An electronically controlled suspension method based on vibration energy recovery

Abstract

The present invention discloses an electronically controlled suspension method based on vibration energy recovery, comprising the following steps: a vehicle-mounted sensor obtains state variables of vehicle body force and vehicle body motion, and transmits the motion state to a controller. Use "ceiling" and "ground shelf" hybrid damping control to generate ideal motion, rely on linear quadratic optimal control to follow the ideal motion, and calculate the optimal control main force. Design a decision system to decide whether to output the main force based on the road surface grade and the amount of electricity. If the electricity is low or the road surface grade is low, the equivalent damping main force is output; if the electricity is high or the road surface grade is high, the optimal control main force is output. Rely on a linear DC motor to output the corresponding main force, and the power can be obtained according to the product of the coil current and the power supply voltage. When the power is positive, energy is consumed, and when it is negative, energy is recovered. The present invention uses a linear DC motor as an actuator for the active suspension, which can achieve self-power supply while controlling the stability of the vehicle body.

Description

Technical Field

[0001] The invention belongs to the field of automobile intelligentization methods, and particularly relates to an electric control suspension method based on vibration energy recovery. Background Art

[0002] Traditional passive suspension systems cannot guarantee both ride comfort and handling stability at the same time, so researchers are committed to the study of active suspension. Active suspension can generate active force and effectively reduce the pitch and tilt of the car during driving, but it consumes a lot of energy to suppress the movement of the car body, which raises the issue of energy saving. Linear DC motors can solve this problem. The present invention uses a linear DC motor as an actuator for the active suspension and designs an electronically controlled suspension method that recovers part of the vehicle's vibration energy, which can achieve self-power supply while controlling the stability of the vehicle body. Summary of the invention

[0003] In order to solve the technical problems mentioned in the above background technology, the present invention proposes an electronically controlled suspension method based on vibration energy recovery.

[0004] In order to achieve the above method objectives, the technical solution of the present invention is:

[0005] An electronically controlled suspension method based on vibration energy recovery comprises the following steps:

[0006] S1, the vehicle-mounted sensor obtains the state variables of the vehicle body force and vehicle body motion, and transmits the motion state to the controller;

[0007] S2, use "ceiling shed" and "ground shed" mixed damping control to produce ideal motion, rely on linear quadratic optimal control to follow the ideal motion, and calculate the optimal control active force;

[0008] S3. Design a decision-making system to determine the output of main power according to the road grade and the amount of electricity;

[0009] The road conditions in the past period of time are obtained by using a laser sensor, and then the road conditions in the future period of time are predicted by using the grey system DGM (2,1) prediction model. The actual power spectrum density is calculated and compared with the standard power spectrum density to estimate the road grade. At the same time, the remaining power in the energy storage battery is detected. If the power is low or the road grade is low, the equivalent damping main force is output; if the power is high or the road grade is high, the optimal control main force is output.

[0010] S4. Rely on the linear DC motor to output the corresponding main force, and get the power according to the product of the coil current and the power supply voltage. When the power is positive, energy is consumed, and when it is negative, energy is recovered.

[0011] Furthermore, the linear quadratic optimal control strategy described in step S2 is as follows:

[0012] 2.1. Establish a 1 / 4 vehicle active suspension model, and the dynamic equation is:

[0013]

[0014] The calculation formulas for spring force and damping force are as follows:

[0015]

[0016] Among them, m1 is the wheel mass, m2 is the suspension mass, z1 is the wheel vertical displacement, z2 is the suspension vertical displacement, q1 is the road surface, and F k1 is the tire force, F k2 is the spring force, F c is the damping force, F u Main power;

[0017] 2.2. Solve the state space equation:

[0018] Define the state variables of the system:

[0019]

[0020] The state space equation of the active suspension is obtained:

[0021]

[0022] Where u = F u , A is the system matrix, B1 and B2 are input matrices, and their internal elements are as follows:

[0023]

[0024] 2.3. Use "ceiling shed" and "ground shed" mixed damping control to produce ideal motion:

[0025]

[0026] Where z 1d is the ideal vertical displacement of the wheel; z 2d is the ideal vertical displacement of the suspension; is the ideal vertical speed of the wheel; is the ideal vertical speed of the suspension; is the ideal vertical acceleration of the wheel; is the ideal vertical acceleration of the suspension; where the “ground shelf” damping force F sky1 and the "ceiling" damping force F sky2 The calculation formula is as follows:

[0027]

[0028] Consider the ideal motion as the input of the actual motion system, and define the input x d for:

[0029]

[0030] 2.4. Linear quadratic optimal control strategy:

[0031] Define the output variable y:

[0032]

[0033] The output equation is:

[0034] y=Cx+Du+Ex d

[0035] The internal elements of C, D, and E are as follows:

[0036]

[0037] The comprehensive performance index function is:

[0038]

[0039] Among them, q1, q2, q3, r0 are the weighted coefficients of the four indicators; t f Indicates the termination time of the control action; define the diagonal matrix Q0:

[0040]

[0041] Convert the performance index function into matrix form:

[0042]

[0043] According to the output equation, the performance index function is further written as:

[0044]

[0045] Define the following matrix:

[0046] Q=C T Q0C R=C T Q0D S=C T QUR

[0047] T=D T Q0E W=E T Q0E V=D T Q0D+r0

[0048] The performance indicator function is simplified to the following form:

[0049]

[0050] Introduce the Hamiltonian function:

[0051]

[0052] Where λ is the undetermined multiplier vector; due to optimal control, we get:

[0053]

[0054] Therefore, the expression of the control quantity u when the optimal control is obtained is:

[0055]

[0056] From the canonical equation we know:

[0057]

[0058]

[0059] Then, a new matrix P,g is introduced to represent λ:

[0060] λ=Px-g

[0061] Substituting this into the canonical equation and taking the derivative with respect to λ, we obtain the Riccati equation and the solution equation for g:

[0062]

[0063]

[0064] So the main force F under optimal control is obtained a ;

[0065] However, if you want to recover energy, the output main force should be opposite to the relative speed of the suspension. Therefore, the output main force should be similar to the damping force. The calculation formula is as follows:

[0066]

[0067] In the formula, c b is the equivalent damping coefficient.

[0068] Furthermore, the decision-making system described in step S3 has a decision-making principle as follows:

[0069] 3.1. Decision based on road surface grade:

[0070] First, the laser sensor is used to obtain the road surface conditions over a period of time in the past, and the original sequence of road surface height is obtained:

[0071] x (0) =[x (0) (1),x (0) (2),x (0) (3),…,x (0) (n)]

[0072] In the formula, n represents the number of road height data obtained;

[0073] Then the grey system DGM (2,1) prediction model is used to predict the road conditions in the future, thus obtaining the prediction sequence:

[0074]

[0075] In the formula, k represents the number of road height data predicted based on the obtained n data;

[0076] Analogously to the power spectral density of continuous data, the power spectral density of discrete data is obtained by Fourier transform:

[0077]

[0078] In the formula, is the number of road roughness data; x m is the value of the mth data; Δt is the interval of emitting laser light; n is the spatial frequency; e is the natural logarithm; i is the imaginary unit;

[0079] The calculation formula of standard power spectrum density is obtained according to the road surface grade:

[0080]

[0081] In the formula, G q (n0) is the road roughness coefficient; n is the spatial frequency; n0 is the spatial reference frequency, which is 0.1m -1 ; w is the frequency index, which is 2;

[0082] Then find the spectral density function of the road surface grade that is closest to the actual power spectral density function;

[0083] The higher the road grade, the smaller the road undulation, and the optimal control main force F is output. a , mainly to suppress the vibration of the vehicle body; if the road surface grade is low, the road surface will be undulating, and the output damping main force F b , mainly to recover energy;

[0084] 3.2. Decision based on the remaining power in the energy storage battery:

[0085] Detect the remaining power in the energy storage battery. If the power is lower than the warning power, the damping active force F is output. b, mainly to recover energy; if the power is higher than the warning power, the optimal control main force F is output a , mainly to suppress vehicle body vibration.

[0086] Furthermore, the model of the linear DC motor described in step S4 is as follows:

[0087] The electromagnetic force equation of a linear DC motor is:

[0088] F m =k b B δ lNi=k m i

[0089] In the formula, k b is the structural coefficient; B δ is the magnetic induction intensity; l is the effective length of each turn of the coil in the magnetic field; N is the number of turns of the coil; k m =k b B δ lN is the force constant of the linear motor; i is the coil current;

[0090] The back electromotive force E of the armature winding is:

[0091]

[0092] In the formula, k E =k b B δ lN is the back electromotive force coefficient; x is the displacement of the mover, that is, the vertical displacement of the wheel;

[0093] The voltage balance equation is:

[0094]

[0095] In the formula, R is the resistance of the coil winding; L is the inductance of the coil winding; the influence of the coil part is very small and can be ignored; i is the coil current; u is the power supply voltage;

[0096] The output power calculation formula of the power supply is:

[0097]

[0098] When the power P is positive, energy is consumed, and when it is negative, energy is recovered.

[0099] The beneficial effects brought by adopting the above technical solution are:

[0100] The present invention adopts a linear DC motor to replace the previous mechanical transmission, improve the reaction speed and reduce energy loss. The "ceiling" and "ground shed" hybrid damping control is used to produce ideal motion and improve the rationality of the control strategy. The use of linear quadratic optimal control can take into account both control performance and the energy consumed to generate control force. The gray prediction model is used to predict the road conditions in the future to improve the accuracy of estimating the road surface grade, and the type of output main power is determined according to the remaining power of the energy storage battery and the road surface grade, so as to control energy recovery and consumption and achieve self-power supply. Therefore, the present invention has strong applicability and good energy recovery effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 A simplified diagram of the steps of the energy recovery electronically controlled suspension method of the present invention;

[0102] Figure 2 A detailed diagram of the steps of the energy recovery electronically controlled suspension method of the present invention;

[0103] Figure 3 The active suspension 1 / 4 car model adopted by the present invention;

[0104] Figure 4 This is the "ceiling shed" and "ground shed" hybrid damping control adopted by the present invention. DETAILED DESCRIPTION

[0105] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0106] like Figure 1 Figure 2 As shown, the present invention designs an electronically controlled suspension method based on vibration energy recovery, comprising the following steps:

[0107] S1. The vehicle-mounted sensor obtains the state variables of the vehicle body force and vehicle body motion, and transmits the motion state to the controller.

[0108] S2. Use the "ceiling shed" and "ground shed" mixed damping control to produce ideal motion, rely on linear quadratic optimal control to follow the ideal motion, and calculate the optimal control main force.

[0109] S3. Design a decision-making system to determine the output of main power according to the road grade and the amount of electricity;

[0110] The road conditions in the past period of time are obtained using a laser sensor, and then the grey system DGM (2,1) prediction model is used to predict the road conditions in the future. The actual power spectrum density is calculated and compared with the standard power spectrum density to estimate the road grade. At the same time, the remaining power in the energy storage battery is detected. If the power is low or the road grade is low, the equivalent damping main force is output; if the power is high or the road grade is high, the optimal control main force is output.

[0111] S4. Rely on the linear DC motor to output the corresponding main force. The power can be obtained according to the product of the coil current and the power supply voltage. When the power is positive, energy is consumed, and when it is negative, energy is recovered.

[0112] First, establish a 1 / 4 vehicle active suspension model, such as Figure 3 As shown, the kinetic equation is:

[0113]

[0114] The calculation formulas for spring force and damping force are as follows:

[0115]

[0116] Among them, m1 is the wheel mass, m2 is the suspension mass, z1 is the wheel vertical displacement, z2 is the suspension vertical displacement, q1 is the road surface, and F k1 is the tire force, F k2 is the spring force, F c is the damping force, F u As the main driving force.

[0117] Then we can define the state variables of the system:

[0118]

[0119] Define the control input u as:

[0120] u=F u

[0121] Define the disturbance input w as:

[0122]

[0123] Therefore, we can get the state space equation of the active suspension:

[0124]

[0125] In the formula, A is the system matrix, B1 and B2 are input matrices, and their internal elements are as follows:

[0126]

[0127] Next, use the "sky shed" damping control and the "ground shed" damping control to mix and match to produce the desired motion, such as Figure 4 As shown, there is no "ceiling" and "ground shelf" on the actual car, so there is no need to design them in reality. We can get the ideal dynamic equation generated by mixed damping control:

[0128]

[0129] Where z 1d is the ideal vertical displacement of the wheel; z 2d is the ideal vertical displacement of the suspension; is the ideal vertical speed of the wheel; is the ideal vertical speed of the suspension; is the ideal vertical acceleration of the wheel; is the ideal vertical acceleration of the suspension; where the “ground shelf” damping force F sky1 and the "ceiling" damping force F sky2 The calculation formula is as follows:

[0130]

[0131] Therefore, we can regard the ideal motion as the input of the actual motion system and define the input x d for:

[0132]

[0133] Combining the ideal motion and the actual motion, we can define the output variable y:

[0134]

[0135] So we can get the output equation:

[0136] y=Cx+Du+Ex d

[0137] The internal elements of C, D, and E are as follows:

[0138]

[0139] In order to balance the control performance and the energy consumed in generating the control force, we decided to adopt the optimal linear quadratic controller LQR, using the output variable y and the active force F u As a performance indicator, the tracking error of the vehicle body acceleration The tracking error of the suspension displacement is z2-z1-(z 2d -z 1d ), the tracking error of the tire type variable z1-q1-(z 1d-q1) and the main power F output by the motor u The comprehensive performance index function is:

[0140]

[0141] Among them, q1, q2, q3, r0 are the weighted coefficients of the four indicators; t f is the termination time of the control action. Define the diagonal matrix Q0:

[0142]

[0143] The performance indicator function can be transformed into a matrix form:

[0144]

[0145] According to the output equation, the performance index function can be further written as:

[0146]

[0147] For convenience, we define the following matrix:

[0148] Q=C T Q0C R=C T Q0D S=C T QUR

[0149] T=D T Q0E W=E T Q0E V=D T Q0D+r0

[0150] The performance indicator function can be simplified as follows:

[0151]

[0152] In order to find the optimal control, we introduce the Hamiltonian function:

[0153]

[0154] Where λ is the undetermined multiplier vector. Due to optimal control, we can obtain:

[0155]

[0156] So we can get the expression of the control quantity u during optimal control:

[0157]

[0158] From the canonical equation we know that:

[0159]

[0160]

[0161] Then, we introduce a new matrix P,g to represent λ:

[0162] λ=Px-g

[0163] Substituting this into the canonical equation, we get:

[0164]

[0165] Taking the derivative of λ, we can get:

[0166]

[0167] Substituting the regularization equation and the expression for λ into this formula, we can obtain:

[0168]

[0169] Comparing the two equations we can get the Ricatti equation and the solution equation for g:

[0170]

[0171]

[0172] Therefore, the entire LQR control strategy is: we first select appropriate parameters. At the same time, the actual motion state and ideal motion state of the vehicle will also generate corresponding data when it is moving. Then, we can get P,g by solving the Ricatti equation and g equation, and then get λ, and finally get the active force F at the optimal control time. a .

[0173] However, if you want to recover energy, the output main force should be opposite to the relative speed of the suspension. Therefore, the output main force should be similar to the damping force. The calculation formula is as follows:

[0174]

[0175] In the formula, c b is the equivalent damping coefficient;

[0176] 3.1. Decision based on road roughness level:

[0177] First, the laser sensor is used to obtain the road surface conditions over a period of time in the past, from which the original sequence of road surface height can be obtained:

[0178] x (0) =[x (0)(1),x (0) (2),x (0) (3),…,x (0) (n)]

[0179] In the formula, n represents the number of road height data obtained;

[0180] Then the grey system prediction model is used to predict the road conditions in the future.

[0181] So we can get its 1-AGO sequence and 1-IAGO as:

[0182] x (1) =[x (1) (1),x (1) (2),x (1) (3),…,x (1) (n)]

[0183] α (1) x (0) =[α (1) x (0) (2),…,α (1) x (0) (n)]

[0184] in

[0185] α (1) x (0) (k) = x (0) (k)-x (0) (k-1),k=2,3,…,n

[0186] then

[0187] α (1) x (0) (k)+ax (0) (k) = b

[0188] It is called the DGM(2,1) model.

[0189] Set parameter column So we can get

[0190]

[0191] in

[0192]

[0193] So the least squares estimate of the DGM(2,1) model satisfies:

[0194]

[0195] So the time response sequence of the DGM(2,1) model is:

[0196]

[0197] Finally we can get the restored value:

[0198]

[0199] We can then get the prediction sequence:

[0200]

[0201] In the formula, k represents the number of road height data predicted based on the obtained n data;

[0202] Since the laser sensor emits laser at a fixed frequency, we can obtain the road roughness x in the past time t1 and the future time t2, that is, the road roughness x in time t, and these data are discrete. The power spectrum density of discrete data can be obtained by analogy with the power spectrum density of continuous data. First, we can regard the non-periodic continuous road roughness function x(t) as a periodic function x T (t) When T→∞, using Fourier transform, we can get the power spectral density function:

[0203]

[0204] Where f is the frequency, which can be positive or negative, but in practice it is meaningless to be negative, so the power spectral density is often expressed as:

[0205]

[0206] By analogy, we can get the power spectral density of discrete data:

[0207]

[0208] In the formula, is the number of road roughness data; x m is the value of the mth data; Δt is the interval of emitting laser light; n is the spatial frequency;

[0209] According to the road surface grade, we can get the calculation formula of standard power spectrum density:

[0210]

[0211] In the formula, G q(n0) is the road roughness coefficient. The road roughness coefficient values ​​for different road grades can be found in the national standard GB7031-87; n is the spatial frequency; n0 is the spatial reference frequency, which is 0.1m -1 ; w is the frequency index, which is 2;

[0212] Therefore, we can find the difference between the actual power spectrum density and the power spectrum density of each road surface grade at different spatial frequencies, and then take the absolute value of the difference between different grades and sum them up. The grade corresponding to the smallest sum is the grade corresponding to the actual road surface. That is, we find the spectral density function of the road surface grade that is closest to the actual power spectrum density function.

[0213] The higher the road grade, the smaller the road undulation, and the optimal control main force F is output. a , mainly to suppress the vibration of the vehicle body; if the road surface grade is low, the road surface will be undulating, and the output damping main force F b , mainly to recover energy.

[0214] 3.2. Decision based on the remaining power in the energy storage battery:

[0215] Detect the remaining power in the energy storage battery. If the power is lower than the warning power, the damping active force F is output. b , mainly to recover energy; if the power is still higher than the warning power, the optimal control main force F is output a , mainly to suppress vehicle body vibration.

[0216] The electromagnetic force equation of a linear DC motor is:

[0217] F m =k b B δ lNi=k m i

[0218] In the formula, k b is the structural coefficient; B δ is the magnetic induction intensity; l is the effective length of each turn of the coil in the magnetic field; N is the number of turns of the coil; k m =k b B δ lN is the force constant of the linear motor; i is the coil current;

[0219] The back electromotive force E of the armature winding is:

[0220]

[0221] In the formula, k E =k b B δ lN is the back electromotive force coefficient; x is the displacement of the mover, that is, the vertical displacement of the wheel; the voltage balance equation when the main force is output is:

[0222]

[0223] Where R is the resistance of the coil winding and L is the inductance of the coil winding. The influence of the coil part is very small and can be ignored.

[0224] The power calculation formula of the power supply is

[0225]

[0226] When the power P is positive, energy is consumed, and when it is negative, energy is recovered.

[0227] The present invention adopts a linear DC motor to replace the previous mechanical transmission, improve the reaction speed and reduce energy loss. The "ceiling" and "ground shed" hybrid damping control is used to produce ideal motion and improve the rationality of the control strategy. The use of linear quadratic optimal control can take into account both control performance and the energy consumed to generate control force. The gray prediction model is used to predict the road conditions in the future to improve the accuracy of estimating the road surface grade, and the type of output main power is determined according to the remaining power of the energy storage battery and the road surface grade, so as to control energy recovery and consumption and achieve self-power supply. Therefore, the present invention has strong applicability and good energy recovery effect.

[0228] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. An electronically controlled suspension method based on vibration energy recovery, characterized in that: The following steps are involved: S1, the vehicle-mounted sensor obtains the state variables of the vehicle body force and vehicle body motion, and transmits the motion state to the controller; S2, use "ceiling" and "ground" mixed damping control to produce ideal motion, rely on linear quadratic optimal control to follow the ideal motion, and calculate the optimal control active force; S3. Design a decision-making system to determine the output of main power according to the road grade and the amount of electricity; The road conditions in the past period of time are obtained by using a laser sensor, and then the road conditions in the future period of time are predicted by using the grey system DGM (2,1) prediction model. The actual power spectrum density is calculated and compared with the standard power spectrum density to estimate the road grade. If the road grade is high, the road surface fluctuation is small, and if the road grade is low, the road surface fluctuation is large. At the same time, the remaining power in the energy storage battery is detected. If the power is low or the road grade is low, the equivalent damping main force is output; if the power is high or the road grade is high, the optimal control main force is output; S4, relying on the linear DC motor to output the corresponding main force, the power is obtained according to the product of the coil current and the power supply voltage, when the power is positive, energy is consumed, and when it is negative, energy is recovered; The linear quadratic optimal control strategy described in step S2 is as follows: 2.

1. Establish a 1 / 4 vehicle active suspension model, and the dynamic equation is: The calculation formulas for spring force and damping force are as follows: Among them, m1 is the wheel mass, m2 is the suspension mass, z1 is the wheel vertical displacement, z2 is the suspension vertical displacement, q1 is the road surface, and F k1 is the tire force, F k2 is the spring force, F c is the damping force, F u Main power; 2.

2. Solve the state space equation: Define the state variables of the system: The state space equation of the active suspension is obtained: Where u = F u , A is the system matrix, B1 and B2 are input matrices, and their internal elements are as follows: 2.

3. Use "ceiling" and "ground" mixed damping control to produce ideal motion: Where z 1d is the ideal vertical displacement of the wheel; z 2d is the ideal vertical displacement of the suspension; is the ideal vertical speed of the wheel; is the ideal vertical speed of the suspension; is the ideal vertical acceleration of the wheel; is the ideal vertical acceleration of the suspension; where the "ground shelf" damping force F sky1 and the "ceiling" damping force F sky2 The calculation formula is as follows: Consider the ideal motion as the input of the actual motion system and define the input x d for: 2.

4. Linear quadratic optimal control strategy: Define the output variable y: The output equation is: y=Cx+Du+Ex d The internal elements of C, D, and E are as follows: The comprehensive performance index function is: Among them, q1, q2, q3, r0 are the weighted coefficients of the four indicators; t f Indicates the termination time of the control action; define the diagonal matrix Q0: Convert the performance index function into matrix form: According to the output equation, the performance index function is further written as: Define the following matrix: Q=C T Q0C R=C T Q0D S=C T Q0E T=D T Q0E W=E T Q0E V=D T Q0D+r0 The performance indicator function is simplified to the following form: Introduce the Hamiltonian function: Where λ is the undetermined multiplier vector; due to optimal control, we get: Therefore, the expression of the control quantity u when the optimal control is obtained is: From the canonical equation we know: Then, a new matrix P,g is introduced to represent λ: λ=Px-g Substituting this into the canonical equation and taking the derivative with respect to λ, we obtain the Riccati equation and the solution equation for g: So the main force F under optimal control is obtained a ; However, if you want to recover energy, the output main force should be opposite to the relative speed of the suspension. Therefore, the output main force should be similar to the damping force. The calculation formula is as follows: In the formula, c b is the equivalent damping coefficient.

2. The electronically controlled suspension method based on vibration energy recovery according to claim 1, characterized in that: The decision-making system described in step S3 has the following decision-making principle: 3.

1. Decision based on road surface grade: First, the laser sensor is used to obtain the road surface conditions over a period of time in the past, and the original sequence of road surface height is obtained: In the formula, n represents the number of road height data obtained; Then the grey system DGM (2,1) prediction model is used to predict the road conditions in the future, thus obtaining the prediction sequence: In the formula, k represents the number of road height data predicted based on the obtained n data; Analogously to the power spectral density of continuous data, the power spectral density of discrete data is obtained by Fourier transform: In the formula, is the number of road roughness data; x m is the value of the mth data; Δt is the interval of emitting laser light; n is the spatial frequency; e is the natural logarithm; i is the imaginary unit; The calculation formula of standard power spectrum density is obtained according to the road surface grade: In the formula, G q (n0) is the road roughness coefficient; n is the spatial frequency; n0 is the spatial reference frequency, which is 0.1m -1 ; w is the frequency index, which is 2; Then find the spectral density function of the road surface grade that is closest to the actual power spectral density function; The higher the road grade, the smaller the road undulation, and the optimal control main force F is output. a , mainly to suppress the vibration of the vehicle body; if the road surface grade is low, the road surface will be undulating, and the output damping main force F b , mainly to recover energy; 3.

2. Decision based on the remaining power in the energy storage battery: Detect the remaining power in the energy storage battery. If the power is lower than the warning power, the damping active force F is output. b , mainly to recover energy; if the power is higher than the warning power, the optimal control main force F is output a , mainly to suppress vehicle body vibration.

3. The electronically controlled suspension method based on vibration energy recovery according to claim 1, characterized in that: The model of the linear DC motor described in step S4 is as follows: The electromagnetic force equation of a linear DC motor is: F m =k b B δ lNi=k m i In the formula, k b is the structural coefficient; B δ is the magnetic induction intensity; l is the effective length of each turn of the coil in the magnetic field; N is the number of turns of the coil; k m =k b B δ lN is the force constant of the linear motor; i is the coil current; The back electromotive force E of the armature winding is: In the formula, k E =k b B δ lN is the back electromotive force coefficient; x is the displacement of the mover, that is, the vertical displacement of the wheel; The voltage balance equation is: In the formula, R is the resistance of the coil winding; L is the inductance of the coil winding; the influence of the coil part is very small and can be ignored; i is the coil current; u is the power supply voltage; The output power calculation formula of the power supply is: When the power P is positive, energy is consumed, and when it is negative, energy is recovered.

Images