Torque feedforward adaptive method based on rotational inertia of elevator system

Abstract

The application is a torque feedforward adaptive method based on the rotational inertia of an elevator system. First, it judges whether the rotational inertia self-learning of the elevator system is completed. If the rotational inertia self-learning of the elevator system is completed, the load information is read and the system rotational inertia is calculated according to the Newton interpolation method according to the read load information, and then the acceleration feedforward torque is calculated as the feedforward to the current controller. If the rotational inertia self-learning of the elevator system is not completed, the car is run up and down in the empty load and full load, and the intermediate step torque current in the up and down running process of the car in the empty load and full load is captured. The empty load system rotational inertia and the full load system rotational inertia are calculated for self-learning. According to the load information of the elevator system in the actual working condition, the feedforward control realizes the adaptive control of the change of the rotational inertia of the elevator system caused by the change of the load, reduces the burden of the feedback control, and realizes the low-bandwidth feedback control configuration.

Description

Technical Field

[0001] This invention relates to the field of elevator technology, and in particular to a torque feedforward adaptive method based on the rotational inertia of an elevator system. Background Technology

[0002] Feedforward control, based on the dynamic equations of the load and traction system, determines the required torque control quantity during control. Feedback control, on the other hand, uses sensor measurements to determine the deviation between the actual system value and a reference value, performing correction control. Precise feedforward control can reduce the burden on feedback, enabling low-bandwidth feedback control configurations.

[0003] Existing feedforward control cannot achieve adaptive control of system inertia changes caused by load variations. In other words, feedforward control cannot achieve real-time load tracking, which will cause a certain deviation between the feedforward torque signal and the actual required torque signal. Therefore, more feedback control is needed to achieve speed and current tracking, and the weight of the feedback signal will increase.

[0004] like Figure 1 As shown, the elevator traction system consists of two parts: a rotary frame and a linear frame. The elevator hoist machine has the car system and counterweight system on either side, connected by a hoist pulley 1 and guide pulleys 2. Depending on the desired lifting ratio, movable pulleys are provided on the car 3 and counterweight system 4. The car 3 rolls or slides on the guide rail via guide wheels and guide shoes. Here, the rotary frame refers to the various pulley systems within the elevator traction system, while the linear frame refers to the car system and counterweight system.

[0005] In summary, an elevator system is a multi-axis system. To simplify calculations, instead of studying the problems on each axis in detail, the motion relationships of other axes are equated to the traction machine axis. A practical multi-axis drive system is simplified into a single-axis rotary drive system. The principle for calculating the load torque is the power components before and after the conversion.

[0006] For rotary motion, power = torque × angular velocity. Considering the transmission efficiency of the transmission mechanism, the losses of the transmission mechanism are borne by the electric motor, thus...

[0007] T g Ω g =T L Ω L ηc

[0008] Among them, T g T represents the actual load torque of the working mechanism. L Ω is the torque of the working mechanism referred to the traction machine shaft. g Ω represents the actual angular velocity of the working mechanism. L Let η be the angular velocity of the traction wheel. c The transmission efficiency of the transmission mechanism is the product of the transmission efficiencies at each stage, η1η2….

[0009] From the above formula, the converted load torque can be obtained as follows:

[0010]

[0011] in Let n be the transmission speed ratio of the transmission mechanism. L n is the rotational speed measured at the equivalent shaft of the traction machine. g This refers to the rotational speed of each stage of the rotating mechanism. For the movable pulley, the transmission speed ratio is 2, which is called the traction ratio in the elevator industry.

[0012] For elevator systems, the load torque T L This actually refers to the unbalanced torque of the car and counterweight system. Specifically,

[0013] When the car is lifted upwards, for a car driven by a moving pulley, the transmission mechanism's losses are directed downwards and borne by the traction machine shaft.

[0014]

[0015] Here This means that a movable pulley requires half the lifting force compared to a fixed pulley, m car R is the mass of the car, g is the acceleration due to gravity, and R is the acceleration due to gravity. car Ω is the radius of the car side moving pulley. car T is the angular velocity of the car's side-moving pulley. L_car The load torque converted from the lifting torque on the car side to the load torque on the traction sheave side. To obtain the efficiency of the car's upward transmission (including bearing losses in the pulley system and frictional losses between the traction steel wire rope and the pulley system), we get...

[0016]

[0017] in This refers to the speed ratio between the car return sheave and the traction sheave.

[0018] For a counterweight driven by a movable pulley, the direction of movement is downward, while the transmission mechanism's losses are upward and borne by the transmission mechanism.

[0019]

[0020] Here This means that a movable pulley requires half the lifting force compared to a fixed pulley, m cwt The mass of the weight, R cwt Ω is the radius of the movable pulley on the counterweight side. cwt For the angular velocity of the counterweight side movable pulley, T L_cwt To calculate the load torque on the traction sheave side by converting the counterweight side descent torque to the load torque on the counterweight side, To obtain the efficiency of the counterweight downshift transmission (including bearing losses in the gear train and frictional losses between the traction wire rope and the gear train),

[0021]

[0022] in, This refers to the speed ratio between the return sheave and the traction sheave.

[0023] When the car descends and releases, for a car driven by a movable pulley, the transmission mechanism's losses are directed downwards and borne by the transmission mechanism side.

[0024]

[0025] get

[0026]

[0027] in The speed ratio between the car return sheave and the traction sheave. The efficiency of the car's downward transmission (including bearing losses in the gear train and frictional losses between the traction steel wire rope and the gear train).

[0028] For a counterweight driven by a movable pulley, the direction of movement is upward, the transmission mechanism's losses are downward, and the load is borne by the traction machine shaft.

[0029]

[0030] get

[0031]

[0032] in To determine the speed ratio between the return sheave and the traction sheave, The efficiency of the counterweight upward transmission (including bearing losses of the gear train and frictional losses between the traction wire rope and the gear train).

[0033] For worm gear type forced drive elevators, since there is no counterweight, η can still demonstrate the self-locking effect of the worm gear mechanism. Specifically...

[0034] When the car travels upwards, the power loss p of the transmission mechanism is reduced because the losses are borne by the traction side. c = Traction machine power - Load power

[0035] From the above, we obtain the load torque from the car side to the traction sheave side.

[0036]

[0037] get

[0038]

[0039] When the car descends, the power loss of the transmission mechanism is borne by the transmission mechanism, and the power loss p of the transmission mechanism is... c = Load power - Traction machine power

[0040] From the above, we obtain the load torque from the car side to the traction sheave side.

[0041]

[0042] get

[0043]

[0044] For the car, the losses in the transmission mechanism can be considered constant during lifting and lowering, thus yielding...

[0045]

[0046] Push

[0047]

[0048] like but This indicates that the load power is insufficient to overcome the losses in the transmission mechanism, therefore a traction machine is still needed to provide power. In other words, without the traction machine's push, the heavy object cannot be lowered; this is the transmission mechanism's operation.

[0049] The self-locking function of the moving mechanism.

[0050] In summary, the load torque equivalent to the traction sheave side is...

[0051] T L =±T L_car ±T L_cwt

[0052] The following section introduces the reduced moment of inertia (or flywheel moment). Moment of inertia is a measure of the inertia of a rigid body rotating about an axis of rotation. It reflects the magnitude of the mechanical inertia of a moving object, and the principle of reduction is to ensure that the kinetic energy remains unchanged before and after reduction. Reductions are included for linear and rotating systems.

[0053] First, we will introduce the calculation of the moment of inertia of the rotating system. The rotating components include the counterweight wheel, the car bottom wheel, and the traction wheel, which are considered as disks with uniformly distributed mass.

[0054] According to the definition of moment of inertia,

[0055] J=∫r 2 dm

[0056] If we consider the disk as being composed of many small rings, and choose a infinitesimal element dm, then we have

[0057]

[0058] Where σ is the density, r is the radius of the infinitesimal element, s is the area of ​​the infinitesimal element, and R is the radius of the disk.

[0059]

[0060] Therefore, we obtain

[0061]

[0062]

[0063]

[0064] Among them, J CWT M is the moment of inertia of the heavy wheel. CWT The mass of the heavy wheel, R CWT J is the radius of the counterweight; CAR It is the moment of inertia of the car's bottom wheel, M. CAR It is the mass of the car's undercarriage wheels, R CAR J is the radius of the car's undercarriage wheels; HOST It is the moment of inertia of the traction sheave, M. HOST It is the mass of the traction sheave, R HOST It is the radius of the traction sheave.

[0065] Adding these three together, we obtain the moment of inertia of the rotating frame.

[0066] J Rotate =J CWT +J CAR +J HOST

[0067] Next, we will introduce the calculation of the moment of inertia of a linear system. For a rotating system, the kinetic energy is... For a linear system, the kinetic energy is Since the two are equal, we can obtain

[0068]

[0069] because

[0070] v = ΩR

[0071] get

[0072] J′=mR 2

[0073] This relationship can also be derived using the parallel axis theorem.

[0074] J′=J+mR 2

[0075] Where J is the moment of inertia of the original axis of rotation, and J′ is the moment of inertia of the new axis of rotation obtained after translating the original axis of rotation by a distance R. Treating the linear system components as point masses, then J = 0, therefore... available

[0076] m=(P+Q+P-kQ+W r1 +W r2 +0.5×W r3 ) / λ

[0077] Where P is the car's own weight, Q is the rated load, k is the balance coefficient, and W... r1 It is the weight of the wire rope, W r2 It is the weight of the compensation chain, W r3 λ is the weight of the accompanying cable, and λ is the traction ratio.

[0078] The total moment of inertia of the system is the sum of the moments of inertia of the rotating frame and the linear frame, that is...

[0079] J System =J Rotate +J′

[0080] The frictional force on the equivalent guide traction side of the elevator traction system is mainly manifested on the guide shoes, and is related to the normal force perpendicular to the guide rail side they experience. This normal force is related to the eccentric load of the car. When the eccentric load of the car is constant, the sliding friction force can be considered to remain constant.

[0081] When the system is moving upwards under no-load conditions, during the uniform speed operation phase, the force situation of the system is as follows: Figure 2 As shown, where f unbalanced This refers to the unbalanced force of the system, f riction It is kinetic friction, T e It is electromagnetic torque

[0082] Since the counterweight mass is heavier than the car mass, the unbalanced torque of the system is upward, and the direction of dynamic friction is opposite to the direction of motion, which is downward. According to Newton's first law, the direction of electromagnetic torque is downward.

[0083] When the system is descending under no-load conditions, during the uniform speed operation phase, the force situation of the system is as follows: Figure 3 As shown, since the counterweight is heavier than the car, the unbalanced torque of the system is upward, and the direction of dynamic friction is opposite to the direction of motion, also upward. According to Newton's first law, the direction of electromagnetic torque is downward. It should be noted that since the directions of friction during unloaded upward and downward travel are opposite, one acts as an assist force and the other as a resistance force relative to the electromagnetic torque. Therefore, the electromagnetic torque required for unloaded downward travel is greater than that for unloaded upward travel.

[0084] When the system is fully loaded and moving upwards, during the uniform speed operation phase, the force situation of the system is as follows: Figure 4 As shown, since the counterweight mass is lighter than the car mass, the unbalanced torque of the system is downward, the direction of the dynamic friction force is opposite to the direction of motion and is downward. According to Newton's first law, the direction of the electromagnetic torque is upward.

[0085] When the system is fully loaded and descending at a constant speed, the force situation of the system is as follows: Figure 5 As shown, since the counterweight mass is lighter than the car mass, the unbalanced torque of the system is downward, and the direction of the dynamic friction force is opposite to the direction of motion, which is upward. According to Newton's first law, the direction of the electromagnetic torque is upward.

[0086] When the car's load is 50%, assuming the elevator system's balance coefficient (i.e., unbalanced mass on both sides of the traction sheave / rated load * 100%) is 50%, then uniform speed operation can be achieved by overcoming only very small frictional resistance. During acceleration and deceleration, the unbalanced torque can be considered zero. When frictional torque is ignored, we can obtain...

[0087]

[0088] because

[0089] T e =K it I

[0090] Where I is the torque current, K it The torque / current coefficient,

[0091] get

[0092]

[0093] When the load is not 50%, the calculation must take into account the unbalanced torque T. unbanlancedThat is, the drag equation is

[0094]

[0095] The plus or minus signs here are determined by different loads and different acceleration / deceleration conditions. For example, when accelerating upwards without a load, the car is lighter than the counterweight, the unbalanced torque is upward, so the electromagnetic torque is downward, and the frictional resistance is downward. If we take upward as the positive direction, then the sign before the electromagnetic torque is negative, and the sign before the unbalanced torque is positive.

[0096] Based on the analysis above, please refer to the table below for specific symbol values ​​(all directions are positive, with the direction of movement being the primary direction).

[0097]

[0098]

[0099] The feedforward torque described in this invention is T obtained through this drive equation. em

[0100] Taking unloaded uplink acceleration as an example, the actual compensation value required is...

[0101]

[0102] in This is an unbalanced torque current, directed downwards; This is the static bias compensation amount, directed downwards, and its value is slightly smaller than the unbalanced torque current. This is the dynamic compensation quantity related to acceleration. Acceleration feedforward compensation and PI regulation are applied to the dynamic compensation quantity, while static bias compensation is controlled by a PI controller and weighing pre-compensation.

[0103] Taking the measured disc traction machine system as an example, the image of the unloaded upward movement is as follows: Figure 6 As shown, Figure 6 Positive values ​​indicate an upward direction, while negative values ​​indicate a downward direction.

[0104] It can be seen that the direction of dynamic acceleration compensation is upward during the acceleration phase and downward during the deceleration phase.

[0105] pass Figure 6 It can be seen that when moving upwards under no-load conditions, the torque current is downwards; when moving at a constant speed, the static bias compensation torque current is downwards; and the unbalanced torque current is downwards. In static conditions, the torque current is the unbalanced torque current; during dynamic operation, the torque current is the vector superposition of the static bias compensation torque current and the acceleration feedforward dynamic compensation current.

[0106] Taking downlink acceleration under no-load conditions as an example, the actual compensation value required is...

[0107]

[0108] in This is an unbalanced torque current, directed downwards; This is the static bias compensation amount, directed downwards, and its value is slightly greater than the unbalanced torque current. This is the dynamic compensation quantity related to acceleration. Acceleration feedforward compensation and PI regulation are applied to the dynamic compensation quantity, while static bias compensation is controlled by a PI controller and weighing pre-compensation.

[0109] Taking the measured disc traction machine system as an example, the image of the unloaded downward movement is as follows: Figure 7 As shown, Figure 7 Positive values ​​indicate an upward direction, while negative values ​​indicate a downward direction.

[0110] It can be seen that the direction of acceleration feedforward dynamic compensation is downward during the acceleration phase and upward during the deceleration phase.

[0111] pass Figure 7 It can be seen that during unloaded downward movement, the torque current is downward; during constant speed movement, the static bias compensation torque current is downward; and the unbalanced torque current is downward. In static conditions, the torque current is the unbalanced torque current; during dynamic operation, the torque current is the vector superposition of the static bias compensation torque current and the acceleration feedforward dynamic compensation current.

[0112] Taking full-load uplink acceleration as an example, the actual compensation value required is...

[0113]

[0114] in This is an unbalanced torque current, directed upwards; This is the static bias compensation amount, directed upwards, and its value is slightly greater than the unbalanced torque current. This is the dynamic compensation quantity related to acceleration. Acceleration feedforward compensation and PI regulation are applied to the dynamic compensation quantity, while static bias compensation is controlled by a PI controller and weighing pre-compensation.

[0115] Taking the measured disc traction machine system as an example, the image of the fully loaded upward movement is as follows: Figure 8 As shown, Figure 8 Positive values ​​indicate upward direction, while negative values ​​indicate downward direction.

[0116] It can be seen that the direction of acceleration feedforward dynamic compensation is upward during the acceleration phase and downward during the deceleration phase.

[0117] pass Figure 8It can be seen that when fully loaded and moving upwards, the torque current is upwards; when moving at a constant speed, the static bias compensation torque current is upwards; and the unbalanced torque current is upwards. In static conditions, the torque current is the unbalanced torque current; during dynamic operation, the torque current is the vector superposition of the static bias compensation torque current and the acceleration feedforward dynamic compensation current.

[0118] Taking full-load downlink acceleration as an example, the actual compensation value required is...

[0119]

[0120] in This is an unbalanced torque current, directed upwards; This is the static bias compensation amount, directed upwards, and its value is slightly smaller than the unbalanced torque current. This is the dynamic compensation quantity related to acceleration. Acceleration feedforward compensation and PI regulation are applied to the dynamic compensation quantity, while static bias compensation is controlled by a PI controller and weighing pre-compensation.

[0121] Taking the measured disc traction machine system as an example, the image of a fully loaded downward traction machine is as follows: Figure 9 As shown, Figure 9 Positive values ​​indicate upward direction, while negative values ​​indicate downward direction.

[0122] It can be seen that the direction of acceleration feedforward dynamic compensation is downward during the acceleration phase and upward during the deceleration phase.

[0123] pass Figure 9 It can be seen that during full-load downward movement, the torque current is upward; during constant speed movement, the static bias compensation torque current is upward; and the unbalanced torque current is upward. In static conditions, the torque current is the unbalanced torque current; during dynamic operation, the torque current is the vector superposition of the static bias compensation torque current and the acceleration feedforward dynamic compensation current.

[0124] From the above examples, it can be concluded that elevator systems are actually four-quadrant control systems, such as... Figure 10 As shown in the electromagnetic torque-angular velocity diagram, the first quadrant represents the motor state, which consumes electrical energy and corresponds to full-load upward movement; the second quadrant represents the generator state, which generates electrical energy and corresponds to full-load downward movement; the third quadrant represents the motor state, which consumes electrical energy and corresponds to no-load downward movement; and the fourth quadrant represents the generator state, which generates electrical energy and corresponds to no-load upward movement.

[0125] System rotational inertia J System It can be obtained from the load using the calculation method described above, or it can be obtained through experimental testing. The experimental method is introduced below.

[0126] The moment of inertia of the system is solved using recursive least squares. The applicable model for ARX (Autoregressive eXogenous out-of-band input) is defined as follows:

[0127] a(z)y k =b(z)u k +e k

[0128] Where e k The noise is Gaussian white noise, meaning it has a mean of 0, is independently and identically distributed, and has a variance of σ. 2 Rearranging items yields...

[0129] y k =(1-a(z))y k +b(z)u k +e k

[0130] Define the prediction model as

[0131]

[0132] Where z is the time-shift operator, this equation is a difference equation, and when rearranged into a matrix, it becomes:

[0133]

[0134] This matrix contains historical information.

[0135] The cost function is designed as the sum of squares of the errors of the real system and the prediction model.

[0136]

[0137] The goal is to find θ that minimizes the evaluation equation.

[0138] The above formula can be expressed in matrix form as follows:

[0139] f(θ) = ε T ε=(y-Φθ) T (y-Φθ)

[0140] =y T yy T Φθ-θ T Φ T y+θ T Φ T Φθ

[0141] =y T y-2θ T Φ T y +θ T Φ T Φθ

[0142] Simplified to:

[0143]

[0144] Where H = 2Φ T Φ (Hessian Matrix / Covariance Matrix, Autocorrelation Matrix), f = -2Φ T y (Cross-correlation matrix)

[0145] Find the partial derivative of f with respect to θ. The optimal coefficient should be found when the partial derivative is 0.

[0146]

[0147] Seeking

[0148]

[0149] Hessian matrix Φ T Calculating and inverting Φ is computationally time-consuming, so a recursive method was proposed to calculate the parameter matrix θ.

[0150] Design a recursive matrix in

[0151]

[0152] At time t, before the parameter matrix is ​​updated

[0153]

[0154]

[0155] so,

[0156] The following matrix is ​​used to update information from time t-1 to t.

[0157]

[0158] Design a recursive update method for these two matrices.

[0159] Autocorrelation matrix

[0160] Cross-correlation matrix

[0161] therefore

[0162]

[0163]

[0164] The inverse of the autocorrelation matrix P is solved using the Small Matrix Inversion Lemma. This lemma states that if the inverse of a high-dimensional matrix A is known, and the matrix A undergoes a very small change (its dimension is much lower than or equal to A), the inverse of the matrix after the small change can be found from the known inverse of A.

[0165] (A+BCD) -1 =A -1 -A -1 B(C -1 +DA -1 B) -1 DA -1 The condition is that the rank of BCD is very small.

[0166] According to this lemma, let P(t) = A + BCD, A = P -1 (t-1), C = 1,

[0167] but

[0168] In summary,

[0169] The recursive algorithm process is as follows:

[0170] A. Initialization

[0171] Initialization requires setting the initial value P0 of the autocorrelation matrix; θ can be estimated based on actual conditions.

[0172] B. Update Prediction Error

[0173]

[0174] C. Calculate the autocorrelation matrix

[0175]

[0176] Where λ is the forgetting factor, and its value ranges from 0.9 to 1.

[0177] D. Update parameters

[0178]

[0179] E. Entering the next cycle

[0180]

[0181] The following is a brief description of the specific applications of the recursive method.

[0182] When moving upwards under no load, record the average torque during the intermediate stage of uniform acceleration and the average torque during the intermediate stage of uniform speed.

[0183] When descending under no-load conditions, record the average torque during the intermediate stage of uniform acceleration (at the same position and with the same uniform acceleration as when ascending under no-load conditions), which satisfies...

[0184]

[0185]

[0186] Where T e_up_empty It is the average torque during the intermediate stage of uniform acceleration when the vehicle is moving upwards under no-load conditions, T e_dn_empty It is the average torque during the intermediate stage of uniform acceleration under no-load downward movement.

[0187] Subtracting the two equations, we get

[0188]

[0189] To capture the average torque during a constant upward motion under no-load conditions, we have...

[0190] f riction =f unbalanced -T e_up_const

[0191] Among them, T e_up_const The average torque when moving upwards at a constant speed under no-load conditions.

[0192] For elevator traction systems, during unloaded upward travel...

[0193]

[0194] Where T e It is electromagnetic torque, T L It is the unbalanced torque, f is the frictional resistance, and ω is the frictional resistance. k It is the current angular velocity, ω k-1 It is the angular velocity of the previous state, T s That is the sampling time.

[0195] Organized

[0196]

[0197] correspond

[0198] v = ay + bu

[0199] in

[0200] a = 1;

[0201] Based on the standard linear system equations

[0202]

[0203] correspond

[0204]

[0205] Then, calculations are performed according to the recursive algorithm described above, and after a certain number of iterations, the coefficient matrix Ξ is obtained. k The system's moment of inertia under no-load conditions is obtained.

[0206] When fully loaded and moving upwards, record the average torque during the intermediate stage of uniform acceleration and the average torque during the intermediate stage of uniform velocity.

[0207] When fully loaded and descending, record the average torque during the intermediate stage of uniform acceleration.

[0208] That is, satisfy

[0209]

[0210]

[0211] Where T e_up It is the average torque during the intermediate stage of uniform acceleration under full load, T e_dn It is the average torque during the intermediate stage of uniform acceleration under full load downward movement.

[0212] Subtracting the two equations, we get

[0213]

[0214] To capture the average torque during a fully loaded, uniformly upward-moving load, we have...

[0215] f riction =T e_up_const -f unbalanced

[0216] Among them, T e_up_const The average torque when fully loaded and moving at a constant speed upwards.

[0217] For elevator traction systems, when fully loaded and moving upwards...

[0218]

[0219] Where T e It is electromagnetic torque, T L It is the unbalanced torque, f is the frictional resistance, and ω is the frictional resistance. k It is the current angular velocity, ωk-1 It is the upper-state angular velocity, T s That is the sampling time.

[0220] Organized

[0221]

[0222] correspond

[0223] y = ay + bu

[0224] in

[0225] a = 1;

[0226] Based on the standard linear system equations

[0227]

[0228] correspond

[0229]

[0230] Then, calculations are performed according to the recursive algorithm described above, and after a certain number of iterations, the coefficient matrix Ξ is obtained. k The system's rotational inertia under full load is obtained.

[0231] The references for this invention are:

[0232] (1) Li Junyuan et al., Fundamentals of Electric Drive, Huazhong University of Science and Technology Press, 1999;

[0233] (2) Chen Min et al., Physics, Higher Education Press, 2012;

[0234] (3) A method and device for identifying the inertia of a permanent magnet synchronous motor, invented by Li Danyun et al. and applied for by China University of Geosciences (Wuhan) under Chinese invention patent authorization announcement number CNl09586645B. Summary of the Invention

[0235] The purpose of this invention is to address the problem that existing feedforward control cannot adaptively control the system inertia due to changes in load, i.e., feedforward control cannot achieve real-time tracking of the load, leading to a certain deviation between the feedforward torque signal and the actual required torque signal. This invention provides a torque feedforward adaptive method based on the rotational inertia of the elevator system. It performs self-learning based on the changes in system rotational inertia caused by different loads, obtaining dynamically changing system rotational inertia for compensation of acceleration dynamic torque current. The current obtained in this process is the torque current to be fed forward to the system.

[0236] To achieve the objective of this invention, the torque feedforward adaptive method based on the rotational inertia of an elevator system includes the following steps:

[0237] Step 1: Determine whether the elevator system's rotational inertia self-learning is complete. If the elevator system's inertia self-learning is complete, proceed to Step 9 below; if the elevator system's inertia self-learning is not complete, proceed to Step 2 below.

[0238] Step 2: With the car unloaded, the car moves upwards to capture the intermediate torque current during the unloaded upward movement.

[0239] Step 3: The car is unloaded and descends, capturing the intermediate torque current during the unloaded descent process;

[0240] Step 4: Based on the intermediate-order torque current captured in Step 2 during the car's unloaded upward movement and the intermediate-order torque current captured in Step 3 during the car's unloaded downward movement, calculate the moment of inertia of the unloaded system or obtain the moment of inertia of the unloaded system through experimental testing.

[0241] Step 5: When the car is fully loaded, it moves upwards, and the intermediate torque current during the upward movement of the fully loaded car is captured.

[0242] Step 6: With the car fully loaded, the car descends, and the intermediate torque current during the descent process is captured;

[0243] Step 7: Based on the intermediate-order torque current captured in Step 5 during the fully loaded upward process and the intermediate-order torque current captured in Step 6 during the fully loaded downward process, calculate the moment of inertia of the fully loaded system or obtain the moment of inertia of the fully loaded system through experimental testing.

[0244] Step 8: After the elevator system's rotational inertia self-learning is complete, return to Step 1;

[0245] Step 9: Read the load information;

[0246] Step 10: Based on the load information read in Step 9, calculate the system's moment of inertia using Newton's interpolation method;

[0247] Step 11: Based on the system moment of inertia calculated in Step 10, calculate the acceleration feedforward torque, which is used as the feedforward current controller.

[0248] In a preferred embodiment of the present invention, steps one to eight are repeated until the elevator system's rotational inertia self-learning is completed.

[0249] In a preferred embodiment of the present invention, the intermediate stage torque current is the torque current of the car's intermediate stage uniform speed running section.

[0250] By adopting the above technical solution, this invention enables the feedforward control to achieve adaptive control of the elevator system's inertia changes caused by load changes, based on the load information of the elevator system under actual working conditions. This reduces the burden on feedback control and achieves a low-bandwidth feedback control configuration. Attached Figure Description

[0251] Figure 1 This is a schematic diagram illustrating the operating principle of an elevator system.

[0252] Figure 2 This diagram illustrates the force distribution of the elevator system during the uniform speed travel phase when the elevator car is traveling upwards without a load.

[0253] Figure 3 This diagram illustrates the force distribution of the elevator system during the uniform speed operation phase when the elevator car is descending unloaded.

[0254] Figure 4 This diagram illustrates the force distribution of the elevator system during the uniform speed travel phase when the elevator car is fully loaded and moving upwards.

[0255] Figure 5 This diagram illustrates the force distribution of the elevator system during the uniform speed operation phase when the elevator car is fully loaded and descending.

[0256] Figure 6 This is a schematic diagram of an empty car traveling upwards, using a measured disc traction machine system as an example.

[0257] Figure 7 The diagram below shows the car descending unloaded, using a measured disc traction machine system as an example.

[0258] Figure 8 This is a schematic diagram of a fully loaded car ascending, using a measured disc traction machine system as an example.

[0259] Figure 9 The diagram below shows a fully loaded car descending, using a measured disc traction machine system as an example.

[0260] Figure 10 This is a schematic diagram of the four-quadrant control of an elevator system.

[0261] Figure 11 This is a flowchart illustrating the torque feedforward adaptive method based on the rotational inertia of an elevator system according to the present invention. Detailed Implementation

[0262] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0263] See Figure 11 The torque feedforward adaptive method based on the rotational inertia of an elevator system of the present invention includes the following steps:

[0264] Step 1: Determine whether the elevator system's rotational inertia self-learning is complete. If the elevator system's inertia self-learning is complete, proceed to Step 9 below; if the elevator system's inertia self-learning is not complete, proceed to Step 2 below.

[0265] Step 2: With the car unloaded, the car moves upwards to capture the intermediate torque current during the unloaded upward movement.

[0266] Step 3: The car is unloaded and descends, capturing the intermediate torque current during the unloaded descent process;

[0267] Step 4: Based on the intermediate-order torque current captured in Step 2 during the car's unloaded upward movement and the intermediate-order torque current captured in Step 3 during the car's unloaded downward movement, calculate the moment of inertia of the unloaded system or obtain the moment of inertia of the unloaded system through experimental testing.

[0268] Step 5: When the car is fully loaded, it moves upwards, and the intermediate torque current during the upward movement of the fully loaded car is captured.

[0269] Step 6: With the car fully loaded, the car descends, and the intermediate torque current during the descent process is captured;

[0270] Step 7: Based on the intermediate-order torque current captured in Step 5 during the fully loaded upward process and the intermediate-order torque current captured in Step 6 during the fully loaded downward process, calculate the moment of inertia of the fully loaded system or obtain the moment of inertia of the fully loaded system through experimental testing.

[0271] Step 8: After the elevator system's rotational inertia self-learning is complete, return to Step 1;

[0272] Step 9: Read the load information;

[0273] Step 10: Based on the load information read in Step 9, calculate the system's moment of inertia using Newton's interpolation method;

[0274] Step 11: Based on the system moment of inertia calculated in Step 10, calculate the acceleration feedforward torque, which is used as the feedforward current controller.

[0275] Steps one through eight above are repeated until the elevator system's rotational inertia self-learning is completed.

[0276] The aforementioned intermediate-stage torque current refers to the torque current during the uniform speed running section of the car's intermediate stage.

Claims

1. A torque feedforward adaptive method based on the rotational inertia of an elevator system, characterized in that, Includes the following steps: Step 1: Determine whether the elevator system's rotational inertia self-learning is complete. If the elevator system's inertia self-learning is complete, proceed to Step 9 below; if the elevator system's inertia self-learning is not complete, proceed to Step 2 below. Step 2: With the car unloaded, the car moves upwards to capture the intermediate torque current during the unloaded upward movement. Step 3: The car is unloaded and descends, capturing the intermediate torque current during the unloaded descent process; Step 4: Based on the intermediate-order torque current captured in Step 2 during the car's unloaded upward movement and the intermediate-order torque current captured in Step 3 during the car's unloaded downward movement, calculate the moment of inertia of the unloaded system or obtain the moment of inertia of the unloaded system through experimental testing. Step 5: When the car is fully loaded, it moves upwards, and the intermediate torque current during the upward movement of the fully loaded car is captured. Step 6: With the car fully loaded, the car descends, and the intermediate torque current during the descent process is captured; Step 7: Based on the intermediate-order torque current captured in Step 5 during the fully loaded upward process and the intermediate-order torque current captured in Step 6 during the fully loaded downward process, calculate the moment of inertia of the fully loaded system or obtain the moment of inertia of the fully loaded system through experimental testing. Step 8: After the elevator system's rotational inertia self-learning is complete, return to Step 1; Step 9: Read the load information; Step 10: Based on the load information read in Step 9, calculate the system's moment of inertia using Newton's interpolation method; Step 11: Based on the system moment of inertia calculated in Step 10, calculate the acceleration feedforward torque, which is used as the feedforward current controller.

2. The torque feedforward adaptive method based on the rotational inertia of an elevator system as described in claim 1, characterized in that, Steps one through eight are repeated until the elevator system's rotational inertia self-learning is completed.

3. A torque feedforward adaptive method based on the rotational inertia of an elevator system as described in claim 1 or 2, characterized in that, The intermediate-stage torque current is the torque current in the intermediate-stage uniform speed running section of the car.

Images