Multimodal transport method and system suitable for salt mist environment

Abstract

The invention discloses a multimodal transport method and system suitable for a salt mist environment, and the method comprises the steps: building a multimodal transport salt mist adaptive transport capacity grid model with city / hub nodes as vertexes and connected road sections as edges; configuring a salt mist adaptation device at the node, and marking the node protection capability, the availability time window and the supply capability; meanwhile, a model hypothesis is set, and cargo transportation integrity, facility availability, transportation rules and a parameter acquisition mode are determined; defining a multi-dimensional decision variable, a parameter and a state quantity; incorporating the salt mist environment influence into cost composition, and constructing a salt mist correlation total cost model; constructing a mixed integer programming model containing path connection and flow conservation, protection capability constraint, energy / endurance constraint and corrosion / exposure constraint by taking the minimum total cost as a target; and designing an improved hybrid immune-particle swarm dual optimization algorithm, and carrying out optimization solution, optimal path output, mode selection and protection parameter configuration through hybrid coding, security repair operators and a memory mechanism.

Description

Technical Field

[0001] This invention belongs to the field of multimodal transport technology, and more specifically, relates to a multimodal transport method and system suitable for salt spray environments. Background Technology

[0002] Multimodal transport, as an integrated logistics organization model that combines two or more modes of transportation such as road, rail, waterway, and air, has become a core support of the modern logistics system due to its efficiency, continuity, and safety throughout the entire cargo transportation process. In recent years, with the expansion of coastal port clusters, the development of island economies, the expansion of cross-border logistics channels along the "Maritime Silk Road," and the densification of sea-rail intermodal transport networks, the demand for multimodal transport of goods in high-salt-fog environments (long-term exposure to coastal areas, ports, islands, or during ship transport) has experienced explosive growth, driving the upgrading of multimodal transport technology towards adaptation to special environments.

[0003] However, current mainstream multimodal transport technologies and supporting equipment are designed based on mild climatic conditions (low humidity, no salt spray corrosion), and their structures, materials, and system configurations do not take into account the special characteristics of salt spray environments. When directly applied to high salt spray scenarios, the air containing high concentrations of chloride ions and the high humidity environment will undergo complex physicochemical reactions with equipment components, causing a series of technical problems. This not only leads to shortened equipment lifespan and a surge in failure rates, but also seriously disrupts the continuity and timeliness of multimodal transport, drives up operation and maintenance costs, and becomes a core bottleneck restricting the development of multimodal transport in salt spray environments.

[0004] First, the equipment and key components lack corrosion resistance, leading to structural reliability failure. In a salt spray environment, the chloride-containing air causes rapid corrosion and aging of metal parts, electronic circuits, and composite material surfaces. In particular, key components such as mechanical seals, bearings, fasteners, and transmission chains are prone to pitting corrosion, stress corrosion cracking, and wear failure, resulting in shortened equipment lifespan and increased failure rate, seriously affecting the long-term operational reliability of the multimodal transport system.

[0005] Secondly, the electrical and power systems are corroded by salt spray, causing a sharp drop in operational stability and safety. Salt spray environments can easily cause poor contact, insulation degradation, and even short circuits in components such as electrical control cabinets, sensors, connectors, and power battery packs. At the same time, the high humidity and high salinity climate exacerbates the electrochemical corrosion process, significantly reducing the working efficiency and safety of the electrical and power systems and increasing the risk of downtime and maintenance during transportation.

[0006] Third, the lack of salt spray adaptability in loading and unloading interfaces and transportation organization leads to disruptions in logistics continuity. At ports, wharves, and maritime transport junctions, salt spray deposition can easily cause failure of seals and sliding structures in container interfaces, rapid docking mechanisms, and loading / unloading equipment, reducing interface precision and loading / unloading efficiency. Simultaneously, the high salt and humidity environment exacerbates infrastructure corrosion, limiting maintenance and support conditions. Traditional multimodal transport organization and scheduling models lack effective emergency response mechanisms in the event of sudden equipment corrosion damage or interface failure, making it difficult to ensure transportation continuity and timeliness reliability.

[0007] The three major issues mentioned above overlap, leading to adverse consequences in multimodal transport under salt spray conditions, such as accelerated equipment corrosion, increased failure rate, reduced transport timeliness, and increased operation and maintenance costs. Therefore, it is necessary to propose targeted technical improvement solutions for corrosion-resistant design, electrical and power protection, and loading and unloading adaptability to break through the bottleneck of existing multimodal transport technologies under salt spray conditions. Summary of the Invention

[0008] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a multimodal transport method and system suitable for salt spray environments. By configuring dedicated salt spray adaptation devices such as anti-corrosion coated cargo holds, salt-resistant sealing connectors, and cathodic protection stations at transport capacity grid nodes, and combining this with quantitative control of parameters such as metal corrosion rate and electrical insulation degradation factor, the system significantly enhances the salt spray corrosion resistance of mechanical components and electrical systems, extends equipment life, reduces failure rate, and significantly improves the system's adaptability and operational reliability in high-salt-spray scenarios such as coastal areas, ports, and islands. Simultaneously, standardized salt-resistant loading and unloading interfaces ensure the accuracy of multimodal transport connections, and an improved hybrid immune-particle swarm optimization algorithm enables dynamic path planning and emergency repair, avoiding loading and unloading failures and transport interruptions caused by salt spray, ensuring the continuity and timeliness of logistics throughout the entire process. Furthermore, by constructing and optimizing a total cost model covering multiple dimensions such as basic transportation, cargo damage, protective energy consumption, and equipment maintenance, the system effectively reduces equipment maintenance expenses, cargo damage losses, and time window penalty costs, significantly improving the economic benefits of multimodal transport in salt spray environments.

[0009] To achieve the above objectives, one aspect of the present invention provides a multimodal transport method suitable for salt spray environments, comprising:

[0010] S1: Construct a multimodal transport salt spray adaptation capacity grid model with city / hub nodes as vertices and connecting road segments as edges; configure salt spray adaptation devices at nodes, and label the node protection capabilities, availability time windows, and replenishment capabilities; at the same time, set model assumptions, and clarify the integrity of cargo transport, facility availability, transport rules, and parameter acquisition methods.

[0011] S2: Define multi-dimensional decision variables, parameters and state variables, and establish a salt spray-related state variable system;

[0012] S3: Incorporate the environmental impact of salt spray into the cost structure, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost;

[0013] S4: With the goal of minimizing total cost, a hybrid integer programming model is constructed, which includes path connectivity and flow conservation, protection capability constraints, energy / endurance constraints, and corrosion / exposure constraints. To address the non-convex nature of the model, an improved hybrid immune-particle swarm optimization algorithm is designed. The solution process is optimized through hybrid encoding, security repair operators, and memory mechanisms, and the optimal path, mode selection, and protection parameter configuration are output.

[0014] Furthermore, the multimodal transport capacity grid model that can simulate high-temperature transport coordination described in step S1 uses several city / hub nodes as vertices and one or more transport modes among highways, railways, waterways, and air connecting sections as edges; each node is equipped with a salt spray / corrosion adaptation device and interface unit; its protection capability, availability time window, and replenishment capability should be marked in the node attributes.

[0015] The city / hub nodes include ports, railway marshalling yards, highway transfer stations, and island supply stations;

[0016] The salt spray / corrosion adapter includes a corrosion-resistant coated cargo hold, a salt-resistant sealing dock, a dehumidification / drying chamber, a corrosion-resistant supply / cathode protection station, a sand and dust filtration unit, and an electrical protection chamber.

[0017] The node protection capabilities include coating grade, dehumidification capacity, cathodic protection capacity, filtration capacity, and electrical sealing grade.

[0018] Furthermore, the model assumptions in step S1 include: goods are transported in containers, and goods within containers are not split for transport (holistic assumption);

[0019] Salt spray adaptation devices and protective facilities marked in the area are in an available state during planning; if a node / facility is unavailable, it is set as unreachable or limited capacity in the model;

[0020] The vehicles operate on the road segment according to the established driving rules, and each transfer node can be loaded and unloaded within the predetermined time window; if a node is temporarily unavailable due to salt spray, it will be reflected by the node reachability parameter.

[0021] All corrosion / energy consumption / failure model parameters, including metal corrosion rate, coating decay rate, insulation degradation factor, protection system power consumption, and component Weibul failure parameters, are estimated through field tests or historical data and substituted into the model.

[0022] Furthermore, the expression for the salt spray-related total cost model described in step S3 is as follows:

[0023] Ctotal =C trans +C transf +C loss +C energy +C maint +C safety +C time (1)

[0024] Among them, C total For total cost, C trans Based on basic transportation costs, C transf For transit costs, C loss For loss / damage costs, C energy For energy consumption / cooling cost, C maint For equipment failure and maintenance costs, C time Penalties and costs associated with time windows;

[0025] The loss / damage cost C loss The expression is:

[0026] C loss =n c ·P unit ·F(T expose S salt (t),prot(t)) (4)

[0027] Where, n c P represents the number of containers used. unit For the value of a single box of goods, T expose S represents the cumulative time that the cargo has been exposed to salt spray. salt (t) represents the salt spray exposure intensity of the path / node at time t; prot(t) represents the timing of the protective measures taken; F(·) represents the loss / failure probability function, with a value of 0–1; Weibull distribution is adopted and a salt spray exposure correction factor a is introduced. salt (S,t), to obtain the corrected Weibull cumulative failure probability F(t), the expression is:

[0028]

[0029] Where η represents the scale parameter; β represents the shape parameter; γ represents the position parameter; a salt (S,t) represents the salt spray acceleration factor.

[0030] Furthermore, the energy consumption / cooling cost C energy The expression is:

[0031] C energy =∑ t (λ elec P elec(t)+λ chem P chem (t)+λ prot P prot (t))Δt (6)

[0032] Among them, C energy λ represents the total energy cost; elec P represents the unit price of electricity. elec (t) represents the power consumption at time t, and P represents the total power consumption of the dehumidifier, fan, and cathodic protection power supply; chem (t) represents the chemical / coating consumption rate at time t; λ chem P is the unit price of chemicals or coating materials. prot (t) represents the power or energy specifically used for protection at time t; λ prot This represents the unit cost of power / energy specifically used for protection; Δt represents the time step.

[0033] Furthermore, the equipment failure and maintenance cost C maint The expression is:

[0034] C maint =∑ comp Cost comp ·Pr[comp fails during mission] (7)

[0035] Pr[comp fails during mission]=F comp (t run η comp / a salt ,β comp ,γ comp (8)

[0036] Among them, C maint Cost represents the total cost of equipment failure and maintenance, calculated based on the expected maintenance cost derived from the failure probability of critical components. comp This represents the direct cost of replacing or repairing a component during the mission period, including material and labor costs; Pr[comp fails during mission] represents the component's failure time t during operation. run The table that malfunctioned internally;

[0037] F comp (t run η comp / a(T),β comp ,γ comp ) represents the Weibull distribution function form used to calculate the probability of component failure; t run (for transport mission operation); ηcomp The characteristic lifespan of a key component; β comp The shape distribution of a key component; γ comp To quickly locate a key component; a salt This is a salt spray acceleration factor, which can be adjusted according to the local salt spray intensity and protection status. The stronger the salt spray or the worse the protection, the larger this factor will be, which will accelerate the failure rate of components.

[0038] If corrosion depth d is used corr (t) can be used as a maintenance trigger indicator:

[0039]

[0040] Where, d corr (t) represents the change in corrosion depth over time; S salt R(τ) is the salt spray concentration at time τ; R(τ) is the temperature-dependent corrosion factor at time τ.

[0041] When d corr (t)≥d crit When a replacement / repair operation is triggered, its expected cost can be assessed based on the trigger probability and added to C. maint ;d crit This represents the critical depth of corrosion.

[0042] Furthermore, the expression for the mixed-integer programming model in step S4 is:

[0043] minC total =C trans +C transf +C loss +C energy +C maint +C time (11)

[0044] The protection capability constraint is:

[0045]

[0046] in, Let be the protection power of node i at time t; Maximum protection power of the node; Indicates whether on-site coating / recoating operations are performed at node i (0 indicates no, 1 indicates yes); Indicates whether the node / vehicle has enabled dehumidification mode (0 indicates no, 1 indicates yes);

[0047] The energy / range constraint is:

[0048] ∑ t P elec(t)Δt≤E available ,∑ t P chem (t)Δt≤Chem avail (14)

[0049] Among them, P elec (t) represents the power consumption at time t; Δt represents the time step; E available P represents the total usable electrical energy. chem (t) represents the consumption rate of the chemical / coating at time t; Chem avail Indicates the total amount of available chemical materials;

[0050] The corrosion / exposure constraint is:

[0051]

[0052] in, The salt spray concentration S represents the concentration of salt spray within the time interval from t0 to t1. salt (t) is the integral over time, representing the cumulative salt spray exposure of the cargo container during that time period; S crit This is the critical threshold for salt spray exposure. When the cumulative salt spray exposure exceeds this value, protective measures or route rerouting are required. coat This is a decision variable for whether to perform on-site coating operations, with a value of 1 indicating execution or 0 indicating non-execution.

[0053] Furthermore, step S4 employs a three-segment hybrid encoding method to construct the antibody / particle structure. The three-segment hybrid encoding means that a string contains three parts: a path-mode binary segment, a protection / module activation segment, and a continuous segment. The specific encoding rules are as follows:

[0054] Path-mode binary segment: Use 0 / 1 binary encoding for each candidate route segment (i,j) and each mode of transport k, or simplify it to compressed encoding of "route node sequence + mode sequence";

[0055] Protection / Module Activation Section: Whether the discrete decision variables specific to the salt spray scenario are enabled is represented by binary encoding; the discrete decision variables specific to the salt spray scenario include whether the node enables coating respray, cathodic protection, dehumidification, filtration, and night operation;

[0056] Continuous Quantity Segment: Real-valued encoding is used for continuous decision variables in the model; the continuous decision variables include protection power. driving speed Chemical protection delivery rate

[0057] Step S4 aims to minimize the total cost, using the reciprocal of the objective function as the basis for the fitness function; at the same time, an indicator function is introduced, which penalizes the fitness if the salt spray exposure of a path or node exceeds the limit.

[0058] Generate an initial population of F individuals, including heuristic and stochastic solutions that prioritize low-salt-spray coastlines, enable intermediate node spraying, and shorten exposure time; the continuous segment variables are initialized based on uniform distribution or engineering experience.

[0059] For discrete segments, immune genetic operations are used to maintain population diversity and perform global search; for continuous segments, particle swarm optimization velocity-position update rules are used to accelerate local accurate search; at the same time, a safety repair operator is introduced. If the solution violates corrosion / safety constraints, it is repaired by adding protection activation bits, increasing dehumidification / cathode protection power, or replacing it with low-exposure road sections, so that the solution is restored to the constraint feasible region.

[0060] The similarity of individuals is calculated based on the Hamming distance of binary segments, maintaining the diversity of the population within the threshold range; the best solutions are stored in the memory bank for subsequent crossover operations to generate offspring, avoiding the algorithm from getting trapped in local optima and improving the global optimization ability.

[0061] Furthermore, the fitness function expression is:

[0062]

[0063] Among them, C total is the total cost in the mixed-integer programming model; M is the penalty coefficient when a path or node exceeds the exposure threshold (such as exceeding the cumulative salt spray exposure limit or electrical insulation degradation exceeding the safety threshold), used to penalize violations of safety constraints and prompt the algorithm to prioritize the solution that meets the safety requirements; 1 (violation / exceedance) represents the indicator function, which takes the value of 1 when a path or node exceeds the cumulative salt spray exposure limit or electrical insulation degradation exceeds the safety threshold, otherwise it takes the value of 0.

[0064] A second aspect of the present invention provides a multimodal transport system suitable for salt spray environments, for implementing the aforementioned multimodal transport method suitable for salt spray environments, comprising:

[0065] The capacity grid model construction module is used to build a multimodal transport salt spray adapted capacity grid model with city / hub nodes as vertices and connecting road segments as edges; configure salt spray adaptation devices at nodes, and mark the node protection capabilities, availability time windows and replenishment capabilities; at the same time, set model assumptions, and clarify the integrity of cargo transport, facility availability, transport rules and parameter acquisition methods.

[0066] The variable and parameter definition module is used to define multi-dimensional decision variables, parameters and state variables, and to establish a salt spray-related state variable system.

[0067] The cost calculation module is used to incorporate the environmental impact of salt spray into the cost composition, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost.

[0068] The optimization model building module is used to construct a mixed integer programming model with the goal of minimizing total cost. This model includes constraints on path connectivity and flow conservation, protection capability, energy / endurance, and corrosion / exposure, enabling global optimization of multimodal transport paths, modes of transport, and protection strategies in salt spray environments.

[0069] An improved hybrid immune-particle swarm optimization (HPS) module is designed to address the non-convex characteristics of the model. An improved HPS algorithm is developed, which optimizes the solution process through hybrid encoding, fitness function design, population initialization, hybrid immune-PSO operation, safety repair operator, and memory mechanism. The algorithm outputs the optimal path, mode selection, and protection parameter configuration.

[0070] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0071] (1) The multimodal transport method and system of the present invention applicable to salt spray environment addresses the core bottleneck of traditional intermodal transport devices in salt spray environment, namely "metal corrosion, electrical failure and interface failure", and achieves a breakthrough through two key designs: On the one hand, special devices such as anti-corrosion coated cargo holds, anti-salt sealed docking devices, cathodic protection stations and electrical protection cabins are configured in the nodes of the transport capacity grid model to block the salt spray erosion path from three levels: physical isolation (sealing), chemical protection (cathodic protection, coating) and environmental control (dehumidification / drying), directly reducing the corrosion rate and failure risk of key components such as mechanical seals, bearings, sensors and power battery packs. On the other hand, by defining parameters such as metal corrosion rate function, electrical insulation degradation factor, and salt spray acceleration factor, a quantitative model containing salt spray acceleration factor and corrosion rate function is constructed, quantifying the impact of salt spray on equipment into calculable model variables. At the same time, decision items such as whether to activate cathodic protection and whether to recoat on-site are incorporated into the optimization model to achieve proactive scheduling of equipment protection. For example, after high salt spray concentration sections, the life of equipment can be extended by coating replenishment at transfer nodes, avoiding emergency repairs after failures caused by traditional passive maintenance, and significantly improving the mean time between failures (MTBF) of multimodal transport equipment. In addition, the model can be flexibly adapted to different salt spray scenarios such as coastal ports, islands, and the Maritime Silk Road, solving the problem that traditional mild environment designs for intermodal transport cannot adapt to special scenarios.

[0072] (2) The multimodal transport method and system of the present invention applicable to salt spray environment ensures the accuracy of loading and unloading connection of multiple transport modes through standardized anti-corrosion interface. Combined with corrosion exposure constraints and safety repair operators, it can automatically avoid failure nodes and switch low salt spray routes to avoid the entire chain from stopping at one fault, thus greatly improving the transport continuity compliance rate. It incorporates the unique cargo damage, protection energy consumption and equipment maintenance costs of salt spray into the total cost model, and achieves the minimization of the entire chain cost of transportation, protection and loss through mixed integer programming. At the same time, it dynamically adjusts the protection power (such as reducing energy consumption during low salt spray periods) and optimizes the transfer scheduling to reduce ineffective expenses and significantly reduce the overall operating cost.

[0073] (3) The multimodal transport method and system of the present invention, applicable to salt spray environments, constructs a total cost model that includes basic transportation costs, transshipment costs, loss / cargo damage costs, energy / cooling costs, equipment failure and maintenance costs, time window penalties and demurrage costs, etc., and optimizes it using an improved algorithm to achieve optimal configuration of transport routes, protection strategies, energy consumption scheduling, etc. This reduces equipment replacement and maintenance costs caused by salt spray corrosion, lowers the risk of cargo damage, and avoids additional costs caused by transportation timeliness issues, significantly improving the economic benefits of multimodal transport in salt spray environments.

[0074] (4) The multimodal transport method and system of the present invention applicable to salt spray environment optimizes the non-convex characteristics of MINLP model by improving the hybrid immune-particle swarm algorithm, taking into account both global search (immune genetic operation) and local precision (PSO operation). Compared with the traditional genetic algorithm, the solution efficiency is greatly improved, and a complete solution of optimal path, protection strategy and control parameters can be output in a short time. At the same time, the algorithm output results can directly correspond to actual operation without secondary conversion, which is convenient for engineers to execute directly. Attached Figure Description

[0075] Figure 1 This is a flowchart of a multimodal transport method applicable to salt spray environments according to an embodiment of the present invention;

[0076] Figure 2 This is a schematic diagram of a multimodal transport salt spray adaptation capacity grid model according to an embodiment of the present invention;

[0077] Figure 3 This is a schematic diagram of the solution process of the improved hybrid immune-particle swarm optimization algorithm according to an embodiment of the present invention;

[0078] Figure 4 This is a schematic diagram of hybrid encoding according to an embodiment of the present invention;

[0079] Figure 5 This is a schematic diagram illustrating the simulation curves of protection and energy consumption under salt spray conditions according to an embodiment of the present invention.

[0080] Figure 6This is a schematic diagram of the structure of a multimodal transport system suitable for salt spray environments according to an embodiment of the present invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0082] like Figure 1 As shown, one aspect of the present invention provides a multimodal transport method suitable for salt spray environments, comprising:

[0083] S1: Construct a multimodal transport salt spray adaptation capacity grid model with city / hub nodes as vertices and connecting road segments as edges; configure salt spray adaptation devices at nodes, and label the node protection capabilities, availability time windows, and replenishment capabilities; at the same time, set 4 model assumptions to clarify the integrity of cargo transport, facility availability, transport rules, and parameter acquisition methods.

[0084] S2: Define multi-dimensional decision variables, parameters and state variables, and establish a salt spray-related state variable system;

[0085] S3: Incorporate the environmental impact of salt spray into the cost structure, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost;

[0086] S4: With the goal of minimizing total cost, a hybrid integer programming model is constructed, which includes path connectivity and flow conservation, protection capability constraints, energy / endurance constraints, and corrosion / exposure constraints. To address the non-convex nature of the model, an improved hybrid immune-particle swarm optimization algorithm is designed. The solution process is optimized through hybrid encoding, security repair operators, and memory mechanisms, and the optimal path, mode selection, and protection parameter configuration are output.

[0087] Furthermore, such as Figure 2 As shown, the multimodal transport salt spray adaptation capacity grid model described in step S1 uses several city / hub nodes N as vertices and several connecting road segments E as edges. The set of transport modes on the edges is denoted as K = {1, ..., K}. Each node is equipped with a salt spray / corrosion adaptation device and an interface unit. The node attributes should indicate its protection capability, availability time window, and replenishment capability.

[0088] The city / hub node N includes ports, railway marshalling yards, highway transfer yards, and island supply stations; the connecting road segment E includes one or more modes of transportation such as highways, railways, waterways, and aviation; the salt spray / corrosion adaptation device includes anti-corrosion coated cargo holds, salt-proof sealing docks, dehumidification / drying chambers, anti-corrosion supply / cathode protection stations, sand and dust filtration units, and electrical protection chambers; the node protection capability includes coating grade, dehumidification capacity, cathodic protection capacity, filtration capacity, and electrical sealing grade;

[0089] Based on the above grid, by selecting nodes, road segments and corresponding transportation modes, the collaborative work of cargo transportation routes and multimodal transport devices / systems under salt spray environment is simulated. The organization mode selection and parameter scheduling are carried out with the total cost / corrosion risk / transportation time as optimization objectives (e.g., whether to activate cathodic protection at a certain node, whether to adopt delayed anti-corrosion, whether to prioritize low salt spray exposure routes, etc.).

[0090] This invention classifies applicable intermodal transport modes based on the salt spray intensity along the route, the protection capabilities of nodes, and the accessibility of resupply, such as "low exposure priority mode (prioritizing inland / barrier corridors)," "mid-way reprotection mode (via nodes with recoating / cathode protection)," and "short-haul high-frequency switching mode (shortening continuous exposure time)." Standard operating procedures (SOPs) are defined for each type of mode, including protection activation thresholds, resupply sequences, and interface checklists, to facilitate rapid on-site execution.

[0091] Furthermore, in step S1, to facilitate modeling and solving, four key assumptions are formulated to simplify the modeling complexity based on the characteristics of the salt spray environment. These assumptions can be relaxed in specific implementations. The model condition assumptions include:

[0092] Goods are transported in containers, and the goods within the containers are not split up for transport (holistic assumption);

[0093] Salt spray adaptation devices and protective facilities marked in the area are in an available state during planning; if a node / facility is unavailable, it is set as unreachable or limited capacity in the model;

[0094] The vehicles operate on the road segment according to the established driving rules, and each transfer node can be loaded and unloaded within the predetermined time window; if a node is temporarily unavailable due to salt spray, it will be reflected by the node reachability parameter.

[0095] All corrosion / energy consumption / failure model parameters (such as metal corrosion rate, coating decay rate, insulation degradation factor, protection system power consumption, component Weibul failure parameters, etc.) can be estimated through field tests or historical data and substituted into the model;

[0096] Furthermore, step S2 includes defining the symbol system for decision variables, parameters, and state variables of the model; the decision variables cover discrete and continuous variables such as road segment-transportation mode selection, node transfer operation, container quantity, and node protection power scheduling; the parameters and state variables include road segment distance, salt spray exposure concentration time series, metal corrosion rate function, and Weibull failure parameters of key components, especially clarifying the correlation between salt spray exposure concentration and corrosion and failure;

[0097] The decision variables and symbols are as follows:

[0098] Represents a set of city / hub nodes, indexed E represents the set of road segments, and a road segment is denoted as (i,j)∈E; Represents a set of transportation modes, index (1: Highway, 2: Railway, 3: Waterway, 4: Aviation, etc.);

[0099] Define decision variables: This indicates that if transportation mode k is used on road segment (i,j), the value is 1; otherwise, it is 0 (road segment - mode selection). This indicates that if a transit / connection operation occurs at node i, changing from mode p to mode q, then the value is 1 (transit selection); n c ∈Z + This represents the number of containers used (which can be expanded to a vector of container counts for each route); This represents the rate at which the power or chemical protection used for corrosion prevention / dehumidification / cathode protection at node i is consumed at time t. This represents the running speed (a controllable continuous variable) of mode k on road segment (i,j); Indicates whether on-site coating / recoating operations are performed at node i (0 indicates no, 1 indicates yes); Indicates whether the node / vehicle has enabled dehumidification mode (0 indicates no, 1 indicates yes);

[0100] Define parameters and state variables; d ij This represents the distance (km) of road segment (i,j); This represents the transportation time (in hours) on road segment (i,j) using method k. This represents the basic unit transportation price (currency / container·km) for a unit container and unit distance under mode k; This represents the transit time (in hours) at node i for transferring from mode p to mode q. This represents the unit transit cost (currency / box) for making this transit at node i; S indicates the volume or mass of a single container of goods (used for heat load calculation); salt (t) represents the time series of salt spray exposure concentrations along the road segment or node (mg / m³). 2• day or related measure); κ corr (S,T) represents the metal corrosion rate function (dependent on salt spray concentration S and temperature and humidity conditions T); α ins (t): Electrical insulation degradation factor as a function of exposure time;

[0101] Define energy / equipment parameters: E rated Indicates the battery's rated capacity, Indicates the rated power of the protective device;

[0102] Define failure / loss parameters: Weibull three parameters (η, β, γ) represent the life / failure distribution of critical components, and the parameters vary with the salt spray exposure acceleration factor;

[0103] This invention adds salt spray-related fields to the node / segment database: typical salt spray deposition rate, tidal / wind direction sensitivity, node protection level (coating life, cathodic protection capacity, dehumidification capacity), and chemical protective material inventory and replenishment delay; it couples the salt spray exposure time series with the vehicle / cargo box transit time to calculate the cumulative exposure and use it as a route feasibility and maintenance trigger condition.

[0104] Furthermore, the expression for the salt spray-related total cost model described in step S3 is as follows:

[0105] C total =C trans +C transf +C loss +C energy +C maint +C safety +C time (1)

[0106] Among them, C total For total cost, C trans Based on basic transportation costs, C transf For transit costs, C loss For loss / damage costs, C energy For energy consumption / cooling cost, C maint For equipment failure and maintenance costs, C time Penalties and costs associated with time windows;

[0107] When mode k is used on road segment (i,j), the basic transportation cost is related to distance and the selected speed (speed affects energy consumption but time); the basic transportation cost C trans The expression is:

[0108]

[0109] Where, n c This refers to the total number of containers transported or the number of containers carried on the corresponding route. The unit transportation price (currency / box·km) for segment k on road segment (i,j); d ij This represents the distance (km) of road segment (i,j); This represents the decision variable; if this road segment / mode is selected, it is 1.

[0110] If a transition from mode p to mode q occurs at node i, the transition cost and time must be factored in, along with a time increment. The transit cost C transf The expression is:

[0111]

[0112] in, This represents the unit transfer cost (currency / box) for node i when transferring from mode p to mode q; This indicates that if the transfer is performed at node i, the value is 1; n c This refers to the total number of containers transported or the number of containers carried on the corresponding route.

[0113] Furthermore, salt spray exposure increases the risk of damage to goods (especially electronics, metal components, packaging seals, and sensitive materials). A failure probability function F(·) influenced by salt spray exposure and protective measures is introduced. Based on this distribution, the loss / damage cost C... loss The expression is:

[0114] C loss =n c ·P unit ·F(T expose S salt (t),prot(t)) (4)

[0115] Where, n c P represents the number of containers used. unit For the value of a single box of goods, T expose S represents the cumulative time that the cargo has been exposed to salt spray. salt (t) represents the salt spray exposure intensity (e.g., mg / m²·day or normalized exponent) of the path / node at time t; prot(t) represents the timing of the protective measures taken (e.g., whether to activate coating re-spraying, dehumidification, cathodic protection, etc., which can reduce the probability of failure); F(·) represents the loss / failure probability function, taking values ​​from 0 to 1, using a Weibull distribution and introducing a salt spray exposure correction factor a. salt (S,t);

[0116] Weibull-type example (with salt spray acceleration factor): To mathematically represent the effects of salt spray, the following modified Weibull cumulative failure probability can be used:

[0117]

[0118] Among them, a salt (S,t)≥1 is the salt spray acceleration factor (the stronger the salt spray or the worse the protection, the greater the acceleration factor). salt The larger the value, the greater the equivalent scaling parameter η / a. salt (S,t) decreases, failure accelerates); η represents the scale parameter (characterizing the lifetime scale, time unit); β represents the shape parameter (>1 indicates that the failure rate increases with time); γ represents the location parameter (failure delay time); a salt (S,t) represents the salt spray acceleration factor, and its functional form can be based on empirical / experimental fitting, for example, a salt (S,t)=1+κ1·S salt (t)+k2. (Lack of protective factors);

[0119] Furthermore, the energy consumption / cooling cost C energy The expression is:

[0120] C energy =∑ t (λ elec P elec (t)+λ chem P chem (t)+λ prot P prot (t))Δt (6)

[0121] Among them, C energy λ represents the total energy cost; elec P represents the unit price of electricity. elec (t) represents the power consumption at time t, and P represents the total power consumption of the dehumidifier, fan, and cathodic protection power supply; chem (t) represents the chemical / coating consumption rate at time t; λ chem P is the unit price of chemicals or coating materials. prot (t) represents the power or energy specifically used for protection at time t; λ prot This represents the unit cost of power / energy specifically used for protection (if combined, it can be combined with λ). elec (To be processed together); Δt represents the time step (hours or days);

[0122] Furthermore, for critical components (electrical connectors, seals, bearings, sensors, etc.), the failure probability is accelerated by salt spray, and should be included in the maintenance / replacement cost based on the expected number of failures or the failure probability; the equipment failure and maintenance cost C maint The expression is:

[0123] C maint =∑comp Cost comp ·Pr[comp fails during mission] (7)

[0124] Pr[comp fails during mission]=F comp (t run η comp / a salt ,β comp ,γ comp (8)

[0125] Among them, C maint Cost represents the total cost of equipment failure and maintenance, calculated based on the expected maintenance cost derived from the failure probability of critical components. comp This represents the direct cost of replacing or repairing a component during the mission period, including material and labor costs; Pr[comp fails during mission] represents the component's failure time t during operation. run The probability of a failure occurring within the system, expressed in terms of Weibull's formula including the salt spray acceleration factor;

[0126] F comp (t run η comp / a(T),β comp ,γ comp ) represents the Weibull distribution function form used to calculate the probability of component failure; t run For the transportation task running time; η comp ,β comp ,γ comp Three Weibull parameters representing the component (estimated from field / experimental data); η comp β is a characteristic life parameter of a key component, reflecting the component's lifespan characteristics under normal temperatures; comp β is a dimensionless shape parameter of a critical component, representing the shape parameter of a Weibull distribution. It determines the shape of the curve showing the failure probability changing over time. If β > 1, it indicates that the component failure probability increases with time; if β = 1, it follows an exponential distribution. comp Let γ be the position parameter of a critical component, representing the minimum time at which the component begins to fail. comp =0, then the distribution accelerates from time zero; a salt This is a salt spray acceleration factor, which can be adjusted according to the local salt spray intensity and protection status. The stronger the salt spray or the worse the protection, the larger this factor will be, which will accelerate the failure rate of components.

[0127] If corrosion depth d is used corr (t) can be used as a maintenance trigger indicator:

[0128]

[0129] Where, d corr (t) represents the change in corrosion depth over time, obtained by analyzing the corrosion rate κ. corr (S salt Integrating (τ),T(τ)) over the time interval [0, t] yields S; salt (τ) is the salt spray concentration at time τ; T(τ) is the temperature at time τ that affects corrosion.

[0130] When d corr (t)≥d crit When a replacement / repair operation is triggered, its expected cost can be assessed based on the trigger probability and added to C. maint ;d crit This is the critical value for corrosion depth;

[0131] Formulas (7) and (8) indicate that the total maintenance cost of the equipment is the sum of the products of the "single component direct maintenance cost" and the "probability of failure of the component during the mission period" of all critical components (such as electrical connectors, seals, bearings, sensors, etc.). Because salt spray will accelerate the failure of these critical components, a salt spray acceleration factor is introduced when calculating the failure probability to more accurately reflect the failure of components in the salt spray environment, and thus obtain the maintenance cost that conforms to the actual salt spray environment.

[0132] Formula (9) means: used to calculate the corrosion depth of the component from time 0 to t; the corrosion rate is affected by factors such as salt spray concentration and temperature. By integrating the corrosion rate over time, the corrosion depth at different times can be dynamically obtained; when the corrosion depth reaches the critical value, the component needs to be replaced or repaired, and the maintenance cost triggered by corrosion will be evaluated and added to the total maintenance cost.

[0133] Furthermore, if a delivery time window exists (e.g., arrival must not be later than T), due Or not earlier than T earliest Then, the cost of detention / penalty is introduced; the time window penalty and the cost of detention C time The expression is:

[0134]

[0135] Where 1{·} is an indicator function, the function value is 1 when the condition in the parentheses is met; T arrive,s T represents the actual time of arrival at station s; due,s P represents the latest allowed arrival time for station s (the upper bound of the time window); exceeding this time is considered late. late,sThe costs of late arrival penalties for site s (e.g., late payment penalties per order, service quality deductions, contract breach costs); T earliest,s This is the earliest allowed arrival time for site s (lower bound of the time window). Arrivals earlier than this time are considered too early and require waiting or storage. store,s The costs of early arrival storage or waiting at site s (e.g., costs incurred for refrigerated goods or waiting for vehicles);

[0136] Furthermore, the expression for the mixed-integer programming model in step S4 is:

[0137] m = C trans +C transf +C loss +C energy +C maint +C time (11)

[0138] The objective function aims to minimize the total comprehensive cost of multimodal transport. By comprehensively considering six types of costs—transportation, transshipment, cargo damage, energy consumption, equipment maintenance, and time-related costs—it seeks to find the optimal solution for the organization and parameter scheduling of the multimodal transport system under salt spray conditions, thereby achieving cost optimization.

[0139] The following constraints are established for the mixed-integer programming model, including:

[0140] Path connectivity and flow conservation (ensuring there is a feasible path from the originating node s to the destination node t):

[0141]

[0142] Where u represents an intermediate node;

[0143] Protection capability constraints (maximum protection power and resource limitations of nodes / vehicles under their load):

[0144]

[0145] in, Let be the protection power of node i at time t; Maximum protection power of the node; Indicates whether on-site coating / recoating operations are performed at node i (0 indicates no, 1 indicates yes); Indicates whether the node / vehicle has enabled dehumidification mode (0 indicates no, 1 indicates yes);

[0146] Energy / range constraints (energy required for protection and dehumidification is limited by battery or fuel):

[0147] ∑ t P elec (t)Δt≤E available ,∑t P chem (t)Δt≤Chem avail (14)

[0148] Among them, P elec (t) represents the power consumed at time t; Δt represents the time step, used to convert the power integral into energy; E available P represents the total usable electrical energy. chem (t) represents the consumption rate of the chemical / coating at time t; Chem avail Indicates the total amount of available chemical materials;

[0149] By limiting the energy required for protection and dehumidification to no more than the available energy provided by batteries or fuel, sufficient energy supply is ensured during transportation and will not be interrupted due to insufficient energy.

[0150] Docking / node capacity constraints (node ​​transfer throughput, maximum power of heating facilities):

[0151]

[0152] in, This indicates whether a transfer from mode p to mode q occurs at node i (a binary variable, 1 indicates occurrence, 0 indicates no occurrence); n c This indicates the number of cargo units transshipped, measured in containers. This represents the maximum transfer capacity of node i, i.e., the number of containers that can be transferred per unit of time.

[0153] Time window constraint (arrival time meets customer time limit):

[0154] T arrive ≤T due +Δ allow ,T arrive ≥T earliest (16)

[0155] Among them, T arrive Indicates the arrival time of the vehicle or goods; T due Indicates the customer's requested delivery deadline; Δ allow Indicates the allowed delay buffer time; T earliest This indicates the earliest time when goods can be received, preventing premature arrival from causing stockpiling or storage costs.

[0156] Binary and physical variable domain constraints:

[0157]

[0158] in, This represents whether to choose the k-th transportation method from node i to node j; it is a binary variable. This indicates whether a mode of transport change occurs at node i; it is a binary variable. This represents the heating power supplied at time t, and it must be non-negative. This represents the speed of transportation mode k on path (i,j), ranging from... This represents the minimum operating speed of transportation mode k on path (i,j); This represents the maximum speed of transportation mode k on path (i,j);

[0159] Corrosion / Exposure Constraints (limiting the cumulative exposure of cargo containers on a certain path segment to not exceed a threshold; otherwise, on-site re-protection or route rerouting must be enforced):

[0160]

[0161] in, The salt spray concentration S represents the concentration of salt spray within the time interval from t0 to t1. salt (t) is the integral over time, representing the cumulative salt spray exposure of the cargo container during that time period; S crit This is the critical threshold for salt spray exposure. When the cumulative salt spray exposure exceeds this value, protective measures or route rerouting are required. coat This is the decision variable for whether to perform on-site coating work, with a value of 1 (execute) or 0 (do not execute);

[0162] Formula (18) represents the cumulative amount of salt spray the cargo container is exposed to during the time interval t0 to t1 (by analyzing the salt spray concentration S). salt (t) Integrating over this time interval yields a result that cannot exceed the critical threshold S. crit If this threshold is exceeded, on-site coating operations need to be forcibly triggered (at this time u coat =1, u coat The decision variable for whether to perform on-site coating operations (1 indicates execution, 0 indicates non-execution) or change the transportation route, thereby controlling the degree of corrosion of the cargo box due to salt spray exposure and ensuring the safety of goods or transportation equipment;

[0163] The mixed-integer programming model is an optimization model that simultaneously includes both discrete and continuous variables. Its objective function or constraint graph is not a convex function; multiple local minima exist on the graph, making it impossible to determine which valley is the true global minimum (i.e., the optimal cost solution). Therefore, the mixed-integer programming model (MINLP) is a non-convex problem. Its main nonlinearity stems from the corrosion-failure model, energy consumption-protection coupling, and indicator function terms. It can be solved using hierarchical optimization, linear approximation, or metaheuristic global search methods.

[0164] Specifically, in step S3, the loss / damage cost C loss And equipment failure and maintenance costs C maint In this study, we used the Weibull failure probability model to calculate the failure probability caused by salt spray. The formula includes "exponential functions" and "power functions," and its graph is a curve rather than a straight line (non-linear). In terms of energy consumption / cooling cost C... energy In this context, the energy consumption of protective measures is "coupled" with the salt spray environment and the protective effect, and this relationship is not linear; the time window penalty and the cost of stay C... time In the process, an "indicator function" is used to determine whether a penalty is triggered. The graph of the indicator function is a "piecewise function". Sudden changes will occur at the critical points of "overdue / not overdue" and "early / not early", which are typical nonlinear terms, making the model unsolvable by conventional linear methods.

[0165] This invention incorporates corrosion-driven cargo damage costs and equipment maintenance costs into the objective function, and includes protection energy consumption and supply logistics in the constraint layer; the constraints mainly reflect the upper limit of protection resources, node recoating time window and energy / chemical supply capacity, and the model solution needs to simultaneously determine the path, node protection activation plan and protection power time series.

[0166] Furthermore, such as Figure 3 As shown, step S4 is to efficiently solve the complex problem of discrete path selection and continuous scheduling decision-making. An improved Hybrid PSO-Immune algorithm is proposed. Through hybrid coding, fitness function design, population initialization, immune-PSO hybrid operation, diversity maintenance and other steps, the optimal intermodal transport scheme is efficiently solved and the output includes the path, transport mode, protection strategy and control parameters.

[0167] Specifically, it includes:

[0168] The antibody / particle structure is constructed using a three-segment hybrid encoding method. This three-segment hybrid encoding means that a string contains three parts: a path-mode binary segment, a protection / module activation segment, and a continuous segment. The specific encoding rules are as follows (e.g., ...). Figure 4 As shown):

[0169] Path-Mode Binary Segment: Each candidate road segment (i,j) and each mode of transport k are encoded using 0 / 1 binary (0 = not selected, 1 = selected), or simplified to a compressed encoding of "route node sequence + mode sequence"; where the route node sequence uses 0 / 1 to mark the selection status of the candidate road segment (i,j), 1 indicates that the road segment is selected, and 0 indicates that it is not selected, and the path continuity is ensured through node connection relationship; the mode of transport sequence uses 1 / 2 / 3 / 4 to correspond to road / rail / waterway / air transport respectively, realizing the mapping of the discrete variable of "road segment-mode of transport selection";

[0170] Protection / Module Activation Section: Whether the discrete decision variables specific to the salt spray scenario are enabled is represented by binary encoding (0 = disabled, 1 = enabled); the discrete decision variables specific to the salt spray scenario include whether the node enables coating respray, cathodic protection, dehumidification, filtration, and night operation;

[0171] Continuous Quantity Segment: Real-valued encoding is used for continuous decision variables in the model, and the value range matches the physical constraints of the variables; the continuous decision variables include protection power. driving speed Chemical protection delivery rate Real-valued data is directly used as the optimization object of the particle swarm optimization algorithm, and the parameters are finely adapted through local optimization. Through this encoding, discrete path selection, protection activation decision and continuous control parameters are uniformly incorporated into the algorithm processing framework.

[0172] With the goal of minimizing total cost, the reciprocal of the objective function is used as the basis for the fitness function. At the same time, an indicator function is introduced. If the salt spray exposure of a path or node exceeds the limit (such as cumulative salt spray exposure or electrical insulation degradation exceeding the safety threshold), the fitness is penalized to ensure that the evaluation results reflect both the degree of cost optimization and the safety constraints under the salt spray environment.

[0173] Generate an initial population of F individuals, including heuristic and stochastic solutions that prioritize low-salt-spray coastlines, enable intermediate node spraying, and shorten exposure time; the continuous segment variables are initialized based on uniform distribution or engineering experience to ensure the diversity and engineering practicality of the initial population.

[0174] For discrete segments (path / module activation), immune genetic operations (including selection, crossover, and mutation) are employed to maintain population diversity and perform global search. For continuous segments (protection power, driving speed, etc.), particle swarm optimization (PSO) velocity-position update rules are used to accelerate local accurate search. At the same time, a safety repair operator is introduced. If the solution violates corrosion / safety constraints, it is repaired by increasing protection activation bits, increasing dehumidification / cathode protection power, or replacing it with low-exposure road segments, so that the solution is restored to the feasible region of the constraints.

[0175] The similarity of individuals is calculated based on the Hamming distance of binary segments to maintain the diversity of the population within the threshold range; the best solutions are stored in the memory bank for subsequent crossover operations to generate offspring, thus avoiding the algorithm from getting stuck in local optima and improving the global optimization ability.

[0176] The algorithm terminates when the iteration limit is reached or the fitness converges; the optimal antibody is extracted from the memory, and the corresponding transportation route, transportation mode selection, transit node protection strategy, and continuous control parameters such as protection power and driving speed are analyzed to serve as the optimal solution for multimodal transport under salt spray environment.

[0177] Furthermore, the fitness function expression is:

[0178]

[0179] Among them, C total is the total cost in the mixed-integer programming (MINLP) model; M is the penalty coefficient when a path or node exceeds the exposure threshold (such as exceeding the cumulative salt spray exposure limit or electrical insulation degradation exceeding the safety threshold), used to penalize violations of safety constraints and prompt the algorithm to prioritize the solution that meets the safety requirements; 1 (violation / exceedance) represents the indicator function, which takes the value of 1 when a path or node exceeds the cumulative salt spray exposure limit or electrical insulation degradation exceeds the safety threshold, otherwise it takes the value of 0;

[0180] Formula (19) is based on the total cost of multimodal transport, and its reciprocal is used as the core calculation part of fitness, while also considering the penalty for breach of contract / over-limitation.

[0181] When there are no defaults / overlimits (1=0), fitness At this point, the total cost C total The smaller the size, the greater the adaptability, which means the solution is more cost-effective;

[0182] When there is a breach / over-limit situation (1=1), fitness At this point, due to the addition of the penalty coefficient M, the fitness will decrease significantly, indicating that the scheme will not be preferred because it violates the safety constraints.

[0183] The present invention also provides a specific implementation step, as follows:

[0184] 1) Regional Capability and Pattern Prediction: Conduct a capacity survey of the target area (coastal / port clusters, islands and inland transshipment stations), record the corrosion protection level of each node (coating level, cathodic protection equipment, dehumidification capacity, chemical protection inventory, online detection capability, etc.), and give the node capability level and preliminary intermodal transport organization mode (e.g., "prioritize low salt spray routes + intermediate node coating regeneration").

[0185] 2) Establish a salt spray transportation capacity grid database: Select several nodes in the study area as grid vertices, and record road segments, transportation modes, and node capabilities (including distance, typical salt spray deposition rate / concentration statistics S). salt (e.g., node protection capabilities, throughput capacity, cost parameters, etc.)

[0186] 3) Construct a corrosion-energy consumption and loss parameter library: calibrate the metal corrosion rate function κ using experimental or historical data. corr (S,T), Electrical insulation degradation curve α ins (t), Weibull parameters of components (η,β,γ), power consumption and efficiency of protective measures (coating life, cathodic protection efficiency, dehumidification power consumption, etc.).

[0187] 4) Establish a joint optimization model: Establish the total cost target and constraints, and set the range and time discretization of decision variables (e.g., in hourly or half-hourly increments). The constraints should include a corrosion exposure calculation module to evaluate the cumulative exposure of the path segment in real time and determine whether to trigger mandatory protection or rerouting.

[0188] 5) Set up and run the hybrid PSO-Immune algorithm to solve the problem: Set the population size F, memory capacity, crossover / mutation probability, PSO parameters (inertia weight, learning factor), and maximum number of iterations G. max Then, run the algorithm to obtain the optimal solution.

[0189] 6) Simulation Verification and Parameter Fine-tuning: Perform corrosion-energy consumption dynamic simulations on candidate schemes in MATLAB / Simulink or Python environments to verify cumulative exposure, component degradation, and energy consumption budget; such as Figure 5 The model parameters or penalty weights are adjusted based on the simulation results, and the solution is obtained again.

[0190] 7) On-site implementation: The organizational mode (path, method, transit node and module activation command) output by the algorithm is sent to the scheduling system, and the node end and vehicle end execute the corresponding protection process (e.g., on-site coating re-spray, enabling cathodic protection, enabling dehumidification / sealing mode).

[0191] 8) Online Re-optimization and Emergency Response: During transportation, the system receives real-time salt spray / humidity / equipment status data, triggering online replanning (which can employ a near-real-time version of Model Predictive Control (MPC) + fast hybrid algorithm) to address sudden node unavailability or protection failure. Emergency plans are established (e.g., temporary relocation to the nearest protection node, activation of emergency coating, isolation and replacement of high-risk electrical units).

[0192] This invention introduces a corrosion-sensitive crossover / mutation strategy during the immune-genetic operation phase: path segments affecting exposure accumulation are given a higher mutation probability to increase the search weight of low-exposure routes. An energy-constrained projection operator is introduced in the continuous segment PSO update to ensure that protection power scheduling conforms to battery / node power supply limits; simultaneously, a safety repair operator is introduced to automatically repair solutions that violate corrosion exposure constraints (by enabling protection modules or replacing road segments). A memory store is used to store several "low-exposure, high-feasibility solutions" and "emergency protection solutions" for rapid online replanning.

[0193] This invention establishes a node-level protection supply chain: standardized mobile coating regeneration vehicles, temporary cathodic protection kits, and quick-replacement connectors, reducing on-site maintenance time. It also establishes salt spray-oriented inspection / replacement cycles (based on actual corrosion rates and life models) to reduce sudden failure rates and optimize spare parts inventory levels.

[0194] This invention calibrates a corrosion / failure model by combining indoor accelerated salt spray testing with long-term field observation, verifying the contribution of coatings / materials and protection strategies to lifespan extension and maintenance cost reduction. A simulation example (a typical coastal A→B→C multimodal transport route) verifies the effectiveness of the proposed model and algorithm in reducing total cost, cargo damage rate, and extending equipment lifespan. Comparative experiments demonstrate the advantages and disadvantages of organizational strategies such as mid-journey re-protection and low-exposure priority.

[0195] like Figure 6 As shown, a second aspect of the present invention provides a multimodal transport system suitable for salt spray environments, for implementing the aforementioned multimodal transport method suitable for salt spray environments, comprising:

[0196] The capacity grid model construction module is used to build a multimodal transport salt spray adapted capacity grid model with city / hub nodes as vertices and connecting road segments as edges; configure salt spray adaptation devices at nodes, and mark the node protection capabilities, availability time windows and replenishment capabilities; at the same time, set model assumptions, and clarify the integrity of cargo transport, facility availability, transport rules and parameter acquisition methods.

[0197] The variable and parameter definition module is used to define multi-dimensional decision variables, parameters and state variables, and to establish a salt spray-related state variable system.

[0198] The cost calculation module is used to incorporate the environmental impact of salt spray into the cost composition, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost.

[0199] The optimization model building module is used to construct a mixed integer programming model with the goal of minimizing total cost. This model includes constraints on path connectivity and flow conservation, protection capability, energy / endurance, and corrosion / exposure, enabling global optimization of multimodal transport paths, modes of transport, and protection strategies in salt spray environments.

[0200] An improved hybrid immune-particle swarm optimization (HPS) module is designed to address the non-convex characteristics of the model. An improved HPS algorithm is developed, which optimizes the solution process through hybrid encoding, fitness function design, population initialization, hybrid immune-PSO operation, safety repair operator, and memory mechanism. The algorithm outputs the optimal path, mode selection, and protection parameter configuration.

[0201] A third aspect of the present invention also provides a computer-readable storage medium on which a computer program is stored, the computer program being executable by a processor. Figure 1 Any step of the multimodal transport method applicable to salt spray environments described herein.

[0202] This invention addresses the problems of accelerated equipment corrosion, electrical and sensor failures, damaged sealing of loading and unloading interfaces, and high long-term maintenance costs in multimodal transport under salt spray environments. It proposes a multimodal transport method and system for coastal ports, islands, and sea-land intermodal transport corridors. This system can simultaneously adapt to road, rail, waterway, and air transport modes, ensuring the long-term reliable operation of equipment and electronic systems in high salt spray and high humidity environments. Through the synergistic optimization of organizational models and protection strategies, it reduces the risk of cargo damage from salt spray and lowers energy consumption and maintenance costs.

[0203] By introducing a multi-layer anti-corrosion wall design (a high-reflectivity / hydrophilic dispersion coating on the surface, a passivation / cathode protection layer in the middle, and a corrosion-resistant substrate in the inner layer), and using homogeneous metal interfaces + cathode / sacrificial anode design at the joints to eliminate crevice corrosion; key load-bearing components and connecting parts are made of high corrosion-resistant alloys or lightweight composite materials that have undergone surface ion implantation / anodic oxidation treatment, the structural life is extended and the corrosion rate is reduced.

[0204] The design features a pluggable salt spray resistant electrical chamber with internal airtight isolation, inert gas filling or continuous dehumidification, and metal-plated connectors with self-recovering sealing structures. The sensors employ packaging redundancy and self-calibration mechanisms, and a distributed sensor array for corrosion rate, salt spray deposition, humidity, and electrical contact resistance is deployed at key nodes to achieve early fault detection and fault-tolerant control.

[0205] Mobile corrosion protection supply stations (including rapid coating regeneration, temporary cathodic protection access, dehumidification and dust filtration modules) are deployed at key nodes and sections of the road to achieve mid-journey re-protection and rapid on-site repair. A self-cleaning and self-sealing mechanism for the docking / unloading interface is developed, utilizing mechanical reset + micro-spray coating technology to reduce the failure rate of docking due to salt accumulation.

[0206] Incorporate salt spray exposure intensity and equipment status into energy management decisions: prioritize the allocation of cathodic protection and dehumidification energy during periods of high exposure, otherwise reduce protection power to save energy; combine weather forecasts and tidal / wind direction information to make short-term protection strategy switching (such as temporarily activating high-power cathodic protection or changing the route to avoid high-exposure sections).

[0207] A capacity grid model is established to consider the cost of salt spray exposure and equipment failure rate. In route selection, the cumulative exposure, re-protection cost, and maintenance risk are taken as decision factors. An improved hybrid optimization algorithm (e.g., Hybrid PSO–Immune modified for salt spray conditions) is proposed: the discrete part is used for route / module activation selection, the continuous part is used for protection power and dehumidification scheduling, and a safety repair operator is introduced (if the solution violates the exposure constraint, protection activation or route replacement is triggered).

[0208] Based on accelerated corrosion tests and field sensor data, a hybrid Weibull / Arrhenius component life model is established and calibrated in real time to enable risk-based maintenance decisions (predictive replacement vs. emergency repair). The system supports online re-optimization (MPC + fast hybrid algorithm) to address node protection failures or sudden salt spray events, ensuring transportation continuity and safety.

[0209] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multimodal transport method suitable for salt spray environments, characterized in that, include: S1: Construct a multimodal transport salt spray adaptive transport capacity grid model with city / hub nodes as vertices and connecting road segments as edges; Salt spray adaptation devices are configured at nodes, and node protection capabilities, availability time windows, and resupply capabilities are marked; at the same time, model assumptions are set to clarify the overall cargo transportation, facility availability, transportation rules, and parameter acquisition methods. S2: Define multi-dimensional decision variables, parameters and state variables, and establish a salt spray-related state variable system; S3: Incorporate the environmental impact of salt spray into the cost structure, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost; S4: With the goal of minimizing total cost, a hybrid integer programming model is constructed, which includes path connectivity and flow conservation, protection capability constraints, energy / endurance constraints, and corrosion / exposure constraints. To address the non-convex nature of the model, an improved hybrid immune-particle swarm optimization algorithm is designed. The solution process is optimized through hybrid encoding, security repair operators, and memory mechanisms, and the optimal path, mode selection, and protection parameter configuration are output.

2. The multimodal transport method suitable for salt spray environments according to claim 1, characterized in that, The multimodal transport capacity grid model that can simulate high-temperature transport coordination described in step S1 uses several city / hub nodes as vertices and one or more transport modes such as highway, railway, waterway, and air connecting sections as edges; each node is equipped with a salt spray / corrosion adaptation device and interface unit; its protection capability, availability time window and replenishment capability should be marked in the node attributes; The city / hub nodes include ports, railway marshalling yards, highway transfer stations, and island supply stations; The salt spray / corrosion adapter includes a corrosion-resistant coated cargo hold, a salt-resistant sealing dock, a dehumidification / drying chamber, a corrosion-resistant supply / cathode protection station, a sand and dust filtration unit, and an electrical protection chamber. The node protection capabilities include coating grade, dehumidification capacity, cathodic protection capacity, filtration capacity, and electrical sealing grade.

3. The multimodal transport method suitable for salt spray environments according to claim 2, characterized in that, The model assumptions mentioned in step S1 include: goods are transported in containers, and goods within containers are not split up for transport (holistic assumption); Salt spray adaptation devices and protective facilities marked in the area are in an available state during planning; if a node / facility is unavailable, it is set as unreachable or limited capacity in the model; The vehicles operate on the road segment according to the established driving rules, and each transfer node can be loaded and unloaded within the predetermined time window; if a node is temporarily unavailable due to salt spray, it will be reflected by the node reachability parameter. All corrosion / energy consumption / failure model parameters, including metal corrosion rate, coating decay rate, insulation degradation factor, protection system power consumption, and component Weibul failure parameters, are estimated through field tests or historical data and substituted into the model.

4. A multimodal transport method suitable for salt spray environments according to any one of claims 1-3, characterized in that, The expression for the salt spray-related total cost model in step S3 is as follows: （1） in, For total cost, Based on transportation costs, For transit costs, For loss / damage costs, For energy consumption / cooling costs, For equipment failure and maintenance costs, Penalties and costs associated with time windows; The cost of loss / damage The expression is: （4） in, The number of containers used. The value of a single box of goods. This refers to the cumulative time that the goods have been exposed to the salt spray environment. This represents the salt spray exposure intensity of the path / node at time t; Indicates the sequence of protective measures taken; The function represents the probability of loss / failure, taking values ​​from 0 to 1; a Weibull distribution is used, and a salt spray exposure correction factor is introduced. Obtain the corrected Weibull cumulative failure probability. The expression is: （5） in Indicates the scale parameter; Indicates shape parameters; Indicates positional parameters; This indicates the salt spray acceleration factor.

5. A multimodal transport method suitable for salt spray environments according to claim 4, characterized in that, Energy consumption / cooling cost The expression is: （6） in, This indicates the total energy cost; The unit price of electricity; For a moment Power consumption is the sum of the power consumption of the dehumidifier, fan, and cathodic protection power supply; Indicates the first The rate of chemical / coating consumption at any given time; The unit price of chemicals or coating materials; Indicates the first Power or energy specifically designed for protection at all times; This indicates the unit cost of power / energy specifically used for protection; Indicates the time step.

6. A multimodal transport method suitable for salt spray environments according to claim 5, characterized in that, The equipment failure and maintenance cost C maint The expression is: C maint =∑ comp Cost comp ·Pr[comp fails during mission] (7) Pr[comp fails during mission]=F comp (t run ;η comp / a salt ,β comp ,γ comp ) (8) Among them, C maint Cost represents the total cost of equipment failure and maintenance, calculated based on the expected maintenance cost derived from the failure probability of critical components. comp This represents the direct cost of replacing or repairing a component during the mission period, including material and labor costs; Pr[comp fails during mission] represents the component's failure time t during operation. run The table that malfunctioned internally; F comp (t run η comp / a(T), β comp γ comp ) represents the Weibull distribution function form used to calculate the probability of component failure; t run (for transport mission operation); η comp The characteristic lifespan of a key component; β comp The shape distribution of a key component; γ comp To quickly locate a key component; a salt This is a salt spray acceleration factor, which can be adjusted according to the local salt spray intensity and protection status. The stronger the salt spray or the worse the protection, the larger this factor will be, which will accelerate the failure rate of components. If corrosion depth d is used corr (t) can be used as a maintenance trigger indicator: Where, d corr (t) represents the change in corrosion depth over time; S salt (τ) is the salt spray concentration at time τ; T(τ) is the temperature at time τ that affects corrosion. When d corr (t)≥d crit When a replacement / repair operation is triggered, its expected cost can be assessed based on the trigger probability and added to C. maint ;d crit This represents the critical depth of corrosion.

7. A multimodal transport method suitable for salt spray environments according to claim 6, characterized in that, The mixed-integer programming model expression in step S4 is: （11） The protection capability constraint is: （13） in, for exist The protection power at all times; Maximum protection power of the node; Represents a node Whether on-site coating / recoating operations are carried out (0 indicates no, 1 indicates yes); Indicates whether the node / vehicle has enabled dehumidification mode (0 indicates no, 1 indicates yes); The energy / range constraint is: （14） in, For a moment Power consumption; Indicates the time step; This represents the total usable energy. Indicates time The rate of consumption of chemicals / coatings; Indicates the total amount of available chemical materials; The corrosion / exposure constraint is: （18） in, Indicates time from arrive Within the range, salt spray concentration The integral over time represents the cumulative salt spray exposure of the cargo container during that time period; This is the critical threshold for salt spray exposure. When the cumulative salt spray exposure exceeds this value, protective measures or route changes are required. This is a decision variable for whether to perform on-site coating operations, with a value of 1 indicating execution or 0 indicating non-execution.

8. A multimodal transport method suitable for salt spray environments according to any one of claims 1-4, 6, and 7, characterized in that, Step S4 employs a three-segment hybrid encoding method to construct the antibody / particle structure. The three-segment hybrid encoding means that a single string contains three parts: a path-mode binary segment, a protection / module activation segment, and a continuous segment. The specific encoding rules are as follows: Path-mode binary segment: Use 0 / 1 binary encoding for each candidate route segment (i,j) and each mode of transport k, or simplify it to compressed encoding of "route node sequence + mode sequence"; Protection / Module Activation Section: Whether the discrete decision variables specific to the salt spray scenario are enabled is represented by binary encoding; the discrete decision variables specific to the salt spray scenario include whether the node enables coating respray, cathodic protection, dehumidification, filtration, and night operation; Continuous variables: Real-valued encoding is used for continuous decision variables in the model; The continuous decision variables include protection power. driving speed Chemical protection delivery rate Step S4 aims to minimize the total cost, using the reciprocal of the objective function as the basis for the fitness function; at the same time, an indicator function is introduced, which penalizes the fitness if the salt spray exposure of a path or node exceeds the limit. Generate an initial population of F individuals, including a heuristic solution that prioritizes low-salt-spray coastlines, enables intermediate node spraying, and shortens exposure time, as well as a stochastic solution. Variables in continuous segments are initialized based on uniform distribution or engineering experience; For discrete segments, immune genetic operations are used to maintain population diversity and perform global search; for continuous segments, particle swarm optimization velocity-position update rules are used to accelerate local accurate search; at the same time, a safety repair operator is introduced. If the solution violates corrosion / safety constraints, it is repaired by adding protection activation bits, increasing dehumidification / cathode protection power, or replacing it with low-exposure road sections, so that the solution is restored to the constraint feasible region. The similarity of individuals is calculated based on the Hamming distance of binary segments, maintaining the diversity of the population within the threshold range; the best solutions are stored in the memory bank for subsequent crossover operations to generate offspring, avoiding the algorithm from getting trapped in local optima and improving the global optimization ability.

9. A multimodal transport method suitable for salt spray environments according to claim 8, characterized in that, The fitness function expression is: Among them, C total is the total cost in the mixed-integer programming model; M is the penalty coefficient when a path or node exceeds the exposure threshold (such as exceeding the cumulative salt spray exposure limit or electrical insulation degradation exceeding the safety threshold), used to penalize violations of safety constraints and prompt the algorithm to prioritize the solution that meets the safety requirements; 1 (violation / exceedance) represents the indicator function, which takes the value of 1 when a path or node exceeds the cumulative salt spray exposure limit or electrical insulation degradation exceeds the safety threshold, otherwise it takes the value of 0.

10. A multimodal transport system suitable for salt spray environments, characterized in that, A method for implementing multimodal transport suitable for salt spray environments as described in any one of claims 1-9, comprising: The capacity grid model construction module is used to build a multimodal transport salt spray adapted capacity grid model with city / hub nodes as vertices and connecting road segments as edges; configure salt spray adaptation devices at nodes, and mark the node protection capabilities, availability time windows and replenishment capabilities; at the same time, set model assumptions, and clarify the integrity of cargo transport, facility availability, transport rules and parameter acquisition methods. The variable and parameter definition module is used to define multi-dimensional decision variables, parameters and state variables, and to establish a salt spray-related state variable system. The cost calculation module is used to incorporate the environmental impact of salt spray into the cost composition, construct a total cost model related to salt spray, derive cost calculation formulas for multiple scenarios, and quantify the impact of salt spray on total cost. The optimization model building module is used to construct a mixed integer programming model with the goal of minimizing total cost. This model includes constraints on path connectivity and flow conservation, protection capability, energy / endurance, and corrosion / exposure, enabling global optimization of multimodal transport paths, modes of transport, and protection strategies in salt spray environments. An improved hybrid immune-particle swarm optimization (HPS) module is designed to address the non-convex characteristics of the model. An improved HPS algorithm is developed, which optimizes the solution process through hybrid encoding, fitness function design, population initialization, hybrid immune-PSO operation, safety repair operator, and memory mechanism. The algorithm outputs the optimal path, mode selection, and protection parameter configuration.

Images