Designing a railway power supply network

Abstract

The invention relates to a method for designing a railway power supply network (BSVN). In the method, a plurality of simulated railway power supply networks (BSVN) are generated by a computer simulation on the basis of a plurality of flexible input parameters (FP). Furthermore, the quality of the simulated railway power supply networks (BSVN) is evaluated by a generated evolutionary algorithm (ALG) on the basis of a result (EG) of an objective function (ZFE) with respect to a property (E) which is to be improved by optimization. On the basis of the result (EG) of the objective function (ZFE), it is also determined whether improvement of the simulated railway power supply networks (BSVN) is possible. Furthermore, if it has been determined that improvement of the simulated railway power supply networks (BSVN) is possible, adapted flexible input parameters (AFP) are determined on the basis of the evaluation by the evolutionary algorithm (ALG). Alternatively, if it has been determined that improvement of the simulated railway power supply networks (BSVN) is no longer possible, the simulated railway power supply networks (BSVN) are output as optimal railway power supply networks (OBSVN). If it has been determined that improvement of the simulated railway power supply networks (BSVN) is possible, the above steps are repeated with the adapted flexible input parameters (AFP) as flexible input parameters (FP). The invention also relates to a design device (30).

Description

[0001] 202316706

[0002] 1

[0003] Description

[0004] Design of a railway power supply network

[0005] The invention relates to a method for designing a railway power supply network. The invention also relates to a design device.

[0006] Railway power supply systems are used to supply rail vehicles with direct current (DC) or alternating current (AC). These systems are currently designed, dimensioned, and configured for operation using simulation software, specifically parameterized. Many input parameters are required for simulation-based dimensioning. Some of these inputs are known, some are based on worst-case scenarios, and others are determined by the engineers' experience, resulting in a usable design. Further optimization of the railway power supply system can be achieved through iterative, manual refinement. This optimization is necessary due to the significant number of free variables and parameters, such as those related to substation location, compliance with relevant voltage limits according to standards, and so on.However, a high level of experience is required when designing a railway power supply system. Furthermore, it is not yet possible to design a technically and economically optimal system from all perspectives. In addition, a significant amount of time is needed to optimize a railway power supply system.

[0007] The design and calculation of a new railway power supply network or system is therefore technically complex, time-consuming, and costly. Due to the aforementioned circumstances, optimizing the design is also complex and thus time-consuming and costly. For this reason, optimization is usually kept very simple, often leaving additional optimization potential untapped. In particular, railway power supply systems could be more energy-efficient in operation, have higher availability, be easier to install, or have lower investment costs if they had an improved, optimized design.

[0008] The task, therefore, is to improve or even optimize railway power supply networks with regard to predefined criteria. 202316706

[0009] 2

[0010] This problem is solved by a method for designing a railway power supply network according to claim 1 and a design device according to claim 11.

[0011] In the inventive method for designing a railway power supply network, a plurality of simulated railway power supply networks are generated by a computer simulation based on a plurality of flexible input parameters. Unlike conventional methods, several variants of railway power supply networks are therefore generated first. The simulated railway power supply networks serve as an "initial population" that is adapted to optimization goals using an evolutionary algorithm.

[0012] The design of a railway power supply network and the parameters considered are described in "Introduction: Structure and Design of AC and DC Railway Power Supply Systems," presented by Martin Altmann and Dr. Alexej Halank. The script of this presentation is incorporated into the present patent application by reference to it.

[0013] Furthermore, the quality of the simulated railway power supply networks is evaluated based on the result of an objective function with respect to a property that is to be improved through optimization, using a generated evolutionary algorithm. The objective function indicates how "well" a predetermined property has been improved.

[0014] Furthermore, based on the result of the objective function, it is determined whether an improvement of the simulated railway power supply networks is possible.

[0015] If it is determined that an improvement of the simulated railway power supply networks is possible, adapted flexible input parameters are generated based on the evaluation by the evolutionary algorithm. In addition to evaluating the generated railway power supply networks, adapted railway power supply networks with adapted flexible input parameters are thus created, which can then be evaluated and optimized again with respect to predetermined properties. An evolutionary algorithm is understood to be an algorithm with which solution candidates, in this case railway power supply networks, are generated for a given task.

[0016] 3. A specific problem can be artificially evolved. Typical operations of such algorithms include selection operators, recombination operators, and mutation operators.

[0017] However, if, after applying the evolutionary algorithm based on the result of the objective function, it is determined that further improvement of the simulated railway power supply networks is no longer possible, a simulated railway power supply network based on the simulated networks is output as the optimal railway power supply network. Typically, the simulated railway power supply networks converge, and based on this convergence, a result in the form of an optimal railway power supply network can be output.

[0018] The preceding steps are repeated with the adapted input parameters as flexible input parameters in the case that, after applying the evolutionary algorithm based on the result of the objective function, it has been determined that a further improvement of the simulated railway power supply network is possible, with the adapted input parameters as flexible input parameters.

[0019] According to the invention, a design optimization of the free, flexible input parameters is performed automatically using a predefined objective function and an evolutionary algorithm. This saves material resources and significantly reduces the time required for design optimization. The complexity of the optimization is represented by the evolutionary algorithm. Furthermore, unconventional optimizations can be performed; that is, the entire solution space is searched, including areas where an expert would not look for an optimal solution. Advantageously, railway power supply networks of existing lines, new lines, or entire rail networks can be optimized with respect to countless definable parameters.

[0020] The design device according to the invention has a generation unit for generating a plurality of simulated railway power supply networks by means of a computer simulation based on a plurality of flexible input parameters.

[0021] Part of the design device according to the invention is also an evaluation unit for 202316706

[0022] 4

[0023] Evaluating the quality of simulated railway power supply networks based on the result of an objective function with respect to a property that can be improved by optimization, using a generated evolutionary algorithm.

[0024] The design device according to the invention also includes a determination unit for determining whether an improvement of the simulated railway power supply networks is possible, based on the result of the objective function.

[0025] Furthermore, the design device according to the invention includes an adaptation unit for determining adapted flexible input parameters based on the evaluation by the evolutionary algorithm, so that an improvement of the simulated railway power supply networks is possible.

[0026] The design device according to the invention further comprises an output interface for outputting the simulated railway power supply networks as optimal railway power supply networks, in the event that it has been determined that a further improvement of the simulated railway power supply networks no longer appears possible.

[0027] Part of the design device according to the invention is also an iteration unit. The iteration unit is configured to transmit the adapted input parameters as flexible input parameters to the generation unit. The adapted input parameters are used to repeat the optimization if the evolutionary algorithm determines that an improvement of the simulated railway power supply networks is possible, whereby the adapted input parameters are then used as flexible input parameters. The design device according to the invention shares the advantages of the inventive method for designing a railway power supply network.

[0028] A large proportion of the aforementioned components of the design device can be implemented wholly or partially as software modules in a processor of a corresponding computer system. A largely software-based implementation has the advantage that even previously used computer systems can be easily retrofitted by a software update to operate in the manner of the invention. In this respect, the problem is also solved by a corresponding computer program product with a computer program that can be directly loaded into a computer system, containing program sections to execute the steps of the inventive method 202316706.

[0029] 5. To execute the design of a railway power supply network when the program is executed in the computer system. Such a computer program product may, in addition to the computer program itself, include additional components such as documentation and / or additional components, including hardware components such as hardware keys (dongles, etc.) for using the software.

[0030] For transport to and / or storage on or in the computer system, a computer-readable medium, such as a memory stick, a hard drive, or other portable or permanently installed data storage device, can be used, on which the program sections of the computer program that can be read and executed by a computer system are stored. The computer system may, for example, have one or more cooperating microprocessors or similar components for this purpose.

[0031] The dependent claims and the subsequent description each contain particularly advantageous embodiments and further developments of the invention. In particular, the claims of one claim category may also be further developed analogously to the dependent claims of another claim category and their descriptive parts. Furthermore, within the scope of the invention, the various features of different embodiments and claims may also be combined to form new embodiments.

[0032] In one variant of the inventive method for designing a railway power supply network, a theoretical model of a railway power supply network with a plurality of flexible parameters is first determined by computer simulation in order to generate the simulated railway power supply networks. The theoretical model represents the most important functional properties of the railway power supply network. In particular, the theoretical model represents the expected energy flow and energy consumption during compliant operation of the railway power supply network.

[0033] In a subsequent step, the flexible parameters of the theoretical model that may be changed through optimization are defined. It is therefore quite possible that some flexible parameters, which in principle can be changed, should not or may not be changed. These then remain unchanged during optimization. 202316706

[0034] 6

[0035] Furthermore, constraints for the optimization are defined as conditions that must not be violated during the optimization process. These constraints are preferably based on limits set for safety reasons or arising from the goal of maintaining the functionality of the overall system.

[0036] Furthermore, an objective function is defined, specifying a result correlated with a property to be improved through optimization. Advantageously, structures and criteria are defined that can be used by the evolutionary algorithm to evaluate and generate adapted railway power supply networks and their associated adapted synthetic parameter values.

[0037] Preferably, the generation of simulated railway power supply networks by computer simulation includes the following steps:

[0038] An evolutionary algorithm generates and populates a data gene pool, preferably randomly selecting different parameter values ​​for the flexible parameters of the theoretical model. This data gene pool serves as the basis for further evolutionary steps. Since the parameter values ​​are preferably random, they are not yet optimal, but preferably provide a broad data foundation for later optimization. It is preferred that the data gene pool initially covers an "allowed" range of parameter values ​​as broadly and evenly as possible, so that the range of parameter values ​​for which an optimum is possible is also included.

[0039] The evolutionary algorithm continues to serve as a tool for finding suitable parameter values ​​for the flexible parameters. Based on these parameter values, railway power networks are simulated using computer simulation, and a result is then assigned to each simulation based on the objective function. Thus, a result is determined based on the objective function, which can be used to assess the quality of the parameter selection.

[0040] To optimize the data gene pool, parameter sets whose assigned result for the objective function deviates most significantly in the undesired direction from the other results are now eliminated from the data gene pool as part of the evaluation of the quality of the simulated railway power supply networks. 202316706

[0041] 7

[0042] Through this selection, "worse" parameter values ​​or sets of parameter values ​​are eliminated, and the data gene pool is gradually improved.

[0043] Once the data gene pool is populated, combinations of the previous input parameter values, or parameter values ​​from all individual gene pool calculations, are continuously generated. A computer simulation is then performed using the new input parameter values, or synthetic parameter values, generated by these combinations. Once the computer simulation is complete, the objective function value is determined. If this value is better than the worst-performing value in the gene pool, the worst-performing value is replaced by this new one. In this way, the gene pool is improved incrementally. Thus, the worst-performing variant is not first discarded and then a new one created, but rather the other way around.

[0044] The simulation and generation of new parameter values ​​and their selection are preferably continued until a termination condition for the optimization is met. This termination condition can, in particular, include the condition that it is determined that further improvement is no longer possible or to be expected. Advantageously, a non-convex complex optimization problem can also be satisfactorily solved.

[0045] In a preferred embodiment of the inventive method for designing a railway power supply network, the flexible parameters of the theoretical model include a substation location. The arrangement of the substations allows for influencing energy consumption. For example, overall energy consumption can be reduced by positioning substations where energy consumption is particularly high. In this way, line losses can be minimized.

[0046] Preferably, in the inventive method for designing a railway power supply network, the constraints include compliance with relevant voltage limits according to standards. Advantageously, this ensures that, despite optimization of the overall system, safety-relevant limits in particular are met.

[0047] In a preferred embodiment of the inventive method for designing a railway power supply network, the theoretical model comprises the following elements:

[0048] - the essential components for supplying railway power,

[0049] - the number and type of rail vehicles used, 202316706

[0050] 8

[0051] - the topography of the route,

[0052] - the timetable.

[0053] The elements mentioned are relevant for a variety of variables that can be optimized, such as energy consumption or resource consumption. An optimal design of these elements can be used to optimize the aforementioned potential target variables.

[0054] In a preferred embodiment of the inventive method for designing a railway power supply network, the constraints include at least one of the following conditions:

[0055] - Overhead line voltages must comply with EN 50 163,

[0056] - Rail potentials must comply with EN 50 122,

[0057] - Arrival and departure times of rail vehicles must be adhered to.

[0058] DIN standard EN 50163 concerns regulations for the use of supply voltages in railway networks. DIN standard 50122 covers regulations for stationary installations in railway transport, relating to electrical safety, earthing, and return conduction.

[0059] It is advantageous that essential framework conditions are left unchanged.

[0060] Preferably, the objective function evaluates at least one of the following properties of the railway power supply network that are to be optimized:

[0061] - a minimum cable diameter,

[0062] - the energy efficiency of the railway power supply network during operation,

[0063] - the initial costs of setting up the railway power supply network,

[0064] - the number of components required for the railway power supply network, in particular the number of substations required.

[0065] A minimal cable diameter allows for savings in costly materials. Optimized energy efficiency conserves resources and protects the environment. Minimizing the initial costs of setting up the railway power supply network reduces the overall system's base costs and improves its profitability and resource efficiency. 202316706

[0066] 9

[0067] It is particularly preferred that the aforementioned properties to be optimized are at least partially combined, so that an optimal compromise between different objectives is achieved.

[0068] Preferably, the evolutionary algorithm is set up to perform one of the following actions:

[0069] - to define a permissible combination of input parameters for the simulation,

[0070] - each independently start a simulation,

[0071] - To evaluate the simulation results with regard to the results of the objective function and the constraints.

[0072] Advantageously, a self-optimizing iterative process is used, with which an optimal combination of parameters and an associated optimal railway power supply network can be determined.

[0073] Preferably, the simulation includes one of the following steps:

[0074] - Determining another input parameter combination using the evolutionary algorithm,

[0075] - Performing another simulation,

[0076] - Evaluating the results of the simulation,

[0077] - Replacing one input parameter combination with another input parameter combination depending on its assigned rating.

[0078] Advantageously, through the regeneration, evaluation and selection of input parameter combinations, optimized combinations of input parameters are continuously enriched in the data gene pool, while unsuitable combinations are eliminated.

[0079] Preferably, steps for generating synthetic parameter values ​​using the evolutionary algorithm based on the data gene pool, for simulating railway power networks based on the synthetic parameter values, and for eliminating sets of parameter values ​​whose associated result for the objective function of the respective railway power network deviates most significantly from the other results are repeated iteratively until the termination condition is met or further improvement of the simulation result is no longer expected. Advantageously, the result can be further optimized in each iteration until the termination condition is met. 202316706

[0080] 10

[0081] The termination condition preferably consists of the expectation that no significant improvement will be achieved with further iterations.

[0082] In one embodiment of the design device according to the invention, the generation unit comprises a model determination unit for determining a theoretical model of a railway power supply network with a plurality of flexible parameters, a parameter setting unit for defining the flexible parameters of the theoretical model that may be modified by optimization, a constraint setting unit for defining constraints for the optimization as conditions that must not be violated during the optimization, and an objective function setting unit for defining an objective function with respect to a property that is to be improved by optimization. Advantageously, criteria for determining an optimized railway power supply network can be defined, which must be fulfilled. It is also advantageously defined which parameters can be varied for optimization.

[0083] In a preferred embodiment of the design device according to the invention, the generation unit comprises a gene pool unit for filling a data gene pool by means of an evolutionary algorithm, wherein different parameter values ​​are selected for the flexible parameters of the theoretical model.

[0084] The design equipment also preferably includes a simulation unit for simulating a railway power supply network based on parameter values ​​of the theoretical model.

[0085] Preferably, the design unit, in particular the evaluation unit, comprises an evolution unit for generating synthetic parameter values ​​using the evolutionary algorithm based on the data gene pool and an elimination unit for eliminating sets of parameter values ​​whose corresponding result for the objective function deviates most significantly from the other results. Advantageously, the data gene pool is expanded to provide the broadest possible basis for selecting suitable railway power supply networks. In this case, the selection is carried out by eliminating unsuitable examples, so that the more suitable ones advantageously remain in the data gene pool and can influence the generation of new synthetic parameter values. 202316706

[0086] 11

[0087] The invention is explained in more detail below with reference to the accompanying figures and exemplary embodiments. The figures show:

[0088] FIG 1 shows a flowchart illustrating a method for designing a railway power supply network according to an embodiment of the invention.

[0089] FIG 2 is a flowchart illustrating a method for designing a railway power supply network according to an alternative embodiment of the invention.

[0090] FIG 3 is a schematic representation illustrating a design device according to an embodiment of the invention,

[0091] FIG 4 is a schematic representation illustrating a design device according to an alternative embodiment of the invention.

[0092] FIG 5 is a diagram illustrating an optimization process based on a data gene pool.

[0093] FIG 1 shows a flowchart 100 illustrating a method for designing a railway power supply network BSVN according to an embodiment of the invention.

[0094] In step 1.1, a plurality of simulated railway power supply networks (BSVN) are generated by a computer simulation based on a plurality of flexible input parameters (FP).

[0095] In step 1.11, the quality of the simulated railway power supply networks BSVN is determined based on a result EG of an objective function ZF. E Regarding a property E, which is to be improved by optimization, determined by a generated evolutionary algorithm ALG.

[0096] In step 1.111, based on the result EG of the objective function ZFE, it is determined whether an improvement V of the simulated railway power supply networks BSVN is possible. 202316706

[0097] 12

[0098] In the event that step 1.111 determines that an improvement of the simulated railway power supply networks BSVN is no longer possible, which is indicated by "n" in FIG 1, the process proceeds to step 1.1V and the simulated railway power supply networks BSVN are output as optimal railway power supply networks OBSVN.

[0099] In the event that step 1.111 determines that an improvement V of the simulated railway power supply networks BSVN is possible, which is marked with “y” in FIG 1, the process proceeds to step 1.V and adapted flexible input parameters AFP are determined based on the evaluation by the evolutionary algorithm ALG.

[0100] Furthermore, steps 1.1 to 1.1V or I .V are repeated with the adapted flexible input parameters AFP as flexible input parameters FP in the event that it has been determined that an improvement of the simulated railway power supply networks BSVN is possible.

[0101] FIG 2 shows a flowchart 200 illustrating a method for designing a railway power supply network according to an alternative embodiment of the invention.

[0102] In step 2.1, a theoretical model TM of a railway power supply network with a plurality of flexible parameters FP is determined.

[0103] In step 2.11, variable flexible parameters VFP of the theoretical model TM, which may be changed through optimization, are defined.

[0104] In step 2.III, constraints (NB) for the optimization are defined as conditions that must not be violated during the optimization.

[0105] In step 2.1V, an objective function ZFE is defined with respect to a property E, which is to be improved through optimization.

[0106] In step 2. V, a data gene pool DGP is processed by an evolutionary algorithm ALG, with different parameter values ​​PW for the flexible parameters FP 202316706.

[0107] 13 of the theoretical model TM is selected and a simulation S of a railway power supply network BSVN is carried out.

[0108] In step 2. VI, synthetic parameter values ​​SPW are generated by the evolutionary algorithm ALG based on the data gene pool DGP, and a simulation S of railway power networks BSVN is carried out based on these synthetic parameter values ​​SPW.

[0109] In step 2. VII, parameter value sets SSPW, whose associated result EG of the objective function ZF are used. E If a parameter set exhibits a value that deviates most significantly from the other results, it is eliminated. It is also possible that the currently investigated set of parameter values ​​will be eliminated again because it is no better than the existing ones.

[0110] In step 2.VIII, it is checked whether a termination condition BA for the optimization is met. If the termination condition is met, which is indicated by "y" in FIG. 1, the process proceeds to step 2.IX and the application of the evolutionary algorithm ALG and the simulation S is terminated, which is symbolized by "EXT" in FIG. 2. If the termination condition is not yet met, which is indicated by "n" in FIG. 1, the process returns to step 2.VI and the application of the evolutionary algorithm ALG and the simulation S continues.

[0111] FIG 3 shows a schematic representation illustrating a design device 30 according to an embodiment of the invention.

[0112] The design device 30 has a generation unit 31 for generating a plurality of simulated railway power supply networks BSVN by means of a computer simulation S based on a plurality of flexible input parameters FP.

[0113] Part of the design device 30 is also an evaluation unit 32 for evaluating the quality of the simulated railway power supply networks BSVN based on a result EG of an objective function ZF. E Regarding a property E of a railway power supply network BSVN, which is to be improved by optimization using an evolutionary algorithm ALG. 202316706

[0114] 14

[0115] The design device 30 also includes a determination unit 33 for determining whether an improvement V of the simulated railway power supply networks BSVN is possible, based on the result EG of the objective function ZFE.

[0116] The design device 30 also includes an adaptation unit 34 for determining adapted flexible input parameters AFP based on the evaluation by the evolutionary algorithm ALG, in the event that it has been determined that an improvement V of the simulated railway power supply networks BSVN is possible.

[0117] The design device 30 includes an output interface 35 for outputting a simulated railway power supply network BSVN as the optimal railway power supply network OBSVN in the event that the evolutionary algorithm ALG has determined that further improvement of the simulated railway power supply networks BSVN is no longer possible. It should be noted that a convergence of the simulated railway power supply networks BSVN towards the optimal railway power supply network OBSVN can be observed during the optimization process.

[0118] Part of the design unit 30 is also an iteration unit 36 ​​for transmitting the adapted flexible input parameters AFP as flexible input parameters FP to the generation unit 31 for repeating the optimization in the case that (not shown in FIG 3, see e.g. FIG 1, FIG 2) it has been determined that an improvement of the simulated railway power supply network BSVN is possible, using the adapted flexible input parameters AFP as flexible input parameters FP. Thus, in one iteration, the adapted flexible input parameters AFP take over the function of the flexible input parameters FP.

[0119] FIG 4 shows a schematic representation illustrating a design device 30 according to an alternative embodiment of the invention.

[0120] In particular, the generation unit 31 and evaluation unit 32, already shown in FIG. 3, are depicted in detail in FIG. 4. Furthermore, FIG. 4 also shows the determination unit 33, the adaptation unit 34, the output interface 35, and the iteration unit 36, all illustrated in FIG. 3. 202316706

[0121] 15

[0122] The generation unit 31 includes a model determination unit 31a. The model determination unit 31a is configured to determine a theoretical model TM of a railway power supply network BSVN with a plurality of flexible parameters FP.

[0123] Part of the design device 30 is a parameter setting unit 31b for setting the flexible parameters VFP of the theoretical model TM, which may be changed by optimization.

[0124] The design device 30 according to the invention also includes a constraint definition unit 31c, which is configured to define constraints NB for the optimization as conditions that must not be violated during the optimization. The constraints NB include, for example, regulations that must be complied with. Such regulations relate in particular to overhead line voltages that must be complied with in accordance with EN 50 163, rail potentials that must be complied with in accordance with EN 50 122, and arrival and departure times of rail vehicles that are prescribed by a timetable.

[0125] Part of the design device 30 is also an objective function definition unit 31 d for defining an objective function ZFE with respect to a property which is to be improved by optimization.

[0126] The design unit 30 also includes a gene pool unit 31e for populating a data gene pool DGP using an evolutionary algorithm ALG, whereby different parameter values ​​PW are selected for the flexible parameters FP of the theoretical model TM. Here, too, a simulation S of a railway power supply network BSVN is performed for each set of different parameter values ​​PW by a simulation unit 31f, which is also part of the generation unit 31, in order to generate a result EG of the objective function ZF. E to obtain.

[0127] The evaluation unit 32, also shown in FIG. 3, includes an evolution unit 32a. Evolution unit 32a is configured to expand the data gene pool DGP by generating synthetic parameter values ​​SPW using the evolutionary algorithm ALG based on the data gene pool DGP. After the evolutionary process, i.e., the application of the evolutionary algorithm ALG, the simulation unit 31f is accessed again, based on the synthetic parameter values ​​SPW 202316706.

[0128] 16 each simulate a railway power supply network BSVN and also generate an associated result EG of the target function ZFE.

[0129] Part of the evaluation unit 32 is also an elimination unit 32b. The elimination unit 32b is designed to eliminate, within the framework of an optimization process, parameter value sets PW, SPW, whose associated result EG of the objective function ZFE has a value that deviates most significantly from the other results.

[0130] As already mentioned, the structure of the remaining components 33, 34, 35, 36 corresponds to the arrangement shown in FIG 3.

[0131] Figure 5 shows diagram 50 with an upper and a lower sub-diagram, illustrating an optimization process based on a data gene pool. The optimization example shown in Figure 5 is a DC railway power supply network with a first substation UWA and a second substation UWB, as required, for example, for a tramway.

[0132] The upper part of FIG. 5 illustrates the results of an objective function ZFE, which was already mentioned in connection with FIG. 1 to FIG. 3. In the example shown in FIG. 5, the objective function ZFE represents the energy required per second. During optimization, this value decreases from over 150,000 kWs to just over 40,000 kWs.

[0133] The lower part of FIG. 5 illustrates a distribution of values ​​of the corresponding adapted flexible input parameters AFP and their number Z(AFP). In the case shown in FIG. 5, the adapted flexible input parameters AFP are the feed-in voltages UE[ ] of two substations UWA and UWB. Symbols marked with a plus sign “+” each represent a feed-in voltage UE[ ] of a first substation UWA, and symbols marked with an x ​​sign “x” each represent a feed-in voltage UE[ ] of a second substation UWB. As can be seen in FIG. 5, the substation voltages or feed-in voltages UE[ ] have values ​​between 500 V and 900 V at the beginning of the optimization, and during the optimization, the substation voltages or feed-in voltages UE[ ] each converge to a value approximately 900 V. 202316706

[0134] 17

[0135] Finally, it should be noted once again that the methods and devices described above are merely preferred embodiments of the invention and that the invention can be varied by a person skilled in the art without departing from the scope of the invention, insofar as it is defined by the claims. For the sake of completeness, it should also be noted that the use of the indefinite articles "a" or "an" does not preclude the possibility that the features in question may be present multiple times. Likewise, the term "unit" does not preclude the possibility that it consists of several components, which may also be spatially distributed. Regardless of the grammatical gender of a particular term, persons of male, female, or other gender identities are included.

Claims

202316706 18 Patent claims 1. A method for designing a railway power supply network (BSVN), comprising the steps of: i) generating a plurality of simulated railway power supply networks (BSVN) by a computer simulation (S) based on a plurality of flexible input parameters (FP), ii) evaluating the quality of the simulated railway power supply networks (BSVN) based on a result (EG) of an objective function (ZF). E ) with regard to a property (E) which is to be improved by optimization, iii) Determine whether an improvement of the simulated railway power supply networks (BSVN) is possible, based on the result (EG) of the objective function (ZF) E), iv) Determining adapted flexible input parameters (AFP) by an evolutionary algorithm (ALG) based on the evaluation, in the case that step iii) determined that an improvement (V) of the simulated railway power supply networks (BSVN) is possible, or outputting the simulated railway power supply networks (BSVN) as optimal railway power supply networks (OBSVN) in the case that step iii) determined that an improvement (V) of the simulated railway power supply networks (BSVN) is no longer possible, v) Repeating steps i) to iv) in the case that an improvement (V) of the simulated railway power supply networks (BSVN) is possible, with the adapted flexible input parameters (AFP) as flexible input parameters (FP).

2. The method of claim 1, wherein the generation of the simulated railway power supply networks (BSVN) by a computer simulation (S) comprises the steps: - Determining a theoretical model (TM) of a railway power supply network (BSVN) with a plurality of flexible input parameters (FP), - Defining the flexible input parameters (FP) of the theoretical model (TM) that may be changed through optimization, - Defining constraints (CS) for optimization as conditions that must not be violated during optimization, - Defining an objective function (IF) E ) with regard to a property (E) which can be improved by optimization. 202316706 19 3. Method according to claim 1 or 2, wherein - the generation of the simulated railway power supply networks (BSVN) by a computer simulation (S) which includes the following steps: - Filling a data gene pool (DGP) using the evolutionary algorithm (ALG), selecting different parameter values ​​(PW) for the flexible parameters (FP) of the theoretical model (TM), - Generating synthetic parameter values ​​(SPW) using the evolutionary algorithm (ALG) based on the data gene pool (DGP), - Generating simulated railway power supply networks (BSVN) through computer simulation (S) based on synthetic parameter values ​​(SPW), and evaluating the quality of the simulated railway power supply networks (BSVN) includes the following step: - Eliminating parameter value sets (PVS) whose assigned result (V) of the objective function (FFE) has a value that differs most from the other results (VS).

4. Method according to any of the preceding claims, wherein the flexible parameters (FP) of the theoretical model (TM) comprise a sub-works position.

5. Method according to claim 1 or 2, wherein the constraints (SC) include compliance with relevant voltage limits according to the standard.

6. A method according to any of the preceding claims, wherein the constraints (CS) comprise at least one of the following: - Overhead line voltages must comply with EN 50 163, - Rail potentials must comply with EN 50 122, - Arrival and departure times of rail vehicles must be adhered to.

7. A method according to any of the preceding claims, wherein the theoretical model (TM) comprises the following elements: - the essential components for supplying railway power, - the number and type of rail vehicles used, 202316706 20 - the topography of a road, - a timetable.

8. Method according to one of the preceding claims, wherein the objective function (ZFE) evaluates at least one of the following properties (E) of the railway power supply network (BSVN) to be optimized: - a minimum cable diameter, - the energy efficiency of the railway power supply network (BSVN) during operation, - the initial costs of setting up the railway power supply network (BSVN), - the number of components required for the railway power supply network (BSVN), in particular the number of substations required.

9. A method according to any of the preceding claims, wherein the evolutionary algorithm (ALG) is configured to - to define a permissible input parameter combination (PW, SPW) for the computer simulation (S), - each independently start a computer simulation (S), - To evaluate the results of the computer simulation (S) with regard to the results of the objective function (ZFE) and the constraints (NB).

10. Method according to any one of the preceding claims, wherein performing the simulation (S) comprises one of the following steps: - Determining another input parameter combination (IPC) using the evolutionary algorithm (ALG), - Performing another computer simulation (S), - Evaluating the result (EG) of the computer simulation (S), - Replacing one input parameter combination (PW) with another input parameter combination (SPW) depending on its assigned rating.

11. Design device (30), comprising: - a generation unit (31) for generating a plurality of simulated railway power supply networks (BSVN) by a computer simulation (S) based on a plurality of flexible input parameters (FP), - an evaluation unit (32) for evaluating the quality of the simulated railway power supply networks (BSVN) based on a result (EG) of an objective function (ZFE) 202316706 21 with regard to a property (E) which is to be improved by optimization by a generated evolutionary algorithm (ALG), - an investigation unit (33) to determine whether an improvement (V) of the simulated railway power supply networks (BSVN) is possible, based on the result (EG) of the objective function (ZFE), - an adaptation unit (34) for determining adapted flexible input parameters (AFP) based on the evaluation by the evolutionary algorithm (ALG) in the event that it has been determined that an improvement (V) of the simulated railway power supply networks (BSVN) is possible, - an output interface (35) for outputting the simulated railway power supply networks (BSVN) as optimal railway power supply networks (OBSVN) in the event that it has been determined that an improvement of the simulated railway power supply networks (BSVN) is no longer possible, - an iteration unit (36) for transmitting the adapted flexible input parameters (AFP) as flexible input parameters (FP) to the generation unit (31) for repeating the computer simulation (S) in the event that it has been determined that an improvement of the simulated railway power supply networks (BSVN) is possible, with the adapted flexible input parameters (AFP) as flexible input parameters (FP).

12. Design device (30) according to claim 11, wherein the generation unit (31) comprises: - a model determination unit (31a) for determining a theoretical model (TM) of a railway power supply network (BSVN) with a plurality of flexible parameters (FP), - a parameter setting unit (31b) for setting the flexible parameters (VFP) of the theoretical model (TM) that may be changed by optimization, - a constraint specification unit (31c) for specifying constraints (CS) for the optimization as conditions that must not be violated during the optimization, - an objective function definition unit (31d) for defining an objective function (ZFE) with respect to a property (E) which is to be improved by optimization.

13. Design device (30) according to claim 11 or 12, wherein - the generation unit (31) has: 202316706 22 - a gene pool unit (31 e) for populating a data gene pool (DGP) by the evolutionary algorithm (ALG), whereby different parameter values ​​(PW) are chosen for the flexible parameters (FP) of the theoretical model (TM), - a simulation unit (31 f) for simulating a railway power supply network (BSVN) based on parameter values ​​(PW, SPW) of the theoretical model (TM), and - the assessment unit (32) exhibits: - an evolutionary unit (32a) for generating synthetic parameter values ​​(SPW) by the evolutionary algorithm (ALG) based on the data gene pool (DGP) and - an elimination unit (32b) for eliminating parameter value sets (SSPW), whose associated result (EG) is the objective function (ZF) E ) shows a value that differs most significantly from the other results (EC).

14. Computer program product comprising instructions which, when the program is executed by a computer, cause it to perform the steps of a method according to any one of claims 1 to 10.

15. Computer-readable storage medium comprising instructions which, when executed by a computer, cause it to perform the steps of a method according to claims 1 to 10.

Images