Topological structure optimization method and system for hydrogen energy driven train system

Abstract

The invention provides a topological structure optimization method and system for a hydrogen energy driven train system, and belongs to the field of simulation and optimization of train systems. According to the optimization method, the hydrogen energy driven train system is divided into a mechanical part, an electrical part, an information part and a gas-liquid part, each part corresponds to one kind of parts, and all elements of the system are divided into four kinds of parts; constructing a mechanical-electrical-information-gas-liquid dependent network model by considering each part and the action relationship among the parts; by considering the physical attributes of the four parts, a weighting method of the action relation between the parts is provided; based on the weight and a mechanical-electrical-information-gas-liquid dependent network model, a weighted topology network reliability calculation method is provided; and taking the action relationship between the components as a decision variable, taking improvement of the reliability of the weighted network as an optimization target, and performing solution based on an improved simulated annealing algorithm to complete optimization of the topological structure. According to the method, the topological structure optimization effect of the hydrogen energy driven train system is improved.

Description

Technical Field

[0001] The present invention belongs to the field of simulation and optimization of train systems, and in particular relates to a topology optimization method and system for a hydrogen-powered train system. Background Art

[0002] As a clean, zero-carbon renewable energy source, hydrogen energy only produces water as a combustion product, which does not pollute the environment. Applying hydrogen energy to the train sector can significantly reduce carbon emissions during train operation and promote the green transformation of the transportation sector. In addition, hydrogen-powered trains have high energy conversion efficiency, which can minimize energy waste and significantly improve the operating efficiency of the train. At the same time, hydrogen trains have a fast hydrogen refueling speed, which greatly shortens the train's dwell time and further improves operational efficiency. However, the normal operation of the hydrogen train system includes components such as the fuel cell stack and hydrogen cylinders, which increases the complexity of the train system in terms of train structure, and thus faces significant risks in terms of reliability. Therefore, the topology of the train system needs to be optimized.

[0003] Existing technologies typically focus on optimizing train topology, specifically the communication network used for data transmission, to ensure compatibility between control commands between various components. However, issues such as the layout of other components and the construction of a global network model remain, including low topology reliability and power conversion efficiency. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, the present invention aims to provide a method and system for optimizing the topological structure of a hydrogen-powered train system, which divides the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and constructs a four-layer dependent network model; based on the physical meaning of the four components, a method for calculating the weighted reliability of the topological network and an optimization model are proposed. By considering the probability of accepting a solution in the simulated annealing algorithm, the simulated annealing algorithm is improved and applied to the optimization problem of the weighted reliability of the topological network, thereby optimizing the topological structure of the hydrogen-powered train system.

[0005] In order to achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for optimizing the topology of a hydrogen-powered train system, the method comprising the following steps:

[0007] Step S1, dividing the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and dividing all components in the system into the corresponding four types of components;

[0008] Step S2: constructing a mechanical network topology model, an electrical network topology model, an information network topology model, and a gas-liquid network topology model based on the four parts of mechanical, electrical, information, and gas-liquid, respectively, according to the components in each component and the interaction relationship between the components;

[0009] Step S3, integrating the four network models to obtain a mechanical-electrical-information-gas-liquid interdependent network model;

[0010] Step S4, respectively calculating the mechanical network node weights, electrical network node weights, information network node weights, and gas-liquid network node weights in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information, and gas-liquid networks;

[0011] Step S5, constructing a weighted matrix based on the four network node weights, wherein the weighted matrix includes intra-network weights and inter-network weights;

[0012] Step S6, screening the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix;

[0013] Step S7: constructing a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction between the components as a decision variable and takes improving the weighted network reliability index as an optimization goal;

[0014] Step S8: Solve the weighted network reliability optimization model based on the improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

[0015] As a preferred embodiment of the present invention, in step S2:

[0016] In a mechanical network MN, nodes represent mechanical components and edges represent mechanical action relationships. The mechanical network is represented by G M =(V M , E M ) is represented by, where V M represents the set of nodes in the mechanical network, E M Represents the set of edges in the mechanical network; in the mechanical network G M In the mechanical network, if a fastener connects the mechanical component α corresponding to node i and the mechanical component β corresponding to node j, and the failure of the mechanical component α will cause the loss of the function of the mechanical component β, then there is an edge from node i to node j in the mechanical network;

[0017] In the electrical network EN, nodes represent electrical components and edges represent electrical interaction relationships. The electrical network is represented by G E =(V E , E E) is represented by, where V E Represents the set of nodes in the electrical network, E E Represents the set of edges in the electrical network; in the electrical network G E In the network, if the electrical component γ corresponding to the node p and the electrical component δ corresponding to the node q have a current action relationship, then there is an undirected edge between the node p and the node q in the electrical network;

[0018] In the information network CN, nodes represent information components and edges represent information interaction relationships. The information network is represented by G C =(V C , E C ) is represented by, where V C represents the set of nodes in the information network, E C Represents the set of edges in the information network; in the information network G C In the information network, if the information component ε of the corresponding node u and the information component ∈ of the corresponding node v have an information interaction relationship, then there is an undirected edge between the node u and the node v in the information network;

[0019] In the gas-liquid network GN, nodes represent components related to gas and liquid and edges represent the input or output of gas or liquid. G =(V G , E G ) is represented by, where V G represents the set of nodes in the gas-liquid network, E G Represents the set of edges in the gas-liquid network; in the gas-liquid network G G In the gas-liquid network, if there is a flow of gas or liquid between the gas-liquid component ζ corresponding to the node s and the gas-liquid component η corresponding to the node t, then there is an undirected edge between the node s and the node t in the gas-liquid network.

[0020] As a preferred embodiment of the present invention, in step S3, when constructing the dependent network model, dependent edges between the networks are added between the same physical elements in the four networks, thereby constructing a mechanical-electrical-information-gas-liquid dependent network model.

[0021] As a preferred embodiment of the present invention, in step S4, when calculating the weight of the mechanical network node, the support of each component in the component is taken into consideration, and the functional support is used as the weight value between the components in the mechanical network; when calculating the weight of the electrical network node, the power of each component in the component is taken into consideration, and the power value is used as the weight value between the components in the electrical network; when calculating the weight of the information network node, the data load of each component in the component is taken into consideration, and the data load is used as the weight value between the components in the information network; when calculating the weight of the gas-liquid network node, the flow of each component in the component is taken into consideration, and the flow value is used as the weight value of each component in the gas-liquid network.

[0022] As a preferred embodiment of the present invention, the weight of node i in the mechanical network The definition is as follows:

[0023]

[0024] In formula (1), a ij represents the element in the i-th row and j-th column of the network adjacency matrix A; for a directed network, if there is an edge from node i to node j, then a ij =1; otherwise a ij =0;N M represents the number of nodes in the mechanical network;

[0025] The weight of the electrical node p It is calculated from its power, which is defined as follows:

[0026]

[0027] In formula (2), u p represents the voltage at electrical node p, I op represents the current flowing out of the electrical node p;

[0028] And according to Kirchhoff's law, the voltage and current of the electrical node are calculated as follows:

[0029] Y*U=I (20)

[0031] In formula (3), Y is the admittance matrix, y pq represents the element in the pth row and qth column of the node admittance matrix in the electrical network, and y pp =-∑ p≠q y pq , U is the voltage vector of the node, I is the current vector of the node; then the current flowing through edge pq is I pq The definition is as follows:

[0032] I pq =(u p -u q)*Y pq (twenty one);

[0034] The weight of information node u Characterized by data traffic load;

[0035] In the gas-liquid network, weights are defined based on the gas and liquid flows of component i in actual operation.

[0036] The weights between mechanical components ij, electrical components pq, information components uv, and gas-liquid components st and The definition is as follows:

[0037]

[0038] In formulas (5)-(8), and Indicates the adjacency relationship between mechanical components ij, electrical components pq, information components uv, and gas-liquid components st. If they are connected, and otherwise and

[0039] As a preferred embodiment of the present invention, the weighting matrix M in step S5 W , defined as follows:

[0040]

[0041] In formula (9), M M , M E , M C and M G Represents the weighted matrix of the four networks of mechanical, electrical, information, and gas-liquid; in the weighted matrix M M , M E , M C and M G The elements in the network are the weights of the relationships between components in the mechanical, electrical, information, and gas-liquid networks. and Weight matrix M ME ,M MC ,M MG ,M EC ,M EG and M CG It represents the importance of the dependent edges between networks and is calculated by multiplying the weight importance of the nodes.

[0042] As a preferred embodiment of the present invention, in step S6, according to the weighting matrix M W, construct the weighted distance matrix M D , where the elements in the weighted distance matrix is defined as follows:

[0043]

[0044] In formula (10), Represents the matrix M W Any element in ;

[0045] According to the value of the weighted distance matrix, by using the Floyd algorithm to calculate the weighted shortest path length between nodes, the weighted network efficiency value can be obtained, that is, the weighted network reliability index NE is as follows:

[0046]

[0047] In formula (11), d hk It represents the weighted shortest path length between nodes hk, and N represents the number of nodes in all networks; nodes hk include any two nodes in the four networks of mechanical, electrical, information, and gas-liquid.

[0048] As a preferred embodiment of the present invention, in step S7, the connection relationship between components is used as a decision variable, and the degree value of the node is used as a constraint to construct a topology optimization model of the hydrogen train system, as shown in formula (12):

[0049]

[0050] The constraints are:

[0051]

[0052] a hh =0 (31)

[0053] a hk ∈{0,1} (32)

[0054] In formulas (12)-(15), d h represents the degree value of node h, a hk It represents the adjacency relationship between any nodes hk in the same network.

[0055] As a preferred embodiment of the present invention, step S8 includes:

[0056] Step S81: for a given mechanical-electrical-information-gas-liquid interdependent network G, M ,p E ,p C and p G The probability of selecting the mechanical, electrical, information, and gas-liquid subnets isM ,p E ,p C and p G The value of is the ratio of the number of edges in the four subnetworks to the total number of edges in the dependent network;

[0057] Step S82: After selecting the corresponding subnet, randomly select two edges in the network to exchange, and obtain a newly generated dependent network G′. Determine whether the newly generated dependent network G′ satisfies constraints (13) and (14). If so, proceed to step S83; otherwise, return to step S81.

[0058] Step S83, calculating the weighted matrix, the weighted distance matrix and the weighted network reliability value NE' according to the new dependency network G';

[0059] Step S84: If the weighted network reliability value NE′ in the new network is better than the weighted network reliability value in the current network, the newly generated network G′ is accepted and the current solution is updated to G′; if the weighted network reliability value NE′ in the new network is not better than the weighted network reliability value in the current network, the probability p of accepting the solution is calculated as follows:

[0060]

[0061] In formula (16), T represents temperature, λ is the parameter value of the accepted solution, and the calculation formula is as follows:

[0062]

[0063] In formula (17), p c represents the critical probability of accepting a worse solution, μ is the critical relative proportion, T max is the maximum temperature;

[0064] Step S85, determine whether the number of iterations t meets the maximum number of iterations; if so, output the optimal network structure; otherwise, set T = T*0.95, t = t+1, and return to step S81.

[0065] In a second aspect, an embodiment of the present invention further provides a topology optimization system for a hydrogen-powered train system, the system comprising a component classification module, a subnetwork construction module, a dependent network construction module, a subnetwork node weight calculation module, a weighted matrix construction module, a reliability index calculation module, an optimization model construction module, and a solution and optimization module; wherein,

[0066] The component classification module is used to divide the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and to divide all components in the system into four types of parts;

[0067] The sub-network construction module is used to construct a mechanical network topology structure model, an electrical network topology structure model, an information network topology structure model, and a gas-liquid network topology structure model based on the four parts of mechanical, electrical, information, and gas-liquid according to the components in each component and the interaction relationship between the components;

[0068] The interdependent network construction module is used to fuse the four network models to obtain a mechanical-electrical-information-gas-liquid interdependent network model;

[0069] The sub-network node weight calculation module is used to calculate the mechanical network node weight, electrical network node weight, information network node weight and gas-liquid network node weight in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information and gas-liquid networks respectively;

[0070] The weighted matrix construction module is used to construct a weighted matrix according to four network node weights, wherein the weighted matrix includes intra-network weights and inter-network weights;

[0071] The reliability index calculation module is used to screen the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix;

[0072] The optimization model construction module is used to construct a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction relationship between the various components as a decision variable and takes improving the weighted network reliability index as an optimization goal;

[0073] The solution and optimization module is used to solve the weighted network reliability optimization model based on an improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

[0074] The technical solution provided by the embodiment of the present invention has the following beneficial effects:

[0075] The topology optimization method and system of the hydrogen-powered train system provided in the embodiment of the present invention divide the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid. Each part corresponds to a category of components, and all elements of the system are divided into four types of components. Considering the interaction relationship between each part and the interaction relationship between parts, a mechanical-electrical-information-gas-liquid interdependent network model is constructed. By considering the physical properties of the four components, a weighted method for the interaction relationship between components is proposed. Based on the weight of the interaction relationship and the mechanical-electrical-information-gas-liquid interdependent network model, a weighted topology network reliability calculation method for the hydrogen-powered train system is proposed. The interaction relationship between components is used as a decision variable, and improving the weighted network reliability is taken as the optimization goal. The solution is based on the improved simulated annealing algorithm to complete the optimization of the topology structure.

[0076] Of course, it is not necessary to achieve all of the advantages described above simultaneously in order to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0078] Figure 1 This is a schematic diagram of the topological structure optimization principle of the hydrogen-powered train system according to an embodiment of the present invention;

[0079] Figure 2 is a flow chart of a method for optimizing the topology of a hydrogen-powered train system according to an embodiment of the present invention;

[0080] Figure 3 is a schematic diagram of the mechanical network topology structure optimized in an embodiment of the present invention;

[0081] Figure 4 is a schematic diagram of the electrical network topology optimized in an embodiment of the present invention;

[0082] Figure 5 is a schematic diagram of the information network topology structure optimized in an embodiment of the present invention;

[0083] Figure 6 Schematic diagram of the gas-liquid network topology optimized in an embodiment of the present invention;

[0084] Figure 7 is a weighted reliability optimization iteration graph of the topology structure in an embodiment of the present invention;

[0085] Figure 8 3 is a comparison diagram of the topological structure of the train system before and after optimization in an embodiment of the present invention. DETAILED DESCRIPTION

[0086] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations. It should be noted that the embodiments of the present invention and the features in the embodiments can also be combined with each other in the absence of conflict.

[0087] It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures. In the description of the present invention, the terms "first," "second," "third," "fourth," etc. are used only to distinguish the description and are not to be understood as indicating or implying relative importance.

[0088] Based on the structural optimization problem of a hydrogen-powered train system, an embodiment of the present invention provides a topology optimization method and system for a hydrogen-powered train system. The hydrogen-powered train system is divided into four parts: mechanical, electrical, information, and gas-liquid. Each part corresponds to a category of components, and all components of the system are divided into four types of components. Considering the interaction relationship between each part and between parts, a mechanical-electrical-information-gas-liquid interdependent network model is constructed. By considering the physical properties of the four components, a weighted method for the interaction relationship between components is proposed. Based on the weight of the interaction relationship and the mechanical-electrical-information-gas-liquid interdependent network model, a weighted topology network reliability calculation method for the hydrogen-powered train system is proposed. The interaction relationship between components is used as a decision variable, and improving the reliability of the weighted network is taken as the optimization goal. The solution is based on an improved simulated annealing algorithm to complete the optimization of the topology structure.

[0089] like Figure 1 and Figure 2 As shown, the topology optimization method of the hydrogen-powered train system provided in the embodiment of the present invention includes the following steps:

[0090] Step S1, divide the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and divide all components in the system into the corresponding four parts.

[0091] Step S2, according to the elements in each component and the interaction relationship between the elements, construct a mechanical network topology model, an electrical network topology model, an information network topology model and a gas-liquid network topology model based on the four parts of mechanical, electrical, information and gas-liquid respectively.

[0092] In this step, in each network topology model, the components constitute the points of the topology model, and the interaction relationships between the components constitute the edges of the topology model. The components refer to the smallest maintainable units in the train system.

[0093] Specifically, if Figure 3 As shown in the figure, in the machine network (MN), nodes represent mechanical parts and edges represent mechanical action relationships. The machine network is represented by G M =(V M , E M ) is represented by V M represents the set of nodes in the mechanical network, EM Represents the set of edges in a mechanical network. Based on engineers' practical experience, if a fastener (such as a bolt, screw, weld, etc.) connects a mechanical component α (corresponding to node i) and a mechanical component β (corresponding to node j), and the failure of mechanical component α will cause the function of mechanical component β to be lost, then there is an edge in the mechanical network pointing from node i to node j. By determining whether the above-mentioned interaction relationship exists between all mechanical components, the mechanical network model G is constructed. M .

[0094] like Figure 4 As shown in the figure, in the electrical network (EN), nodes represent electrical components and edges represent electrical interaction relationships. The electrical network is represented by G E =(V E , E E ) is represented by V E Represents the set of nodes in the electrical network, E E Represents the set of edges in the electrical network. If there is a current interaction relationship between electrical component γ (corresponding to node p) and electrical component δ (corresponding to node q), then there is an undirected edge between node p and node q in the electrical network. By determining whether the above interaction relationship exists between all electrical components, the electrical network model G is constructed. E .

[0095] like Figure 5 As shown in the figure, in the information network (CN), nodes represent information components and edges represent information interaction relationships. The information network is represented by G C =(V C , E C ) is represented by V C Represents the set of nodes in the information network, E C Represents the set of edges in the information network. If there is an information interaction relationship between information component ε (corresponding to node u) and information component ∈ (corresponding to node v), then there is an undirected edge between node u and node v in the information network. By determining whether the above interaction relationship exists between all information components, the information network model G is constructed. C .

[0096] like Figure 6 As shown in the figure, in the gas-liquid network (GN), nodes represent components related to gas and liquid and edges represent the input or output of gas or liquid. G =(V G , E G ) is represented by V Grepresents the set of nodes in the gas-liquid network, E G Represents the set of edges in the gas-liquid network. If there is a flow of gas or liquid between the gas-liquid component ζ (corresponding to node s) and the gas-liquid component η (corresponding to node t), then there is an undirected edge between nodes s and t in the gas-liquid network. By determining whether the above-mentioned interaction relationship exists between all gas-liquid components, the gas-liquid network model G is constructed. G .

[0097] In step S3, the four network models are integrated to obtain a mechanical-electrical-information-gas-liquid interdependent network model.

[0098] In this step, when constructing the dependent network model, dependent edges between the networks are added between the same physical elements in the four networks to construct a mechanical-electrical-information-gas-liquid dependent network model.

[0099] Specifically, a dependent network is a whole composed of two or more networks with different characteristics, and there are dependencies between the networks. Because some components in the hydrogen-powered train system may simultaneously have mechanical, electrical, information, and gas-liquid interactions, which in turn leads to coupling mechanisms between different networks, the failure of one network may affect the working status of other networks. Therefore, this invention constructs these four networks into a whole for modeling.

[0100] In order to model the hydrogen-powered train system, the present invention constructs a four-layer dependent network model based on the four subnets of mechanical, electrical, information, and gas-liquid, namely the mechanical-electrical-information-gas-liquid dependent network model. Among them, if the node h and the node k from different subnets represent the same actual physical components, there is an edge between the networks between the node h and the node k and the two nodes are dependent nodes. Through the dependent edge, it can be expressed that there is a mutual functional dependency between the nodes, and the nodes hk from the two networks are also coupled. In order to distinguish between the edges within the subnet and the edges between the subnets, the interaction relationship within the mechanical, electrical, information, and gas-liquid networks is called a homogeneous edge (i.e., the edge within the subnet), and the interaction relationship between the subnets is called a dependent edge.

[0101] Step S4 calculates the mechanical, electrical, information, and gas-liquid network node weights in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information, and gas-liquid networks. The network node weights here refer to the weights of the nodes within the network. Based on these weights, the weights of the edges within the network and the weights of the edges between networks can be calculated.

[0102] In this step, when calculating the weight of the mechanical network node, the support of each component in the component is taken into account, and the functional support is used as the weight value between the components in the mechanical network; when calculating the weight of the electrical network node, the power of each component in the component is taken into account; when calculating the weight of the information network node, the data load of each component in the component is taken into account; when calculating the weight of the gas-liquid network node, the flow of each component in the component is taken into account.

[0103] Specifically, in a mechanical network, functional metrics are mainly used to quantify the performance levels of different types of components. Therefore, the present invention constructs the functional importance of three types of components based on their characteristics. In terms of a mechanical network, the more out-degrees a mechanical component has, the more functional support relationships it provides to other components. Therefore, the weight of node i in a mechanical network is can be defined as follows:

[0104]

[0105] In formula (1), a ij represents the element in the i-th row and j-th column of the network adjacency matrix A. For a directed network, if there is an edge from node i to node j, then a ij =1; otherwise a ij =0;N M Indicates the number of nodes in the mechanical network.

[0106] In the electrical network, the purpose of the node is to provide electrical energy to the train system or convert electrical energy into kinetic energy. Electrical nodes include power generation nodes and point nodes. Therefore, the weight of the electrical node p is It can be calculated from its power, which is defined as follows:

[0107]

[0108] In formula (2), u p represents the voltage at electrical node p, I op represents the current flowing out of the electrical node p. According to Kirchhoff's law, the voltage and current of an electrical node are calculated as follows:

[0109] Y*U=I (37)

[0111] In formula (3), Y is the admittance matrix (y pq represents the element in the pth row and qth column of the node admittance matrix in the electrical network, and y pp =-∑ p≠q y pq ), U is the voltage vector of the node, and l is the current vector of the node. Then the current I flowing through edge pq is pq can be defined as follows:

[0112] I pq =(u p -u q )*Y pq (38)

[0114] In order to model the operation mechanism of information networks, this paper assumes that all data packets are transmitted along the shortest path and adopts a data traffic model to characterize the importance of node functions. Considering the complexity of data transmission modeling and the fact that node betweenness can approximate data load, the weight of information node u is Can be characterized by data traffic load.

[0115] In the gas-liquid network, the hydrogen storage system and the fuel cell system are connected via hydrogen hard pipes, while the fuel cell system and the radiator are connected via rubber hoses. Therefore, in order to quantify the functional properties of the components in the pipeline network, the weights of the components can be defined based on the gas and liquid flows in actual operation.

[0116] In order to measure the importance of the relationship between mechanical, electrical, information, and gas-liquid networks, the weights between mechanical components ij, between electrical components pq, between information components uv, and between gas-liquid components st are and The definition is as follows:

[0117]

[0118] In formulas (5)-(8), and Indicates the adjacency relationship between mechanical components ij, electrical components pq, information components uv, and gas-liquid components st. If they are connected, and otherwise and

[0119] Step S5: constructing a weighting matrix based on the four network node weights, wherein the weighting matrix includes intra-network weights and inter-network weights.

[0120] In this step, the inter-network weights refer to the weights of the interdependence edges between the mechanical and electrical, mechanical and information, mechanical and gas-liquid, electrical and information, electrical and gas-liquid, and information and gas-liquid networks. The weights of the interdependence edges are determined by multiplying the node weights between the networks.

[0121] Specifically, a network weighted matrix of the hydrogen-powered train system is constructed based on the weights of the interaction relationships in the mechanical, electrical, information, and gas-liquid networks and the dependencies between the two networks. In order to construct the strength of the interaction relationships in the weighted mechanical-electrical-information-gas-liquid dependency network model, a weighted matrix M is proposed. W , defined as follows:

[0122]

[0123] In formula (9), M M , M E , M C and M G Represents the weighted matrix of the four networks: mechanical, electrical, information, and gas-liquid. M , M E , M C and M G The elements in the network are the weights of the relationships between components in the mechanical, electrical, information, and gas-liquid networks. and In addition, the weight matrix M ME ,M Mc ,M MG ,M EC ,M EG and M CG It represents the importance of the dependent edges between networks, and can also be calculated by multiplying the weight importance of the nodes.

[0124] Step S6: screening the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix.

[0125] In this step, the weights between the interaction relationships represent the strength of the interaction relationships between components. W , we can construct the weighted distance matrix M D , where the elements in the weighted distance matrix is defined as follows.

[0126]

[0127] Obviously, this shows that the greater the weight between two components, the closer the interaction relationship.

[0128] According to the value of the weighted distance matrix, by using the Floyd algorithm to calculate the weighted shortest path length between nodes, the weighted network efficiency value can be obtained, that is, the weighted network reliability index NE is defined as follows:

[0129]

[0130] In formula (11), dhk represents the weighted shortest path length between nodes hk, and N represents the number of nodes in all networks. Nodes hk here include any two nodes in the mechanical, electrical, information, and gas-hydraulic networks.

[0131] Step S7, constructing a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction between the components as a decision variable and takes improving the weighted network reliability index as the optimization goal.

[0132] In this step, the connection relationship between components is used as the decision variable, and the node degree (i.e., the number of connected nodes) is used as the constraint. Then, the following hydrogen train system topology optimization model can be constructed:

[0133]

[0134] The constraints are:

[0135]

[0136] a hh =0 (48)

[0137] a hk ∈{0,1} (49)

[0138] In formulas (12)-(15), d h represents the degree value of node h, a hk It represents the adjacency relationship between any nodes hk in the same network.

[0139] Step S8: Solve the weighted network reliability optimization model based on the improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

[0140] In this step, an improved simulated annealing algorithm is used for solving the problem, wherein the algorithm steps of the simulated annealing are as follows:

[0141] Step S81: for a given mechanical-electrical-information-gas-liquid interdependent network G, M ,p E ,p C and p G The probability of selecting the mechanical, electrical, information, and gas and liquid subnets. M ,p E ,p C and p G The value of is the ratio of the number of edges in the four subnetworks to the total number of edges in the dependent network;

[0142] In step S82, after selecting the corresponding subnet, two edges in the network are randomly selected for exchange, and a newly generated dependent network G′ is obtained. It is determined whether the newly generated dependent network G′ satisfies constraints (13) and (14); if so, proceed to step S83; otherwise, return to step S81.

[0143] Step S83 , calculating a weighted matrix, a weighted distance matrix and a weighted network reliability index NE′ based on the new dependency network G′.

[0144] Step S84: If the weighted network reliability index NE′ in the new network is better than the weighted network reliability index in the current network, the newly generated network G′ is accepted and the current solution is updated to G′. If the weighted network reliability index NE′ in the new network is not better than the weighted network reliability index in the current network, the probability p of accepting the solution is calculated as follows:

[0145]

[0146] In formula (16), T represents temperature, λ is the parameter value of the accepted solution, and the calculation formula is as follows:

[0147]

[0148] In formula (17), p c represents the critical probability of accepting a worse solution, μ is the critical relative proportion, T max is the highest temperature.

[0149] Step S85, determine whether the number of iterations t meets the maximum number of iterations; if so, output the optimal network structure; otherwise, set T = T*0.95, t = t+1, and return to step S81.

[0150] The topology optimization method for a hydrogen-powered train system described in this embodiment is used to optimize a train system. Table 1 shows a partial list of components extracted from the mechanical, electrical, information, and gas-liquid components in step S1.

[0151] Table 1 Some components in mechanical, electrical, information, and gas-liquid parts

[0152]

[0153] In order to obtain the functional importance of components in the electrical network and the gas-liquid network, it is assumed that the element Y in the admittance matrix of the electrical network is ij The value of is 10, the voltage of the power generation node is 1, and the current of the user node is 1; in the gas-liquid network, assuming that the flow rate of the liquid flowing through the component is 1m 3 / h, the gas flows through the parts, the flow rate is 2m 3 / h. Since both gas and liquid flow through the stack, the flow rate is assumed to be 3m 3 / h. In the simulated annealing algorithm, the critical probability p of accepting a worse solution c is 0.8, the critical relative ratio μ is 0.005, and the maximum temperature T max The maximum number of iterations is 20,000. By taking the above steps, the weighted matrix and weighted network reliability in the mechanical-electrical-information-gas-liquid network are calculated, and the improved simulated annealing algorithm is used to solve the problem. The iterative results are as follows: Figure 7 As shown in the figure, the weight importance of the edges in the network before and after optimization is compared, as shown in the figure. Figure 8 shown.

[0154] Depend on Figure 7 and Figure 8 It can be seen that the topology optimization method of the hydrogen-powered train system provided by the embodiment of the present invention is Figure 7 It can be seen that by adjusting the topological structure of the weighted network, the network weighted reliability has been significantly improved, from the initial 0.0140 to 0.0187, an increase of 33.57%. Figure 8 It can be seen that in the optimized mechanical, electrical, and gas-hydraulic networks, edges with higher importance have higher weights, indicating that components with higher importance tend to connect to each other.

[0155] Based on the same idea, an embodiment of the present invention also provides a topology optimization system for a hydrogen-powered train system, which includes a component classification module, a subnetwork construction module, a dependent network construction module, a subnetwork node weight calculation module, a weighted matrix construction module, a reliability index calculation module, an optimization model construction module, and a solution and optimization module; wherein,

[0156] The component classification module is used to divide the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and to divide all components in the system into four types of parts;

[0157] The sub-network construction module is used to construct a mechanical network topology structure model, an electrical network topology structure model, an information network topology structure model, and a gas-liquid network topology structure model based on the four parts of mechanical, electrical, information, and gas-liquid according to the components in each component and the interaction relationship between the components;

[0158] The interdependent network construction module is used to fuse the four network models to obtain a mechanical-electrical-information-gas-liquid interdependent network model;

[0159] The sub-network node weight calculation module is used to calculate the mechanical network node weight, electrical network node weight, information network node weight and gas-liquid network node weight in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information and gas-liquid networks respectively;

[0160] The weighted matrix construction module is used to construct a weighted matrix according to four network node weights, wherein the weighted matrix includes intra-network weights and inter-network weights;

[0161] The reliability index calculation module is used to screen the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix;

[0162] The optimization model construction module is used to construct a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction relationship between the various components as a decision variable and takes improving the weighted network reliability index as an optimization goal;

[0163] The solution and optimization module is used to solve the weighted network reliability optimization model based on an improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

[0164] In this embodiment, each module is implemented by a processor, and a memory is appropriately added when storage is required. The processor may be, but is not limited to, a microprocessor MPU, a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), other programmable logic devices, discrete gates, transistor logic devices, discrete hardware components, etc. The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk storage. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0165] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode.

[0166] It should also be noted that the topology structure optimization system of the hydrogen-powered train system described in this embodiment corresponds to the topology structure optimization method of the hydrogen-powered train system. The description and limitation of the method are also applicable to the system and will not be repeated here.

[0167] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. It is not intended to limit the scope of the invention to be protected, but merely represents a preferred embodiment of the present invention. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the inventive concept. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work shall fall within the scope of protection of the present invention.

Claims

1. A method for optimizing the topology of a hydrogen-powered train system, characterized in that: The method comprises the following steps: Step S1, dividing the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and dividing all components in the system into the corresponding four types of components; Step S2: constructing a mechanical network topology model, an electrical network topology model, an information network topology model, and a gas-liquid network topology model based on the four parts of mechanical, electrical, information, and gas-liquid, respectively, according to the components in each component and the interaction relationship between the components; Step S3, integrating the four network models to obtain a mechanical-electrical-information-gas-liquid interdependent network model; Step S4, respectively calculating the mechanical network node weights, electrical network node weights, information network node weights, and gas-liquid network node weights in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information, and gas-liquid networks; Step S5, constructing a weighted matrix based on the four network node weights, wherein the weighted matrix includes intra-network weights and inter-network weights; Step S6, screening the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix; Step S7: constructing a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction between the components as a decision variable and takes improving the weighted network reliability index as an optimization goal; Step S8: Solve the weighted network reliability optimization model based on the improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

2. The method according to claim 1, characterized in that In step S2: In a mechanical network MN, nodes represent mechanical components and edges represent mechanical action relationships. The mechanical network is represented by G M =(V M , E M ) is represented by V M represents the set of nodes in the mechanical network, E M Represents the set of edges in the mechanical network; in the mechanical network G M In the mechanical network, if a fastener connects the mechanical component α corresponding to node i and the mechanical component β corresponding to node j, and the failure of the mechanical component α will cause the loss of the function of the mechanical component β, then there is an edge from node i to node j in the mechanical network; In the electrical network EN, nodes represent electrical components and edges represent electrical interaction relationships. The electrical network is represented by G E =(V E , E E ) is represented by V E Represents the set of nodes in the electrical network, E E Represents the set of edges in the electrical network; in the electrical network G E In the network, if the electrical component γ corresponding to the node p and the electrical component δ corresponding to the node q have a current action relationship, then there is an undirected edge between the node p and the node q in the electrical network; In the information network CN, nodes represent information components and edges represent information interaction relationships. The information network is represented by G C =(V C , E C ) is represented by V C represents the set of nodes in the information network, E C Represents the set of edges in the information network; in the information network G C In the information network, if the information component ε of the corresponding node u and the information component ∈ of the corresponding node v have an information interaction relationship, then there is an undirected edge between the node u and the node v in the information network; In the gas-liquid network GN, nodes represent components related to gas and liquid and edges represent the input or output of gas or liquid. G =(V G , E G ) is represented by V G represents the set of nodes in the gas-liquid network, e G Represents the set of edges in the gas-liquid network; in the gas-liquid network G G In the gas-liquid network, if there is a flow of gas or liquid between the gas-liquid component ζ corresponding to the node s and the gas-liquid component η corresponding to the node t, then there is an undirected edge between the node s and the node t in the gas-liquid network.

3. The method according to claim 1, characterized in that In step S3, when constructing the dependent network model, dependent edges between the networks are added between the same physical elements in the four networks, thereby constructing a mechanical-electrical-information-gas-liquid dependent network model.

4. The method according to claim 1, wherein In step S4, when calculating the weight of the mechanical network node, the support of each component in the component is taken into account, and the functional support is used as the weight value between the components in the mechanical network; when calculating the weight of the electrical network node, the power of each component in the component is taken into account, and the power value is used as the weight value between the components in the electrical network; when calculating the weight of the information network node, the data load of each component in the component is taken into account, and the data load is used as the weight value between the components in the information network; when calculating the weight of the gas-liquid network node, the flow of each component in the component is taken into account, and the flow value is used as the weight value of each component in the gas-liquid network.

5. The method according to claim 4, characterized in that The weight of node i in the mechanical network The definition is as follows: In formula (1), a ij Represents the element in row i and column j of the network adjacency matrix A; For a directed network, if there is an edge from node i to node j, then a ij =1; otherwise a ij =0;N M represents the number of nodes in the mechanical network; The weight of the electrical node p It is calculated from its power, which is defined as follows: In formula (2), u p represents the voltage at electrical node p, I op represents the current flowing out of the electrical node p; And according to Kirchhoff's law, the voltage and current of the electrical node are calculated as follows: Y*U=I (3) In formula (3), Y is the admittance matrix, y pq represents the element in the pth row and qth column of the node admittance matrix in the electrical network, and y pp =-∑ p≠q y pq , U is the voltage vector of the node, I is the current vector of the node; then the current flowing through edge pq is I pq The definition is as follows: I pq =(u p -u q )*Y pq (4); The weight of information node u Characterized by data traffic load; In the gas-liquid network, weights are defined based on the gas and liquid flows of component i in actual operation. The weights between mechanical components ij, electrical components pq, information components uv, and gas-liquid components st and The definition is as follows: In formulas (5)-(8), and Indicates the adjacency relationship between mechanical components ij, electrical components pq, information components uv, and gas-liquid components st. If they are connected, and otherwise and 6. The method according to claim 5, characterized in that In step S5, the weight matrix M W , defined as follows: In formula (9), M M , M E , M C and M G Represents the weighted matrix of the four networks of mechanical, electrical, information, and gas-liquid; in the weighted matrix M M , M E , M C and M G The elements in the network are the weights of the relationships between components in the mechanical, electrical, information, and gas-liquid networks. and Weight matrix M ME ,M MC ,M MG ,M EC ,M EG and M CG It represents the importance of the dependent edges between networks and is calculated by multiplying the weight importance of the nodes.

7. The method according to any one of claims 1 to 6, characterized in that In step S6, according to the weight matrix M W , construct the weighted distance matrix M D , where the elements in the weighted distance matrix is defined as follows: In formula (10), Represents the matrix M W Any element in ; According to the value of the weighted distance matrix, by using the Floyd algorithm to calculate the weighted shortest path length between nodes, the weighted network efficiency value can be obtained, that is, the weighted network reliability index NE is as follows: In formula (11), d hk It represents the weighted shortest path length between nodes hk, and N represents the number of nodes in all networks; nodes hk include any two nodes in the four networks of mechanical, electrical, information, and gas-liquid.

8. The method according to claim 7, characterized in that In step S7, the connection relationship between components is used as the decision variable, and the degree value of the node is used as the constraint to construct the topology optimization model of the hydrogen train system, as shown in formula (12): The constraints are: a hh ＝0 (14) a hk ∈{0,1} (15) In formulas (12)-(15), d h represents the degree value of node h, a hk It represents the adjacency relationship between any nodes hk in the same network.

9. The method according to claim 1, characterized in that Step S8 includes: Step S81: for a given mechanical-electrical-information-gas-liquid interdependent network G, M ,p E ,p C and p G The probability of selecting the mechanical, electrical, information, and gas-liquid subnets is M ,p E ,p C and p G The value of is the ratio of the number of edges in the four subnetworks to the total number of edges in the dependent network; Step S82: After selecting the corresponding subnet, randomly select two edges in the network to exchange, and obtain a newly generated dependent network G′. Determine whether the newly generated dependent network G′ satisfies constraints (13) and (14). If so, proceed to step S83; otherwise, return to step S81. Step S83, calculating the weighted matrix, the weighted distance matrix and the weighted network reliability value NE' according to the new dependency network G'; Step S84: If the weighted network reliability value NE′ in the new network is better than the weighted network reliability value in the current network, the newly generated network G′ is accepted and the current solution is updated to G′; if the weighted network reliability value NE′ in the new network is not better than the weighted network reliability value in the current network, the probability p of accepting the solution is calculated as follows: In formula (16), T represents temperature, λ is the parameter value of the accepted solution, and the calculation formula is as follows: In formula (17), p c represents the critical probability of accepting a worse solution, μ is the critical relative proportion, T max is the maximum temperature; Step S85, determine whether the number of iterations t meets the maximum number of iterations; if so, output the optimal network structure; otherwise, set T = T*0.95, t = t+1, and return to step S81.

10. A topology optimization system for a hydrogen-powered train system, characterized in that: The system includes a component classification module, a sub-network construction module, a dependent network construction module, a sub-network node weight calculation module, a weighted matrix construction module, a reliability index calculation module, an optimization model construction module, and a solution and optimization module; wherein, The component classification module is used to divide the hydrogen-powered train system into four parts: mechanical, electrical, information, and gas-liquid, and to divide all components in the system into four types of parts; The sub-network construction module is used to construct a mechanical network topology structure model, an electrical network topology structure model, an information network topology structure model, and a gas-liquid network topology structure model based on the four parts of mechanical, electrical, information, and gas-liquid according to the components in each component and the interaction relationship between the components; The interdependent network construction module is used to fuse the four network models to obtain a mechanical-electrical-information-gas-liquid interdependent network model; The sub-network node weight calculation module is used to calculate the mechanical network node weight, electrical network node weight, information network node weight and gas-liquid network node weight in the dependent network model based on the physical characteristics of the components in the mechanical, electrical, information and gas-liquid networks respectively; The weighted matrix construction module is used to construct a weighted matrix according to four network node weights, wherein the weighted matrix includes intra-network weights and inter-network weights; The reliability index calculation module is used to screen the mechanical-electrical-information-gas-liquid weighted network reliability index according to the hydrogen-powered train system network weighted matrix; The optimization model construction module is used to construct a mechanical-electrical-information-gas-liquid weighted network reliability optimization model based on the reliability index; the optimization model uses the interaction relationship between the various components as a decision variable and takes improving the weighted network reliability index as an optimization goal; The solving and optimization module is used to solve the weighted network reliability optimization model based on an improved simulated annealing algorithm, and optimize the topology structure according to the solution structure.

Images