CN-122020929-A - Hydrogen energy network node arrangement method utilizing traffic data and gas pipe network data

CN122020929ACN 122020929 ACN122020929 ACN 122020929ACN-122020929-A

Abstract

The invention discloses a method for arranging hydrogen energy network nodes by using traffic data and gas pipe network data, which comprises the steps of firstly obtaining the traffic data and the gas pipe network data of a target area, carrying out pretreatment and unified mapping, mapping the data of different sources to unified network node space, constructing a node multidimensional characteristic data set, determining a candidate space, executing principal component analysis to extract coupling structure characteristic components, executing cluster analysis to form node types, constructing an operation scene set containing operation condition changes, constructing a robust optimization model under the scene set and adopting a maximum and minimum criterion to solve, and outputting a node arrangement scheme which continuously meets physical safety constraint and meets the robustness requirement under the scene set. The method is suitable for hydrogen energy network planning and infrastructure configuration in different areas and under different operating conditions.

Inventors

PEI ZHI
JIN BINGCAN
LU HAIMIN
CHEN YONG
LI SHIYUN
YI WENCHAO

Assignees

浙江工业大学

Dates

Publication Date: 20260512
Application Date: 20260206

Claims (10)

1. The hydrogen energy network node arrangement method utilizing traffic data and gas pipe network data is characterized by comprising the following steps of: S1, acquiring traffic data and gas pipe network data of a target area; S2, preprocessing and uniformly mapping the traffic data and the gas pipe network data, mapping the data from different sources to a uniform network node space J, and preparing input parameters and intermediate result caches required by mixed bearing upper limit and safety clearance judgment; s3, constructing a node multidimensional feature data set based on the uniformly mapped data, determining a candidate space omega, and executing principal component analysis to extract coupling structure feature components; s4, performing cluster analysis on the nodes based on the coupling structure characteristic components to form node types; s5, constructing an operation scene set containing operation condition changes, constructing a robust optimization model under the scene set, and solving by adopting a maximum and minimum criterion; Invoking a physical safety feasibility checking mechanism SafeOracle in the iterative solving process, only performing hard constraint checking of the mixed bearing upper limit and the safety backspace on a selected node subset in the current scheme, and if a violation is detected, removing and replacing the violation node in the current scheme or continuing iteration after applying a constraint for prohibiting selection on the violation node; and S6, outputting a node arrangement scheme which continuously meets physical security constraint and meets robustness requirements under the scene set.
2. The method of claim 1, wherein the traffic data comprises commercial vehicle movement tracks, traffic flow or accessibility information, and the gas pipe network data comprises pipe network connection structures and physical constraint information under the condition of hydrogen mixed energy supply; Before the unified mapping, the coordinate reference system unification, the dimension unit unification, the deduplication cleaning and the missing field marking are performed on the traffic data and the gas pipe network data, and the initial candidate points are combined through the combination threshold value to construct a unified node set.
3. The method of claim 1, wherein the unified mapping of traffic data comprises defining a traffic mapping function The original traffic record set is D T , and any traffic record Mapping any traffic record o to the nearest node in the unified node set J, and meeting the constraint of a preset traffic mapping radius R T , wherein the mapping and effectiveness are determined as follows: ; And is also provided with ; Unified node set is J, each node Is marked as u j , any traffic record Having a spatial position g (o), dist (·, ·) is a spatial distance measure, and R T is a preset traffic mapping radius; And when the radius constraint is not met, marking the traffic records as unmapped records and not participating in the calculation of the traffic characteristics of the subsequent nodes, and aggregating the traffic records mapped to the same node according to a preset aggregation operator Agg (-) in a fixed time window to form the node traffic characteristic vector.
4. The method of claim 1, wherein the unified mapping of the gas pipe network data comprises selecting, for each candidate node j, a pipe segment closest to the node geometry and having a distance not exceeding a preset pipe network mapping radius R p as an associated pipe segment m (j), wherein the mapping and validity determination is as follows: And (2) and ; And inherit associated pipe segment attributes to nodes after mapping is completed, wherein For the pipe section index associated with node j, p m is the pipe section operating pressure, D m is the pipe section pipe diameter, p m(j) is the pipe section operating pressure associated with node j, For the pipe diameter of the pipe section associated with the node j, defining the pressure and the pipe diameter associated with the node as follows: ; The pipe network mapping is only used for determining the index of the associated pipe section and inheriting the attribute of the index, u j is the original space coordinate of the node j, and projection or adsorption modification is not carried out on the node space position u j ; The pipe sections of the gas pipe network are assembled into M, each pipe section Properties with geometry Γ m , operating pressure p m , pipe diameter D m ; The geometric distance metric from the spatial position u j of the node j to the pipe segment geometry Γ m is represented, and R P is a preset pipe network mapping radius.
5. The method of claim 1, wherein step S2 further comprises performing coarse-grained prescreening to form an initial candidate subset without affecting a final decision mechanism The coarse-particle size pre-screening comprises at least: Upper limit pre-screening of mixed bearing, when the mixed proportion parameter b j of the node is obviously beyond the allowable upper limit When the node is marked as unavailable and does not incorporate F 0 ; Safety backspace prescreening, marking node j as unavailable and not incorporating F 0 when the distance delta j of the node j to its associated pipe segment is significantly less than the safety backspace threshold r j ; And marking the node which is not determinable due to the lack of the key attribute as not available and not including F 0 ; for the rest nodes, reserving an initial candidate subset F 0 ; Step S3 the candidate space 。
6. The method of claim 1, wherein intermediate calculation results required by node mapping relationships, key pipe network attributes and physical security constraint determination are cached, and only newly added or changed node triggers SafeOracle are checked when a solution is locally changed in the solving process, so that repeated calculation cost is reduced.
7. The method of claim 1, wherein the node multidimensional feature dataset includes at least traffic features for characterizing traffic load and reachability, and pipe network features for characterizing pipe network load-bearing boundaries and safety boundaries, and wherein the principal component analysis is configured to extract coupling feature components reflecting traffic load and pipe network load-bearing boundary and safety boundary related structures.
8. The method of claim 1, wherein the set of operating scenarios is obtained by combining a demand scenario with an operating condition scenario, wherein the pipe network operating pressure level, the allowable upper limit of mixing, and the safety factor value rule k (·) involved in the scenario are model inputs given by exogenous sources rather than decision variables, and the value range is determined according to available operating data, design data, and current standard specifications; Wherein exogenous given model input refers to parameters set directly by the external environment or decision maker before the model is run; the demand of the demand scene refers to the demand strength scene of the demand unit on the hydrogenation service, which is the situation that the demand weight distribution changes along with the traffic state.
9. The method of claim 1, wherein the robust optimization uses a max-min criterion to satisfy a robust value And taking the maximization Z (y) as a target to obtain a node arrangement scheme with the maximum system performance lower bound under the scene set, wherein Z s (y) is a system performance index of the scheme y under the scene s.
10. The method of claim 1, wherein the physical security feasibility checking mechanism SafeOracle comprises at least a mixed loading condition check and a safe ranging condition check; The mixed bearing condition check is to check the selected node under each scene , Wherein b j (s) is the hydrogen-gas mixing ratio of the node j in the scene s, The maximum mixing proportion of hydrogen and gas allowed by the node j in the scene s; the safety clearance checking is based on the mixing ratio, the pipe pressure and the pipe diameter ; ; Wherein c is a unit scale calibration coefficient; wherein k (·) is a safety factor value rule related to the mixing ratio, used for adjusting the safety back-off calculation, m (j) is a pipe section index associated with the node j, D m(j) is the pipe diameter of the pipe section, which is the operating pressure of the pipe section in the scene s; When the node is selected, let Then it should satisfy Wherein Selecting decision variables for nodes when When it indicates that node j is forbidden to be selected, when Time represents the selection node j; Representing the actual shortest distance between a node j and its associated pipe segment; A minimum security fallback threshold for node j in scene s, and optionally checking for map accessibility, such that To satisfy a preset pipe network mapping radius constraint.

Description

Hydrogen energy network node arrangement method utilizing traffic data and gas pipe network data Technical Field The invention relates to the technical field of hydrogen energy infrastructure construction, in particular to a hydrogen energy network node arrangement method utilizing traffic data and gas pipe network data. Background Hydrogen energy is widely focused on as a low-carbon alternative energy source in the background of emission reduction and decarburization in the traffic field. In particular, hydrogen fuel driven applications are gradually advancing in the field of large commercial vehicles with high demands on endurance mileage and energy density. Along with this is the planning and construction requirements of the hydrogen energy infrastructure, where how to determine the key nodes of the hydrogen energy network in the target area and form a practical network configuration scheme becomes an engineering technical problem to be solved. In the initial stage of the construction of the hydrogen energy infrastructure, as the special hydrogen transmission and distribution network is not perfect, a hydrogen-gas mixing (Blending) mode of mixing and conveying hydrogen and gas by utilizing the existing natural gas pipe network is widely discussed and applied as a transitional energy supply means. However, under the condition of mixed energy supply, physical parameters such as topological connection structure, operation pressure, pipe diameter, hydrogen mixing proportion and the like of the gas pipe network directly influence allowable operation boundaries, and hard constraint is formed on node selection and network configuration. If the above physical constraints are not fully considered in the node configuration process, the candidate nodes may be not implemented in engineering or trigger a security risk in operation, so that the network scheme is difficult to land. In the existing node arrangement or network construction method, one type of method is mainly used for selecting and configuring addresses based on indexes such as distance, accessibility, traffic flow or demand intensity, and the like, and the other type of method is used for introducing pipe network related information, but the pipe network is often regarded as general connection resources or cost factors, and physical safety boundaries induced by factors such as pressure, pipe diameter, mixing proportion, and the like under the condition of mixed energy supply are not used as core constraints of feasibility judgment and optimization decision. Thus configuration results that are "plausible at the planning level but not feasible at the physical and security level" may occur. In addition, although some methods attempt to integrate traffic flow characteristics and pipe network characteristics to develop multi-index decisions or optimizations, a simple weighting or index combination mode is often adopted, and correlation and structural characteristics between different indexes cannot be systematically described, so that internal modes of network behaviors in a hybrid energy supply environment are difficult to reflect. Under complex operating scenarios (e.g., supply and load fluctuations, operating pressure variations, mixing ratio adjustments, etc.), the configuration results of the above-described methods are prone to exhibiting instability, and it is difficult to ensure that feasibility and safety requirements are met under different operating conditions. Meanwhile, the hydrogen fuel large-scale commercial vehicle is still in the initial popularization stage, the actual hydrogenation record and the actual measurement energy consumption data are limited, and the difficulty of making a network configuration decision under the incomplete observation condition is further increased. Therefore, a method for disposing hydrogen energy network nodes is needed to comprehensively utilize traffic data and gas pipe network data under mixed energy supply condition under the condition that the observed data are limited and the operation condition is uncertain, and to incorporate key physical safety constraint into feasibility screening and configuration decision process, so as to obtain practical and stable hydrogen energy network node arrangement method. Disclosure of Invention Aiming at the technical problems existing in the prior art, the invention aims to provide a hydrogen energy network node arrangement method utilizing traffic data and gas pipe network data. The invention arranges the hydrogen energy network nodes by processing traffic data and gas pipe network data. The method is particularly suitable for processing the constitution and arrangement of the hydrogen energy network in the running environment of the large commercial vehicle adopting the hydrogen fuel by combining the physical characteristics of the traffic flow structure and the gas pipe network under the condition of hydrogen mixed energy supply. The invention provides a technical sch