Electro-hydraulic servo control method for working device of electric forklift

Abstract

The invention provides an electric forklift working device electro-hydraulic servo control method which comprises the following steps: firstly, constructing an electro-hydraulic servo control system, then optimizing parameters of a linear active-disturbance-rejection controller by adopting a big cane-mouse algorithm, and searching optimal parameters of the linear active-disturbance-rejection controller; and then position closed-loop control of the hydraulic cylinder and pressure closed-loop control of the outlet of the metering pump can be realized by means of the optimized linear active disturbance rejection controller. Parameters of the linear active-disturbance-rejection controller are optimized by adopting a big cane-mouse algorithm, optimal parameters of the linear active-disturbance-rejection controller are searched, self-adaptive adjustment of system parameters is achieved, and timely output of pressure and flow of the servo system is effectively controlled.

Description

Technical Field

[0001] The invention relates to the field of electro-hydraulic servo control, in particular to an electro-hydraulic servo control method for an electric forklift working device. Background Art

[0002] With the rapid rise of my country's manufacturing and e-commerce industries, electric forklifts have been favored by the market for their excellent loading and unloading, stacking capabilities and the convenience of short-distance transportation. During the operation of electric forklifts, they need to be frequently started and stopped to lift or lower goods. The hydraulic pump of a traditional electric forklift runs at a constant speed, ignoring the actual load demand, resulting in energy waste; further, although the valve control system has precise control and quick response, the throttling and overflow losses also reduce energy efficiency. Therefore, optimizing the forklift hydraulic system and improving energy efficiency have become issues that need to be urgently addressed in the current industry.

[0003] In the field of modern industrial automation, electro-hydraulic servo systems are widely used due to their high precision, fast response and strong load driving capabilities. The electro-hydraulic servo control system is a typical nonlinear system, and its nonlinearity is mainly manifested in valve port flow and valve core displacement, oil pressure and flow, hydraulic cylinder and servo valve nonlinear friction, hydraulic system flow leakage, servo valve dynamic characteristics, etc. Improving the control accuracy of the electro-hydraulic servo control system and enhancing its anti-interference ability have always been important concerns in the field of engineering control.

[0004] Active Disturbance Rejection Controller (ADRC) has good anti-disturbance performance and simple controller structure. In recent years, it has been widely studied and applied in many fields. Its characteristics are that the total uncertainty (or total disturbance) is estimated and compensated in real time and actively through the extended state observer, and all uncertain disturbances are integrated and compensated in the system by the feedback controller. However, the traditional ADRC contains many nonlinear elements, and its parameter tuning process is very complicated, which has become the main obstacle to the widespread application of the ADRC algorithm. The linear active disturbance rejection controller (LADRC) linearizes the extended state observer and nonlinear combination in ADRC, and simplifies the parameter tuning process through the pole configuration method, so that the number of controller parameters is reduced. Each parameter has a clear physical meaning, which is more convenient for engineering application, but it still has the problem of manual parameter tuning. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the present invention provides an electro-hydraulic servo control method for an electric forklift working device, which optimizes the parameters of a linear active disturbance rejection controller by adopting a large cane rat algorithm, finds the optimal parameters of the linear active disturbance rejection controller, realizes adaptive adjustment of system parameters, and effectively controls the timely output of pressure and flow of the servo system.

[0006] In order to achieve the above object, the specific scheme adopted by the present invention is:

[0007] An electro-hydraulic servo control method for an electric forklift working device mainly comprises the following steps: S1. Construct an electro-hydraulic servo control system, which includes a servo motor, a metering pump, an oil tank, a relief valve, a control valve, a hydraulic cylinder, a displacement sensor, a pressure sensor, a hydraulically controlled one-way valve, and a pressure signal selector; the hydraulic cylinder includes a lifting hydraulic cylinder and a tilting hydraulic cylinder; the servo motor drives the metering pump to rotate through a coupling, the metering pump delivers the hydraulic oil in the oil tank to the control valve, the relief valve maintains the pressure stability of the system, the control valve guides the hydraulic oil into the hydraulic cylinder, drives the load to move by pushing the piston, the hydraulically controlled one-way valve maintains the system pressure, and the displacement sensor and the pressure sensor feedback the real-time status of the system to the controller to achieve closed-loop control of the system; S2, using the large cane rat algorithm to optimize the parameters of the linear active disturbance rejection controller and find the optimal parameters of the linear active disturbance rejection controller; S3, the position controller performs closed-loop judgment of the hydraulic cylinder position according to the command of the forklift controller. When the hydraulic cylinder performs closed-loop position control, the position controller converts the target position of the hydraulic cylinder movement according to the hydraulic cylinder position control command sent by the forklift controller. The position controller compares the actual position information of the hydraulic cylinder collected by the position sensor with the target position to obtain the hydraulic cylinder position adjustment information. The linear anti-disturbance control controller optimized in step S2 can adjust the valve core action of the control valve based on the hydraulic cylinder position adjustment information, thereby realizing the closed-loop position control of the hydraulic cylinder. S4. When the metering pump performs pressure closed-loop control, the pressure signal selector determines the pressure of the hydraulic cylinder oil inlet chamber at the start of the action according to the target position of the hydraulic cylinder, and adds it to the preset pressure margin to obtain the target pressure. The pressure controller compares the actual pressure at the metering pump outlet collected by the pressure sensor with the target pressure to obtain pressure regulation information. The linear anti-disturbance control controller optimized in step S2 can adjust the speed of the servo motor based on the pressure regulation information, thereby realizing pressure closed-loop control at the metering pump outlet.

[0008] Furthermore, in step S4, the preset pressure margin is 15 to 25 bar.

[0009] Furthermore, in step S2, the linear active disturbance rejection controller includes a tracking differentiator TD, a linear extended state observer LESO and a linear state error feedback control law LSEF.

[0010] Further, in step S2, the method for optimizing the parameters of the linear active disturbance rejection controller using the large cane rat algorithm is: The mathematical model of the cane rat algorithm was established, and the population was randomly initialized. Each cane rat represented a parameter to be optimized. Where X represents the entire GCR population, and the current GCR position x is randomly generated using the following formula: i,j , n and d represent the population size and problem dimension, respectively; x i,j =rand×(UB j -LB j )+LB j Where rand is a random number between 0 and 1, UB and LB are the upper and lower limits of the search space; The intelligent foraging path and behavior of cane rats during the mating season are mathematically modeled. The male with the highest fitness is considered to be the dominant male k, which guides the evolution of the population. The position of the dominant male is represented by X. k Indicates that the positions of other rats are adjusted according to the position of the dominant male rat; the position adjustment formula is as follows: In the formula, represents the new GCR position, x i,j Indicates the current GCR position, x k,j represents the dominant male rat in the jth dimension; The cane rat algorithm determines whether the population should enter the exploration phase or the exploitation phase based on the ρ value. ρ is a variable that determines whether it is the rainy season and takes a value of 0.5 to balance exploration and exploitation. If rand < ρ, the exploration phase begins. The new position of the remaining mice in the search space is determined based on the position of the dominant male mouse. The formula is as follows: Where C is a random number defined within the boundary of the problem space, simulating scattered food sources and shelters, and r 1 To simulate the effect of abundant food sources, the formula is as follows: In the formula, is the target value for dominant males, C iter is the current iteration number, Max iter is the maximum number of iterations; If rand ≥ ρ, the development phase begins. The simulation starts with a randomly selected female m, so that m ≠ k. The position update formula is as follows: In the formula, x m,j represents the position of a randomly selected female in the jth dimension, μ is a random value from 1 to 4, simulating the number of pups produced by each female GCR per year; In this phase of the GCR behavior simulation, if the objective function value of another rat is better, the dominant male rat is updated; and the positions of the other rats are adjusted according to the newly established dominant male rat; the position movement strategy formula is as follows: Where α is the coefficient for simulating the reduction of food sources, α = 2×r 1 ×rand-r 1 , β is the coefficient that encourages GCR to migrate to other abundant food sources available in the breeding area, β = 2×r 1 ×μ-r 1 ; The objective function of the large cane rat algorithm is: In the formula, J (t) is the comprehensive performance index; e(t) is the system feedback error; M p is the system overshoot; t r Adjust the time for the system; μ 1 , μ 2 , μ 3 is the weight coefficient; In order to avoid falling into the local optimum, an improved crossover operator is introduced to update the individual position, and the current individual is allowed to perform a crossover operation with the optimal individual in the population with a certain probability Pt. t When , the crossover operation formula is as follows: In the formula, X A1 , X A2 are the positions of the two offspring individuals produced by the crossover, X A is the current individual position, X best is the optimal individual position of the current population, L is a random number between 0 and 1. In order to enable the offspring to obtain more excellent genes of the parents and maintain population diversity, the Laplace coefficient is introduced to improve the parameter L. The formula is as follows: In the formula, μ 4 and λ are the Laplace coefficients, μ 4 Take a natural number to control the position, λ>0 to control the scale, r 2 is a random number in [0,1]; the Laplace coefficient improvement parameter L adjusts the distance between the parent and the offspring by introducing the coefficient λ. The smaller λ is, the closer the offspring is to the parent, and the higher the population diversity is. At the same time, the Laplace coefficient μ 4 Under the joint regulation of and λ, the offspring can selectively obtain more excellent parent position information.

[0011] Furthermore, in step S2, the parameters to be optimized in the linear ADRC are: system parameter b0 , controller parameter ω c , observer bandwidth ω 0 , speed factor r, filter factor h 0 ,The Great Cane Rat Algorithm is used to optimize the five parameters simultaneously to obtain the optimal value of each parameter.

[0012] Further, in step S2, j is 5.

[0013] Beneficial effects:

[0014] (1) The present invention applies electro-hydraulic servo technology to the working device of an electric forklift, and sets a position controller and a pressure controller through a control strategy of pump-valve joint control to achieve dual closed-loop control of the position and pressure of the hydraulic system of the working device of the electric forklift.

[0015] (2) The electro-hydraulic servo system of the working device of the electric forklift is controlled by a linear anti-disturbance control controller which is more convenient for practical engineering applications. The present invention combines the large cane rat algorithm to optimize the linear anti-disturbance control controller, realizes adaptive adjustment of system parameters, effectively controls the pressure and flow of the servo system to output in time, and achieves energy saving.

[0016] (3) In order to prevent the optimization algorithm from falling into a local optimum, the present invention introduces an improved crossover operator, which allows the current individual to perform a crossover operation with the optimal individual in the population with a certain probability. At the same time, in order to maintain population diversity, the Laplace coefficient improvement parameter is introduced so that the offspring can selectively obtain the location information of more excellent parents. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural diagram of the electro-hydraulic servo system of the electric forklift working device of the present invention.

[0018] Figure 2 It is a schematic diagram of the electro-hydraulic servo control method of the electric forklift working device of the present invention.

[0019] Figure 3 It is a working principle diagram of the linear active disturbance rejection controller described in the present invention.

[0020] Figure 4 It is a flow chart of optimizing parameters by using the great cane rat algorithm in the present invention.

[0021] Figure 5 It is a three-dimensional graph of each test function of the present invention and a comparison graph of parameter optimization effects of corresponding optimization algorithms.

[0022] Among them, 1. servo motor, 2. metering pump, 3.1. first oil tank, 3.2. second oil tank, 3.3. third oil tank, 3.4. fourth oil tank, 4.1. first overflow valve, 4.2. second overflow valve, 5.1. first control valve, 5.2. second control valve, 6.1. first lifting hydraulic cylinder, 6.2. second lifting hydraulic cylinder, 7.1. first tilting hydraulic cylinder, 7.2. second tilting hydraulic cylinder, 8.1. first displacement sensor, 8.2. second displacement sensor, 9.1. first pressure sensor, 9.2. second pressure sensor, 9.3. third pressure sensor, 9.4. fourth pressure sensor, 9.5. fifth pressure sensor, 10.1. first hydraulically controlled one-way valve, 10.2. second hydraulically controlled one-way valve. DETAILED DESCRIPTION

[0023] The technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] The invention discloses an electro-hydraulic servo control method for an electric forklift working device, and each step is described in detail below.

[0025] S1. Build an electro-hydraulic servo control system.

[0026] like Figure 1 As shown, the electro-hydraulic servo system of the working device of the electric forklift includes a linear anti-disturbance controller, a servo motor 1, a metering pump 2, a position controller, a pressure controller, four oil tanks (a first oil tank 3.1, a second oil tank 3.2, a third oil tank 3.3, and a fourth oil tank 3.4), two overflow valves (a first overflow valve 4.1, a second overflow valve 4.2), two control valves (a first control valve 5.1, a second control valve 5.2), two lifting hydraulic cylinders (a first lifting hydraulic cylinder 6.1, a second lifting hydraulic cylinder 6.2), two tilting hydraulic cylinders (a first tilting hydraulic cylinder 7.1, a second tilting hydraulic cylinder 7.2), two displacement sensors (a first displacement sensor 8.1, a second displacement sensor 8.2), five pressure sensors (a first pressure sensor 9.1, a second pressure sensor 9.2, a third pressure sensor 9.3, a fourth pressure sensor 9.4, and a fifth pressure sensor 9.5), two hydraulically controlled one-way valves (a first hydraulically controlled one-way valve 10.1, a second hydraulically controlled one-way valve 10.2), etc.

[0027] Two control valves (first control valve 5.1, second control valve 5.2), two displacement sensors (first displacement sensor 8.1, second displacement sensor 8.2), two hydraulically controlled one-way valves (first hydraulically controlled one-way valve 10.1, second hydraulically controlled one-way valve 10.2) and a position controller constitute a valve control subsystem, which is configured for position servo control of lifting and tilting of the working device. Servo motor 1, quantitative pump 2, five pressure sensors (first pressure sensor 9.1, second pressure sensor 9.2, third pressure sensor 9.3, fourth pressure sensor 9.4, fifth pressure sensor 9.5), pressure signal selector and pressure controller constitute a pump control subsystem, which is configured for pressure servo control of lifting and tilting of the working device.

[0028] The servo motor 1 drives the metering pump 2 to rotate through the coupling. The metering pump 2 delivers the hydraulic oil in the oil tank to the control valve. The overflow valve maintains the pressure stability of the system. The control valve guides the hydraulic oil into the hydraulic cylinder, drives the load to move by pushing the piston, and the hydraulic control check valve maintains the system pressure. The displacement sensor and pressure sensor feedback the real-time status of the system to the corresponding controller to achieve closed-loop control of the system. For details, please refer to Figure 1 The metering pump 2 is connected to the servo motor 1, the first oil tank 3.1, the first overflow valve 4.1, the second overflow valve 4.2, and the first control valve 5.1. The first overflow valve 4.1 is connected to the second oil tank 3.2. A fifth pressure sensor 9.5 is provided between the metering pump 2 and the first control valve 5.1. The first control valve 5.1 is connected to the third oil tank 3.3 and the first hydraulically controlled one-way valve 10.1. A first pressure sensor 9.1 and a second pressure sensor 9.2 are provided between the first hydraulically controlled one-way valve 10.1 and the first control valve 5.1. The first hydraulically controlled one-way valve 10.1 is connected to the first lifting hydraulic cylinder 6.1 and the second lifting hydraulic cylinder 6.2. A first displacement sensor 8.1 is provided on the piston rod of the first lifting hydraulic cylinder 6.1. The second overflow valve 4.2 is connected to the second control valve 5.2, the second control valve 5.2 is connected to the fourth oil tank 3.4 and the second hydraulically controlled one-way valve 10.2, a third pressure sensor 9.3 and a fourth pressure sensor 9.4 are arranged between the second hydraulically controlled one-way valve 10.2 and the second control valve 5.2, the second hydraulically controlled one-way valve 10.2 is connected to the first tilt hydraulic cylinder 7.1 and the second tilt hydraulic cylinder 7.2, and a second displacement sensor 8.2 is arranged on the piston rod of the second tilt hydraulic cylinder 7.2.

[0029] S2. Use the large cane rat algorithm to optimize the parameters of the linear active disturbance rejection controller and find the optimal parameters of the linear active disturbance rejection controller.

[0030] Please refer to Figure 3 ,The linear active disturbance rejection controller (LADRC) includes a tracking differentiator (TD), a linear extended state observer (LESO) and a linear state error feedback control law (LSEF).

[0031] TD aims to obtain a smooth input signal and its differential signal by arranging the transition process. Its expression is as follows: In the formula: fhan[v 1 (k)-v 0 ,v 2 (k),r,h 0 ] is the fastest comprehensive function; v 0 、v 1 (k), v 2 (k) are the signals v R 、v 1 、v 2 The discrete form of k is the order of the derivative; r is the speed factor; h 0 is the filtering factor; h is the integral step length during operation, v 1 (k+1) is the tracking signal output by TD, v 2 (k+1) is the differential of the tracking signal.

[0032] The specific expression of the fastest comprehensive function fh is as follows: Where: a, a 0 , d, d 0 is the intermediate variable, y is the expected input, x 1 is the estimate of y, x 2 is the differential with respect to y.

[0033] LESO can estimate the system state and the comprehensive disturbance caused by the internal uncertainty (internal disturbance) and external disturbance (external disturbance) of the system in real time. Its expression is as follows: Where: z 1 、z 2 For x 1 、x 2 The estimate of z 3 is the interference estimator; For x 1 、x 2 The estimated rate of change of is the change rate of the disturbance estimation; u is the total control quantity of the LADRC controller, b 0 is the system parameter; observer gain ω 0 is the observer bandwidth; b is the system gain.

[0034] LSEF uses the PD controller to linearly combine the signal generated by TD with the state estimated by ESO to obtain the virtual control quantity. Its expression is as follows: u 0 =k p (v 1 -z 1 )+k d (v 2 -z 2 ) (4) Where: u 0 Virtual control quantity calculated by LSEF; controller gain k d =2ω c ,ω c is the controller coefficient. It can be seen that the parameters that need to be adjusted in LADRC are r, h 0 ,ω 0 ,ω c , b 0 .

[0035] Specifically, Figure 4 As shown in FIG. 1 , the method for optimizing the parameters of the linear active disturbance rejection controller using the cane rat algorithm is as follows: a mathematical model of the cane rat algorithm is established, and the population is randomly initialized, and each cane rat represents a parameter to be optimized. In the formula, X represents the entire GCR population, and the current GCR position x is randomly generated using formula (6): i,j , n and d represent the population size and problem dimension, respectively. It should be noted that GCR stands for Greater Cane Rat. x i,j =rand×(UB j -LB j )+LB j (6) Where rand is a random number between 0 and 1, and UB and LB are the upper and lower limits of the search space.

[0036] The intelligent foraging path and behavior of cane rats during the mating season are mathematically modeled. The male with the highest fitness is considered to be the dominant male k, which guides the evolution of the population. The position of the dominant male is represented by X. k Indicates that the positions of other rats are adjusted according to the position of the dominant male rat. The position adjustment formula is as follows: In the formula, represents the new GCR position, x k,j represents the dominant male rat in the jth dimension.

[0037] The Greater Cane Rat Algorithm (GCRA) determines whether the population should enter the exploration phase or the exploitation phase according to the ρ value. ρ is a variable that determines whether it is the rainy season and takes a value of 0.5 to balance exploration and exploitation.

[0038] If rand < ρ, the exploration phase begins. The new position of the remaining mice in the search space is determined based on the position of the dominant male mouse. The formula is as follows: Where C is a random number defined within the boundary of the problem space, simulating scattered food sources and shelters, and r 1 To simulate the effect of abundant food sources, the formula is as follows: In the formula, is the target value for dominant males, C iter is the current iteration number, Max iter is the maximum number of iterations.

[0039] If rand ≥ ρ, the development phase begins, and the simulation starts with a randomly selected female m, so that m ≠ k (dominant male). The position update formula is as follows: In the formula, x m,j represents the position of a randomly selected female in the jth dimension, and μ is a random value from 1 to 4, simulating the number of pups produced by each female GCR each year.

[0040] In this phase of the GCR behavior simulation, if the objective function value of another rat is better, the dominant male rat is updated, and the positions of the other rats are adjusted according to the newly established dominant male rat. The position movement strategy formula is as follows: Where α is the coefficient for simulating the reduction of food sources, α = 2×r 1 ×rand-r 1 , β is the coefficient that encourages GCR to migrate to other abundant food sources available in the breeding area, β = 2×r 1 ×μ-r 1 .

[0041] The objective function of the large cane rat algorithm is: In the formula, J (t) is the comprehensive performance index; e(t) is the system feedback error; M p is the system overshoot; t r Adjust the time for the system; μ 1 , μ 2 , μ 3is the weight coefficient; t represents the system running time.

[0042] In order to avoid falling into the local optimum, an improved crossover operator is introduced to update the individual position with a certain probability P t Let the current individual perform a crossover operation with the optimal individual in the population. When P < P t When , the crossover operation formula is as follows: Where, X A1 , X A2 are the positions of the two offspring individuals produced by the crossover, X A is the current individual position, X best is the optimal individual position of the current population, L is a random number between 0 and 1. In order to enable the offspring to obtain more excellent genes of the parents and maintain population diversity, the Laplace coefficient is introduced to improve the parameter L. The formula is as follows: In the formula, μ 4 and λ are the Laplace coefficients, μ 4 Take a natural number to control the position, λ>0 to control the scale, r 2 is a random number between [0,1]. The improved parameter L adjusts the distance between the parent and the offspring by introducing the coefficient λ. The smaller λ is, the closer the offspring is to the parent, and the higher the population diversity is. 4 Under the joint regulation of and λ, the offspring can selectively obtain more excellent parent position information.

[0043] The present invention controls the electro-hydraulic servo system of the working device of the electric forklift through a linear active disturbance rejection controller which is more convenient for practical engineering applications, and optimizes the linear active disturbance rejection controller in combination with a large cane rat algorithm, thereby realizing adaptive adjustment of system parameters, effectively controlling the pressure and flow of the servo system to output in time, and achieving energy saving.

[0044] In order to prevent the optimization algorithm from falling into the local optimum, an improved crossover operator is introduced to allow the current individual to cross with the optimal individual in the population with a certain probability. At the same time, in order to maintain the diversity of the population, the Laplace coefficient is introduced to improve the parameters so that the offspring can selectively obtain more excellent parent position information.

[0045] S3. The position controller performs closed-loop judgment on the hydraulic cylinder position according to the command of the forklift controller. When the hydraulic cylinder performs closed-loop position control, the position controller converts the target position of the hydraulic cylinder movement according to the hydraulic cylinder position control command sent by the forklift controller. The position controller compares the actual position information of the hydraulic cylinder collected by the position sensor with the target position to obtain the hydraulic cylinder position adjustment information. The linear anti-disturbance control controller optimized in step S2 can adjust the valve core action of the control valve based on the hydraulic cylinder position adjustment information, thereby realizing closed-loop position control of the hydraulic cylinder.

[0046] S4. When the metering pump performs pressure closed-loop control, the pressure signal sensor determines the pressure of the hydraulic cylinder oil inlet chamber at the start of the action according to the target position of the hydraulic cylinder, and adds it to the preset pressure margin to obtain the target pressure. The pressure controller compares the actual pressure at the metering pump outlet collected by the pressure sensor with the target pressure to obtain pressure regulation information. The linear anti-disturbance control controller optimized in step S2 can adjust the speed of the servo motor based on the pressure regulation information, thereby realizing pressure closed-loop control at the metering pump outlet.

[0047] like Figure 5 (A-1), (B-1), (C-1), and (D-1) respectively represent the three-dimensional graphs of the unimodal test function Sphere, the unimodal test function Schwefel's problem 2.22, the multimodal test function Ackley, and the multimodal test function GeneralizedPenalized. Figure 5 (A-2), (B-2), (C-2), and (D-2) are the convergence curve comparison diagrams of the parameter optimization method (GCRA) used in the present invention, the Grey Wolf Algorithm (GWO), and the Particle Swarm Optimization (PSO) under four test functions. In this experiment, the maximum number of iterations is set to 500. Figure 5 It can be seen that under the tests of four standard test functions, GCRA shows faster convergence speed and higher solution accuracy than GWO and PSO. GCRA has better solution ability than GWO and PSO.

[0048] The present invention applies electro-hydraulic servo technology to the working device of an electric forklift, and sets a position servo controller and a pressure servo controller through a pump-valve joint control strategy to achieve double closed-loop control of the position and pressure of the hydraulic system of the working device of the electric forklift.

[0049] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any equivalent changes or modifications made according to the essence of the present invention should be included in the protection scope of the present invention.

Claims

1. An electro-hydraulic servo control method for an electric forklift working device, characterized in that: The main steps are as follows: S1. Construct an electro-hydraulic servo control system, which includes a linear anti-disturbance controller, a servo motor, a metering pump, an oil tank, a relief valve, a control valve, a hydraulic cylinder, a displacement sensor, a pressure sensor, a hydraulically controlled one-way valve, and a pressure signal selector; the hydraulic cylinder includes a lifting hydraulic cylinder and a tilting hydraulic cylinder; the servo motor drives the metering pump to rotate through a coupling, the metering pump delivers the hydraulic oil in the oil tank to the control valve, the relief valve maintains the pressure stability of the system, the control valve guides the hydraulic oil into the hydraulic cylinder, drives the load to move by pushing the piston, the hydraulically controlled one-way valve maintains the system pressure, and the displacement sensor and the pressure sensor feedback the real-time status of the system to the corresponding controller to achieve closed-loop control of the system; S2, using the large cane rat algorithm to optimize the parameters of the linear active disturbance rejection controller and find the optimal parameters of the linear active disturbance rejection controller; S3, the position controller performs closed-loop judgment of the hydraulic cylinder position according to the command of the forklift controller on the electric forklift. When the hydraulic cylinder performs closed-loop position control, the position controller converts the target position of the hydraulic cylinder movement according to the hydraulic cylinder position control command sent by the forklift controller. The position controller compares the actual position information of the hydraulic cylinder collected by the position sensor with the target position to obtain the hydraulic cylinder position adjustment information. The linear anti-disturbance control controller optimized in step S2 can adjust the valve core action of the control valve based on the hydraulic cylinder position adjustment information, thereby realizing the closed-loop position control of the hydraulic cylinder. S4. When the metering pump performs pressure closed-loop control, the pressure signal selector determines the pressure of the hydraulic cylinder oil inlet chamber at the start of the action according to the target position of the hydraulic cylinder, and adds it to the preset pressure margin to obtain the target pressure. The pressure controller compares the actual pressure at the metering pump outlet collected by the pressure sensor with the target pressure to obtain pressure regulation information. The linear anti-disturbance control controller optimized in step S2 can adjust the speed of the servo motor based on the pressure regulation information, thereby realizing pressure closed-loop control at the metering pump outlet.

2. The electro-hydraulic servo control method of an electric forklift working device according to claim 1, characterized in that: In step S4, the preset pressure margin is 15-25 bar.

3. The electro-hydraulic servo control method of an electric forklift working device according to claim 1, characterized in that: In step S2, the linear active disturbance rejection controller includes a tracking differentiator TD, a linear extended state observer LESO and a linear state error feedback control law LSEF.

4. The electro-hydraulic servo control method of an electric forklift working device according to claim 1, characterized in that: In step S2, the method for optimizing the parameters of the linear active disturbance rejection controller using the large cane rat algorithm is: The mathematical model of the cane rat algorithm was established, and the population was randomly initialized. Each cane rat represented a parameter to be optimized. Where X represents the entire GCR population, and the current GCR position x is randomly generated using the following formula: i,j , n and d represent the population size and problem dimension, respectively; x i,j =rand×(UB j -LB j )+LB j Where rand is a random number between 0 and 1, UB and LB are the upper and lower limits of the search space; The intelligent foraging path and behavior of cane rats during the mating season are mathematically modeled. The dominant male rat k is considered to have the highest fitness and guides the evolution of the population. The position of the dominant male rat k is represented by X. k Indicates that the positions of other rats are adjusted according to the position of the dominant male rat; the position adjustment formula is as follows: In the formula, represents the new GCR position, x i,j Indicates the current GCR position, x k,j represents the dominant male rat in the jth dimension; The cane rat algorithm determines whether the population should enter the exploration phase or the exploitation phase based on the ρ value. ρ is a variable that determines whether it is the rainy season and takes a value of 0.5 to balance exploration and exploitation. If rand < ρ, the exploration phase begins. The new position of the remaining mice in the search space is determined based on the position of the dominant male mouse. The formula is as follows: Where C is a random number defined within the boundary of the problem space, simulating scattered food sources and shelters, and r1 simulates the impact of abundant food sources. The formula is as follows: In the formula, is the target value for dominant males, C iter is the current iteration number, Max iter is the maximum number of iterations; If rand ≥ ρ, the development phase begins. The simulation starts with a randomly selected female m, so that m ≠ k. The position update formula is as follows: In the formula, x m,j represents the position of a randomly selected female in the jth dimension, μ is a random value from 1 to 4, simulating the number of pups produced by each female GCR per year; In this phase of the GCR behavior simulation, if the objective function value of another rat is better, the dominant male rat is updated; and the positions of the other rats are adjusted according to the newly established dominant male rat; the position movement strategy formula is as follows: Where α is the coefficient simulating the reduction of food sources, α = 2 × r1 × rand-r1, β is the coefficient that prompts GCR to migrate to other abundant food sources available in the breeding area, β = 2 × r1 × μ-r1; The objective function of the large cane rat algorithm is: J (t) =μ1∫0t|e(t)|dt+μ2M p +μ3t r In the formula, J (t) is the comprehensive performance index; e(t) is the system feedback error; M p is the system overshoot; t r is the system adjustment time; μ1, μ2, μ3 are weight coefficients; In order to avoid falling into the local optimum, an improved crossover operator is introduced to update the individual position, and the current individual is allowed to perform a crossover operation with the optimal individual in the population with a certain probability Pt. t When , the crossover operation formula is as follows: Where, X A1 , X A2 are the positions of the two offspring individuals produced by the crossover, X A is the current individual position, X best is the optimal individual position of the current population, L is a random number between 0 and 1. In order to enable the offspring to obtain more excellent genes of the parents and maintain population diversity, the Laplace coefficient is introduced to improve the parameter L. The formula is as follows: In the formula, μ4 and λ are Laplace coefficients, μ4 is a natural number that controls the position, λ>0 controls the scale, and r2 is a random number in [0,1]. The improved parameter L adjusts the distance between the parent and the offspring by introducing the coefficient λ. The smaller λ is, the closer the offspring is to the parent, and the higher the population diversity. At the same time, under the joint regulation of the Laplace coefficients μ4 and λ, the offspring can selectively obtain more excellent parent location information.

5. The electro-hydraulic servo control method of an electric forklift working device according to claim 4, characterized in that: In step S2, the parameters that need to be optimized in the linear ADRC are: system parameter b0, controller parameter ω c , observer bandwidth ω0, speed factor r, filter factor h0, and the great cane rat algorithm are used to optimize the five parameters simultaneously to obtain the optimal value of each parameter.

6. The electro-hydraulic servo control method of an electric forklift working device according to claim 5, characterized in that: In step S2, j is 5.

Images