Solar energy and hydrogen energy hybrid power hovercar energy management method based on solar energy distribution map

Abstract

The invention provides a solar energy distribution map-based solar energy and hydrogen energy hybrid power hovercar energy management method, and belongs to the field of solar energy and hydrogen energy hybrid power hovercar energy management, and the method comprises the following steps: S1, predicting solar energy radiation intensity distribution by using a BP neural network; s2, determining the track of the hovercar by using an Astar algorithm improved based on solar radiation; s3, establishing a trajectory optimization and energy management coupling model; and S4, solving the coupling model by using a sequential quadratic programming algorithm. By reasonably distributing the power of each power source of the hybrid power hovercar, the charge state of the power battery of the hydrogen fuel cell hovercar is kept balanced, the hydrogen energy consumption is reduced, and the drivability and riding comfort of the hydrogen fuel cell hovercar are improved.

Description

Technical Field

[0001] The invention relates to the technical field of energy management of solar-hydrogen hybrid flying cars, and in particular to an energy management method of solar-hydrogen hybrid flying cars based on a solar energy distribution map. Background Art

[0002] Against the backdrop of an increasingly severe energy crisis, solar energy and hydrogen energy, as two clean and renewable energy sources, have shown great potential. Solar energy is directly converted into electricity through sunlight, and hydrogen energy is generated through water electrolysis and other methods. Both do not rely on fossil fuels and can significantly reduce greenhouse gas emissions. The combination of solar energy and hydrogen energy can not only provide a stable and reliable energy supply, but also reduce dependence on traditional energy sources such as oil, and improve energy independence and security. This combination provides a viable way to cope with the energy crisis and promote the world towards a more sustainable and environmentally friendly future. The combination of solar energy and hydrogen fuel cell flying cars represents the development direction of future transportation. Solar panels can collect solar energy during flight or when parked, and convert it into electricity to drive the car or charge the hydrogen fuel cell. Hydrogen fuel cells provide high energy density and long endurance, ensuring that flying cars can still operate efficiently at night or on cloudy days when there is insufficient solar energy. This energy combination not only significantly reduces carbon emissions and environmental pollution, but also improves energy efficiency, reduces dependence on fossil fuels, and lays the foundation for building a more sustainable and environmentally friendly transportation system.

[0003] However, most of the existing flying cars rely on traditional mechanical or electrical power, and lack the development of solar energy and hydrogen flying cars, which makes it difficult to improve their operating efficiency and sustainability, and also cannot reduce energy consumption and environmental impact, and thus it is difficult to provide strong support for the development of green transportation. Therefore, a solar hydrogen hybrid flying car energy management method based on solar energy distribution map is proposed to solve the above problems. Summary of the invention

[0004] In view of this, an object of the present invention is to provide an energy management method for a solar-hydrogen hybrid flying car based on a solar energy distribution map, so as to at least solve the above problems.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A method for energy management of a solar-hydrogen hybrid flying car based on a solar energy distribution map, the method comprising the following steps:

[0007] S1, predicting the solar radiation intensity distribution by using BP neural network;

[0008] S2, determining the trajectory of the flying car by using the improved Astar algorithm based on solar radiation;

[0009] S3, establish a coupled model of trajectory optimization and energy management;

[0010] S4. Use the sequential quadratic programming algorithm to solve the coupled model.

[0011] Furthermore, in step S1, the solar radiation intensity distribution is predicted by using the BP neural network as follows:

[0012] By dividing the BP neural network into l layers, and setting the output of any node l to O i , for the mth input, the output of the i-th node in the (l-1)th layer is f (l) () represents the activation function, and the connection weight between the l-th node in the l-1th layer and the j-th node in the l-1th layer is The number of nodes in the first layer is d l-1 , the output is Then the first layer input of the jth node is Then the output of the jth node in the i-th layer is The specific formula is as follows:

[0013]

[0014] Furthermore, the input parameters of the BP neural network are time, flying car accuracy, latitude, pressure, humidity, and solar radiation parameters, and the output parameter of the BP neural network is solar radiation intensity.

[0015] Furthermore, in step S2, the following steps are also included:

[0016] S21, gridding the area from the starting point to the end point of the flying car;

[0017] S22, define OPEN and CLOSED lists, initialize them to empty;

[0018] S23, starting point A is assigned to the OPEN list, and obstacles are assigned to the CLOSED list;

[0019] S24, calculate the f(s) value of the node in the OPEN list;

[0020] S25, find the node with the minimum f(s) value, add it as the parent node to the CLOSED list, and add its adjacent nodes to the OPEN list;

[0021] S26, determine whether the CLOSED list contains the target node, if yes, execute step S27, if no, execute step S24;

[0022] S27. Calculate the optimal path from the starting point to the target location in the CLOSED list to complete the path planning.

[0023] Further, in step S25, the node with the minimum f(s) value is found, and it is added to the CLOSED list as a parent node, and its adjacent nodes are added to the OPEN list. Specifically:

[0024] The calculation is done by the following formula:

[0025] f(s)=g(s)+h(s)+λ sr ·e(s)

[0026] Where: f(s) is the estimated cost from the initial state via state s to the target state, g(s) is the actual cost from the initial state to state s in the state space, h(s) is the estimated cost of the best path from state n to the target state, e(s) is the solar radiation cost from the initial state to state s in the state space, specifically expressed as the difference in radiation intensity, λ sr The value represents the radiation equivalence factor, which indicates whether the path planning tends to rely on the solar radiation value. The h(s) formula uses the Euclidean distance as the cost of moving between two nodes:

[0027]

[0028] e(s)=e(i)-e(j)

[0029] where x i (i=1,2) is the horizontal coordinate of node i, y i (i=1,2) is the ordinate of node i, e(i) is the radiation intensity of node i, and e(j) is the radiation intensity of node j.

[0030] Furthermore, in step S3, the trajectory optimization and energy management coupling model is established as follows:

[0031] By building a solar energy, hydrogen energy, and battery hybrid system and setting the hybrid system state variables to The control variable is set to u = [T, n v ,n h ,P fc ,P b ], the equation of state as follows

[0032]

[0033] Among them, the three-dimensional position of the flying car is defined by the following formula (r x ,r y ,r h), the flight speed is defined by the following formula V, the flight path angle is defined by the following formula γ represents the angle between the flight speed and the altitude. The heading angle is defined by the following formula χ, represents the flight direction of the flying car, the thrust of the flying car during flight is T, and the vertical load n v and horizontal load n h , represents the load factor in the direction corresponding to the flight path, m represents the mass of the flying car and S represents the reference wing area, the coefficient k e λ represents the parasitic drag coefficient and aerodynamic coefficient, g and ρ represent the gravitational acceleration and atmospheric density respectively. pv is the angle between the sun's rays and the normal vector of the flying car, S pv is the PV area, Q irr is the solar irradiance in the environment, η pv is the photovoltaic conversion efficiency from solar energy to electrical energy,

[0034] Minimizing the energy consumption of the hybrid system and transferring it to optimizing the FC hydrogen consumption and the battery equivalent hydrogen consumption, the loss function is:

[0035]

[0036] where λ eq is the battery equivalent hydrogen consumption coefficient. The battery power consumption can be calculated by λ eq Converted into hydrogen consumption, λ eq According to the dynamic change of battery power SOC, PID controller is used to control λ eq The change of t f The final flight time, trajectory optimization and energy management coupling model is:

[0037] Problem Ι:minJ

[0038]

[0039] Where P D is the driving force behind the demand for flying cars. b , P fc , P pv are the battery output power, fuel cell output power and solar panel output power, η DC and η MPPT They are DC / DC conversion efficiency and maximum power point tracker conversion efficiency respectively.

[0040] Furthermore, in step S4, the sequential quadratic programming algorithm is used to solve the coupling model as follows:

[0041] Set to minimize an objective function J while satisfying the constraints in is the optimization variable,

[0042] (1) By constructing the Lagrangian function: The Lagrangian function is defined as follows:

[0043]

[0044] where λ and v are Lagrange multipliers used to handle the inequality constraints g i (x)≤0 and the equality constraint h j (x) = 0;

[0045] (2) Calculate the gradient and Hessian matrix: Calculate the gradient of the objective function J And the constraint function g i (x) and h j The gradient of (x) and their Hessian matrices

[0046] (3) Construct a quadratic programming subproblem: At the current iteration point x k , construct the quadratic programming subproblem:

[0047]

[0048] (4) Solve the quadratic programming problem: By solving the above quadratic programming problem, we can obtain the search direction p k ;

[0049] (5) Update the iteration point: calculate the next iteration point x k+1 =x k +p k , and update the Lagrange multiplier λ,v; (6) Iteration: Repeat steps 2 to 5 until the convergence criterion is met, that is, the objective function changes very little.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention proposes an energy management method for a solar-hydrogen hybrid flying car based on a solar energy distribution map. The combination of trajectory optimization and energy management of a solar-hydrogen flying car can significantly improve its operating efficiency and sustainability. Through advanced trajectory optimization technology, the flying car can choose the most energy-efficient route and maximize the use of solar energy and hydrogen energy resources. At the same time, the intelligent energy management system can monitor and adjust the energy distribution of solar panels and hydrogen fuel cells in real time to ensure optimal energy utilization under different flight conditions. This comprehensive optimization not only improves the endurance and reliability of the flying car, but also reduces energy consumption and environmental impact, providing strong support for the development of green transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 It is a schematic diagram of the overall process proposed in the embodiment of the present invention.

[0054] Figure 2 It is a schematic diagram of the overall structure of using BP neural network to predict solar radiation intensity distribution in an embodiment of the present invention.

[0055] Figure 3 It is a schematic diagram of the overall structure of determining the trajectory of a flying car using the improved Astar algorithm based on solar radiation according to an embodiment of the present invention.

[0056] Figure 4 It is a schematic diagram of the sub-step flow of step S2 of an embodiment of the present invention.

[0057] Figure 5 It is a schematic diagram of a trajectory optimization and energy management coupling model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The enumerated embodiments are only used to explain the present invention but not to limit the scope of the present invention.

[0059] Reference Figures 1 to 4 The present invention provides an energy management method for a solar-hydrogen hybrid flying car based on a solar energy distribution map, the method comprising the following steps:

[0060] S1, predicting the solar radiation intensity distribution by using BP neural network;

[0061] S2, determining the trajectory of the flying car by using the improved Astar algorithm based on solar radiation;

[0062] S3, establish a coupled model of trajectory optimization and energy management;

[0063] S4. Use the sequential quadratic programming algorithm to solve the coupled model.

[0064] In step S1, the solar radiation intensity distribution is predicted by using the BP neural network as follows:

[0065] By dividing the BP neural network into l layers, and setting the output of any node l to O i , for the mth input, the output of the i-th node in the (l-1)th layer is f (l) () represents the activation function, and the connection weight between the l-th node in the l-1th layer and the j-th node in the l-1th layer is The number of nodes in the first layer is d l-1 , the output is Then the first layer input of the jth node is Then the output of the jth node in the i-th layer is The specific formula is as follows:

[0066]

[0067] The input parameters of the BP neural network are time, flying car accuracy, latitude, pressure, humidity, solar radiation and other parameters, and the output parameter of the BP neural network is the solar radiation intensity.

[0068] In step S2, the following steps are also included:

[0069] S21, gridding the area from the starting point to the end point of the flying car;

[0070] S22, define OPEN and CLOSED lists, initialize them to empty;

[0071] S23, starting point A is assigned to the OPEN list, and obstacles are assigned to the CLOSED list;

[0072] S24, calculate the f(s) value of the node in the OPEN list;

[0073] S25, find the node with the minimum f(s) value, add it as the parent node to the CLOSED list, and add its adjacent nodes to the OPEN list;

[0074] S26, determine whether the CLOSED list contains the target node, if yes, execute step S27, if no, execute step S24;

[0075] S27. Calculate the optimal path from the starting point to the target location in the CLOSED list to complete the path planning.

[0076] In step S25, find the node with the minimum f(s) value, add it as the parent node to the CLOSED list, and add its adjacent nodes to the OPEN list. Specifically:

[0077] The Astar algorithm process improved based on solar radiation intensity is as follows Figure 2As shown in the figure, the algorithm starts from the starting point and expands to the surrounding areas. The cost value of each surrounding node is calculated by the evaluation function. The node with the minimum cost is selected as the next expansion node. This process is repeated until the target point is reached to generate the final path. During the search process, since each node on the path has the minimum cost, the total cost of the path is the minimum. The evaluation function f(n) of the Astar algorithm is expressed as:

[0078] f(s)=g(s)+h(s)+λ sr ·e(s)

[0079] Where: f(s) is the estimated cost from the initial state via state s to the target state, g(s) is the actual cost from the initial state to state s in the state space, h(s) is the estimated cost of the best path from state n to the target state, e(s) is the solar radiation cost from the initial state to state s in the state space, specifically expressed as the difference in radiation intensity, λ sr The value represents the radiation equivalence factor, which indicates whether the path planning tends to rely on the solar radiation value. The h(s) formula uses the Euclidean distance as the cost of moving between two nodes:

[0080]

[0081] e(s)=e(i)-e(j)

[0082] where x i (i=1,2) is the horizontal coordinate of node i, y i (i=1,2) is the ordinate of node i, e(i) is the radiation intensity of node i, and e(j) is the radiation intensity of node j.

[0083] In step S3, the trajectory optimization and energy management coupling model is established as follows:

[0084] By building a solar energy, hydrogen energy, and battery hybrid system and setting the hybrid system state variables to The control variable is set to u = [T, n v ,n h ,P fc ,P b ], the equation of state as follows

[0085]

[0086] Among them, the three-dimensional position of the flying car is defined by the following formula (r x ,r y ,r h), the flight speed is defined by the following formula V, the flight path angle is defined by the following formula γ represents the angle between the flight speed and the altitude. The heading angle is defined by the following formula X, represents the flight direction of the flying car, the thrust of the flying car during flight is T, and the vertical load n v and horizontal load n h , represents the load factor in the direction corresponding to the flight path, m represents the mass of the flying car and S represents the reference wing area, the coefficient k e λ represents the parasitic drag coefficient and aerodynamic coefficient, g and ρ represent the gravitational acceleration and atmospheric density respectively. pv is the angle between the sun's rays and the normal vector of the flying car, S [v is the PV area, Q irr is the solar irradiance in the environment, η pv is the photovoltaic conversion efficiency from solar energy to electrical energy,

[0087] Minimizing the energy consumption of the hybrid system and transferring it to optimizing the FC hydrogen consumption and the battery equivalent hydrogen consumption, the loss function is:

[0088]

[0089] where λ eq is the battery equivalent hydrogen consumption coefficient. The battery power consumption can be calculated by λ eq Converted into hydrogen consumption, λ eq According to the dynamic change of battery power SOC, PID controller is used to control λ eq The change of t f The final flight time, trajectory optimization and energy management coupling model is:

[0090]

[0091] Where P D is the driving force behind the demand for flying cars. b , P fc , P pv are the battery output power, fuel cell output power and solar panel output power, η DC and η MPPT They are DC / DC conversion efficiency and maximum power point tracker conversion efficiency respectively.

[0092] In step S4, the sequential quadratic programming algorithm is used to solve the coupling model as follows:

[0093] Set to minimize an objective function J while satisfying the constraint g i (x)≤0,h j (x) = 0, where is the optimization variable,

[0094] (1) By constructing the Lagrangian function: The Lagrangian function is defined as follows:

[0095]

[0096] where λ and v are Lagrange multipliers used to handle the inequality constraints g i (x)≤0 and the equality constraint h j (x) = 0;

[0097] (2) Calculate the gradient and Hessian matrix: Calculate the gradient of the objective function J And the constraint function g i (x) and h j The gradient of (x) and their Hessian matrices

[0098] (3) Construct a quadratic programming subproblem: At the current iteration point x k , construct the quadratic programming subproblem:

[0099]

[0100]

[0101] (4) Solve the quadratic programming problem: By solving the above quadratic programming problem, we can obtain the search direction p k ;

[0102] (5) Update the iteration point: calculate the next iteration point x k+1 =x k +p k , and update the Lagrange multiplier λ,v;

[0103] (6) Iteration: Repeat steps 2 to 5 until the convergence criterion is met, that is, the objective function changes very little.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A solar-hydrogen hybrid flying car energy management method based on solar energy distribution map, characterized in that: The method comprises the following steps: S1, predicting the solar radiation intensity distribution by using BP neural network; S2, determining the trajectory of the flying car by using the improved Astar algorithm based on solar radiation; S3, establish a coupled model of trajectory optimization and energy management; S4. Use the sequential quadratic programming algorithm to solve the coupled model.

2. The energy management method of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 1 is characterized in that: In step S1, the solar radiation intensity distribution is predicted by using the BP neural network as follows: By dividing the BP neural network into l layers, and setting the output of any node l to O i , for the mth input, the output of the i-th node in the (l-1)th layer is f (l) () represents the activation function, and the connection weight between the l-th node in the l-1th layer and the j-th node in the l-1th layer is The number of nodes in the first layer is d l-1 , the output is Then the first layer input of the jth node is Then the output of the jth node in the i-th layer is The specific formula is as follows:

3. The energy management method of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 2 is characterized in that: The input parameters of the BP neural network are time, flying car accuracy, latitude, pressure, humidity, and solar radiation parameters, and the output parameter of the BP neural network is solar radiation intensity.

4. The energy management method of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 3 is characterized in that: In step S2, the following steps are also included: S21, gridding the area from the starting point to the end point of the flying car; S22, define OPEN and CLOSED lists, initialize them to empty; S23, starting point A is assigned to the OPEN list, and obstacles are assigned to the CLOSED list; S24, calculate the f(s) value of the node in the OPEN list; S25, find the node with the minimum f(s) value, add it as the parent node to the CLOSED list, and add its adjacent nodes to the OPEN list; S26, determine whether the CLOSED list contains the target node, if yes, execute step S27, if no, execute step S24; S27. Calculate the optimal path from the starting point to the target location in the CLOSED list to complete the path planning.

5. The energy management method of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 4 is characterized in that: In step S25, find the node with the minimum f(s) value, add it as the parent node to the CLOSED list, and add its adjacent nodes to the OPEN list. Specifically: The calculation is done by the following formula: f(s)=g(s)+h(s)+λ sr ·e(s) Where: f(s) is the estimated cost from the initial state via state s to the target state, g(s) is the actual cost from the initial state to state s in the state space, h(s) is the estimated cost of the best path from state n to the target state, e(s) is the solar radiation cost from the initial state to state s in the state space, specifically expressed as the difference in radiation intensity, λ sr The value represents the radiation equivalence factor, which indicates whether the path planning tends to rely on the solar radiation value. The h(s) formula uses the Euclidean distance as the cost of moving between two nodes: e(s)=e(i)-e(j) where x i (i=1,2) is the horizontal coordinate of node i, y i (i=1,2) is the ordinate of node i, e(i) is the radiation intensity of node i, and e(j) is the radiation intensity of node j.

6. The method for energy management of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 5, characterized in that: In step S3, the trajectory optimization and energy management coupling model is established as follows: By building a solar energy, hydrogen energy, and battery hybrid system and setting the hybrid system state variables to The control variable is set to u = [T, n v ,nh,P fc ,P b ], the equation of state as follows Among them, the three-dimensional position of the flying car is defined by the following formula (r x ,r y ,rh), the flight speed is defined by the following formula V, the flight path angle is defined by the following formula γ represents the angle between the flight speed and the altitude. The heading angle is defined by the following formula χ, represents the flight direction of the flying car, the thrust of the flying car during flight is T, and the vertical load n v and horizontal load nh, which represents the load factor in the direction corresponding to the flight path, m represents the mass of the flying car and S represents the reference wing area, the coefficient k e λ represents the parasitic drag coefficient and aerodynamic coefficient, g and ρ represent the gravitational acceleration and atmospheric density respectively. pv is the angle between the sun's rays and the normal vector of the flying car, S pv is the PV area, Q irr is the solar irradiance in the environment, η pv is the photovoltaic conversion efficiency from solar energy to electrical energy, Minimizing the energy consumption of the hybrid system and transferring it to optimizing the FC hydrogen consumption and the battery equivalent hydrogen consumption, the loss function is: where λ eq is the battery equivalent hydrogen consumption coefficient. The battery power consumption can be calculated by λ eq Converted into hydrogen consumption, λ eq According to the dynamic change of battery power SOC, PID controller is used to control λ eq The change of t f The final flight time, trajectory optimization and energy management coupling model is: Where P D is the driving force behind the demand for flying cars. b , P fc , P pv are the battery output power, fuel cell output power and solar panel output power, η DC and η MPPT They are DC / DC conversion efficiency and maximum power point tracker conversion efficiency respectively.

7. The method for energy management of a solar-hydrogen hybrid flying car based on a solar energy distribution map according to claim 6, characterized in that: In step S4, the sequential quadratic programming algorithm is used to solve the coupling model as follows: Set to minimize an objective function J while satisfying the constraint g i (x)≤0,h j (x) = 0, where is the optimization variable, (1) By constructing the Lagrangian function: The Lagrangian function is defined as follows: where λ and v are Lagrange multipliers used to handle the inequality constraints g i (x)≤0 and the equality constraint h j (x) = 0; (2) Calculate the gradient and Hessian matrix: Calculate the gradient of the objective function J And the constraint function g i (x) and h j (x) Gradient and their Hessian matrices (3) Construct a quadratic programming subproblem: At the current iteration point x k , construct the quadratic programming subproblem: (4) Solve the quadratic programming problem: By solving the above quadratic programming problem, we can obtain the search direction p k ; (5) Update the iteration point: calculate the next iteration point x k+1 =x k +p k , and update the Lagrange multiplier λ,v; (6) Iteration: Repeat steps 2 to 5 until the convergence criterion is met, that is, the objective function changes very little.

Images