A high-safety and stable control system for a flywheel-motor-associated vehicle-mounted flywheel battery system

Abstract

This invention provides a highly safe and stable control system for a flywheel-motor-linked on-board flywheel battery system. The system includes a direct torque control loop and a speed correction loop. The direct torque control loop uses stator current and voltage to control the output voltage of flux observation correction, which is then fed into the direct torque control. The direct torque control also inputs a speed error and outputs a switching signal to control an AC-DC-AC matrix converter, forming a flux loop. The speed correction loop is used to obtain the fundamental component of the flywheel speed and vibration signal. The flywheel speed, the fundamental component of the vibration signal, a given speed, and the unbalanced current of the flywheel motor are fed into a speed cutoff negative feedback module to obtain the speed error. This invention can significantly improve the response speed and safety and stability control of the flywheel battery system.

Description

Technical Field

[0001] The present invention belongs to the field of control system design for vehicle-mounted flywheel batteries, and in particular relates to a high-safety and stable control system for a flywheel-motor associated vehicle-mounted flywheel battery system. Background Art

[0002] Flywheel batteries transcend the limitations of chemical batteries, offering advantages such as environmental friendliness, high energy storage efficiency, high instantaneous power, and a long service life. Their use in electric vehicles, combined with the primary power battery to form a composite power source, can significantly improve the vehicle's energy efficiency and performance. Flywheels experience a gyroscopic effect at high speeds, particularly during vehicle operation. This effect is exacerbated by the various forces generated during the transition between onboard, flywheel, and normal operating conditions, as well as the bandwidth limitations of amplifiers, filters, and sensors, and the effects of nonlinear factors. This gyroscopic effect can negatively impact the stability of the entire flywheel battery system. Furthermore, despite ideal flywheel rotor operation, various potential rotor-related failures pose a significant risk to the safe operation of the equipment. Therefore, ensuring the high safety and stable operation of automotive flywheel battery systems is crucial for their practical application.

[0003] The flywheel motor and the flywheel rotor supported by magnetic bearings are two key components of an on-board flywheel battery system. Designing a suitable control system for these two components directly determines the safety and stability of the system. Currently, achieving safe and stable operation of the flywheel rotor typically relies solely on the magnetic bearing control system. It is worth noting that the flywheel rotor is driven by the flywheel motor. In particular, for traditional flywheel battery systems with an inertial spindle, since the magnetic bearings and motor are located at different locations along the flywheel battery system's inertial spindle, the rotating shaft at the magnetic bearing location can cause deviation or vibration at the motor during adjustment, thus impacting the safety and stability of the flywheel motor. Without a corresponding safety and stability control system to ensure the motor's stability, the motor could be damaged and prone to failure. Furthermore, passive motor eccentricity or rotor vibration could affect the stability of the inertial spindle, further reducing the time and difficulty required to adjust the flywheel rotor back to its equilibrium position. While classic fault-tolerant control strategies can ensure safe motor operation, the design of motor safety control systems often fails to consider the impact of the flywheel rotor system on the flywheel motor, making it impossible to guarantee safe and stable operation of the entire system. Therefore, the control system of the flywheel motor and the flywheel controlled by the magnetic bearing influence each other and are inseparable.

[0004] At the same time, traditional thinking fails to effectively integrate stability control and safety control. For example, in the flywheel rotor, the mainstream approach is to use magnetic bearings to control the flywheel's position when the rotor is subject to external disturbances, thereby reducing vibration during operation. However, this approach attributes flywheel vibration to external disturbances under varying operating conditions, ignoring internal factors that contribute to flywheel vibration. While the vibration of the flywheel rotor is determined by external disturbances, internal factors also contribute. While these internal factors can be ignored during normal operation, they become a key factor in vibration during faults and should not be neglected. The concept of stable control does not fully consider system safety, and failures can have serious consequences. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a high-safety and stability control system for a flywheel-motor associated vehicle-mounted flywheel battery system, which significantly improves the response speed and safety and stability control effect of the flywheel battery system.

[0006] The present invention achieves the above technical objectives through the following technical means.

[0007] A high-safety and stable control system for a flywheel-motor-associated vehicle-mounted flywheel battery system, comprising:

[0008] Direct torque control loop, the sampled stator current and stator voltage control flux observation and correction output voltage u α 、u β , the output voltage u α 、u β Input direct torque control, direct torque control also input speed error △n*, output switching signal S(n) to control AC-DC-AC matrix converter to form a magnetic link; the speed error △n* is composed of flywheel speed n, the fundamental component A1 of flywheel shaft vibration signal A after filtering, given speed n* and unbalanced current i of flywheel motor. n Input speed cutoff negative feedback module to obtain;

[0009] A speed correction loop includes an eddy current displacement sensor, a saturable vibration suppression controller, a first amplifier, a first filter, and a speed cutoff negative feedback module. The eddy current displacement sensor is used to measure the flywheel speed n and the flywheel shaft vibration signal A. The vibration signal A is subjected to vibration reduction by the saturable vibration suppression controller to obtain a saturation error value e. The saturation error value e is sequentially passed through the first amplifier and the first filter to obtain the fundamental wave component A1.

[0010] The speed cutoff negative feedback module obtains a deceleration vibration amplitude limit value A1* corresponding to the speed based on the flywheel speed n; the speed cutoff negative feedback module obtains a power-off vibration amplitude limit value A2* based on the database of the signal processor; a critical stable amplitude error ΔA1 is obtained by subtracting A1* from the fundamental wave component A1, and a critical runaway amplitude error ΔA2 is obtained by subtracting A2* from the fundamental wave component A1; the critical stable amplitude error ΔA1 is amplified to obtain a speed correction value Δn;

[0011] When the given speed n* can be connected to the speed comparison, the speed correction value Δn and the speed n are subtracted from the given speed n* to obtain Δn*.

[0012] In the above technical solution, the speed cutoff negative feedback module is provided in the flywheel motor control system, and the speed cutoff negative feedback module is controlled by the flywheel rotor vibration reduction system.

[0013] In the above technical solution, the AC-DC-AC matrix converter commutates the five-phase electricity in the main circuit of the motor; the AC-DC-AC matrix converter includes an insulated gate bipolar transistor and a five-phase full-bridge rectifier circuit. The five-phase electricity is commutated through five groups of insulated gate bipolar transistors connected in reverse order to form a DC voltage U dc , the U dc The current signal obtained by the five-phase full-bridge rectifier circuit is input to the stator side of the flywheel motor as the stator current.

[0014] In the above technical solution, the A1* is calculated by connecting the speed n to the speed-vibration function Y(n). The acquisition process of the speed-vibration function Y(n) is as follows: adjust n* to obtain multiple groups of speeds n and their corresponding A max , taking the speed n as the independent variable, A max The speed-vibration function Y(n) is fitted as the dependent variable.

[0015] In the above technical solution, the A2*=min{A i *, A n *}, where A i * is the current-vibration allowable value, which is determined by looking up the table based on the unbalanced current and the replacement range time T. n * is the speed-vibration allowable value, which is determined by looking up the table based on the speed n and the replacement range time T.

[0016] In the above technical solution, the deceleration vibration error ΔA1' is obtained by subtracting the A1* from the fundamental wave component A1. The ΔA1' is filtered and then the critical stable amplitude error ΔA1 is obtained by taking only positive values ​​using the light emitting diode D.

[0017] In the above technical solution, the critical out-of-control amplitude error ΔA2 is used as a control signal to be connected to the relay KA, and the relay KA is used to control the on and off of the main circuit of the motor.

[0018] In the above technical solution, whether the given speed n* can be connected to the speed comparison is determined by the following process: the fundamental component A1 is subjected to signal processing and phase judgment to obtain the vibration phase error Δφ. When the inverted Δφ is 0, the given speed n* can be connected to the speed comparison.

[0019] In the above technical solution, different state types of the flywheel battery system are judged according to the values ​​of the critical stable amplitude error ΔA1, the critical out-of-control amplitude error ΔA2, and the vibration phase error Δφ, specifically:

[0020] When ΔA1=0, ΔA2=0 and Δφ=0, the flywheel battery system is in normal operation;

[0021] When ΔA1>0, ΔA2=0 and Δφ=0, the flywheel battery system is in a fault-suppressible state;

[0022] When ΔA1>0, ΔA2=0 or 1, and Δφ=1, the flywheel battery system is in an uncontrollable mechanical fault state;

[0023] When ΔA1>0, ΔA2=1 and Δφ=0, the flywheel battery system is in an uncontrollable electromagnetic fault state.

[0024] In the above technical solution, the easily saturated vibration suppression controller is a vibration suppression structure.

[0025] The beneficial effects of the present invention are:

[0026] (1) The present invention combines the safety control function of the flywheel rotor with the fault-tolerant control function of the flywheel motor, and uses them in the control system of the flywheel motor. It uses vibration detection technology to control the common variable speed of the two, and uses speed cutoff negative feedback to adjust the overall speed when the vibration is too large, thereby protecting the safety of the system; the safety control of the flywheel motor and the flywheel rotor are integrated to achieve the integrated safety control of the flywheel rotor and the motor part, saving relatively many components, reducing costs and losses, simplifying the structure and improving the reliability of the system; it solves the problem of the traditional flywheel system safety control idea of ​​controlling the flywheel motor and the flywheel rotor separately, resulting in the serious consequences of the flywheel motor continuing to work when the flywheel rotor fails, causing the system motor to stall or the flywheel rotor to run away.

[0027] (2) The present invention integrates vibration reduction control and safety assurance and realizes action scheduling and complementary action. The present invention uses the error signal of vibration reduction control to start the safety control system, which can start the safety control when the vibration reduction is out of control, realizes the action scheduling of the safety and stability control, and makes the overall structure more reasonable; eliminates the structural redundancy caused by the common ideas of the two control methods in the general distinction of actions; at the same time, the sampling signal of the safety control part of the present invention is different from that of the vibration reduction system, and the action point is placed on the flywheel motor, so that the action signal is changed from five position signals to one speed signal; while simplifying the control part, the two can complement each other; solves the problem that when the general safety control and stability control are running in parallel, the failure of the sampling point and the action point will cause both to be unable to act.

[0028] (3) The safety control used in the present invention places the safety control of the flywheel part in the control structure of the motor, uses the speed-vibration function to determine the deceleration vibration limit value A1* under normal circumstances, uses the relationship between the speed, unbalanced current and the replacement range time, and looks up the table to obtain the power-off vibration limit value A2*. Then, the empirical mode decomposition algorithm is used to judge the phase change fault, which can distinguish the faults of the flywheel motor and the flywheel rotor, and realizes a huge improvement in the system's analysis and processing capabilities for various faults, thereby greatly improving the system safety; and solves the problem that the general fault judgment system has no subsequent corresponding action, cannot clearly distinguish between various faults, and has incomplete fault judgment.

[0029] (4) The present invention is to provide a relay KA for controlling the five-phase power supply at the action point of the motor part, to set the speed, to set the speed correction value, and to use the deceleration vibration limit value A1*, the power-off vibration limit value A2* and the vibration phase error Δφ for control; therefore, it is possible to distinguish the control to ensure rationality while ensuring the reliability of the safety control; it realizes the distinction between major and minor faults, and mechanical and electromagnetic faults; compared with general safety control, it not only solves the problem of not distinguishing actions between major and minor faults, but also solves the problem that a failure of the safety control part will cause the entire safety control system to be unable to operate.

[0030] (5) The saturable vibration suppression controller of the present invention can be any one or several vibration suppression structures. Reducing the saturation error value can weaken its vibration suppression performance, thereby realizing the combined operation of the motor vibration detection technology and the flywheel vibration reduction technology, solving the contradiction and mutual influence between the two technologies; solving the conflict that reducing vibration will affect the accuracy of vibration detection, and ensuring the accuracy of vibration detection requires ensuring vibration.

[0031] (6) The present invention designs a matrix converter based on direct torque control and a five-phase permanent magnet synchronous motor, realizing the combination of the two. While retaining the strong robustness and simple calculation of direct torque control, it also accommodates the advantages of the five-phase permanent magnet synchronous motor, which is vibration-free, fault-tolerant, and has low torque pulsation. At the same time, the matrix converter has a high safety factor and low loss, solving the problems of large torque pulsation, unsafe commutation, and severe loss in general flywheel motor systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a block diagram of the high-safety and stable control system of the flywheel-motor associated vehicle-mounted flywheel battery system of the present invention;

[0033] Figure 2 This is a schematic diagram of the internal structure of the high-safety and stable control system of the flywheel-motor associated vehicle-mounted flywheel battery system of the present invention;

[0034] Figure 3 This is the speed cut-off negative feedback structure diagram of the present invention;

[0035] Figure 4 This is a structural diagram of the AC-DC-AC matrix converter of the present invention;

[0036] Figure 5 This is a simulation model diagram of the overall structure of the flywheel battery system described in the present invention. DETAILED DESCRIPTION

[0037] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the protection scope of the present invention is not limited thereto.

[0038] like Figure 1 As shown, the high-safety and stability control system of the present invention combines the flywheel rotor safety control and the flywheel motor fault tolerance, and places them together in the control system of the flywheel motor; the flywheel motor fault tolerance system is set in the flywheel motor control system to control the safety of the flywheel motor, and the flywheel rotor vibration reduction system is set in the flywheel rotor control system to control the stability of the flywheel rotor. In addition, since the safety control and stability control of the flywheel rotor have the same signal source and action point, the flywheel rotor vibration reduction system can control the vibration detection and speed cutoff negative feedback, and the vibration detection and speed cutoff negative feedback are set in the flywheel motor control system. Since the flywheel rotor and the flywheel motor have the same operating environment and speed, the safety of the flywheel rotor and the flywheel motor can be guaranteed at the same time.

[0039] See also Figure 2 The present invention is a three-loop safety and stability control system based on direct torque double closed-loop control.

[0040] 1. Direct Torque Control Loop

[0041] The inner and middle loops act as direct torque control loops to control the five-phase permanent magnet synchronous motor (FPMSM). In the main circuit, to achieve the effect of modulating the five-phase power and thus controlling the FPMSM, the five-phase power is commutated through an AC-DC-AC matrix converter. The design of the AC-DC-AC matrix converter is as follows: Figure 4 As shown, the five-phase electricity is commutated through five groups of insulated gate bipolar transistors IGBTs connected in reverse order to form a DC voltage U dc , U dc The current signal obtained by the five-phase full-bridge rectifier circuit is input into the stator side of the flywheel motor as the stator current. The matrix converter has little problem of safe commutation and can realize multiple power conversion functions such as frequency conversion, rectification and inversion.

[0042] The direct torque control flux loop of the flywheel motor is equipped with flux observation correction to achieve fault-tolerant control of the flywheel motor. The stator current (i a 、i b 、i c 、i d 、i e ) and stator voltage (u a 、u b 、u c 、u d 、u e ) Control the output voltage u of the flux observation correction α 、u β , output voltage u α 、u β Input direct torque control, direct torque control also input speed error △n*, output switching signal S(n) to control AC-DC-AC matrix converter, forming a flux loop. Among them, stator current and stator voltage control flux observation correction output voltage u α 、u β , output voltage u α 、u β The process of inputting direct torque control, inputting a speed error Δn*, and outputting a switching signal S(n) is all prior art.

[0043] The X and Y axis directions of the flywheel shaft are both equipped with eddy current displacement sensors (this embodiment uses KD2446 eddy current displacement sensors). The eddy current displacement sensors in both directions measure the speed n and input it into the speed cutoff negative feedback module. At the same time, the fundamental wave component A1 of the vibration signal A measured by the eddy current displacement sensor is filtered and input into the speed cutoff negative feedback module. In addition, the five-phase current signal i of the flywheel motor is a 、i b 、i c 、i d 、i e The unbalanced current i is obtained after the summation of then Input the speed cutoff negative feedback module, the cutoff negative feedback module inputs the given speed n* for correction, outputs the speed error △n*, and forms a middle loop with the direct torque control of the inner loop.

[0044] 2. Speed ​​Correction Ring

[0045] like Figure 2 As shown, the outer loop is the speed correction loop, which includes an eddy current displacement sensor, a saturable vibration suppression controller, a first amplifier, a first filter, and a speed cutoff negative feedback module.

[0046] The eddy current displacement sensor in the X and Y directions of the flywheel shaft can not only measure the speed n, but also the vibration signal. The measured vibration signal parameters are first sent to the saturable vibration suppression controller to reduce the vibration and obtain the saturation error value e. The first amplifier amplifies e into the signal E required for the first filter to turn on. The amplification factor of the first amplifier is K. a5 The settings are as shown in (1):

[0047] K a5 =E g / e min (1)

[0048] Among them: e min is the error value e when saturated, E g is the minimum signal required for the first filter to conduct. When E>E g When the first filter is turned on;

[0049] The first filter is a low-pass filter, which is used to filter out the fundamental wave component A1 from the measured vibration signal A after the first filter is turned on.

[0050] In this embodiment, the saturable vibration suppression controller can be any one or more vibration suppression structures. After receiving the vibration signal parameters measured by the eddy current displacement sensor, the saturable vibration suppression controller will feed back the corrected displacement signal to the flywheel shaft. The corrected displacement signal is set to an output limit, and the difference between the vibration signal parameters and the corrected displacement signal is obtained to obtain e, so that combined operation can be achieved. This embodiment takes the classic proportional-integral-differential (PID) control as an example. Assuming that the PID inputs a vibration signal parameter r(t) and feeds back the corresponding corrected displacement signal s(t), the function e(t) of e and time is:

[0051]

[0052] Where K1 is the proportional coefficient of PID, T t is the integral time constant of PID, T Dis the differential time constant of PID; adjust e min It can not only achieve the vibration reduction effect but also prevent fault vibration from being incorrectly weakened.

[0053] 3. Speed ​​cut-off negative feedback

[0054] Figure 3 It is the internal structure of the speed cutoff negative feedback module, which consists of a speed-vibration function Y(n), a second amplifier, a phase judgment algorithm, a signal processor, a second filter, an inverter, a latch, a light-emitting diode D and a relay KA. Relay KA is also used to shut down the five-phase permanent magnet synchronous motor FPMSM.

[0055] Connect the speed n to the speed-vibration function Y(n) and calculate the corresponding speed deceleration vibration limit value A1*; according to n, i n The value of and the range change time T are used to determine the power-off vibration limit value A2* by looking up the table; the function of the signal processor is to obtain the amplitude and phase of the input signal and output it, or to obtain the difference by subtracting it from the given value. The difference can be output as an analog quantity or a digital quantity. A1* and A2* are input into the signal processor as given values. The signal processor subtracts the deceleration vibration limit value A1* from the fundamental component A1 to obtain the deceleration vibration error ΔA1'. ΔA1' is filtered through the second filter to eliminate the ripple, and then the light-emitting diode D is used to take only the positive value to obtain the critical stable amplitude error ΔA1. ΔA1 is an analog quantity. At the same time, the signal processor subtracts A2* from A1 and converts the difference result into a digital quantity ΔA2. ΔA2 represents the critical out-of-control amplitude error. The critical out-of-control amplitude error ΔA2 is then used as a control signal to connect to the relay KA, which is used to control the on and off of the main circuit of the motor.

[0056] The analog value ΔA1 is connected to the second amplifier to amplify K P After multiplying, the speed correction value Δn is obtained. Δn is negatively fed back to the speed comparison. This link is called speed cutoff negative feedback, that is, the speed decreases as the vibration increases. In the amplification function of the second amplifier, K P Size design:

[0057]

[0058] At the same time, the signal processor derives the vibration phase φ based on the fundamental component A1. After the vibration phase φ is input into the phase judgment, the vibration phase error Δφ is obtained. Δφ is a digital quantity. If Δφ is 1, that is, φ has changed significantly, indicating that there is a crack in the rotor or a foreign object. After the vibration phase error Δφ is inverted by the inverter, if it is 1, the latch is blocked. After the latch is blocked, n* cannot be connected to the speed comparison. If it is 0, it is directly connected to the speed comparison. The speed correction value Δn and the speed n are subtracted from n* to obtain Δn*.

[0059] The process of determining A1*, A2*, and Δφ will be specifically described below.

[0060] (1) Use Adams to build a simulation model of the overall structure of the flywheel battery system (as Figure 5 shown), apply random noise excitation to the flywheel rotating shaft, and obtain the maximum value A of A1 according to the signal processor max ; adjust n*, obtain multiple sets of rotational speeds n and their corresponding A max , take n as the independent variable and A max as the dependent variable to fit the rotational speed-vibration function Y(n). Input the rotational speed n into the rotational speed-vibration function Y(n) to obtain the corresponding A1*.

[0061] (2) It is planned to determine A2* using a three-segment protection time limit standard, and the time elapsed in the replacement range is selected as 0 seconds, 0.5 seconds, and 1 second;

[0062] Let:

[0063] A2* = min{A i *, A n *} (4) <000027i3>Among them: A i * is the current-vibration allowable value, which is determined by the relationship between the current and the replacement range time T; A n * is the rotational speed-vibration allowable value, which is determined by the relationship between the rotational speed and the replacement range time T. The determination methods of A i *, A n * are shown in Table 1 and Table 2.

[0065] Table 1 A i * look-up table values

[0066]

[0067] Table 2 A n * look-up table values

[0068]

[0069]

[0070] A in the two tables m refers to the vibration mechanical allowable value of the flywheel rotor, which is generally equal to the air gap size between the mechanical protection bearing and the flywheel rotor in the vehicle-mounted flywheel battery system. I1 in Table 1 represents the upper limit of the rated unbalanced current. According to the calculation of the maximum unbalanced current of the flywheel motor, the magnitude of I1 is 0.5 A. When i n < I1, there is no power failure in the system; I2 represents the maximum open-circuit phase current. According to the measurement of the phase current magnitude under the maximum operating state of the motor, I2 = 10I1 = 5 A. When i nWhen I2, there is no short - circuit fault in the system. N1 in Table 2 represents the first upper - limit speed. According to the probability estimation of the flywheel rotor falling into chaotic motion, N1 is set to 25000 r / s; N2 represents the second upper - limit speed, and N2 is set to 40000 r / s. Beyond this speed, the flywheel rotor will not be a rigid body; N n represents the minimum rated speed of the flywheel rotor, and N n is set to 14000 r / s. When it is lower than this speed, the rotor is in a stall state or a starting state.

[0071] According to n, i n and the replacement - range time T, query the values in Tables 1 and 2, and combine with Equation (4) to output A2*. <>

[0072] (3) The vibration phase φ obtains the vibration - phase error Δφ through the phase - judgment algorithm. The specific steps are as follows:

[0073] Step 1), use the vibration signal when there is a crack in the flywheel and the orbital vibration - phase signal of the normal flywheel as the selection target of the training samples; collect the training - sample signals and perform improved empirical - mode (Empirical Mode Decomposition, EMD) decomposition;

[0074] Step 1.1), use the empirical - mode - decomposition algorithm to decompose φ n(i) =φ n ±β i w n(i) to obtain the first IMF component imf1 of the improved EMD:

[0075]

[0076] where, E 1+ (φ n(i) ) is the first IMF component obtained by using EMD decomposition of φ n(i) =φ n +β i w n(i) E 1- (φ n(i) ) is the first IMF component obtained by using EMD decomposition of φ n(i) =φ n -β i w n(i) And let w n(i) be a group of white - noise signals with a mean of zero and an amplitude standard deviation of a constant C. The crack or the vibration - phase signal as the normal state and the original training sample is φ n , and its signal after the i - th noise addition is φ n(i) , and the imf component of the signal after the i - th noise addition is imf i, I is the total number of times noise is added, β i is the white noise intensity;

[0077] Step 1.2), calculate the residual r1 of the first IMF component imf1:

[0078] r1=φ n -imf1 (6)

[0079] Step 1.3), use EMD decomposition to calculate r1±β i w n(i) The first mode of is defined as the second IMF component imf2 of the improved EMD decomposition:

[0080]

[0081] Among them, E 1- (r1-β i w (i) ) is r1+β i w (i) The first IMF component obtained after EMD decomposition, E 1- (r1-β i w (i) ) is r1-β i w (i) The first IMF component obtained after EMD decomposition;

[0082] Step 1.4), and so on, until the vibration phase signal φ is obtained n The Kth IMF component of n The number of IMF components obtained by the final decomposition.

[0083] Step 2) Select the decomposed oscillation function with time-varying frequency, that is, the IMF component imf j ;

[0084] Step 2.1), let the training sample vibration phase signal φ n The discretized signal sequence is Φ={φ(1),φ(2)...φ(n)}, imf i The discrete signal sequence is {y i (1),y i (2)..y i (n)}, define the mutual correlation coefficient ρ between the two xyi for:

[0085]

[0086] Solve all imfs i and the vibration phase signal φ nThe mutual correlation coefficient of K IMF components is obtained, and the standard deviation SD of the mutual correlation coefficient of K IMF components is obtained;

[0087] Step 2.2), assume For {y i (1),y i (2)..y i (n)}, define imf i The signal kurtosis value Kurtosis is:

[0088]

[0089] Calculate all IMFs i The signal kurtosis value is calculated and the threshold ε of the signal kurtosis value is set according to the result;

[0090] Step 2.3) Select and retain all signals with kurtosis values ​​greater than the threshold ε and with a correlation coefficient ρ xyi The IMF components that are larger than the standard deviation SD are rearranged in order of their original numbers and are recorded as {imf j}, where j = 1, 2....m, and m ≤ K.

[0091] Step 3), use the selected IMF component imf j The corresponding time-frequency feature quantity is solved and used as the input signal of the training sample; the fault feature corresponding to the training sample signal is selected as the label value of the training sample to generate the training sample data.

[0092] Step 4), performing improved BP (back propagation) neural network training to obtain a flywheel rotor crack model and perform flywheel rotor crack identification;

[0093] The input sample and the corresponding training sample label value are used as unit sample data, which are sent to the BP neural network for training, and the BP neural network is trained iteratively based on the batch gradient descent method, where the weight w ij The iterative formula for (t) is:

[0094]

[0095] Where: η(t) is the learning rate, is the gradient function, J(t) is the cost function, t is the number of iterative training, and t=0, 1...t max ;

[0096] Define η(0) as the initial custom learning rate. When performing the first gradient descent, record all w ij (t) corresponding When t=1,2...t max When, define Using the learning rate and gradient function, w ij (t) Iterate.

[0097] The trained BP neural network is used as the phase judgment algorithm, the vibration phase φ is input and the IMF quantities are calculated. The corresponding time-frequency feature quantities are solved using the selected IMF components and used as the input signal of the training sample; the fault feature corresponding to the training sample signal is selected as the label value of the training sample to generate the training sample data, specifically as follows: for all the IMF components imf selected in step 2.3) j The kurtosis factor, energy, and energy moment are evaluated (the solution process is the existing technology), and their numerical results are used as input samples in the form of K-dimensional vectors; the label value is set to 1 dimension, and 0 is set for a normal flywheel and 1 is set for a cracked flywheel, and the value is output as Δφ.

[0098] 4. Distinguishing and judging the status types of flywheel battery systems

[0099] Based on the above-mentioned high-safety and stability control system, a standard is formulated to distinguish the status types of the flywheel battery system. The specific status types are as follows:

[0100] W1: It is called normal operating state, including flywheel rotor scraping, flywheel rotor misalignment, and flywheel motor current resonance;

[0101] W2: called suppressible fault state, including vacuum chamber leakage, flywheel motor demagnetization and flying fault;

[0102] W3: It is called uncontrollable mechanical failure state, including flywheel rotor crack failure and foreign matter scratching inside the rotor;

[0103] W4: It is called the unsuppressible electromagnetic fault state, including the stator side short circuit of the flywheel motor and the inter-turn short circuit fault of the flywheel motor;

[0104] Table 3 shows the relationship between the state type and the critical stability amplitude error ΔA1, critical loss of control amplitude error ΔA2, and vibration phase error Δφ in the high safety and stability control system:

[0105] Table 3 Relationship between state type and ΔA1, ΔA2, Δφ

[0106] type <![CDATA[ΔA1]]> <![CDATA[ΔA2]]> Δφ <![CDATA[W1]]> 0 0 0 <![CDATA[W2]]> >0 0 0 <![CDATA[W3]]> >0 1 / 0 1 <![CDATA[W4]]> >0 1 0

[0107] Therefore, the different status types of the flywheel battery system can be determined according to the values ​​of ΔA1, ΔA2, and Δφ.

[0108] The embodiments described are preferred implementations of the present invention, but the present invention is not limited to the above-mentioned implementations. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention are within the scope of protection of the present invention.

Claims

1. A high safety and stability control system for a flywheel-motor-associated vehicle-mounted flywheel battery system, characterized in that: include: Direct torque control loop, the sampled stator current and stator voltage control flux observation and correction output voltage u α 、u β , the output voltage u α 、u β Input direct torque control, direct torque control also input speed error △n*, output switching signal S(n) to control AC-DC-AC matrix converter to form a magnetic link; the speed error △n* is composed of flywheel speed n, the fundamental component A1 of flywheel shaft vibration signal A after filtering, given speed n* and unbalanced current i of flywheel motor. n Input speed cutoff negative feedback module to obtain; A speed correction loop includes an eddy current displacement sensor, a saturable vibration suppression controller, a first amplifier, a first filter, and a speed cutoff negative feedback module. The eddy current displacement sensor is used to measure the flywheel speed n and the flywheel shaft vibration signal A. The vibration signal A is subjected to vibration reduction by the saturable vibration suppression controller to obtain a saturation error value e. The saturation error value e is sequentially passed through the first amplifier and the first filter to obtain the fundamental wave component A1. The speed cutoff negative feedback module obtains a deceleration vibration amplitude limit value A1* corresponding to the speed based on the flywheel speed n; the speed cutoff negative feedback module obtains a power-off vibration amplitude limit value A2* based on the vibration amplitude upper limit database; the signal processor subtracts A1* from the fundamental wave component A1 to obtain a critical stable amplitude error ΔA1, and the signal processor subtracts A2* from the fundamental wave component A1 to obtain a critical runaway amplitude error ΔA2; the critical stable amplitude error ΔA1 is amplified by a second amplifier to obtain a speed correction value Δn; When the given speed n* can be connected to the speed comparison, the speed correction value Δn and the speed n are subtracted from the given speed n* to obtain Δn*; The A1* is calculated by connecting the speed n to the speed-vibration function Y(n). The process of obtaining the speed-vibration function Y(n) is as follows: adjusting n*, obtaining multiple sets of speed n and the corresponding A max , take the speed n as the independent variable, A max The speed-vibration function Y(n) is fitted as the dependent variable; A2*=min{A i *, A n *}, where A i * is the current-vibration allowable value, which is determined by looking up the table based on the unbalanced current and the replacement range time T. n * is the speed-vibration allowable value, which is determined by looking up the table based on the speed n and the replacement range time T; The deceleration vibration error ΔA1' is obtained by subtracting the A1* from the fundamental wave component A1. The ΔA1' is filtered by a second filter and then the critical stable amplitude error ΔA1 is obtained by taking only positive values ​​using the light-emitting diode D. The critical out-of-control amplitude error ΔA2 is used as a control signal to connect to the relay KA, and the relay KA is used to control the on and off of the main circuit of the motor; Whether the given speed n* can be connected to the speed comparison is determined by the following process: the fundamental component A1 is judged by signal processing and phase judgment algorithm to obtain the vibration phase error Δφ. After the Δφ is inverted by the inverter, if it is 1, the latch is blocked. If it is 0, the given speed n* can be connected to the speed comparison.

2. The high safety and stability control system for a flywheel-motor-associated vehicle-mounted flywheel battery system according to claim 1, characterized in that: The speed cutoff negative feedback module is arranged in the flywheel motor control system, and the speed cutoff negative feedback module is controlled by the flywheel rotor vibration reduction system.

3. The high safety and stability control system for a flywheel-motor-associated vehicle-mounted flywheel battery system according to claim 1, characterized in that: The AC-DC-AC matrix converter commutates the five-phase electricity in the main circuit of the motor; the AC-DC-AC matrix converter includes an insulated gate bipolar transistor and a five-phase full-bridge rectifier circuit. The five-phase electricity is commutated through five groups of insulated gate bipolar transistors connected in reverse order to form a DC voltage U dc , the U dc The current signal obtained by the five-phase full-bridge rectifier circuit is input to the stator side of the flywheel motor as the stator current.

4. The high safety and stability control system for a flywheel-motor-associated vehicle-mounted flywheel battery system according to claim 1, characterized in that: According to the values ​​of the critical stable amplitude error ΔA1, the critical out-of-control amplitude error ΔA2, and the vibration phase error Δφ, different state types of the flywheel battery system are judged, specifically: When ΔA1=0, ΔA2=0 and Δφ=0, the flywheel battery system is in normal operation; When ΔA1>0, ΔA2=0 and Δφ=0, the flywheel battery system is in a fault-suppressible state; When ΔA1>0, ΔA2=0 or 1, and Δφ=1, the flywheel battery system is in an uncontrollable mechanical fault state; When ΔA1>0, ΔA2=1 and Δφ=0, the flywheel battery system is in an uncontrollable electromagnetic fault state.

5. The high safety and stability control system for a flywheel-motor-associated vehicle-mounted flywheel battery system according to claim 1, characterized in that: The easily saturated vibration suppression controller is a vibration suppression structure.

Images