Logistics park multi-capability complementary system management and control method and service platform

Abstract

The application relates to a logistics park multi-capability complementary system management and control method and a service platform. The method comprises the following steps: using a particle swarm optimization algorithm to perform population search iteration on multiple device configuration particle individuals to generate multiple first configuration particle individuals; using a target solver to perform daily operation scheduling planning solving on all particles of the first configuration particle individuals to generate corresponding daily scheduling response data, and calculating corresponding configuration fitness based on scheduling fitness of the daily scheduling response data and a target function associated with the first configuration particle individuals; based on the configuration fitness, repeatedly performing the steps of generating corresponding first configuration particle individuals and daily scheduling response data until multiple second configuration particle individuals meeting preset fitness defined configuration fitness are obtained; using a TOPSIS method to select a target device configuration particle individual from the multiple second configuration particle individuals, and using the target solver to solve target scheduling response data to obtain a corresponding optimization result.

Description

Technical Field

[0001] This application relates to the fields of smart logistics and energy management technology, particularly to the management and control methods and service platforms for multi-energy complementary systems in logistics parks. Background Technology

[0002] Green energy management measures in logistics parks are mostly focused on the configuration of photovoltaic energy storage systems and the greening of freight vehicles. Multi-energy complementary systems that allow for refined management of multiple energy forms have not yet been widely applied.

[0003] In related technologies, the low-carbon transformation of logistics parks requires increasing the application of clean energy and the integration of new energy logistics vehicles, while simultaneously planning multi-energy complementary systems to achieve efficient energy allocation and optimized scheduling. Furthermore, logistics parks experience high freight volume and traffic volume. With the increasing penetration rate of new energy vehicles, the large-scale integration of these vehicles will lead to a surge in electricity consumption, posing new challenges to the load management of the park and even the city's power grid. The charging and hydrogen refueling of new energy vehicles are characterized by randomness; without proper guidance and management, this could further widen the peak-valley difference in the power grid, affecting its safety and stability.

[0004] Therefore, how to effectively manage the load of new energy vehicles and achieve coordinated development of new energy vehicles, power grids, and multi-energy complementary systems has become an urgent management problem to be solved. Summary of the Invention

[0005] This application provides a management and control method and service platform for a multi-energy complementary system in a logistics park, which at least solves the problem of poor coordinated scheduling between new energy vehicles, the power grid, and the multi-energy complementary system in logistics parks in related technologies.

[0006] In a first aspect, embodiments of this application provide a management and control method for a multi-energy complementary system in a logistics park, comprising: using a particle swarm optimization algorithm to perform a population search iteration on multiple equipment configuration particles to generate multiple first configuration particles corresponding to the current iteration, wherein the equipment configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of freight and warehousing services in the target logistics park within the target planning period, each particle of the equipment configuration particles is associated with a type of energy supply equipment, and the particle position of the particle is used to characterize the decision of the capacity configuration of a type of energy supply equipment; using a target solver, performing a daily operation scheduling plan solution based on mixed integer linear programming on all the particles of each first configuration particle to generate daily scheduling response data corresponding to each first configuration particle, and based on the scheduling fitness corresponding to the daily scheduling response data and the associated energy supply equipment of the first configuration particle. The objective function is used to calculate the configuration fitness of the corresponding first configuration particle individual, wherein the daily scheduling response data is used to characterize the power supply conditions for various power supply equipment configurations; based on the configuration fitness, the steps of generating multiple corresponding first configuration particle individuals using the particle swarm optimization algorithm and generating corresponding daily scheduling response data based on the objective solver are repeatedly executed until the corresponding configuration fitness meets the preset fitness limit, resulting in multiple second configuration particle individuals; using the TOPSIS method, a target equipment configuration particle individual is selected from the multiple second configuration particle individuals, and the daily scheduling response data corresponding to the target equipment configuration particle individual is solved daily within the target planning period using the objective solver, resulting in target scheduling response data, and the particle position of each particle of the target equipment configuration particle individual and all the target scheduling response data are used as the optimization result.

[0007] Secondly, embodiments of this application provide a service platform, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the logistics park multi-energy complementary system management and control method as described in the first aspect.

[0008] Compared to related technologies, the logistics park multi-energy complementary system management method and service platform provided in this application utilizes a particle swarm optimization algorithm to perform a population search iteration on multiple equipment configuration particles, generating multiple first configuration particles corresponding to the current iteration. These equipment configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of freight and warehousing services in the target logistics park within the target planning period. Each particle in the equipment configuration particle is associated with a type of energy supply equipment, and the particle position is used to characterize the decision on the capacity configuration of that energy supply equipment. Using a target solver, a daily operation scheduling plan based on mixed-integer linear programming is solved for all particles of each first configuration particle, generating daily scheduling response data corresponding to each first configuration particle. Based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle, the configuration fitness corresponding to the first configuration particle is calculated. The daily scheduling response data is used to characterize the multi-energy complementary system management method and service platform. The energy supply equipment is configured with the following operating conditions: Based on the configuration fitness, the steps of generating multiple first configuration particles using the particle swarm optimization algorithm and generating corresponding daily scheduling response data based on the target solver are repeatedly executed until the corresponding configuration fitness meets the preset fitness limit, resulting in multiple second configuration particles; Using the TOPSIS method, a target equipment configuration particle is selected from the multiple second configuration particles, and the daily scheduling response data corresponding to the target equipment configuration particle is solved using the target solver within the target planning period, resulting in target scheduling response data. The particle position of each particle of the target equipment configuration particle and all target scheduling response data are used as optimization results. This solves the problem of poor coordinated scheduling effect between new energy vehicles, power grid, and multi-energy complementary systems in logistics parks in related technologies, realizes supply and demand matching, improves the economy, low carbon emissions, and sustainability of multi-energy complementary system operation, and provides an important decision-making basis for the operation and management of multi-energy complementary systems in logistics parks.

[0009] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0010] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a hardware structure block diagram of the terminal of the logistics park multi-energy complementary system management and control method according to an embodiment of this application; Figure 2 This is a flowchart of a multi-energy complementary system management and control method for logistics parks according to an embodiment of this application; Figure 3 This is a schematic diagram of load scheduling for electric hydrogen logistics vehicles according to an embodiment of this application; Figure 4 This is a structural block diagram of a logistics park multi-energy complementary system control device according to an embodiment of this application. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated 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 application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application. Furthermore, it is understood that although the efforts made in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, modifications to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0012] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0013] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "a," "an," "an," "the," and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms "comprising," "including," "having," and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. "Multiple stages" used in this application refers to two or more stages. "And / or" describes the relationship between related objects, indicating that three relationships may exist; for example, "A and / or B" can represent: A alone, A and B simultaneously, and B alone. The terms "first," "second," "third," etc., used in this application are merely to distinguish similar objects and do not represent a specific ordering of objects.

[0014] Before providing a detailed description of the embodiments of this application, the relevant technologies involved in this application are described below:

[0015] Particle Swarm Optimization Algorithm

[0016] The algorithm initially generates a random swarm of particles, forming a random initial scheme. After multiple iterations, it finds the optimal solution. In each iteration, the next move of each particle in the swarm is determined by two main factors: the individual particle's search for a locally optimal solution and the search for the global optimal solution for the entire swarm. This updates the particle's search direction and speed accordingly. Specifically, assuming a D-dimensional optimization target search space, the swarm consists of N particles, where the i-th particle has D-dimensional attributes, represented as a D-dimensional vector, denoted as x. i =(x i1 x i2 x i3 , ..., x iD ), i = 1, 2, 3, ..., N. Similarly, the flight velocity of the i-th particle can also be regarded as a D-dimensional vector, denoted as: v i =(v i1 v i2 v i3 , ..., v iD Let i = 1, 2, 3, ..., N. Then, the optimal position (individual extreme value) currently found by the i-th particle is denoted as p. best =(p i1 p i2 p i3, ..., p iD Let g be the optimal position (population extreme value) currently found by the entire particle swarm, where i = 1, 2, 3, ..., N. best =(g i1 g i2 g i3 , ..., g iD Let i = 1, 2, 3, ..., N. When the i-th particle evolves to the (t+1)-th generation through these two extreme values ​​after the t-th iteration, the "flight" speed of the i-th particle is updated according to the formula: v ij (t+1)=ωv ij (t)+c1r1[p ij (t)-x ij (t)]+c2r2[p ij (t)-x ij [(t)], where c1 represents the self-learning factor and c2 represents the group learning factor; r1 and r2 are two random functions with values ​​in the range [0, 1] to increase the randomness of the search; ω represents the inertia weight, a non-negative number used to adjust the search range of the solution space, ωv ij (t) represents the inertia of the particle's flight velocity, indicating the tendency of the particle to maintain its previous state of motion; c1r1[p ij (t)-x ij [(t)] represents the self-awareness part, indicating the particle's tendency to move towards its historical best position; c2r2[p ij (t)-x ij [(t)] represents the "social" part, indicating the trend of particles moving towards their historical best position within the particle swarm; finally, after the t-th iteration, the latest self-position of the i-th particle is updated according to the formula, denoted as: x ij (t+1)=x ij (t)+v ij (t+1).

[0017] Multi-objective decision making

[0018] Multi-objective decision-making problems arise from the diversity of human needs and the multi-criteria nature of socio-economic activities undertaken to satisfy these needs. A single problem may require consideration of multiple evaluation criteria, such as cost and quality, and the establishment of work objectives, with the corresponding goals being minimum cost and maximum quality. Simply put, it involves solving for two or more objectives under the same constraints. The "ideal state" of solving multi-objective decision-making problems refers to satisfying all objective needs under objective conditions. However, in reality, this "ideal state" is difficult to achieve because the various objective needs are generally contradictory. Optimizing one objective may lead to a decrease in the value of another. It is precisely because of the conflict and contradiction among multiple objectives that people are motivated to seriously study scientific decision-making theories and methods.

[0019] The general form of a multi-objective programming problem is: , where f i (x), i=1,…,p is the objective function, g j Let (x)≥0, j=1,…,m be constraints, and x be the decision variable. Let R={x|g j (x)≥0,j=1,…,m,x∈E n Let R be the set of feasible solutions (decision space) of problem (VP), and F(R)={f(x)|x∈R} be the image set (objective space) of problem (VP).

[0020] Two-layer planning

[0021] Bilevel programming is a two-level decision-making mathematical model for system optimization problems based on a two-level hierarchical optimization structure. The upper-level programming problem and the lower-level programming problem have their own different objective functions, constraints, and decision variables. They are independent of each other but also influence each other. The two levels of programming make decisions in a prescribed order. The upper-level programming problem is the core and will give the decision solution first. The lower-level programming problem makes its own decision based on the decision solution given by the upper-level programming problem. However, the decision solution of the lower-level programming problem will be passed up and affect the decision of the upper-level programming problem.

[0022] The mathematical model of the bilevel programming model is expressed as: (L)min x F(x, y(x)) and stG(x, y(x)) ≤ 0, where F(x, y(x)) are the objective functions of the upper-level programming problem, and stG() are the constraints of the upper-level programming problem; the optimal solution y(x) of the lower-level programming problem is: (F)min y f(x, y(x)) and st g(x, y(x)) ≤ 0, where f(x, y(x)) is the lower-level objective function and st g() is the lower-level constraint condition.

[0023] In this embodiment, a two-level programming model is used to make energy allocation decisions for the logistics park. The upper level aims to minimize the total annual cost, the annual carbon emissions, and the renewable energy penetration rate. The upper-level programming problem is solved using the particle swarm optimization algorithm. The lower level is based on the mixed integer linear programming method and uses a commercial solver to make operation scheduling planning decisions, so as to achieve coordinated optimization of warehousing and freight operations within the logistics park.

[0024] The Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS), also known as the approximation-to-ideal-solution ranking method, is a multi-attribute decision-making method based on distance. It ranks and selects solutions by constructing positive and negative ideal solutions and calculating the distance between each solution and these two solutions. A positive ideal solution represents the optimal combination of all attributes, while a negative ideal solution represents the worst combination of all attributes. The basic steps of the TOPSIS method are as follows:

[0025] Step 1: Data standardization. Transform the raw data into dimensionless values, typically to the range of 0 to 1, to ensure fair comparison of different indicators. For each indicator, X = {x1, x2, ..., x...} n Where, x i The index value r of the i-th sample is the standardized value. i It can be calculated using the following formula: This allows all indicator values ​​to be converted to a range between 0 and 1.

[0026] Step 2: Calculate the distance between each solution and the ideal and negative ideal solutions using Euclidean distance or other distance metrics. This step quantifies the difference between each solution and the ideal and worst-case states. For each sample j, calculate its distance from the positive ideal solution S. + and negative ideal solution S - Euclidean distance and ,in, and These are the maximum value (corresponding to the positive ideal solution) and minimum value (corresponding to the negative ideal solution) of the k-th index, respectively.

[0027] Step 3: Calculate the relative closeness (the relative proximity of each index to the positive ideal solution). For each index k, calculate the relative closeness (or goodness coefficient) of index i relative to the positive ideal solution based on the Euclidean distance. It is usually defined as: , where d iNIS Denotes the negative ideal solution, d iPIS This represents the ideal solution. The closer to 0, the closer index i is to the positive ideal solution, meaning that the performance across all evaluation indicators is more ideal.

[0028] Step 4: Sort the options according to their relative proximity and select the one with the highest proximity as the best choice.

[0029] The method embodiments provided in this example can be executed on a terminal, computer, or similar computing device. Taking running on a terminal as an example, Figure 1 This is a hardware structure block diagram of the terminal of the logistics park multi-energy complementary system management method according to an embodiment of this application. For example... Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. Optionally, the terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0030] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the logistics park multi-energy complementary system management method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0031] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0032] This embodiment provides a management and control method for a multi-energy complementary system in a logistics park operating on the aforementioned terminal. Figure 2 This is a flowchart of a multi-energy complementary system management and control method for logistics parks according to an embodiment of this application, such as... Figure 2 As shown, the process includes the following steps:

[0033] Step S201: Using the particle swarm optimization algorithm, a population search iteration is performed on multiple device configuration particles to generate multiple first configuration particles corresponding to the current iteration. The device configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of the target logistics park for freight and warehousing services within the target planning period. Each particle of the device configuration particles is associated with a type of energy supply device, and the particle position is used to characterize the decision on the capacity configuration of a type of energy supply device.

[0034] In this embodiment, the executing entity of the method of this application embodiment includes, but is not limited to, the terminal at the highest level of the multi-level system (relative to the upper and lower decision-making layers). The decision-making layers involved in this application embodiment include the upper decision-making layer for equipment capacity planning and the lower decision-making layer for operation scheduling optimization. The energy supply equipment to be configured in this application embodiment is the energy supply equipment to be deployed, constructed, or purchased in the target logistics park (e.g., a zero-carbon logistics park). The corresponding energy supply equipment provides four types of energy: electricity, cooling, heating, and hydrogen. In this embodiment, the energy supply equipment of the four types of energy constitutes an electricity supply subsystem, a heating subsystem, a cooling subsystem, and a hydrogen supply subsystem.

[0035] In this embodiment, the power supply equipment of the power supply system includes photovoltaic equipment, gas turbine, hydrogen fuel cell and energy storage battery. The photovoltaic equipment, gas turbine and hydrogen fuel cell generate electricity using solar energy, natural gas and hydrogen energy respectively, and the energy storage battery consumes excess electrical energy.

[0036] In this embodiment, the mathematical model for the output of the photovoltaic equipment is as follows: , where P PV，t P represents the output of the photovoltaic power generation system at time t. pv f represents the rated power of a photovoltaic module under standard conditions.PV R is the performance degradation parameter; k is the power temperature coefficient of the photovoltaic module, and R s T represents the intensity of solar radiation received by a photovoltaic module under standard conditions. s R represents ambient temperature. a (t) represents the actual solar radiation intensity received by the photovoltaic module at a specific moment, T a (t) represents the surface temperature of the photovoltaic power generation system in actual operation, T a (t) = T ae (t) + 30R a (t) / 1000, T ae (t) represents the actual ambient temperature of the photovoltaic system at that moment.

[0037] In this embodiment, the mathematical model for the gas turbine output is: ; Among them, P GT，t Let Q represent the power generation capacity of the gas turbine at time t, where ẟ is the minimum combustion value of natural gas; gas，t and Q h β1 and ℇ represent the amount of natural gas consumed and waste heat generated by the gas turbine, respectively, and their power generation efficiency and heat loss coefficients.

[0038] In this embodiment, the mathematical model for the output power of the hydrogen fuel cell is: ,in, This represents the output power of the hydrogen fuel cell at time t; This represents the input hydrogen power of the hydrogen fuel cell at time t; The energy conversion efficiency of hydrogen fuel cells; This refers to the electrical energy that can be released per kilogram of hydrogen gas.

[0039] In this embodiment, the mathematical model for the output power of the energy storage battery is: ; in, and These represent the charging and discharging power of the energy storage battery at time t, respectively. and The charging and discharging efficiency of energy storage batteries; and These represent the energy storage capacity of the battery at times t+1 and t, respectively.

[0040] In this embodiment, the energy supply equipment of the heating subsystem includes an electric boiler and a waste heat boiler. The electric boiler is an electrothermal coupling device that provides heat energy by consuming electrical energy, and the waste heat boiler uses high-temperature flue gas generated by a gas turbine for heating.

[0041] In this embodiment, the mathematical model for the output of the electric boiler is: ,in, and Let η be the input power and output power of the electric boiler at time t. EB This refers to the conversion efficiency of the electric boiler.

[0042] In this embodiment, the mathematical model for the output of the waste heat boiler is as follows: ,in, η represents the heat output of the waste heat boiler. WHB Q represents the heat production efficiency of the waste heat boiler. h,t This refers to the waste heat input power of the waste heat boiler.

[0043] In this embodiment, the energy supply equipment of the cooling subsystem includes an electric chiller and an absorption chiller. The electric chiller consumes electrical energy to meet the cooling load supply demand, while the absorption chiller can absorb the waste heat of the flue gas generated by the gas turbine for cooling, thus meeting the cooling load demand.

[0044] In this embodiment, the mathematical model for the output of the electric chiller is: ,in, and These represent the cooling power and power consumption of the electric chiller at time t; COP EC This is the coefficient of performance (COP) of the electric refrigeration unit.

[0045] In this embodiment, the mathematical model for the output of the absorption chiller is: ,in, η is the cooling power output of the absorption chiller at time t. AC The refrigeration efficiency of an absorption chiller. Let t be the thermal power input to the absorption chiller at time t.

[0046] In this embodiment, the energy supply equipment of the hydrogen supply subsystem includes an electrolyzer and a hydrogen storage tank. The electrolyzer produces hydrogen using electricity, and the hydrogen storage tank is used for hydrogen storage. On the one hand, it provides hydrogen refueling for hydrogen-powered heavy trucks, and on the other hand, it is used for multi-energy complementarity to provide power generation for fuel cells.

[0047] In this embodiment, the mathematical model for the output power of the electrolytic cell is: ,in, Let t be the hydrogen production power of the electrolyzer at time t. The hydrogen production efficiency of the electrolyzer. Let be the power consumption of the electrolytic cell at time t. This refers to the mass of hydrogen produced per kilowatt of electrical energy.

[0048] In this embodiment, the mathematical model for the output power of the hydrogen storage tank is: ; Among them, Phst,t Let be the output power of the hydrogen storage tank at time t. Let be the hydrogen charging power of the hydrogen storage tank at time t. Let be the hydrogen release power of the hydrogen storage tank at time t. The hydrogen filling efficiency of the hydrogen storage tank. S represents the hydrogen release efficiency of the hydrogen storage tank. hst,t+1 and S hst,t These represent the capacities of the hydrogen storage tank at times t+1 and t, respectively.

[0049] In this embodiment, configuration and optimization refer to the selection of energy supply equipment types and capacity configuration for the target logistics park during the target planning period (e.g., year), and based on the configured energy supply equipment and capacity, the scheduling optimization of which energy supply equipment will provide energy and the energy supply time periods. For example, to meet the energy demand of the target logistics park within one year, the following equipment is installed in the target logistics park: photovoltaic equipment a with a capacity of A, gas turbine b with a capacity of B, hydrogen fuel cell c with a capacity of C, energy storage battery d with a capacity of D, electric boiler e with a capacity of E, waste heat boiler f with a capacity of F, electric chiller g with a capacity of G, absorption chiller h with a capacity of H, electrolyzer j with a capacity of J, and hydrogen storage tank k with a capacity of K. The capacity is the upper limit of the energy supply power provided by the corresponding energy supply equipment. Then, under the corresponding capacity configuration, the daily operation scheduling plan of the target logistics park is carried out. For example, photovoltaic equipment a operates all day during the day. The system provides electrical power not exceeding the power corresponding to capacity A, and gas turbine b operates at a power not exceeding the power corresponding to capacity B (B-2) within the nm time period. The configuration in this embodiment is not a selection of equipment, but a scheme selection based on the known energy supply equipment that can participate in the energy supply, and by considering the three aspects of economic benefits, environmental effects, and energy benefits, the system determines the optimal comprehensive benefits (e.g., minimizing annual comprehensive cost, minimizing annual carbon emissions, and maximizing energy penetration rate) based on the daily operation scheduling optimization of the corresponding configuration scheme after selecting the capacity of the energy supply equipment. In this embodiment, the scheduling optimization is to determine the optimal minimum daily operating cost after the daily operation scheduling optimization decision of the energy supply equipment. The minimum daily operating cost is used to participate in the calculation of the minimum annual comprehensive cost and is also the coupling variable between the upper-level decision and the lower-level decision involved in this embodiment.

[0050] In this embodiment, a particle swarm optimization (PSO) algorithm is used to make energy supply capacity configuration decisions. It is understood that the population search optimization step performed by the PSO algorithm in this embodiment is a known prior art. In this embodiment, when using the PSO algorithm to make energy supply capacity configuration decisions, a randomly generated initial solution (corresponding to a single device configuration particle) is a random allocation of the energy supply capacity of the energy supply devices. The number of particles in a single device configuration particle corresponds to the type of energy supply device, and the particle position corresponds to the configured capacity. For example, a single device configuration particle might be {a}. i A, b i :B, c i :C,d i :D, e i :E, f i :F, g i :G,h i :H,j i :J,k i :K},a i For photovoltaic equipment, the corresponding configuration capacity is A, where A represents the corresponding particle position. In this embodiment, one allocation corresponds to one decision variable, which maps or associates with multiple corresponding decision items. In this embodiment, the decision item is used to characterize one of the following: minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the renewable energy penetration rate. After randomly generating the initial solution, the particles of each equipment configuration particle individual fly based on the current optimal position and particle velocity, which is to perform population search iteration, update the current optimal position (individual extreme value) of the particles of the equipment configuration particle individual and the optimal positions (population extreme value) of multiple equipment configuration particle individuals, and obtain the current corresponding equipment configuration particle population, which is the multiple first delivery particle individuals corresponding to the current iteration.

[0051] Step S202: Using the objective solver, perform daily operation scheduling planning based on mixed integer linear programming for all particles of each first configuration particle individual to generate daily scheduling response data corresponding to each first configuration particle individual. Based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle individual, calculate the configuration fitness corresponding to multiple first configuration particle individuals. The daily scheduling response data is used to characterize the energy supply conditions configured for multiple energy supply devices.

[0052] In this embodiment, the main objective is to optimize the operation status and energy flow of various energy supply equipment in the target logistics park on a typical day, based on the determined upper-level capacity configuration results, thereby obtaining the optimal operation plan of the system. Based on the typical day, the annual planning cycle is divided into three seasonal planning cycles, with one typical day corresponding to one seasonal planning cycle. Specifically, the annual planning cycle is divided into three sub-planning cycles: summer, winter, and transitional season. During the optimization scheduling, the typical day's operation is divided into 24 scheduling periods using an hourly time scale. Within each period, the output of various energy supply equipment is optimized based on conditions such as energy supply and demand balance, equipment operation constraints, and the refueling needs of hydrogen-electric logistics vehicles. This ultimately yields the optimal operation strategy for each energy supply equipment on a typical day and the daily operating cost for each operating day, thus obtaining the total operating cost. It can be understood that the optimized scheduling operation can be abstracted as a Mixed Integer Linear Programming (MILP) problem. MILP is a type of optimization problem that introduces integer variables into traditional linear programming. Its mathematical model can typically be expressed as: min c T x Where x is the decision variable vector, containing both continuous and integer variables; c is the objective function coefficient vector; A and A eq These are the coefficient matrices for the inequality constraints and equality constraints, respectively; b and b eq These are the constant vectors corresponding to the constraints; x min and x max These are the upper and lower bounds of the variable, respectively; x j Representing 0-1 variables, these are typically used to describe discrete decision variables such as the start-up and shutdown states of energy supply equipment. By establishing a mathematical model of this form, the operational optimization problem can be transformed into a standard MILP problem for solution. In this embodiment, to efficiently solve this mixed-integer linear programming problem, the mathematical programming solver CPLEX is used to solve the model. In this embodiment, daily operational scheduling optimization is the decision for lower-level operational optimization. First, a corresponding mathematical model is established based on the operational characteristics of the multi-energy complementary system of the target logistics park, clarifying the system's decision variables, objective function, and various constraints in the lower-level model. Then, the model is constructed and the CPLEX solver is called to solve it. By inputting the model parameters into the solver and performing optimization calculations, the daily scheduling response data is finally output, which is the operational status (whether it is started), energy allocation (power supply), and objective function value (corresponding to scheduling fitness) of the energy supply equipment in the target logistics park during each energy supply period, thereby obtaining the optimal solution to the lower-level operational optimization problem.

[0053] In this embodiment, after obtaining the optimal solution for lower-level operation optimization, the daily scheduling response data obtained from the operation scheduling optimization is fed back to the upper-level decision-making layer. This allows the upper-level decision-making layer to calculate the function value of the objective function corresponding to the first configuration particle individual based on the energy allocation and scheduling fitness in the daily scheduling response data, thereby evaluating the quality of the first configuration particle individual generated in the current iteration. In this embodiment, the objective function corresponding to the first configuration particle individual is constructed from three parameters: minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the energy penetration rate. Upon receiving the feedback from the lower layer regarding the minimized daily operating cost, the corresponding annual operating cost within the target logistics park can be calculated. Combined with the corresponding equipment investment cost, the minimized annual comprehensive cost can be determined, and the minimized annual carbon emissions... The function value corresponding to the quantity can be calculated based on the annual carbon emissions from the corresponding purchased energy consumption in the energy allocation. Maximizing the energy penetration rate is calculated based on the proportion of the energy provided by the photovoltaic equipment (that is, the electrical energy produced by the photovoltaic equipment started in the operation scheduling) to the total load energy (corresponding to the total energy supply load) output by all energy supply equipment started in the operation scheduling for electricity, cooling, heating and hydrogen supply. It should be noted that in this embodiment, because the energy output by some energy supply equipment will be used for other energy supply equipment, for example, part of the electrical energy provided by photovoltaic is used for electric cooling, and this part of electrical energy is converted into cold energy, when calculating the total load energy, the electrical energy provided by photovoltaic equipment needs to be deducted from the electrical energy used for cooling, and the cold energy provided by the cooling equipment needs to be included in the total load capacity.

[0054] Step S203: Based on the configuration fitness, repeatedly execute the steps of generating multiple first configuration particles using the particle swarm optimization algorithm and generating corresponding daily scheduling response data based on the objective solver, until the corresponding configuration fitness meets the preset fitness limit, and obtain multiple second configuration particles.

[0055] In this embodiment, after the first configuration particle completes one population search iteration, the particle positions corresponding to each particle in all the first configuration particles are obtained. These particle positions are used as the result of the upper-level decision after one iteration, and this result is substituted into the objective function corresponding to the first configuration particle that has completed one iteration, i.e., the configuration fitness of the first configuration particle that has completed one iteration is calculated. After calculating the configuration fitness of the first configuration particle after one iteration, it is compared with the configuration fitness after the previous iteration. If it is better, then all the first configuration particles and their corresponding particle positions after this iteration are used as the current target configuration particle. The appropriate particle swarm is configured, and then the particle swarm optimization algorithm is used to perform a swarm search iteration on the first configured particle. At the same time, after completing one swarm search iteration for the first configured particle, the upper-level capacity configuration result corresponding to the first configured particle is sent to the lower-level decision layer. Based on the determined upper-level capacity configuration result, the operating status and energy flow of various energy supply equipment in the target logistics park during a typical day are optimized and scheduled. That is, the corresponding daily scheduling response data is generated based on the target solver. In this way, the collaborative decision-making of upper-level capacity configuration and lower-level motion scheduling optimization is repeated multiple times to obtain multiple second configured particle individuals that meet the preset requirements.

[0056] Step S204: Using the TOPSIS method, select the target device configuration particle from multiple second configuration particle individuals. Using the target solver, solve for the daily scheduling response data of the target device configuration particle individual within the target planning period, obtain the target scheduling response data, and use the particle position of each particle of the target device configuration particle individual and all target scheduling response data as the optimization result.

[0057] In this embodiment, multiple second configuration particles that meet preset requirements are obtained, which is a set of optimal upper-layer capacity configuration solutions. However, since each second configuration particle is associated with three dimensions of objectives—minimizing annual comprehensive cost, minimizing annual carbon emissions, and maximizing renewable energy penetration rate—the TOPSIS method is used to comprehensively evaluate each optimal upper-layer capacity configuration solution. The optimal second configuration particle, which is the target equipment configuration particle, is selected from the multiple second configuration particles. Then, based on the upper-layer capacity configuration result (the particle position of each particle in the target equipment configuration particle) corresponding to the target equipment configuration particle, the target solver is used to optimize the scheduling of the operating status and energy flow of various energy supply equipment in the target logistics park during a typical day, and obtain the target scheduling response data.

[0058] Through steps S201 to S204, a particle swarm optimization algorithm is used to perform a population search iteration on multiple equipment configuration particles, generating multiple first configuration particles corresponding to the current iteration. These equipment configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of the target logistics park's freight and warehousing services within the target planning period. Each particle in the equipment configuration particle is associated with a type of energy supply equipment, and the particle position represents the decision on the capacity configuration of that energy supply equipment. Using a target solver, a daily operation scheduling plan based on mixed-integer linear programming is solved for all particles of each first configuration particle, generating daily scheduling response data corresponding to each first configuration particle. Based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle, the configuration fitness corresponding to the first configuration particle is calculated. The daily scheduling response data represents the energy supply capacity configuration for multiple energy supply equipment. The process involves repeatedly executing the steps of generating multiple first configuration particles using a particle swarm optimization algorithm and generating corresponding daily scheduling response data based on a target solver, based on configuration fitness, until the corresponding configuration fitness meets a preset fitness limit, resulting in multiple second configuration particles. Using the TOPSIS method, target equipment configuration particles are selected from these second configuration particles. The target solver is then used to solve for the daily scheduling response data of the target equipment configuration particles within the target planning period, obtaining target scheduling response data. The particle position of each particle in the target equipment configuration particles and all target scheduling response data are used as the optimization result. This approach addresses the problem of poor coordinated scheduling between new energy vehicles, the power grid, and multi-energy complementary systems in logistics parks, achieving supply-demand matching and improving the economy, low-carbon nature, and sustainability of multi-energy complementary system operation. This provides a valuable decision-making basis for the operation and management of multi-energy complementary systems in logistics parks.

[0059] It should be noted that the multi-energy complementary system configuration and optimization method in this application takes into account three objectives: economy, environment and energy. It uses the particle swarm optimization algorithm, which is widely used in multi-objective solutions, to obtain Pareto front solutions and a set of equipment capacity configuration schemes for the three objectives. The lower-level scheduling model is a mixed-integer linear programming problem, which is solved by calling the CPLEX commercial solver. Finally, the TOPSIS method based on entropy weight is used to evaluate the solution set, select a set of optimal solutions from the Pareto front solution set, and determine the final planning scheme.

[0060] In some embodiments, the target device configuration particle individual is selected from a plurality of second configuration particle individuals using the TOPSIS method, which is achieved through the following steps:

[0061] Step 21: Obtain multiple target decision parameters associated with the objective function corresponding to each second configuration particle individual, and preprocess the multiple target decision parameters to generate multiple standard decision parameters. Among them, the target decision parameters include minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the renewable energy penetration rate. The minimizing of the annual comprehensive cost is determined based on the scheduling fitness corresponding to the daily scheduling response data. The preprocessing includes homing processing and normalization processing.

[0062] In this embodiment, the goal of maximizing renewable energy penetration is oriented in the same direction, for example, by taking the negative value, so that the two minimization objectives remain unchanged, resulting in oriented parameters. Then, the oriented parameters are normalized to obtain a variety of standard decision parameters.

[0063] Step 22: Among all the standard parameter values ​​corresponding to each standard decision parameter, determine the positive ideal solution value with the largest parameter value and the negative ideal solution value with the smallest parameter value, and determine the Euclidean distance between each standard parameter value and the positive ideal solution value and the negative ideal solution value for each standard decision parameter, so as to obtain the first Euclidean distance and the second Euclidean distance corresponding to each standard parameter value.

[0064] Step 23: After determining the relative proximity of the corresponding standard decision parameter based on the ratio of the second Euclidean distance to the total distance, the weight of each standard decision parameter is determined based on the entropy weight method. Then, the relative proximity of the various standard decision parameters is weighted according to the weights to obtain the total proximity of each second configuration particle. The total distance is used to represent the sum of the first Euclidean distance and the second Euclidean distance, and the relative proximity is used to represent the relative importance of the corresponding standard decision parameter among all standard decision parameters.

[0065] In this embodiment, when determining the corresponding weights, the entropy value of each standard decision parameter is first calculated, and the corresponding entropy value is calculated according to the following formula: , Where i is the index of the second configured particle, and k is the index of the type of standard decision parameters. The standard parameter value of the k-th standard decision parameter in the i-th second-configuration particle individual; P ik As the middle, E k Let P be the information entropy corresponding to the k-th standard decision parameter. ik It is the median value, and P ik When =0, P ik lnP ik =0; then, calculate the weight of each standard decision parameter according to the following formula: , λ k Let be the weight of the k-th standard decision parameter, and m be the maximum value of the types of standard decision parameters.

[0066] Step 24: Select the second configuration particle with the highest total proximity from the multiple second configuration particle individuals sorted from low to high according to total proximity, and obtain the target device configuration particle individual.

[0067] Through steps 21 to 24 above, the TOPSIS method is used to evaluate the various target decision parameters of the second configuration particle individual, which can objectively and realistically reflect the advantages and disadvantages of each candidate scheme, thereby determining the final optimization result and improving the accuracy of upper-level capacity configuration decision.

[0068] In some embodiments, the following steps are performed before encoding and generating multiple device configuration particle individuals:

[0069] Step 31: Obtain the building energy load demand of the target logistics park in multiple seasonal sub-cycles of the target planning period. The energy supply corresponding to the building energy load demand includes the first power supply demand, heating demand, and cooling demand.

[0070] In this embodiment, when configuring a multi-energy complementary system, capacity configuration decisions need to be made based on the energy load demand within the target logistics park. Furthermore, this embodiment also categorizes the upper-level capacity configuration results and the operating status and energy flow scheduling of various energy supply equipment within the target logistics park on a typical day into three scenarios, matching the capacity configuration and operation scheduling results corresponding to summer, winter, and transitional seasons. Therefore, when determining the corresponding load, energy load demand is also confirmed based on a typical day. It is understood that the building load demand in this embodiment includes the energy load demand of office buildings and the energy load demand of warehouse buildings, and the corresponding energy load demand includes the first power supply demand, heating demand, and cooling demand.

[0071] Step 32: Using the Monte Carlo simulation method, generate the freight load demand in each seasonal sub-cycle, and determine the corresponding hydrogen supply demand and second power supply demand.

[0072] In some alternative implementations, the freight load demand for each seasonal sub-cycle is generated using Monte Carlo simulation methods, through the following steps:

[0073] Step 32-1: Obtain the target information and operating parameters of the target vehicles participating in the freight of the target logistics park. The target information is used to characterize the type and number of target vehicles. The operating parameters include the entry time, exit time and initial energy state parameters of the target vehicles when entering the park. The initial energy state parameters include the state of charge and the hydrogen storage state. Each operating parameter is associated with a corresponding probability distribution.

[0074] Step 32-2: Based on the corresponding probability distribution, use Monte Carlo simulation to randomly sample the operating parameters of all target vehicles, and calculate the refueling time and hourly load of the target vehicles in the set operating scenario based on the sampled operating parameters.

[0075] Step 32-3: Based on the refueling duration and hourly load of the target vehicles, the hourly load of all target vehicles is superimposed in time sequence to obtain the total hydrogen refueling load and total charging load in each seasonal sub-cycle. The hydrogen supply demand includes the total hydrogen refueling load, and the second power supply demand includes the total charging load.

[0076] In some preferred embodiments, in order to match the actual freight load demand, this embodiment establishes a reasonable new energy logistics vehicle load model. In this embodiment, a probability distribution model is adopted to characterize the travel behavior and other related characteristics of the target vehicles (including light electric logistics vehicles, electric heavy trucks and hydrogen-powered heavy trucks) in the logistics park, and key parameters such as the arrival time of the target vehicle in the target logistics park and the next travel time are generated by Monte Carlo sampling, so as to calculate the charging and hydrogen refueling load of the target vehicle.

[0077] In this embodiment, by setting sampling judgment logic, the operation is divided into two modes: daytime short-term stay and nighttime stay, corresponding to the vehicle's next trip of the day and the next trip of the following day, respectively. For electric heavy trucks and hydrogen-powered heavy trucks used for transportation, it is assumed that they will make their next trip the following day after completing their tasks for the day. The load influencing factors of new energy logistics vehicles include arrival time at the park, next trip time, length of stay in the park, state of charge (SOC) and state of hydrogen (SOH) upon arrival at the park, target SOC and SOH, and charging and hydrogen refueling power.

[0078] (1) Arrival time

[0079] Based on the big data of new energy electric truck travel in location S, a Gaussian mixed model (GMM) is used to represent the time characteristics of new energy logistics vehicles, such as the end time of the task and the departure time of the task.

[0080] Assuming the time a vehicle finishes its task is its arrival time, and its departure time is its departure time, let's consider the arrival time of electric logistics vehicles used for urban delivery. Arrival time of heavy trucks for trunk line transportation It follows a Gaussian mixture distribution as follows: , Where s represents the number of Gaussian components in the mixture model, and m i μ represents the weight of the i-th Gaussian component. i and σi Let represent the mean and variance of the i-th Gaussian component, respectively.

[0081] (2) Departure time For the travel time of electric logistics vehicles in urban delivery Travel times of heavy trucks for trunk line transportation They respectively follow the following Gaussian mixture distributions: , .

[0082] (3) Duration of stay Electric logistics vehicles used for urban delivery determine their dwell time based on sampled arrival times and next departure times. The definition is as follows: like If so, then the duration of their stay on their next trip that day will be considered. for: ; like If so, then it is assumed that the duration of their stay will be the next day or the next trip. for: ; For electric heavy-duty trucks and hydrogen-powered heavy-duty trucks, assuming they both travel the following day, their dwell time is... for: .

[0083] (4) Initial SOC and SOH When choosing between charging and hydrogen refueling, the acceptable thresholds for initial SOC and SOH of new energy logistics vehicles are approximately 30%. The initial SOC and SOH of the three types of new energy logistics vehicles all follow a normal distribution. , where μ y and σ y These are the corresponding mean and standard deviation, respectively.

[0084] (5) Target SOC and SOH In this embodiment, it is assumed that the target SOC and SOH of each new energy logistics vehicle are 100%. If the time required to complete the refueling is greater than the vehicle's dwell time, the target SOC and SOH are calculated as the product of the dwell time and the corresponding charging power or hydrogen refueling power.

[0085] (6) Charging method / hydrogen refueling power In this embodiment, light-duty electric logistics vehicles and electric heavy-duty trucks select their charging methods based on the vehicle's dwell time and the required charging time, prioritizing slow charging. If the required slow charging time is less than the dwell time, slow charging is selected, and the slow charging power is [not specified]. and If slow charging cannot meet the target power requirement, then fast charging should be selected, with a power of [missing information]. and If fast charging still does not meet the target charge level, continue charging using fast charging until the vehicle leaves the park. Hydrogen-powered heavy-duty trucks refuel relatively quickly, typically in 5-15 minutes, with a refueling power V... ht Constant.

[0086] In some alternative implementations, the following steps are taken to generate freight load demand for each seasonal sub-cycle using Monte Carlo simulation:

[0087] Step 1: Input the basic parameters and relevant probability distributions of the model, and initialize the number of each type of new energy logistics vehicle.

[0088] Step 2: Use Monte Carlo simulation to perform random sampling and select the end time of each car's mission. , and When leaving the park , and And the initial SOC / SOH value.

[0089] Step 3: Calculate the dwell time of each car based on its end time and departure time from the park. , and .

[0090] Step 4: After obtaining the relevant load influencing factors for each new energy logistics vehicle through Monte Carlo sampling, calculate the charging amount and hydrogen refueling amount.

[0091] The charging amount for light-duty electric logistics vehicles (LELV) and electric heavy-duty trucks (ET) is determined based on the vehicle's dwell time and target energy replenishment needs. The charging capacity of electric heavy trucks Represented as: ,in, and The initial SOC of the i-th LELV and ET when they are charging, respectively; and The target SOCs of the i-th LELV and ET are respectively; and These are the maximum energy storage capacities of LELV and ET, respectively.

[0092] Hydrogen refueling capacity Q of hydrogen-powered heavy-duty truck HT ht,i Represented as: ,in, The initial SOH for hydrogenation of the i-th HT vehicle; Let the target SOH of the i-th HT at the departure time be; This represents the maximum hydrogen storage capacity of HT.

[0093] Step 5: Calculate the charging time and hydrogen refueling time.

[0094] In this embodiment, LELV and ET select the charging method based on the dwell time, prioritizing slow charging. If the slow charging time exceeds the dwell time, fast charging is selected. If the fast charging time still exceeds the dwell time, the charging time is the dwell time, and the charging time t is the dwell time. lelv,i and t et,i They are respectively: ; ; The hydrogenation time for HT is: .

[0095] Step 6: After determining the charging power and charging time of each vehicle, add them up one by one to obtain the total load of the system: , , , where P lelvload，t， P etload，t P htload，t P represents the total electrical load or total hydrogen load of LELV, ET, and HT at time t. lelv，i，t， P et，i，t and P ht，i，t N represents the charging power or hydrogen refueling power of the i-th vehicle at time t. lelv N et and N ht These represent the quantities of LELV, ET, and HT, respectively.

[0096] Step 33: After merging the first power supply demand and the second power supply demand into the corresponding power supply demand, determine the capacity configuration range corresponding to the power supply equipment for each type of power supply demand according to the preset margin rule. The power supply demand includes the following energy demands: electrical energy, thermal energy, cold energy, and hydrogen energy. Each type of power supply demand is associated with at least one power supply equipment. The capacity configuration range is used to determine the particle position.

[0097] In some optional implementations, the capacity configuration range corresponding to the energy supply equipment for each type of energy demand is determined according to a preset margin rule, including the following steps:

[0098] Step 33-1: Determine the energy supply equipment to be configured for each type of energy demand within the target logistics park. The energy supply equipment includes power supply equipment, heating equipment, cooling equipment, and hydrogen supply equipment. The power supply equipment includes photovoltaic equipment and at least one first power supply equipment.

[0099] Step 33-2: Based on the building physical parameters and photovoltaic equipment parameters within the target logistics park, determine the corresponding assembly capacity of the photovoltaic equipment. The building physical parameters are used to characterize the roof area of ​​the building where the photovoltaic equipment is installed, and the equipment parameters include the area of ​​the photovoltaic modules and the power output of a single photovoltaic device.

[0100] In this embodiment, the upper bound of the search corresponding to the installed capacity of the photovoltaic equipment is determined by the physical constraint of the actual available roof area of ​​the target logistics park. For example, the total area of ​​the park is set to approximately 128,000 m². 2 The effective roof area available for photovoltaic installation accounts for approximately 60%, combined with the rated power of a single photovoltaic module (280W) and the module area (1.6m²). 2 The theoretical maximum installed capacity is approximately 13,440 kW. Based on this, the upper limit of the search for the installed capacity of photovoltaic equipment is set at 10,000 kW, which is within the physical constraints.

[0101] Step 33-3: Obtain the peak load of various energy demands in each seasonal sub-cycle, and determine the energy supply for each energy demand based on the peak load and the preset margin coefficient. The energy supply corresponding to the first power supply equipment is the amount of power supplied by the removal and assembly capacity.

[0102] In this embodiment, the upper limit of the search for the energy supplied by other power supply equipment is determined by multiplying the typical daily peak value of various types of loads by a margin coefficient. Specifically, based on the typical daily peak value of various types of loads such as electricity, cooling, heating, and hydrogen, and combined with the functional role of each equipment (main power supply equipment or auxiliary equipment), 1.0 to 1.5 times the peak load is taken as the upper limit of the search for the corresponding equipment capacity. It can be understood that the setting of this margin coefficient is intended to ensure that the algorithm optimizes within a sufficient search space, covers the structural adjustment requirements of load prediction deviation and multi-equipment collaborative configuration, and at the same time constrains the search space within an effective area that matches the actual load level, so as to ensure the convergence efficiency and solution quality of the multi-objective particle swarm optimization algorithm.

[0103] Step 33-4: Use the assembly capacity as the upper limit of the capacity configuration range corresponding to the photovoltaic equipment, and use the corresponding energy supply as the upper limit of the capacity configuration range corresponding to the first power supply equipment, heating equipment, cooling equipment and hydrogen supply equipment in sequence to obtain the corresponding capacity configuration range.

[0104] Through steps 31 to 33 above, the capacity configuration range corresponding to the energy supply equipment for each type of energy demand is determined, providing a data basis for the capacity configuration of upper-level equipment.

[0105] In some embodiments, a particle swarm optimization algorithm is used to generate multiple second-configuration particle individuals, which is achieved through the following steps:

[0106] Step 41: Determine the initial particle position and preset initial particle velocity of each device configuration particle individual, wherein the capacity configuration value corresponding to the initial particle position belongs to the capacity configuration range.

[0107] Step 42: Based on the preset dynamic adaptive change method, determine the dynamic adaptive inertia factor, the first acceleration coefficient and the second acceleration coefficient corresponding to the current particle search. The dynamic adaptive inertia factor, the first acceleration coefficient and the second acceleration coefficient are generated by adaptively changing the configuration fitness and the number of search iterations corresponding to each first configuration particle.

[0108] Step 43: Based on the initial particle position, initial particle velocity, dynamic adaptive inertia factor, first acceleration coefficient and second acceleration coefficient, the multi-objective particle swarm optimization algorithm is used to perform at least one population search iteration on the device configuration particle individuals until the configuration fitness corresponding to the generated candidate configuration particle individuals meets the preset fitness limit, and the same number of candidate configuration particle individuals as the device configuration particle individuals are obtained, wherein the second configuration particle individuals include the candidate configuration particle individuals.

[0109] In this embodiment, the preset fitness limit for configuring fitness can be set to the number of iterations reaching a set number, or the corresponding configured fitness being greater than a set fitness threshold. Regardless of the method used, it is clear and achievable for those skilled in the art. In this embodiment, the implementation method of the particle swarm optimization algorithm is understandable and known to those skilled in the art. However, this application adopts a processing method different from conventional particle swarm optimization algorithms, namely, using a dynamic adaptive inertia factor and a dynamically adjusted acceleration coefficient. The dynamic adaptive inertia factor and the dynamically adjusted acceleration coefficient are configured and adjusted during the search process so that they can adaptively change according to the fitness value of the chromosome and the number of iterations. Compared with a fixed inertia factor, the dynamic adaptive inertia factor can better balance global and local search capabilities. By dynamically adjusting the acceleration factor, the diversity of the particle swarm in the early search stage and the convergence of the algorithm in the later search stage are achieved.

[0110] Through steps 41 to 43 above, the search and iteration of the optimal configuration particle individual is realized, and a set of Pareto front solutions and equipment capacity configuration schemes that meet the preset requirements are obtained. Then, the TOPSIS method can be used to determine the equipment capacity configuration scheme corresponding to the target configuration particle individual, realize supply and demand matching, and improve the economy, low carbon and sustainability of the multi-energy complementary system operation.

[0111] In some embodiments, the configuration fitness corresponding to the first configuration particle is calculated based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle, including:

[0112] Step 51: Calculate the minimum daily total operating cost corresponding to the daily dispatch response data according to the sub-objective function corresponding to the daily dispatch response data. Based on the minimum daily total operating cost, determine the total operating cost in multiple seasonal sub-cycles. Also, determine the annual comprehensive cost corresponding to the first configuration particle individual based on the equipment cost and total operating cost of various energy supply devices. The dispatch fitness includes minimizing the daily total operating cost. The sub-objective function is constructed based on the energy supply device operating cost parameters, energy purchase cost parameters, and photovoltaic equipment power generation cost parameters generated under the daily energy supply conditions of the energy supply devices. The energy supply conditions include the type of energy supply device being dispatched, the dispatch operation time, and the energy supply power during operation.

[0113] In this embodiment, the minimum daily operating cost is calculated using the following formula: minF sys =f ode +f buy +f pv、 、 、 , where f ode This represents the operating cost of all power supply equipment; f buy Indicates energy purchase cost; f pv Indicates the power generation cost of photovoltaic equipment; C oj Qo represents the unit operation and maintenance cost of the j-th type of energy supply equipment; j P represents the operating power of the j-th type of power supply device at time t; buy_e，t Indicates the amount of electricity purchased externally; P buy_gas，t Indicates gas consumption; C e，t C is the purchase price of electricity. gas，t Gas purchase price refers to the price of natural gas, C pv Price of penalty for abandoning light per unit. P is the predicted maximum output power of the photovoltaic system at time t. pv，t Let be the output power of the photovoltaic system at time t.

[0114] Step 52: Determine the gas consumption required for the first power supply equipment based on gas power supply, and determine the purchased electricity in the electricity consumption required for the heating equipment, cooling equipment and hydrogen supply equipment based on electric power supply. Calculate the annual carbon emissions corresponding to the first configuration particle individual based on the gas consumption, purchased electricity and the corresponding unit carbon emission factor.

[0115] Step 53: Obtain the daily power output of the photovoltaic equipment and the daily power supply of all energy supply equipment from all daily dispatch response data, and determine the renewable energy penetration rate corresponding to the first configuration particle based on the ratio of the daily power output to the daily power supply. The daily power output is determined based on the dispatch operation time and the power supply during operation of the photovoltaic equipment, and the daily power supply is used to characterize the daily load demand in the target logistics park.

[0116] Step 54: Normalize and weight the annual comprehensive cost, annual carbon emissions and renewable energy penetration rate corresponding to the first configuration particle individual to generate the corresponding configuration fitness.

[0117] In this embodiment, the objective function set at the upper level includes minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the renewable energy penetration rate. Specifically,

[0118] (1) Minimize annual comprehensive cost

[0119] In this embodiment, the annual comprehensive cost consists of equipment purchase cost and annual operating cost, which can be expressed as: Where F1 is the annual comprehensive cost, f inv f represents the annual investment cost of the equipment. sys This refers to the daily operating cost of the system.

[0120] The annual investment cost is the initial investment cost of all equipment divided by year, which can be expressed as: Among them, C j and Q j Let be the unit capacity investment cost and capacity of energy supply equipment j, respectively; I be the investment discount rate, which is taken as 0.05 in this paper; and r be the equipment life cycle.

[0121] (2) Minimize annual carbon emissions In this embodiment, carbon emissions come from the purchase of natural gas from the natural gas grid and the purchase of electricity from the upstream power grid. The annual carbon emissions can be expressed as: Where F2 represents the total annual carbon emissions. and These represent the electricity and gas purchased at time t, respectively. and These are the unit carbon emission factors for electricity and natural gas purchases, respectively.

[0122] (3) Maximize the penetration rate of renewable energy In this embodiment, renewable energy penetration is the percentage of renewable energy output in the total system load, which can be expressed as: Where F3 is the renewable energy penetration rate, and P pv,t Let P be the operating power of the photovoltaic device at time t. e_load,t Ph_load,t P c_load,t and P hy_load,t These represent the electrical, thermal, cold, and hydrogen loads at time t, respectively.

[0123] In this embodiment, the constraint considered at the upper level is the investment capacity constraint of various types of equipment, which can be expressed as: 0≤Q j ≤Q j,max , where Q j,max Maximum installed capacity of each energy device.

[0124] Through steps 51 to 54 above, the configuration fitness is calculated.

[0125] In some embodiments, a target solver is used to solve a daily scheduling plan based on mixed-integer linear programming for all particles of each first configuration particle individual, generating daily scheduling response data corresponding to each first configuration particle individual, through the following steps: Step 61: Among all the power supply devices associated with the target logistics park, select the second power supply device and the first hydrogen supply unit for supplying power and hydrogen to the target vehicles involved in freight transportation, and determine the corresponding charging power supply range and hydrogen supply range based on the particle position of the particles in the first configuration particle individual that correspond to the second power supply device and the first hydrogen supply unit.

[0126] Step 62: After determining the dynamic scheduling energy supply period corresponding to the second power supply equipment and the first hydrogen supply unit under the preset scheduling mode, determine the building energy supply equipment and the first energy consumption interval determined according to the particle position of the particle corresponding to each building energy supply equipment in the first configuration particle individual. The scheduling mode includes one of the following: disordered scheduling, flexible scheduling, and the dynamic scheduling energy supply period is used to characterize the charging period or hydrogen refueling period allocated to the target vehicle under the corresponding scheduling mode.

[0127] It should be noted that the large-scale introduction of new energy logistics vehicles will bring new challenges to the planning and operation of multi-energy complementary systems. On the one hand, the concentrated charging of electric vehicles will become a new source of electricity load growth in the park; on the other hand, in parks that have already built their own hydrogen refueling stations, the electrolyzers used for hydrogen refueling of hydrogen fuel cell vehicles also rely on the power system for hydrogen production. With the continuous increase in the number of new energy logistics vehicles, the park may face "peak-on-peak" pressure during peak electricity consumption periods. Vehicles parked within the park are essentially flexible resources with scheduling potential; their charging and hydrogen refueling strategies can be adjusted based on objectives such as grid operation safety and the stability of the multi-energy complementary system. In this embodiment, referring to... Figure 3 Considering the flexibility of new energy vehicle load, this application considers two different scheduling scenarios to simulate the operation scheduling of the second power supply equipment and the first hydrogen supply unit under a preset scheduling mode, such as... Figure 3 As shown, for vehicles whose dwell time at the station exceeds the charging and refueling time, their load can be shifted from the original time period to other periods of low load or when renewable energy output is abundant by adjusting their charging and refueling times. This process achieves orderly charging and refueling of new energy vehicles without affecting their planned travel schedules. Specifically,

[0128] Disorder scheduling 1. In unordered scheduling scenarios, such as Figure 3 As shown in Figure a, new energy logistics vehicles adopt an on-demand charging method, with the start of charging and hydrogen refueling coinciding with their entry into the park. Once the vehicle arrives at the park, it begins charging or refueling, and the refueling process continues until completion, without unified coordination at the system level. In this scenario, the load of the new energy logistics vehicles participates as an exogenous variable in the configuration and optimization of the multi-energy complementary system.

[0129] 2. Flexible scheduling refer to Figure 3 As shown in b, in this mode, new energy logistics vehicles participate in the optimization of the park's multi-energy complementary system and are subject to the unified scheduling of the system, achieving coordinated matching between charging load and hydrogen refueling load and the park's energy demand. This scheduling method is based on the system's total load and operational objectives, comprehensively considering the vehicle's access time, vehicle load status, and expected departure time, dynamically adjusting the vehicle's charging and hydrogen refueling plans to achieve precise matching between the load of new energy vehicles and the overall energy demand within the park.

[0130] Under the flexible scheduling mode, the load of new energy logistics vehicles and the multi-energy complementary system can be optimized collaboratively. The load of new energy vehicles is no longer treated as a fixed load, but as an endogenous variable, participating in the overall scheduling optimization of the multi-energy complementary system, and obeying the unified scheduling of the integrated energy control center for charging or hydrogen refueling.

[0131] In the coordinated scheduling optimization of vehicle load and multi-energy complementary system, the mathematical model of the load is as follows: , , , , , Among them, P lelv,i,t P et,i,t and P ht,i,t Let x represent the electrical load of the i-th LELV and ET vehicles at time t, and the hydrogen load of the HT vehicle. lelv,i,t x et,i,t and x ht,i,t V represents the state variables of the i-th LELV, ET, and HT vehicles at time t, respectively. These are binary variables, indicating whether the vehicles are charging or refueling with hydrogen at that time. lelv,i,t Vet,i,t and V ht,i,t Let P be the charging power of the i-th LELV and ET vehicles at time t, and the hydrogen refueling power of the HT vehicle; lelvload,t P etload,t and P htload,t Let t represent the total electrical load of LELV, the total electrical load of ET, and the total hydrogen load of HT at time t.

[0132] Step 63: Using the energy supply conditions of the building energy supply equipment, the second power supply equipment, and the first hydrogen supply unit as lower-level decision variables, the objective solver is used to solve for the energy supply condition planning results that satisfy the objective constraints. The daily dispatch response data includes the energy supply condition planning results, which include at least one of the following: the first energy consumption of the building energy supply equipment in each corresponding dispatch period of the day, the charging power supply of the second power supply equipment in each charging period of the corresponding charging period, and the hydrogen supply of the first hydrogen supply unit in each hydrogen refueling period of the corresponding hydrogen refueling period.

[0133] In this embodiment, the energy supply condition as a lower-level decision variable refers to configuring the corresponding energy supply from the energy supply power corresponding to the corresponding energy supply equipment. For example, the first energy consumption selected by the building energy supply equipment from the first energy consumption range, the charging power supply selected by the second power supply equipment from the charging power supply range, and the hydrogen supply selected by the first hydrogen supply unit from the hydrogen refueling power supply range.

[0134] Step 64: Determine the minimum daily total operating cost corresponding to the energy supply condition planning result according to the sub-objective function, and repeat the step of using the objective solver to solve the energy supply condition planning result that satisfies the objective constraints based on the determined minimum daily total operating cost, until the target energy supply condition planning result that satisfies the scheduling adaptability limit is obtained.

[0135] The objective function of the lower-level model is to minimize the operating cost. The calculation of the total daily operating cost can be found in step 51 using the formula for minimizing the total daily operating cost, which will not be repeated here.

[0136] The lower-level model considers the following constraints: equipment output constraints, energy purchase constraints from outside the system, energy storage device constraints, energy power balance constraints, and vehicle state constraints, as detailed below: (1) Equipment output constraints The operating power of energy equipment in a multi-energy complementary system in a logistics park cannot exceed its capacity limit, which can be expressed as: P j，min ≤P j，t ≤P j，max , where P j，t P represents the operating power of each power supply device at time t. j，min and P j，maxThese represent the upper and lower limits of the device's operating power, with the upper limit determined by the capacity of various devices generated at the upper layer.

[0137] (2) Energy purchase constraints The amount of energy a logistics park can purchase from the upstream power grid and gas network in a multi-energy complementary system has a certain upper limit, which can be expressed as: , ,in, and These represent the maximum power capacity for purchasing electricity and gas from the power grid and gas grid, respectively.

[0138] (3) Constraints of energy storage equipment The relevant constraints for energy storage devices are as follows: , , , , ;in, and These are the upper limits of the charging and discharging power of the energy storage battery, respectively. and All are binary variables, representing the charge and discharge state parameters of the energy storage battery during time period t, with the aim of preventing the energy storage battery from being in both charge and discharge states simultaneously. and These represent the upper and lower limits of the energy storage battery capacity. Considering the continuity of the energy storage dispatch cycle, the energy storage status should remain consistent at the beginning and end of the dispatch cycle. and These represent the energy storage capacity of the energy storage batteries at the beginning and end of the scheduling cycle, respectively.

[0139] The relevant constraints for hydrogen energy storage devices are as follows: , , , , ,in, and These are the upper limits for the charging and discharging power of the hydrogen storage tank, respectively. and These are all binary variables, representing the charging and discharging state parameters of the hydrogen storage tank during time period t. and These represent the upper and lower limits of the hydrogen storage tank capacity. and These represent the hydrogen storage capacity of the hydrogen storage tanks at the beginning and end of the scheduling cycle, respectively.

[0140] (4) Energy supply and demand balance constraints The energy supply and demand balance constraint mainly considers the coupling relationship between various energy sources in the multi-energy complementary system of the logistics park. Among them, the power balance constraint mainly consists of the power supply and demand balance constraints of four energy sources: electricity, cooling, heating, and hydrogen, which are expressed as follows: .

[0141] (5) Vehicle state constraints Vehicle power constraints: Based on vehicle technical specifications and operational limitations of charging and hydrogen refueling stations, there are upper limits on the charging and refueling power of each vehicle, as shown in the following constraints: , , ,in, , and These are the limits for the maximum charging power and hydrogen refueling power of light electric logistics vehicles, electric heavy trucks, and hydrogen-powered heavy trucks, respectively.

[0142] Vehicle operation constraints: To ensure the energy needs of vehicles for daily logistics operations, vehicle load should remain consistent before and after dispatching, as constrained by the following: , , , where x lelv，i，t , x et，i，t and x ht，i，t These are binary variables, representing whether the vehicle is charging or refueling with hydrogen at the corresponding moment.

[0143] Maximum power constraints for charging / hydrogen refueling: Considering the infrastructure capacity constraints within the park, the total load of all vehicles at any given time must be controlled within the maximum energy supply capacity of charging and hydrogen refueling. The constraint formula is as follows: , ,in, and These represent the maximum polymerization power for charging and hydrogenation, respectively.

[0144] It should be noted that this application embodiment first employs the Monte Carlo simulation method to construct stochastic load models for two types of new energy logistics vehicles: electric and hydrogen. Then, combining the flexible demand response characteristics of the vehicles, a multi-objective bi-level optimization model is established. The upper-level model aims to minimize total annual cost, minimize annual carbon emissions, and maximize renewable energy penetration rate, conducting research on system capacity configuration. The lower-level model, based on mixed-integer linear programming, achieves coordinated optimization of system operation strategies and vehicle scheduling schemes. For this bi-level optimization problem, this application embodiment uses a multi-objective particle swarm optimization algorithm for solution, and combines the entropy weight method-TOPSIS combined evaluation method to select the optimal compromise solution.

[0145] This embodiment also provides a multi-energy complementary system control device for logistics parks. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0146] Figure 4 This is a structural block diagram of a logistics park multi-energy complementary system control device according to an embodiment of this application, such as... Figure 4 As shown, the device includes an iteration module 41, a scheduling module 42, a processing module 43, and a filtering module 44, wherein... The iteration module 41 is used to perform population search iteration on multiple device configuration particles using the particle swarm optimization algorithm to generate multiple first configuration particles corresponding to the current iteration. The device configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of the target logistics park for freight and warehousing services within the target planning period. Each particle of the device configuration particles is associated with a type of energy supply device, and the particle position is used to characterize the decision of the capacity configuration of a type of energy supply device. The scheduling module 42, coupled to the iteration module 41, is used to solve the daily operation scheduling plan based on mixed integer linear programming for all particles of each first configuration particle individual using the objective solver, generate daily scheduling response data corresponding to each first configuration particle individual, and calculate the configuration fitness corresponding to the first configuration particle individual based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle individual. The daily scheduling response data is used to characterize the energy supply conditions configured for multiple energy supply devices. The processing module 43, coupled to the iteration module 41 and the scheduling module 42, is used to repeatedly execute the steps of generating multiple first configuration particles using the particle swarm optimization algorithm and generating corresponding daily scheduling response data based on the objective solver, based on the configuration fitness, until the corresponding configuration fitness meets the preset fitness limit, and thus obtain multiple second configuration particles. The filtering module 44, coupled to the processing module 43, is used to select target device configuration particles from multiple second configuration particles using the TOPSIS method, solve the daily scheduling response data of the target device configuration particles within the target planning period using the target solver, obtain the target scheduling response data, and use the particle position of each particle of the target device configuration particles and all target scheduling response data as the optimization result.

[0147] In some embodiments, the filtering module 44 further includes: The acquisition unit is used to acquire multiple target decision parameters associated with the objective function corresponding to each second configuration particle individual, and to preprocess the multiple target decision parameters to generate multiple standard decision parameters. Among them, the target decision parameters include minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the renewable energy penetration rate. The minimizing of the annual comprehensive cost is determined based on the scheduling fitness corresponding to the daily scheduling response data. The preprocessing includes homogenization processing and normalization processing. The calculation unit, coupled to the acquisition unit, is used to determine the positive ideal solution value with the largest parameter value and the negative ideal solution value with the smallest parameter value among all standard parameter values ​​corresponding to each standard decision parameter, and to determine the Euclidean distance between each standard parameter value and the positive ideal solution value and the negative ideal solution value respectively, so as to obtain the first Euclidean distance and the second Euclidean distance corresponding to each standard parameter value. The determination unit, coupled to the calculation unit, is used to determine the relative proximity of the corresponding standard decision parameter based on the ratio of the second Euclidean distance to the total distance. Then, based on the entropy weight method, it determines the weight of each standard decision parameter and performs a weighted average of the relative proximity of the various standard decision parameters to obtain the total proximity of each second configuration particle. Here, the total distance represents the sum of the first and second Euclidean distances, and the relative proximity represents the relative importance of the corresponding standard decision parameter among all standard decision parameters. The first selection unit, coupled to the determining unit, is used to select the second configuration particle with the highest total proximity from a plurality of second configuration particle individuals sorted from low to high according to total proximity, to obtain the target device configuration particle individual.

[0148] Before generating multiple equipment configuration particles, the multi-energy complementary system control device for this logistics park also acquires the building energy load demand of the target logistics park in multiple seasonal sub-cycles within the target planning period. The energy supply corresponding to the building energy load demand includes a first power supply demand, a heating demand, and a cooling demand. Using Monte Carlo simulation, the freight load demand in each seasonal sub-cycle is generated, and the hydrogen supply demand and the second power supply demand corresponding to the freight load demand are determined. After merging the first and second power supply demands into the corresponding power supply demand, the capacity configuration range corresponding to the energy supply equipment for each type of energy supply demand is determined according to a preset margin rule. The energy supply demand includes the following energy requirements: electricity, heat, cooling, and hydrogen. Each energy supply demand is associated with at least one energy supply equipment, and the capacity configuration range is used to determine the particle position.

[0149] The multi-energy complementary system control device of the logistics park is also used to acquire target information and operating parameters of target vehicles participating in freight transportation in the target logistics park. The target information is used to characterize the type and number of target vehicles. The operating parameters include the entry time, exit time, and initial energy state parameters of the target vehicles upon entry. The initial energy state parameters include the state of charge and the hydrogen storage state. Each operating parameter is associated with a corresponding probability distribution. Based on the corresponding probability distribution, Monte Carlo simulation is used to randomly sample the operating parameters of all target vehicles. Based on the sampled operating parameters, the refueling time and hourly load of the target vehicles under the set operating scenario are calculated. Based on the refueling time and hourly load of the target vehicles, the hourly load of all target vehicles is superimposed in time sequence to obtain the total hydrogen refueling load and total charging load in each seasonal sub-cycle. The hydrogen supply demand includes the total hydrogen refueling load, and the second power supply demand includes the total charging load.

[0150] The multi-energy complementary system control device for the logistics park is also used to determine the energy supply equipment configured for each type of energy demand within the target logistics park. This energy supply equipment includes power supply equipment, heating equipment, cooling equipment, and hydrogen supply equipment. The power supply equipment includes photovoltaic equipment and at least one first power supply equipment. Based on the building physical parameters within the target logistics park and the equipment parameters of the photovoltaic equipment, the device determines the corresponding assembly capacity of the photovoltaic equipment. The building physical parameters characterize the roof area used for photovoltaic equipment installation, and the equipment parameters include the photovoltaic module area and the power supply capacity of a single photovoltaic device. The device obtains the peak load of various energy demands within each seasonal sub-cycle and determines the energy supply for each type of energy demand based on the peak load and a preset margin coefficient. The energy supply corresponding to the first power supply equipment is the amount of electricity supplied after deducting the electrical energy provided by the assembly capacity. The assembly capacity is used as the upper limit of the capacity configuration range corresponding to the photovoltaic equipment, and the corresponding energy supply is used sequentially as the upper limit of the capacity configuration range corresponding to the first power supply equipment, heating equipment, cooling equipment, and hydrogen supply equipment, thus obtaining the corresponding capacity configuration ranges.

[0151] The processing module 43 further includes: The processing unit is used to determine the initial particle position and preset initial particle velocity of each device-configured particle individual, wherein the capacity configuration value corresponding to the initial particle position belongs to the capacity configuration range. The generation unit, coupled to the processing unit, is used to determine the dynamic adaptive inertia factor, the first acceleration coefficient, and the second acceleration coefficient corresponding to the current particle search based on a preset dynamic adaptive change method. The dynamic adaptive inertia factor, the first acceleration coefficient, and the second acceleration coefficient are generated by adaptively changing the configuration fitness and the number of search iterations corresponding to each first configuration particle. The search unit, coupled to the generation unit, is used to perform at least one population search iteration on the device configuration particle individuals based on the initial particle position, initial particle velocity, dynamic adaptive inertia factor, first acceleration coefficient and second acceleration coefficient, using a multi-objective particle swarm optimization algorithm, until the configuration fitness corresponding to the generated candidate configuration particle individuals meets the preset fitness limit, thereby obtaining the same number of candidate configuration particle individuals as the device configuration particle individuals, wherein the second configuration particle individuals include the candidate configuration particle individuals.

[0152] The scheduling module 42 further includes: The first calculation unit is used to calculate the minimum daily total operating cost corresponding to the daily scheduling response data according to the sub-objective function corresponding to the daily scheduling response data, and to determine the total operating cost in multiple seasonal sub-cycles based on the minimum daily total operating cost, and to determine the annual comprehensive cost corresponding to the first configuration particle individual according to the equipment cost and total operating cost of various energy supply equipment. The scheduling fitness includes minimizing the daily total operating cost. The sub-objective function is constructed based on the energy supply equipment operating cost parameters, energy purchase cost parameters and photovoltaic equipment power generation cost parameters generated under the daily energy supply conditions of the energy supply equipment. The energy supply conditions include the type of energy supply equipment to be scheduled, the scheduling operation time and the energy supply power during operation. The second calculation unit is used to determine the amount of gas required for the first power supply equipment based on gas power supply, and to determine the purchased electricity in the amount of electricity required for the heating equipment, cooling equipment and hydrogen supply equipment based on electric power supply. Based on the amount of gas, the purchased electricity and the corresponding unit carbon emission factor, the annual carbon emission of the first configuration particle is calculated. The third calculation unit is used to obtain the daily power output of the photovoltaic equipment and the daily power supply of all energy supply equipment from all daily dispatch response data, and to determine the renewable energy penetration rate corresponding to the first configuration particle based on the ratio of the daily power output and the daily power supply. The daily power output is determined based on the dispatch operation time and the power supply during operation of the photovoltaic equipment, and the daily power supply is used to characterize the daily load demand in the target logistics park. The weighting unit, coupled to the first, second, and third operation units, is used to normalize and weight the annual comprehensive cost, annual carbon emissions, and renewable energy penetration rate corresponding to the first configuration particle individual to generate the corresponding configuration fitness.

[0153] The first computing unit is also used to calculate the minimum total daily operating cost according to the following formula: minF sys =f ode +f buy +f pv , , , , where f ode This represents the operating cost of all power supply equipment; f buy Indicates energy purchase cost; f pv Indicates the power generation cost of photovoltaic equipment; Co j Q represents the unit operation and maintenance cost of the j-th type of energy supply equipment; oj P represents the operating power of the j-th type of power supply device at time t; buy_e，t Indicates the amount of electricity purchased externally; P buy_gas，t Indicates gas consumption; C e，t C is the purchase price of electricity. gas，t Gas purchase price refers to the price of natural gas, C pv Price of penalty for abandoning light per unit. P is the predicted maximum output power of the photovoltaic system at time t. pv，t Let be the output power of the photovoltaic system at time t.

[0154] The scheduling module 42 further includes: The second selection unit is used to select, from all the power supply equipment associated with the target logistics park, a second power supply device and a first hydrogen supply unit for supplying power and hydrogen to the target vehicles involved in freight transportation, and to determine the corresponding charging power supply range and hydrogen supply range based on the particle position of the particle corresponding to the second power supply device and the first hydrogen supply unit in the first configuration particle individual. The first scheduling unit, coupled to the second selection unit, is used to determine the building energy supply equipment and the first energy consumption interval determined according to the particle position of the particle corresponding to each building energy supply equipment in the first configuration particle individual after determining the dynamic scheduling energy supply period corresponding to the second power supply equipment and the first hydrogen supply unit under the preset scheduling mode. The scheduling mode includes one of the following: disordered scheduling and flexible scheduling. The dynamic scheduling energy supply period is used to characterize the charging period or hydrogen refueling period allocated to the target vehicle under the corresponding scheduling mode. The second scheduling unit, coupled to the first scheduling unit, is used to solve for the energy supply condition planning results that satisfy the objective constraints using the energy supply conditions of the building energy supply equipment, the second power supply equipment, and the first hydrogen supply unit as lower-level decision variables and an objective solver. The daily scheduling response data includes the energy supply condition planning results, which include at least one of the following: the first energy consumption of the building energy supply equipment in each corresponding scheduling period of the day, the charging power supply of the second power supply equipment in each charging period of the corresponding charging period, and the hydrogen supply of the first hydrogen supply unit in each hydrogen refueling period of the corresponding hydrogen refueling period. The solver unit, coupled to the first and second scheduling units, is used to determine the minimum daily total operating cost corresponding to the energy supply condition planning result according to the sub-objective function. Based on the determined minimum daily total operating cost, the solver is used to repeatedly solve the energy supply condition planning result that satisfies the objective constraints until the target energy supply condition planning result that satisfies the scheduling adaptability limit is obtained.

[0155] This embodiment also provides a service platform, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0156] Optionally, the service platform may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0157] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program: S1. Using the particle swarm optimization algorithm, a population search is performed on multiple device configuration particles to generate multiple first configuration particles corresponding to the current iteration. The device configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of the target logistics park for freight and warehousing services within the target planning period. Each particle of the device configuration particles is associated with a type of energy supply device, and the particle position is used to characterize the decision of the capacity configuration of a type of energy supply device.

[0158] S2, using the objective solver, performs daily operation scheduling planning based on mixed integer linear programming for all particles of each first configuration particle individual, generates daily scheduling response data corresponding to each first configuration particle individual, and calculates the configuration fitness corresponding to the first configuration particle individual based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle individual. The daily scheduling response data is used to characterize the energy supply conditions configured for multiple energy supply devices.

[0159] S3, based on configuration fitness, repeatedly execute the steps of generating multiple first configuration particles using the particle swarm optimization algorithm and generating corresponding daily scheduling response data based on the objective solver, until the corresponding configuration fitness meets the preset fitness limit, and obtain multiple second configuration particles.

[0160] S4. Using the TOPSIS method, target device configuration particles are selected from multiple second configuration particles. Using the target solver, the daily scheduling response data of the target device configuration particles is solved for each day within the target planning period. The target scheduling response data is obtained, and the particle position of each particle of the target device configuration particles and all target scheduling response data are used as the optimization results.

[0161] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0162] Furthermore, in conjunction with the logistics park multi-energy complementary system management and control method in the above embodiments, this application embodiment can provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the logistics park multi-energy complementary system management and control methods in the above embodiments.

[0163] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0164] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A management and control method for a multi-energy complementary system in a logistics park, characterized in that, include: Using the particle swarm optimization algorithm, a population search iterative process is performed on multiple device configuration particles to generate multiple first configuration particles corresponding to the current iteration. The device configuration particles are randomly encoded based on the load demand corresponding to the energy supply demand of the target logistics park for freight and warehousing services within the target planning period. Each particle of the device configuration particles is associated with a type of energy supply device, and the particle position is used to characterize the decision of the capacity configuration of an energy supply device. Using the objective solver, a daily operation scheduling plan based on mixed integer linear programming is solved for all the particles of each first configuration particle individual to generate daily scheduling response data corresponding to each first configuration particle individual. Based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configuration particle individual, the configuration fitness corresponding to the first configuration particle individual is calculated. The daily scheduling response data is used to characterize the energy supply conditions for various energy supply equipment configurations. Based on the configuration fitness, the steps of generating multiple first configuration particles using the particle swarm optimization algorithm and generating the corresponding daily scheduling response data based on the target solver are repeated until the corresponding configuration fitness meets the preset fitness limit, thus obtaining multiple second configuration particles. Using the TOPSIS method, a target device configuration particle is selected from multiple second configuration particle individuals. Using the target solver, the daily scheduling response data of the target device configuration particle individual is solved for each day within the target planning period to obtain the target scheduling response data. The particle position of each particle of the target device configuration particle individual and all the target scheduling response data are used as the optimization result.

2. The method according to claim 1, characterized in that, Using the TOPSIS method, a target device configuration particle individual is selected from multiple second configuration particle individuals, including: The objective decision parameters associated with the objective function corresponding to each individual second configuration particle are obtained, and the objective decision parameters are preprocessed to generate a variety of standard decision parameters. The objective decision parameters include minimizing the annual comprehensive cost, minimizing the annual carbon emissions, and maximizing the renewable energy penetration rate. The minimizing of the annual comprehensive cost is determined based on the scheduling fitness corresponding to the daily scheduling response data. The preprocessing includes homing processing and normalization processing. Among all standard parameter values ​​corresponding to each of the standard decision parameters, the positive ideal solution value with the largest parameter value and the negative ideal solution value with the smallest parameter value are determined, and the Euclidean distance between each standard parameter value corresponding to each of the standard decision parameters and the positive ideal solution value and the negative ideal solution value are determined respectively, so as to obtain the first Euclidean distance and the second Euclidean distance corresponding to each standard parameter value. After determining the relative proximity of the corresponding standard decision parameter based on the ratio of the second Euclidean distance to the total distance, the weight of each standard decision parameter is determined based on the entropy weight method. Then, the relative proximity of the various standard decision parameters is weighted according to the weights to obtain the total proximity of each second configuration particle. The total distance represents the sum of the first Euclidean distance and the second Euclidean distance, and the relative proximity represents the relative importance of the corresponding standard decision parameter among all the standard decision parameters. From a plurality of second configuration particle individuals sorted by total proximity from low to high, the second configuration particle individual with the highest total proximity is selected to obtain the target device configuration particle individual.

3. The method according to claim 1, characterized in that, Before encoding and generating multiple individual device configuration particles, the method further includes: The building energy load demand of the target logistics park is obtained in multiple seasonal sub-cycles of the target planning period, wherein the energy supply corresponding to the building energy load demand includes the first power supply demand, heating demand, and cooling demand. Using the Monte Carlo simulation method, the freight load demand in each seasonal sub-cycle is generated, and the corresponding hydrogen supply demand and second power supply demand are determined. After merging the first power supply demand and the second power supply demand into a corresponding power supply demand, the capacity configuration range corresponding to the power supply equipment for each type of power supply demand is determined according to a preset margin rule. The power supply demand includes the following energy demands: electrical energy, thermal energy, cold energy, and hydrogen energy. Each type of power supply demand is associated with at least one of the power supply equipment. The capacity configuration range is used to determine the particle position.

4. The method according to claim 3, characterized in that, Using Monte Carlo simulation, freight load demand is generated for each seasonal sub-cycle, including: Obtain target information and operational parameters of target vehicles participating in freight transportation in the target logistics park. The target information is used to characterize the type and number of target vehicles. The operational parameters include the entry time, exit time, and initial energy state parameters of the target vehicles upon entry. The initial energy state parameters include the state of charge and the hydrogen storage state. Each operational parameter is associated with a corresponding probability distribution. Based on the corresponding probability distribution, Monte Carlo simulation is used to randomly sample the operating parameters of all the target vehicles, and the refueling time and hourly load of the target vehicles in the set operating scenario are calculated according to the sampled operating parameters. Based on the refueling time of the target vehicles and the hourly load, the hourly loads of all the target vehicles are superimposed in time sequence to obtain the total hydrogen refueling load and total charging load in each seasonal sub-cycle. The hydrogen supply demand includes the total hydrogen refueling load, and the second power supply demand includes the total charging load.

5. The method according to claim 4, characterized in that, According to preset margin rules, determine the capacity configuration range corresponding to the energy supply equipment for each type of energy demand, including: The energy supply equipment to be configured for each of the energy supply needs within the target logistics park is determined, wherein the energy supply equipment includes power supply equipment, heating equipment, cooling equipment and hydrogen supply equipment, and the power supply equipment includes photovoltaic equipment and at least one first power supply equipment; Based on the building physical parameters within the target logistics park and the equipment parameters of the photovoltaic equipment, the corresponding assembly capacity of the photovoltaic equipment is determined. The building physical parameters are used to characterize the roof area of ​​the building where the photovoltaic equipment is installed, and the equipment parameters include the area of ​​the photovoltaic modules and the power supply of a single photovoltaic device. Obtain the peak load of various energy demands within each seasonal sub-cycle, and determine the energy supply for each energy demand based on the peak load and a preset margin coefficient, wherein the energy supply corresponding to the first power supply equipment is the amount of power supply after deducting the power supplied by the assembly capacity. The assembly capacity is used as the upper limit of the capacity configuration range corresponding to the photovoltaic equipment, and the corresponding energy supply is used as the upper limit of the capacity configuration range corresponding to the first power supply equipment, the heating equipment, the cooling equipment and the hydrogen supply equipment in sequence, so as to obtain the corresponding capacity configuration range.

6. The method according to claim 5, characterized in that, Using the particle swarm optimization algorithm, multiple second-configuration particle individuals are generated, including: Determine the initial particle position and preset initial particle velocity of each of the device configuration particle individuals, wherein the capacity configuration value corresponding to the initial particle position belongs to the capacity configuration range; Based on a preset dynamic adaptive change method, the dynamic adaptive inertia factor, the first acceleration coefficient, and the second acceleration coefficient corresponding to the current particle search are determined respectively. The dynamic adaptive inertia factor, the first acceleration coefficient, and the second acceleration coefficient are generated by adaptively changing the configuration fitness and the number of search iterations corresponding to each first configuration particle individual. Based on the initial particle position, the initial particle velocity, the dynamic adaptive inertia factor, the first acceleration coefficient, and the second acceleration coefficient, a multi-objective particle swarm optimization algorithm is used to perform at least one population search iteration on the device configuration particle individuals until the configuration fitness corresponding to the generated candidate configuration particle individuals meets a preset fitness limit, thereby obtaining the same number of candidate configuration particle individuals as the device configuration particle individuals, wherein the second configuration particle individuals include the candidate configuration particle individuals.

7. The method according to claim 6, characterized in that, Based on the scheduling fitness corresponding to the daily scheduling response data and the objective function associated with the first configured particle, the configuration fitness corresponding to the first configured particle is calculated, including: The minimum daily total operating cost corresponding to the daily scheduling response data is calculated according to the sub-objective function corresponding to the daily scheduling response data. Based on the minimum daily total operating cost, the total operating cost in multiple seasonal sub-cycles is determined. The annual comprehensive cost corresponding to the first configuration particle is determined according to the equipment cost of various energy supply devices and the total operating cost. The scheduling fitness includes the minimum daily total operating cost. The sub-objective function is constructed based on the energy supply device operating cost parameters, energy purchase cost parameters and photovoltaic device power generation cost parameters generated under the daily energy supply conditions of the energy supply devices. The energy supply conditions include the type of energy supply device being scheduled, the scheduling operation time and the energy supply power during operation. The amount of gas required for the first power supply equipment based on gas power supply is determined, and the amount of purchased electricity required for the heating equipment, cooling equipment and hydrogen supply equipment based on electric power supply is determined. Based on the amount of gas used, the amount of purchased electricity and the corresponding unit carbon emission factor, the annual carbon emission of the first configured particle is calculated. From all the daily scheduling response data, the daily power output of the photovoltaic equipment and the daily power supply of all the energy supply equipment are obtained. Based on the ratio of the daily power output to the daily power supply, the renewable energy penetration rate corresponding to the first configuration particle is determined. The daily power output is determined based on the scheduling operation time and the power supply during operation of the photovoltaic equipment. The daily power supply is used to characterize the daily load demand in the target logistics park. The annual comprehensive cost, annual carbon emissions, and renewable energy penetration rate corresponding to the first configured particle are normalized and weighted to generate the corresponding configuration fitness.

8. The method according to claim 7, characterized in that, Calculate the minimum total daily operating cost using the following formula: minF sys =f ode +f buy +f pv Among them, f ode This represents the operating cost of all the aforementioned power supply equipment; f buy Indicates energy purchase cost; f pv This indicates the power generation cost of the photovoltaic equipment; Co j Qo represents the unit operation and maintenance cost of the j-th type of energy supply equipment; j P represents the operating power of the j-th type of power supply device at time t; buy_e，t Indicates the amount of electricity purchased externally; P buy_gas，t Indicates gas consumption; C e，t C is the purchase price of electricity. gas，t Gas purchase price refers to the price of natural gas, C pv Price of penalty for abandoning light per unit. P is the predicted maximum output power of the photovoltaic system at time t. pv，t Let be the output power of the photovoltaic system at time t.

9. The method according to claim 8, characterized in that, Using the objective solver, a daily scheduling plan based on mixed-integer linear programming is solved for all particles of each first-configuration particle individual, generating daily scheduling response data corresponding to each first-configuration particle individual, including: Among all the energy supply devices associated with the target logistics park, a second power supply device and a first hydrogen supply unit are selected for supplying power and hydrogen to the target vehicles involved in freight transportation. Based on the particle position of the particle in the first configured particle individual that corresponds to the second power supply device and the first hydrogen supply unit, the corresponding charging power supply range and hydrogen refueling power supply range are determined. After determining the dynamic scheduling energy supply period corresponding to the second power supply equipment and the first hydrogen supply unit under the preset scheduling mode, the building energy supply equipment and the first energy consumption interval determined according to the particle position of the particle corresponding to each building energy supply equipment in the first configured particle individual are determined. The scheduling mode includes one of the following: disordered scheduling and flexible scheduling. The dynamic scheduling energy supply period is used to characterize the charging period or hydrogen refueling period allocated to the target vehicle under the corresponding scheduling mode. Using the energy supply conditions of the building energy supply equipment, the second power supply equipment, and the first hydrogen supply unit as lower-level decision variables, the target solver is used to solve for the energy supply condition planning results that satisfy the target constraints. The daily dispatch response data includes the energy supply condition planning results, which include at least one of the following: the first energy consumption of the building energy supply equipment in each corresponding dispatch period of the day, the charging power supply of the second power supply equipment in each charging period in the corresponding charging period, and the hydrogen supply of the first hydrogen supply unit in each hydrogen refueling period in the corresponding hydrogen refueling period. According to the sub-objective function, determine the minimum daily total operating cost corresponding to the energy supply condition planning result, and repeat the step of using the objective solver to solve the energy supply condition planning result that satisfies the objective constraints based on the determined minimum daily total operating cost, until the target energy supply condition planning result that satisfies the scheduling adaptability limit is obtained.

10. A service platform, comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the steps of the logistics park multi-energy complementary system management method according to any one of claims 1 to 9.

Images