Hybrid energy store with predictive maintenance

Abstract

The invention relates to a method for predicting the lifetime of a hybrid energy store (1), comprising the following steps: providing a hybrid energy store, comprising a high-power energy store (2), a long-term energy store (3), and a converter (4); applying a load (5) with a load profile to the hybrid energy store; calculating a moving average of the load profile, wherein a difference between the load profile and the moving average of the load profile forms a manipulated variable for the converter (4), wherein the converter (4) distributes the load to the high-power energy store (2) and / or the long-term energy store (3) on the basis of the manipulated variable, wherein the long-term energy store (3) is discharged substantially constantly within a discharge cycle; comparing an instantaneous discharge behaviour of the long-term energy store (3) with a known discharge behaviour and predicting the lifetime of the long-term energy store (3) on the basis of the comparison.

Description

[0001] Hybrid energy storage system with predictive maintenance

[0002] The invention relates to a method for predicting the lifetime of a hybrid energy storage system, comprising a high-performance energy storage system, a long-term energy storage system and a converter.

[0003] Background of the invention

[0004] Hybrid energy storage systems are systems that combine high-performance energy storage and long-term energy storage to take advantage of both technologies.

[0005] For example, a supercapacitor can serve as a high-performance energy storage device, and a conventional battery as a long-term energy storage device. These systems are not limited to battery and supercapacitor technologies but can also include fuel cells, electrolytic capacitors, and other technologies. The main focus is on combining the complementary properties of two or more different technologies. In the system example mentioned above, the battery can provide the capacity for a long-lasting energy supply, while the supercapacitor can cover peak loads. Examples of this include starting machines, rapid load changes, or recuperation processes such as braking.

[0006] The main advantage of this combination lies in the improved power density and more efficient energy handling. Long-term energy storage devices, such as batteries, have a high energy density, meaning they can store a lot of energy in a given mass, but they are less efficient when it comes to quickly supplying or absorbing large amounts of energy at high power. In this case, heat is also generated, which accelerates battery aging. High-performance energy storage devices, such as supercapacitors, on the other hand, offer a high power density, making them ideal for quickly supplying or absorbing large amounts of energy, as required during power peaks. However, they have the disadvantage of not having as high an energy density.

[0007] To effectively combine the two technologies in a hybrid energy storage system, a bidirectional converter is used. This acts as a bridge between the long-term energy storage and the high-performance energy storage, regulating the energy flow between them to ensure optimal performance. The converter ensures that energy is quickly provided by the high-performance energy storage when high power is needed immediately (filtering power peaks), while the long-term energy storage provides a continuous supply with minimal power variation. The lifespan of long-term and high-performance energy storage depends on various factors. In the case of a battery, the type of battery, its manufacturing process, its use, environmental conditions, and the number of charge and discharge cycles play a crucial role.Therefore, predicting the lifespan of the hybrid energy storage system requires a reliable prediction of the lifespan of its components.

[0008] Prior art includes, for example, CN 112 564252 A. This document describes a hybrid energy storage system and a model-based predictive energy control method. The hybrid energy storage system combines batteries and supercapacitors, with the battery connected to a DC bus via a DC / DC converter and the supercapacitor connected directly in parallel. The system aims to increase efficiency and minimize energy losses by enabling optimized energy distribution between the battery and the supercapacitor. The invention includes models for predicting losses and a comprehensive optimization function for controlling the system. System efficiency is improved by a model-based predictive energy control method; however, a detailed analysis or forecast of battery life is not described.

[0009] US Patent 2023 / 241984 AI deals with a hybrid energy storage system and a method for energy control. The focus is on the efficient management and distribution of energy, specifically with regard to handling and controlling power peaks within the system. A system is described that is capable of predicting potential power peaks and reacting accordingly to ensure optimal energy distribution.

[0010] EP 3 772 148 Al describes a system for power distribution and control in aircraft, which includes a hybrid energy storage system (HESS). The HESS contains at least two energy storage subsystems with different power and energy density characteristics. The system's power management controller is configured to interact with the HESS and perform model predictive controls based on the aircraft engine's power demand to dynamically adjust electrical energy flows.

[0011] The CN 116647040 A generally shows a controllable electricity distribution system with carbon emissions that can be automatically adjusted based on electricity grid data.

[0012] Mesbahi et al., “Optimal Energy Management for a Li-Ion Battery / Supercapacitor Hybrid Energy Storage System Based on a Particle Swarm Optimization Incorporating Nelder-Mead Simplex Approach,” IEEE Transactions on Intelligent Vehicles, IEEE, Vol. 2, No. 2, June 1, 2017, pp. 99–110, DOI: 10.1109 / TIV.2017.2720464, describes minimizing battery stress to improve the lifetime of a hybrid energy storage system (HESS). This is achieved by coupling a rule-based approach, based on knowledge of battery and supercapacitor efficiency, with a hybrid particle swarm-Nelder-Mead optimization algorithm.

[0013] Rade, “Design and Development of Hybrid Energy Storage System for Electric Vehicle”, 2018 International Conference on Information, Communication, Engineering and Technology (ICICET), IEEE, August 29, 2018, pages 1-5, DOI: 10.1109 / ICICET.2018.8533757 describes design and dimensioning calculations of hybrid energy storage systems based on theoretical concepts.

[0014] Brief description of the invention

[0015] The object of the present invention is to provide a method for predicting the lifetime of a hybrid energy storage device, which alleviates or eliminates at least some disadvantages of the prior art.

[0016] This problem is solved by a method according to claim 1. Preferred embodiments are specified in the dependent claims, the description, and the drawings.

[0017] According to the invention, a method for predicting the lifetime of a hybrid energy storage device is provided. The method comprises the following steps:

[0018] - Provision of a hybrid energy storage system comprising a high-performance energy storage system, a long-term energy storage system, and a converter,

[0019] - Applying a load with a load profile to the hybrid energy storage system,

[0020] - Calculating a moving average of the load profile, wherein a difference between the load profile and the moving average of the load profile forms a control variable for the converter, wherein the converter distributes the load to the high-performance energy storage and / or the long-term energy storage based on the control variable, wherein the long-term energy storage is discharged at a substantially constant rate within a discharge cycle,

[0021] - Comparing the current discharge behavior of the long-term energy storage system with a known discharge behavior and predicting the service life of the long-term energy storage system based on this comparison. A significant advantage of the method according to the invention is that the long-term energy storage system is discharged essentially constantly within a discharge cycle through its interaction with the high-performance energy storage system. This discharge behavior can, for example, approximate manufacturers' cycle tests and / or specific standards for testing the cells, which is why the manufacturers' reliability tests can be used to determine the service life of the long-term energy storage system. It is also possible to use in-house cycle tests that were carried out according to the same specific standards. This also allows for better maintenance planning.The use of the hybrid energy storage system now allows for the establishment of a highly reliable system with predictive maintenance capabilities. The high-performance energy storage system is monitored in a manner familiar to specialists.

[0022] The high-performance energy storage device could be, for example, a supercapacitor, while the long-term storage device could be a battery. However, other combinations of high-performance and long-term energy storage devices are also possible. Hybrid energy storage systems can utilize electrolytic capacitors, supercapacitors, electric double-layer capacitors (EDLCs), batteries, and fuel cells. For instance, a fuel cell can be used as a long-term energy storage device in combination with a faster storage device with higher power density, such as a supercapacitor or a battery. Another example would be the combination of an electrolytic capacitor with a supercapacitor, where the supercapacitor would serve as the long-term storage device.Furthermore, a first hybrid energy storage system can be used as a high-performance energy storage device, and a second hybrid energy storage device can be used as a long-term energy storage device for a third hybrid energy storage device. An example of this is the combination of an electrolyte capacitor and a double-layer capacitor (first hybrid energy storage device) with a double-layer capacitor and a battery (second hybrid energy storage device), which together form a third hybrid energy storage device. It is therefore evident that there are many combinations for realizing a hybrid energy storage system, up to and including a potentially multiple, interconnected hybrid energy storage system.

[0023] In a first step of the method according to the invention, a hybrid energy storage system is provided, comprising a high-performance energy storage system, a long-term energy storage system, and a converter. The high-performance energy storage system and the long-term energy storage system can be implemented, as described above, by various combinations of technologies known per se. The converter, also known as a DC-DC converter, is responsible for adjusting the voltage and controlling the energy flow between the two storage systems and the connected load. Since the high-performance energy storage system and the long-term energy storage system have different electrical properties, the converter ensures that the strengths of both energy storage systems can be used efficiently and function as a single unit.

[0024] Additionally, the converter's control system monitors and regulates the state of charge of both energy storage devices. Precise control ensures that the high-performance energy storage device is not overcharged and that the state of charge of the long-term energy storage device remains within safe limits. This is essential for preventing thermal and electrical overload, which could lead to accelerated aging or even safety hazards.

[0025] In the next step, a load is applied to the hybrid energy storage system, with the load having a load profile. This load profile is a measure of how much power the hybrid energy storage system must provide at a given time. The load profile can be, for example, white noise, a sinusoidal profile, a pulsed profile, a triangular profile, or a constant profile.

[0026] In a next step, a moving average of the load profile is calculated. The difference between the instantaneous load profile and the moving average of the load profile forms a control variable for the converter. Based on this control variable, the converter distributes the load to the high-performance energy storage device and / or the long-term energy storage device, whereby the long-term energy storage device is discharged at a substantially constant rate within a discharge cycle. This step of the method according to the invention allows a reliable comparison with known discharge behavior, since in routine discharge tests the long-term energy storage devices under test are generally discharged at a constant rate (e.g., according to a standard that must be adhered to).

[0027] In a further step of the inventive method, the instantaneous discharge behavior of the long-term energy storage device is compared with a known discharge behavior in order to predict its lifetime. The discharge behavior according to the invention thus closely resembles known cycle tests, which is why these reliability tests can be used to determine the lifetime of the long-term energy storage device.

[0028] In a preferred embodiment, the comparison step is performed while the load is applied to the hybrid energy storage system. The ability to predict the lifetime of a long-term energy storage system while it is under load offers both practical and economic advantages. Real-time assessment of the long-term energy storage system's condition allows for more accurate and timely decisions regarding maintenance and replacement, thus improving the reliability and safety of the entire hybrid energy storage system. Particularly in applications where long-term energy storage systems operate under variable or high loads—such as in electric vehicles, industrial machinery, or medical devices—early detection of potential failures is crucial to avoid unplanned downtime.

[0029] This predictive lifetime determination enables optimal utilization of the long-term energy storage system by operating it for as long as possible without risking a sudden loss of capacity. The ability to monitor wear and aging under real-world load conditions also supports more precise energy management. This allows power output to be adjusted according to the long-term energy storage system's condition, increasing efficiency and further extending its lifespan. Economically, this prediction helps reduce maintenance costs and optimize operational planning, as replacement of the long-term energy storage system and system failures can be proactively avoided. Thus, lifetime prediction under load contributes to maximum availability and efficiency of hybrid energy storage systems.

[0030] In a preferred embodiment, the known discharge behavior is based on test cycle data, in particular test cycle data from a manufacturer of the long-term energy storage system. Test cycles from a manufacturer of the long-term energy storage system used in the hybrid energy storage system provide a reliable source for predicting the lifetime of the long-term energy storage system.

[0031] In a preferred embodiment, the load has at least one load peak that is at least 5 times, preferably at least 10 times, and particularly preferably at least 20 times greater than the moving average of the load. This embodiment opens up a wide range of applications, especially in areas where high power surges are required or load requirements vary considerably. Applications in the fields of mobility, industry, infrastructure, and consumer electronics benefit significantly from this power flexibility. A hybrid energy storage system that can absorb extreme load peaks thus opens up new application possibilities in areas that require rapid and intensive energy output. In electromobility, for example, it enables powerful acceleration and the towing of heavy loads without overloading the battery.In robotics, it enables significantly faster and more precise movements, such as rapid acceleration and deceleration, while simultaneous energy recuperation increases efficiency. This leads to increased productivity without affecting battery lifespan. In rail and watercraft, it allows for efficient maneuvering and starting, and in industrial machinery like cranes, it provides the necessary power for lifting heavy loads. Thus, the hybrid storage system enables a wide range of applications that would be virtually impossible with conventional batteries alone.

[0032] In a preferred embodiment, the comparison step involves comparing a number of discharge cycles of the long-term energy storage device. This embodiment uses a particularly straightforward model that assumes long-term energy storage devices must be replaced after a certain number of charge and discharge cycles because they lose capacity with each cycle.

[0033] In a preferred embodiment, the ambient temperature and / or the cell temperature of the long-term energy storage device are taken into account during the comparison step. Ambient and cell temperatures have a significant impact on the performance, lifespan, and safety of long-term energy storage devices. In the case of batteries, strong temperature fluctuations can negatively affect the electrochemical processes within the battery and, in the long term, lead to performance losses and accelerated aging. Repeatedly subjecting a battery to excessively low temperatures can lead to a loss of capacity over the long term, as certain chemical processes, such as the formation of lithium metal deposits on the anode in lithium-ion batteries, are promoted, which permanently impairs the charging and discharging processes.Excessively high temperatures, on the other hand, can accelerate chemical reactions in the cell, which may lead to higher performance in the short term, but at the same time significantly intensifies the aging processes in the battery.

[0034] In this embodiment, an empirical model using data from real-world applications is employed in the comparison step to predict the lifespan of the long-term energy storage system. This model takes into account the number of charge and discharge cycles, ambient temperature, and the type of load. For specific applications, this model can provide highly accurate predictions.

[0035] In a preferred embodiment, the comparison step involves generating a stochastic prediction of the discharge behavior based on the current discharge behavior and comparing this prediction with a known discharge behavior. Such a stochastic model assumes that the lifetime of the long-term energy storage device fluctuates due to random, unpredictable events. The model uses probabilities to make predictions rather than calculating a precise lifetime. Stochastic models are particularly useful when many factors can influence the lifetime and when it is difficult to accurately measure all of these factors.

[0036] In a preferred embodiment, feedforward control is used to transmit the manipulated variable directly to the converter. Feedforward control is particularly advantageous in the case of extreme load peaks. The required manipulated variable is sent directly to the converter without first having to generate it via a current controller. Thus, the converter can react almost simultaneously with the current measurement, while the remaining manipulated variable for precise control is generated by the standard current controller.

[0037] In a preferred embodiment, the hybrid energy storage system is used in a mobile robotic system, an e-scooter, an e-bike, an electric vehicle, a mobile crane, a lifting platform, or an electric boat. In mobile robotic systems, the efficient and rapid provision of energy for dynamic movements is essential. Since robots frequently perform rapid changes of direction, sudden starts, and stops, a high-performance energy storage system can provide the necessary energy pulses, while the long-term energy storage system covers the robot's baseline energy requirements. This reduces wear on the long-term energy storage system and ensures consistent performance, thereby increasing the robot's overall operating time and lowering maintenance costs.

[0038] For lightweight electric vehicles like e-scooters and e-bikes, a hybrid energy storage system significantly improves range and energy efficiency. For example, energy recovered during braking or downhill driving can be stored in the high-performance energy storage system and used for the next acceleration phase. This reduces energy consumption and extends the lifespan of the long-term energy storage system, which is particularly beneficial in urban environments with frequent acceleration and braking cycles. For the user, this translates to longer battery life and potentially fewer charging stops, increasing the practicality and appeal of these vehicles.

[0039] In larger electric vehicles, such as electric cars, the advantage of a hybrid energy storage system is particularly pronounced, as high power demands occur during acceleration and recuperation. While a long-term energy storage system ages and loses capacity more quickly under intensive load demands, a high-performance energy storage system can absorb these intense loads. It absorbs the high charging pulses generated during braking and efficiently makes the energy available for the next acceleration. This protects the long-term energy storage system, resulting in greater range and a longer lifespan. Furthermore, the rapid availability of power allows the vehicle to be operated more dynamically and efficiently.

[0040] In the field of heavy mobile machinery, such as mobile cranes or lifting platforms, the high peak power output of a long-term energy storage system plays a crucial role. When lifting or lowering heavy loads, the long-term energy storage system can provide the required energy instantly, enabling more precise and faster control. By decoupling peak loads from the long-term energy storage system, its thermal stress is reduced, which increases operational reliability and lowers maintenance costs in work environments requiring continuous operation.

[0041] Electric boats also benefit from hybrid energy storage, especially in harbors or during maneuvers that require frequent power changes. The high-performance energy storage system can absorb the loads from acceleration and braking, thus conserving battery power. This type of storage improves the boat's overall energy efficiency and extends the lifespan of the long-term energy storage system, resulting in greater operational reliability and lower costs.

[0042] Similar advantages arise with industrial stationary energy storage systems, particularly hybrid energy storage systems. In production lines, hybrid energy storage systems can not only provide the required peak energy during acceleration and deceleration processes, but also efficiently recuperate excess energy. Decoupling peak loads from the individual storage components reduces their thermal stress, which increases operational reliability and lowers maintenance costs. Additionally, integrated preventive maintenance indicators enable the early detection of wear or performance degradation in the storage system, contributing to predictable and efficient maintenance and minimizing unplanned downtime.

[0043] In a preferred embodiment, energy recovered through recuperation is fed into the high-performance energy storage system. Storing the recuperated energy in the high-performance energy storage system relieves the load on the long-term energy storage system, thus extending its service life, as it is not subjected to the short, intensive charging cycles caused by recuperation, which can temporarily overload its storage capacity. Simultaneously, thermal loads on the long-term energy storage system are reduced, since the peak loads resulting from rapid charging and discharging processes are absorbed by the high-performance energy storage system. Furthermore, storing the recuperated energy in the high-performance energy storage system improves the overall system efficiency by minimizing energy losses that could occur during the conversion and storage of short-term energy peaks in long-term energy storage systems such as batteries.The converter controls the recuperation process and first directs the recovered energy to the high-performance energy storage system, which can efficiently absorb short-term peak loads before they are needed for the next load change. This targeted energy distribution protects the long-term energy storage system from unnecessary (partial) charging cycles, thus extending its lifespan and improving system efficiency. Thanks to this efficient energy distribution, braking resistors are only needed as an emergency system or can be eliminated entirely.

[0044] Detailed description of the invention

[0045] Advantageous and non-restrictive embodiments of the invention described in the claims are explained in more detail below with reference to the drawings.

[0046] Fig. 1 shows a schematic representation of a hybrid energy storage system with a load. Fig. 2 shows a flowchart of the method according to the invention.

[0047] Reference is now made in detail to embodiments, examples of which are shown in the accompanying figures. The effects and features of the embodiments, as well as their implementation methods, are described with reference to the accompanying figures. In the figures, the same reference numerals denote the same elements, and redundant descriptions are omitted. However, the present invention can be implemented in several different forms and is not to be understood as limited to the embodiments shown here. Rather, these embodiments are provided as examples to ensure that this disclosure is thorough and complete and fully conveys the aspects and features of the present invention to those skilled in the art. Methods, elements, and techniques that are not necessary for a skilled person to fully understand the aspects and features of the present invention are not described.In the figures, the relative sizes of elements, layers, and areas may be exaggerated for clarity. The embodiments described below are merely illustrative of the principles of the present invention. It is understood that changes and deviations of the arrangements and the details described herein will be apparent to other persons skilled in the art. It is therefore intended to limit ourselves to the scope of the pending patent claims and not to the specific details included herein for the description and explanation of the embodiments. Fig. 1 shows a schematic representation of a hybrid energy storage device 1, comprising a supercapacitor 2 as a high-performance energy storage device, a battery 3 as a long-term energy storage device, and a (schematically represented) converter 4. A load 5 is connected to the hybrid energy storage device 1. The load profile of the load 5 can, in principle, assume any conceivable form.The examples mentioned herein, such as white noise, a sinusoidal profile, a pulsed profile, a triangular profile, a constant profile, or a combination thereof, serve only for illustration and are listed as examples to simulate various critical test situations. The load profile of Load 5 can exhibit load peaks that are at least 5 times, at least 10 times, or at least 20 times larger than the moving average of Load 5. This opens up a wide range of applications, especially in areas where high power surges are required or load requirements vary significantly.

[0048] The first arrow represents the energy flow from battery 3 to both supercapacitor 2 and load 5. The initial charging of supercapacitor 2 is carried out via battery 3, so battery 3 is not subjected to any additional load during operation. A second arrow 7 shows that supercapacitor 2 supplies energy to load 5, which in turn can also supply energy to supercapacitor 2, for example by supplying energy recovered through recuperation.

[0049] The hybrid energy storage system 1 shown can be used in a mobile robot system, an e-scooter, an e-bike, an electric vehicle, a mobile crane, a lifting platform, or an electric boat. However, the system is not limited to mobile applications. It can also be used in stationary applications, such as stationary robot systems. The hybrid energy storage system 1 is particularly well-suited for mobile applications due to its dynamic load profiles and recuperation capabilities.

[0050] The combination of supercapacitor 2 and battery 3 shown represents only one possible embodiment of a hybrid energy storage device 1. Other combinations of high-performance energy storage and long-term energy storage are also possible. Another example would be the combination of electrolyte capacitor with a supercapacitor, where the supercapacitor would be the long-term storage device. Furthermore, a first hybrid energy storage device can be used as the high-performance energy storage device, and a second hybrid energy storage device can be used as the long-term energy storage device for a third hybrid energy storage device. An example of this is the combination of electrolyte capacitor and double-layer capacitor (first hybrid energy storage device) with a double-layer capacitor and battery (second hybrid energy storage device), which together form a third hybrid energy storage device.It is therefore evident that there are several combinations to realize a hybrid energy storage system, up to and including a, possibly multiple, chaining of hybrid energy storage systems.

[0051] A moving average is calculated based on the load profile, with the difference between the actual load profile and the calculated moving average serving as a control variable for converter 4. Other filtering methods can also be used, such as those that only become active above a certain threshold or that react adaptively to the dynamics of the load 5. However, the moving average method is preferred because it offers a simple and robust solution. Based on this control variable, converter 4 distributes the load 5 to the supercapacitor 2 and / or the battery 3. Converter 4 ensures that the battery 3 is discharged at a substantially constant rate within a discharge cycle.

[0052] This discharge behavior closely resembles the manufacturers' cycle tests, which is why the manufacturers' reliability tests can be used to determine the battery's lifespan by comparing its current discharge behavior with a known discharge behavior. For this purpose, data from test cycles, particularly data from the battery's manufacturer, can be used.

[0053] Fig. 2 shows a flowchart of the method according to the invention. In a first step S1 of the method according to the invention, a hybrid energy storage device 1 is provided, comprising a high-performance energy storage device 2, a long-term energy storage device 3, and a converter 4. The high-performance energy storage device 2 and the long-term energy storage device 3 can be implemented by various combinations of technologies known per se, as described above. The converter 4, also referred to as a DC-DC converter, is responsible for adjusting the voltage and controlling the energy flow between the two storage devices 2, 3 and the connected consumer (e.g., a load 5). Since the high-performance energy storage device 2 and the long-term energy storage device 3 have different electrical properties, the converter 4 ensures that the strengths of both energy storage devices can be used efficiently.Converter 4 can be considered a type of valve that handles energy distribution, while the control function is implemented in a controller. In its simplest form, the controller is a computing unit that executes specific control algorithms to monitor and control the energy flow and operating parameters in the hybrid energy storage system 1. Additionally, converter 4 contributes to monitoring and controlling the state of charge of both energy storage systems 2 and 3. Through precise control, it ensures that the high-performance energy storage system 2 is not overcharged and that the state of charge of the long-term energy storage system 3 remains within safe limits. This is essential for preventing thermal and electrical overload, which could lead to accelerated aging or even safety hazards.

[0054] In the next step S2, a load 5 is applied to the hybrid energy storage system 1, where the load 5 has a load profile. This load profile is a measure of how much power the hybrid energy storage system 1 must provide at a given time. The load profile is detected and filtered by the controller so that the system can operate efficiently regardless of the specific operating state. It results from the respective application and not the other way around – the application is not defined by the load profile. Examples of possible load profiles are white noise, a sinusoidal profile, a pulsed profile, a triangular profile, or a constant profile, or a combination thereof.

[0055] In a subsequent step S3, a moving average of the load profile is calculated. The difference between the instantaneous load profile and the moving average of the load profile forms a control variable for the converter 4. Based on this control variable, the converter 4 distributes the load to the high-performance energy storage device 2 and / or the long-term energy storage device 3, whereby the long-term energy storage device 3 is discharged at a substantially constant rate within a discharge cycle. This step of the method according to the invention allows a reliable comparison with known discharge behavior, since in routine discharge tests the long-term energy storage devices under test are generally discharged at a constant rate (e.g., according to a standard that must be adhered to).

[0056] In a subsequent step S4 of the inventive method, the instantaneous discharge behavior of the long-term energy storage device 3 is compared with a known discharge behavior in order to predict the lifetime of the long-term energy storage device 3. The discharge behavior according to the invention thus closely resembles known cycle tests, which is why these reliability tests can be used to determine the lifetime of the long-term energy storage device 3.

[0057] Step S4 of the comparison can be performed while load 5 is applied to the hybrid energy storage system 1. The ability to predict the lifetime of a long-term energy storage system 3 while it is under load offers numerous advantages. A real-time assessment of the long-term energy storage system 3 allows for more accurate and timely decisions regarding maintenance and replacement, thus improving the reliability and safety of the entire hybrid energy storage system 1.

[0058] Step S4 of the comparison can be performed in different ways. In a simple variant, a number of discharge cycles of the long-term energy storage device 3 are compared. This assumes that long-term energy storage devices 3 must be replaced after a certain number of charge and discharge cycles, as they lose capacity with each cycle.

[0059] Alternatively or additionally, an empirical model can be used, taking into account the ambient temperature and / or the cell temperature of the long-term energy storage device 3. This empirical model considers the number of charge and discharge cycles, the ambient temperature, and the type of load. The ambient temperature and the cell temperature have a significant influence on the performance, lifespan, and safety of long-term energy storage devices 3. In the case of batteries, strong temperature fluctuations can negatively affect the electrochemical processes within the battery and, in the long term, lead to performance losses and an accelerated aging process.

[0060] Alternatively or additionally, a stochastic prediction of the discharge behavior can be generated based on the current discharge behavior and compared with a known discharge behavior. Such a stochastic model assumes that the lifetime of the long-term energy storage device 3 fluctuates due to random, unpredictable events. The model uses probabilities to make predictions instead of calculating a precise lifetime. Stochastic models are particularly useful when there are many factors that can influence the lifetime and when it is difficult to measure all of these factors accurately.

[0061] Alternatively or additionally, a machine learning-based AI model can be used to derive patterns from the current discharge behavior and to generate a stochastic prediction of future discharge behavior. Such models combine probabilistic approaches with data-driven learning to assess the impact of complex, difficult-to-measure factors on the lifetime of the long-term energy storage system 3 and to enable predictions with higher accuracy.

[0062] If a particularly fast response of the converter 4 to a change in the load 5 or the load profile is required, a feedforward control can be used to transmit the manipulated variable directly to the converter 4. Feedforward control is particularly advantageous in the case of extreme load peaks. In this case, the required manipulated variable is sent directly to the converter 4 without first having to generate it via a current controller. Thus, the converter 4 can react almost simultaneously with the current measurement, while the remaining manipulated variable for precise control is generated via the normal current controller. It is evident that various features are combined in a single embodiment to simplify the presentation of the invention. This type of disclosure is not to be understood as meaning that the claimed embodiments require more features than are expressly stated in the individual claims.As the claims show, the subject matter of the invention lies in fewer than all features of a single disclosed embodiment.

Claims

REQUIREMENTS 1. Method for predicting the lifetime of a hybrid energy storage system (1), comprising the following steps: - Provision of a hybrid energy storage system (1) comprising a high-performance energy storage system (2), a long-term energy storage system (3), and a converter (4), - Applying a load (5) with a load profile to the hybrid energy storage system (1), - Calculating a moving average of the load profile, wherein a difference between the load profile and the moving average of the load profile forms a control variable for the converter (4), wherein the converter (4) distributes the load to the high-performance energy storage system (2) and / or the long-term energy storage system (3) based on the control variable, wherein the long-term energy storage system (3) is discharged at a substantially constant rate within a discharge cycle, - Comparing an instantaneous discharge behavior of the long-term energy storage device (3) with a known discharge behavior and predicting the lifetime of the long-term energy storage device (3) based on the comparison.

2. Method according to claim 1, wherein the comparison step is performed while the load (5) is applied to the hybrid energy storage device (1).

3. Method according to one of the preceding claims, wherein the known discharge behavior is based on data from test cycles, in particular on data from test cycles of a manufacturer of the long-term energy storage device (3).

4. Method according to any of the preceding claims, wherein the load (5) has at least one load peak which is at least 5 times, preferably at least 10 times, particularly preferably at least 20 times greater than the moving average of the load (5).

5. Method according to one of the preceding claims, wherein in the step of comparing a number of discharge cycles of the long-term energy storage device (3) is compared.

6. Method according to one of the preceding claims, wherein in the comparison step an ambient temperature and / or a cell temperature of the long-term energy storage device (3) is taken into account.

7. Method according to one of the preceding claims, wherein in the step of comparison, a stochastic prediction of the discharge behavior is created starting from the current discharge behavior and the stochastic prediction of the discharge behavior is compared with a known discharge behavior.

8. Method according to one of the preceding claims, wherein a feedforward control is used to transmit the manipulated variable directly to the converter (4).

9. Method according to one of the preceding claims, wherein the hybrid energy storage device (1) is used in a mobile robot system, in an e-scooter, in an e-bike, in an electric vehicle, in a mobile crane, in a lifting platform or in an electric boat.

10. Method according to one of the preceding claims, wherein energy obtained through recuperation is supplied to the high-performance energy storage device (2).

Images