CN115079553B - Time self-correction method for hot runner time schedule controller - Google Patents
Time self-correction method for hot runner time schedule controller Download PDFInfo
- Publication number
- CN115079553B CN115079553B CN202210747309.1A CN202210747309A CN115079553B CN 115079553 B CN115079553 B CN 115079553B CN 202210747309 A CN202210747309 A CN 202210747309A CN 115079553 B CN115079553 B CN 115079553B
- Authority
- CN
- China
- Prior art keywords
- time
- value
- timer
- data
- calibration coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G04—HOROLOGY
- G04G—ELECTRONIC TIME-PIECES
- G04G5/00—Setting, i.e. correcting or changing, the time-indication
Abstract
The invention provides a time schedule controller of a hot runnerThe self-correcting method comprises the following steps: step 1: the method comprises the steps that on the basis of a microcontroller included by a time sequence controller, the start time correction of a timer in the time sequence controller is controlled, and after the start time correction of the timer, the internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated are recorded; step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment; and step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And performing compensation correction on the time of the timer. The invention continuously corrects the crystal oscillator in the running process of the time sequence controller, and ensures that the time accuracy of the time sequence controller is controlled within an allowable range. Especially in long-term operation, the deviation can not be accumulated, and the normal operation of the time schedule controller is ensured.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a time self-correcting method for a hot runner time schedule controller.
Background
The time sequence controller is applied to time sequence control of a hot runner system of an injection mold, and can eliminate the welding line or change the position of the welding line of the visible surface of a product or the weak structure part of the product, eliminate the flash and fill deficiency, improve the injection molding speed of a sprue, reduce flow marks and the like by controlling the injection molding time sequence, thereby effectively improving the quality of injection molded parts.
The time schedule controller has a plurality of working modes, realizes the subsection time delay and the opening, realizes the opening and closing control of the sprue by setting the time of the time delay and the opening, and therefore, the time precision is very important for the time schedule controller. The controller core circuit adopts a crystal oscillator to provide a reference clock, and the accuracy and the stability of the crystal oscillator directly influence the time accuracy of the time schedule controller. In order to ensure the time accuracy, the frequency of the crystal oscillator generally needs to be calibrated through a standard clock source when the crystal oscillator leaves a factory. However, the time still deviates due to the temperature influence, the aging of the device and the like during the use process of the equipment. In order to eliminate time deviation and ensure long-term reliable operation of equipment, the invention provides a method for self-calibration of a time schedule controller clock.
Disclosure of Invention
The invention provides a time self-correcting method of a hot runner time schedule controller, which is used for solving the problem of time deviation caused by temperature influence, device aging and the like in the running process of the controller. The invention continuously corrects the crystal oscillator in the running process of the time sequence controller, and ensures that the time accuracy of the time sequence controller is controlled within an allowable range. Especially when the time schedule controller runs for a long time, the deviation can not be accumulated, and the normal running of the time schedule controller is ensured.
The invention provides a time self-correction method of a hot runner time schedule controller, which comprises the following steps:
step 1: the method comprises the steps that on the basis of a microcontroller included by a time sequence controller, the start time correction of a timer in the time sequence controller is controlled, and after the start time correction of the timer, the internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated are recorded;
and 2, step: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment;
and 3, step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And performing compensation correction on the time of the timer.
Preferably, before the timer starts time correction, the method further includes: based on the time schedule controller, the time for opening and closing the sprue of the hot runner system is set, and the method specifically comprises the following steps:
according to the nominal value f of the crystal oscillator frequency 0 Determining the oscillation period of the time sequence controller, and controlling a timer in the time sequence controller to generate 1ms interrupt;
obtaining a first millisecond count value T of the time schedule controller through interrupt processing ms ；
Counting the first millisecond value T ms Accumulating to obtain a first second count value T for controlling the opening and closing of the pouring gate by the time schedule controller s 。
Preferably, the step 1: the method comprises the following steps of controlling the correction of the starting time of a timer in a time sequence controller based on the microcontroller included by the time sequence controller, and recording internal time data of the microcontroller corresponding to the generation of each zero-crossing interrupt after the correction of the starting time of the timer, wherein the specific steps comprise:
step 1.1: detecting a zero crossing point of an input power frequency signal by using an alternating current detection module, generating a zero crossing interrupt signal, and sending the zero crossing interrupt signal to a microcontroller;
step 1.2: controlling a timer in the time schedule controller to start time correction;
step 1.3: recording corresponding internal time data when the microcontroller receives a zero-crossing interrupt signal every time, wherein the internal time data comprises a second counting value T si A second millisecond count value T msi And the current count value T of the timer usi 。
Preferably, the step 1.1: the method comprises the following steps of detecting the zero crossing point of an input power frequency signal by using an alternating current detection module, and generating a zero crossing interrupt signal, wherein the method comprises the following steps:
acquiring a signal image of the input power frequency signal, and marking a first position corresponding to a zero-crossing point of the input power frequency signal on the signal image;
acquiring a logic level diagram corresponding to the input power frequency signal based on the signal image, and marking a second position generated by interruption on the logic level diagram;
aligning a signal image with the logic level image to obtain an aligned image, and obtaining a position difference between the first position and the second position based on the aligned image;
judging whether an interference signal exists in the input power frequency signal or not based on the position difference, judging that no interference signal exists in the input power frequency signal when the position difference is always consistent, normally detecting a zero crossing point of the alternating current detection module, and generating a zero crossing interrupt signal based on the zero crossing point;
when the position difference is inconsistent, when an interference signal exists in the input power frequency signal, acquiring alignment position change data of the zero-crossing point and the interrupt signal based on the alignment image;
meanwhile, standard data of zero crossing point detection are obtained, and the change data of the standard data are compared to obtain a comparison result;
determining data fluctuation amplitude according to the comparison result, judging that zero crossing point detection of the alternating current detection module is abnormal when the data fluctuation amplitude is larger than or equal to a preset value, displaying the abnormal zero crossing point detection based on a display module, and stopping generating a zero crossing interrupt signal;
and when the data fluctuation amplitude is smaller than a preset value, judging that the zero crossing point detection of the alternating current detection module is normal, and generating a zero crossing interrupt signal based on the zero crossing point.
Preferably, the internal data is stored in a ring-shaped storage buffer, the internal data corresponds to the zero-crossing interrupt signal one to one, when a data sequence formed by the internal data is discontinuous, it is determined that the zero-crossing signal is lost in the current recording process, and the algorithm program is controlled to restart a new recording process.
Preferably, the step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment, and the specific steps comprise:
obtaining ith internal data of a data sequence, and calculating a time stamp t of an ith zero-crossing signal corresponding to the ith internal data i ：
t i ＝T si +T msi +T usi /TIM_Period
TIM _ Period represents the automatic reloading value of a timer register in the microcontroller; after all internal data in the data sequence are processed, a continuous time stamp t is formed 0 、t 1 、t 2 ……t n The interval Δ t between adjacent time stamps is a fixed value;
selecting a selected time interval on the time stamp sequence, and acquiring the starting time stamp t of the selected time interval 1 And an end timestamp t 2 Calculating a time difference DeltaT between the timer and the selected period 1-2 ：
ΔT 1-2 ＝t 2 -t 1
Calculating a calibration coefficient K of the timer under the current environment according to the time difference corresponding to the timer and the selected time interval and the time length of the selected time interval:
K＝ΔT 1-2 /ΔT
where Δ T represents the selected period time length.
Preferably, the step 2 further includes:
acquiring historical timestamp sequences corresponding to a plurality of historical data sequences, and comparing each timestamp in the historical timestamp sequences with adjacent timestamps thereof respectively to obtain a plurality of timestamp deviation values;
when all timestamp deviation values corresponding to the historical timestamp sequence are smaller than a set deviation value, taking the historical timestamp sequence as a first to-be-selected number sequence;
when a timestamp deviation value which is not smaller than a set deviation value exists in all timestamp deviation values corresponding to the historical timestamp sequence, marking a timestamp of which the timestamp deviation value is smaller than the set deviation value on the historical timestamp sequence to obtain a marking timestamp;
judging whether the adjacent marking time stamps exist in the marking time stamps, and if not, deleting the marks of the marking time stamps;
if yes, judging that the marking timestamps are on continuous marking lines, determining the head position of the continuous marking lines, deleting the timestamp marks on the continuous marking lines, selecting marks at the head position, and deleting the selected marks when the first time length between the selected marks is smaller than a preset threshold value;
taking the historical timestamp sequence with the selected mark as a second to-be-selected number sequence;
wherein each selected mark has one and only one corresponding selected mark;
establishing a training set based on the first to-be-selected number sequence and the second to-be-selected number sequence, training a deep learning neural network through the training set, and obtaining a selected time period identification model;
acquiring a time period to be selected of the current timestamp sequence according to the selected time period identification model;
acquiring data acquisition time of internal data corresponding to the time period to be selected, and screening out the latest time period to be selected and the next time period to be selected based on the data acquisition time;
respectively acquiring a second time length of the latest time period to be selected and a third time length of the next time period to be selected;
when the second time length is greater than or equal to a third time length, taking the latest time period to be selected as a final selected time period;
and when the second time length is smaller than the third time length, taking the time period to be selected newly as a final selected time period.
Preferably, the step 2 further includes:
filtering the calibration coefficient K of the timer under the current environment, and specifically comprising the following steps:
obtaining the optimal estimated value of the calibration coefficient at the last moment
And its covariance matrix K m-1 Optimal estimated covariance matrix of calibration coefficients at the current time
Speed of change of calibration coefficient v m ＝v m-1 + a Δ t, to obtain
Corresponding to
Wherein v is m Representing the speed of change of the calibration coefficient at the current time; v. of m-1 Representing the rate of change of the calibration coefficient at the previous time;
an optimal estimation value representing a calibration coefficient at the current time; f m A deviation matrix representing the optimal estimated value of the calibration coefficient at the previous moment and the optimal estimated value of the calibration coefficient at the current moment; a transpose matrix representing a deviation matrix of the optimal estimated value of the calibration coefficient at a time and the optimal estimated value of the calibration coefficient at the current time; a represents the acceleration of the change of the calibration coefficient; b is m A deviation matrix representing the speed of change of the calibration coefficient at the previous time and the speed of change of the calibration coefficient at the current time;
an estimation value of the calibration coefficient at the current time in an estimation value space;
when there is a covariance of Q m When the noise of (2) interferes with the signal, the following relationship is obtained:
correcting the optimal estimation value by using the observation data of the sensor, and mapping the estimation value space to an observation value space to obtain:
wherein H m A mapping matrix representing a mapping of an estimated value space to an observed value space;
representing an estimated valueA transpose of a mapping matrix spatially mapped to an observation space; r m Representing the covariance of the noise in the observation space;
representing an observed value;
optimal estimated value to be processed by filtering
As a final calibration factor i
Preferably, the step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And compensating and correcting the time of the timer, which comprises the following specific steps:
according to the nominal value f of the crystal oscillator frequency 0 And a calibration coefficient, calculating a frequency deviation value delta f between the nominal frequency and the actual frequency of the crystal oscillator:
Δf＝f 0 ×(1-K * )
and performing compensation correction on the timer according to the frequency deviation value delta f.
Preferably, when the frequency deviation value Δ f is greater than a preset deviation value, the timer is compensated in a stepwise compensation manner.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a hot runner timing controller timing self-calibration method according to the present invention;
FIG. 2 is a schematic diagram of a circuit layout of a hot runner timing controller according to the present invention;
FIG. 3 is a timing control flow diagram of an injection mold hot runner system;
FIG. 4 is a flowchart of step 1 of a method for time self-calibration of a hot runner timing controller according to the present invention;
FIG. 5 is a schematic diagram of probability distributions of estimated values and observed values.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it should be understood that they are presented herein only to illustrate and explain the present invention and not to limit the present invention.
Example 1:
the invention provides a time self-correcting method of a hot runner time schedule controller, as shown in figure 1, comprising the following steps:
step 1: controlling the starting time correction of a timer in the time sequence controller based on the microcontroller included by the time sequence controller, and recording internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated after the starting time correction of the timer is performed;
step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment;
and step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And performing compensation correction on the time of the timer.
In this embodiment, as shown in fig. 2, the timing controller includes: the device comprises a power module, an alternating current/direct current power supply detection circuit, a 32-bit microcontroller, a digital signal input detection module, an output control module, a display module and the like. The power supply module can convert externally input 220V alternating current into 24V/3.3V current voltage used by the controller; the digital signal input detection module is used for detecting an injection molding signal input signal and triggering the microcontroller to interrupt a signal; the output module is used for outputting through a relay to realize the opening/closing control of the pouring gate; the display module can display information such as delay/start time, working mode and the like of a plurality of channels of the time schedule controller; the microcontroller is used as a main control unit of the time schedule controller to realize all functions of input/output control, equipment parameter setting, clock calibration and the like.
In the embodiment, the zero crossing point detection precision can be controlled within 0.5mS, the window time of the sliding arithmetic mean value calculation is 10 minutes, and the crystal oscillator frequency precision can be ensured to reach 0.8PPM.
The beneficial effects of the above technical scheme are that: the invention is based on the microcontroller included in the time sequence controller, controls the start time correction of the timer in the time sequence controller, and records the internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated after the start time correction of the timer, thereby providing reliable data for calculating the calibration coefficient; processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment and provide data support for time self-correction of the time sequence controller; according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And compensating and correcting the timer to finish the time self-correction of the clock system of the time schedule controller. The invention continuously corrects the crystal oscillator in the running process of the time sequence controller, and ensures that the time accuracy of the time sequence controller is controlled within an allowable range. Especially in long-term operation, the deviation can not be accumulated, and the normal operation of the time schedule controller is ensured.
Example 2:
on the basis of embodiment 1, before the timer starts time correction, the method further includes: based on the time schedule controller, the time for opening and closing the sprue of the hot runner system is set, and the method specifically comprises the following steps:
step 0.1: according to the nominal value f of the crystal oscillator frequency 0 Determining the oscillation period of the time schedule controller, and controlling a timer in the time schedule controller to generate 1ms interruption;
step 0.2: obtaining the first of the timing controller by interrupt processingOne millisecond count value T ms ；
Step 0.3: counting the first millisecond value T ms Accumulating to obtain a first second count value T for controlling the opening and closing of the pouring gate by the time schedule controller s 。
In this embodiment, the first millisecond count value is a millisecond count value before the timing controller does not perform the time correction.
In this embodiment, the first second count value is a second count value before the timing controller does not perform time correction.
The beneficial effects of the above technical scheme are that: the invention depends on the nominal value f of the crystal oscillator frequency 0 Determining the oscillation period of the time sequence controller, controlling a timer in the time sequence controller to generate 1ms interrupt, and obtaining a first millisecond count value T of the time sequence controller through interrupt processing ms For the first millisecond count value T ms Accumulating to obtain a first second count value T for controlling the opening and closing of the gate by the time schedule controller s The control of opening and closing of the sprue is realized, and the quality of the injection molding part is effectively improved.
Example 3:
on the basis of example 1, the step 1: based on a microcontroller included in a timing controller, controlling the start time correction of a timer in the timing controller, and recording internal time data of the microcontroller corresponding to each time when zero-crossing interrupt is generated after the start time correction of the timer, as shown in fig. 3, the specific steps include:
step 1.1: detecting a zero crossing point of an input power frequency signal by using an alternating current detection module, generating a zero crossing interrupt signal, and sending the zero crossing interrupt signal to the microcontroller;
step 1.2: controlling a timer in the time schedule controller to start time correction;
step 1.3: recording corresponding internal time data when the microcontroller receives a zero-crossing interrupt signal every time, wherein the internal time data comprises a second counting value T si The second millisecond count value T msi And the current count value T of the timer usi 。
In this embodiment, the internal data refers to a second count value, a millisecond count value and a current count value of a timer corresponding to each zero-crossing interrupt signal of a microcontroller in the timing controller.
In this embodiment, the second millisecond count value refers to a millisecond count value collected after the microcontroller starts the clock calibration.
In this embodiment, the second counting value refers to a second counting value collected after the microcontroller starts the clock calibration.
The beneficial effects of the above technical scheme are that: the invention uses an alternating current detection module to detect the zero crossing point of an input power frequency signal, generates a zero crossing interrupt signal, sends the zero crossing interrupt signal to a microcontroller, and acquires internal time data corresponding to each zero crossing interrupt signal after the microcontroller does not start clock calibration.
Example 4:
on the basis of example 3, the step 1.1: the method comprises the following steps of detecting the zero crossing point of an input power frequency signal by using an alternating current detection module, and generating a zero crossing interrupt signal, wherein the method comprises the following steps:
acquiring a signal image of the input power frequency signal, and marking a first position corresponding to a zero-crossing point of the input power frequency signal on the signal image;
acquiring a logic level diagram corresponding to the input power frequency signal based on the signal image, and marking a second position generated by interruption on the logic level diagram;
aligning a signal image with the logic level image to obtain an aligned image, and obtaining a position difference between the first position and the second position based on the aligned image;
judging whether an interference signal exists in the input power frequency signal or not based on the position difference, judging that no interference signal exists in the input power frequency signal when the position difference is always consistent, normally detecting a zero crossing point of the alternating current detection module, and generating a zero crossing interrupt signal based on the zero crossing point;
when the position difference is inconsistent, when an interference signal exists in the input power frequency signal, acquiring alignment position change data of the zero-crossing point and the interrupt signal based on the alignment image;
meanwhile, standard data detected by the zero crossing point are obtained, and the change data of the standard data are compared to obtain a comparison result;
determining data fluctuation amplitude according to the comparison result, judging that zero crossing point detection of the alternating current detection module is abnormal when the data fluctuation amplitude is larger than or equal to a preset value, displaying the abnormal zero crossing point detection based on a display module, and stopping generating a zero crossing interrupt signal;
and when the data fluctuation amplitude is smaller than a preset value, judging that the zero-crossing point detection of the alternating current detection module is normal, and generating a zero-crossing interrupt signal based on the zero-crossing point.
In this embodiment, the signal image refers to a voltage waveform diagram of an input power frequency signal
In this embodiment, the first position is a position where a zero point passes on a signal image of the input power frequency signal.
In this embodiment, the logic level diagram refers to a logic image in which only high and low levels represent the input power frequency signal.
In the present embodiment, the second position refers to a position marked on the logic level diagram where the interrupt signal is generated.
In this embodiment, the alignment image refers to an image obtained by aligning a signal image and a logic level diagram.
In this embodiment, the position difference refers to a direct difference between the first position and the second position.
In this embodiment, the alignment position change data refers to data corresponding to a position of a zero crossing point and a position where an interrupt signal is generated, where a logic level diagram is changed due to existence of an interference signal, a time difference between the interrupt signal and the zero crossing point is not fixed, time intervals between the interrupt signal and between the zero crossing point and the zero crossing point are not fixed, and the position of the zero crossing point and a position where the interrupt signal is generated at this time.
In this embodiment, the standard data indicates that, when no interference signal exists in the input power frequency signal, time intervals between the interrupt signal and between the zero-crossing point and the zero-crossing point are fixed, and a position difference between the zero-crossing point and the interrupt signal is also fixed.
In this embodiment, the data fluctuation amplitude number refers to a time interval change amplitude between the interrupt signal and between the zero crossing point and the zero crossing point caused by the interference signal.
The beneficial effects of the above technical scheme are that: the invention determines the zero crossing point position through the signal image of the input power frequency signal, and determines the interrupt generation position through the logic level diagram corresponding to the input power frequency signal, thereby being beneficial to rapidly determining the position difference between the input power frequency signal and the interrupt generation position; and judging whether an interference signal exists in the input power frequency signal according to whether the position difference changes along with time according to the change, so that the time interval for sending the interrupt signal is kept consistent, and the data accuracy is improved.
Example 5:
on the basis of embodiment 3, the internal data is stored in a ring-shaped storage buffer, the internal data corresponds to the zero-crossing interrupt signals one by one, when a data sequence formed by the internal data is discontinuous, it is determined that the zero-crossing signal is lost in the current recording process, and the control algorithm program restarts a new recording process.
The beneficial effects of the above technical scheme are that: the internal data is stored in the annular storage buffer area, so that the acquisition and the processing of the internal data are realized at the same time, and the continuous correction of the crystal oscillator can be realized in the running process of the time schedule controller; when a data sequence formed by internal data is discontinuous, judging that zero-crossing signals are lost in the current recording process, controlling an algorithm program to restart a new recording process, ensuring the accuracy of data acquisition and the accuracy of calibration coefficient calculation.
Example 6:
on the basis of example 3, the step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment, and specifically comprising the following steps of:
obtaining the ith internal data of the data sequence, and calculatingThe time stamp t of the ith zero-crossing signal corresponding to the ith internal data i ：
t i ＝T si +T msi +T usi /TIM_Period
TIM _ Period represents the automatic reloading value of a timer register in the microcontroller; after all internal data in the data sequence are processed, a continuous time stamp t is formed 0 、t 1 、t 2 ……t n The interval Δ t between adjacent time stamps is a fixed value;
selecting a selected time interval on the time stamp sequence, and acquiring a starting time stamp t of the selected time interval 1 And an end timestamp t 2 Calculating a time difference DeltaT between the timer and the selected period 1-2 ：
ΔT 1-2 ＝t 2 -t 1
Calculating a calibration coefficient K of the timer under the current environment according to the time difference corresponding to the timer and the selected time interval and the time length of the selected time interval:
K＝ΔT 1-2 /ΔT
where Δ T represents the selected period time length.
The beneficial effects of the above technical scheme are that: the invention calculates the time stamp t of the ith zero-crossing point signal corresponding to the ith internal data i Forming a succession of time stamps t 0 、t 1 、t 2 ……t n Then selecting a selected time interval Delta T to calculate the time difference Delta T between the timer and the selected time interval 1-2 And finally, obtaining a calibration coefficient K of the timer under the current environment, and providing theoretical data support for time self-correction of the time schedule controller.
Example 7:
on the basis of the embodiment 1, the step 2 further includes:
acquiring historical timestamp sequences corresponding to a plurality of historical data sequences, and comparing each timestamp in the historical timestamp sequences with adjacent timestamps thereof respectively to obtain a plurality of timestamp deviation values;
when all timestamp deviation values corresponding to the historical timestamp sequence are smaller than a set deviation value, taking the historical timestamp sequence as a first to-be-selected number sequence;
when the timestamp deviation value which is not smaller than the set deviation value exists in all timestamp deviation values corresponding to the historical timestamp sequence, marking the timestamp of which the timestamp deviation value is smaller than the set deviation value on the historical timestamp sequence to obtain a marked timestamp;
judging whether the adjacent marking time stamps exist in the marking time stamps, and if not, deleting the marks of the marking time stamps;
if yes, judging that the marking timestamps are on continuous marking lines, determining the head position of the continuous marking lines, deleting the timestamp marks on the continuous marking lines, selecting marks at the head position, and deleting the selected marks when the first time length between the selected marks is smaller than a preset threshold value;
taking the historical timestamp sequence with the selected mark as a second to-be-selected number sequence;
wherein each selected mark has one and only one corresponding selected mark;
establishing a training set based on the first to-be-selected number sequence and the second to-be-selected number sequence, training a deep learning neural network through the training set, and obtaining a selected time period identification model;
acquiring a time period to be selected of the current timestamp sequence according to the selected time period identification model;
acquiring data acquisition time of internal data corresponding to the time period to be selected, and screening out the latest time period to be selected and the next time period to be selected based on the data acquisition time;
respectively acquiring a second time length of the latest time period to be selected and a third time length of the next time period to be selected;
when the second time length is greater than or equal to a third time length, taking the latest time period to be selected as a final selected time period;
and when the second time length is smaller than the third time length, taking the time period to be selected newly as a final selected time period.
In this embodiment, the historical data sequence refers to that the data sequences formed by the internal data corresponding to the last sliding arithmetic mean calculation window and the previous sliding arithmetic mean calculation window are historical data sequences, and a new group of historical data sequences is generated every time the sliding arithmetic mean calculation window is updated.
In this embodiment, the historical timestamp sequence refers to a timestamp sequence corresponding to the historical data sequence.
In this embodiment, the timestamp deviation value refers to a difference between any one timestamp and its neighboring timestamp in the historical data sequence.
In this embodiment, the first to-be-selected sequence refers to a historical timestamp sequence in which the timestamp deviation values of all timestamps are smaller than the set deviation value.
In this embodiment, the time stamp marking means that there is a time stamp having a time stamp deviation value greater than or equal to the time stamp deviation value in the historical time stamp sequence in which the deviation value is set, and the time stamp deviation value is smaller than the time stamp in which the deviation value is set.
In this embodiment, the continuous marking line refers to a continuous line segment in a shape in which a plurality of time stamp marks each having a time stamp are connected.
In this embodiment, the selected mark refers to a selected time interval mark performed at the head position of the continuous mark line.
In this embodiment, the second candidate sequence refers to a history data sequence with a selected flag.
In this embodiment, the selected time period identification model is used to select a selected time period on the sequence of time stamps.
In this embodiment, the first time length is a data acquisition time length corresponding to a selected time period between selected markers on the historical data sequence.
In this embodiment, the time period to be selected refers to a plurality of time periods which are selected on the current timestamp sequence by using the selected time period recognition model and can be used for calculating the calibration coefficient.
In this embodiment, the latest time period to be selected refers to a time period to be selected that is closest to the current data processing time; the next time period to be selected refers to the time period to be selected which is next closest to the current data processing time.
In this embodiment, the second time length refers to a data acquisition duration corresponding to the latest time period to be selected.
In this embodiment, the third time length is a data acquisition time length corresponding to the next time period to be selected.
The beneficial effects of the above technical scheme are that: according to the method, historical timestamp sequences corresponding to a plurality of historical data sequences are obtained, and selected time periods are marked on the historical timestamp sequences respectively to obtain selected time periods with different lengths, so that a training set constructed by the historical timestamp sequences has diversity, and a selected time period training model obtained through training can find time periods to be selected more accurately; screening the first time length between the selected marks in the process of processing the second sequence to be selected to ensure that the selected time period identified by the training model in the selected time period all meets the minimum precision requirement of the calculation of the calibration coefficient; after the time period to be selected is obtained, screening out the latest time period to be selected and the next time period to be selected according to the data acquisition time of the internal data corresponding to the time period to be selected, and ensuring that the data adopted by the calculation of the calibration coefficient is closest to the real situation of the current environment; and a selected time period with a higher time length is selected, so that the calculation accuracy of the calibration coefficient is improved.
Example 8:
on the basis of embodiment 6, the step 2 further includes:
filtering the calibration coefficient K of the timer under the current environment, and specifically comprising the following steps:
obtaining the optimal estimated value of the calibration coefficient at the last moment
And its covariance matrix K m-1 Optimal estimated covariance matrix of calibration coefficients at the current time
Speed of change of calibration coefficient v m ＝v m-1 + a Δ t, to obtain
Corresponding to
Wherein v is m Representing the speed of change of the calibration coefficient at the current time; v. of m-1 Representing the rate of change of the calibration coefficient at the previous time;
an optimal estimation value representing a calibration coefficient at the current time; f m A deviation matrix representing the optimal estimation value of the calibration coefficient at the previous moment and the optimal estimation value of the calibration coefficient at the current moment; a transpose matrix representing a deviation matrix of the optimal estimated value of the calibration coefficient at a time and the optimal estimated value of the calibration coefficient at the current time; a represents the acceleration of the change of the calibration coefficient; b is m A deviation matrix representing the speed of change of the calibration coefficient at the previous time and the speed of change of the calibration coefficient at the current time;
an estimation value of the calibration coefficient at the current time in an estimation value space;
when there is a covariance of Q m When the noise is interfered, the following relationship is obtained:
correcting the optimal estimation value by using the observation data of the sensor, and mapping the estimation value space to an observation value space to obtain:
wherein H m A mapping matrix representing a mapping of an estimated value space to an observed value space;
a transposed matrix representing a mapping matrix from the estimated value space to the observed value space; r m Representing the covariance of the noise in the observation space;
representing an observed value;
optimal estimated value to be processed by filtering
As a final calibration factor i
In this embodiment, when the acceleration a =0 of the change of the calibration coefficient, the optimal estimated covariance matrix K of the calibration coefficient at the current time is m-1 ＝K m-1 +Δtv m-1 Speed of change of calibration coefficient v m ＝v m-1 To obtain
In this embodiment, the estimation value space is mapped to the observation value space, as shown in the following formula:
assuming an observed value of
Meanwhile, as the observation data also has the noise interference problem, such as sensor noise and the like, the distribution of the noise is used as the covariance R m And (4) showing. At this time, the observed value
And the estimated value
In the same state space but with different probability distributions, as shown in fig. 5 below.
It is believed that the overlapping portions of the two probability distributions will more closely approach the true data of the system, i.e., have higher confidence.
Here, the observed value is multiplied by the gaussian distribution of the two distributions of the estimated value, and the resulting gaussian distribution is described as follows:
∑＝∑ 0 -∑ 0 (∑ 0 +∑ 1 ) -1 ∑ 0
wherein: Σ represents the covariance of the gaussian distribution of the estimate,
represents the mean of the gaussian distribution of the estimate,
an estimate value representing a previous time;
an estimate value representing a current time; sigma 0 A covariance representing the gaussian distribution of the estimated value at the previous time; sigma 1 Representing the covariance of the gaussian distribution of the estimate at the current time.
Then, the estimated value is obtained
And observed value
Substituting to obtain:
simplifying to obtain:
the beneficial effects of the above technical scheme are as follows: the invention carries out filtering processing on the calibration coefficient K of the microcontroller clock system under the current environment, and avoids the condition of inaccurate time calibration caused by sudden change of the calibration coefficient caused by noise signals.
Example 9:
on the basis of example 1, the step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And compensating and correcting the time of the timer, specifically comprising the following steps of:
according to the nominal value f of the crystal oscillator frequency 0 And a calibration coefficient, calculating a frequency deviation value delta f between the nominal frequency and the actual frequency of the crystal oscillator:
Δf＝f 0 ×(1-K * )
and performing compensation correction on the timer according to the frequency deviation value delta f.
The beneficial effects of the above technical scheme are that: the invention is based on the nominal value f of the crystal oscillator frequency 0 The calibration coefficient is used for calculating the frequency deviation value delta f of the nominal frequency and the actual frequency of the crystal oscillator and providing an objective basis for a timer of a time schedule controller time clock system; and compensating and correcting the timer of the time sequence controller clock system according to the frequency deviation value delta f, so that the self-correction of the time sequence controller clock system is realized.
Example 10:
on the basis of embodiment 9, when the frequency deviation Δ f is greater than a preset deviation value, the timer is compensated in a stepwise compensation manner.
The beneficial effects of the above technical scheme are that: when the frequency deviation value delta f is larger than the preset deviation value and the timer is compensated, the invention adopts a step-by-step compensation mode, thereby avoiding jump caused by compensation.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A time self-correcting method of a hot runner time schedule controller is characterized by comprising the following steps:
step 1: controlling the starting time correction of a timer in the time sequence controller based on the microcontroller included by the time sequence controller, and recording internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated after the starting time correction of the timer is performed;
step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment;
and step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 Performing compensation correction on the time of the timer;
wherein, the step 2: processing the internal time data by adopting sliding arithmetic mean filtering to obtain a calibration coefficient of the timer under the current environment, and specifically comprising the following steps of:
acquiring ith internal time data of a data sequence, and calculating a time stamp t of an ith zero-crossing signal corresponding to the ith internal time data i ：
t i ＝T si +T msi +T usi /TIM_Period
TIM _ Period represents the automatic reloading value of a timer register in the microcontroller; after all internal time data in the data sequence are processed, a continuous time stamp t is formed 0 、t 1 、t 2 ……t n The interval Δ t between adjacent time stamps is a fixed value; t is si Indicating a second count value; t is msi Represents a second millisecond count value; t is usi Indicating the current count value of the timer;
selecting a selected time interval on the time stamp sequence, and acquiring a starting time stamp t of the selected time interval 1 And an end timestamp t 2 Calculating a time difference DeltaT between the timer and the selected period 1-2 ：
ΔT 1-2 ＝t 2 -t 1
Calculating a calibration coefficient K of the timer under the current environment according to the time difference corresponding to the timer and the selected time interval and the time length of the selected time interval:
K＝ΔT 1-2 /ΔT
wherein Δ T represents a selected period time length;
the step 2 further comprises:
filtering the calibration coefficient K of the timer under the current environment, and specifically comprising the following steps:
obtaining the optimal estimated value of the calibration coefficient at the last moment
And its covariance matrix K m-1 Optimal estimated covariance matrix of calibration coefficients at the current time
Speed of change of calibration coefficient v m ＝v m-1 + a Δ t, to obtain
Corresponding to
Wherein v is m Representing the speed of change of the calibration coefficient at the current time; v. of m-1 Representing the rate of change of the calibration coefficient at the previous time;
an optimal estimation value representing a calibration coefficient at the current time; f m A deviation matrix representing the optimal estimated value of the calibration coefficient at the previous moment and the optimal estimated value of the calibration coefficient at the current moment; a transpose matrix representing a deviation matrix of the optimal estimated value of the calibration coefficient at a time and the optimal estimated value of the calibration coefficient at the current time; a represents the acceleration of the change of the calibration coefficient; b is m A deviation matrix representing the speed of change of the calibration coefficient at the previous time and the speed of change of the calibration coefficient at the current time;
an estimation value of the calibration coefficient at the current time in an estimation value space;
when there is a covariance of Q m When the noise is interfered, the following relationship is obtained:
correcting the optimal estimation value by using the observation data of the sensor, and mapping the estimation value space to an observation value space to obtain:
wherein H m A mapping matrix representing a mapping of an estimated value space to an observed value space;
a transposed matrix representing a mapping matrix from the estimated value space to the observed value space; r m Representing the covariance of the noise in the observation space;
representing an observed value;
optimal estimation value to be processed by filtering
As a final calibration factor i
2. The method of claim 1, further comprising, before the timer starts time calibration: based on the time schedule controller, the time for opening and closing the sprue of the hot runner system is set, and the method specifically comprises the following steps:
according to the nominal value f of the crystal oscillator frequency 0 Determining the oscillation period of the time sequence controller, and controlling a timer in the time sequence controller to generate 1ms interrupt;
obtaining a first millisecond count value T of the timing controller through interrupt processing ms ；
Counting the first millisecond value T ms Accumulating to obtain a first second count value T for controlling the opening and closing of the gate by the time schedule controller s 。
3. The method of claim 1, wherein the step 1: based on a microcontroller included by a time schedule controller, controlling the correction of the starting time of a timer in the time schedule controller, and recording the internal time data of the corresponding microcontroller when each zero-crossing interrupt is generated after the correction of the starting time of the timer, wherein the specific steps comprise:
step 1.1: detecting a zero crossing point of an input power frequency signal by using an alternating current detection module, generating a zero crossing interrupt signal, and sending the zero crossing interrupt signal to a microcontroller;
step 1.2: controlling a timer in the time sequence controller to start time correction;
step 1.3: recording corresponding internal time data when the microcontroller receives a zero-crossing interrupt signal every time, wherein the internal time data comprises a second counting value T si A second millisecond count value T msi And the current count value T of the timer usi 。
4. The method of claim 3, wherein the method further comprises: step 1.1: the method comprises the following steps of detecting the zero crossing point of an input power frequency signal by using an alternating current detection module, and generating a zero crossing interrupt signal, wherein the method comprises the following steps:
acquiring a signal image of the input power frequency signal, and marking a first position corresponding to a zero-crossing point of the input power frequency signal on the signal image;
acquiring a logic level diagram corresponding to the input power frequency signal based on the signal image, and marking a second position generated by interruption on the logic level diagram;
aligning a signal image with the logic level image to obtain an aligned image, and obtaining a position difference between the first position and the second position based on the aligned image;
judging whether an interference signal exists in the input power frequency signal or not based on the position difference, judging that no interference signal exists in the input power frequency signal when the position difference is always consistent, normally detecting a zero crossing point of the alternating current detection module, and generating a zero crossing interrupt signal based on the zero crossing point;
when the position difference is inconsistent, when an interference signal exists in the input power frequency signal, acquiring alignment position change data of the zero-crossing point and the interrupt signal based on the alignment image;
meanwhile, standard data of zero crossing point detection are obtained, and the change data of the standard data are compared to obtain a comparison result;
determining data fluctuation amplitude according to the comparison result, judging that zero crossing point detection of the alternating current detection module is abnormal when the data fluctuation amplitude is larger than or equal to a preset value, displaying the abnormal zero crossing point detection based on a display module, and stopping generating a zero crossing interrupt signal;
and when the data fluctuation amplitude is smaller than a preset value, judging that the zero crossing point detection of the alternating current detection module is normal, and generating a zero crossing interrupt signal based on the zero crossing point.
5. The method of claim 3, wherein the method further comprises:
the internal time data are stored in an annular storage buffer area, the internal time data correspond to zero-crossing interrupt signals one by one, when a data sequence formed by the internal time data is discontinuous, the fact that zero-crossing signals are lost in the current recording process is judged, and an algorithm program is controlled to restart a new recording process.
6. The method of claim 1, wherein the method further comprises: the step 2 further comprises:
acquiring historical timestamp sequences corresponding to a plurality of historical data sequences, and comparing each timestamp in the historical timestamp sequences with adjacent timestamps thereof respectively to obtain a plurality of timestamp deviation values;
when all timestamp deviation values corresponding to the historical timestamp sequence are smaller than a set deviation value, taking the historical timestamp sequence as a first to-be-selected number sequence;
when the timestamp deviation value which is not smaller than the set deviation value exists in all timestamp deviation values corresponding to the historical timestamp sequence, marking the timestamp of which the timestamp deviation value is smaller than the set deviation value on the historical timestamp sequence to obtain a marked timestamp;
judging whether the adjacent marking time stamps exist in the marking time stamps, and if not, deleting the marks of the marking time stamps;
if yes, judging that the marking timestamps are on continuous marking lines, determining the head position of the continuous marking lines, deleting the timestamp marks on the continuous marking lines, selecting marks at the head position, and deleting the selected marks when the first time length between the selected marks is smaller than a preset threshold value;
taking the historical timestamp sequence with the selected mark as a second to-be-selected number sequence;
wherein each selected mark has one and only one corresponding selected mark;
establishing a training set based on the first number series to be selected and the second number series to be selected, and training a deep learning neural network through the training set to obtain a selected time period identification model;
acquiring a time period to be selected of the current timestamp sequence according to the selected time period identification model;
acquiring data acquisition time of internal time data corresponding to the time period to be selected, and screening out the latest time period to be selected and the next time period to be selected based on the data acquisition time;
respectively acquiring a second time length of the latest time period to be selected and a third time length of the next time period to be selected;
when the second time length is greater than or equal to a third time length, taking the latest time period to be selected as a final selected time period;
and when the second time length is smaller than the third time length, taking the time period to be selected newly as a final selected time period.
7. The method of claim 1, wherein the step 3: according to the calibration coefficient and the nominal value f of the crystal oscillator frequency 0 And compensating and correcting the time of the timer, which comprises the following specific steps:
according to the nominal value f of the crystal oscillator frequency 0 And a calibration coefficient, calculating a frequency deviation value delta f between the nominal frequency and the actual frequency of the crystal oscillator:
Δf＝f 0 ×(1-K * )
performing compensation correction on the timer according to the frequency deviation value delta f; k is * The final calibration coefficients are represented.
8. The method of claim 7, wherein the method further comprises:
and when the frequency deviation value delta f is larger than a preset deviation value, compensating the timer in a step-by-step compensation mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210747309.1A CN115079553B (en) | 2022-06-28 | 2022-06-28 | Time self-correction method for hot runner time schedule controller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210747309.1A CN115079553B (en) | 2022-06-28 | 2022-06-28 | Time self-correction method for hot runner time schedule controller |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115079553A CN115079553A (en) | 2022-09-20 |
| CN115079553B true CN115079553B (en) | 2023-02-28 |
Family
ID=83255660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210747309.1A Active CN115079553B (en) | 2022-06-28 | 2022-06-28 | Time self-correction method for hot runner time schedule controller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115079553B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4942559A (en) * | 1988-07-08 | 1990-07-17 | Siemens Aktiengesellschaft | Counter/timer circuit for a microcontroller |
| US5191336A (en) * | 1991-08-29 | 1993-03-02 | Hewlett-Packard Company | Digital time interpolation system |
| US5253179A (en) * | 1990-12-24 | 1993-10-12 | Xerox Corporation | Compensation for line frequency variation |
| CN103138677A (en) * | 2011-11-29 | 2013-06-05 | 江苏林洋电子股份有限公司 | Method, device and system of calibration of baud rate |
| WO2018159726A1 (en) * | 2017-02-28 | 2018-09-07 | 住友重機械工業株式会社 | Correcting device, injection molding system, and correcting method |
| CN109062021A (en) * | 2018-09-21 | 2018-12-21 | 珠海格力电器股份有限公司 | It is a kind of to interrupt the method and air conditioner for generating time base by zero passage |
| CN109581856A (en) * | 2018-12-13 | 2019-04-05 | 国电南瑞科技股份有限公司 | A kind of punctual method of clock synchronization based on the calibration of high-performance crystal oscillator frequency |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004064469A (en) * | 2002-07-30 | 2004-02-26 | Mitsubishi Electric Corp | Timing correcting circuit and receiver |
| JP5306512B1 (en) * | 2012-04-27 | 2013-10-02 | ラピスセミコンダクタ株式会社 | Semiconductor device, measuring instrument, and correction method |
| CN111319210B (en) * | 2018-12-14 | 2021-08-10 | 广州捷克易自动化设备有限公司 | Intelligent hot runner temperature control method and temperature control system |
-
2022
- 2022-06-28 CN CN202210747309.1A patent/CN115079553B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4942559A (en) * | 1988-07-08 | 1990-07-17 | Siemens Aktiengesellschaft | Counter/timer circuit for a microcontroller |
| US5253179A (en) * | 1990-12-24 | 1993-10-12 | Xerox Corporation | Compensation for line frequency variation |
| US5191336A (en) * | 1991-08-29 | 1993-03-02 | Hewlett-Packard Company | Digital time interpolation system |
| CN103138677A (en) * | 2011-11-29 | 2013-06-05 | 江苏林洋电子股份有限公司 | Method, device and system of calibration of baud rate |
| WO2018159726A1 (en) * | 2017-02-28 | 2018-09-07 | 住友重機械工業株式会社 | Correcting device, injection molding system, and correcting method |
| CN109062021A (en) * | 2018-09-21 | 2018-12-21 | 珠海格力电器股份有限公司 | It is a kind of to interrupt the method and air conditioner for generating time base by zero passage |
| CN109581856A (en) * | 2018-12-13 | 2019-04-05 | 国电南瑞科技股份有限公司 | A kind of punctual method of clock synchronization based on the calibration of high-performance crystal oscillator frequency |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115079553A (en) | 2022-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR20060050551A (en) | Method for measuring rotation angle position of camschaft of reciprocating piston-combustion engine relative to crankshaft | |
| CN107797446B (en) | Model prediction control device, control method, and recording medium | |
| CN107782421B (en) | Calibration method for water meter metering threshold value based on nonmagnetic NB-IoT (NB-IoT) internet of things | |
| US20130081477A1 (en) | Flow meter device | |
| US10135440B2 (en) | Magnetic field triggering | |
| CN115079553B (en) | Time self-correction method for hot runner time schedule controller | |
| US20190064281A1 (en) | Sensor module and residual battery capacity monitoring method using sensor module | |
| ATE498108T1 (en) | SYSTEM AND METHOD FOR COMPENSATING FOR MAGNETIC INTERFERENCES ON A COMPASS THROUGH A MOVABLE VEHICLE PART | |
| CN116451014B (en) | Real-time monitoring method for temperature of movable part in wireless passive equipment | |
| US9966753B2 (en) | Power switching device operating time prediction equipment and method | |
| US8719614B2 (en) | Apparatus and method for generating a timing signal | |
| US5508698A (en) | Vehicle detector with environmental adaptation | |
| KR101100128B1 (en) | Method and arrangement for frequency synchronization of a mobile station with a base station in a mobile communication system | |
| JP2008096328A (en) | Charging rate estimation method, charging rate estimating device, and secondary battery power supply system | |
| CN116717938B (en) | Method and system for detecting and controlling internal environment of refrigerator | |
| US20230010155A1 (en) | Time correction apparatus, time correction method, and computer readable medium | |
| KR101076310B1 (en) | Conveying device and computer program | |
| US20150015172A1 (en) | Motor controller | |
| CN111897202A (en) | RTC (real time clock) calibration circuit and method of smoke detector MCU (microprogrammed control Unit) | |
| US7106054B2 (en) | Device for sensing a rotary member such as a water meter turbine | |
| US20140114593A1 (en) | Apparatus and method for calculating electrical charge quantity or electrical discharge quantity for battery unit according to calibration parameter used for adjusting time interval of electrical charge/discharge calculation, and method for calibrating integral time interval | |
| US11636001B2 (en) | Method and system for determining an error threshold value for machine failure prediction | |
| CN106843051B (en) | A kind of FPGA time-delay mechanism and method | |
| CN110618604A (en) | Method and device for improving time keeping precision by using NTP auxiliary source | |
| JP6550036B2 (en) | Secondary battery state detection device and secondary battery state detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |