DE102015015976A1 - Method and device for determining an operating strategy - Google Patents

Abstract

The invention relates to a method for determining an operating strategy for the use of electrical energy for a drive of a hybrid vehicle (10), with which a predetermined distance can be traversed time-optimized, the distance is divided into sections, the operating strategy is determined by means of a self-learning iteration process, which starts as a first iteration step (14) from a purely internal combustion engine drive over the entire route, and a device (11) for carrying out a method according to the invention for determining an operating strategy for a hybrid vehicle (10), comprising at least one arithmetic unit (12) for a calculation of manipulated variables for determining the operating strategy is provided, and with at least one memory unit (13), which is provided for a storage of the manipulated variables during an iteration.

Description

The invention relates to a method for determining an operating strategy for using electrical energy for a drive of a hybrid vehicle and to an apparatus for carrying out a method.

Methods are already known for determining an operating strategy for the use of electrical energy for a drive of a hybrid vehicle, with which a predetermined travel distance can be traversed in a time-optimized manner. In the known methods, the operating strategy is determined by a discrete dynamic programming, in which the distance, manipulated variables for the drive of the hybrid vehicle and state variables are discretized, are arranged in a state grid and final costs for all points of the state grid are calculated. Then, by means of backward recursion, lowest residual costs for a transition to the next state point and transition costs for all state points are determined. Subsequently, the manipulated variables for the operating strategy are determined by means of forward recursion, based on the initial costs, taking advantage of the remaining costs from the backward recursion and newly calculated transition costs. Here, the transition costs for all states must be calculated, resulting in a high computational effort with sufficiently fine discretization.

The invention is in particular the object of providing an accurate method with a low computational cost. It is achieved by an inventive method according to claim 1 and a device according to the invention according to claim 6. Further developments of the invention will become apparent from the dependent claims.

The invention is based on a method for determining an operating strategy for the use of electrical energy for a drive of a hybrid vehicle, with which a predetermined distance can be traversed in a time-optimized manner, the distance being divided into sections.

It is proposed that the operating strategy be determined by means of a self-learning iteration process, which starts as a first iteration step from a pure combustion engine drive over the entire route. As a result, an accurate method for determining the operating strategy can be achieved with little computational effort. In this context, an "operating strategy" should be understood as a strategy that determines when and how much electrical energy is used over the route to drive the hybrid vehicle. In this context, a "self-learning iteration process" should be understood to mean an iteration process in which, within the iteration process itself, the sections in which the use of electrical energy is particularly lucrative are determined step by step. In particular, the self-learning iteration process calculates how a travel time through the section changes by using electrical energy in a section, calculates an available electrical energy, applies a portion of the available electrical energy in at least one section, calculates the change in travel times and the available over the route electrical energy and determined in further iterative steps further gaining time by using additional electrical energy and other sections in which electrical energy is used.

Furthermore, it is proposed that in the first iteration step for all sections, a first map with a required or recuperated electrical energy is created as a function of an inserted electrical power and an electrical energy ΔE _Batt obtained over all sections is calculated. As a result, an electrical energy which is available for use in reducing the travel time can be calculated in the method even with little computation effort. A "map" is to be understood in this context, a field that represents a functional relationship between two sizes for the section. By evaluating the first map for all sections under the condition of the first iteration step that a purely internal combustion engine ride is assumed and thus electrical energy can be obtained in all sections only by Rekuperationsvorgänge, the electrical energy obtained over all sections .DELTA.E is calculated. This energy is available for use to reduce travel time.

It is also proposed that, in the first iteration step, a second characteristic map with a travel time through the section as a function of an inserted electrical energy and from the second characteristic map a time gain is calculated by using electrical energy for all sections and at least one section determined for a second iteration step in which a portion of the electrical energy ΔE _{Batt is used} to reduce travel time. As a result, the time saved by using electrical energy in different sections can be determined in the method even with little computational effort. In the method, therefore, particularly lucrative sections are determined for a reduction of the travel time, before, in a subsequent iteration step, the time gained through the use of electrical energy in determined in these sections. Thus, a calculation of the time gain in all possible sections can be dispensed with, whereby a small amount of computation is achieved.

Furthermore, that in the second iteration for all sections on the basis of the first map, the second map, and the proportion of electrical energy .DELTA.E _Batt used in at least a portion of the recovered over all the sections electrical energy .DELTA.E _Batt is calculated and for a subsequent iteration step, at least it is proposed a section is determined in which a proportion of the obtained electrical energy ΔE _{Batt is used} to reduce the travel time. The method thus only calculates the changes with respect to the previous iteration step and thus achieves a low computation outlay.

It is also proposed that an electrical energy that can be used in a section be limited to an overload protection value. As a result, an operating strategy can be achieved in which a drive train of the hybrid vehicle is spared.

Furthermore, the invention relates to a device for carrying out a method according to the invention for determining an operating strategy for a hybrid vehicle, with at least one arithmetic unit which is provided for a calculation of manipulated variables for determining the operating strategy, and with at least one memory unit which is used for a storage of the manipulated variables an iteration is provided. As a result, a simple device for carrying out an accurate method for determining the operating strategy can be achieved with a low computation effort.

Further advantages will become apparent from the following description of the figures. In the figures, an embodiment of the invention is shown. The figures, the description of the figures and the claims contain numerous features in combination. The person skilled in the art will expediently also consider the features individually and combine them into meaningful further combinations.

Showing:

1 1 is a schematic representation of a hybrid vehicle having an apparatus for carrying out a method for determining an operating strategy for using electrical energy for a drive of a hybrid vehicle;

2 a schematic representation of maps that are used to determine the operating strategy, and

3 a schematic representation of two iterative steps of the method for determining the operating strategy for using electrical energy for the drive of the hybrid vehicle.

The 1 shows a section of a hybrid vehicle 10 with a device 11 for carrying out a method for determining an operating strategy for using electrical energy for a drive of the hybrid vehicle 10 , with which a given distance can be traversed time-optimized. The device 11 includes a computing unit 12 , which is provided for a calculation of manipulated variables for determining the operating strategy, and at least one memory unit 13 , which is intended to store the manipulated variables during an iteration. The operating strategy determined in the method determines areas of travel in which electrical energy is used to drive the hybrid vehicle 10 is used, and also sets an amount of energy used in the areas. The electrical energy can improve longitudinal dynamics by increasing the electrical energy in a boost to increase a forward speed of the hybrid vehicle 10 is used, or to improve a lateral dynamics, in which a driving stability of the hybrid vehicle 10 influences and, for example, a cornering speed of the hybrid vehicle 10 is increased, are used. The improvement of the lateral dynamics can, for example, by an electric drive a rear axle of the hybrid vehicle 10 during cornering with internal combustion engine drive a front axle of the hybrid vehicle 10 be achieved. The route, which is to be traversed time-optimized, can be a single, irregular route or a route that is to be repeated several times in succession, such as a circuit.

In the method, the route is divided into sections. The distance is divided into segments in this embodiment, wherein a start and an end of the segments by a course of characteristics, such as may be predetermined by a course of a speed. For example, in the exemplary embodiment presented, a circular path is used as the route, which is subdivided into segments which each extend from a local speed minimum to a next local speed minimum. In the method, the segments are used as sections in which an application of electrical energy for time-optimized driving through the route is calculated. The segments are thus used as the basis of the subdivision of the route into sections for the determination of the operating strategy. Alternatively, the segments can be divided again, for example, fragmentary subsections, wherein a first fragment of the segment extends from a velocity maximum within the segment to the end of the segment, and further fragments equidistantly subdivide the segment from the beginning to the velocity maximum.

In the method, the operating strategy is determined by means of a self-learning iteration process, the first iteration step 14 assumes a purely internal combustion engine drive over the entire route (cf. 3 ). The self-learning process optimizes manipulated variables for the drive of the hybrid vehicle 10 , in particular an inserted electrical energy and a portion of the path in which the electrical energy is used, on its own. Based on the results of the first iteration step 14 the method determines the manipulated variables which are in a second iteration step 20 be incorporated. In the second iteration step 20 are based on the manipulated variables in the first iteration step 14 be determined, further manipulated variables for further iteration steps 20 Developed. The process continues until only little or no shortening of the travel time can be obtained by further iteration steps or all available electrical energy has been used. The available electrical energy may be limited by an energy stored in a battery at the beginning of the travel and an energy recuperated during the travel by braking, or external specifications for the process may be made for the process. Such specifications may, for example, in a specification of an energy reserve, which should be present after passing the distance in the battery, or in a specification of a batch-sustaining strategy in which a battery charge after passing the distance of the battery charge at the beginning of the route should correspond consist. In the described embodiment, a batch-sustaining strategy is used.

In the first iteration step 14 becomes a first map for all sections 15 created with a required or recuperated electrical energy E in a function of an applied electrical power P (see. 2 ). In the first iteration step 14 becomes an electric energy ΔE _Batt 17 calculated (cf. 3 ). The recovered electrical energy ΔE _Batt 17 forms a manipulated variable of the method. A calculation of the electrical energy ΔE _Batt 17 is in 3 in a diagram 16 shown. In the diagram 16 is the electric energy E _{Batt of} the battery of the hybrid vehicle 10 applied in the respective sections. Because in the first iteration step 14 is assumed by a pure internal combustion engine operation over the entire route, no electrical energy to drive the hybrid vehicle 10 used and it is only electrical energy recuperated. This energy stands as gained electrical energy ΔE _Batt 17 for use in a subsequent iteration step.

In the first iteration step 14 becomes a second map 18 calculated with a travel time t through the section as a function of an inserted electrical energy E (see. 2 ). From the second map 18 becomes in the first iteration step 14 a time gain is calculated by using electrical energy for all sections. In 3 is in a second diagram 19 a calculation of the time gain by using electrical energy for all sections shown. The time gain is calculated by forming a derivative of the travel time after the electrical energy used .DELTA.t / .DELTA.E.

For a second iteration step 20 At least one section is determined in which a portion of the electrical energy obtained ΔE _Batt 17 is used to reduce the travel time. In the described embodiment of the method, two sections having a maximum value of the derivation of the travel time after the inserted electrical energy are selected and for the following second iteration step 20 the fraction of electrical energy gained is ΔE _Batt 17 divided into the two sections. A number of sections in which the proportion of electrical energy gained ΔE _Batt 17 can be specified for the procedure from the outside; For example, the proportion of electrical energy gained can be ΔE _Batt 17 in only one section or in three or more sections in the following second iteration step 20 be used. The sections in which electrical energy is used form a manipulated variable of the method.

In the second iteration step 20 is for all sections based on the first map 15 , the second map 18 and the proportion of recovered electrical energy ΔE _Batt used in at least one section 17 the electrical energy ΔE gained across all sections 21 calculates and determines at least one section for a subsequent iteration step, in which a proportion of the electrical energy obtained .DELTA.E _Batt 21 is used to reduce the travel time. Because in the second iteration step 20 already the proportion of the gained electrical energy ΔE _Batt 17 from the first iteration step 14 is used, the recovered electrical energy is ΔE 21 in the second iteration step 20 less than in the first iteration step 14 , The first map 15 and the second map 18 already exist, so that in the second iteration step 20 directly taking into account the proportion of energy produced ΔE _Batt 17 from the first iteration step 14 the electrical energy gained over all sections ΔE _Batt 21 is calculated. A calculation of the obtained electrical energy ΔE _Batt 21 is in 3 in a diagram 22 shown. In the diagram 22 is the electric energy E _{Batt of} the battery of the hybrid vehicle 10 applied in the respective sections. Alternatively, the first map 15 and the second map 18 also be recalculated. Once again, at least two sections are determined, in which a proportion of the electrical energy obtained ΔE _Batt 21 is used to reduce the travel time. Such a determination of the sections in which a proportion of the electrical energy obtained ΔE _Batt 21 is used is in 3 in a diagram 23 shown. The one for the second iteration step 20 used proportion of recovered electrical energy ΔE _Batt 17 in two sections is preserved. As in the first iteration step 14 two sections are selected which have a maximum value of the derivation of the travel time according to the electrical energy used. Since electrical energy is already being used in two sections, these are in the second iteration step 20 For the following, third iteration step, certain sections are not necessarily identical to the sections that are for the second iteration step 20 were determined.

In the third and further subsequent iteration steps, the same procedure is used as in the first iteration step 14 and the second iteration step 20 applied. The self-learning iteration method is continued over further iteration steps until no further sufficient improvement of the travel time in the iteration steps can be achieved or all electrical energy that is available has been used. For the method, it was previously determined from which threshold time gain by additional electrical energy is no longer considered sufficient to continue the iteration.

In the method, an electric energy that can be used in a section is limited to an overload protection value. The overload protection value takes into account a thermal load on electrical components due to high electrical power and is designed in such a way that no thermal damage occurs when the overload protection value is passed through. If a larger electrical energy is used in the section in an iteration step, it is set to the overload protection value for the following iteration steps and by the arithmetic unit 12 In the following iteration steps, this section will not be selected for use of additional electrical energy.

The described method can also be carried out by using the fragments as sections. In this case, by the arithmetic unit 12 In the iteration steps it is also calculated whether and how an output velocity from the fragment and an input velocity into the subsequent fragment changes as a result of the use of electrical energy in certain fragments and whether and how the travel time thereby changes.

LIST OF REFERENCE NUMBERS

10

hybrid vehicle

11

contraption

12

computer unit

13

storage unit

14

iteration

15

map

16

diagram

17

gained electrical energy

18

map

19

diagram

20

iteration

21

gained electrical energy

22

diagram

23

diagram

Claims (6)

Method for determining an operating strategy for using electrical energy for a drive of a hybrid vehicle ( 10 ), with which a predetermined route can be traversed in a time-optimized manner, the route being subdivided into sections, characterized in that the operating strategy is determined by means of a self-learning iteration process, which is used as a first iteration step ( 14 ) emanates from a pure combustion engine drive over the entire route. Method according to claim 1, characterized in that in the first iteration step ( 14 ) for all sections a first map ( 15 ) is created with a required or recuperated electrical energy as a function of an inserted electrical power and an electrical energy ΔE ( 17 ) is calculated. Method according to claim 2, characterized in that in the first iteration step ( 14 ) a second map ( 18 ) with a travel time through the section as a function of an inserted electrical energy and from the second characteristic map ( 18 ) a time gain is calculated by using electrical energy for all sections and for a second iteration step ( 20 ) at least one section is determined in which a portion of the electrical energy obtained is ΔE _Batt ( 17 ) is used to reduce the travel time. Method according to claim 3, characterized in that in the second iteration step ( 20 ) for all sections based on the first map ( 15 ), the second map ( 18 ) and of the portion of the electrical energy ΔE ( 17 ) the electrical energy ΔE _Batt ( 21 ) and at least one section is determined for a subsequent iteration step, in which a portion of the electrical energy obtained is ΔE _Batt (FIG. 21 ) is used to reduce the travel time. Method according to one of the preceding claims, characterized in that a usable in a section of electrical energy is limited to an overload protection value. Device for carrying out a method for determining an operating strategy for a hybrid vehicle ( 10 ) according to one of the preceding claims, with at least one arithmetic unit ( 12 ), which is provided for a calculation of manipulated variables for determining the operating strategy, and with at least one memory unit ( 13 ), which is provided for a storage of the manipulated variables during an iteration.

Images