Non-overshoot motor speed given soft start method

Abstract

The invention relates to an overshoot-free motor speed given soft start method, which comprises the following steps of: performing real-time feedback observation on an output value Y and first-order differential output YA, dynamically calculating an output amplitude and a switching moment of second-order differential output YB, and dynamically adjusting the output of the YB when a given X is continuously changed, so that the output value Y not only meets the rising / falling time requirement, but also meets the switching moment of the first-order differential output YA. The optimal adjustment of control output is ensured; and by performing real-time feedback judgment on the given value X, the output value Y and the first-order differential value YA, the output amplitude and the switching time of the first-order differential output YA are dynamically adjusted, and it is ensured that the output value Y cannot generate overshoot, undershoot or oscillation. According to the invention, the characteristics of no overshoot and no oscillation can be realized when the acceleration is paused and the given value is changed.

Description

Technical Field

[0001] The invention belongs to the technical field of electrical control, and in particular to a motor speed setting soft starting method without overshoot. Background Art

[0002] In an electrical control system, when the operator gives a step command signal (such as speed, torque or position), the controlled object in the electrical control system is generally required to reach the command signal value according to a predetermined function generator. The purpose is to limit the rate of change of the adjustment process, reduce electrical and mechanical shocks, and achieve the control target safely, stably and quickly.

[0003] Currently, the function generators used in given links mainly include: first-order inertia function generator, ordinary ramp function generator and ramp function generator with rounded corners.

[0004] After the step command signal is input to the ramp function generator with rounded corners, the starting and ending segments of the output signal are arc-shaped waveforms, and the middle segment is a fixed slope waveform, which makes the starting and ending segments smooth, greatly reducing electrical and mechanical shocks. Therefore, this type of ramp function generator is widely used. The ramp function generator with rounded corners is based on pre-calculation. The principle is to calculate the slope of different segments according to the set acceleration / deceleration time and rounding time to generate a curve. However, if the input value changes in real time, the function generator cannot respond to the change in real time, which will cause overshoot, undershoot and oscillation. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and to provide a motor speed setting soft starting method without overshoot, which can achieve the characteristics of no overshoot and no oscillation when the speed increase is paused and the given value is changed.

[0006] The present invention solves the technical problem by adopting the following technical solutions:

[0007] A motor speed setting soft starting method without overshoot comprises the following steps:

[0008] Step 1: Input the given value X into the function generator. k , Rise time T U , Fall time T D , Start rounding time T UR and the end rounding time T DR ;

[0009] Step 2: According to the rise time T U , Fall time T D Calculate the maximum first-order differential output Y during acceleration A1 , maximum first-order differential output Y during deceleration A2 And the second-order differential output YB1 , Y B2 , Y B3 and Y B4 ;

[0010] Step 3: Combine the current input with the given value X k , output value Y k-1 And the first-order differential output Y A.k-1 , the difference between input and output Δ k =X k -Y k-1 , the difference between two consecutive inputs in the current sampling period E k =X k -X k-1 , real-time calculation of the acceleration value Y output by the second-order differential output sampling period B.k ; Wherein, the variable subscript K represents the current sampling period value, and the variable subscript K-1 represents the previous sampling period value;

[0011] Step 4: The acceleration value Y outputted during the second-order differential output sampling period calculated in step 3 B,k , calculate the first-order differential output process variable Y′ A.K , where Y′ A.K =Y B,k +Y A.K-1 ;

[0012] Step 5: Determine E k Is it equal to 0? k ≠0, judge Δ k With Y B.k , Y A,k polarity and make Y A ' .K Perform corresponding operations and obtain Y A ;

[0013] Step 6: Based on the Y obtained in step 5 A , calculate Y K ;

[0014] Step 7: If you need to pause the acceleration, use X K =Y K-1 , T U =T D =0, and return to step 1.

[0015] Moreover, the specific implementation method of step 2 is:

[0016] Y A1 =1 / T U

[0017] Y A2 =1 / T D

[0018] Y B1 =1 / T U *1 / T UR

[0019] Y B2 =1 / T U *1 / T DR

[0020] Y B3 =1 / T D *1 / T UR

[0021] Y B4 =1 / T D *1 / T DR .

[0022] Moreover, the specific implementation method of step 3 is:

[0023] ①When Y k-1 >0,Y A.k-1 >0 hours;

[0024] If Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 <Y A1 , output Y B.k =Y B1 ;

[0025] If Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 >=Y A1 , output Y B.k =0;

[0026] If Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Δ k >0.5*Y 2 A.k-1 / Y B2 Output Y B.k =0;

[0027] If Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / YB2 , Δ k ≤ 0.5 * Y 2 A.k-1 / Y B2 Output Y B.k = -Y B2 ;

[0028] ② When Y k-1 > 0, Y A.k-1 ≤ 0;

[0029] If -Δ k > 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 , Y A.k-1 > -Y A2 , output Y B.k = -Y B3 ;

[0030] If -Δ k > 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 , Y A.k-1 ≤ -Y A2 , output Y B.k = 0;

[0031] If -Δ k ≤ 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 , -Δ k > 0.5 * Y 2 A.k-1 / Y B4 , output Y B.k = 0;

[0032] If -Δ k ≤ 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 , -Δ k ≤ 0.5 * Y 2 A.k-1 / Y B4 , output Y B.k = Y B4 ;

[0033] ③ When Y k-1 ≤ 0, Y A.k-1 > 0;

[0034] If Δ k>0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 ,Y A.k-1 <Y A2 ,output Y B.k =Y B3 ;

[0035] If Δ k >0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 ,Y A.k-1 >=Y A2 ,output Y B.k =0;

[0036] If Δ k <=0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 ,Δ k >0.5*Y 2 A.k-1 / Y B3 output Y B.k =0;

[0037] If Δ k <=0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 ,Δ k <=0.5*Y 2 A.k-1 / Y B3 output Y B.k =-Y B4 ;

[0038] ④When Y k-1 <=0,Y A.k-1 <=0;

[0039] If -Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 ,Y A.k-1 >-Y A1 ,output Y B.k =-Y B1 ;

[0040] If -Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / YB2 , Y A.k-1 <=-Y A1 , output Y B.k =0;

[0041] If -Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , -Δ k >0.5*Y 2 A.k-1 / Y B2 , output Y B.k =0;

[0042] If -Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , -Δ k <=0.5*Y 2 A.k-1 / Y B2 , output Y B.k =Y B2 .

[0043] Moreover, the specific implementation method of step 4 is:

[0044] Y′ A.K =Y B,k +Y A.K-1

[0045] Moreover, the specific method of step 5 is:

[0046] If E k ≠0, then judge Δ k ;

[0047] If Δ k With Y B.k Opposite polarity or Δ k With Y A ' .K The polarity is opposite, making Y A ' .K =0, to prevent overshoot;

[0048] When |Δ k |>max(Y B3 ,Y B4 ), then Y A.K =Y A ' .K ,

[0049] When |Δ k |<=max(YB3 ,Y B4 ), then Y A.K =Δ k .

[0050] Moreover, the specific implementation method of step 6 is:

[0051] Y K =Y K-1 +Y A.K .

[0052] The advantages and positive effects of the present invention are:

[0053] The present invention performs the following steps on the output value Y and the first-order differential output Y. A Perform real-time feedback observation and dynamically calculate the second-order differential output Y B The output amplitude and switching time of the controller ensure that Y is dynamically adjusted when X changes continuously. B Output, so that the output value Y not only meets the rise / fall time requirements, but also ensures the best control output regulation; and by the given value X, output value Y, first-order differential value Y A Perform real-time feedback judgment and dynamically adjust the first-order differential output Y A The output amplitude and switching time of the invention ensure that the output value Y will not produce overshoot, undershoot or oscillation. Figure 4 Compared with the conventional method (as shown in Figure 3 As shown in the figure, it can achieve the characteristics of no overshoot and no oscillation when the speed is paused and the given value is changed. For example, in the metallurgical rolling process, if an abnormality is found during the speed increase and a pause is required, the output value of the patented method will immediately stop at the current value, while the conventional method will continue to increase the given value; and when changing the given value, the patented method can achieve that the output value will not overshoot under any given value change. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of a ramp function generator with rounded corners;

[0055] Figure 2 Schematic diagram of the transmission function of the algorithm of the present invention;

[0056] Figure 3 This is a schematic diagram of the conventional method of speed increase and pause;

[0057] Figure 4 This is a schematic diagram of the speed increase and pause of this method;

[0058] Figure 5 The calculation flow chart of the algorithm of the present invention is

[0059] Figure 6 A schematic diagram comparing the speed-up and pause effects of the present invention and the traditional method;

[0060] Figure 7 This is a schematic diagram comparing the effects of changing given values ​​between the present invention and the traditional method.

[0061] Figure 8 Schematic diagram for comparison of the extreme cases of different methods. DETAILED DESCRIPTION

[0062] The present invention is further described in detail below with reference to the accompanying drawings.

[0063] The working principle of the ramp function generator with rounded corners is as follows Figure 1 As shown, the relevant parameters are defined as follows: input given value X, output value Y, rise time T U , fall time T D , starting rounding time T UR , end rounding time T DR , T U.eff =T U +T UR / 2+T DR / 2 is the time it takes for the output absolute value |Y| to increase from 0% to 100% of the maximum command value, T D.eff =T D +T UR / 2+T DR / 2 is the time it takes for the output absolute value |Y| to decrease from the maximum command value 100% to 0%, Y A is the first-order differential output, Y B is the second-order differential output.

[0064] The principle of the present invention to solve the real-time response of the ramp function generator is: real-time monitoring of the output Y of the function generator, the first-order differential output Y A , switch Y at the appropriate time A and Y B The value of can make Y reach the given X value accurately without overshoot, undershoot or oscillation. When this method is used for speed setting, it can achieve the characteristic of no overshoot when the speed is increased and the given value is changed.

[0065] Depend on Figure 1 It can be seen that in order to meet the requirement that Y reaches X at the end of acceleration / deceleration, it is necessary to meet the requirement that Y A =0, so it is necessary to switch Y at the appropriate time B , combined with physical principles, we can get the following formula:

[0066] Y A / Y B =t

[0067] 0.5*Y B *t 2 =XY

[0068] When observing Y = X-0.5*Y A *Y A / Y B When the switching output corresponds to the second-order differential output Y B , achieve Y A When it drops to 0, the output value Y is equal to the given value X.

[0069] When the output Y is about to reach the input given X, it is necessary to observe X, Y and Y in real time. A The value of Y is adjusted in real time A value to ensure that the output Y accurately reaches the given value X.

[0070] When calculating with discrete algorithm, Y A.K =Y B.K +Y A.K-1 , Y K =Y A.K +Y K-1 , all variables with subscript K represent the current sampling period value, and variables with subscript K-1 represent the previous sampling period value.

[0071] A motor speed setting soft start method without overshoot, the transmission function diagram is as follows Figure 2 As shown, the algorithm flow chart is as follows Figure 5 As shown, the following steps are included:

[0072] Step 1: Input the given value X into the function generator. k , rise time T U , fall time T D , starting rounding time T UR , end rounding time T DR .

[0073] Step 2: According to the rise time T U , Fall time T D Calculate the maximum first-order differential output Y during acceleration A1 , maximum first-order differential output Y during deceleration A2 And the second-order differential output Y B1 , Y B2 , Y B3 and Y B4 .

[0074] The specific implementation method of step 2 is:

[0075] Y A1 =1 / T U

[0076] Y A2 =1 / T D

[0077] Y B1 =1 / TU *1 / T UR

[0078] Y B2 =1 / T U *1 / T DR

[0079] Y B3 =1 / T D *1 / T UR

[0080] Y B4 =1 / T D *1 / T DR .

[0081] Step 3: Combine the current input with the given value X k , output value Y k-1 And the first-order differential output Y A.k-1 , the difference between input and output Δ k =X k -Y k-1 , the difference between two consecutive inputs in the current sampling period E k =X k -X k-1 Real-time calculation of the second-order differential output of the acceleration value Y output in the current sampling period B.k ; Among them, the variable subscript K represents the current sampling period value, and the variable subscript K-1 represents the previous sampling period value.

[0082] The specific implementation method of step 3 is:

[0083] ①When Y k-1 >0,Y A.k-1 >0 hours;

[0084] If Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 <Y A1 , output Y B.k =Y B1 ;

[0085] If Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 >=Y A1 , output Y B.k =0;

[0086] If Δ k<= 0.5 * (Y A.k-1+ Y B1 ) 2 / Y B2 ,Δ k > 0.5 * Y 2 A.k-1 / Y B2 Output Y B.k = 0;

[0087] If Δ k <= 0.5 * (Y A.k-1+ Y B1 ) 2 / Y B2 ,Δ k <= 0.5 * Y 2 A.k-1 / Y B2 Output Y B.k = -Y B2 ;

[0088] ② When Y k-1 > 0, Y A.k-1 <= 0;

[0089] If -Δ k > 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 ,Y A.k-1 > -Y A2 ,Output Y B.k = -Y B3 ;

[0090] If -Δ k > 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 ,Y A.k-1 <= -Y A2 ,Output Y B.k = 0;

[0091] If -Δ k <= 0.5 * (Y A.k-1+ Y B3 ) 2 / Y B4 ,-Δ k > 0.5 * Y 2 A.k-1 / Y B4 ,Output Y B.k = 0;

[0092] If -Δ k <= 0.5 * (Y A.k-1+ Y B3) 2 / Y B4 ,-Δ k ≤0.5*Y 2 A.k-1 / Y B4 , output Y B.k =Y B4 ;

[0093] ③ When Y k-1 ≤0, Y A.k-1 >0;

[0094] If Δ k >0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 , Y A.k-1 <Y A2 , output Y B.k =Y B3 ;

[0095] If Δ k >0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 , Y A.k-1 ≥Y A2 , output Y B.k =0;

[0096] If Δ k ≤0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 , Δ k >0.5*Y 2 A.k-1 / Y B3 Output Y B.k =0;

[0097] If Δ k ≤0.5*(Y A.k-1+ Y B3 ) 2 / Y B4 , Δ k ≤0.5*Y 2 A.k-1 / Y B3 Output Y B.k =-Y B4 ;

[0098] ④ When Y k-1 ≤0, Y A.k-1 ≤0;

[0099] If -Δk >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 >-Y A1 , output Y B.k =-Y B1 ;

[0100] If -Δ k >0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , Y A.k-1 <=-Y A1 , output Y B.k =0;

[0101] If -Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , -Δ k >0.5*Y 2 A.k-1 / Y B2 , output Y B.k =0;

[0102] If -Δ k <=0.5*(Y A.k-1+ Y B1 ) 2 / Y B2 , -Δ k <=0.5*Y 2 A.k-1 / Y B2 , output Y B.k =Y B2 .

[0103] Step 4: The second-order differential output Y calculated in step 3 B,k , calculate the first-order differential output process variable Y′ A.K .

[0104] The specific implementation method of step 4 is:

[0105] Y A ' .K =Y B,k +Y A.K-1

[0106] Step 5: Determine E k Is it equal to 0? k ≠0, judge E k With Y B.kpolarity and make Y A,k Perform corresponding operations. k = 0, according to M k Calculate Y A .

[0107] The specific method of step 5 is: if E k ≠0, then judge Δ k , if Δ k With Y B.k Opposite polarity or Δ k With Y A ' .K The polarity is opposite, making Y A ' .K =0, to prevent overshoot;

[0108] When |Δ k |>max(Y B3 ,Y B4 ), then Y A.K =Y A ' .K

[0109] When |Δ k |<=max(Y B3 ,Y B4 ), then Y A.K =Δ k

[0110] Step 6: Based on the Y obtained in step 5 A , calculate Y K .

[0111] The specific implementation method of step 6 is:

[0112] Y K =Y K-1 +Y A.K .

[0113] Step 7: If you need to pause the acceleration, Figure 4 , so that X K =Y K-1 , T U =T D = 0. Step 7 only changes the input conditions of the system, and the system still executes the operations according to steps 1 to 6 in a loop.

[0114] The present invention (such as Figure 4 Compared with the conventional method (as shown in Figure 3 As shown in the figure, it can achieve the characteristics of no overshoot and no oscillation when the speed is paused and the given value is changed. For example, in the metallurgical rolling process, if an abnormality is found during the speed increase and a pause is required, the output value of the patented method will immediately stop at the current value, while the conventional method will continue to increase the given value (such as Figure 6 When changing the given value, the patented method can ensure that the output value will not overshoot under any given value change (such as Figure 7 As shown); in the output under extreme conditions (small given value, short rise and fall time), the patented method can achieve no overshoot and no oscillation (such as Figure 8 shown).

[0115] It should be emphasized that the embodiments described in the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation manner. Any other implementation manners derived by those skilled in the art based on the technical solution of the present invention also fall within the scope of protection of the present invention.

Claims

1. A motor speed setting soft starting method without overshoot, characterized in that: The following steps are involved: Step 1: Input the given value X into the function generator. k , Rise time T U , Fall time T D , Start rounding time T UR and the end rounding time T DR ; Step 2: According to the rise time T U , Fall time T D Calculate the maximum first-order differential output Y during acceleration A1 , maximum first-order differential output Y during deceleration A2 And the second-order differential output Y B1 , Y B2 , Y B3 and Y B4 ; Step 3: Combine the current input with the given value X k , output value Y k-1 And the first-order differential output Y A.k-1 , the difference between input and output Δ k =X k -Y k-1 , the difference between two consecutive inputs in the current sampling period E k =X k -X k-1 , real-time calculation of the acceleration value Y output by the second-order differential output sampling period B.k ; Wherein, the variable subscript K represents the current sampling period value, and the variable subscript K-1 represents the previous sampling period value; Step 4: The acceleration value Y outputted during the second-order differential output sampling period calculated in step 3 B,k , calculate the first-order differential output process variable Y′ A.K , where Y′ A.K =Y B,k +Y A.K-1 ; Step 5: Determine E k Is it equal to 0? k ≠0, judge Δ k With Y B.k , Y A,k polarity and make Y′ A.K Perform corresponding operations and obtain Y A ; Step 6: Based on the Y obtained in step 5 A , calculate Y K ; Step 7: If you need to pause the acceleration, use X K =Y K-1 , T U =T D =0, and return to step 1.

2. The method for soft starting a motor with given speed without overshoot according to claim 1, characterized in that: The specific implementation method of step 2 is: Y A1 =1 / T U Y A2 =1 / T D Y B1 =1 / T U *1 / T UR Y B2 =1 / T U *1 / T DR Y B3 =1 / T D *1 / T UR Y B4 =1 / T D *1 / T DR 。 3. The motor speed setting soft starting method without overshoot according to claim 1 is characterized in that: The specific implementation method of step 3 is: ①When Y k-1 >0,Y A·k-1 >0 hours; If Δ k >0.5*(Y A·k-1+ Y B1 ) 2 / Y B2 ,Y A·k-1 <Y A1 ,output Y B.k =Y B1 ; If Δ k > 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , Y A·k-1 >= Y A1 , output Y B.k = 0; If Δ k <= 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , Δ k > 0.5 * Y 2 A·k-1 / Y B2 Output Y B.k = 0; If Δ k <= 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , Δ k <= 0.5 * Y 2 A·k-1 / Y B2 Output Y B.k = -Y B2 ; ②When Y k-1 >0,Y A·k-1 <=0; If -Δ k > 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 ,Y A·k-1 > -Y A2 ,output Y B.k = -Y B3 ; If -Δ k >0.5*(Y A·k-1+ Y B3 ) 2 / Y B4 ,Y A·k-1 <= -Y A2 ,output Y B.k = 0; If -Δ k <= 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , -Δ k > 0.5 * Y 2 A·k-1 / Y B4 , output Y B.k = 0; If -Δ k <= 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , -Δ k <= 0.5 * Y 2 A·k-1 / Y B4 , output Y B.k = Y B4 ; ③When Y k-1 <=0, Y A·k-1 >0 hours; If Δ k > 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , Y A·k-1 < Y A2 , output Y B.k = Y B3 ; If Δ k > 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , Y A·k-1 >= Y A2 , output Y B.k = 0; If Δ k <= 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , Δ k > 0.5 * Y 2 A·k-1 / Y B3 Output Y B.k = 0; If Δ k <= 0.5 * (Y A·k-1+ Y B3 ) 2 / Y B4 , Δ k <= 0.5 * Y 2 A·k-1 / Y B3 Output Y B.k = -Y B4 ; ④When Y k-1 <=0, Y A·k-1 <=0; If -Δ k > 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , Y A·k-1 > -Y A1 , output Y B.k = -Y B1 ; If -Δ k > 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 ,Y A·k-1 <= -Y A1 ,output Y B.k = 0; If -Δ k <= 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , -Δ k > 0.5 * Y 2 A·k-1 / Y B2 , output Y B.k = 0; If -Δ k <= 0.5 * (Y A·k-1+ Y B1 ) 2 / Y B2 , -Δ k <= 0.5 * Y 2 A·k-1 / Y B2 , output Y B.k = Y B2 .

4. The method for soft starting a motor with given speed without overshoot according to claim 1, characterized in that: The specific implementation method of step 4 is: AND' A.K =And B,k +Y A.K-1 5. The method for soft starting a motor with given speed without overshoot according to claim 1, characterized in that: The specific method of step 5 is: If E k ≠0, then judge Δ k ; If Δ k With Y B.k Opposite polarity or Δ k and Y′ A.K The polarity is opposite, making Y′ A.K =0, to prevent overshoot; When |Δ k | > max(Y B3 , Y B4 ), then Y A.K = Y' A.K , When |D k |<=max(Y B3 ,Y B4 ), Y A.K =D k 。 6. A motor speed setting soft starting method without overshoot according to claim 5, characterized in that: The specific implementation method of step 6 is: AND K =And K-1 +Y A.K 。

Images