target value ^ ௗ at the mileage ^ as follows: ^ ^ ప െ ^ ^ௗ ൌ ^ ^ ప െ ^ ாை^ െ ^ ^ ^^ െ ^ ^ ^ ^Ǥ With each new effective the acceptance is therefore re-determined and the aforementioned update 73 is carried out. In principle, the acceptance can also be parameterized using a non-linear relationship. The degradation estimate 61 is based on vehicle usage data. In the illustrated embodiment, the degradation estimate results from the course of the state of charge of the drive battery 12 since the last determination. The degradation can then be calculated using an aging model. A suitable tool for this purpose is, for example, the freely available tool OpenSesame, which enables the simulation of Li-ion batteries (M. Beyeler, SS Bhoir, S. Broennimann, Y. Moullet, Open-SESAME-Battery, 28115 WO PR / zol Keller Schneider Patent- 08. Februar 2024 und Markenanwälte AG V1.1, https: / / gitlab.ti.bfh.ch / oss / esrec / open-sesame (2022)).In addition to information about the battery (capacity, cell chemistry) and the starting state (SoC, SOH), time series with values ​​of the state of charge (SoC), the input and output power (which, with sufficient time resolution, can generally be calculated from the SoC curve), and the cell temperature are incorporated into the simulation. The result includes, among other things, an estimate of battery degradation for the given cycle. The quality of this estimate is iteratively improved based on model calibration 72: For this purpose, the estimated degradation ^^ is calculated for each SOH measurement 71. ^^ ^ െ ^ ^ െ ^ ^^ ^ ^ ^ seit the previous SOH measurement (as the sum of the estimated degradations) with the difference ^ ^ ^^ െ ^^ െ ^ ^ ^^^ between the effectively measured SOH values. For this purpose, a degradation error is first calculated as follows: ൫ ^ ^ ^^ െ ^^ െ ^ ^^^^൯ െ ^^ ^^ ^ ^ ^^ ^ ^ ^^^^ ^^^ ൌെ ^ െ ^ ^ ^ ^^ ^ ^^ െ ^^ െ ^ ^ ^^^^ Ǥ This error value ^ ^ at the time ^ ^ to be determined as follows: ^ ^ ൌ ^ ^ ^ ெெ ^^ ௗ^^ ^^^ ^ ^ ் ಾಾ σ ^ ^ ୀ^ ^ ௗ^^ ^^^ ή ^^ ^^^^ ^^^ െ ^ ^^^^ ^^ െ ^^^^ , where The correction factor is used to correct the result ^^ ^^ ^ െ ^ ^ െ ^ ^^ ^ ^ ^^ derDegradation estimate 61 is multiplied by this before summation with the predetermined degradations to obtain the updated SOH estimate 62. At the same time, with each new SOH measurement 71, the sum of the degradations is reset to zero, i.e., further degradation estimates 61 are made based on the last determined effective SOH value. The results from the SOH comparison 63 are now taken into account within the framework of vehicle control 21. In particular, depending on the requested power 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG (drive power and power for other consumers), it is determined which proportion of the power is taken from the drive battery 12 and which is taken directly from the overhead line via the pantograph. If the overhead line is not available in a certain operating section, additional operating parameters (e.g., the power of the auxiliary consumers 15) can be varied.However, these operating parameters can also be adjusted if the overhead line is available. The well-known adaptive ECMS approach is used to consider the results from the SOH comparison 63 (L. Guzzella, A. Sciarretta, Vehicle Propulsion Systems, Springer-Verlag Berlin Heidelberg, 2013; A. Sciarretta, L. Guzzella, Control of hybrid electric vehicles, IEEE Control Systems Magazine 27 (2) (2007) 60-70; L. Serrao, S. Onori, G. Rizzoni, ECMS as a realization of Pontryagin's minimum principle for HEV control, in: 2009 American Control Conference, 2009, pp. 3964-3969). The approach was developed for vehicles with hybrid-electric drive, but can also be used for vehicles with a vehicle-mounted drive battery and an additional source of electrical energy (e.g. from a fuel cell or overhead line), as shown below.In this approach, the Hamiltonian function ^^ή^ is derived by applying Pontryagin's minimum principle to the optimal control problem. This function depends on the values ​​of two so-called equivalence factors ^. క ^^^ and ^ ஏ ^^^, which represent the value of the two states SOC and SOH relative to energy consumption. The value of ^ ஏ^^^ is determined in the second level 40.2, as described below. Based on the values ​​of the two equivalence factors, the distribution between battery and overhead line current is determined, which results in an optimal balance between the lowest possible total energy consumption and the maximum battery lifetime, taking into account the target battery lifetime. The distribution is achieved by minimizing the local Hamiltonian function ^^ή^. The specific form of the Hamiltonian function depends on the respective vehicle and drive. An example of the illustrated embodiment is given below. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG The operating strategy is now adjusted using a PI controller by ^ క ^^^ is adjusted depending on the current value ^^^^ of the SOC: If ^^^^ a constant reference value ^ ^^^exceeds the reference value, the operating strategy favors discharging the traction battery 12, while if the reference value is undershot, electrical energy is preferentially drawn from the overhead line. For this purpose, the equivalence factor for the SOC is calculated as follows: ^ క ^ ^ ^ ൌ ^^^^ െ ^ ^ ^ ^ ǡ Ǥ Thereby of the PI controller. This can depend on the SOC to determine the aggressiveness of the controller near the SOC- while in the SOC ranges in between is reduced. The ^ should be chosen slower than a corresponding time constant of the expected feed-forward component ^ కǡி ൫^ ^ ^ ^ ൯ is ி ஏadded to compensate for the influence of the battery health control loop (Level 40.2) on the operating strategy control loop (Level 40.1). Simulations have shown that the feedforward component can be defined, for example, as follows: ^ కǡிி ^^ ஏ ^ ൌ ^Ǥ^ ή ^^ ି^^ ή ^ ஏ ^ ͺ^^Ǥ This definition reflects the fact with a high value for the equivalence factor ^ ஏ for battery health also a higher value for the other ^ క is necessary to ensure charge-maintaining operation of the drive battery. At the first level 40.1, the controller must be able to respond to changes in the requested power; the update interval can be set accordingly, for example, to 1 s. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG The control loop for battery health in the second level 40.2 serves to keep the SOH trajectory of the drive battery 12 close to the reference curve. At this level, an update after each driving phase is sufficient; an update therefore usually occurs every few hours. Due to the slow battery aging, deviations from the SOH reference trajectory also occur comparatively slowly. With the aforementioned model for battery aging, the trajectory ^ ^ of the SOC since the last update of the value ^^ ^ for the degradation assessment 61. This is adjusted with the mentioned correction factor ^ ^ multiplied. The value ^ ^for the updated SOH estimate 62 is then obtained by summing 61. The deviation of the current SOH estimate 62 from the corresponding reference value is given as ^ ஏ ^ ^ ^ ൌ ^^^^ ^ ^ ^ െ ^ ^^ ^ ^ ǡ which is converted into a PI controller as follows ^ ^ ^ ஏ ^^^ ൌ ^ ^ ஏ ^^ ஏ ^^^ ^ ^^ ஏ ^^^ ^^ ௧^ ^^^ െ ^ ௧^ ^^ െ ^^^ ^Ǥ Again and ^ the time constant. It has been shown that the sensitivity of the operating strategy with respect to a change in the equivalence factor ^ ஏ is best represented on a logarithmic scale, accordingly the result of the second level 40.2 at a time ^^^^ as defined as follows: ^ ஏ ^^^ ൌ ^^ ௨ಇ^^^Ǥ 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG Within the framework of a simulation study, the inventive method was tested using a trolleybus that, with the help of its pantograph, can draw power from a DC overhead line and simultaneously store electrical energy in its traction battery 12. Due to its unidirectional DC-DC converter, feeding energy back into the grid is not possible. The drive train is therefore topologically very similar to a series hybrid drive train, where a motor-generator plays the role of the grid for recharging the traction battery. The trolleybus in question has the following characteristics: Vehicle length 18.7 m As part of the simulation study, the bus's operation on various routes was examined in realistic, everyday driving missions. The underlying model is described in the article by F. Widmer, A. Ritter, P. Duhr, CH Onder, "Battery lifetime extension through optimal design and control of traction and heating systems in hybrid drivetrains," eTransportation 14 (2022) 100196, https: / / doi.org / 10.1016 / j.etran.2022.100196, Chapter 2. The power balance takes into account the components schematically shown in Figure 3: - a feed point 1, from which the trolleybus 10 is supplied with power via a grid 2; - the DC-DC converter 18; 28115 WO PR / zol Keller Schneider Patent- February 8, 2024 and Trademark Attorneys AG - the power distributor 16, through which the power flows between the individual sources and consumers; - the traction battery 12; - the drive unit 11 and the auxiliary consumers 15; and - the braking resistor 17.The arrows indicate the potential energy flow. This results in the following power balance: ^. ^ǡ^௨௧ ^ ^ ^ ^^ ^^ ^ ^ ^ ൌ ^^^^ ^ ^ ^ ^ ^^^^ ^ ^ ^ ǡ (1) where ^^ǡ^௨௧ ^^^ is the power at the output of the DC-DC converter 18, ^ ^ ^^^ the discharge power of the drive battery 12, ^^^^ ^^^ the required power of the drive unit 11 and the auxiliary consumers 15 and ^^^^ ^^^ the power consumed in the braking resistor 17. Assuming that ^^^^ ^^^ resulting from the driving mission and other requirements (HVAC, etc.), there are two degrees of freedom, namely the power consumption from the traction battery 12 and the energy to be dissipated in the braking resistor 17. For the application to the trolleybus, therefore, the parameters ^^ ^^^ and ^^^^^^^ considered as input variables. The power of the braking resistor ^^^^ ^^^ is not negative: ^ ^^^ ^ ^ ^ ^ ^Ǥ The battery can store energy As stated above, the lithium titanate oxide (LTO) battery cells of the traction battery 12 of trolleybus 10 have a capacity of 23 Ah. The traction battery 12 consists of 1296 cells in a 324-series-4-parallel configuration. The battery model used includes an ideal voltage source ^ ^ǡை^ ^^^ in series with an internal resistance ^ ^ ^^^. The value of both the open-circuit voltage and the internal resistance 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Trademark Attorneys AG vary with the state of charge (SOC). The corresponding curves are derived from the aforementioned OpenSESAME package, scaled for the specific battery configuration. The current required to deliver a specific discharge power^ ^^^^ is required, is as follows: ^ ^ ^ ^ ^ǡை^ ^^^ െ ^ ^ ^ǡை^ ^^^ଶ െ ^ ή ^^ ^^^ ή ^ ^ ^^^ ^ ^ ൌ ʹ Ǥ The charge level current charge and maximum charge capacity ^ ^ ^ ^ ^ . It results from the following differential equation: ^ ^ ^ ^ ^ ^ ^ ^ ^ ^^^^ ^ ^ ^^ ൌ െ ^ ^^^^ ൌ െ ^ ^^^ wobei ^ ^ ^ ^ wie oben In addition to the decreasing capacity, the aging of the drive battery 12 also manifests itself in an increased internal resistance. This effect is also taken into account in the simulation study. Instead of calculating the SOR curve separately, however, it is assumed here for the sake of simplicity that the SOR is linearly related to the SOH. This linear relationship was determined based on previous experiments with the aforementioned OpenSESAME package. However, the aforementioned OpenSESAME package also easily allows for the calculation of the internal resistance, so that this parameter can be integrated into the simulation or an operating method according to the invention if necessary.Ultimately, OpenSESAME enables the determination of the degradation of the traction battery over a specific cycle, depending on the charge / discharge power, SOC, and cell temperature curves. The battery's usage cycle is divided into an equivalent set of subcycles, and the degradation is determined as a superposition of cycle and calendar aging. 28115 WO PR / zol Keller Schneider Patent Attorneys and Trademark Attorneys AG February 8, 2024 The simulated trolleybus 10 includes a powerful system for temperature control of the battery cells. With currently available systems, the cell temperature can be maintained within a relatively narrow range, e.g., between 10 °C and 25 °C, although the aging of LTO cells is not very sensitive to different temperatures within this range. For a given outside temperature, the resulting cell temperature can also be maintained within a narrow window of approximately 2-3 °C.Changes in battery temperature resulting from varying outside temperatures can thus be considered as external disturbances and integrated into the driving mission. To calculate the chronological aging of the traction battery 12, idle times of the trolleybus 10 outside of the driving missions, during which the vehicle is usually in the depot, are modeled with a constant, specified temperature. In an extension of the described method, however, the temperature management of the battery can, for example, be integrated into the operating procedure with additional control parameters. The grid 2, which transmits electrical power from the feed point 1 via the pantograph 13 to the DC-DC converter 18 of the trolleybus 10, is used as the ideal voltage source. ^^ǡை^ in series with a line resistance ^ ^ modeled. Accordingly, the power reaching the DC-DC converter 18 is: ^ ^ ^ǡ^^ ^ ^^ ൌ ^^ ^ ^ ^ െ ^ ଶ ή ^^ ^ ^ ^ଶ ǡ where ^^ ^^^ the performance The mains current ^ ^ ^^^ is limited to prevent excessive wear due to heat build-up between the pantograph 13 and the overhead line. The upper limit ^ ഥ ^ depends on the vehicle speed ^^^^ and compliance with this is ensured by the vehicle control 21. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG The route may include sections where no overhead line is available. This is taken into account with a two-valued indicator function ^^^^, which takes the value 1 if an overhead line is available and the value 0 if this is not the case. This results in the following conditions: ^^ ^^^^^ ^ ^^^^ ή ^ ഥ ^൫^^^^൯ ή ^^ǡை^ǤThe unidirectional The power available at its output is given by ^ ^ǡ^௨௧ ^ ^ ^ ൌ ^^ ή ^^ǡ^^ ^ ^ ^ െ ^^ǡ (6) where the constants ^ ^ and^ ^The converter efficiency and the no-load losses, respectively. According to the ECMS approach, the optimal power split is calculated by minimizing the so-called Hamiltonian function, which results from the application of the Pontryagin Minimum Principle (PMP). To derive the Hamiltonian function, as mentioned above, only the state variables SOC and SOH are considered within the model, so that operating conditions that are particularly detrimental to the internal resistance are not specifically prevented. However, due to the aforementioned correlation with operating conditions that are detrimental to the capacity-related SOH, this does not usually lead to problems. For the state update function of the SOH, the time-based instantaneous degradation ^^^^ is introduced as follows: ^ ^ ^ ^^^^ ൌ െ^^^^ǤIn the article F. Widmer, A. Ritter, P. Battery lifetime extension through optimal design and control of traction and heating systems in hybrid drivetrains, 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Markenanwälte AG eTransportation 14 (2022) 100196, https: / / doi.org / 10.1016 / j.etran.2022.100196, describes how the cycle-based approach of OpenSESAME can be transformed to obtain the following estimate for the degradation: ȁ^^^^ȁ ή ^^ ^^^^ ^ ^ ൫^^^^൯ ή ^ ್ ^^ ^^ ௬^ǡ^ ^^ ^௬^ǡ^ ^௬^ǡక ^௬^ǡణ൫^ ^ ^൯ ή ^ ^௬^ǡఋ ൫^^൯ ή ^ ^ ൫^^൯ ή ʹ ^^^^ ^^ ^௬^ǡ^ a reference cycle depending on the SOC, the cell temperature, the (depth of discharge, DOD) ^ ^ ^ ^ and the C-rate ^ ^ ^ ^ the drive battery.Similarly, the scaling factors ^ ^^^ǡక and ^ ^^^ǡణ್ the reference value ^ ^^^ǡ^ for calendar aging in the SOC and the Cell temperature is scaled. The C-rate is defined relative to the nominal battery capacity, i.e. ^ ^^^ ൌ ^^^^^ ^ Ǥ When determining the optimal at the time ^ can ^^^^, ^ ^ ^^^ and ^(t) can be considered as given. In addition, as also described in the aforementioned article by F. Widmer et al., a constant value ^ can be used for the depth of discharge (DOD) as a simplifying assumption. ^ be chosen. Thus, the following two new variables can be introduced: ^^ ^ ൫^^^^ǡ ^^^^ǡ ^^^^൯ ൌ ^ ^ή^ ή ^ ^ή^ ή ^ ^^ ^௬^ǡ^ ௗ^^ ^ ^௬^ǡక ^௬^ǡణ್ ^௬^ǡఋή ή ʹ ^^^^ ǡ From this 28115 WO PR / zol Keller Schneider Patent Attorneys and Trademark Attorneys AG February 8, 2024 The aforementioned power distribution is performed by minimizing the Hamiltonian function. However, the cycle-based OpenSESAME models are used directly to estimate degradation at the second level. (The OpenSESAME models are also used in the simulation study presented here to calculate the actual battery aging.) The following results for the state update function: ^ ^^^^^ ^ ^^^^ ൨ ൌ ^െ ^ ^^^ ^Ǥ With regard to the The Pontryagin minimum principle for the Hamilton function yields the following: ^ ^ ^ή^ ൌ ^ ^ ^ െ ^ǡ in favor of Dependencies were omitted. The equivalence factors ^ క ^^^ and ^ ஏ^^^ are introduced as negative values ​​of the corresponding co-state variables ^^^^. This results in the following minimization problem: ^^^^^^ ^ ^ ^ǡ ^^^^ ൭ ^^ ൫ ^^^^ ǡ ^^ ǡ ^^^^ ൯ ^ క ^ ^ ή ^ ^ ^ ^^ ǡ ^ǡ ^ǡ^^ ^ ^ ஏ ή ^^ ^^ ǡ ^ǡ ^ ^ ǡ ^ǡ^^^ǡ where the The result corresponds to the optimal power distribution. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG The arguments arise from the above equations: - ^^ ൫ ^^^^ ǡ ^^ ǡ ^^^^ ൯ from equations (1), (4) and (6); - ^ ^ ^ ^^ ǡ ^ǡ ^ǡ^^ from equation (3); - ^^ ^^ ǡ ^ǡ ^ ^ ǡ^ǡ^^ from equation (8). In addition, the minimization is subject to the conditions according to equations (2) and (5),^ ^^^ ,^^ and the upper limit^ ഥ ^ for the power supply from the grid are derived from the driving mission and the equivalence factors ^ క and ^ ஏ are set by the corresponding controls. The minimization can be solved on a commercially available personal computer with an Intel Core i7-8565U processor in approximately 20 ms. The described method can therefore be readily used to regularly determine the power split at the lowest level 40.1 of the inventive method (update rate in the range of 1 s). As described above, the selected operating strategy depends primarily on the deviation of the estimated health status of the traction battery from a target state: If the health status is better, an operating strategy is selected that favors minimizing the total energy; if the health status is worse, an operating strategy is selected that favors maximizing the service life of the traction battery. Figures 4A and 4B show the resulting power distribution for an operating strategy with a high weighting of a low total energy consumption (Fig. 4A, equivalence factor ^ ஏൌ ͵Ǥʹ ή ^^ ^^ ) and with an operating strategy with a high weighting of a high battery life (Fig. 4B, equivalence factor ^ ஏ ൌ ͵Ǥʹ ή ^^ ^ଷ ). Shown are each depending on the required power ^^^^ axis, in kW) the power values ^^ for the energy supply from the drive battery or the energy storage in the drive battery (medium grey point cloud 91.1, 91.2), the performance values ^^ǡ^௨௧ for the energy supply from the overhead line (dark grey point cloud 28115 WO PR / zol Keller Schneider Patent- 08 February 2024 and Trademark Attorneys AG 92.1, 92.2) and the performance values ^^^^ for the dissipation in the braking resistor (light grey point cloud 93.1, 93.2), each on the vertical axis, in kW. The finely dashed horizontal lines represent the maximum power from the primary power source. ഥ ^^at 50 km / h, the coarse dashed lines represent the requested power ^^^^. It is immediately clear from the illustrations that the braking resistor is not used in the energy-oriented operating strategy (Fig. 4A), whereas it is used in the battery health-oriented operating strategy (Fig. 4B) to avoid excessively high charging power of more than approximately 180 kW. In the energy-oriented operating strategy (Fig. 4A), energy is drawn from the overhead line in a uniform manner, with maximum powers that are significantly lower than those in the battery health-oriented operating strategy (Fig. 4B). This avoids energy losses due to the significantly higher resistance of the overhead line compared to the internal resistance of the traction battery. If the overhead line can supply sufficient power to cover demand in the battery health-oriented operating strategy (here up to approximately130 kW), the energy requirement is primarily covered by the overhead line and the traction battery is used primarily when higher power is required. In the example shown, the battery health-oriented operating strategy leads to a reduction in battery degradation of 31.4% and to an increased overall energy consumption of 5.1%. The operating method according to the invention was tested in several simulations. One of the simulations was based on 16 different driving missions on four different bus routes, the relevant characteristics of which were determined by measurements on a real vehicle. The routes had different availability of electrical energy from overhead lines, between approximately 50% and 100% of the route traveled. The routes were also characterized by different elevation differences and maximum permissible speeds. The selection of driving missions also covered all 28115 WO PR / zol Keller Schneider Patent- 08.February 2024 and Markenanwälte AG seasons and thus different uses of the HVAC system. The missions were defined by data with a time resolution of 1 s. Figure 5 shows the results of a lifetime simulation, comparing the results of the inventive method with those of a method without regular recalibration. The lifetime simulation was based on the following scenario: - SOH measurements are carried out after 180 missions, i.e. after approximately 6 months. - The SOH measurements are superimposed with Gaussian-distributed noise (mean 0, standard deviation ^ ൌ ^Ǥ^^^). - The target health status after 1.2 million kilometers of travel is 80%. - In an initial operational phase, the simulated trolleybus is used on three routes, with a network coverage of 90-100%. After approximately 0.3 million km, the bus is used for 0.1 million km on a route with a network coverage of only approximately 50%.The previous route selection is then used again. Figure 5 shows the results of a simulation with active recalibration according to the invention and, in comparison, the results of a simulation without regular recalibration. The three diagrams (a) - (c) show the course of various parameters as a function of the driving distance d (in million km). The top diagram (a) shows the battery health (SOH) value ^. The circles correspond to the periodic SOH measurements 81. The curve 82 (solid line) shows the actual battery health curve with active recalibration. It should be noted that in the real application of the operating method according to the invention, this value is only effectively known at the times of the SOH measurements 81. Between these times, operation is based on an estimate 83 (dashed line).The reference curve 84 (dotted line) is updated with each SOH measurement. As can be clearly seen in Figure 5 (a), all these curves are close together, and at the target service life of the traction battery of 1.2 million km, the traction battery has a health status of 80.3%, very close to the target of 80%. Without recalibration, the battery health in the presented case is permanently underestimated because the degradation is overestimated. This is because the degradation is then estimated according to approximation (7), which does not take into account effects dependent on the depth of discharge (DOD). In addition, the most influential scaling factor ^. ^௬^ǡ^by a piecewise linear function for positive and negative rates of C. This results in an actual (unknown in the real case) curve 86 (solid line) and the battery health at 1.2 million kilometers of travel is over 85%. Analogously, the total energy required was 8.8% higher than in the case of the inventive operating method. The estimate 87 (dashed line) and the reference curve 88 (dotted line) are significantly lower in this case than in the case with recalibration, but are unable to compensate for the estimation errors that continuously accumulate during the simulation, and the difference to the actual SOH increases continuously. The bottom panel (c) of Figure 5 shows the curve 101 of the parameter ^ with recalibration and the curve 102 of the parameter ^ without recalibration. It can be seen that it changes at values ​​of 0.6-0.8 and the estimated degradation is adjusted downwards accordingly. In contrast, in a simulation without recalibration, the parameter ^ remains (trivially) at a value of 1. The middle panel (b) of Figure 5 shows the curves of the equivalence factor ^. ஏfor battery health (in J) with recalibration (curve 111) and without recalibration (curve 112). Due to the correction of the overestimated battery degradation, significantly lower values ​​of the equivalence factor for battery health can be selected in the first case than in the case without correction. In the second case, great importance is consistently attached to battery protection; in the stressful operating interval between 0.3 and 0.4 million km, the increased battery aging cannot be fully compensated within the interval; however, compensation is reliably carried out in the model with recalibration 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Markenanwälte AG after returning to the previous route selection.This shows that when a fleet with several vehicles operated according to the invention is included, the different loads associated with operation on different routes can be balanced out in the long term through targeted route allocation of the vehicles. Further simulations have shown that the method according to the invention, due to the regular determination of the effective health status of the drive battery, the recalibration of the degradation estimates and the recalculation of the reference curves, also reliably takes into account other influences, e.g. incorrect estimations of the calendar degradation or a sudden drop in the health status, e.g. due to longer downtimes or improper handling of the battery. The invention is not limited to the exemplary embodiment shown. The method can therefore be combined with other control parameters, e.g. relating to charging in the stationary state (e.g.in the depot between driving missions) and / or the control of the HVAC system or other components. Furthermore, the method is applicable not only to trolleybuses, but also, for example, to hybrid vehicles or purely electric vehicles with stationary charging processes. In summary, the invention creates a method for optimizing the service life of a vehicle's traction battery, which ensures that a specified service life of the traction battery is achieved with minimized adverse effects on overall energy consumption and / or user comfort. 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG.

Claims

Patent claims 1.^ Method for optimizing the service life of a drive battery of a vehicle, comprising the following steps: a) determining an effective state of health of the drive battery with a first periodicity; b) determining an estimated state of health of the drive battery with a second periodicity, starting from a last determined effective state of health, taking into account usage data of the vehicle since a last determination of the estimated state of health, by applying an aging model for the drive battery; c) operating the vehicle based on the estimated state of health and a current drive energy requirement, wherein at least one battery load-relevant control parameter is redetermined with at least a third periodicity; wherein the aging model for the drive battery is calibrated after determining the effective state of health. 2.^ Method according to claim 1, characterized in that the control parameter is determined based on a deviation of the estimated state of health from a reference curve. 3.^ Method according to claim 2, characterized in that the reference curve is determined such that, for a predetermined mileage, the drive battery has a predetermined state of health. 4.^ Method according to claim 2 or 3, characterized in that the reference curve is recalculated after the actual state of health has been determined. 28115 WO PR / zol Keller Schneider Patent- und Markenanwälte AG February 8, 2024. 5.^ Method according to one of claims 1 to 4, characterized in that the usage data taken into account for determining the estimated state of health comprise a battery charge level history since the last determination. 6.^ Method according to one of claims 1 to 5, characterized in that, for calibrating the aging model, the estimated state of health of the drive battery at the time of determining the actual state of health is compared with the determined actual state of health. 7.^ Method according to one of claims 1 to 6, characterized in that, for determining the control parameter, an optimization is carried out with a view to maximizing a first target variable corresponding to a service life of the drive battery. 8.^ Method according to claim 7, characterized in that, with a view to maximizing the first target variable, an influence of operating parameters on the state of health of the drive battery is determined based on the vehicle usage data and the recorded effective state of health. 9.^ Method according to one of claims 1 to 8, characterized in that, to determine the control parameter, an optimization is carried out with a view to minimizing a second target variable corresponding to an energy consumption of the vehicle. 10.^ Method according to claims 2, 7 or 8, and 9, characterized in that, depending on the deviation of the estimated state of health from the reference curve, a weighting of the first and second target variables is adjusted in the optimization. 11.^Method according to one of claims 1 to 10, wherein a first period of the first periodicity is longer than a second period of the second periodicity and wherein the 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 and Markenanwälte AG. second period is longer than the third period of the third periodicity and wherein in particular - the first period is 14 days to 400 days, - the second period is 1 hour to 24 hours, and / or - the third period is 0.1 to 10 s. 12.^Method according to one of claims 1 to 11, characterized in that the vehicle is a battery-electric vehicle. 13.^Method according to claim 12, characterized in that the vehicle comprises a current collector to absorb electrical energy from an overhead line during driving operation. 14.^Method according to one of claims 1 to 13, characterized in that the vehicle is used in scheduled service and forecast driving data of a line being traveled or to be traveled is taken into account when determining the control parameter. 15.^Method according to one of claims 1 to 14, characterized in that the control parameter comprises one or more of the following parameters: a) a state of charge; b) a charging power; c) a reference power from the drive battery; d) a maximum acceleration of the vehicle; e) a maximum recuperation current of the vehicle; f) a consumption power for a braking resistor; g) an operating parameter of a heating, ventilation and / or air conditioning system of the vehicle; 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG. h) an operating parameter of an interior lighting and / or an interior information system and / or other auxiliary consumers of the vehicle; i) an operating parameter of a thermal management system for the drive battery. 16.^An arrangement for operating an electrically powered vehicle, comprising a) the electrically powered vehicle with a drive battery for driving at least one drive unit of the vehicle; and b) a measuring device for determining an effective state of health of the drive battery of the vehicle; wherein the vehicle comprises the following: c) a processing device for determining an estimated state of health of the drive battery; d) a calibration device; e) a vehicle control system for operating the vehicle;wherein - the processing device is configured to receive the most recently recorded effective health status from the measuring device, to receive vehicle usage data from the vehicle controller, and to determine the estimated health status based on the effective health status and taking into account the usage data since the last determination of the estimated health status by applying an aging model for the drive battery; - the calibration device is configured to calibrate the aging model based on the recorded effective health status received from the device; and 28115 WO PR / zol Keller Schneider Patent- 08. February 2024 und Markenanwälte AG; - the vehicle control system is configured to determine at least one battery load-relevant control parameter based on the estimated state of health and a current drive energy requirement. 28115 WO PR / zol Keller Schneider Patent- und Markenanwälte AG February 8, 2024

Images