CN101654838A - Electromagnetically needle selecting method of computerized flat knitting machine based on Q15 fixed-point algorithm - Google Patents

Abstract

The invention belongs to the technical field of automatic knitting machines, in particular to an electromagnetically needle selecting method of a computerized flat knitting machine based on a Q15 fixed-point algorithm. The method is characterized by comprising the following steps: (1) reading needle selecting information; (2) storing information of needle pressing states of blades on a needle selector; (3) reading selected needles in a nose direction; (4) utilizing a coder to feed back pulse signals and adopting the Q15 fixed-point algorithm to carry out real-time data processing so as to determine positions of the blades of the needle selector; (5) calculating state values of the blades of the needle selector, i.e. determining the state values of the positions of the blades of the needleselector by judging a state value of the machine head direction, i.e. controlling 0.1 values of the blades; (6) comparing the states of the blades; (7) driving the blades with different state values to act; and (8) judging whether a needle selecting process is ended or not. On one hand, the needle selecting method improves the speed and the precision of needle selection, and on the other hand, solves the problem of reliability of the needle selection, thereby not only saving manufacturing cost, but also effectively overcoming technical contradiction of the speed and the reliability of the needle selection.

Description

Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm

Technical field

The invention belongs to technical field of automatic knitting machines, particularly relate to a kind of Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm.

Background technology

The control of Computerized flat knitting machine selecting mechanism is the key that can whole straight-bar machines operate as normal, and it can select each the root pin on the straight-bar machines, thereby has many advantages.In the electromagnetic needle selection system, electromagnetic needle selection is at all needle selection information computings later final step that finishes, and therefore reads the accuracy of pin and the real-time of needle selection information processing, will influence the accuracy and the reliability of needle selection control.

External colleague adopts the tooth bar copperhead to read the skill of handling needles basically, tooth bar promptly is installed on needle-bar, the corresponding one piece of pin of one tooth, produce sine wave signal and numeration signal that one group of two-way phase difference is 90 degree by detection to the Computerized flat knitting machine upper rack, by to this Signal Processing, thereby realize control to the straight-bar machines eedle.The method is comparatively reliable, also simple in theory, can improve loom speed, avoid the long-term transmission of belt to cause error,, cost an arm and a leg mainly by fewer companies productions such as Japan, Germany but read pin device (copperhead, read the pin copperhead), can increase the cost of tooth bar and copperhead, and its pairing tooth bar of the straight-bar machines of differing needles distance is also different, brings a lot of inconvenience also for production and management.

Summary of the invention

The object of the present invention is to provide that a kind of needle selection precision is accurate, speed is fast, the brain straight-bar machines electromagnetic needle selection method based on the Q15 fixed-point algorithm of good reliability.

The present invention seeks to adopt such technical scheme to realize: to it is characterized in that said method comprising the steps of:

1. read needle selection information

2. preserve the information of each blade pressing state on the needle selection device;

3. read heading by needle selection,

4. utilize the encoder feedback pulse signal, adopt the Q15 fixed-point algorithm to carry out data in real time and handle, determine needle selection device blade position;

5. calculate needle selection device blade state value; By judging the heading state value, determine the state value of each blade position of needle selection device, promptly control 0.1 value of this group blade;

6. compare the blade state;

7. the blade action that the driving condition value is different;

8. judge whether needle selection finishes;

The present invention utilizes the built-in encoder of servomotor to read pin, adopt the software resolution policy of advanced Q15 fixed-point algorithm, the optimization that replaces floating number by integer is calculated to realize the fast processing of data, promote the speed and the precision of needle selection on the one hand, solve the integrity problem of needle selection on the other hand, not only saved manufacturing cost but also effectively overcome this technical contradiction of the speed and the reliability of needle selection.

Description of drawings

Fig. 1 is a schematic flow sheet of the present invention

The specific embodiment

With reference to Fig. 1: the method for the invention may further comprise the steps:

1. read needle selection information; At first utilize the built-in encoder of servomotor to read pin, DSP obtains through calculating the needle selection information of intercepting, this information has been 0,1 two-value that directly can be used for controlling blade, with the information of " 1 " expression needle selection device blade pressing, with the information of " 0 " expression needle selection device blade mis-pressing;

2. preserve the information of each blade pressing state on the needle selection device; The information of each blade pressing state on the needle selection device is kept in one 16 the array;

3. read heading by needle selection, according to the arrangement mode and the head traffic direction needle selection of six sections needle selection stitch on the needle-bar; Because six pieces is one group knitting needle oblique arrangement, therefore, Nose Left motion and right motion mutually, the needle selection blade on the corresponding needle selection device is different; The arrangement mode of six sections needle selection stitch is always arranged to upper right corner from the lower left corner on the needle-bar, so when Airplane Nose Right was capable, always uppermost that needle selection electromagnet ran into the pin that will select earlier in every group of needle selection electromagnet, and choosing more successively thereafter; When Nose Left moved, always nethermost that electromagnet ran into the pin that will select earlier in every group of needle selection electromagnet, choosing more thereafter;

4. utilize the encoder feedback pulse signal, adopt the Q15 fixed-point algorithm to carry out data in real time and handle, determine needle selection device blade position;

5. calculate needle selection device blade state value; By judging the heading state value, determine the state value of each blade position of needle selection device, promptly control 0.1 value of this group blade;

6. compare the blade state; By comparing of each blade pressing state on 5. calculating needle selection device blade state value and 2. preserving the needle selection device, draw 0,1 status switch value of control blade movement, and then 7. drive different blade actions set by step;

7. the blade action that the driving condition value is different; According to the duty of the knitting program control blade of working out, the blade action that the driving condition value is different;

8. judge whether needle selection finishes; Current pin position braiding is finished the back and is judged the head position whether beyond stroke, if not, then make head needle selection device counting add one, wait for that position counting interrupts producing next time, carry out next pin position braiding:, if then represent this row braiding end, this row needle selection end.

Q15 fixed-point algorithm of the present invention carries out the data in real time processing and replaces floating number optimization calculating to realize quick computing by integer: for fixed-point number, number range and precision are a pair of contradiction, a variable is wanted to represent bigger number range, must be cost to sacrifice precision; And to improve precision, then numerical representation scope just correspondingly reduces; The present invention by the test and the analysis of reality, adopts the Q15 fixed-point algorithm under the condition that guarantees needle selection speed, and the number range of Q15 is-1 to 0.9999695, and precision is 0.00003051, and the system control accuracy of making reaches optimum performance, guarantee the reliability of needle selection, its operational formula is:

Floating number (x) is converted to fixed-point number (x _q): x _q=(int) x*2 ^Q

Floating number (x) is converted to fixed-point number (x _q): x _q=(int) x*2 ^Q

Embodiment 1:

The present invention is directed to weft-knitted mode is that d pin, fabric width are that the homemade straight-bar machines control system of l inch adopts fixed-point algorithm realization needle selection to do following analysis:

So-called d pin is meant that in one inch distance of needle-bar d piece of pin arranged, i.e. needle gage d; The fabric width of straight-bar machines is meant the Breadth Maximum l of cross machine knitting.This project utilizes the built-in encoder of servomotor to read pin, and it is pulse that hypothesis driven servomotor mobile fabric width on needle plate need send umber of pulse; Eedle adds up to total_pins.

Eedle sum=machine length (inch) * needle gage can get thus,

Eedle sum: total_pins=l*d

The umber of pulse of every piece of corresponding pin is: pin_plse=pulse/total_pins

Suppose that the umber of pulse currency that a certain moment DSP reads by encoder is current_pulse, the needle selection position that then current needle selection device is determined is:

current_pin＝current_pulse/pin_pulse； (1)

In order to make the C compiler produce optimum code, improve arithmetic speed, the division of (1) should be converted into multiplying, make compiler produce the MAC instruction, and with the operand of MAC is that local variable is assigned in the register, carries out quick computing with the hardware multiplier resource that makes full use of DSP.Can get thus:

current_pin＝current_pulse*(l/pin_pulse) (2)

The value of l/pin_pulse is a floating number in the formula (2), therefore will be translated into fixed-point calculation, and its value size is relevant with the needle gage and the servomotor of straight-bar machines move a fabric width on needle plate required pulse number.

Formula (2) floating-point multiplication is converted into fixed-point multiplication and is analyzed as follows:

If the expression formula of floating-point multiplication is:

float?x，y，z，current_pin，current_pulse，l/pin_pulse；

z＝current_pin；

y＝l/pin_pulse；

x＝current_pulse；

z＝xy；

The scaled values of supposing x is Qx, and the scaled values of y is Qy, and the scaled values of product z is Qz, then

$z=\mathrm{xy}\⇒$

$z_q\·2^{-Q_z}=x_q\·y_q\·2^{-(Q_x+Q_y)}\⇒$

$z_q=\left(x_q{y}_q\right)2^{Q_z-(Q_x+Q_y)}---\left(3\right)$

So the multiplication of fixed-point representation is:

int?x，y，z；

long?temp；

temp＝(long)x；

Can get by formula (3),

z＝(temp×y)＞＞(Qx+Qy-Qz)； (4)

Because current_pin and current_pulse are integer, therefore, scaled values is Qx=Qz=0, determines that for the scaled values of floating-point variable l/pin_pulse method is as follows:

In fact the Qy value of determining variable is exactly to determine the dynamic range of variable, as long as dynamic range has determined that then the Q value has also just been determined.If the maximum of the absolute value of variable is | max|, wherein | max| must be less than or equal to 32767.Get an Integer n, it is satisfied

2 ^n-1＜|max|＜2 ⁿ

Then have

2 ^-Q＝2 ^-15×2 ⁿ＝2 ^-(15-n)

Q＝15-n

Therefore, having determined variable | max| just can determine its Q value, and variable | we determine the max| value by statistical analysis method: carry out statistical analysis by the needle gage of all kinds straight-bar machines and the umber of pulse of servomotor mobile fabric width on needle plate, we determine the dynamic range of variable l/pin_pulse, can get | max|＜1, so n=0, Q=15-n=15.

Embodiment 2:

Illustrate that the Q15 fixed-point algorithm is that 12 pins, fabric width are the concrete application on 52 inches straight-bar machines control systems in weft-knitted mode.

Eedle sum: total_pins=l*d=52*12=624;

It is pulse=26416 that driving servomotor mobile one-period on needle plate need send umber of pulse

The umber of pulse of every piece of corresponding pin is: pin_pulse=pulse/total_pins=26416/624=42.33;

The pulsed quantity that moves the corresponding encoder in arbitrary position for head is current_pulse,

Suppose current_pulse=1000,

current_pin＝current_pulse*(l/pin_pulse)＝1000*(l/pin_pulse)

Make l/pin_pulse=pp;

pp＝l/pin_pulse＝1/42.33＝0.0236

According to above analysis, in conjunction with (4) formula Qx=0 as can be known, Qy=15, Qz=0, therefore, and x=1000, y=0.0236*32767=773 promptly handles by Q15 the y floating point values.

temp＝1000L

z＝(1000L*773)＞＞(0+15-0)＝24

Promptly when current_pulse=1000 that encoder reads, corresponding pin position is 24.

The floating-point arithmetic library that the present invention utilizes the C compiler to carry compares the result of floating-point arithmetic and fixed-point algorithm, finds through test, and the Q15 fixed-point algorithm realizes that a needle selection calculating process only needs 6.67ns, and will reach millimeter grade with time of floating-point operation.

In order to improve arithmetic speed, guarantee validity and the real-time calculated, this algorithm is further optimized: at first be that to be configured to latent period be in zero the internal memory for the intermediate variable that will often use; Next is to adopt FLASH speed technology (machine-processed FLASH pipeline mode is realized referring in advance in the ENPIPE position that enables the FOPT register), can reach the disposal ability of 100～120MIPS like this, be much higher than the reading capability of itself giving tacit consent to, effectively raise the needle selection speed issue in this way.

Claims (6)

1, based on the Computerized flat knitting machine electromagnetic needle selection method of Q15 fixed-point algorithm, it is characterized in that said method comprising the steps of: 1. read needle selection information; 2. preserve the information of each blade pressing state on the needle selection device; 3. read heading by needle selection, 4. utilize the encoder feedback pulse signal, adopt the Q15 fixed-point algorithm to carry out data in real time and handle, determine needle selection device blade position; 5. calculate needle selection device blade state value; By judging the heading state value, determine the state value of each blade position of needle selection device, promptly control 0.1 value of this group blade; 6. compare the blade state; 7. the blade action that the driving condition value is different; 8. judge whether needle selection finishes.

2, the Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm according to claim 1, it is characterized in that the described needle selection information that reads is to utilize the built-in encoder of servomotor to read pin, DSP obtains through calculating the needle selection information of intercepting, this information has been 0,1 two-value that directly can be used for controlling blade, with the information of " 1 " expression needle selection device blade pressing, with the information of " 0 " expression needle selection device blade mis-pressing.

3, the Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm according to claim 1, the information that it is characterized in that each blade pressing state on the described preservation needle selection device is that information with each blade pressing state on the needle selection device is kept in one 16 the array.

4, the Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm according to claim 1 is characterized in that the described encoder feedback pulse signal that utilizes is to adopt the Q15 fixed-point algorithm to carry out data in real time to handle, and determines needle selection device blade position.

5, the Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm according to claim 1, it is characterized in that described relatively blade state step is 5. to calculate needle selection device blade state value and step by step 2. to preserve comparing of each blade pressing state on the needle selection device, draw 0,1 status switch value of control blade movement.

6, the Computerized flat knitting machine electromagnetic needle selection method based on the Q15 fixed-point algorithm according to claim 6, it is characterized in that described Q15 fixed-point algorithm carries out the data in real time processing and replaces floating number optimization calculating to realize quick computing by integer: under the condition that guarantees needle selection speed, test and analysis by reality, the number range of Q15 is-1 to 0.9999695, precision is 0.00003051, the system control accuracy of making reaches optimum performance, guarantees the reliability of needle selection, and its operational formula is:

Floating number (x) is converted to fixed-point number (x _q): x _q=(int) x*2 ^Q

Floating number (x) is converted to fixed-point number (x _q): x _q=(int) x*2 ^Q

Images