Thermal Management Systems for Electric Vehicles

Abstract

The present invention relates to a thermal management system for an electric vehicle with various defined components related to thermal management, in particular a high-voltage store and an electric machine, comprising at least one thermal module for each defined component, controllable by a control module, comprising a navigation system and at least one electronic control unit having a control module and a prediction module, the thermal management system being capable of detecting, during the journey, the following conditions by suitable configuration of the prediction module: - based on a plurality of thermal management-related data of the navigation system acquired during a set period, at least one historical heating or cooling action progression (heating or cooling action progression) related to the route section is identified over time; - during the same set period, historical temperature trends over time (temperature trends) are acquired for each component by sensors, which relate to the route section, - determining at least one predicted heating or cooling progression associated with a route segment based on predictable thermal management related data of the navigation system for at least one defined horizon; A predicted temperature transition is calculated for each component based on the historical heating or cooling action over time, the historical temperature transition over time, and the predicted heating or cooling action over time.

Description

[Technical field]

[0001] The present invention relates to a thermal management system for an electric vehicle, particularly including a high voltage storage unit and various other components related to thermal management. [Background technology]

[0002] Thermal management systems are already partly known, for example in the form of heater installations and / or climate control installations for thermostating (heating and / or cooling) the interior space and / or the high-voltage store, in particular of electric vehicles. Thermal management for the interior space and the high-voltage store of electric vehicles is known, for example, from DE 199 1 43 336 A1.

[0003] To drive electric or hybrid vehicles, such vehicles have a drive train with an energy store for energy supply. The energy store is usually a suitably dimensioned high-voltage battery, also referred to below as high-voltage store. The high-voltage store usually heats up during the charging or discharging process, which, if excessive, can, among other things, lead to a permanent deterioration in performance or a shortened service life of the high-voltage store. For this reason, the high-voltage store is usually cooled appropriately during operation and is often connected to the vehicle's environmental temperature circuit, which is also used for the environmental temperature regulation (air conditioning) of the interior space. This environmental temperature circuit has a certain performance, i.e. a certain maximum cooling capacity that can be relied on to cool the interior space and the high-voltage store. Depending on the cooling needs of these two components, a kind of competition can arise in which the cooling capacity is not sufficient to cool the high-voltage store and the interior space as much as required. In this case, depending on the priority with which the cooling capacity is allocated, one must either accept an increased thermal load on the high-voltage store or a loss of comfort in the interior space.

[0004] In order to reduce the energy consumption during the conditioning of the interior space of an electric or hybrid vehicle and thus to extend the vehicle's range by reducing the energy draw from the high-voltage store, the prior art provides that a device for conditioning the passenger compartment climate and an energy store are thermally coupled to each other, for example for the exchange of coolant. This makes it possible, in certain circumstances, to first exchange heat between these two components instead of activating the device for conditioning the climate. For example, thermal energy, in particular waste heat, is received from the energy store and transferred to the device for conditioning the passenger compartment climate. This is done as long as the actual temperature in the passenger compartment is within a predefined temperature range. In this way, cooling of the energy store is achieved without the device for conditioning the climate. The removed heat is released to the passenger compartment, but only if its temperature is within a predefined temperature range.

[0005] In the above-mentioned prior art, an electric vehicle or hybrid vehicle includes an interior space and a high-voltage storage unit, both of which can be conditioned by an environmental temperature control device of the vehicle, and the environmental temperature control device has a certain cooling capacity. Here, the high-voltage storage unit (HVS) has an actual HVS temperature, and the interior space has an actual interior space temperature. In a pre-conditioning mode, the high-voltage storage unit is sub-cooled by the environmental temperature control device to an HVS temperature lower than an HVS operating temperature for pre-conditioning the high-voltage storage unit.

[0006] Thus, the high-voltage store is cooled by the climate control even if there is currently no need for cooling the high-voltage store and the actual HVS temperature indicates a value below the HVS operating temperature. In this way, a subcooling of the high-voltage store below the HVS operating temperature of the high-voltage store is advantageously achieved. This so-called preconditioning advantageously creates a cold buffer, which delays the time when the need for cooling of the high-voltage store may occur. Thanks to the cold buffer, the temperature of the high-voltage store can be increased without performance degradation due to an excessive temperature increase, and there is no need to use the climate control for cooling the high-voltage store. This makes it possible, in particular, to use all cooling capacity exclusively for cooling the interior space of the vehicle. In this way, the high-voltage store also becomes a cold storage tank with respect to its own climate control. The preconditioning of the high-voltage store is carried out in particular in anticipation in advance at a stage in which no or only little cooling of the high-voltage store would normally take place.

[0007] Outside the preconditioning mode, the HVS temperature is specifically controlled to an HVS operating temperature that is within the appropriate HVS operating temperature range to avoid performance degradation or damage.

[0008] In summary, in Patent Document 1, a pre-conditioning of the high voltage store is performed, considering the high voltage store as a cold / hot buffer, in order to reduce the load of the energy demand of the environmental temperature control equipment for the interior space of the vehicle during driving.

[0009] At this time, the possibility of taking into account the future temperatures of the interior space and the high-voltage storage unit in the event of overcooling based on the navigation data is examined in advance.

[0010] Furthermore, from US Pat. No. 6,299,633 it is known to predict the need for pre-adjustment based on available data such as the expected duration of stay, weather, etc. The vehicle user is notified and must confirm the recommended pre-adjustment.

[0011] Finally, Patent Document 3 is a prior art that applies the so-called “reinforcement learning” method in combination with thermal management of environmental temperature control equipment in order to fundamentally improve the quality of temperature control.

[0012] The applicant's unpublished German patent application No. 102021101513 describes a thermal management system as an environmental temperature regulation system for an electric vehicle with an interior space and a high-voltage store, which has an environmental temperature regulation facility and an electronic control unit, the environmental temperature regulation facility being designed to regulate the environmental temperature of both the interior space and the high-voltage store, the control unit having a preconditioning module for executing a preconditioning mode while charging the vehicle when it is stopped before starting the journey. The preconditioning module is designed to be able to predict at least the length of the travel route and the outside air temperature along the travel route, and to use the high-voltage store as a heat or cold storage depending on the prediction. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] DE 102014226514 [Patent Document 2] International Publication No. 2019 / 238389 [Patent Document 3] US Patent Publication No. 2019 / 0390867 Summary of the Invention [Problem to be solved by the invention]

[0014] The object of the present invention is to improve the thermal management of the various energy consuming components in an electric vehicle, including the high voltage storage, with regard to efficiency and optimization during operation. [Means for solving the problem]

[0015] This problem is solved according to the invention by the features of the independent patent claims. Advantageous embodiments, developments and modifications are the subject of the dependent claims.

[0016] The present invention relates to a thermal management system for an electric vehicle with various defined components related to thermal management, in particular a high-voltage store and an electric machine, comprising at least one thermal module for each defined component, controllable by a control module, comprising a navigation system and at least one electronic control unit having a control module and a prediction module, the thermal management system being capable of detecting, during the journey, the following conditions by suitable configuration of the prediction module: - determining at least one historical heating or cooling progression associated with a route segment based on a plurality of thermal management related data of the navigation system acquired during a set time period; - during the same set period, historical temperature trends are acquired for each component by sensors relating to the route section, - at least one predicted heating or cooling progression related to a route section is determined based on predictable thermal management related data of the navigation system for at least one defined horizon, A predicted temperature transition is calculated for each component based on the historical heating or cooling action over time, the historical temperature transition over time, and the predicted heating or cooling action over time. The thermal management system is configured as follows:

[0017] The present invention is based on the following idea:

[0018] The present invention preferably makes use of the so-called "Reinforcement Learning", which is essentially a machine learning method, thus aiming to predict and control when to (need to) cool or heat the high voltage storage and other thermal management related hardware components of the on-board energy system.

[0019] The heat flows and the resulting component and interior space temperatures are highly dependent on the (individual) driving profile and the use of the vehicle with external influences (e.g. due to road profile and weather conditions). Furthermore, components and passengers have different requirements for their optimal operating temperatures as well as different thermal masses (or time constants). The efficiency of heat transfer between these components and the power coefficients of cooling bodies and heat sources are also determined by various internal and external influences.

[0020] Therefore, the optimized thermal management of the vehicle interior and the hardware components of the on-board energy system must be subject to complex interdependencies, including external influences, that are not taken into account, or at least are only incompletely taken into account, by the current driving strategy. The interdependencies can be taken into account, preferably by using a reinforcement learning approach, to find an optimal thermal management strategy for the external influences on the vehicle and the driving profile. Some of the thermal management related components may be the high voltage storage, the electric motor, the vehicle passenger cabin, and further heating and cooling elements. In a real thermal management system, these heating and cooling components may be electric heaters, pumps, electric refrigerant compressors, intake dampers, and cooling fans.

[0021] Technical issues: Current thermal management strategies are not optimally adapted to the customer and his driving characteristics (e.g. duration, acceleration). They do not take into account all (combinations of) external influences during the journey, such as information about the upcoming route or driver characteristics. A more complex thermal management strategy must be derived, taking into account the individual and partly conflicting requirements for optimal temperature of components and passengers.

[0022] Current approaches, such as using individual characteristic maps for each component in thermal management, waste energy because they do not consider the characteristics and needs of the entire system and its complex interactions with external influences. Furthermore, current systems do not focus on the optimum of total energy consumption across all components, because they ignore the dependence of the coefficient of performance (COP) of these components and the dependence of the efficiency of heat transfer between them on external factors.

[0023] The large number of combinations and situations that must be addressed by thermal management strategies increases the demand for intelligent algorithms that can identify the optimal strategy for each situation. In contrast, current implementations use highly simplified functions that cannot accurately model interdependencies and address them appropriately and flexibly. However, the current system has a limited set of possible combinations and triggers, which prevents in-depth consideration of the diversity of requirements and situations. Furthermore, the current strategy of delivering optimal driving performance to all customers, up to highly dynamic driving characteristics, lacks optimization of efficiency at moderate acceleration requests. Therefore, individual strategies must be derived for each driver.

[0024] Basic principle of the present invention (basic concept): The basic idea according to the present invention is to model the problem as a reinforcement learning problem in order to predict when to actively cool or warm the hardware components of the on-board energy system.

[0025] In “reinforcement learning”, so-called “agents” learn to take actions based on rewards and punishments. (Figure 4 illustrates the per se known principles of “reinforcement learning” (RL) as a mathematical method.)

[0026] The demands as described in the "Technical Problem" section can be modeled with individual rewards, for example with a negative reward for temperatures to be avoided and a (higher) positive reward for optimal temperatures. This means, for example, that a high-voltage battery has an increased internal resistance at low temperatures, i.e. a lower available power output, and ages faster at high temperatures. In addition to meeting the demands of the components, the total energy consumption must be minimized and the heat transfer efficiency must also be considered. Thermal management strategies can include measures such as the activation and control of components of the thermal management system that result in the cooling or heating of the system or parts of it. Actions of an intelligent algorithm (e.g. by reinforcement learning) can initiate a combination of these components depending on the situation. Such situations are defined by various ambient parameters or conditions, which can include weather information, navigation data, information about the cabin and the components.

[0027] Examples of practical applications of the present invention: In the following, only the main aspects are described. Advantageous embodiments of the invention are explained in more detail with reference to the drawings, in which: - The surroundings are modeled by several on-board signals. - Based on the current and previous state of affairs, the “Agent” takes measures according to guidelines (Richtlinie). - Compensation is influenced by several factors. - To learn the guidelines, we propose to train a network (DQN; neural network) that outputs actions based on states (by estimating q-values ​​for state, action pairs).

[0028] The present invention will now be described in detail with reference to the accompanying drawings based on embodiments. [Brief description of the drawings]

[0029] [Figure 1] 1 is a highly simplified block diagram of a thermal management system according to the present invention; [Diagram 2] FIG. 13 shows an example of a predicted temperature progression for a high voltage store based on historical heating or cooling activity over time (from NAV_hist and NAV_praed), historical temperature progression over time, and predicted heating or cooling activity over time for the high voltage store. [Diagram 3] FIG. 2 shows an example of a predicted driving route having several determined route segments. [Figure 4] FIG. 1 shows a schematic overview of the scheme of “reinforcement learning” (basically prior art). [Diagram 5] FIG. 2 is a diagram showing a signal, an action, and a reward when using a "reinforcement learning" scheme according to the present invention. [Figure 6] FIG. 8 is a diagram for comparing the control flow and its operation of a thermal module according to the prior art with the control flow and its operation of the thermal module according to the present invention shown in FIG. 7. [Figure 7] FIG. 2 is a diagram showing the control flow and operation of the thermal module according to the present invention; [Figure 8] FIG. 13 illustrates a possible processing sequence for thermal management related navigation data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] FIG. 1 shows a schematic block diagram of a thermal management system according to the present invention.

[0031] The thermal management system according to the invention is provided for a number of different defined components K1, K2 related to thermal management (here for example a high voltage storage part HV and an electric machine EM). The thermal management system comprises for each defined component HV and EM, in this example, two thermal modules TM_HV and TM_EM for heating and / or cooling, which can be controlled by a control module (here called "agent" from "reinforcement learning"). Furthermore, the thermal management part includes a navigation system NAV and an electronic control unit SE, which incorporates the control module "agent" and the prediction module PM.

[0032] The prediction module PM is in particular configured by suitable programming (computer program product) so that (see also Fig. 2 and Fig. 3): during the journey: - based on a number of data "State NAV_hist" related to thermal management of the navigation system NAV acquired during a set period ("History" in Fig. 2; H1, H2 in Fig. 3), at least one historical chronological progression of heating or cooling action (1) related to the route section is identified; - during the same set period (history), historical temperature progressions (2) related to the route section are acquired for each of the components HV and EM by sensors ("condition sensors"); - on the basis of the predictable thermal management related data State NAV_praed ("ae" with an umlaut a) of the navigation system NAV for the first horizon H1 and for the second horizon H2, at least one predicted heating or cooling effect progression (3) related to the route section is identified, - Based on the historical heating or cooling effect over time (1), the historical temperature change over time (2) and the predicted heating or cooling effect over time (3), a predicted temperature change (4) is calculated for each component HV and EM.

[0033] “Thermal management related data” “State NAV” is a root attribute, e.g.: - Vehicle speed - Road type (including road cover / surface irregularities) - Incline / gradient - Driving downhill or uphill - Curve Radius - Outside temperature - Weather (sun, ice, snow, ...) - Tunnel Driving - RTTI (Real Time Traffic Information) (traffic jams, danger zones and further warnings) - Energy consumption - and so on

[0034] A "thermal module" is a controllable module (eg, a "heat exchanger") that can cool or heat an attached component.

[0035] A "route segment" is a determined travel route segment that can be predicted by the navigation system NAV, such as, for example, S1 to S4 in Fig. 3. "Related to a route segment" means associated with such segments S1 to S4.

[0036] The prediction module PM can comprise, for each component HV and EM, partial prediction modules P_HV and P_EM or partial components for multivariate prediction.

[0037] As shown in FIG. 3, the predicted self-heating of the components HV and EM is taken into account in determining the predicted heating or cooling course (3).

[0038] By "heating or cooling transition" is meant in particular the influence on temperature by the root attributes, which act, so to speak, like a virtual additional thermal module that is independent of the component.

[0039] The predicted temperature progression (4) is preferably determined in the form of a probability distribution W (over time).

[0040] Here, preferably, "neural network" and "reinforcement learning" are used as the mathematical function modules.

[0041] The control module “agent” is appropriately configured to determine, in particular, stored heating and / or cooling thresholds T_HV and T_EM for controlling the thermal modules TM_HV and TM_EM of the components HV and EM. hvs_S and T em_S can be modified in proportion to the expected temperature evolution (4).

[0042] In FIG. 3 an example of a driving route is shown with exemplary route attributes for the necessary adaptation of cooling hysteresis or temperature thresholds for the thermal modules of the components during the journey based on the predicted temperature progression (4):

[0043] A trip under changing environmental conditions is shown diagrammatically. The route segments S1 to S4 are assigned specific route attributes or thermal management related data or features M1, M2, M3 and M4, e.g.: M1: Climbing: High self-heating M2: Downhill driving: Low self-heating M3: High-speed route (e.g. highway): high self-heating M4: City driving: Low self-heating

[0044] Additionally, other route attributes P1 and P2 can be considered, for example: P1: End of ride or break: Low self-heating P2: Charging process: High self-heating

[0045] For different components K1, K2, K3 etc. different prediction horizons H1 (eg for the electric motor EM) and H2 (for the high voltage store HV) can be set.

[0046] The cooling hysteresis changes A1, A2 and A3 are made by the control module (or AI module, "agent" or controller) based on their respective prediction horizons. A1: Next comes the low self-heating downhill run: Higher cooling hysteresis A2: If self-heating becomes too high on the highway afterwards: Lower the cooling hysteresis A3: Low self-heating Subsequent city driving or end of driving: Higher cooling hysteresis

[0047] Analysis of previous, determined vehicle usage data can be performed without entering a route destination in the navigation system.

[0048] For example, previous, determined and stored vehicle usage data can be analyzed to predict the shortest possible driving route.

[0049] The application of the reinforcement learning method to the basic idea of ​​the present invention is explained in Figures 4 and 5. Figure 4 shows a rough overview of the principle of "reinforcement learning". Figure 5 shows a detailed example of possible signals, actions, and "rewards" when the "reinforcement learning" scheme is applied according to the present invention.

[0050] Expected environment: State S t+1 ”: “State”: Input signal (sensor signal, navigation output data, ...) Component Temperature - High Voltage Storage (HVS) - Electronic Control Module - Electric motor - Power Electronics - Charging unit - Current Cable Cabin Temperature - Actual temperature - Target temperature Route information - Tilt - Speed ​​Limit - Estimated speed (e.g. traffic conditions) - Biography - Navigation in progress (travel time, charging terminals, destinations) Weather Information - Ambient temperature - Humidity - Wind - Sunshine Energy consumption - All components, especially those specified in “Action”.

[0051] Reward function “Rewardr t+1 ”: Component Temperature & Cabin Temperature Reward Function - Adherence to current and foreseen temperature windows and avoidance of temperature spikes. - Consideration of temperature-dependent aging and its effect on power limits. The "reward function" may, for example, take into account current and foresight considerations of individual efficiency and total energy consumption.

[0052] The so-called “agents” evolve better learned policies that generate control interventions (:= “actions”) based on the determined next state and a “reward function”, much like a controller with a predefined policy.

[0053] In this application, the so-called “action” is a newly learned temperature threshold (e.g., T hvs_S ,T em_S ) to operate the thermal modules TM_HV and TM_EM of the components K1 and K2 for heating or cooling.

[0054] 6 and 7 provide a rough explanation of the difference between the thermal module control flow and its operation according to the prior art (FIG. 6) and the thermal module control flow and its operation according to the present invention, which preferably employs "Reinforcement Learning" (RL) (FIG. 7).

[0055] FIG. 8 shows one possible processing sequence of thermal management related navigation data “State NAV” (see also FIG. 1) as a basis for determining historical and predicted heating or cooling trends (1) and (3) (see also FIG. 2).

Claims

1. A thermal management system for an electric vehicle with various defined components (K1, K2; HV, EM) relevant for thermal management, in particular a high voltage store (HV) and an electric machine (EM), comprising: A thermal management system comprising at least one thermal module (TM_HV, TM_EM) for each determined component (HV, EM) controllable by a control module ("agent"), a navigation system (NAV), and at least one electronic control unit (SE) having said control module ("agent") and a prediction module (PM), By suitable configuration of said prediction module (PM), during the journey: - based on a plurality of thermal management related data (State NAV_hist) of the navigation system (NAV) acquired during a set period, at least one historical heating or cooling action progression (1) related to a route section is identified over time, During the same set period, historical temperature progressions (2) relating to the route section are acquired for each component (HV, EM) by sensors, at least one predicted heating or cooling action progression (3) related to a route section is determined based on predictable thermal management related data (State NAV_praed) of said navigation system (NAV) for at least one defined horizon (H1, H2), Based on the time series heating or cooling action transition (1) according to the history, the time series temperature transition (2) according to the history, and the predicted heating or cooling action transition (3), a predicted temperature transition (4) is calculated for each component (HV, EM). It is configured as follows: Thermal management system.

2. 2. A thermal management system according to claim 1, characterized in that the predicted self-heating of each of the components (HV, EM) is taken into account when determining the predicted heating or cooling action progression (3).

3. 3. A thermal management system according to claim 1 or 2, characterized in that for each component (HV, EM) the predicted temperature progression (4) is determined in the form of a probability distribution (W).

4. In a thermal management system according to claim 1 or 2, the control module ("agent") is suitably configured to determine stored heating and / or cooling thresholds (T_HV, TM_EM) for the control of the thermal modules (TM_HV, TM_EM) of the components (HV, EM). ｈｖｓ _ Ｓ , T ｅｍ _ Ｓ ) is variable in proportion to the predicted temperature progression (4).

5. An electronic control unit (SE) for a thermal management system according to claim 1 or 2.

6. A vehicle comprising a thermal management system according to claim 1 or 2.

Images