Working umbrella control method for high-altitude wind power

Abstract

The invention relates to a working umbrella control method for high-altitude wind power, which comprises the following steps: when working umbrellas reach the lowest umbrella opening height, the working umbrellas are sequentially opened from top to bottom, and the umbrella opening opportunity and the umbrella opening number are adaptively adjusted; after all the acting umbrellas are opened, entering the next stage; a work doing umbrella control model is constructed and trained, the operation angle of the work doing umbrella is controlled, and after the work doing umbrella enters the set height range, the next stage is started; the acting umbrellas are sequentially closed from bottom to top, and the umbrella closing opportunity and the number of the closed umbrellas are adjusted in a self-adaptive mode till all the acting umbrellas are closed. The probability that the cable structure is damaged due to sudden increase of pulling force in the parachute opening process is greatly reduced, the service life of the cable is prolonged, and the output power in each power generation cycle is improved; the wind power captured by the acting umbrella in the acting stage can reach the maximum value, so that the output power of the high-altitude wind power device can reach the maximum value; and it is ensured that part of wind energy can be converted into electric energy in the umbrella closing stage, and the power generation efficiency in each power generation cycle is improved.

Description

Technical Field

[0001] The invention relates to the technical field of high-altitude wind power generation, in particular to a working parachute control method for high-altitude wind power generation. Background Art

[0002] High-altitude wind power is a technology that uses lightweight aircraft to collect high-altitude high-quality wind energy for power generation. Tethered aircraft such as kites and light airships drive cables and drag generators to generate electricity. The process of high-altitude wind power is roughly divided into two parts. The wind drives the aircraft to rise, and when the cable reaches the maximum length, the cable or the operating status of the aircraft is adjusted to control the aircraft to descend, and then enter the next cycle again.

[0003] The use of parachutes is also a common form of aircraft for high-altitude wind power generation. Parachutes are used to capture high-altitude wind energy to drive the cables up, and then drag the generator to generate electricity. In engineering power generation, it is usually necessary to control the aircraft to repeat the power generation cycle. Affected by environmental factors, there are differences in the optimal parachute opening and closing phases during each power generation cycle. Existing parachute ladder high-altitude wind power systems usually have fixed parachute opening and closing times, and all working parachutes are opened or closed in sequence at the same time, and the adaptability to different environments is insufficient. Therefore, it is of great significance to adjust the operating status of high-altitude wind power aircraft according to environmental conditions.

[0004] Therefore, there is an urgent need for a high-altitude wind power device that can adaptively adjust the operating state of the high-altitude aircraft according to environmental changes to improve the power generation efficiency to solve the above problems. Summary of the invention

[0005] In order to solve the problem that the prior art cannot adjust the operating state of a parachute-ladder aircraft according to the environment, the purpose of the present invention is to provide a working parachute control method for high-altitude wind power, which can adaptively adjust the power generation cycle according to the environment, quickly adjust the operating state of the aircraft according to the environment, and improve the power generation efficiency of the system.

[0006] To achieve the above object, the present invention adopts the following technical solution: a working parachute control method for high-altitude wind power, the method comprising the following steps in sequence:

[0007] (1) Adaptive parachute opening stage: After the parachute ladder type high-altitude wind turbine starts to operate, when the working parachute reaches the minimum parachute opening height, the working parachute opens from top to bottom in sequence, and the controller adaptively adjusts the opening time and number of parachutes according to the environmental conditions; when all the working parachutes are opened, it enters the next stage;

[0008] (2) Maximum power working stage: construct and train a working parachute control model, and control the operating angle of the working parachute through the trained working parachute control model. When the working parachute enters the set altitude range, it enters the next stage; the set altitude range is 2300 meters to 2600 meters;

[0009] (3) Adaptive parachute closing stage: The working parachutes are closed from bottom to top in sequence, and the controller adaptively adjusts the closing timing and number of parachutes according to environmental conditions until all the working parachutes are closed.

[0010] The step (1) specifically refers to: launching the lift balloon, collecting the working parachute operation data in real time, the working parachute operation data including the height, speed, air pressure, wind speed and wind direction of the working parachute in the environment where the working parachute is located; when the working parachute reaches the minimum parachute opening height, the minimum parachute opening height is 400 meters, entering the adaptive parachute opening stage, the working parachute is opened from top to bottom in sequence, the controller controls the number and timing of the working parachute opened each time, when the wind speed is high, the number of parachutes opened each time is controlled to be small; when the wind speed is low, the number of parachutes opened is controlled to be large; the opening time interval of the working parachute is a fixed value T 1 , when the first batch of power parachutes are opened, after a time interval T 1 , open the second batch of working parachutes again until all the working parachutes are fully opened.

[0011] The step (2) specifically refers to: the work parachute control model adopts a BP neural network model, and the BP neural network model is trained using the Adam method;

[0012] The BP neural network model comprises an input layer, a first hidden layer, a second hidden layer and an output layer, wherein the information input to the input layer includes the wind speed v w , wind direction d and air pressure p:

[0013]

[0014] in, represents the input of the input layer, l = 1, 2, 3; the output of the BP neural network model is the opening angle of the power parachute, which is adjusted by controlling the parachute rope of the power parachute. The output information includes the deviation angle of the parachute surface relative to the cable. With the goal of achieving maximum power generation efficiency for the parachute ladder type high altitude wind power device, the performance index function of the BP neural network model is:

[0015] E(k)=[y 1_0 (k)-y 1 (k)] 2

[0016] Among them, y 1_0 is the ideal deviation angle when actually opening the parachute, y 1 is the deviation angle of the umbrella surface relative to the cable during the actual operation of the umbrella-ladder high-altitude wind turbine device, and E is the output error;

[0017] The calculation formula of the performance indicator function gradient is as follows:

[0018] Δw 3,m,n =(y n -y n_0 )·y n (1-y n )·h 2,m

[0019]

[0020] Among them, w 3,m,n 、w 2,k,m and w 1,l,k are the weight coefficients from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; Δw 3,m,n , Δw 2,k,m and Δw 1,l,k are the weight coefficient adjustments from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; h 1,k represents the output of the first hidden layer, h 2,m represents the output of the second hidden layer; l, k, m, n represent the dimensions of the input layer, the first hidden layer, the second hidden layer, and the output layer, respectively; y n Represents the value of the output layer output, y n_0 Represents the reference value of the output layer; y j Indicates the solution of Δw 2,k,m , Δw 1,l,k In the process, the output layer outputs a value, y j_0 Indicates the solution of Δw 2,k,m , Δw 1,l,k The reference value of the output value of the output layer during the process; i represents the reference value of Δw 1,l,k Related w 2,k,m The number and j represent the relationship with Δw 2,k,m Related w 3,m,n Number;

[0021] Adma method is used to update weights, and the first-order moment estimation and second-order moment estimation of gradient are used to dynamically adjust the weights in the first hidden layer and the second hidden layer; 3,m,n The update process is:

[0022]

[0023] m t =β 1 m t-1 +(1-β 1 ) t

[0024] v t =β 2 vt-1 +(1-β 2 ) t 2

[0025]

[0026] Among them, ε is a very small number used to prevent Δw 3,m,n In the calculation process, the denominator is zero, and the value is 10e -8 ; α is the learning rate, the value is 0.001; m t 、v t are the first-order momentum term and the second-order momentum term respectively; are the deviation correction values ​​of the first-order momentum term and the second-order momentum term respectively; β 1 With β 2 Both are hyperparameters, with values ​​of 0.9 and 0.999 respectively; g t is the gradient of the objective function to be optimized; Δw 3,m,n * w 3,m,n The amount of adjustment.

[0027] The step (3) specifically refers to: when the active parachute enters the set height range, it enters the adaptive parachute closing stage, the active parachute is closed from bottom to top in sequence, the controller controls the number and timing of the active parachutes closed each time, when the wind speed is high, controls the number of parachutes closed each time to be large, when the wind speed is low, controls the number of parachutes closed each time to be small, to ensure that the maximum power is output before the cable length is exhausted; the closing time interval of the active parachute is T 2 , after closing the first batch of parachutes, the interval time T 2 Then close the second batch of working parachutes until all the working parachutes are closed.

[0028] The transfer function of the controller is:

[0029]

[0030] Among them, F u * (s) is the reference value of the parachute line tension output by the controller, k up With k ui are the weight coefficients of the controller, a u0 * is the acceleration reference value during the operation of the power parachute, a u F is the acceleration of the power parachute during operation, u0 The tension provided by the cable for the ground power generation system, f s The lift provided by the lift balloon; s is a complex variable representing the frequency in the Laplace transform domain.

[0031] It can be seen from the above technical scheme that the beneficial effects of the present invention are as follows: First, in the parachute opening stage, the present invention can ensure that the tension on the cable is maintained within the specified range by adaptively controlling the number of working parachutes opened, greatly reducing the probability of damage to the cable structure due to a sudden increase in tension during the parachute opening process, and increasing the service life of the cable. At the same time, within the tension range that the cable can withstand, the ascent rate is maximized, the cable length required for opening all working parachutes of the high-altitude wind power device is reduced, and sufficient cable length is ensured for collecting high-altitude wind energy in the subsequent maximum power working stage, thereby increasing the output power of each power generation cycle; second, the present invention has a During the high-power working stage, the opening angle of the working parachute is controlled by using a pre-trained BP neural network model to ensure that the wind power captured by the working parachute during the working stage can reach the maximum value, thereby enabling the output power of the high-altitude wind power device to reach the maximum value; thirdly, the present invention closes the working parachutes in sequence from bottom to top during the parachute closing stage. The number of working parachutes closed each time is determined according to the parachute speed of the working parachute at this time and the remaining length of the cable. All working parachutes are closed before the cable length is exhausted to avoid damage to equipment such as winches. At the same time, it is ensured that part of the wind energy can still be converted into electrical energy during the parachute closing stage, thereby improving the power generation efficiency in each power generation cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flow chart of the method of the present invention;

[0033] Figure 2 It is a schematic structural diagram of the high altitude wind power device of the present invention. DETAILED DESCRIPTION

[0034] like Figure 1 As shown, a method for controlling a working parachute for high-altitude wind power generation comprises the following steps in sequence:

[0035] (1) Adaptive parachute opening stage: After the parachute ladder type high-altitude wind turbine starts to operate, when the working parachute 3 reaches the minimum parachute opening height, the working parachute 3 is opened in sequence from top to bottom, and the controller adaptively adjusts the opening timing and number of parachutes according to the environmental conditions; when all the working parachutes 3 are opened, it enters the next stage;

[0036] (2) Maximum power working stage: construct and train the working parachute control model, and control the operating angle of the working parachute 3 through the trained working parachute control model. When the working parachute 3 enters the set altitude range, it enters the next stage; the set altitude range is 2300 meters to 2600 meters; the working parachute control model is loaded on the controller;

[0037] (3) Adaptive parachute closing stage: the working parachutes 3 are closed in sequence from bottom to top, and the controller adaptively adjusts the closing timing and the number of closed parachutes according to the environmental conditions until all the working parachutes 3 are closed.

[0038] The step (1) specifically refers to: releasing the lift balloon and collecting the working parachute operation data in real time, wherein the working parachute operation data includes the height and speed of the working parachute 3, the air pressure, wind speed and wind direction in the environment where the working parachute 3 is located; when the working parachute 3 reaches the minimum parachute opening height, which is 400 meters, the adaptive parachute opening stage is entered, and the working parachute 3 is opened from top to bottom in sequence. The controller controls the number and timing of the working parachute 3 opened each time. When the wind speed is high, the number of parachutes opened each time is controlled to be small; when the wind speed is low, the number of parachutes opened is controlled to be large; the opening time interval of the working parachute 3 is a fixed value T 1 , when the first batch of power parachutes are opened, after a time interval T 1 , open the second batch of working parachutes again until all working parachutes 3 are fully opened.

[0039] The step (2) specifically refers to: the work parachute control model adopts a BP neural network model, and the BP neural network model is trained using the Adam method;

[0040] The BP neural network model comprises an input layer, a first hidden layer, a second hidden layer and an output layer, wherein the information input to the input layer includes the wind speed v w , wind direction d and air pressure p:

[0041]

[0042] in, represents the input of the input layer, l = 1, 2, 3; the output of the BP neural network model is the opening angle of the power parachute 3, the opening angle of the power parachute 3 is adjusted by controlling the parachute rope of the power parachute 3, and the output information includes the deviation angle of the parachute surface relative to the cable With the goal of achieving maximum power generation efficiency for the parachute ladder type high altitude wind power device, the performance index function of the BP neural network model is:

[0043] E(k)=[y 1_0 (k)-y 1 (k)] 2

[0044] Among them, y 1_0 is the ideal deviation angle when actually opening the parachute, y 1 is the deviation angle of the umbrella surface relative to the cable during the actual operation of the umbrella-ladder high-altitude wind turbine device, and E is the output error;

[0045] The calculation formula of the performance indicator function gradient is as follows:

[0046] Δw 3,m,n =(y n -y n_0 )·y n (1-yn )·h 2,m

[0047]

[0048] Among them, w 3,m,n 、w 2,k,m and w 1,l,k are the weight coefficients from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; Δw 3,m,n , Δw 2,k,m and Δw 1,l,k are the weight coefficient adjustments from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; h 1,k represents the output of the first hidden layer, h 2,m represents the output of the second hidden layer; l, k, m, n represent the dimensions of the input layer, the first hidden layer, the second hidden layer, and the output layer, respectively; y n Represents the value of the output layer output, y n_0 Represents the reference value of the output layer; y j Indicates the solution of Δw 2,k,m , Δw 1,l,k In the process, the output layer outputs a value, y j_0 Indicates the solution of Δw 2,k,m , Δw 1,l,k The reference value of the output value of the output layer during the process; i represents the reference value of Δw 1,l,k Related w 2,k,m The number and j represent the relationship with Δw 2,k,m Related w 3,m,n Number;

[0049] Adma method is used to update weights, and the first-order moment estimation and second-order moment estimation of gradient are used to dynamically adjust the weights in the first hidden layer and the second hidden layer; 3,m,n The update process is:

[0050]

[0051] m t =β 1 m t-1 +(1-β 1 ) t

[0052] v t =β 2 v t-1 +(1-β 2 ) t 2

[0053]

[0054] Among them, ε is a very small number used to prevent Δw 3,m,n In the calculation process, the denominator is zero, and the value is 10e -8 ; α is the learning rate, the value is 0.001; m t 、v t are the first-order momentum term and the second-order momentum term respectively; are the deviation correction values ​​of the first-order momentum term and the second-order momentum term respectively; β 1 With β 2 Both are hyperparameters, with values ​​of 0.9 and 0.999 respectively; g t is the gradient of the objective function to be optimized; Δw 3,m,n * w 3,m,n The amount of adjustment.

[0055] The operating data of the working parachute and the data used in training the model are obtained by sensors arranged on the parachute, and the opening and closing of the working parachute 3 are controlled by a controller on the parachute.

[0056] Adaptive control is used to adjust the opening and closing of the active parachute 3 in both the adaptive parachute opening stage and the adaptive parachute closing stage. By ensuring that the parachute speed change rate after the system opens or closes the parachute is maintained within a certain range, the system speed is prevented from suddenly increasing or decreasing to damage the structure of the cable and the parachute rope, thereby improving the reliability and stability of the high-altitude wind power device. When the active parachute 3 is opened, the lift obtained when the active parachute 3 is opened can be calculated according to the sensor measurement data. The order of opening the active parachute 3 is to open from top to bottom, and the order of closing the active parachute 3 is to close from bottom to top.

[0057] When in the maximum power working stage, the data measured by the sensor is input into the BP neural network model, and the operating angle of the working parachute 3 is calculated by the BP neural network model. The system power generation efficiency is judged by the average power generated by the generator over a period of time. The time for measuring the average power must be an integer multiple of the power generation cycle of the working parachute 3, where the calculation of the average power of the generator needs to consider the energy generated by the working parachute in the working stage and the energy consumed in the recovery stage of the working parachute. When training the BP neural network control model, it is necessary to collect data of the high-altitude wind power device under different operating conditions. The operating conditions include, under the same environmental conditions, the output power of the generator when the working parachute 3 is opened at different times and at different angles, and the energy consumed by the working parachute 3 when the working parachute 3 is closed at different times and at different angles. At the same time, it is also necessary to repeat the above process under different environmental conditions to collect the operating data of the working parachute 3 under different environmental conditions.

[0058] The step (3) specifically refers to: when the active parachute 3 enters the set height range, it enters the adaptive parachute closing stage, the active parachute 3 is closed from bottom to top in sequence, and the controller controls the number and timing of the active parachute 3 closed each time. When the wind speed is high, the number of parachute closed each time is controlled to be large, and when the wind speed is low, the number of parachute closed each time is controlled to be small, so as to ensure that the maximum power is output before the cable length is exhausted; the closing time interval of the active parachute 3 is T 2 , after closing the first batch of parachutes, the interval time T 2 Then close the second batch of working parachutes until all the working parachutes are closed.

[0059] The transfer function of the controller is:

[0060]

[0061] Among them, F u * (s) is the reference value of the parachute line tension output by the controller, k up With k ui are the weight coefficients of the controller, a u0 * is the acceleration reference value during the operation of the power parachute, a u F is the acceleration of the power parachute during operation, u0 The tension provided by the cable for the ground power generation system, F s The lift provided by the lift balloon; s is a complex variable representing the frequency in the Laplace transform domain.

[0062] like Figure 2 As shown, the high-altitude wind power device of the present invention comprises a ground power generation system 1 located on the ground, a cable 2 connected to the ground power generation system 1, a working parachute 3 for capturing wind energy, and a lifting balloon 4 for building a cable track. Before starting the high-altitude wind power device, it is necessary to release the lifting balloon 4 to drive the cable 2 into the air and provide an ascending track for the working parachute 3.

[0063] The ground power generation system 1 includes a controller, a frequency converter, a winch, a cable groove and a universal pulley. One end of the cable is connected to the working parachute 3, and the other end is turned through the universal pulley and then connected to the drum of the winch through the cable groove. The drum of the winch rotates to drive the motor of the winch to generate electricity, which is transmitted to the output of the frequency converter through the cable. The controller receives the signal sent by the sensor, which is installed on the working parachute 3. The sensor collects the operating data of the working parachute in real time. The controller processes the received signal and sends a control signal to the working parachute 3 to control the opening and closing of the working parachute 3.

[0064] In summary, in the parachute opening stage, the present invention can ensure that the tension on the cable is maintained within the specified range by adaptively controlling the number of open working parachutes 3, greatly reducing the probability of damage to the cable structure due to a sudden increase in tension during the parachute opening process, and increasing the service life of the cable. At the same time, within the tension range that the cable can withstand, the ascent rate is maximized, reducing the cable length required for opening all working parachutes 3 of the high-altitude wind power device, ensuring that the subsequent maximum power working stage has sufficient cable length for collecting high-altitude wind energy, and increasing the output power of each power generation cycle; in the maximum power working stage, the present invention adopts The opening angle of the working parachute 3 is controlled by using a pre-trained BP neural network model to ensure that the wind power captured by the working parachute 3 in the working stage can reach the maximum value, so that the output power of the high-altitude wind power device can reach the maximum value; the present invention closes the working parachute 3 from bottom to top in sequence in the parachute closing stage, and the number of working parachutes 3 closed each time is determined according to the parachute speed of the working parachute 3 at this time and the remaining length of the cable. All the working parachutes 3 are closed before the cable length is exhausted to avoid damage to equipment such as winches. At the same time, it is ensured that part of the wind energy can still be converted into electrical energy in the parachute closing stage, thereby improving the power generation efficiency in each power generation cycle.

Claims

1. A method for controlling a working parachute for high-altitude wind power, characterized in that: The method comprises the following steps in order: (1) Adaptive parachute opening stage: After the parachute ladder type high-altitude wind turbine starts to operate, when the working parachute reaches the minimum parachute opening height, the working parachute opens from top to bottom in sequence, and the controller adaptively adjusts the opening time and number of parachutes according to the environmental conditions; when all the working parachutes are opened, it enters the next stage; (2) Maximum power working stage: construct and train a working parachute control model, and control the operating angle of the working parachute through the trained working parachute control model. When the working parachute enters the set altitude range, it enters the next stage; the set altitude range is 2300 meters to 2600 meters; (3) Adaptive parachute closing stage: The working parachutes are closed from bottom to top in sequence, and the controller adaptively adjusts the closing timing and number of parachutes according to environmental conditions until all the working parachutes are closed.

2. The working parachute control method for high-altitude wind power according to claim 1 is characterized in that: The step (1) specifically refers to: launching the lift balloon and collecting the working parachute operation data in real time, wherein the working parachute operation data includes the height, speed, air pressure, wind speed and wind direction of the working parachute in the environment where the working parachute is located; when the working parachute reaches the minimum parachute opening height, the minimum parachute opening height is 400 meters, and enters the adaptive parachute opening stage, the working parachutes are opened in sequence from top to bottom, and the controller controls the number and timing of the working parachutes opened each time, when the wind speed is large, the number of parachutes opened each time is controlled to be small; when the wind speed is small, the number of parachutes opened is controlled to be large; the opening time interval of the working parachutes is a fixed value T1, when the first batch of working parachutes are opened, after the time interval T1, the second batch of working parachutes are opened again until all the working parachutes are fully opened.

3. The working parachute control method for high-altitude wind power according to claim 1 is characterized in that: The step (2) specifically refers to: the work parachute control model adopts a BP neural network model, and the BP neural network model is trained using the Adam method; The BP neural network model comprises an input layer, a first hidden layer, a second hidden layer and an output layer, wherein the information input to the input layer includes the wind speed v w , wind direction d and air pressure p: in, represents the input of the input layer, l = 1, 2, 3; the output of the BP neural network model is the opening angle of the power parachute, which is adjusted by controlling the parachute rope of the power parachute. The output information includes the deviation angle of the parachute surface relative to the cable. With the goal of achieving maximum power generation efficiency for the parachute ladder type high altitude wind power device, the performance index function of the BP neural network model is: E(k)=[y 1_0 (k)-y1(k)] 2 Among them, y 1_0 is the ideal deviation angle when the parachute is actually opened, y1 is the deviation angle of the parachute surface relative to the cable during the actual operation of the parachute ladder type high-altitude wind turbine device, and E is the output error; The calculation formula of the performance indicator function gradient is as follows: Δw 3,m，n (and n -and n_0 )·and n (1-and n )·h 2,m Among them, w 3,m，n 、w 2,k，m and w 1,l，k are the weight coefficients from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; Δw 3,m，n , Δw 2,k，m and Δw 1,l,k are the weight coefficient adjustments from the second hidden layer to the output layer, from the first hidden layer to the second hidden layer, and from the input layer to the first hidden layer; h 1，k represents the output of the first hidden layer, h 2,m represents the output of the second hidden layer; l, k, m, n represent the dimensions of the input layer, the first hidden layer, the second hidden layer, and the output layer, respectively; y n Represents the value of the output layer output, y n_0 Represents the reference value of the output layer; y j Indicates the solution of Δw 2,k,m , Δw 1,l，k In the process, the output layer outputs a value, y j_0 Indicates the solution of Δw 2,k,m , Δw 1,l,k The reference value of the output value of the output layer during the process; i represents the reference value of Δw 1,l，k Related w 2,k,m The number and j represent the relationship with Δw 2,k，m Related w 3,m，n Number; Adma method is used to update weights, and the first-order moment estimation and second-order moment estimation of gradient are used to dynamically adjust the weights in the first hidden layer and the second hidden layer; 3,m，n The update process is: m t =β1m t-1 +(1-β1)g t v t =β2v t-1 +(1-β2)g t 2 Among them, ε is a very small number used to prevent Δw 3,m,n In the calculation process, the denominator is zero, and the value is 10e -8 ; α is the learning rate, the value is 0.001; m t 、v t are the first-order momentum term and the second-order momentum term respectively; are the deviation correction values ​​of the first-order momentum term and the second-order momentum term respectively; β1 and β2 are both hyperparameters, with values ​​of 0.9 and 0.999 respectively; g t is the gradient of the objective function to be optimized; Δw 3,m, n * w 3,m,n The amount of adjustment.

4. The working parachute control method for high-altitude wind power according to claim 1 is characterized in that: The step (3) specifically refers to: when the active parachute enters the set height range, it enters the adaptive parachute closing stage, and the active parachutes are closed in sequence from bottom to top. The controller controls the number and timing of the active parachutes closed each time. When the wind speed is high, the number of parachutes closed each time is controlled to be large, and when the wind speed is low, the number of parachutes closed each time is controlled to be small, so as to ensure that the maximum power is output before the cable length is exhausted; the closing time interval of the active parachutes is T2. After the first batch of active parachutes are closed, the second batch of active parachutes are closed after the interval T2, until all the active parachutes are closed.

5. The working parachute control method for high altitude wind power according to claim 2 or 4, characterized in that: The transfer function of the controller is: Among them, F u * (s) is the reference value of the parachute rope tension output by the controller, k up With k ui are the weight coefficients of the controller, a u0 * is the acceleration reference value during the operation of the power parachute, a u F is the acceleration of the power parachute during operation, u0 The tension provided by the cable for the ground power generation system, F s The lift provided by the lift balloon; s is a complex variable representing the frequency in the Laplace transform domain.

Images