WO2012104542A2

WO2012104542A2 - Method and system for managing the energy of a rail vehicle

Info

Publication number: WO2012104542A2
Application number: PCT/FR2012/050205
Authority: WO
Inventors: Alain Jeunesse; Yannick EVAIN; Florian JOFFRIN
Original assignee: Societe Nationale Des Chemins De Fer Francais Sncf
Priority date: 2011-02-01
Filing date: 2012-01-31
Publication date: 2012-08-09
Also published as: FR2970912A1; WO2012104542A3; FR2970912B1; EP2670618A2

Abstract

The invention relates to a method for managing the energy of a rail vehicle (2) comprising a plurality of on-board sources (12, 14) and electrical energy consumers (6, 20), said sources comprising energy storage means (14). Said method comprises the steps of: collecting information relating to the available quantities of electrical energy that can be produced by the storage means (14) and to the maximum electrical power that may be absorbed by the consumers (6, 20); calculating an autonomy of the storage means (14), said autonomy being equal to the ratio between the electrical energy available in the storage means (14) and the maximum electrical power collected; and optimising the use of the sources (12, 14) for supplying the electrical energy according to the calculated autonomy.

Description

Method and system for managing the energy of a railway vehicle

The present invention relates to a method for managing the energy of a railway vehicle. It also relates to a corresponding management system. The invention is more particularly concerned with the field of rail transport, particularly in the field of hybrid railway vehicles.

Currently, two main railway gear architectures are operated in Europe. The first architecture is based on electric locomotives and the second architecture is based on diesel locomotives. On board electric locomotives, energy is distributed by sampling over a catenary through pantographs, transformers (in the case of alternative networks) and converters. On a diesel locomotive, energy is produced on site by a generator associated with a heat engine. Both architectures have a common part consisting of traction motors and auxiliaries.

Previously, railway engines powered by an electric motor had only one source of energy such as a catenary, a third rail, a heat engine, and so on. This source was sized to provide the maximum power needed to ensure the traction of the machine and the

20 needs of auxiliaries such as passenger comfort, air production, cooling of power organs, etc. Thus, at any moment, the need for electrical power of the mission of the machine is ensured by the only source of energy present.

There are currently two railway sources

25 of electrical energy combining a main power source and an electrical storage means. For these machines, the implementation of a storage means in addition to the main source requires a slight evolution of the existing control principles to take into account the characteristics of the storage means.

As an example of such a command, the main source reloads the storage means from a predefined threshold. This command aligns the power generation with the speed of the machine and the state charge of the storage means.

However, this type of energy management of a railway vehicle is not suitable for a railway vehicle comprising a plurality of on-board electrical energy sources of different types.

The present invention aims to improve the situation.

To this end, the invention relates first of all to a method for managing the energy of a railway vehicle comprising a plurality of on-board energy sources and consumers of electrical energy, said sources comprising storage means for energy, said method comprising the steps of:

o - collection of information relating to the quantities of electrical energy available that can be provided by the storage means and to the maximum electrical power likely to be absorbed by the consumers;

calculation of autonomy of the storage means, said autonomy being equal to the ratio of the electric energy available in the means of storage;

Storage and the maximum electrical power collected; and

- optimization of the use of sources for the supply of electrical energy according to the calculated autonomy.

The use of the autonomy of the electrical energy storage means makes it possible to optimize the management of the energy of the machine by making it possible to optimally use the energy storage means.

Advantageously, the sources comprise, in addition to the energy storage means, a main energy source comprising a generator and / or a fuel cell.

This onboard main energy source allows the production of the energy necessary for the operation of the machine and the load of the storage means.

Advantageously, the optimization step comprises a step of controlling the start of the main energy source if the autonomy is below a determined threshold.

The optimization step thus makes it possible to limit the use of the main energy source and favors the taking of energy from the storage means.

Preferably, the threshold is between 20 and 40 seconds, in particular equal to 30 seconds. This threshold corresponds to the start-up and power-up time of the main energy source.

Advantageously, the start control step comprises a step of transmitting an electrical energy instruction to the main energy source.

In a preferred embodiment, the optimization step comprises the steps of:

- Definition of at least two frequency areas of operation of the machine;

allocating at least one storage means among the storage means to each defined frequency domain; and

- In each frequency domain, control of the machine to use in priority the storage means allocated to said frequency domain during operation of the machine in this area.

This makes it possible to optimize the use of the on-board storage means by exploiting the adequacy of each type of storage means to a particular frequency domain of operation. The frequency domain of operation is directly related to the variation of the electric power demand of the vehicle during this operation.

By way of example, during a maneuver, the power demanded shows large amplitude variations, brief and fast, thus corresponding to a frequency domain operating on the high frequency side.

Conversely, during local services, the power required by the operation of the machine is more continuous, its amplitude can be

25 important and not very variable. The frequency domain of operation is, in this case, located on the low frequency side.

Preferably, the defined frequency domains comprise:

a first frequency range of less than ten MHz (millihertz); and

A second frequency range greater than about 20 MHz.

The first frequency domain, close to the continuum, is an image of the mean value of the mission of the machine while the second frequency domain corresponds to very variable regimes.

Preferably, the storage means allocated to the first frequency domain is a battery pack.

The characteristics of the batteries indeed require operation in steady state for optimum use in terms of efficiency, consumption, pollutant emission lifetime, etc. The batteries are thus able to withstand operating frequencies located in the low frequencies, from continuous to a few mHz.

Advantageously, the storage means allocated to the second frequency domain is a block of supercapacitors.

Indeed, supercapacitors are able to withstand operating cycles of about one hundred MHz to a few Hz. In other words, their charge and discharge cycles have a frequency adapted to the second frequency range.

According to a preferred embodiment, the step of collecting information relating to the maximum electrical power likely to be absorbed by the consumers comprises a step of predicting this maximum power from recorded values of the electrical power absorbed by the consumers. during previous operations of the machine.

Indeed, during each operation of the machine, the values of the electrical power absorbed by consumers are recorded. A statistical study of these values recorded for different types of mission (maintenance operation, service between two given stations, etc.) makes it possible to predict the maximum power likely to be absorbed by the consumers during a future mission of the mission. machine.

The invention also relates to a system for managing the energy of a railway vehicle comprising a plurality of on-board energy sources and consumers of electrical energy, said sources comprising energy storage means, said system including means of:

- gathering information on the quantities of available electrical energy that can be provided by the storage means and the power electrical power likely to be absorbed by consumers;

calculating an autonomy of the storage means, said autonomy being equal to the ratio between the electrical energy available in the storage means and the maximum electrical power collected; and

5 - optimization of the use of sources for the supply of electrical energy according to the calculated autonomy.

Exemplary embodiments of the invention will now be described more specifically, but not limitatively, with reference to the accompanying drawings in which:

- Figure 1 is a diagram illustrating the electrical structure of a railway vehicle according to one embodiment of the invention;

FIG. 2 is a diagram illustrating the structure and operation of the energy management system according to one embodiment of the invention;

FIG. 3 is a diagram illustrating the detailed structure of the processing means of the management system of FIG. 2 according to one embodiment of the invention;

FIG. 4 is a diagram illustrating the operation of the energy management method according to one embodiment of the invention; and

FIG. 5 is an enlarged view of part of the diagram of FIG. 4.

FIG. 1 illustrates the electrical network of a railway vehicle 2 of the hybrid type. The machine 2 is provided with a traction unit 4 for driving the machine. The traction unit 4 comprises, by way of non-limiting example, four electric traction motors 6.

The traction unit 4 also comprises a power supply 8 for energizing the motors 6.

Preferably, the traction diagram of the vehicle 2 allows braking by energy recovery.

These motors 6 consume electrical energy produced by a plurality of electrical sources on board the vehicle 2. These onboard sources 30 produce or store energy on board the craft 2.

In the example of Figure 1, these sources comprise a generator 12. The generator 12 includes a diesel engine providing a power equal to 230 kW, for example. It is the main source of electrical energy.

Alternatively, this main source may comprise a fuel cell, or a combination of a generator with a fuel cell.

The sources on board the craft 2 also include energy storage means 14.

The energy storage means 14 comprise in particular a battery pack 16 and a block of supercapacitors 18.

The battery pack 16 preferably comprises Ni-Cd batteries

(cadmium - nickel). For example, the battery pack 16 comprises 2 groups of 6 modules of 48 cells of battery cells with a capacity of 135 Ah.

The supercapacitor block 18 comprises a plurality of supercapacitors, for example 8 modules of 200 5000F / 2.5V supercapacitors connected in series. The total capacity of the block of supercapacitors 18 is then equal, in this example to 200 F.

It is also possible to use supercapacitors of different capacity, for example 2600F or 9000F.

The machine 2 also comprises a set of auxiliaries 20 which comprises, in particular, fans 22 of the traction motors 6, an air compressor 24 for the operation of the brakes of the machine 2, and a battery charger 26 coupled to an electric accumulator supplying energy to a low voltage circuit (72 V) of the machine 2.

All the sources of electrical energy, ie the generator group 12, the battery pack 16 and the block of supercapacitors 18, and consumers of electrical energy, that is to say the traction motors 6 and the auxiliaries 20 are connected to each other via a bus or high-voltage electrical network 28, otherwise called a power bus.

The electric power sources 12, 16, 18 and the consumers 6, 20 are further connected to each other via a CAN computer network 30 to which a supervisor 32 is connected. The main function of the supervisor 32 is to optimally distribute the energy supplied from the available energy sources to the consumers and the energy supplied by the generator set 12 to the battery pack 16 and the supercapacitor block 18.

Supervisor 32 is the component in which the steps of the energy management method of the invention are implemented. In other words, it is this component which comprises the means of the energy management system of the invention.

The supervisor 32 comprises a controller 34 and a real-time calculator 36.

The controller 34 performs control control functions of the machine 2 by implementing a digitized logic.

The computer 36, which is the subject of the invention, ensures the optimized management of the energy on board the vehicle 2 by developing voltage or current instructions that the sources and the consumers must supply and / or absorb. depending on the rail mission to be performed and according to the state of the equipment (in service or not, for example).

A human machine interface HMI 38 is also provided. This interface

HMI 38 connected to the network CAN 30 is able to show a driver 20 or maintenance of the machine 2, the state of energy sources, their autonomy, energy flows, instructions issued, etc. This information can be given as numeric values and / or as animation.

The HMI interface 38 is implemented either traditionally by conventional commands to a control panel, or by a touch screen 25 connected to the control system and / or maintenance by a computer link.

The structure of the energy management system of the invention, implemented by the computer 36, is described with reference to FIG. 2.

The energy management system manages energy flows between the plurality of sources and consumers. It ensures the energy needs of the mission of the machine 2 while optimizing the consumption, pollution and the life of the component components. By observing the mission of the machine and the state sources, including storage means 14, it decides at any time the contribution of each source according to its availability and its capacity in terms of autonomy and maximum power including, as will be described in the following description .

The energy management system comprises information acquisition means 40. These means include means for decoding the information available on the CAN 30 computer network and means for scaling this information. for example means for standardizing the units of measurement, etc. The information acquired o are analog or Boolean. By way of nonlimiting examples, this information can be a state of charge of the batteries and / or supercapacitors, a quantity of energy stored in the storage means, an instantaneous power consumed and / or supplied, effort instructions , equipment availability, geolocation data from

15 the machine in the railway field, a speed of the machine, a fuel level in the tank that feeds the generator, etc.

The energy management system also includes means 42 for processing the information acquired.

The energy management system also includes means of transmitting

20 44 of instructions. These means are capable of performing post-processing of the information received from the processing means 42. These post-processing operations include the scaling of this information and their encoding into messages transmitted on the CAN 30 network.

The structure and operation of the processing means 42 are detailed

With reference to FIG.

The processing means 42 include means 46 for collecting information relating to the quantities of available electrical energy that can be supplied by the sources and to the maximum electrical power likely to be absorbed by the consumers from the information acquired by the users.

30 means for acquiring information 40.

More particularly, the collection means 46 receive means 40 for acquiring information relating to the power of the generator set. 12, to the charge states of the battery pack 16 and / or the block of supercapacitors 18, to a ramping slope limitation of the storage means 14, to a quantity of available energy stored in the storage means 14, to a power necessary for the mission, that is to say 5 to be provided to the traction motors 6, etc.

The processing means 42 also comprise means 48 for calculating autonomy of the storage means 14 from the information collected by the collection means 46. The autonomy is defined as the ratio between the electrical energy available in the storage means 14 and the maximum electrical power likely to be absorbed by the consumers.

The energy available in each of the storage means 14 is related to a maximum discharge depth tolerated by said storage means taking into account the number of cycles of the storage means for a defined lifetime.

The maximum power likely to be absorbed by the consumers is proportional to the sum of the maximum currents consumed by the auxiliaries 20 and the traction motors 6.

The calculated autonomy is transmitted to control means 50 for starting up the generator set 12. These control means decide to start up the generator set 12 when the value of the calculated autonomy is lower than a determined threshold, preferably between 20 and 40 seconds, in particular equal to 30 seconds.

This threshold of autonomy is determined by taking into account the composition of the train towed by the machine and its mission, in terms of speed to meet, line profile, etc., and the characteristics of the means of storage in order to optimize their use in terms of energy efficiency, pollution, lifespan, etc.

The control means 50 transmit an electrical energy setpoint 52 to be supplied by the generator set 12 and a corresponding electric power setpoint 54 to be supplied by the generator set 12. The processing means 42 furthermore comprise means 56 for calculating the electric current to be supplied by the storage means 14 from the stream 57 to be supplied to the consumers for carrying out the mission, the value of which is received from the means 40 acquisition of information, and the electrical current setpoint 54 to be supplied by the generator 12 received from the control means 50.

The calculation means 56 implement the principle of conservation of the currents (node law), the electric current 58 to be provided by the storage means 14 then being equal to the difference between the current 57 to o be provided to consumers for carrying out the mission and the electrical current setpoint 54 to be supplied by the generator 12.

The processing means 42 also comprise frequency discriminating means 60. These frequency discriminating means 60 distribute the current 58 to be supplied by the storage means 14 to at least two frequency domains according to the characteristics of the desired mission. gear 2.

More particularly, the frequency discrimination means 60 define three frequency domains 62, 64, 66 for operating the machine 2.

The first frequency domain 62 is close to the continuous. It corresponds to the average value of the mission of the vehicle 2. A local service is an example of an operation of the vehicle 2 in this frequency domain. Indeed, during such a service, the power requested by the mission is almost continuous with amplitudes that can be important and not very variable. This first frequency domain 62 preferably includes frequencies below 10 mHz.

The second frequency domain 64 corresponds to a very variable operating mode of the vehicle 2. A maneuver is an example of an operation of the vehicle 2 in this frequency domain. Indeed, during such a maneuver, the power requested by the mission shows large amplitude variations, brief and fast. This second frequency domain 64 preferably includes frequencies above 20 mHz. The third frequency domain 66 is intermediate to the first and second frequency domain. It preferably includes frequencies between 10 mHz and 20 mHz.

The frequency discriminating means 60 allocate to each defined frequency domain 5 at least one source capable of supplying electrical energy during operation of the machine in said domain.

Remarkably, the inventors have discovered that the intrinsic temporal properties of the energy storage components project at locations distinct from the frequency axis. More particularly, the batteries are located in the field of low frequencies, the continuous to a few mHz while the supercapacitors are able to withstand operating cycles of about one hundred MHz to a few Hz. The characteristics of flywheels of inertia place them between the batteries and the supercapacitors.

Thus, the frequency discrimination means 60 allocate the battery pack 16 to the first frequency domain 62 first. The generator set 12 is also adapted to this frequency domain 62.

Moreover, the frequency discrimination means 60 allocate the block of supercapacitors 18 to the second frequency domain 64.

In addition, the frequency discrimination means 60 allocate a flywheel 20 and / or the battery pack 16 and / or the block of supercapacitors 18 to the third frequency domain 66, as the case may be, according to the amount of stored energy available. and their state of charge.

Electrical energy instructions 68, 70, 72 to be provided by the storage means allocated to the first, second and third frequency domains are transmitted by the frequency discrimination means 60.

The processing means 42 also include mission carry-over means 74. These transfer means 74 receive the electrical energy instructions 68, 70, 72 of the three frequency domains 62, 64, 66 transmitted from the frequency discrimination means 60. They also receive the electrical energy instruction 52 to be supplied by the generator 12 from the control means 50 for starting the generator 12. The reporting means 74 also receive information relating to the functional characteristics of the different sources from the information collection means 46.

These functional characteristics include the physical limitations of the sources.

In the case of the generator 12, such limitations are for example the maximum power that can be provided and / or the minimum downtime between two successive operations.

In the case of the storage means 14, such a limitation is, for example, the level of charge, the maximum current of discharge or charge.

Remarkably, the carryover means 74 determines the amount of energy that can actually be provided by each source from the physical limitations.

Then, for each source, the transfer means 74, compare the electrical energy setpoint to be supplied by the source and the determined amount of energy that can actually be supplied by the same source. If the setpoint is greater than the determined amount of energy, the transfer means 74 selects another source and controls the other source to provide the energy difference between the setpoint and the determined amount of energy.

For example, if the electrical energy setpoint to be supplied by the sources allocated to the second frequency domain 64 is equal to 70% of the total energy required for the operation of the machine, whereas these sources can only supply 30% of the total energy, the transfer means 74 control

25 to the energy sources allocated to the first and third frequency domains to provide the energy difference equal to 40%.

Thus, in order to ensure correct operation of the machine, the transfer means 74 can transfer part of the mission from one frequency domain to another frequency domain when the first domain can not ensure

30 totally the mission assigned to it by means of frequency discrimination. For example, the transfer strategy implemented when selecting another source promotes the use of the block of supercapacitors 18 which have a longer life in terms of the number of charge and discharge cycles.

Following the implementation of the carry-over by the carry-over means 74, new electrical energy orders 76, 78, 80 to be provided by the storage means allocated to the first, second and third frequency domains are transmitted from reporting means 74.

In addition, an instruction to limit the traction force 81 is emitted from the transfer means 74, to the transmission means 44 setpoints, when the sum of the electric energy can be provided by the different sources n is not sufficient to ensure the operation required by the mission.

In addition, the processing means 42 comprise balancing means 82, 84, 86 of the use of the sources allocated to the frequency domain 62, 64, 66 respectively. The balancing means balance the requested task, that is to say the electrical energy setpoint to be provided by the sources allocated to the frequency domain concerned, between each of the sources allocated to said domain.

More particularly, for storage means of the same type, the balancing means ensure the balancing of the load levels.

For example, the battery pack 16 is allocated to the first frequency range 62. The balancing means 82 then seek to balance the charge levels of the batteries of the battery pack 16. This avoids that one of the batteries are completely discharged while another would be fully charged.

In the same way, the block of supercapacitors 18 is allocated to the second frequency domain 64. The balancing means 84 then seek to balance the charge levels of the supercapacitors of the block of 30 supercapacitors 18. This prevents one of the supercapacitors completely unloaded while another would be fully loaded. Following the implementation of the balancing of the use of sources by the balancing means 82, 84, 86, new electrical energy instructions 88, 90, 92 to be provided by the allocated storage means respectively the first, second and third frequency domains are transmitted from the balancing means 82, 84, 86 to the transmission means 44 setpoints.

The diagram of FIG. 4 illustrates an example of management of the energy of the vehicle 2 by the processing means 42 when the vehicle 2 performs a local service between two stations. This diagram shows the evolution of the power sources supplied by the sources 12, 16, 18 and the power absorbed by the traction motors 6.

The diagram of FIG. 4 comprises 6 curves 100, 102, 104, 106, 108,

1 10.

The curve 100 represents the power absorbed by the traction motors 6 expressed in kilowatts (kW) as a function of time in seconds (s). This power is of negative sign since it is absorbed.

The curve 102 represents the power supplied by the generator set expressed in kilowatts (kW) as a function of time in seconds (s).

The curve 104 represents the power provided by the battery pack 16 expressed in kilowatts (kW) as a function of time in seconds (s).

Curve 106 represents the power provided by the block of supercapacitors 18 expressed in kilowatts (kW) as a function of time in seconds (s).

Furthermore, the curve 108 represents the speed of the vehicle in km / h as a function of time and the curve 1 represents the fuel consumption by the vehicle in liters per hour as a function of time.

The diagram of FIG. 5 is an extract of the diagram of FIG. 4 for the period of time between 0 and approximately 180 seconds.

At first, between 0 and 50 seconds, the machine is stopped. Indeed, the speed 108 of the machine 2 and the power 100 absorbed by the traction motors 6 are zero. Meanwhile, the battery pack 16 (negative power on curve 104) is recharged by the block of supercapacitors 18 (positive power on curve 106). The generator 12 is stopped during this time (zero power on the curve 102).

Then, in a second time between 50 and 75 seconds approximately, the machine 2 starts. The speed 108 is almost zero and the power 100 absorbed by the traction motors 6 increases slowly. This operation falls within the first frequency domain 62. Thus, in accordance with the strategy adopted by the frequency discrimination means 60, it is the battery block 16 that is biased (positive power on the curve 104 while the power is zero on the curves 102 and 106).

Then, in a third time between about 75 and 100 seconds, the traction demand becomes greater. The speed 108 increases and the power 100 absorbed by the traction motors 6 increases more rapidly. The battery pack 16 can no longer supply the energy necessary for the mission of the machine 2. The transfer means 74 then command the block of

Supercapacitors 18 complete the power supply although operation is still in the first frequency domain 62. The power provided by the battery pack 16 is at a maximum. There is indeed a plateau at the curve 104. The power provided by the block of supercapacitors 18 increases.

Then, in a fourth time between about 100 and 150 seconds, the power absorbed by the traction (curve 100) is constant while the speed of the machine (curve 108) increases. The battery pack 16 and the block of supercapacitors 18 provide the power necessary for the mission. There are indeed two levels at the curves 104 and 106.

At the time of approximately 150 seconds, the autonomy of the storage means calculated by the autonomy calculation means 48 becomes less than the threshold of 30 seconds. More particularly, as pointed out by the arrows 120, the autonomy of the block of supercapacitors 18 becomes insufficient to ensure the need for the traction mission of the machine 2. The control means 50

30 then control the start of the generator 12. The power supplied by the generator 12 (curve 102) increases gradually. The generator 12 first takes over the block of supercapacitors 18 and then the battery pack 16.

During the time period of approximately 400 to 700 seconds, the generator fully ensures the traction mission of the machine, the blocks of 5 batteries 16 and supercapacitors 18 providing no power.

At about 700 seconds, the traction is cut to reduce the speed of the machine 2 to park. As pointed by the arrows 130, the available energy of the generator set 12 allows charging of the battery packs 16 and supercapacitors 18 while the machine 2 is still in motion (curve 108). Charging supercapacitors is much faster than batteries.

Of course, other embodiments can still be envisaged.

It is thus possible to provide in the energy management system of the invention means of "eco-driving". By knowing the topology 15 of the line, the journey time, the stops in the station to be made, the composition of the craft, the expected traffic and in real time, such means have the function of anticipating the actions. the energy management system to achieve even more economical driving. For example, when approaching a descent, such "eco-driving" means can control the recharging of the storage means not by the generator but by a regenerative braking.

The energy management method described in the present application comprises the steps of:

- Definition of at least two frequency domains (62,64,66) of operation of the machine (2);

Allocating at least one of the plurality of sources to each defined frequency domain; and

- In each frequency domain, control of the machine (2) to use in priority the source allocated to said frequency domain during operation of the machine (2) in this area.

Said sources comprising energy storage means (14), said method comprises the steps of: - collecting information on the quantities of available electrical energy that can be provided by the storage means (14) and the maximum electrical power that can be absorbed by consumers (6,20);

5 - calculating an autonomy of the storage means (14), said autonomy being equal to the ratio between the electrical energy available in the storage means (14) and the maximum electrical power collected; and

- optimization of the use of sources (12,14) for the supply of electrical energy according to the calculated autonomy.

In addition, said method comprises, for each source, the steps of:

- definition of an electrical energy setpoint to be provided by said source;

acquiring information relating to at least one functional characteristic of said source;

Determining a quantity of electrical energy to be supplied by said source as a function of the functional characteristic of said source; and

if the setpoint is greater than the quantity of energy determined,

- selection of another source; and

controlling the other source to provide the difference in energy between the setpoint and the determined amount of energy.

Claims

A method for managing the energy of a railway vehicle (2) comprising a plurality of on-board power sources (12,14) and electrical energy consumers (6,20), said sources comprising means for storage

Of energy (14), said method comprising the steps of:

- collecting information on the quantities of available electrical energy that can be provided by the storage means (14) and the maximum electrical power that can be absorbed by consumers (6,20);

calculating an autonomy of the storage means (14), said autonomy being equal to the ratio between the electrical energy available in the storage means (14) and the maximum electrical power collected; and

optimization of the use of sources (12, 14) for the supply of electrical energy as a function of the calculated autonomy, this optimization step comprising the steps of:

Defining at least two frequency domains (62, 64, 66) for operating the machine (2);

allocating at least one storage means among the storage means (14) to each defined frequency domain; and

- In each frequency domain, control of the machine (2) to use in priority the storage means allocated to said frequency domain during operation of the machine (2) in this area.

2. Method according to claim 1, wherein the sources (12,14) comprise, besides the energy storage means (14), a source of energy

Main unit comprising a generator (12) and / or a fuel cell.

3. Method according to claim 2, wherein the optimization step comprises a step of controlling the start of the main energy source (12) if the autonomy is below a determined threshold.

30

4. The method of claim 3, wherein the threshold is between 20 and 40 seconds, preferably equal to 30 seconds.

The method of claim 3 or 4, wherein the step of controlling start includes a step of transmitting an energy command.

(52) electrical to the main power source (12).

The method of any one of the preceding claims, wherein the defined frequency domains comprise:

a first domain (62) with frequencies of less than about 10 MHz; and

a second domain (64) with frequencies greater than about 20 MHz.

5

The method of claim 6, wherein the storage means allocated to the first frequency domain is a battery pack (16).

The method of claim 6 or 7, wherein the storage means allocated to the second frequency domain is a block of supercapacitors.

(18).

The method of any one of the preceding claims, wherein the step of collecting information relating to the maximum electrical power.

15 likely to be absorbed by consumers (6,20) comprises a step of predicting this maximum power from recorded values of the electrical power absorbed by consumers during previous operations of the machine (2).

10. A system for managing the energy of a railway vehicle (2) comprising a plurality of on-board energy sources (12, 14) and consumers (6, 20) of electrical energy, said sources comprising energy storage means (14), said system comprising means for:

collecting (46) information relating to the quantities of available electrical energy that can be provided by the storage means (14) and to the maximum electrical power likely to be absorbed by the consumers (6, 20);

calculation of autonomy of the storage means (14), said autonomy being equal to the ratio between the electric energy available in the means of

Storage (14) and the maximum electrical power collected; and

optimization of the use of sources (12, 14) for the supply of electrical energy as a function of the calculated autonomy, said optimization means comprising:

allocating at least one storage means among the storage means (14) to each defined frequency domain; and - In each frequency domain, control of the machine (2) to use primarily the storage means allocated to said frequency domain during operation of the machine (2) in this area.