JP2008274460A - Method for winding roving yarns in roving frame - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for winding a roving yarn in a roving frame by which a wound roving yarn (a roving package) having an expected shoulder shape can be formed stably without causing the collapse of the shoulder. <P>SOLUTION: The growth ΔΦ of a roving yarn-winding diameter at layer change required when winding the roving yarn so as to form the expected shoulder shape is estimated from the information of a roving tension detector, and the wound roving yarn is formed by determining the inversion position of a bobbin rail by using the estimated growth ΔΦ of the roving yarn-winding diameter. The ΔΦ-estimating model is formed by assuming that the growth ΔΦ of the roving yarn-winding diameter monotonously increases according to the increase of the wound diameter of the roving yarn. The growth ΔΦ of the wound diameter of the roving yarn is calculated from an updated model by updating the ΔΦ-estimating model so that the difference between the growth ΔΦp of the wound diameter of the roving yarn estimated based on the roving tension signal obtained from the roving tension detector, and the growth ΔΦq of the wound diameter of the roving yarn calculated by the model before the update may be reduced. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method for winding a roving yarn in a roving machine.

  In general, in a roving machine, a roving yarn fed at a constant speed from a front roller is twisted on the bobbin wheel while twisting the roving yarn by a rotational speed difference between a flyer rotating at a constant speed and a bobbin rotating at a higher speed. The bobbin is wound up to form a roving yarn (coarse yarn package). The bobbin is mounted on the spindle and moved up and down together with the bobbin rail. Each time the direction of the bobbin rail lifting / lowering motion changes, that is, each time a new roving layer is formed on the surface of the roving bobbin, the moving distance (elevating stroke) of the bobbin rail is shortened, and both ends of the roving bobbin are moved. Coarse yarn is wound so as to have a frustum shape. The shape of the shoulder of the roving bobbin, that is, the shape of the frustum-shaped portion at both ends of the roving bobbin, corresponds to the roasting bobbin diameter (coarse bobbin winding diameter) that increases each time the winding layer (winding layer) is switched (layer change). Thus, the bobbin rail reversing position is gradually changed to the direction in which the lift length becomes shorter with respect to the lift length at the start of winding until the full pipe is formed.

  In order to realize the planned shoulder shape, it is important to predict the increase amount (ΔΦ) of the coarse yarn winding diameter with respect to the current layer change, which is an element that determines the inversion position of the bobbin rail. Conventionally, in order to estimate this ΔΦ, a roving machine in which a database capable of calculating ΔΦ from spinning conditions such as fiber type, roving weight, number of rotations, number of presser windings and the like is incorporated in a main control program has been used.

  In addition to the main motor that drives the draft part to rotate, it has a control motor that can change the bobbin rotation independently. The bobbin rail can be moved up and down at the up / down switching position, and the bobbin winding diameter (coarse yarn winding diameter) There has been proposed a roving winding method for winding a roving on a bobbin by controlling the rotation of a control motor so as to achieve an appropriate bobbin rotation according to the above (for example, see Patent Document 1). The method for winding the roving yarn of Patent Document 1 is to predict the bobbin winding diameter of the roving yarn bobbin (coarse yarn winding) after the next raising / lowering switching until the raising / lowering switching of the bobbin rail is performed and the next raising / lowering switching position is reached. The control amount of the bobbin rotation corresponding to is obtained in advance, and the bobbin rotation is controlled by the control amount at the same time as the next elevation switching.

  Specifically, one layer thickness of the roving yarn is corrected based on the roving yarn tension from the roving yarn tension detecting device until the bobbin rail switches the raising / lowering direction and reaches the next raising / lowering switching position. Based on the corrected one layer thickness of the roving and the current bobbin winding diameter, the bobbin winding diameter after the next elevation switching is predicted. Then, the frequency dividing ratio corresponding to the bobbin winding diameter after the switching is obtained and stored, and the frequency dividing ratio is set in the frequency divider at the next elevation switching, and the bobbin rotation is switched. There has been proposed a roving winding method for controlling rotation corresponding to the bobbin winding diameter.

  The correction of the thickness of one layer of the roving is to calculate a correction value (takes positive and negative values) ε for the thickness of the roving (a constant increment Δd) based on the roving tension, and is set in advance. It is calculated by multiplying the deviation between the tension target value and the detected detected value by a certain coefficient. The next bobbin winding diameter D is calculated by the following equation.

D (next bobbin winding diameter) = D (current bobbin winding diameter) + 2 × (Δd−ε)
JP-A-8-232123

  If the full bobbin (full coarse yarn winding) diameter and lift length are the same and the spinning raw material is the same, the amount (winding volume) of the coarse yarn wound around the full bobbin is determined by the shoulder shape, and the desired winding If the winding is performed so that the shoulder shape corresponds to the volume, the planned winding volume can be secured.

  As shown in FIG. 4, generally, the shoulder shape of the roving bobbin F forms a certain angle θ with the central axis direction of the bobbin B. In order to increase the volume of the roving yarn wound per roving yarn winding, it is necessary to increase the shoulder angle θ of the roving yarn winding F. However, there is an upper limit on the shoulder angle θ of the roving yarn F that can be formed without causing shoulder collapse depending on the spinning conditions such as the spinning roving weight and fiber type. Therefore, there is a case where the roving yarn winding volume is increased by winding the roving yarn so that the shoulder shape becomes an outwardly convex curved surface.

  Realizing the planned shoulder shape secures the planned take-up volume. However, the amount of increase (ΔΦ) in the diameter of the roving yarn is affected by the spinning conditions such as the roving yarn type (raw material), selenium, the number of twists, the rotational speed of the flyer, the winding tension, the winding pitch, and the temperature and humidity. It is said. Then, when test spinning (trial spinning) is performed in advance and appropriate spinning conditions are determined and the roving is wound under the spinning conditions, the roving is wound with a predetermined shoulder shape. It becomes possible. However, performing the trial spinning every time the spinning conditions are changed has a problem that the raw materials and labor of the trial spinning are wasted and the operation efficiency of the rough spinning machine is lowered. And the said problem becomes more remarkable by changing a spinning condition frequently corresponding to multi-product small lot production.

  In order to smoothly wind up the roving, it is necessary to prevent the roving on the shoulder of the bobbin from being peeled off from the roving of the front layer (so-called shoulder collapse). Although it depends on the exit conditions, the accuracy of the bobbin rail reversal position of about 0.1 mm is required. If the prediction accuracy of the amount of increase in the roving yarn diameter (ΔΦ) is poor, shoulder collapse is likely to occur.

  The inventor of the present application uses a database capable of calculating the amount of increase in the diameter of a roast yarn ΔΦ from the spinning conditions such as fiber type, roving weight, number of rotations, number of presser turns, etc. The predicted value (estimated value) of ΔΦ was compared with the actually measured value. As a result, it was found that the actual measured values tend to increase monotonously with the increase in the diameter of the wound yarn, although there is variation. Therefore, depending on the spinning conditions, it is necessary to change the shoulder angle θ small in order to avoid shoulder collapse. Further, when the shoulder angle θ is changed to be small, the shoulder angle θ is set to be small in order to avoid the collapse of the shoulder, and the amount of roving of the full roving is reduced.

  In the roving yarn winding method disclosed in Patent Document 1, the information on the roving yarn tension detector is used to correct the one-layer thickness of the roving yarn. The correction method of Patent Document 1 calculates a correction value (takes a positive or negative value) ε for a single layer thickness (a constant increment Δd) of the roving yarn based on the roving yarn tension. It is calculated by multiplying the deviation between the detected tension target value and the detected detection value by a certain coefficient. That is, in Patent Document 1, the information of the roving yarn tension detector is used to correct the constant increment Δd every time the roving layer increases. However, the correction value ε is obtained by simply multiplying the deviation between the target tension value and the detected detection value by a certain coefficient, and the inventor finds that the increase amount ΔΦ of the coarse yarn winding diameter increases with the increase of the coarse yarn winding diameter. No consideration has been given to the monotonous increase.

  In Patent Document 1, in a configuration in which bobbin rail up / down switching is counted and shift control is performed for bobbin rotation corresponding to the bobbin winding diameter to be wound each time the switching is performed, calculation of the bobbin rotational speed generates a switching signal. This can be executed only after the operation has been completed, and the purpose is to solve the problem that the control of bobbin rotation is delayed. In order to achieve the object, there is no problem even if it is not considered that the increase amount ΔΦ of the roving yarn diameter found by the present inventor increases monotonously with the increase of the roving yarn winding diameter. However, it is difficult to stably form a shoulder-shaped roving wound with no shoulder collapse without trial spinning according to the spinning conditions.

  The present invention has been made in view of the above-described conventional problems, and an object of the present invention is to stably form a planned shoulder-shaped roving (coarse yarn package) without causing shoulder collapse. An object of the present invention is to provide a method for winding a roving yarn in a roving machine.

  In order to achieve the above-mentioned object, the invention according to claim 1 is characterized in that the coarse yarn winding diameter increase amount ΔΦ at the time of layer change necessary for winding up the coarse yarn so as to have a preset shoulder shape is defined as: A roving yarn winding method in a roving machine, which estimates from the roving yarn tension information of a tension detector and determines the reversal position of the bobbin rail using this estimated roving winding diameter increase amount ΔΦ to form the roving yarn winding. A model assuming that the amount of increase of the roving yarn diameter ΔΦ monotonously increases with the increase of the roving yarn diameter is prepared, and for each layer change, the amount of increase of the roving yarn diameter ΔΦp estimated from the roving yarn tension information and the above-mentioned before update The model is updated so that the difference from the coarse yarn winding diameter increase amount ΔΦq calculated using the model is small, and the estimated coarse yarn winding diameter increase amount ΔΦ is calculated from the updated model.

  In the present invention, when forming the roving, the roving yarn diameter increase amount ΔΦ at the time of layer change necessary for winding the roving so as to have a preset shoulder shape is obtained from the information of the roving tension detector. presume. The estimated position of the bobbin winding diameter increase ΔΦ is used to determine the reversal position of the bobbin rail to form the roving bobbin. In order to obtain the estimated coarse yarn winding diameter increase amount ΔΦ, a model is prepared on the assumption that the coarse yarn winding diameter increase amount ΔΦ monotonously increases as the coarse yarn winding diameter increases. Then, for each layer change, the model is adjusted so that the difference between the roving yarn diameter increase amount ΔΦp estimated from the roving yarn tension information and the roving yarn winding diameter increase amount ΔΦq calculated using the model before update is reduced. The estimated coarse yarn winding diameter increase amount ΔΦ is calculated from the updated model. Therefore, the planned shoulder-shaped roving (coarse yarn package) can be stably formed without causing shoulder collapse.

  The invention according to claim 2 is the invention according to claim 1, wherein the monotonous increase is a primary increase. The increase amount ΔΦ of the roving yarn diameter accompanying the increase in the roving yarn diameter is not exactly a primary increase, but there is no problem even if it is assumed to be a primary increase. In the present invention, since the primary increase is made, the calculation is simplified.

  According to a third aspect of the present invention, in the invention according to the first or second aspect, the roving machine is configured so that the rotational speed of the flyer increases with the increase of the roving yarn diameter after the roving yarn winding diameter reaches a predetermined value. In the above model, it is assumed that the increase amount ΔΦ of the coarse yarn diameter during the shift is proportional to the value obtained by dividing the initial speed of the flyer rotational speed by the current rotational speed. In this invention, even when the rotational speed of the flyer is controlled so as to decelerate with an increase in the roving yarn diameter after the roving yarn winding diameter reaches a predetermined value, the roving yarn winding can be stably formed. Can do.

  According to the present invention, it is possible to stably form a planned shoulder-shaped roving (coarse yarn package) without causing shoulder collapse.

Hereinafter, an embodiment embodying the present invention will be described with reference to FIGS.
As shown in FIG. 1, the front roller 11 rotates through a gear train (none of which is shown) disposed between one end of the rotating shaft 11a and a driving shaft that is driven to rotate by the main motor M. It is designed to be driven. A driven gear 13 is fitted and fixed to the upper portion of the flyer 12 so as to be integrally rotatable. The driven gear 13 and the flyer 12 are rotationally driven through the drive gear 15 fitted to the rotary shaft 14 by the rotation of the rotary shaft 14 through which the rotation of the driving shaft is transmitted via a belt transmission mechanism (not shown). Is done. On the other hand, a driven gear 17 a is fixed to a bobbin wheel 17 provided on the bobbin rail 16. On the rotating shaft 19 to which the driving gear 18 meshing with the driven gear 17a is fitted and fixed, the rotating force of the driving shaft and the rotating force of the winding motor 21 that is driven to change speed through the inverter 20b are the differential gears. It is synthesized by the mechanism 22 and transmitted.

  A lifter rack 23 is fixed to the bobbin rail 16. The rotation of the drive shaft 27 is transmitted to the rotary shaft 25 fitted with the gear 24 meshing with the lifter rack 23 via the switching mechanism 28 and the gear train. The drive shaft 27 is driven by an elevating motor 26 that is driven to change speed via an inverter 20c. The switching mechanism 28 includes an intermediate shaft 29, a pair of gear trains 30, 31 provided between the intermediate shaft 29 and the drive shaft 27, and an electromagnetic clutch that transmits the rotation of the gear trains 30, 31 to the intermediate shaft 29. 32, 33. The rotational direction of the rotary shaft 25, that is, the direction of up-and-down movement of the bobbin rail 16 is changed by excitation and demagnetization of the electromagnetic clutches 32 and 33. A rotary encoder 34 as a sensor for detecting the moving direction of the bobbin rail 16 is connected to the end of the rotating shaft 25.

  Between the front roller 11 and the flyer top 12a, a roving yarn tension detector 35 for detecting the roving yarn tension by continuously detecting the position of the roving yarn R from the front roller 11 to the flyer top 12a without contact. Is provided. The roving yarn tension detector 35 has, for example, a configuration similar to that disclosed in Japanese Examined Patent Publication No. 63-47811, and includes a light emitting portion and a light receiving portion arranged to face each other, and includes a light emitting portion and a light receiving portion. The roving yarn R is positioned between them. The light emitting part is constituted by an infrared light emitting diode array, and the light receiving part has a light receiving element array composed of a large number of light receiving elements arranged in parallel in the vertical direction at a pitch about half the diameter of the roving yarn R. Each light receiving element outputs an electrical signal corresponding to the intensity of light received. That is, when the roving yarn R blocks the light from the light emitting portion, the light receiving element corresponding to the position of the roving yarn R does not receive light, so the position of the roving yarn R is obtained by detecting the light receiving element, The roving yarn tension is required.

  Next, an electrical configuration for driving and controlling the drive system will be described with reference to FIG. The microcomputer 38 constituting the control device 37 includes a CPU (central processing unit) 39, a program memory 40 including a read only memory (ROM) storing a control program, and a work memory 42. The work memory 42 includes a readable and rewritable memory (RAM) that temporarily stores input data input by the input device 41 and an arithmetic processing result in the CPU 39. The CPU 39 operates based on program data stored in the program memory 40.

  An input device 41 for inputting spinning conditions such as spinning roving weight, fiber type, flyer rotation speed, twisting number, bobbin diameter, and shoulder angle of the roving winding is integrated into the control device 37 as a keyboard. Output signals from the rotary encoder 34 and the roving yarn tension detector 35 are input to the CPU 39 via the input interface 43.

  The excitation and demagnetization of the electromagnetic clutches 32 and 33 is controlled via the electromagnetic clutch excitation / demagnetization circuit 44 based on a signal from the CPU 39, and the bobbin rail 16 is switched up and down. The bobbin rail 16 is moved upward when the first electromagnetic clutch 32 is excited, and is moved downward when the second electromagnetic clutch 33 is excited. Both electromagnetic clutches 32 and 33 are not excited simultaneously.

  The CPU 39 drives and controls the main motor M, the winding motor 21 and the elevating motor 26 via the output interface 45, motor drive circuits 46a, 46b, 46c and inverters 20a, 20b, 20c.

  The rotary encoder 34 outputs different pulse signals corresponding to the forward rotation and reverse rotation of the rotary shaft 25, so that the distinction between the rise and fall of the bobbin rail 16 can be confirmed from the pulse signal. Yes. The CPU 39 operates based on program data stored in the program memory 40. The CPU 39 calculates the lifting speed and position of the bobbin rail 16 based on the output signal from the rotary encoder 34.

  The CPU 39 estimates the current coarse yarn winding diameter Φp based on the coarse yarn tension signal obtained from the coarse yarn tension detector 35, and estimates the coarse yarn winding diameter increase amount ΔΦp at the time of layer change from the current coarse yarn winding diameter Φp. The CPU 39 uses a model that assumes that the amount of increase in the roving yarn diameter ΔΦ monotonously increases with the increase in the roving yarn winding layer, and the amount of increase in the roving yarn diameter estimated based on the roving yarn tension obtained from the roving yarn tension detector 35. Update the model so that the difference between ΔΦp and the coarse yarn winding diameter increase amount ΔΦq calculated in the previous model is updated, and calculate the coarse yarn winding diameter increase amount ΔΦ relative to the current coarse yarn winding diameter Φp from the updated model. To do.

  Next, the operation of the apparatus configured as described above will be described. Prior to the operation of the machine base, spinning conditions such as spinning roving weight, fiber type, flyer rotation number, twisting number, and roving shoulder angle are input by the input device 41.

  The CPU 39 determines a shift curve for setting the rotational speed of the flyer 12 from the start of winding to the full pipe based on the spinning conditions. The shift curve has a predetermined winding diameter from the start of winding in order to improve productivity and to prevent the winding of the roving from being hindered by the centrifugal force accompanying the rotation of the flyer 12 when the roving is grown. For example, the rotation speed is maintained at a constant high speed until reaching 80 to 85 mm, and thereafter gradually decelerated as the roving thread diameter increases. The CPU 39 determines the rotational speed of the main motor M in accordance with the current roving winding diameter, and determines the spinning speed based on the set number of twists. Further, the CPU 39 determines the rotation speed of the winding motor 21 from the spinning speed and the current roving diameter, and further determines the rotation speed of the lifting motor 26 from the number of layers.

  When the operation of the machine base is started, the main motor M is driven, and the driving of the main motor M causes the front roller 11 and the flyer 12 to rotate. The winding motor 21 and the elevating motor 26 are driven simultaneously with the start of the machine base, and the rotational force of the main motor M input to the differential gear mechanism 22 and the rotational force of the winding motor 21 are the differential gears. The rotary shaft 19 is driven by the combined rotational force of the mechanism 22 and the bobbin wheel 17 is rotationally driven. As a result, the roving yarn R spun from the front roller 11 is twisted by the flyer 12 and wound in layers on the bobbin B that rotates at a higher speed than the flyer 12. Further, the bobbin rail 16 is moved up and down together with the lifter rack 23 through the switching mechanism 28, the rotating shaft 25, and the like by driving the lifting motor 26. An output signal from the rotary encoder 34 is input to the CPU 39.

  Further, a signal indicating the tension state of the roving yarn R is input to the CPU 39 by the roving yarn tension detector 35, and the CPU 39 determines from the signal whether the tension of the roving yarn R is in an appropriate state and deviates from the appropriate tension. In this case, the winding motor 21 is controlled so that the roving yarn tension returns to an appropriate value by adjusting the bobbin rotational speed.

  The winding speed and the raising / lowering speed of the bobbin rail 16 are gradually reduced as the roving yarn layer increases by changing the rotational speeds of the winding motor 21 and the raising / lowering motor 26. The CPU 39 always calculates the position of the bobbin rail 16 based on the output signal from the rotary encoder 34.

  Next, the operation of the control device 37 for realizing the set shoulder angle θ will be described. As shown in FIG. 4, the shoulder angle θ of the roving bobbin is expressed by the following equation, where the amount of increase in roving bobbin diameter per layer is ΔΦ and the amount of decrease in lift per layer is ΔYL.

θ = tan ⁻¹ (ΔΦ / 2ΔYL) (1)
The amount of increase in the roving yarn diameter ΔΦ is not constant and increases monotonically with the increase in the roving yarn winding diameter Φ. Therefore, in order to form the roving bobbin F with the set shoulder angle θ, the roving bobbin diameter increase amount ΔΦ is calculated every time the roving bobbin layer increases, and the equation (1) is established correspondingly. It is necessary to determine the lift reduction amount ΔYL and to reverse the bobbin rail 16 so that the lift reduction amount ΔYL is obtained.

  The CPU 39 determines the coarse yarn winding diameter Φ, the coarse yarn winding diameter increase amount ΔΦ, and the reversal target position according to the flowchart of FIG. 3 and controls the reversal of the bobbin rail 16. As shown in FIG. 4, the position of the bobbin rail 16 is a position coordinate set so as to increase in the ascending direction of the bobbin rail 16 with the lower end of the position where the bobbin B can be wound up as the origin (0). Expressed as a value. In this embodiment, the winding of the roving yarn is started from an intermediate position of the bobbin B, and the roving yarn is wound so that the winding position gradually moves toward the lower side of the bobbin B. After the moving direction of the bobbin rail 16 is reversed at the position, the roving yarn is wound so that the winding position gradually moves toward the upper side of the bobbin B. That is, the bobbin rail 16 is moved up and down, and then lifted and lowered so as to reverse and move downward.

  Therefore, the bobbin rail 16 is reversed when the roving yarn layer wound around the bobbin B changes from the odd-numbered winding state to the even-numbered winding state in both the upper and lower turning positions. The lower side reversal position means a position where the bobbin rail 16 is reversed so that the winding of the roving yarn changes from the state of moving toward the lower side to the state of moving toward the upper side. Changes the moving direction from a state of moving up to a state of moving down. On the contrary, the upper reversal position means a position where the moving direction is changed from a state where the bobbin rail 16 is moving downward to a state where the bobbin rail 16 is moving upward. The initial lower reversal target position of the bobbin rail 16 is determined by the lift length LL input by the input device 41, and the first upper reversal target position is determined by the margin Lm input by the input device 41.

  When the operation of the roving machine is started, the CPU 39 determines in step S1 whether or not the first layer is wound (winding). The diameter is set to the bobbin bare diameter, and the reversal target position is determined. The reversal target position is determined from the output signal of the rotary encoder 34 whether the bobbin rail 16 is ascending or descending. If the bobbin rail 16 is ascending, the reversing target position is determined from the lift length LL. A reverse target position is determined. Even if the electromagnetic clutches 32 and 33 are switched at the reverse target position, the moving direction of the bobbin rail 16 is not changed immediately, but the moving direction is changed after overrun. Therefore, the reverse target position is determined in consideration of the overrun amount of the bobbin rail 16.

  Next, the CPU 39 proceeds to step S <b> 3 and determines whether or not the reverse target position has been reached based on the output signal of the rotary encoder 34. When the CPU 39 reaches the reversal target position, it outputs an excitation / demagnetization switching signal to the electromagnetic clutches 32 and 33. In addition, the operation information for each layer is updated. Specifically, the current flyer rotation speed Sou [n] Vf and the current coarse yarn winding diameter Sou [n] Φ are stored in the work memory 42 as operation information for each layer. The excitation / demagnetization state of the electromagnetic clutches 32 and 33 is switched by the excitation / demagnetization switching signal, and the moving direction of the bobbin rail 16 is reversed. The number of output pulses from the rotary encoder 34 from when the excitation / demagnetization switching signal is output until the bobbin rail 16 is reversed is counted. The CPU 39 obtains the overrun amount from the count number and stores it in the work memory 42.

  Next, the CPU 39 proceeds to step S5 and processes the filter coefficient update data. The filter coefficient means a coefficient of a ΔΦ estimation model used for estimating the coarse yarn winding diameter increase amount ΔΦ. Specifically, the estimated model is expressed by the following equation as a function of the current coarse yarn winding diameter and the flyer rotation speed Vf.

ΔΦ = coefficient A × current coarse thread winding diameter + coefficient B
This model assumes the following two.
1. When the flyer rotation speed Vf is constant, the roving yarn diameter increase amount ΔΦ is a linear expression (monotonically increasing) of the roving yarn winding diameter.

2. The coefficient A has a relationship of coefficient A = (initial speed of the flyer rotational speed) / (current flyer rotational speed) when the flyer rotational speed Vf is shifting, that is, after the shift point.
The CPU 39 uses the coarse yarn winding diameter data Cln [n] Φ, the flyer rotation speed data Cln [n] Vf, and the coarse yarn winding diameter increase amount data Cln [n] ΔΦ for updating the filter coefficient for each previous four layers. Refer to the information and calculate using the following formula. However, when it is less than 4 layers, it is calculated using the layer information that can be calculated. The layer information that can be calculated is data corresponding to each layer of the coarse yarn winding diameter and the coarse yarn winding diameter increase amount stored in advance in the program memory 40 or a nonvolatile memory (not shown) as a database.

Cln [n] Φ = (Sou [n] Φ + Sou [n−4] Φ) / 2
Cln [n] Vf = (Sou [n] Vf + Sou [n-4] Vf) / 2
Cln [n] ΔΦ = (Sou [n] ΔΦ + Sou [n−4] ΔΦ) / 4
For example, when the roving layer is the first layer, Cln [1] Φ is the one-layer roving diameter obtained from the spinning conditions, Cln [1] Vf is the flyer rotation speed immediately before reversal, [1] ΔΦ is set to an amount of increase ΔΦ of one layer of roving yarn diameter obtained from spinning conditions. Similarly, until n is up to 3, that is, until the winding is reversed from the third layer to the fourth layer, Cln [n] Φ is set to the coarse yarn winding diameter of the n layer obtained from the spinning conditions, and Cln [n] Vf Is set to the flyer rotation speed immediately before the reversal, and Cln [n] ΔΦ is set to a one-layer coarse yarn diameter increase amount ΔΦ obtained from the spinning conditions.

First, the ΔΦ estimation model is set to y = Ax + B, and the initial value A0 is obtained by the method of least squares based on the data obtained at the first predetermined number of times of up / down switching.
A0 = Σ deviation xy / Σ deviation ^{x 2}
Σ deviation x ² = (inX−ave X0) ²
Σ deviation xy = (inX−ave X0) * (inY−ave Y0)
The initial value P0 of the variance is the reciprocal of the variance of inX.

P0 = 1 / Σ deviation x ²
However, inX is the filter coefficient update data Cln [n] Φ, inY is the filter coefficient update data Cln [n] ΔΦ, ave X0 is the reference average value of the coarse yarn winding diameter, and ave Y0 is the increase amount of the coarse yarn winding diameter. Reference average value. ave X0 and ave Y0 are calculated from the data after the sixth layer of the filter coefficient update data Cln. Specifically, ave X0 is an average value of Cln [n] Φ from the sixth layer Cln [6] Φ to the nth layer, and ave Y0 is nth layer from the sixth layer Cln [6] ΔΦ. Is the average value of Cln [n] ΔΦ.

Next, the CPU 39 proceeds to step S6 and performs filter coefficient update (ΔΦ estimation model update).
The CPU 39 uses an online method of parameter estimation to estimate the amount of increase in the diameter of the wound yarn ΔΦp estimated based on the amount of the roast yarn tension signal obtained from the roving yarn tension detector 35, and the amount of increase in the amount of roast yarn winding diameter calculated by the model before the update. The filter coefficient is updated so that the difference from ΔΦq becomes small. That is, the ΔΦ estimation model is linearized as inY = A * inX, and the coarse yarn winding diameter increase obtained based on the calculation results of inX = current coarse yarn winding diameter−ave X0 and inY = current coarse yarn winding diameter increase amount−ave Y0 The coefficient A (inclination) is updated using the amount ΔΦp and the coarse yarn winding diameter increase amount ΔΦq calculated based on the coefficient A (inclination) of the pre-update model.

Next, based on the updated coefficient A, the CPU 39 estimates the coarse yarn winding diameter increase amount ΔΦ using the following equation in step S7, that is, calculates the ΔΦ predicted value.
ΔΦ predicted value = A * inX * VfRate + ave Y0
However, VfRate is the initial speed Vf ₀ of the flyer rotational speed / current rotational speed Vf _now.
Next, the CPU 39 calculates and determines the next reversal target position in step S8.

  The reversal target position is a value obtained by subtracting ΔYL from the previous reversal position when the reversal target position is the lower reversal position, that is, when the moving direction of the bobbin rail 16 is changed from ascending to descending, and the reversing target position is reversed to the upper side. If the movement direction is changed from the position, that is, the bobbin rail 16 from descending to ascending, the value is obtained by adding ΔYL to the previous inversion position. Since ΔYL = ΔΦ / (2 tan θ), ΔYL can be calculated from the result of step S7, and the previous reverse position is stored in the work memory 42. Therefore, the CPU 39 determines the reverse target position from the previous reverse position and ΔYL. Determine by calculating.

  Next, the CPU 39 updates the roving yarn winding diameter Φ in step S9. Thereafter, the CPU 39 controls the speeds of the winding motor 21 and the elevating motor 26 based on the updated roving winding diameter. And it returns to step S1 again. In the case of winding up after the second layer, the process proceeds from step S1 to step S3, and each step after step S3 is performed.

  Therefore, unlike the conventional case, every time the coarse yarn winding layer increases, the coarse yarn winding diameter increase amount ΔΦ is estimated by a more accurate model updated using the online method of parameter estimation, and based on the coarse yarn winding diameter increase amount ΔΦ. Thus, since the reversal target position is determined, it is possible to wind up the roving yarn so as to obtain a preset shoulder shape without performing trial spinning corresponding to each spinning condition.

  In the case where the roving yarn is wound a plurality of times under the same spinning conditions, in the initial winding (first time) and the second and subsequent windings, in the initial stage of winding, for example, in step S7 at the time of winding up to the sixth layer The estimation method of ΔΦ is different. Specifically, in the case of the first winding, ΔΦ is not estimated from the estimation model, but the filter coefficient updating processed data processed in step S5 is directly used. On the other hand, in the second and subsequent windings, since the initial value of the filter coefficient is fixed in the previous winding, ΔΦ is estimated from the first layer using the estimation model.

According to this embodiment, the following effects can be obtained.
(1) A rough yarn winding diameter increase amount ΔΦ at the time of layer change required when winding the roving yarn R so as to have a preset shoulder shape is estimated from information of the roving yarn tension detector 35, and this estimated roughing The reversing position of the bobbin rail 16 is determined by using the amount of increase φΔ of the bobbin diameter to form the roving bobbin (coarse yarn package) F. Then, a model is prepared on the assumption that the amount of increase in the roving yarn diameter ΔΦ increases monotonically with the increase in the roving yarn diameter, and the amount of increase in the roving yarn diameter estimated based on the roving yarn tension signal obtained from the roving yarn tension detector 35. The model is updated so that the difference between ΔΦp and the coarse yarn winding diameter increase amount ΔΦq calculated in the model before update is small, and the coarse yarn winding diameter increase amount ΔΦ is calculated from the updated model. As a result, the value of the coarse yarn winding diameter increase amount ΔΦ obtained from (1) approaches the optimum value, and the model parameter is suppressed from abruptly changing due to fluctuations in the tension signal. Therefore, it is possible to stably form the planned shoulder-shaped roving winding F without causing shoulder collapse. As a result, it is possible to easily derive an indication of the production length that can be produced from the winding volume due to the shoulder shape.

  (2) When the flyer rotation speed is constant, the monotonous increase is assumed to be a primary increase, and the coarse yarn winding diameter increase amount ΔΦ is estimated. Although the amount of increase in the roving yarn diameter ΔΦ accompanying the increase in the roving yarn diameter is not exactly a primary increase, there is no problem even if it is assumed to be a primary increase, and the calculation is simplified by assuming a primary increase.

  (3) The roving machine is speed-controlled so that the rotational speed of the flyer 12 is reduced as the roving yarn diameter increases after the roving yarn diameter reaches a predetermined value. It is assumed that the diameter increase amount ΔΦ is proportional to a value obtained by dividing the initial speed of the flyer rotational speed by the current rotational speed. Therefore, even when the rotational speed of the flyer 12 is controlled to be decelerated as the roving yarn diameter increases after the roving yarn diameter reaches a predetermined value, the roving yarn winding F can be stably formed. Can do.

  (4) The CPU 39 controls the speeds of the winding motor 21 and the lifting motor 26 based on the roving yarn diameter updated in step S9. Therefore, compared with the case where the speeds of the winding motor 21 and the elevating motor 26 are controlled based on the shift curve of the flyer 12 set based on the spinning conditions, the winding control is stabilized (especially the winding control). Early).

The embodiment is not limited to the above, and may be embodied as follows, for example.
○ The shoulder shape that is set in advance is not limited to the constant shoulder angle θ, and the shoulder angle θ is a shape that gradually decreases with an increase in the diameter of the bobbin winding, that is, a curved surface having an outwardly convex shoulder shape. Or a shape in which the shoulder angle θ changes in a plurality of stages. Even in these shapes, when determining each reversal target position, ΔYL is obtained from the corresponding shoulder angles θ and ΔΦ in the layer, and the current reversal target position is calculated from the ΔYL and the previous reversal target position. can do.

  ○ The shoulder shape may be a shape in which the upper shoulder and the lower shoulder are different. For example, the shoulder angle θ may be a different value between the upper shoulder and the lower shoulder, or the curvature may be a different value between the upper shoulder and the lower shoulder when the shoulder shape is a curved shape.

  When determining the reversal target position of the bobbin rail 16 in step S8, the current reversal target position is calculated from the next ΔYL obtained from ΔΦ and shoulder angle θ estimated in step 7 and the previous reversal target position. It is not limited to the method. For example, a relational expression between the reversal target position and the coarse yarn winding diameter Φ may be obtained corresponding to a preset shoulder shape, and the reversal target position may be determined using the expression. As a relational expression between the reversal target position and the roving yarn winding diameter Φ, for example, the following expression is given.

Reversal target position = 0.5 * K1 (coarse yarn winding diameter−bobbin bare diameter) ² + K2 (coarse yarn winding diameter—bobbin bare diameter) + K3
However, K1: Inversion position change curve slope gradient, K2: Inversion position change curve slope initial value K3: Overrun correction offset amount ○ The inversion target position is determined using a relational expression that is a function of the coarse thread winding diameter Φ. In this case, in the second and subsequent windings, the reverse target position may be determined by estimation using the ΔΦ estimation model from the beginning of winding using the optimized parameter and reference position. In this case, it is not necessary to determine the initial value of the winding start by reusing the actual data.

  O The starting position for winding the roving may not start from the middle position of the bobbin B but may start from the upper end of the bobbin B where the roving can be wound. In this case, the odd-numbered layers of the roving layer are wound so as to go from the upper side to the lower side of the bobbin B, and the even-numbered layers are wound so as to go from the lower side of the bobbin B to the upper side.

○ The relational expression of the inversion target position may be divided into an even layer and an odd layer.
○ In the first winding, the initial ΔΦ may be set to any appropriate value instead of the database. In this case, it is not necessary to maintain a database for estimating ΔΦ. That is, if an initial value of the amount of increase in the diameter of the roving yarn winding ΔΦ can be given, the roving yarn winding can be performed only by the measurement information of the roving yarn tension detector 35 and the flyer rotation speed without setting the fiber type, the presser winding number, and the roving weight. The diameter increase amount ΔΦ can be estimated, and it is not necessary to have a database related to spinning conditions such as fiber type and roving weight.

  The inversion target position of the bobbin rail 16 may be determined by predictive feedback using the bobbin rail moving speed as an element for the overrun amount when the bobbin rail is reversed. This is effective when the change in the moving speed of the bobbin rail 16 is large at the start and end of winding and the difference in the overrun amount affects the bobbin shape.

○ After completion of the first winding, the parameters may be optimized using appropriate layer data.
○ For each spinning condition, a parameter group of the ΔΦ estimation model may be stored and used from the beginning of the next winding. In this case, it is not necessary to determine the initial value of the winding start by reusing the actual data.

  ○ When determining the reversal target position in step S8, as the overrun amount value, the overrun amount at the time of the reversal of the bobbin rail 16 two times before, that is, the latest actual value at the time of reversal of the bobbin rail 16 in the same movement direction. The overrun amount may be used, or the overrun amount at the previous inversion of the bobbin rail 16 may be used.

When determining the reversal target position in step S8, an average value of the overrun amounts of the past multiple times (for example, several times) may be used as the value of the overrun amount.
○ When the flyer rotation speed Vf is constant, the roving yarn diameter increase amount ΔΦ may be a relationship that increases monotonically with respect to the roving yarn winding diameter instead of the primary expression of the roving yarn winding diameter. Good.

The following technical idea (invention) can be understood from the embodiment.
(1) In the invention according to any one of claims 1 to 3, a ΔΦ estimation model for predicting the coarse yarn winding diameter increase amount ΔΦ is defined as ΔΦ = coefficient A × current coarse thread winding diameter + coefficient B, In the second winding, the initial ΔΦ is set to any appropriate value without using a database.

  (2) In the invention according to any one of claims 1 to 3, a parameter group of the ΔΦ estimation model is stored for each spinning condition, and the parameter is set at the next winding of the same spinning condition. Use the group from the beginning.

  (3) In the invention according to any one of claims 1 to 3 and the technical idea (1), after the first winding, parameters are optimized using data of each layer. The parameter is used for the second and subsequent windings.

Schematic of the driving system of the roving machine. The block diagram which shows the electric constitution of a control apparatus. The flowchart which shows the setting procedure of a reversal command output position. The partial fracture | rupture schematic diagram which shows the shoulder angle of a roving thread winding.

Explanation of symbols

  F: Coarse yarn winding, R ... Coarse yarn, 12 ... Flyer, 16 ... Bobbin rail, 35 ... Coarse yarn tension detector.

Claims (3)

Estimate the coarse yarn winding diameter increase ΔΦ at the time of layer change required when winding the coarse yarn so as to have a preset shoulder shape from the coarse yarn tension information of the coarse yarn tension detector, and estimate this coarse yarn winding diameter. A roving winding method in a roving machine for determining a reversing position of a bobbin rail using an increase amount ΔΦ to form a roving winding,
Create a model that assumes that the estimated coarse yarn winding diameter increase amount ΔΦ monotonously increases as the coarse yarn winding diameter increases, and for each layer change, the coarse yarn winding diameter increase amount ΔΦp estimated from the coarse yarn tension information and the pre-update The model is updated so that the difference from the coarse yarn winding diameter increase amount ΔΦq calculated using the model is reduced, and the estimated coarse yarn winding diameter increase amount ΔΦ is calculated from the updated model. The roving winding method in the roving machine.   The roving winding method for a roving machine according to claim 1, wherein the monotonic increase is a primary increase.   The roving machine is speed-controlled so that the rotational speed of the flyer is reduced as the roving yarn diameter increases after the roving yarn diameter reaches a predetermined value. The method for winding a roving yarn in a roving machine according to claim 1 or 2, wherein is assumed to be proportional to a value obtained by dividing an initial speed of a flyer rotation speed by a current rotation speed.

Images