Hydrogen energy hybrid unmanned aerial vehicle capacity configuration method and system based on two-stage optimization

Abstract

The invention discloses a hydrogen energy hybrid unmanned aerial vehicle capacity configuration method and system based on two-stage optimization. The method comprises the following steps: constructing an energy system model, a wind uncertainty model, a multi-time scale comprehensive evaluation index and a corresponding objective function of a hydrogen energy hybrid unmanned aerial vehicle; based on the hydrogen energy hybrid unmanned aerial vehicle energy system model, comprehensive evaluation indexes and corresponding objective functions, constructing a two-stage stochastic optimization model of hydrogen energy hybrid unmanned aerial vehicle capacity configuration considering uncertain factors; and solving the two-stage stochastic programming model in combination with a wind uncertainty model to obtain a capacity configuration result of the hydrogen energy hybrid unmanned aerial vehicle. The research of the invention provides a theoretical basis for the two-stage optimization capacity configuration method in the field of hydrogen energy hybrid unmanned aerial vehicles, the hybrid integer linear programming and the random field model are combined, the uncertainty factors such as environmental wind field and gravity change in the flight process of the unmanned aerial vehicle are considered, and the accuracy and economy of capacity configuration are improved.

Description

Technical Field

[0001] The present invention belongs to the field of UAV energy system engineering, and in particular relates to a hydrogen hybrid UAV capacity configuration method based on two-stage optimization. Background Art

[0002] Compared with traditional lithium battery drones, hydrogen hybrid drones have longer cruising time; compared with diesel or gasoline drones, they are more energy-efficient and environmentally friendly; compared with pure hydrogen drones, they have faster power response under transient load changes. Therefore, designing a hydrogen fuel cell / lithium battery hybrid system is the key to realizing cruising drones, low emissions, and fast response. In configuration optimization research, equipment characteristic modeling refers to the establishment of mathematical models related to the capacity, performance parameters, and operating output parameters of the equipment, which is the basis of the integrated configuration model. According to the degree of refinement of the equipment model in the configuration optimization model, it can be divided into two forms: steady-state standard condition model and variable operating condition model. A two-stage stochastic programming method is used to optimize the design and operation of the energy storage subsystem in the renewable energy system. The random factors are expressed in the form of equivalent scenarios, and the optimization problem is converted into a mixed integer linear programming (MILP) problem for solution.

[0003] At present, some studies have applied the two-stage stochastic programming method to the capacity configuration of energy systems, but the technical problems and difficulties of existing technologies include: it is difficult to generate representative uncertainty scenarios, it is difficult to achieve efficient coupling optimization on multiple time scales, and it is difficult to maintain a balance between economic and environmental benefits. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the present invention provides a hydrogen hybrid UAV capacity configuration method based on two-stage optimization to address the defects in the prior art that it is difficult to generate representative uncertainty scenarios, difficult to achieve efficient coupling optimization on multiple time scales, and difficult to maintain a balance between economic and environmental benefits.

[0005] The present invention adopts the following technical solution.

[0006] A method for configuring the capacity of a hydrogen hybrid UAV based on two-stage optimization includes the following steps:

[0007] Step 1: Construct the energy system model, wind uncertainty model, multi-time scale comprehensive evaluation index and corresponding objective function of hydrogen hybrid UAV;

[0008] Step 2: Based on the hydrogen hybrid UAV energy system model, comprehensive evaluation indicators and corresponding objective functions, a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration considering uncertain factors is constructed;

[0009] Step 3: Solve the two-stage stochastic programming model in combination with the wind uncertainty model to obtain the capacity configuration result of the hydrogen hybrid drone.

[0010] Preferably, the hydrogen hybrid UAV energy system mainly includes a hydrogen fuel cell, a lithium battery, a hydrogen storage tank, and an engine.

[0011] Preferably, the constructed wind uncertainty model is as follows:

[0012]

[0013] In the formula, -ΔW G，j (r, t+τ|τ) represents the wind uncertainty prediction value based on time t at position vector r and forecast look-ahead time t+T; It represents the time-delayed RUC data error at the forecast look-ahead time τ based on the position vector r at time t; They represent the initial measurement error and nominal wind error respectively.

[0014] Preferably, the step 1 further comprises:

[0015] The RUC data of wind speed and direction are collected and processed by interpolation method, and the processed data are used as the data source of wind uncertainty model;

[0016] Among them, the preprocessing of RUC data by interpolation method specifically includes:

[0017] The constructed wind uncertainty model is optimized by combining spline interpolation with inverse distance weighted interpolation: first, the inverse distance weighted interpolation method is used to reduce the data to a scale of 1 km, and then the cubic spline interpolation method is used to reduce the data after inverse distance weighted interpolation to a horizontal resolution of 5 m.

[0018] Preferably, the constructed multi-time scale comprehensive evaluation index includes short-term cost and long-term cost, and the corresponding objective function is as follows:

[0019] min J=ω1C short +ω2C long

[0020] Where J is the target value of comprehensive optimization, C short is the short-term cost, including fuel consumption and maintenance costs; C long is the long-term cost, including equipment investment and service life discount; ω1 and ω2 are the weight coefficients of short-term cost and long-term cost, respectively, which are used to balance the relative importance of short-term and long-term goals.

[0021] Preferably, the short-term cost C short The calculation formula is as follows:

[0022] C short =k1C f +k2C m

[0023] In the formula, C f represents the hydrogen fuel consumption cost, C m represents the regular maintenance cost, k1 and k2 represent the weight coefficients of hydrogen fuel consumption cost and regular maintenance cost respectively;

[0024] Long term cost long The calculation formula is as follows:

[0025] C long =k3C e +k4C l

[0026] In the formula, C e Indicates the one-time investment cost of the equipment, C l represents the service life conversion, k3 and k4 represent the weight coefficients of the one-time investment cost and service life conversion of the equipment respectively.

[0027] Preferably, a two-stage stochastic optimization problem for hydrogen hybrid UAV capacity configuration is constructed, specifically including:

[0028] Step 2.1, construct the constraints of the energy system of the hydrogen-powered hybrid UAV, including energy balance constraints, capacity constraints, and energy storage constraints;

[0029] Step 2.2, construct a two-stage stochastic planning model for hydrogen hybrid drone capacity configuration.

[0030] Preferably, the energy balance constraint includes: the electrical energy obtained in the hydrogen hybrid UAV energy system should be greater than or equal to the electrical energy consumption to maintain electrical balance:

[0031]

[0032] In the formula, E FC Represents the electrical energy provided by hydrogen fuel cells, Indicates the electrical energy output by the lithium battery, E M Indicates the electrical energy consumed by the DC motor, E F Indicates the electric energy consumed by the fan, E CD Indicates the power consumed by the communication equipment, Indicates the electrical energy input by the lithium battery;

[0033] The capacity constraints include: the operating output power of various equipment in the hydrogen hybrid drone system should not be greater than the equipment capacity at the time of design, and the minimum operating power of the hydrogen fuel cell should not be less than 30% of the equipment capacity to avoid running in an underload state. Therefore, the capacity constraints for hydrogen fuel cells and fans are as follows:

[0034] Y FC，t ×0.3Cap FC ≤E FC ≤Y FC，t ×Cap FC

[0035] 0≤E F ≤Y F，t ×Cap F

[0036] In the formula, represents a Boolean variable used to determine whether the hydrogen fuel cell is running, and its value is 0 or 1;

[0037] Cap FC Indicates the maximum output power of hydrogen fuel cells;

[0038] Y F，t Indicates a Boolean variable used to determine whether the fan is running, with a value of 0 or 1;

[0039] Cap F Indicates the maximum output power of the fan;

[0040] The energy storage constraints include electric energy storage device constraints and hydrogen energy storage device constraints;

[0041] The energy storage device is a lithium battery, which meets the following requirements: the lithium battery has stored energy before the drone takes off; when the hydrogen fuel is fully supplied, the lithium battery does not need to be recharged, and the lithium battery power at the end of the previous flight is equal to the starting power at the next flight; the storage and use of the lithium battery do not occur at the same time and are not greater than the maximum storage and use power of the lithium battery; the lithium battery storage energy is within the range of 20% to 80% based on the design capacity;

[0042] The hydrogen energy storage device is a hydrogen storage tank, which meets the following requirements: before the UAV takes off, the hydrogen storage tank has stored hydrogen energy; no refueling is performed when the hydrogen fuel is not less than 20% of the maximum capacity, and the amount of hydrogen in the hydrogen storage tank at the end of the previous voyage is equal to the amount of hydrogen at the beginning of the next voyage; the storage and use of the hydrogen storage tank do not occur at the same time, and the use rate of the hydrogen storage tank is not greater than the maximum use power of the hydrogen storage tank.

[0043] Preferably, a two-stage stochastic planning model is constructed by combining the equipment capacity and the operating variables during the operation of the hydrogen hybrid UAV. The capacity of all equipment is the first-stage variable; the operating variables are the second-stage variables, including the output power of all equipment at each moment, the amount of electricity stored in the power storage facility at each moment, and the amount of hydrogen stored in the hydrogen storage facility at each moment. The constructed two-stage stochastic planning model is as follows:

[0044] min(ω1C short +ω2C long )

[0045] =min[ω1(k1C f +k2C m )+ω2(k3C e +k4C l )]

[0046]

[0047]

[0048] In the formula, f d is the deterministic part of the objective function, that is, the investment cost C corresponding to the selected equipment capacity inv , the random part is composed of f s The expectation representation, f s represents the cost associated with the uncertainty parameter, including operation and maintenance costs C CM and the cost of purchasing resources C tari ; d is a continuous variable determined in the design process, ξ is an uncertain parameter, including wind speed and load size; ψ dc To be the inequality constraints in the design process; and ψ oc are the equality and inequality constraints during operation respectively; x t and t They are continuous variables and 0-1 variables involved in the operation process, which exist in every time interval t during the project life cycle; Points represent the feasible domain of continuous variables involved in the operation process; n represents the number of 0-1 variables.

[0049] Preferably, the step 3 specifically includes:

[0050] The two-stage stochastic programming model is solved by combining the wind speed and direction data output by the wind uncertainty model. The obtained UAV capacity configuration results include the maximum power and rated power of the fuel cell, the battery capacity of the lithium battery, the heat dissipation power of the cooling system, and the hydrogen storage capacity of the hydrogen storage tank.

[0051] The present invention also proposes a hydrogen hybrid UAV capacity configuration system based on two-stage optimization, which is used to implement the hydrogen hybrid UAV capacity configuration method based on two-stage optimization, including: a UAV related parameter construction module, a random optimization model construction module and a capacity configuration solution module;

[0052] The UAV-related parameter construction module is used to construct the hydrogen hybrid UAV energy system model, wind uncertainty model, multi-time scale comprehensive evaluation index and corresponding objective function;

[0053] The stochastic optimization model building module is used to build a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration taking into account uncertain factors based on the hydrogen hybrid UAV energy system model, comprehensive evaluation indicators and corresponding objective functions;

[0054] The capacity configuration solution module is used to solve the two-stage stochastic programming model in combination with the calculation results of the wind uncertainty model to obtain the capacity configuration results of the hydrogen hybrid UAV.

[0055] The present invention also provides a terminal, comprising a processor and a storage medium;

[0056] The storage medium is used to store instructions;

[0057] The processor is used to operate according to the instructions to execute the steps of the hydrogen hybrid drone capacity configuration method based on two-stage optimization.

[0058] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the hydrogen hybrid UAV capacity configuration method based on two-stage optimization.

[0059] The beneficial effects of the present invention are that, compared with the prior art, the present invention has at least the following beneficial effects:

[0060] 1. The present invention proposes a capacity configuration method for hydrogen hybrid UAV based on two-stage optimization. It combines mixed integer linear programming (MILP) and random field models, takes into account uncertain factors such as environmental wind field and gravity changes during the flight of the UAV, improves the accuracy of capacity configuration, can better cope with random fluctuations in the flight environment, provides a more reliable capacity configuration solution, extends the cruising time of the UAV, and ensures the economy of energy use. It is particularly suitable for capacity configuration of UAVs with long flight time;

[0061] 2. The present invention constructs a multi-time scale comprehensive evaluation index and objective function for the capacity configuration of hydrogen hybrid UAVs, weighs short-term fuel costs and maintenance costs with long-term equipment investment and service life, realizes the organic combination of short-term decision-making and long-term planning, and provides a more economical and sustainable capacity configuration solution;

[0062] 3. The present invention proposes an optimized wind field uncertainty random field modeling method based on historical RUC data. On the basis of the existing method based on historical RUC data, it combines inverse distance weighted interpolation (IDW) and spline interpolation to construct more refined wind field data, reducing the minimum data spatial resolution of 13km to the minimum data spatial resolution of 5m, more accurately reflecting the characteristics of the environmental wind field during the flight of the UAV, improving the spatial and temporal resolution of the wind field data, and improving the modeling accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of the hydrogen hybrid UAV capacity configuration method based on two-stage optimization in the present invention;

[0064] Figure 2 It is a two-stage stochastic programming flow chart in the present invention;

[0065] Figure 3 It is a structural diagram of the hydrogen hybrid UAV capacity configuration system based on two-stage optimization in the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0067] like Figure 1 As shown, a hydrogen hybrid UAV capacity configuration method based on two-stage optimization of the present invention comprises the following steps:

[0068] Step 1: Construct a hydrogen hybrid UAV energy system model and a wind uncertainty model, and construct a multi-time scale comprehensive evaluation index and corresponding objective function;

[0069] Furthermore, in step 1, the hydrogen hybrid UAV energy system includes a hydrogen fuel cell, a lithium battery, a hydrogen storage tank, and an engine, and the constructed energy system model includes energy models of the hydrogen fuel cell, the lithium battery, the hydrogen storage tank, and the engine.

[0070] Among them, hydrogen fuel cells:

[0071] The calculation formula for proton exchange membrane fuel cell hydrogen consumption is as follows:

[0072]

[0073] Among them, H FC is the hydrogen consumption rate, artificial working current, nn FC is the number of batteries;

[0074] The fuel cell reaction rate can be expressed by the current density j:

[0075]

[0076] Among them, A cell is the reaction area of ​​the battery cell;

[0077] Thermal power output of fuel cells It is expressed as:

[0078]

[0079] Among them, η T is the thermal efficiency of the fuel cell, η E is the electrical efficiency of the fuel cell.

[0080] Lithium battery:

[0081] Lithium batteries operate in battery pack mode, and their total power is expressed as:

[0082]

[0083] in, is the total power output of the battery pack, is the battery unit power, n1 and n2 are the number of battery rows and columns respectively;

[0084] The current of the battery pack is as follows:

[0085]

[0086] Among them, the open-loop voltage V OC and the battery cell internal resistance R b All can be regarded as fixed values;

[0087] The dynamic characteristic SOC of the battery pack can be expressed as:

[0088]

[0089] Where t is the scheduling time and satisfies 0≤t≤T, T is the scheduling period, Q B is the battery cell capacity, I b，kis the current, charging is negative and discharging is positive;

[0090] The energy currently stored in the lithium battery is related to the energy stored at the previous moment and the charging and discharging of energy between the two moments. The upper and lower limits of the electric energy that can be stored are determined by the equipment capacity, and the charging and discharging power also has upper and lower limit constraints. Due to the continuity of equipment operation, the initial storage state of the equipment each time is equal to the final storage state after the previous cruise. When typical cruise data is used as input, it can be considered that the storage state at the initial and final moments of each cruise is continuous.

[0091] The energy storage state model of lithium battery can be expressed by the following formula (1):

[0092]

[0093] Among them, E BT,t is the energy storage state of the lithium battery, in kWh; σ BT is the self-loss coefficient of the lithium battery; and They are the charging and discharging power of lithium batteries, in kW; and They are the charge and discharge efficiency of lithium batteries; and They are the lower and upper limits of the battery storage state, in kWh; Pe BT is the maximum power of the lithium battery; Δt is the time interval.

[0094] Hydrogen storage tank:

[0095] The hydrogen storage tank uses a high-pressure hydrogen storage tank model (HST) to provide a source of hydrogen for fuel cell power generation. The pressure inside the hydrogen storage tank changes dynamically with the release of hydrogen, expressed as:

[0096]

[0097] in, and are the internal pressure and hydrogen flow rate of the hydrogen storage tank at time t, in mol / s, V H is the volume of the hydrogen storage tank, T H is the operating temperature, R is the universal gas constant, z t is the compressibility factor of hydrogen, which can be calculated based on NIST data and the Lemmon equation:

[0098]

[0099] Among them, a i 、b i 、c iAll are constants, T0 = 100K, R = 8.3145 J / mol·K;

[0100] Table 1: Constant a i 、b i 、c i Value table

[0101]

[0102]

[0103] Electric Motor:

[0104] The hydrogen hybrid UAV of the present invention adopts a brushless DC motor as the main power output, and has the characteristics of low power loss, long life, and low maintenance.

[0105] The equivalent voltage and equivalent current of the equivalent brushless DC motor are as follows:

[0106]

[0107] Among them, I moter is the equivalent current of the motor, U moter is the equivalent voltage of the motor, U m0 is the no-load voltage of the motor, I m0 is the no-load current of the motor, K V0 is the no-load speed of the motor, R moter is the motor internal resistance, M moter is the load torque of the motor, N mote r is the speed of the motor.

[0108] Furthermore, in step 1, the wind field influences considered in constructing the wind uncertainty model include the influence of wind speed and wind direction, specifically:

[0109] During the cruising of hydrogen-powered hybrid UAVs, appropriate wind uncertainty can reflect the basic statistical characteristics of the wind field and the needs of comprehensive trajectory application, which is very important for the capacity configuration of hydrogen-powered hybrid UAVs. For local analysis of UAV trajectories that need to maintain intervals, it is critical to capture the spatial and temporal correlation of the wind field. Usually in this case, it is reasonable to assume that the wind field is stationary, uniform and isotropic. However, for medium and long-distance trajectory predictions that support air traffic control (ATC) and decision support tools (DSTs), the model needs to reflect the non-stationarity, non-uniformity, weather dependence and possible non-Gaussianity of the wind field. This model meets the needs of hydrogen-powered hybrid UAV capacity configuration.

[0110] A convenient method for modeling wind uncertainty is to reasonably utilize a large amount of historical RUC data (numerical weather forecast system data provided by the National Weather Service of the United States). In the present invention, historical RUC data of wind direction and wind speed are obtained for processing.

[0111] However, since RUC data are only valid at discrete network points and specific time points, an appropriate interpolation scheme is needed to obtain samples of wind uncertainty at common locations and times, so that the spatial and temporal correlations of the wind error field need to be preserved.

[0112] This patent constructs a random field model of uncertainty, namely the wind uncertainty model: the wind uncertainty samples for a general location and forward time period are obtained by weighted summation of a sufficient number of historical RUC wind errors at neighboring locations. The locations of these neighboring points are selected based on the spatial range of the current location. For each location, the historical RUC wind error data is calculated from the time-lagged RUC forecasts for the same time period in the same year and season under similar conditions, and the forward time interval is within the time correlation time range of the specified forecast time. The weight selection in the model is inversely proportional to the square of the relative distance and inversely proportional to the absolute value of the forward time interval difference. At the same time, the model also adds two purely random correction terms to take into account other sources of uncertainty.

[0113]

[0114] Where j = 1, ..., N s , j represents the sample number of wind uncertainty, N s is the total number of samples;

[0115] ΔWG,j(r, τ) represents the wind uncertainty prediction value at position vector r and forecast look-ahead time τ;

[0116] represents the wind component provided by RUC;

[0117] The position vector is r, the sample number of wind uncertainty is j, and the forecast look-ahead time is τ k , a pure random variable;

[0118] represents a pure random variable with position vector r and sample number j of wind uncertainty;

[0119] r and ri represent the position vectors, respectively, and the grid point index is the position vector of i;

[0120] τ、τ k They represent the forecast look-ahead time, τ k =0,1,...,9;

[0121] m represents the mass of the aircraft.

[0122] And C i (r) and L k (τ) is as follows

[0123]

[0124] In the formula, C i (r) and L k (τ) represents weight, C i (r) is related to the position, L k (τ) is related to the forecast look-ahead time.

[0125] Since the RUC data show variations in different geographical areas within each forecast and between different forecasts, the resulting model is naturally non-stationary, non-uniform, non-Gaussian and weather-dependent. In addition, uncertainties due to initial measurements and nominal wind field determinations are also appropriately incorporated into the model.

[0126]

[0127] In the formula, -ΔW G，j (r, t+t|t) represents the wind uncertainty prediction value based on time t at position vector r and forecast look-ahead time t+τ;

[0128] It represents the time-delayed RUC data error at the forecast look-ahead time τ based on the position vector r at time t;

[0129] They represent the initial measurement error and nominal wind error respectively.

[0130] Equation (1.6) shows that the initial measurement inputs to the RUC model introduce uncertainty. The basis of the measurement is to relate some physical properties of the wind field to wind speed and direction. Therefore, the measurement errors can be divided into: equipment calibration errors and errors caused by uncertainties in the underlying physical processes. Although further research is needed, it can be assumed that the latter follows a similar pattern to the wind errors. Therefore, the purely random term is added to the model to take into account the correlated part of the measurement error, while another purely random term is added outside the weighted sum to represent the measurement calibration error.

[0131] In contrast, the uncertainty introduced in the nominal wind speed determination process depends on the specific interpolation or approximation method. When interpolation is performed, the error is zero at the RUC grid points, which are 20 km apart horizontally and 25 hPa apart vertically. Therefore, the horizontal spatial scale of the error is about 40 km. This is significantly smaller than the spatial correlation of the wind errors and is therefore considered a purely random term. In addition, the residuals of the weighted sum model are random because the summation already takes into account the spatial and temporal correlations. Therefore, the residuals in equation (1.4) are random. represents a purely random correction for measurement calibration errors, nominal wind field determination errors and model residuals.

[0132] When constructing the above wind uncertainty model, it is assumed that the measurement error, nominal wind field determination error and RUC system prediction error are independent of each other, and the following assumptions are also included:

[0133] (1) The wind uncertainty components along the three perpendicular directions are uncorrelated. Therefore, the corresponding 3×3 correlation matrix is ​​a diagonal matrix. The components in the three directions can be analyzed separately. It should be noted that different coordinate systems affect the correlation of different wind volumes. This patent uses wind directions in three perpendicular directions.

[0134] (2) Compared with the horizontal component, the vertical wind component W h The actual vertical wind volume may not be zero, and its existence may affect the accuracy of the predicted trajectory.

[0135] Furthermore, the collected RUC data of wind speed and wind direction are first processed by the interpolation method proposed in the present invention and then input into the constructed wind uncertainty model, thereby obtaining the wind speed and wind direction data output by the model;

[0136] The interpolation method in the stroke uncertainty model in step 1 is further elaborated.

[0137] In meteorological models, commonly used interpolation methods include linear interpolation, bilinear interpolation, spline interpolation, inverse distance weighted interpolation (IDW), etc. However, for wind fields of wind speed and direction, inverse distance weighted interpolation is difficult to capture complex nonlinear changes, so this patent adopts a method combining spline interpolation and inverse distance weighted interpolation.

[0138] Inverse distance weighted interpolation method

[0139] Inverse distance weighted interpolation (IDW) is an interpolation method based on Tolbrand's first law. The distance between the measured point and the measured value is used as the weight factor, and the unknown point attributes are inferred based on the known point attributes. The expression is as follows:

[0140]

[0141] Among them, Z(x) represents the estimated value of the interpolation point, Z(x i ) represents the value of the i-th interpolation point, n is the number of interpolation points involved in the interpolation, W i Represents the weight value of the i-th interpolation point, calculated as follows:

[0142]

[0143] Where a is the distance index value, is the i-th interpolation point x i The distance to the point x to be measured, weight W i It decreases with increasing distance. In addition, the larger the distance index a is, the more weights are given to interpolation points that are closer to the interpolation point, making the interpolation value close to the value of the nearest interpolation point, and vice versa, closer to the average value of the surrounding interpolation points.

[0144] Spline interpolation method

[0145] Spline interpolation simulates the shape of a given curve by inserting a spline curve between known points to achieve a smoother and more accurate effect.

[0146] Suppose interpolation is performed in the interval [a, b], a=x1<x2,...,<x n = b, each interpolation point has a corresponding function value, let the function F(x i )=y i (i=1,2,...,n) satisfies that they are all polynomials of degree not higher than 3 in [a,b], and F(x) has a second-order derivative in the interval [a,b], then F(x) can be called a cubic spline interpolation function. i , x i+1 ], there exists a cubic polynomial, which can be set to F i (x) = a i x 3 +b i x 2 +c i x+d i , where F(x i )=y i ,F(x i -0)=F(x i +0), F'(x i -0)=F'(x i +0), F”(x i -0)=S”(x i +0), i=(1,2,...,n-1).

[0147] The two methods are combined to adopt a multi-scale interpolation method. Since the data scale range provided by the RUC model is too large, the minimum data spatial resolution is 13km, which cannot be used for hydrogen hybrid drones. First, the IDW is used to reduce the data to a scale of 1km, and then the cubic spline interpolation is used to reduce the data processed by the inverse distance weighted interpolation method to a horizontal resolution of 5m to ensure that the slight airflow changes are captured and meet the size scale of the drone.

[0148] According to the wind uncertainty model and RUC data proposed in the present invention, a set of possible wind speed and wind direction data sets are calculated, as shown in Table 1 below.

[0149] Table 1: Data table of wind speed and wind direction obtained by wind uncertainty model

[0150]

[0151] Furthermore, in step 1, a comprehensive evaluation index of multiple time scales and the corresponding objective function are constructed.

[0152] The comprehensive evaluation indicators of multiple time scales mainly refer to the coupling between short-term decision-making and long-term planning, among which short-term decision-making refers to short-term costs and long-term planning refers to long-term costs.

[0153] Short-term decisions focus on immediate benefits. For hydrogen hybrid drones, short-term costs include hydrogen fuel consumption costs and regular maintenance costs. Long-term planning focuses on the sustainability of the system. For hydrogen hybrid drones, long-term costs include one-time investment in equipment and the service life of the equipment.

[0154] Use weighted summation to construct the corresponding objective function:

[0155] min J=ω1C short +ω2C long (1.9)

[0156] Among them, J is the target value of comprehensive optimization; C short is the short-term cost, including fuel consumption and maintenance costs; C long is the long-term cost, including equipment investment and service life discount; ω1 and ω2 are weight coefficients, balancing the relative importance of short-term and long-term goals.

[0157] Short-term costs:

[0158] C short =k1C f +k2C m (1.10)

[0159] Among them, C f represents the hydrogen fuel consumption cost, C m represents the regular maintenance cost, k1 and k2 represent the weight coefficients.

[0160] C f =P f Q f (1.11)

[0161] Among them, P f Indicates the unit price of hydrogen fuel, unit yuan / L, Q f Indicates the fuel cell hydrogen consumption in a single flight, in L, Q f =f(I FC , t), I FC represents the fuel cell operating current, and t represents the single flight time.

[0162]

[0163] Among them, N m Indicates the number of flights within the regular maintenance cycle, N m Indicates the number of devices that require regular maintenance for hydrogen hybrid drones, K n Represents the equipment maintenance cost coefficient, C n Indicates device parameters. Taking lithium batteries as an example, K n Indicates the unit price during maintenance or inspection, in RMB / kWh, C n Indicates the capacity of lithium battery in kWh.

[0164] Long-term costs:

[0165] C long =k3C e +k4C l (1.13)

[0166] Among them, C e Indicates the one-time investment cost of the equipment, C l represents the service life conversion, k3 and k4 represent the weight coefficients.

[0167]

[0168] Among them, N e Indicates the number of equipment that needs to be invested at one time, P n Indicates the cost of the equipment.

[0169]

[0170] Among them, N l Indicates the number of equipment parts that need to be life-saving, P n represents the cost of the equipment, P n，ev Residual value, that is, the remaining value of the equipment at the end of its service life, N u Indicates service life (number of uses here).

[0171] In this embodiment, the parameter settings are shown in Table 2.

[0172] Table 2: Objective function parameter settings

[0173]

[0174] Step 2: Considering uncertain factors, a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration is constructed;

[0175] Among them, the hydrogen hybrid UAV energy system model is used to consider uncertain factors and construct a two-stage stochastic optimization problem for the capacity configuration of hydrogen hybrid UAVs. The comprehensive evaluation index of multiple time scales and the corresponding objective function are used as the objective function for the capacity configuration optimization in step 2.

[0176] Furthermore, step 2 specifically includes considering uncertainty factors and constructing a two-stage stochastic optimization problem for hydrogen hybrid drone capacity configuration:

[0177] Considering that hydrogen hybrid drones are greatly affected by wind speed and wind direction, and the fluctuation of climate conditions is often random and difficult to predict, the capacity configuration results obtained by optimizing the total system cost through a deterministic model may be inaccurate. This patent proposes a two-stage stochastic programming method to study the optimization problem of hydrogen hybrid drone capacity configuration.

[0178] Furthermore, step 2 constructs a two-stage stochastic optimization problem for hydrogen hybrid drone capacity configuration:

[0179] Step 2.1, construct energy system constraints, mainly including energy balance constraints, capacity constraints, and energy storage constraints.

[0180] Energy balance constraint: In the hydrogen hybrid drone energy system, the main concern is the electrical load demand. During the operation of the system, the amount of electrical energy obtained by the system should be greater than or equal to the amount of electrical energy consumed. The devices that obtain electrical energy in the system include: hydrogen fuel cells, lithium batteries, and the devices that consume electrical energy include: DC motors, fans, communication equipment, and lithium batteries. The electrical balance is expressed by the following formula:

[0181]

[0182] Among them, E FC Represents the electrical energy provided by hydrogen fuel cells, Indicates the electrical energy output by the lithium battery, E M Indicates the electrical energy consumed by the DC motor, E F Indicates the electric energy consumed by the fan, E CD Indicates the power consumed by the communication equipment, Indicates the electrical energy input by the lithium battery.

[0183] Capacity constraints: For hydrogen hybrid drone systems, the operating output power of all types of equipment should not be greater than the equipment capacity at the time of design, and for hydrogen fuel cells, the minimum operating power should not be less than 30% of the equipment capacity to avoid operating in an underloaded state. Therefore, the capacity constraints for hydrogen fuel cells and fans are as follows:

[0184] Y FC，t ×0.3Cap FC ≤E FC ≤Y FC，t ×Cap FC (1.17)

[0185] 0≤E F ≤Y F，t ×Cap F (1.18)

[0186] Where Y FC，t Represents a Boolean variable used to determine whether the hydrogen fuel cell is running, with a value of 0 or 1;

[0187] Cap Fc Indicates the maximum output power of hydrogen fuel cells;

[0188] Y F，t Indicates a Boolean variable used to determine whether the fan is running, with a value of 0 or 1;

[0189] Cap F Indicates the maximum output power of the fan.

[0190] Energy storage constraints: The following assumptions and constraints apply to lithium batteries, i.e., electrical energy storage devices:

[0191] (1) Before the hydrogen hybrid drone takes off, the lithium battery has stored a certain amount of electricity;

[0192] (2) Since hydrogen hybrid drones have continuity when performing multiple flight missions, when the hydrogen fuel supply is sufficient, the lithium battery does not need to be recharged, so the lithium battery power at the end of the last flight is equal to the starting power at the next flight;

[0193] (3) The storage and use of lithium batteries do not occur at the same time and do not exceed the maximum storage and use power of the lithium battery.

[0194] (4) The storage energy of lithium batteries should be within the range of 20% to 80% based on the design capacity.

[0195] Therefore, lithium batteries need to meet the following requirements:

[0196] SOC0=SOC 360 (1.19)

[0197] 0≤SOCin ≤Y SOC，t ×P in，max (1.20)

[0198] 0≤SOC out ≤(1-Y sOC，t )×P out，max (1.21)

[0199] 0.2Cap soc ≤SOC t ≤0.8Cap soc (1.22)

[0200] Where Y SOC，t Represents a Boolean variable used to determine whether the lithium battery is charged or discharged, with a value of 0 or 1;

[0201] P in，max Indicates the maximum charging power of lithium battery;

[0202] P out，max Indicates the maximum discharge power of lithium battery;

[0203] Cap SOC Indicates the capacity of lithium battery.

[0204] The following assumptions and constraints apply to hydrogen storage tanks, i.e., hydrogen energy storage devices: (1) Before the hydrogen hybrid drone takes off, the hydrogen storage tank already has a certain amount of hydrogen energy; (2) Since the hydrogen hybrid drone has continuity when performing multiple flight missions, it does not need to be refilled when the hydrogen fuel is not less than 20% of the maximum capacity, so the amount of hydrogen in the hydrogen storage tank at the end of the last voyage is equal to the amount of hydrogen at the beginning of the next voyage; (3) The storage and use of the hydrogen storage tank do not occur at the same time. Since the filling process does not affect the capacity configuration of the hydrogen hybrid drone, more consideration is given to the use of the hydrogen storage tank, i.e., the use rate is not greater than the maximum use power of the hydrogen storage tank. Therefore, for the hydrogen storage tank:

[0205]

[0206] 0≤TP out ≤(1-Y TP，t )×H out，max (1.24)

[0207] Where Y TP，t A Boolean variable representing whether the hydrogen storage tank is releasing hydrogen, with a value of 0 or 1;

[0208] H out，max Indicates the maximum hydrogen discharge power of the hydrogen storage tank.

[0209] Step 2.2, construct a two-plan stochastic planning model for hydrogen hybrid UAV capacity configuration.

[0210] The variables mentioned in step 1 and step 2.1 include: design variables, i.e., equipment capacity, and operation variables, i.e., variables during the operation of the hydrogen hybrid drone. The MILP problem is constructed for the above variables:

[0211]

[0212] Among them, d is a continuous variable determined in the design process; x t and t are the continuous variables and 0-1 variables involved in the operation process, respectively, which exist in every time interval t during the project life cycle; f is the objective function, that is, the total cost of the system throughout its life cycle in formula 1.9; ψ dc It is an inequality constraint in the design process and is only related to d; and ψ oc are the equality and inequality constraints during operation, and d and x t and t All are related; They respectively represent the feasible domain of the determined continuous variables and the feasible domain of the continuous variables involved in the operation process; n represents the number of dimensions, that is, the number of 0-1 variables.

[0213] When considering the impact of uncertainty parameters on the optimization problem, the above problem becomes a typical two-stage resource problem. For deterministic MILP problems, design variables and operating variables can be determined and optimized at the same time; however, for stochastic optimization problems, design variables are independent of uncertainty parameters and can be determined first (here-and-now), while operating variables need to wait until all random parameters are actually presented to determine the optimal (wait-and-see), that is, the two types of variables need to be determined at different stages of the optimization problem. Therefore, when it is necessary to optimize the design variables and operating variables of the system at the same time to optimize the full-cycle cost, this study adopts a two-stage stochastic programming method. All design variables, that is, the capacity of all equipment, are classified as first-stage variables; all operating variables, that is, the output power of all equipment at each moment, the amount of energy purchased or sold to the outside world at each moment, and the amount of hydrogen stored in the hydrogen storage facility at each moment, are classified as second-stage variables. In combination with the above-mentioned MILP problem, the two-stage stochastic programming model constructed by the present invention can be expressed as an objective function in the following form:

[0214]

[0215] Among them, the above objective function is represented by two parts, f d is the deterministic part of the objective function, that is, the investment cost C corresponding to the selected equipment capacity inv , random part E[f s (d, ξ)] is given by f sThe expectation representation, f s Represents the cost associated with the uncertainty parameter, i.e., the operation and maintenance cost C OM and the cost of purchasing resources C tari ξ is the uncertainty parameter, including wind speed and load size.

[0216] In actual flight scenarios, wind speed and direction are random, volatile, and intermittent. The optimization of the total system cost by the deterministic model does not take these uncertain factors into account, resulting in a large deviation between the optimization results and the actual situation. Therefore, a two-stage stochastic programming model is constructed to achieve uncertainty optimization of the capacity configuration of hydrogen hybrid drones.

[0217] Step 3: Use the MILP solver in Matlab to solve the two-stage stochastic programming model and obtain the capacity configuration result of the hydrogen hybrid UAV;

[0218] Specifically, under the energy system constraints proposed in step 2, the two-stage stochastic programming model is solved in combination with the wind speed and direction data output by the wind uncertainty model. The obtained UAV capacity configuration results include the maximum power and rated power of the fuel cell, the battery capacity of the lithium battery, the heat dissipation power of the cooling system, and the hydrogen storage capacity of the hydrogen storage tank.

[0219] Specifically, in the solution process, the present invention proposes a two-stage random programming model using a particle swarm algorithm to optimize the equipment capacity of the first stage; the second stage is based on the random wind speed and wind direction scenarios of the Monte Carlo method, combined with the optimal equipment capacity of the first stage, to obtain the expected optimal cost of the system under different scenarios, which is used as the fitness function of the particle swarm algorithm to continuously evolve the optimization process of the first stage. CPLEX usually uses the branch-cutting plane method to solve the MILP model, and integrates multiple methods such as preprocessing and heuristics. Therefore, the CPLEX solver can be used to solve the second stage MILP problem. See the specific process. Figure 2 .

[0220] This process is based on solving mixed integer linear programming problems and particle swarm algorithm, using Monte Carlo simulation to generate random scenarios of the system and taking into account the uncertainty in the system. It mainly includes particle swarm generation, fitness calculation, particle position update, particle velocity update, out-of-bounds judgment and termination judgment.

[0221] like Figure 2 As shown, the specific calculation steps in step 3 are as follows:

[0222] Step 1: Initial particle swarm generation: define the spatial range in which particles can move, and randomly generate particle position vectors and velocity vectors within the range.

[0223] Step 2: Fitness calculation: For each individual in the particle swarm, substitute each scenario generated by Monte Carlo to obtain the expected value under each scenario, that is, the fitness of the individual. For some situations where there is no solution or the solution fails, a larger value can be added to the fitness function to reduce the fitness of the individual, so as to complete further screening.

[0224] Step 3: Update the global optimal particle and the individual optimal particle: According to the size of the fitness function, select the particle with the best fitness in the current individual and the global optimal particle.

[0225] Step 4: Out-of-bounds determination: Verify the speed and position of the updated particle range, and re-randomly generate the position and speed of the particles that exceed the range.

[0226] Step5: Repeat Step2-Step4 until the iteration condition is met, exit the loop, and obtain the optimal solution.

[0227] Furthermore, in step 3, the MILP solver in Matlab is used to solve the capacity configuration result of the hydrogen hybrid UAV.

[0228] Furthermore, in a possible embodiment, the capacity configuration of the hydrogen hybrid UAV is solved using the MILP solver in Matlab, and the results are shown in Table 3 below.

[0229] Table 3: Capacity configuration results

[0230]

[0231] like Figure 3 As shown, the present invention also proposes a hydrogen hybrid UAV capacity configuration system based on two-stage optimization, which is used to implement the above-mentioned hydrogen hybrid UAV capacity configuration method based on two-stage optimization. The system includes: a UAV related parameter construction module, a random optimization model construction module and a capacity configuration solution module;

[0232] Specifically, the UAV-related parameter construction module is used to construct the hydrogen hybrid UAV energy system model, wind uncertainty model, multi-time scale comprehensive evaluation index and corresponding objective function;

[0233] The stochastic optimization model building module is used to build a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration taking into account uncertain factors based on the hydrogen hybrid UAV energy system model, comprehensive evaluation indicators and corresponding objective functions;

[0234] The capacity configuration solution module is used to solve the two-stage stochastic programming model in combination with the calculation results of the wind uncertainty model to obtain the capacity configuration results of the hydrogen hybrid UAV.

[0235] The beneficial effect of the present invention is that, compared with the prior art, the present invention aims to configure a hydrogen hybrid drone by establishing a hydrogen hybrid drone energy system model including hydrogen fuel cells and lithium batteries, and configuring the capacity of the model based on a two-stage optimal capacity configuration. The two-stage optimal capacity configuration method takes into account uncertain factors such as the environmental wind field and the gravity change of the hydrogen hybrid drone during navigation, and considers power constraints and energy constraints. It uses a mixed integer linear programming algorithm, takes the system comprehensive evaluation index as the optimization target, and finds the optimal capacity configuration under the optimization target in a random load fluctuation scenario. The research of the present invention provides a theoretical basis for the two-stage optimal capacity configuration method in the field of hydrogen hybrid drones.

[0236] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0237] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive lists) of computer readable storage medium include: portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), static random access memories (SRAM), portable compact disk read-only memories (CD-ROM), digital versatile disks (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or convex structures in grooves on which instructions are stored, and any suitable combination thereof. Computer readable storage medium used here is not interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagated by waveguides or other transmission media (e.g., light pulses by optical fiber cables), or electrical signals transmitted by wires.

[0238] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0239] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages-such as Smalltalk, C++, etc., and conventional procedural programming languages-such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network-including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA) may be personalized by utilizing the state information of a computer-readable program instruction, and the electronic circuit may execute a computer-readable program instruction, thereby realizing various aspects of the present disclosure.

[0240] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for hydrogen hybrid UAV capacity configuration based on two-stage optimization, characterized in that: The steps include: Step 1: Construct the energy system model, wind uncertainty model, multi-time scale comprehensive evaluation index and corresponding objective function of hydrogen hybrid UAV; Step 2: Based on the hydrogen hybrid UAV energy system model, comprehensive evaluation indicators and corresponding objective functions, a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration considering uncertain factors is constructed; Step 3: Solve the two-stage stochastic programming model in combination with the wind uncertainty model to obtain the capacity configuration result of the hydrogen hybrid drone.

2. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 1 is characterized in that: The hydrogen hybrid UAV energy system mainly includes a hydrogen fuel cell, a lithium battery, a hydrogen storage tank, and an engine.

3. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 1 is characterized in that: The constructed wind uncertainty model is as follows: In the formula, -ΔW G，j (r, t+τ|t) represents the wind uncertainty prediction value based on time t at position vector r and forecast look-ahead time t+τ; It represents the time-delayed RUC data error at the forecast look-ahead time τ based on the position vector r at time t; They represent the initial measurement error and nominal wind error respectively.

4. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 3 is characterized in that: The step 1 also includes: The RUC data of wind speed and direction are collected and processed by interpolation method, and the processed data are used as the data source of wind uncertainty model; Among them, the preprocessing of RUC data by interpolation method specifically includes: The constructed wind uncertainty model is optimized by combining spline interpolation with inverse distance weighted interpolation: first, the inverse distance weighted interpolation method is used to reduce the data to a scale of 1 km, and then the cubic spline interpolation method is used to reduce the data after inverse distance weighted interpolation to a horizontal resolution of 5 m.

5. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 1 is characterized in that: The constructed multi-time scale comprehensive evaluation index includes short-term cost and long-term cost, and the corresponding objective function is as follows: minJ=ω1C short +ω2C long In the formula, J is the target value of comprehensive optimization, C short is the short-term cost, including fuel consumption and maintenance costs; C long is the long-term cost, including equipment investment and service life discount; ω1 and ω2 are the weight coefficients of short-term cost and long-term cost, respectively, which are used to balance the relative importance of short-term and long-term goals.

6. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 5 is characterized in that: Short-term costs short The calculation formula is as follows: C short =k1C f +k2C m In the formula, C f represents the hydrogen fuel consumption cost, C m represents the regular maintenance cost, k1 and k2 represent the weight coefficients of hydrogen fuel consumption cost and regular maintenance cost respectively; Long term cost C long The calculation formula is as follows: C long =k3C e +k4C l In the formula, C e Indicates the one-time investment cost of the equipment, C l represents the service life conversion, k3 and k4 represent the weight coefficients of the one-time investment cost and service life conversion of the equipment respectively.

7. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 1 is characterized in that: A two-stage stochastic optimization problem for hydrogen hybrid UAV capacity configuration is constructed, which includes: Step 2.1, construct the constraints of the energy system of the hydrogen-powered hybrid UAV, including energy balance constraints, capacity constraints, and energy storage constraints; Step 2.2, construct a two-stage stochastic planning model for hydrogen hybrid drone capacity configuration.

8. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 7 is characterized in that: The energy balance constraints include: the electrical energy obtained in the hydrogen hybrid UAV energy system should be greater than or equal to the electrical energy consumption to maintain electrical balance: In the formula, E FC Represents the electrical energy provided by hydrogen fuel cells, Indicates the electrical energy output by the lithium battery, E M Indicates the electrical energy consumed by the DC motor, E F Indicates the electric energy consumed by the fan, E CD Indicates the power consumed by the communication equipment, Indicates the electrical energy input by the lithium battery; The capacity constraints include: the operating output power of various equipment in the hydrogen hybrid drone system should not be greater than the equipment capacity at the time of design, and the minimum operating power of the hydrogen fuel cell should not be less than 30% of the equipment capacity to avoid running in an underload state. Therefore, the capacity constraints for hydrogen fuel cells and fans are as follows: <h2 style=";text-align:left;direction:ltr">Y<h2 style=";text-align:left;direction:ltr"> FC，t <h2 style=";text-align:left;direction:ltr"> ×0.3 Cap<h2 style=";text-align:left;direction:ltr"> FC <h2 style=";text-align:left;direction:ltr"> ≤E<h2 style=";text-align:left;direction:ltr"> FC <h2 style=";text-align:left;direction:ltr"> ≤Y<h2 style=";text-align:left;direction:ltr"> FC，t <h2 style=";text-align:left;direction:ltr"> ×Cap<h2 style=";text-align:left;direction:ltr"> FC 0≤E F ≤Y F，t ×Cap F In the formula, represents a Boolean variable used to determine whether the hydrogen fuel cell is running, and its value is 0 or 1; Cap Fc Indicates the maximum output power of hydrogen fuel cells; Y F,t Indicates a Boolean variable used to determine whether the fan is running, with a value of 0 or 1; Cap F Indicates the maximum output power of the fan; The energy storage constraints include electric energy storage device constraints and hydrogen energy storage device constraints; The energy storage device is a lithium battery, which meets the following requirements: the lithium battery has stored energy before the drone takes off; when the hydrogen fuel is fully supplied, the lithium battery does not need to be recharged, and the lithium battery power at the end of the previous flight is equal to the starting power at the next flight; the storage and use of the lithium battery do not occur at the same time and are not greater than the maximum storage and use power of the lithium battery; the lithium battery storage energy is within the range of 20% to 80% based on the design capacity; The hydrogen energy storage device is a hydrogen storage tank, which meets the following requirements: before the UAV takes off, the hydrogen storage tank has stored hydrogen energy; no refueling is performed when the hydrogen fuel is not less than 20% of the maximum capacity, and the amount of hydrogen in the hydrogen storage tank at the end of the previous voyage is equal to the amount of hydrogen at the beginning of the next voyage; the storage and use of the hydrogen storage tank do not occur at the same time, and the use rate of the hydrogen storage tank is not greater than the maximum use power of the hydrogen storage tank.

9. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 7 is characterized in that: A two-stage stochastic planning model is constructed by combining the equipment capacity and the operating variables during the operation of the hydrogen hybrid drone. The capacity of all equipment is the first-stage variable; the operating variables are the second-stage variables, including the output power of all equipment at each moment, the amount of electricity stored in the power storage facility at each moment, and the amount of hydrogen stored in the hydrogen storage facility at each moment. The constructed two-stage stochastic planning model is as follows: In the formula, f d is the deterministic part of the objective function, that is, the investment cost C corresponding to the selected equipment capacity inv , the random part is composed of f s The expectation representation, f s represents the cost associated with the uncertainty parameter, including operation and maintenance costs C OM and the cost of purchasing resources C tari ; d is a continuous variable determined in the design process, ξ is an uncertain parameter, including wind speed and load size; ψ dc To be the inequality constraints in the design process; and ψ oc are the equality and inequality constraints during operation respectively; x t and t They are continuous variables and 0-1 variables involved in the operation process, which exist in every time interval t during the project life cycle; Points represent the feasible domain of continuous variables involved in the operation process; n represents the number of 0-1 variables.

10. The hydrogen hybrid UAV capacity configuration method based on two-stage optimization according to claim 1 is characterized in that: The step 3 specifically includes: The two-stage stochastic programming model is solved by combining the wind speed and direction data output by the wind uncertainty model. The obtained UAV capacity configuration results include the maximum power and rated power of the fuel cell, the battery capacity of the lithium battery, the heat dissipation power of the cooling system, and the hydrogen storage capacity of the hydrogen storage tank.

11. A hydrogen hybrid UAV capacity configuration system based on two-stage optimization, used to implement the hydrogen hybrid UAV capacity configuration method based on two-stage optimization as described in any one of claims 1 to 10, characterized in that: include: UAV related parameter construction module, random optimization model construction module and capacity configuration solution module; The UAV-related parameter construction module is used to construct the hydrogen hybrid UAV energy system model, wind uncertainty model, multi-time scale comprehensive evaluation index and corresponding objective function; The stochastic optimization model building module is used to build a two-stage stochastic optimization model for hydrogen hybrid UAV capacity configuration taking into account uncertain factors based on the hydrogen hybrid UAV energy system model, comprehensive evaluation indicators and corresponding objective functions; The capacity configuration solution module is used to solve the two-stage stochastic programming model in combination with the calculation results of the wind uncertainty model to obtain the capacity configuration results of the hydrogen hybrid UAV.

12. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images