Gradient adjustment synchronous control method for cylindrical valve servomotor

Abstract

The invention relates to the technical field of cylindrical valve control, in particular to a gradient adjustment synchronous control method of a cylindrical valve servomotor, according to the maximum value of the position difference between slave shafts fed back in each period, three gradients are divided, different motion adjustment methods are adopted respectively, in the first gradient, each period is controlled through closed-loop proportional feedback, and in the second gradient, each period is controlled through closed-loop proportional feedback. And in combination with a feedforward compensation value calculation model, adjusting the motion increment of the slave shaft in the next period. Through the method, the current movement state of the cylindrical valve can be accurately recognized, a proper adjusting strategy is selected, and the control precision and flexibility are improved.

Description

Technical Field

[0001] The invention relates to the technical field of cylindrical valve control, and in particular to a gradient regulating synchronous control method of a cylindrical valve servomotor. Background Art

[0002] The cylinder valve is a new type of water inlet valve for turbines, usually arranged between the fixed guide vanes and movable guide vanes of the turbine, and the opening and closing of the cylinder valve are controlled by 6 servomotors. The synchronous movement of the servomotors determines the quality of the opening and closing of the cylinder valves, and is the key core technology of the synchronous control of cylinder valves. When the synchronization error of the servomotors increases, it is easy to cause fault alarms such as jamming and loss of step, which may cause irreversible damage to the mechanical structure in severe cases, causing serious safety hazards.

[0003] In the prior art, a Chinese invention patent document with publication number CN115750191A and publication date March 7, 2023 is proposed. In the technical solution disclosed in the patent document, the synchronization control method is roughly divided into two types: speed regulation control and position correction control.

[0004] Speed ​​regulation control is a method of dynamically clearing the synchronization error by timely adjusting the movement speed of the servo. This control method requires frequent triggering of the motion control function block, which causes the floating point calculation error of trajectory planning to accumulate, which may affect the motion accuracy. In addition, the speed stability is poor during the speed regulation process, and the acceleration will frequently change suddenly, which is easy to cause impact.

[0005] Position correction control is a control method that inserts a correction process into the original motion process. This method will terminate the current motion and cause the motion process to be discontinuous.

[0006] In order to achieve synchronous control, in other technical fields, a method for synchronous control of master and slave axes using virtual master axis and PI control has been proposed. For example, the Chinese invention patent document with publication number CN115750191A and publication date April 14, 2020; the Chinese invention patent document with publication number CN118034190A and publication date May 14, 2024; and the Chinese invention patent document with publication number CN110737193A and publication date September 19, 2019.

[0007] However, when the above technical solution is applied to the synchronous control of the cylindrical valve, due to the friction between the cylindrical valve relay and the cylinder wall, simple traditional feedback control methods such as P, PI or PID cannot completely eliminate the master-slave axis following error. Steady-state errors will be formed over time, and the control accuracy is difficult to meet the control accuracy requirements of the cylindrical valve relay.

[0008] In addition, the overall mechanical structure mass and volume of the cylindrical valve relay control are relatively large, and the adjustment force needs to be properly controlled during synchronization control. Too aggressive or too gentle adjustment methods will cause excessive synchronization errors and lead to serious consequences. Summary of the invention

[0009] In order to solve the above technical problems, the present invention proposes a gradient adjustment synchronous control method for a cylindrical valve relay. According to the maximum value of the position difference between the slave axes fed back in each cycle, it is divided into three gradients and different motion adjustment methods are adopted respectively. This method can accurately identify the current motion state of the cylindrical valve, select the appropriate adjustment strategy, and improve the control accuracy and flexibility.

[0010] The present invention is achieved by adopting the following technical solutions:

[0011] A gradient regulation synchronous control method for a cylindrical valve servomotor comprises the following steps:

[0012] Step S1. Establish a virtual main axis, use multiple relays as slave axes that follow the main axis, bind the master-slave axis relationship and calculate the initial slave axis feedback mean;

[0013] Step S2. Generate a feedforward compensation value calculation model;

[0014] Step S3. The spindle starts to move and records the initial position;

[0015] Step S4. Calculate the feedforward compensation value of the next cycle according to the feedforward compensation value calculation model and the current spindle position;

[0016] Step S5. Calculate the deviation value between each slave axis;

[0017] Step S6. Determine whether the deviation value between each slave axis exceeds SYNCMAX, where SYNCMAX represents the maximum value of the deviation between each slave axis allowed; if the deviation values ​​between all slave axes are less than SYNCMAX, proceed to step S7; if the deviation between any two slave axes is greater than SYNCMAX, then the process ends and all axes stop;

[0018] Step S7. Determine whether the deviation value between each slave axis exceeds SYNCMID, where SYNCMID represents the normal value of the acceptable deviation between each slave axis; if the deviation value between all slave axes is less than SYNCMID, the first gradient adjustment method is adopted; if the deviation value between any two slave axes is greater than SYNCMID, the second gradient adjustment method is adopted;

[0019] The first gradient adjustment method refers to: based on the slave axis feedback mean increment and the main axis motion increment, a proportional feedback adjustment method is used in combination with a feedforward compensation value to calculate the slave axis motion increment, and the slave axis is controlled to move according to the calculated motion increment; the second gradient adjustment method refers to an autonomous deviation correction adjustment method;

[0020] Step S8. Update the initial slave axis feedback mean;

[0021] Step S9: Determine whether the spindle is in place. If not, proceed to step S3. If yes, end and all axes stop.

[0022] The feedforward compensation value calculation model is a piecewise cubic Hermite interpolator, and the generation method of the piecewise cubic Hermite interpolator is:

[0023] Step S 21. Record the current spindle position MasterPos;

[0024] Step S 22. The main axis starts to move, driving the slave axis to move according to the process without feedforward adjustment;

[0025] Step S 23. After this cycle is over, the position difference between the master and slave axes is calculated and sent to the piecewise cubic Hermite interpolator in one-to-one correspondence with MasterPos;

[0026] Step S 24. Determine whether the spindle stroke is completed. If the spindle stroke is completed, exit or repeat the process again to take the average. If the stroke is not completed, enter the next cycle.

[0027] In step S6, if the deviation between any two slave axes is greater than SYNCMAX, a third gradient adjustment method is used for adjustment. The third gradient adjustment method specifically includes: reporting a card group failure and manual adjustment.

[0028] The first gradient adjustment method specifically comprises the following steps:

[0029] Step S 711. Save the current slave axis feedback average feedAverage1 and master axis position MasterPos1;

[0030] Step S 712. Calculate the slave axis feedback mean increment feedAverageDelta and the master axis motion increment MasterPosDelta;

[0031] Step S 713. Calculate the slave axis motion increment SlavePosDelta:

[0032] SlavePosDelta=SYNCAVR+feedAverageDelta*Kp+MasterPosDelta

[0033] In the formula, SYNCAVR is the feedforward compensation value, and Kp is the proportional gain coefficient;

[0034] Step S 714. Control the slave axis to move according to the calculated motion increment.

[0035] The second gradient adjustment method specifically comprises the following steps:

[0036] Step S 721. The spindle stops moving and the binding state between the spindle and the slave axis is released;

[0037] Step S 722. Each slave axis performs alignment movement based on the current position of the master axis;

[0038] Step S 723. After each slave axis is in place, the master axis and slave axis states are re-bound, and the master axis is triggered again at the original target position. 722 Specifically include:

[0039] Based on the current spindle position, calculate the slave axis motion increment:

[0040] SlavePosDelta[i]=MasterPos-SlavePos[i]

[0041] Where, SlavePosDelta[i] represents the motion increment of slave axis i, MasterPos represents the current master axis position, and SlavePos[i] represents the current position of slave axis i.

[0042] Each slave axis moves separately, and after the movement is completed, the position of each slave axis basically coincides with that of the master axis.

[0043] Step S8 specifically refers to: taking the current initial slave axis feedback average value feedAverage0 and the current master axis motion increment MasterPosDelta as the initial slave axis feedback average value feedAverage0′ of the next cycle:

[0044] feedAverage0′=feedAverage0+MasterPosDelta.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] 1. Aiming at the difficulty of synchronous control of the above-mentioned cylindrical valve servomotor, the present invention proposes a gradient adjustment synchronous control method for cylindrical valve servomotor. According to the maximum value of the position difference between the slave axes fed back in each cycle, different gradients are divided into different motion adjustment methods, which can accurately identify the current cylindrical valve motion state, select the appropriate adjustment strategy, and improve the control accuracy and flexibility of the system.

[0047] By using multiple cylindrical valve servomotors as slave axes and binding them to the virtual master axis, and following the movement of the master axis, multiple slave axes have uniform motion, which can fundamentally reduce synchronization errors. When the virtual axis is used as the master axis, it has no actual physical structure and is not affected by friction and mechanical conditions. The motion increment it gives is highly credible and has small deviation.

[0048] Among them, during the first gradient, the adjustment method is mainly aimed at adjusting the following accuracy of the slave axis to the main axis. The improvement of the following accuracy can ensure the repeatable positioning accuracy of the entire system, and to a certain extent, it can also improve the synchronization accuracy of the cylindrical valve servomotor. Specifically, through closed-loop proportional feedback control in each cycle, the movement increment of the slave axis in the next cycle is adjusted, which reduces the impact caused by the sudden change in acceleration, reduces the synchronization error between the cylindrical valve servomotors, and improves the smoothness of the overall operation of the equipment. Combined with the feedforward compensation value calculation model, the steady-state error formed in the process of the slave axis chasing the main axis can be effectively reduced, the following accuracy of the slave axis can be improved, and the robustness of the system can be enhanced.

[0049] During the second gradient, the adjustment method mainly improves the synchronization accuracy between the cylindrical valve relays, enables the slave axis to quickly catch up with the main axis, realizes online deviation correction, and enhances the flexibility of control.

[0050] The third gradient is an abnormal situation and requires manual adjustment.

[0051] The gradient adjustment method can reflect the synchronization accuracy between the slave axes, and the feedforward compensation value can reflect the following accuracy of the slave axes to the master axis. Through the combination of the above schemes, the present invention can effectively improve the position control accuracy and synchronization accuracy of the cylindrical valve servomotor, thereby ensuring the precise operation of the cylindrical valve servomotor.

[0052] 2. In the present invention, the feedforward compensation value calculation model is a piecewise cubic Hermite interpolator, which can adjust the feedforward compensation value in real time as the main axis moves, has stronger adaptability, can improve the performance of feedforward regulation to a greater extent, and improve the following accuracy of the slave axis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, wherein:

[0054] Figure 1 It is a schematic diagram of the process of the present invention;

[0055] Figure 2 Schematic diagram of the generation process of the piecewise cubic Hermite interpolator in the present invention;

[0056] Figure 3 is a control block diagram of the first gradient adjustment method in the present invention;

[0057] Figure 4 It is a system structure diagram suitable for the control method in the present invention. DETAILED DESCRIPTION

[0058] Example 1

[0059] As a basic embodiment of the present invention, the present invention includes a gradient regulation synchronous control method of a cylindrical valve servomotor, comprising:

[0060] Step S1. Establish a virtual master axis, use multiple relays as slave axes that follow the movement of the master axis, bind the master-slave axis relationship and calculate the initial slave axis feedback mean.

[0061] Step S2: Generate a feedforward compensation value calculation model, wherein the feedforward compensation value calculation model may adopt a conventional feedforward control system in the art.

[0062] Step S3: The spindle starts to move and records the initial position.

[0063] Step S4: Calculate the feedforward compensation value of the next cycle according to the feedforward compensation value calculation model and the current spindle position.

[0064] Step S5: Calculate the deviation value between each slave axis.

[0065] Step S6. Determine whether the deviation value between each slave axis exceeds SYNCMAX, where SYNCMAX represents the maximum allowed deviation value between each slave axis; if the deviation values ​​between all slave axes are less than SYNCMAX, proceed to step S7; if the deviation between any two slave axes is greater than SYNCMAX, the process ends and all axes stop.

[0066] Step S7. Determine whether the deviation value between each slave axis exceeds SYNCMID, where SYNCMID represents the normal value of the acceptable deviation between each slave axis; if the deviation value between all slave axes is less than SYNCMID, the first gradient adjustment method is adopted; if the deviation value between any two slave axes is greater than SYNCMID, the second gradient adjustment method is adopted.

[0067] The first gradient adjustment method refers to: based on the slave axis feedback mean increment and the main axis motion increment, a proportional feedback adjustment method is used in combination with a feedforward compensation value to calculate the slave axis motion increment, and the slave axis is controlled to move according to the calculated motion increment. The second gradient adjustment method refers to an autonomous deviation correction adjustment method.

[0068] Step S8: Update the initial slave axis feedback mean value.

[0069] Step S9: Determine whether the spindle is in place. If not, proceed to step S3. If yes, end and all axes stop.

[0070] Example 2

[0071] As a preferred embodiment of the present invention, the present invention includes a gradient adjustment synchronous control method of a cylindrical valve servomotor, comprising:

[0072] Step S1. Establish a virtual master axis, use multiple relays as slave axes that follow the movement of the master axis, bind the master-slave axis relationship and calculate the initial slave axis feedback mean.

[0073] Step S2: Generate a feedforward compensation value calculation model, wherein the feedforward compensation value calculation model may be a piecewise cubic Hermite interpolator.

[0074] Step S3: The spindle starts to move and records the initial position.

[0075] Step S4: Calculate the feedforward compensation value of the next cycle according to the feedforward compensation value calculation model and the current spindle position.

[0076] Step S5: Calculate the deviation value between each slave axis.

[0077] Step S6. Determine whether the deviation value between each slave axis exceeds SYNCMAX, where SYNCMAX represents the maximum deviation value allowed between each slave axis; if the deviation values ​​between all slave axes are less than SYNCMAX, proceed to step S7. If the deviation between any two slave axes is greater than SYNCMAX, the third gradient adjustment method is used to adjust and then terminate, and all axes stop. The third gradient adjustment method specifically includes: reporting a card group failure and manual adjustment.

[0078] Step S7. Determine whether the deviation value between each slave axis exceeds SYNCMID, where SYNCMID represents the normal value of the acceptable deviation between each slave axis; if the deviation value between all slave axes is less than SYNCMID, the first gradient adjustment method is adopted; if the deviation value between any two slave axes is greater than SYNCMID, the second gradient adjustment method is adopted.

[0079] Among them, the first gradient adjustment method refers to: based on the slave axis feedback mean increment and the main axis motion increment, a proportional feedback adjustment method is used in combination with a feedforward compensation value to calculate the slave axis motion increment, and the slave axis is controlled to move according to the calculated motion increment.

[0080] The second gradient adjustment method refers to an autonomous deviation correction adjustment method. The autonomous deviation correction adjustment method can adopt a conventional deviation correction method in the art. The main axis pauses and waits for each slave axis to perform an alignment movement based on the main axis position. After alignment, the main axis is triggered again at the original target position.

[0081] Step S8: Update the initial slave axis feedback mean value.

[0082] Step S9: Determine whether the spindle is in place. If not, proceed to step S3. If yes, end and all axes stop.

[0083] Example 3

[0084] As the best embodiment of the present invention, the present invention includes a gradient adjustment synchronous control method of a cylindrical valve relay. The method sets two variables: SYNCMAX and SYNCMID, which are used to divide three gradients. SYNCMAX represents the maximum value of the deviation between each slave axis allowed; SYNCMID represents the normal value of the deviation between each slave axis that is acceptable. The three divided gradients can be:

[0085] First gradient: 0<|position deviation value between slave axes| <SYNCMID。

[0086] Second gradient: SYNCMID<|position deviation value between slave axes| <SYNCMAX。

[0087] First gradient: |Position deviation value between slave axes|>SYNCMAX.

[0088] Refer to the instruction manual Figure 1 , the method specifically comprises the following steps:

[0089] Step S1. Establish a virtual master axis, use multiple relays as slave axes that follow the movement of the master axis, bind the master-slave axis relationship and calculate the initial slave axis feedback mean.

[0090] In this embodiment, the number of the relays can be 6. The 6 relays are used as slave axes, bound to the virtual main axis through electronic gears, and follow the movement of the main axis. At this time, the 6 slave axes have motion unity, which can fundamentally reduce the synchronization error. When the virtual axis is used as the main axis, there is no actual physical structure, and it is not affected by friction and mechanical conditions. The motion increment it gives is highly credible and has a small deviation.

[0091] The average value of the encoder feedback values ​​of the six slave axes in the initial state is taken and recorded as feedAverage0.

[0092] Step S2. Generate a feedforward compensation value calculation model. The feedforward compensation value calculation model is a piecewise cubic Hermite interpolator, which is used to obtain the feedforward compensation value SYNCAVR in each subsequent cycle. The implementation principle is:

[0093] The main axis position and the master and slave axis position difference generated in the current cycle are abstracted into a piecewise cubic Hermite function relationship. Any position in the main axis motion process corresponds to a position difference result, which is used as the feedforward compensation value for the next cycle.

[0094] This method can adjust the feedforward compensation value in real time with the movement of the main axis, has stronger adaptive ability, can greatly improve the performance of feedforward adjustment, and improve the following accuracy of the slave axis. The generation process of the segmented cubic Hermite interpolator needs to eliminate the feedforward link and perform the remaining motion process once or multiple times. For details, refer to the attached manual. Figure 2 , the generation method of piecewise cubic Hermite interpolator is:

[0095] Step S 21. Record the current spindle position MasterPos.

[0096] Step S 22. The main axis starts to move, driving the slave axis to move according to the process without feedforward adjustment.

[0097] Step S 23. After this cycle is over, the position difference between the master and slave axes is calculated and sent to the piecewise cubic Hermite interpolator in one-to-one correspondence with MasterPos.

[0098] Step S 24. Determine whether the spindle stroke is completed. If the spindle stroke is completed, exit or repeat the process again to take the average. If the stroke is not completed, enter the next cycle.

[0099] Y represents the output position difference result, x represents the current spindle position, and the standard formula of the piecewise cubic Hermite function is rewritten as:

[0100] Y3(x)=γ0α0(x)+γ1α1(x)+γ0′β0(x)+γ1′β1(x)

[0101] In the formula, γ0 and γ1 are the function values ​​corresponding to x=x0 and x=x1, and γ0′ and γ1′ are the first-order derivative values ​​corresponding to x=x0 and x=x1. α0(x), α1(x), β0(x), and β1(x) are the expressions of the multiple power terms in the function, and the calculation method is as follows:

[0102]

[0103] Substituting the one-to-one position difference and spindle position into the formula, a piecewise cubic Hermite interpolator suitable for the present invention can be obtained. In the subsequent movement process, the spindle position of the current cycle is sent to the piecewise cubic Hermite interpolator to obtain the feedforward compensation value of the next cycle.

[0104] Step S3: The spindle starts to move and records the initial position.

[0105] Select the motion control function block according to the actual situation, control the spindle to move, and record the initial position of the spindle during movement, recorded as MasterPos0, and this value will be refreshed when entering the next program cycle. During movement, since the spindle is a virtual axis and has no physical mechanical structure, physical shock and mechanical damage can be avoided.

[0106] Step S4: Calculate the feedforward compensation value of the next cycle according to the feedforward compensation value calculation model and the current spindle position.

[0107] In the absence of feedforward regulation, feedback regulation in each cycle can effectively reduce the error between the master and slave axes, but relying solely on traditional feedback algorithms cannot effectively reduce the following error caused by friction in the cylindrical valve servomotor. The slave axis is in a state of catching up with the master axis for a long time, which is prone to form steady-state errors. Adding feedforward compensation values ​​to the process can effectively reduce steady-state errors, improve control accuracy, and enhance the robustness of the entire system.

[0108] The spindle position of the current cycle is substituted into the piecewise cubic Hermite interpolator in step S2, and the interpolation result SYNCAVR is obtained as the feedforward compensation value of the next cycle.

[0109] Step S5: Calculate the deviation value between each slave axis.

[0110] The encoder feedback values ​​of the six slave axes are subtracted in pairs to obtain the deviation value between each axis. The deviation value is an important indicator for judging motion synchronization error.

[0111] Step S6. Determine whether the deviation between the slave axes exceeds SYNCMAX, where SYNCMAX represents the maximum deviation allowed between the slave axes. If the deviation between all the slave axes is less than SYNCMAX, proceed to step S7. If the deviation between any two slave axes is greater than SYNCMAX, directly use the third gradient adjustment method, report a jam error, and stop all axes, exit the process and wait for manual adjustment before re-running.

[0112] Step S7: Determine whether the deviation value between each slave axis exceeds SYNCMID, where SYNCMID represents the normal value of the acceptable deviation between each slave axis. If the deviation value between all slave axes is less than SYNCMID, the first gradient adjustment method is adopted. If the deviation value between any two slave axes is greater than SYNCMID, the second gradient adjustment method is adopted.

[0113] Among them, the first gradient adjustment method refers to: based on the slave axis feedback mean increment and the main axis motion increment, a proportional feedback adjustment method is used in combination with a feedforward compensation value to calculate the slave axis motion increment, and control the slave axis to move according to the calculated motion increment. The second gradient adjustment method refers to an autonomous deviation correction adjustment method. Specifically, the first gradient adjustment method specifically includes the following steps:

[0114] Step S 711. Save the current slave axis feedback average feedAverage1 and master axis position MasterPos1.

[0115] Step S 712. Calculate the slave axis feedback mean increment feedAverageDelta and the master axis motion increment MasterPosDelta. The calculation method is as follows:

[0116] feedAverageDelta=feedAverage1-feedAverage0

[0117] MasterPosDelta=MasterPos1-MasterPos0

[0118] Step S 713. Calculate the slave axis motion increment SlavePosDelta.

[0119] Refer to the instruction manual Figure 3 In this application, the first gradient adjustment method is mainly divided into two parts:

[0120] Proportional feedback adjustment part: The current slave axis feedback mean increment is fed back to the output process through proportional adjustment to compensate for the slave axis following accuracy.

[0121] A feedforward adjustment method based on piecewise cubic Hermite interpolator: The master axis position and the master-slave axis position difference are abstracted into a piecewise cubic Hermite function relationship to obtain the feedforward compensation for eliminating steady-state errors.

[0122] Therefore, the slave axis motion increment SlavePosDelta is:

[0123] SlavePosDelta=SYNCAVR+feedAverageDelta*Kp+MasterPosDelta

[0124] Where SYNCAVR is the feedforward compensation value; Kp is the proportional gain coefficient, which can be set by the user according to actual usage.

[0125] Step S 714. Control the slave axis to move according to the calculated increment.

[0126] The second gradient adjustment method specifically comprises the following steps:

[0127] Step S 721. The spindle stops moving and the binding state between the spindle and the slave axis is released.

[0128] When using the second gradient adjustment method, first, the main axis movement must be paused and the binding relationship between the master and slave axes must be released, so that the slave axis can quickly catch up with the main axis. At this time, the slave axis is in a free state and can trigger the motion control function block to move freely.

[0129] Since the main axis is a virtual axis, after the binding relationship is released, the suspension of the main axis will not affect the movement of the slave axes, and their movement continuity will not be affected by the main axis.

[0130] Step S 722. Each slave axis performs alignment motion based on the current position of the master axis:

[0131] Each slave axis is currently in a free state. The slave axis motion increment is calculated based on the current master axis position:

[0132] SlavePosDelta[i]=MasterPos-SlavePos[i]

[0133] Where SlavePosDelta[i] represents the motion increment of slave axis i, MasterPos represents the current master axis position, and SlavePos[i] represents the current position of slave axis i; i = 1, 2, 3, 4, 5, 6.

[0134] Each slave axis moves separately, and after the movement is completed, the position of each slave axis basically coincides with that of the master axis.

[0135] Step S 723. After each slave axis is in place, the master and slave axis states are re-bound, and the master axis is triggered again at the original target position.

[0136] At this time, the master and slave axes are basically overlapped, and the deviations between the slave axes are basically eliminated. Rebind the master and slave axis relationship, and the master axis triggers the motion control function block again at the original target position to re-realize the synchronous motion of the master and slave axes.

[0137] Step S8: Update the initial slave axis feedback mean value.

[0138] The current initial slave axis feedback average value feedAverage0 and the current master axis motion increment MasterPosDelta are used as the initial slave axis feedback average value feedAverage0′ of the next cycle:

[0139] feedAverage0′=feedAverage0+MasterPosDelta

[0140] Step S9. Determine whether the main axis is in place. Determine whether the control method is finished according to the main axis status. If the main axis completes the movement or stops due to a fault, the slave axis also stops and the process ends. If the main axis does not reach the target position, return to step S3 of the process and enter the next operation cycle. The position parameters related to the main axis and the slave axis are updated.

[0141] Therefore, under normal processes, the movement adjustment mode of the cylindrical valve servomotor alternates between the first gradient and the second gradient, with the first gradient being the majority. The most effective synchronization adjustment can be performed according to actual conditions, and the synchronization error can be controlled within a very small range.

[0142] Furthermore, this embodiment also includes a gradient adjustment synchronous control system for a cylindrical valve servomotor, which is used to implement the above control method. Figure 4 The control system may include two parts: PLC and cylindrical valve machinery. The PLC part is responsible for the operation of motion control program, binding of master and slave axes, fault alarm, etc. The cylindrical valve machinery is mainly responsible for receiving PLC commands and executing actual motion, mainly including servo system, cylinder, hydraulic relay, etc. If there is a software or hardware alarm during operation, the PLC part will give a corresponding alarm code or alarm flag.

[0143] In summary, after reading the present invention document, ordinary technicians in this field can make various other corresponding transformation schemes based on the technical scheme and technical concept of the present invention without creative mental labor, which all fall within the scope of protection of the present invention.

Claims

1. A gradient regulation synchronous control method for a cylindrical valve servomotor, characterized in that: The following steps are involved: Step S1. Establish a virtual main axis, use multiple relays as slave axes that follow the main axis, bind the master-slave axis relationship and calculate the initial slave axis feedback mean; Step S2. Generate a feedforward compensation value calculation model; Step S3. The spindle starts to move and records the initial position; Step S4. Calculate the feedforward compensation value of the next cycle according to the feedforward compensation value calculation model and the current spindle position; Step S5. Calculate the deviation value between each slave axis; Step S6. Determine whether the deviation value between each slave axis exceeds SYNCMAX, where SYNCMAX represents the maximum value of the deviation between each slave axis allowed; if the deviation values ​​between all slave axes are less than SYNCMAX, proceed to step S7; if the deviation between any two slave axes is greater than SYNCMAX, then the process ends and all axes stop; Step S7. Determine whether the deviation value between each slave axis exceeds SYNCMID, where SYNCMID represents the normal value of the acceptable deviation between each slave axis; if the deviation value between all slave axes is less than SYNCMID, the first gradient adjustment method is adopted; if the deviation value between any two slave axes is greater than SYNCMID, the second gradient adjustment method is adopted; The first gradient adjustment method refers to: based on the slave axis feedback mean increment and the main axis motion increment, a proportional feedback adjustment method is used in combination with a feedforward compensation value to calculate the slave axis motion increment, and the slave axis is controlled to move according to the calculated motion increment; the second gradient adjustment method refers to an autonomous deviation correction adjustment method; Step S8. Update the initial slave axis feedback mean; Step S9: Determine whether the spindle is in place. If not, proceed to step S3. If yes, end and all axes stop.

2. A gradient adjustment synchronous control method for a cylindrical valve servomotor according to claim 1, characterized in that: The feedforward compensation value calculation model is a piecewise cubic Hermite interpolator, and the generation method of the piecewise cubic Hermite interpolator is: Step S 21. Record the current spindle position MasterPos; Step S 22. The main axis starts to move, driving the slave axis to move according to the process without feedforward adjustment; Step S 23. After this cycle is over, the position difference between the master and slave axes is calculated and sent to the piecewise cubic Hermite interpolator in one-to-one correspondence with MasterPos; Step S 24. Determine whether the spindle stroke is completed. If the spindle stroke is completed, exit or repeat the process again to take the average. If the stroke is not completed, enter the next cycle.

3. The gradient adjustment synchronous control method of a cylindrical valve servomotor according to claim 1, characterized in that: In step S6, if the deviation between any two slave axes is greater than SYNCMAX, a third gradient adjustment method is used for adjustment. The third gradient adjustment method specifically includes: reporting a card group failure and manual adjustment.

4. A gradient adjustment synchronous control method for a cylindrical valve servomotor according to any one of claims 1 to 3, characterized in that: The first gradient adjustment method specifically comprises the following steps: Step S 711. Save the current slave axis feedback average feedAverage1 and master axis position MasterPos1; Step S 712. Calculate the slave axis feedback mean increment feedAverageDelta and the master axis motion increment MasterPosDelta; Step S 713. Calculate the slave axis motion increment SlavePosDelta: SlavePosDelta=SYNCAVR+feedAverageDelta*Kp+MasterPosDelta In the formula, SYNCAVR is the feedforward compensation value, and Kp is the proportional gain coefficient; Step S 714. Control the slave axis to move according to the calculated motion increment.

5. A gradient adjustment synchronous control method for a cylindrical valve servomotor according to any one of claims 1 to 3, characterized in that: The second gradient adjustment method specifically comprises the following steps: Step S 721. The spindle stops moving and the binding state between the spindle and the slave axis is released; Step S 722. Each slave axis performs alignment movement based on the current position of the master axis; Step S 723. After each slave axis is in place, the master and slave axis states are re-bound, and the master axis is triggered again at the original target position.

6. A gradient adjustment synchronous control method for a cylindrical valve servomotor according to claim 5, characterized in that: The step S 722 Specifically include: Based on the current spindle position, calculate the slave axis motion increment: SlavePosDelta[i]=MasterPos-SlavePos[i] Where, SlavePosDelta[i] represents the motion increment of slave axis i, MasterPos represents the current master axis position, and SlavePos[i] represents the current position of slave axis i. Each slave axis moves separately, and after the movement is completed, the position of each slave axis basically coincides with that of the master axis.

7. The gradient adjustment synchronous control method of a cylindrical valve servomotor according to claim 1, characterized in that: Step S8 specifically refers to: taking the current initial slave axis feedback average value feedAverage0 and the current master axis motion increment MasterPosDelta as the initial slave axis feedback average value feedAverage0′ of the next cycle: feedAverage0′=feedAverage0+MasterPosDelta.

Images