CN109902335B - Filtering method and system for realizing on-line control model process stability - Google Patents
Filtering method and system for realizing on-line control model process stability Download PDFInfo
- Publication number
- CN109902335B CN109902335B CN201910027096.3A CN201910027096A CN109902335B CN 109902335 B CN109902335 B CN 109902335B CN 201910027096 A CN201910027096 A CN 201910027096A CN 109902335 B CN109902335 B CN 109902335B
- Authority
- CN
- China
- Prior art keywords
- bandwidth
- filtering
- value
- data point
- unit
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention provides a filtering method and a system for realizing the process stability of an online control model, which comprises the following steps: setting a bandwidth period; collecting data points output by the online control model along with time; taking a steady-state data point which is not less than a bandwidth period, and obtaining the maximum value and the minimum value of the steady-state data point, wherein the difference between the maximum value and the minimum value is a filtering bandwidth; filtering each data point of the online control model by using a filtering bandwidth, wherein the filtering bandwidth comprises the following steps: setting the maximum value and the minimum value of the initial bandwidth as any value under the condition that the filtering bandwidth is not changed; judging whether the first data point is in the range of the initial bandwidth maximum value and the initial bandwidth minimum value; filtering the first data point to a bandwidth target value within the range, wherein the bandwidth target value is any value in the range; if not, the filtering bandwidth is wholly translated, so that the first data point is filtered into a translated bandwidth target value within the bandwidth maximum value and the bandwidth minimum value range corresponding to the translated filtering bandwidth; and repeating the steps to sequentially filter the data points.
Description
Technical Field
The invention relates to the technical field of ferrous metallurgy, in particular to a filtering method and a filtering system for realizing on-line control of model process stability.
Background
In the online control model of the continuous casting process, most of the results are calculated based on the temperature field model, such as: calculating the surface temperature, the solidification tail end, the shell thickness and the like of the dynamic secondary cooling model based on the temperature field; the dynamic reduction model is based on the solid phase rate and the like of the solidification tail end and different positions of a casting blank, and the temperature field model is a discrete solution of a continuous heat transfer differential equation, more importantly, discrete gridding is also carried out along the blank drawing direction, management and calculation are carried out in a billet mode, so that the calculation result of the temperature field model also has oscillation even under the steady state condition, the dynamic secondary cooling water and pressing-down process is formulated by the calculation result with the oscillation, and the oscillation of process parameters is inevitably brought.
For example, the rolling reduction of the rolling roller of the soft-reduction model is mainly formulated according to the central solid phase rate corresponding to the solidification tail end or the rolling roller, the solidification tail end calculated by the temperature field model and the central solid phase rate corresponding to the rolling roller still vibrate even under the steady state condition, so that the rolling reduction of the soft-reduction model is continuously vibrated, the continuously vibrated rolling reduction is used for controlling equipment, and the equipment is naturally and always in a frequent adjustment process, so that the process is unstable, and the frequent adjustment of a control system brings influence on the service life of electric appliances and equipment.
The steady state refers to the condition that parameters for calculating the temperature field model such as the pulling speed, the tundish temperature, the crystallizer water quantity and temperature difference, the secondary cooling water quantity and the like are not changed, the non-steady state refers to the condition that the parameters for calculating the temperature field model are changed, and the change of the pulling speed has the most obvious influence on the calculation result of the temperature field.
The existing filtering method is commonly used by a least square fitting method and a mean value filtering method, the least square fitting algorithm can only smooth data to a certain extent, and the filtering method has the characteristic of retaining detail signals, so when noise signals are too strong, the smoothing effect of the least square fitting method is not obvious, the data after filtering still retains too much noise, and the requirements are difficult to meet. The average filtering method has the greatest characteristic of obtaining good smoothing effect, but the smoothing effect is at the cost of time lag and loss of detail signals. Therefore, under the condition of low signal-to-noise ratio, the mean filtering of a long data segment must be adopted in order to obtain a satisfactory smoothing effect, and as a result, detail information at some key positions is lost, and the hysteresis phenomenon is obvious, that is, the mean filtering can be performed on the basis of the least square fitting technology to form an improved comprehensive filtering algorithm, as shown in fig. 1, although the improved comprehensive filtering algorithm can correct the dead pixel to a certain extent when the data has the dead pixel, under the condition that the original data has fluctuation, the processing result of the existing filtering method still remains in the fluctuation state process, and after the objective change occurs, the processing result has a certain hysteresis, which is the characteristic of the filtering processing.
In a steady state situation, the change of the calculation result of the model is completely non-objective and is caused by the model, and noise is taken as a main part, while in a non-steady state situation, the change of the calculation result is objective after the noise caused by the model is removed, and the change is not processed by hysteresis, and the current method for filtering the signal is not suitable. The existing filtering method cannot meet the requirement of a stable process.
The existing filtering method is difficult to ensure that noise can be completely filtered under the steady state condition, so that the stability of a model calculation result is ensured, the oscillation range of the calculation result can be only reduced, and the requirement of process stability cannot be met. More importantly, if the corresponding solid phase ratio of the casting blank at the solidification end or the rolling roller is necessarily changed in an unsteady state, such as a continuous increasing process of the drawing speed, the change is a practical situation, but the existing filtering method has more or less hysteresis, and the hysteresis is greatly related to the steepness degree of the unsteady state change, which is not consistent with the practical situation.
Disclosure of Invention
In view of the foregoing problems, an object of the present invention is to provide a filtering method and system for online control model process stability, which achieves a function of oscillation removal of steady-state and unsteady-state process model calculation results.
According to one aspect of the invention, a filtering method for realizing the process stability of an online control model is provided, which comprises the following steps:
setting a bandwidth period, wherein the bandwidth period is a set number of data points;
collecting data points output by the online control model along with time;
taking a steady-state data point output by the online control model with a bandwidth period not less than time to obtain the maximum value and the minimum value of the steady-state data point, wherein the difference between the maximum value and the minimum value is a filtering bandwidth;
filtering each data point of the online control model by using the filtering bandwidth, including:
setting the maximum value and the minimum value of the initial bandwidth as any value under the condition that the difference between the maximum value and the minimum value of the initial bandwidth is the filtering bandwidth;
judging whether the value of a first data point output by the online control model along with time is within the range of the maximum value of the initial bandwidth and the minimum value of the initial bandwidth;
if the first data point is in the range, filtering the value of the first data point into a bandwidth target value, wherein the bandwidth target value is any value in the range of the initial bandwidth maximum value and the initial bandwidth minimum value;
if the first data point is not in the range, integrally translating the filtering bandwidth, so that the numerical value of the first data point is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth, and filtering the numerical value of the first data point into a translated bandwidth target value, wherein the translated bandwidth target value is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth;
and repeating the steps, and sequentially filtering the data points output by the online control model along with the time.
Preferably, the bandwidth target value is set to be a bandwidth minimum value, or a bandwidth maximum value, or an average value of the bandwidth maximum value and the bandwidth minimum value, or a golden section point of the bandwidth minimum value and the bandwidth maximum value.
Further, preferably, the method for setting the bandwidth cycle includes:
judging whether the model output time sequence has periodic fluctuation in a steady state;
if no periodic fluctuation exists, setting a bandwidth period, wherein the bandwidth period is the number of data points with a set number;
and if the periodic fluctuation exists, determining a bandwidth period according to the periodic fluctuation, wherein the bandwidth period is the number of data points of one fluctuation period of the periodic fluctuation.
Preferably, the filtering method for realizing the process stability of the online control model further includes:
acquiring a parameter time sequence of parameters for a temperature field model along with time, wherein the parameters for the temperature field model comprise a pulling speed, a tundish temperature, a crystallizer water quantity, a temperature difference and a secondary cooling water quantity, the parameters for the temperature field model are taken as a stable state when not changed, and the parameters for the temperature field model are taken as an unstable state when changed;
and acquiring a model output time sequence of an online control model along with time corresponding to the parameter time sequence, wherein the online control model comprises one or two of a dynamic secondary cooling model and a dynamic reduction model, and the model output comprises one or more of solid phase rate, surface temperature, shell thickness and solidification tail end position of different positions of the casting blank.
Preferably, the filtering bandwidth is adaptively changed according to a model output time sequence of the online model output along with time, and the method comprises the following steps:
setting the queue length of a filtering bandwidth queue as m, wherein m is greater than the number n of data points in a bandwidth period;
searching a first data point of the model output time series in a steady state, and adding the first data point into the filtering bandwidth queue;
judging whether a data point subsequent to the first data point in the model output time series is also in a steady state;
when the latter data point of the first data point is also in a steady state, adding the latter data point into the filter bandwidth queue;
when the latter data point of the first data point is not in a steady state, emptying the filter bandwidth queue, and searching the first data point in the steady state again;
repeating the steps until the queue length of the filter bandwidth queue reaches m, and taking the difference value between the maximum value and the minimum value in the filter bandwidth queue as the filter bandwidth;
and emptying the filter bandwidth queue with the queue length reaching m, returning to the step of searching the first data point in a steady state, and updating the filter bandwidth.
Preferably, the bandwidth period is a set time length or a time length of a fluctuation period in which the model output time series fluctuates periodically.
According to another aspect of the present invention, there is provided a filtering system for realizing process stability of an online control model, comprising:
the bandwidth period setting module is used for setting a bandwidth period, wherein the bandwidth period is a set number of data points;
the acquisition module acquires data points output by the online control model along with time;
the filter bandwidth obtaining module is used for obtaining a steady-state data point output by the online control model with the bandwidth period not less than the time to obtain the maximum value and the minimum value of the steady-state data point, and the difference between the maximum value and the minimum value is the filter bandwidth;
a filtering module for filtering each data point of the online control model using the filtering bandwidth,
wherein the filtering module comprises:
a fourth setting unit that sets the maximum value of the initial bandwidth and the minimum value of the initial bandwidth as any value under the condition that the difference between the maximum value of the initial bandwidth and the minimum value of the initial bandwidth is satisfied as the filter bandwidth;
the third judging unit is used for judging whether the value of the first data point output by the online control model along with the time is in the range of the maximum value of the initial bandwidth and the minimum value of the initial bandwidth; if the first data point is within the range, sending a signal to a filtering unit, and if the first data point is not within the range, sending a signal to a translation unit;
the translation unit is used for translating the whole filtering bandwidth to ensure that the numerical value of the first data point is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth, and sending signals to the filtering unit and the fourth setting unit, and the fourth setting unit is used for updating the maximum initial bandwidth value and the minimum initial bandwidth value according to the signals of the translation unit;
the filtering unit is used for receiving the signal of the judging unit and filtering the numerical value of the first data point into a bandwidth target value, wherein the bandwidth target value is any value in the range of the initial bandwidth maximum value and the initial bandwidth minimum value; receiving a signal of a translation unit, and filtering the value of the first data point into a translated bandwidth target value, wherein the translated bandwidth target value is within a bandwidth maximum value and a bandwidth minimum value range corresponding to the translated filtering bandwidth;
and the second counting unit counts the data points filtered by the filtering unit and sends the next data point serving as the first data point to the third judging unit.
Preferably, the bandwidth target value is set to be a bandwidth minimum value, or a bandwidth maximum value, or an average value of the bandwidth maximum value and the bandwidth minimum value, or a golden section point of the bandwidth minimum value and the bandwidth maximum value.
Preferably, the bandwidth cycle setting module includes:
the first judging unit is used for judging whether the model output time sequence has periodic fluctuation in a steady state or not, and sending a signal to the first setting unit if the model output time sequence does not have the periodic fluctuation; if the periodic fluctuation exists, sending a signal to a second setting unit;
the device comprises a first setting unit, a second setting unit and a control unit, wherein the first setting unit sets a bandwidth period which is a set number of data points;
and the second setting unit is used for determining a bandwidth period according to the periodic fluctuation, wherein the bandwidth period is the number of data points of one fluctuation period of the periodic fluctuation.
Preferably, the filtering bandwidth obtaining module includes a fixed bandwidth setting part or a bandwidth adaptation part,
the fixed bandwidth setting part randomly selects continuous steady-state data points output by the online control model with the bandwidth period not less than the bandwidth period along with time to obtain the maximum value and the minimum value of the continuous steady-state data points, the difference between the maximum value and the minimum value is a filtering bandwidth, and the filtering module adopts the filtering bandwidth obtained by the fixed bandwidth setting part to filter each data point output by the online control model along with the time with the fixed bandwidth;
wherein the bandwidth adaptation part includes:
a third setting unit, setting the queue length of the filter bandwidth queue as m, wherein m is greater than the number n of data points in the bandwidth period;
the searching unit is used for searching a first data point of the model output time sequence in a steady state;
the construction unit is used for adding the first data point into the filtering bandwidth queue;
the second judging unit is used for judging whether a data point behind the first data point in the model output time sequence is also in a stable state or not, and sending a signal to the constructing unit when the data point behind the first data point is also in the stable state, and the constructing unit adds the data point behind the first data point into the filtering bandwidth queue; when a data point behind the first data point is not in a stable state, sending a signal to an emptying unit and a searching unit, wherein the emptying unit empties a filtering bandwidth queue, and the searching unit searches the first data point in the stable state again;
the first counting unit is used for counting the queue length of the filtering bandwidth queue constructed by the construction unit, and sending a signal to the bandwidth obtaining unit when the queue length reaches m;
and the bandwidth obtaining unit is used for sending a signal to the emptying unit and the searching unit by taking the difference value between the maximum value and the minimum value in the filtering bandwidth queues built by the building unit as the filtering bandwidth after receiving the signal of the first counting unit, wherein the emptying unit empties the filtering bandwidth queue with the queue length reaching m in the building unit, and the searching unit re-searches the first data point in a steady state in the data points of the model output time sequence after the filtering bandwidth queue with the queue length reaching m is added.
The filtering method and the system for the process stability of the online control model are passivation filtering methods and systems suitable for the online control model, and can realize the function of removing oscillation of the calculated results of the steady-state process model and the unsteady-state process model, thereby ensuring the stability of the online given process, improving the process stability of the equipment control system and prolonging the service life of the control system and the equipment system. Meanwhile, the online statistical method of the bandwidth is provided, and the online bandwidth parameter self-adjustment can be realized.
Drawings
Other objects and results of the present invention will become more apparent and more readily appreciated as the same becomes better understood by reference to the following description taken in conjunction with the accompanying drawings. In the drawings:
FIG. 1 is a graph of prior art pull rate filtering with integrated enhancement filtering;
FIG. 2 is a filtering method for realizing the process stability of an online control model according to the present invention;
FIG. 3 is a schematic diagram of the passivation filtering method of the present invention;
FIG. 4 is a block diagram of a filtering system for realizing the process stability of an online control model according to the present invention;
FIG. 5 is a graph of the solidification end as a function of time at steady state;
FIG. 6 is a schematic representation of the coagulation tip before and after filtering over a filter bandwidth;
FIG. 7 is a schematic illustration of one manner in which the coagulation tip does not move bandwidth within the filter bandwidth;
FIG. 8 is a schematic illustration of one manner in which the coagulation tip does not move bandwidth within the filter bandwidth;
FIG. 9 is a graph of pre-filtering and post-filtering of the coagulation tip at steady state;
FIG. 10 is a graph of pre-filtering and post-filtering of the coagulation tip at non-steady state;
FIG. 11 is a graph of a long time temperature field calculation of solidification end over time without passivation filtering;
FIG. 12 is a graph of the time-dependent change in the temperature field over time of the coagulation tip after passivation filtering;
FIG. 13 is a graph of comparative plots of calculated solidification end time-dependent changes in the long-term temperature field without and with passivation filtering;
FIG. 14 is a graph of the change in solid fraction with time;
FIG. 15 is a graph of a comparison of the pre-and post-passivation filtering of a casting blank solid fraction time series at one position in a short time;
FIG. 16 is a graph of a comparison of the slab solid fraction time series before filtering and after passivation filtering at one of the positions over time;
FIG. 17 is a graph of a comparison of pre-and post-passivation filtering of an ingot solid fraction time series at another location.
Detailed Description
In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. It may be evident, however, that such embodiment(s) may be practiced without these specific details. Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
Fig. 2 shows a schematic diagram of a filtering method for realizing the process stability of an online control model according to the present invention, and as shown in fig. 2, the filtering method includes:
step S1, setting a bandwidth period, wherein the bandwidth period is a set number of data points;
s2, collecting data points output by the online control model along with time;
s3, taking a steady-state data point output by the online control model with the bandwidth period not less than the time to obtain the maximum value and the minimum value of the steady-state data point, wherein the difference between the maximum value and the minimum value is the filter bandwidth;
step S4, filtering each data point of the online control model by using the filtering bandwidth, specifically, filtering by using a "passivation method", which is to set a bandwidth for a variable in general terms, as shown in fig. 3, the bandwidth parameter includes a bandwidth minimum value Bmin, a bandwidth maximum value Bmax, and a bandwidth target value Aim, where Bmin and Bmax form a bandwidth range (filtering bandwidth), and the bandwidth target value is a filtering value (the bandwidth target value may be an average of the bandwidth maximum value Bmax and the bandwidth minimum value Bmin, or may be obtained according to the bandwidth maximum value Bmax and the bandwidth minimum value Bmin by combining a setting algorithm according to an actual situation, such as a golden section point of the bandwidth minimum value and the bandwidth maximum value). If the real-time variable value output by the online control model is within the bandwidth range, the variable is always filtered to be a bandwidth target value; if the real-time variable value is not in the bandwidth range, corresponding change is carried out, specifically, the bandwidth is translated according to a set rule (namely, the maximum value and the minimum value of the bandwidth are simultaneously increased or decreased), so that the real-time variable value is in the translated filtering bandwidth, and the variable filtering value is filtered into a translated bandwidth target value Aim, and the filtering method is called as 'passivation filtering'.
In an optional embodiment, in step S4, the method for filtering each data point of the online control model with the filter bandwidth includes:
step S41, under the condition that the difference between the maximum value of the initial bandwidth and the minimum value of the initial bandwidth is satisfied as the filtering bandwidth (the difference value is not changed, for example, the filtering bandwidth in a steady state), setting the maximum value of the initial bandwidth and the minimum value of the initial bandwidth as any value to form an initial value range, and obtaining a bandwidth target value, wherein the bandwidth target value is in the range;
step S42, judging whether the value of a first data point output by the online control model along with time is in the range of the maximum value of the initial bandwidth and the minimum value of the initial bandwidth;
if the first data point is within the range, in step S43, filtering the value of the first data point into a bandwidth target value, where the bandwidth target value is any value within the range of the initial bandwidth maximum value and the initial bandwidth minimum value, for example, the bandwidth target value may be set as a bandwidth minimum value, or a bandwidth maximum value, or an average value of the bandwidth maximum value and the bandwidth minimum value, or a golden section point of the bandwidth minimum value and the bandwidth maximum value;
if the first data point is not within the range, in step S44, translating the entire filtering bandwidth, so that the value of the first data point is within the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth, and filtering the value of the first data point into a translated bandwidth target value, where the translated bandwidth target value is within the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth;
step S45, taking the bandwidth maximum value and the bandwidth minimum value range of the first data point as the initial bandwidth maximum value and the initial bandwidth minimum value range of the second data point (if the value of the first data point is within the initial value range, the initial value range is still used as the initial value range of the second data point, and if the value of the first data point is not within the initial value range, the range formed by the bandwidth maximum value and the bandwidth minimum value after the bandwidth translation is filtered is used as the initial data range of the second data point), repeating the above steps S41-S44, and sequentially filtering the data points output by the online control model over time.
In another alternative embodiment, in step S41, the adaptive variation of the filtering bandwidth according to the model output time series of the online model output over time includes:
setting the queue length of a filtering bandwidth queue as m, wherein m is greater than the number n of data points in a bandwidth period;
searching a first data point of the model output time sequence in a steady state, and adding the first data point into the filtering bandwidth queue;
judging whether a data point subsequent to the first data point in the model output time series is also in a steady state;
when the latter data point of the first data point is also in a steady state, adding the latter data point into the filter bandwidth queue;
when the latter data point of the first data point is not in a steady state, emptying the filter bandwidth queue, and searching the first data point in the steady state again;
repeating the steps until the queue length of the filter bandwidth queue reaches m, and taking the difference value between the maximum value and the minimum value in the filter bandwidth queue as the filter bandwidth;
and emptying the filter bandwidth queue with the queue length reaching m, returning to the step of searching the first data point in a steady state, and updating the filter bandwidth.
In an optional embodiment, in step S1, the method for setting a bandwidth cycle includes:
judging whether the model output time sequence has periodic fluctuation in a steady state;
if no periodic fluctuation exists, setting a bandwidth period, wherein the bandwidth period is the number of data points with a set quantity;
and if the periodic fluctuation exists, determining a bandwidth period according to the periodic fluctuation, wherein the bandwidth period is the number of data points of one fluctuation period of the periodic fluctuation.
In an optional embodiment of the present invention, the filtering method may be applied to a temperature field model and an online control model related to the temperature field model, and therefore, the filtering method further includes:
acquiring a parameter time sequence of parameters for a temperature field model along with time, wherein the parameters for the temperature field model comprise a pulling speed, a tundish temperature, a crystallizer water quantity, a temperature difference and a secondary cooling water quantity, the parameters for the temperature field model are taken as a stable state when not changed, and the parameters for the temperature field model are taken as an unstable state when changed;
and acquiring a model output time sequence of an online control model along with time corresponding to the parameter time sequence, wherein the online control model comprises one or two of a dynamic secondary cooling model and a dynamic reduction model, and the model output comprises one or more of solid phase rate, surface temperature, shell thickness and solidification tail end position at different positions of a casting blank.
Fig. 4 is a filtering system for realizing the process stability of the online control model according to the present invention, and as shown in fig. 4, the filtering system includes:
a bandwidth cycle setting module 10, configured to set a bandwidth cycle, where the bandwidth cycle is a set number of data points;
the acquisition module 20 is used for acquiring data points output by the online control model along with time;
a filtering bandwidth obtaining module 30, which obtains a steady-state data point output by the online control model with a bandwidth period not less than the time, and obtains a maximum value and a minimum value of the steady-state data point, wherein the difference between the maximum value and the minimum value is a filtering bandwidth;
a filtering module 40 for filtering each data point of the online control model using the filtering bandwidth,
wherein the filtering module 40 includes:
a fourth setting unit 41, configured to set the initial bandwidth maximum value and the initial bandwidth minimum value as any value under the condition that the difference between the initial bandwidth maximum value and the initial bandwidth minimum value is satisfied as the filter bandwidth;
the third judging unit 42 is used for judging whether the value of the first data point output by the online control model along with the time is in the range of the initial bandwidth maximum value and the initial bandwidth minimum value; if the first data point is within the range, sending a signal to a filtering unit, and if the first data point is not within the range, sending a signal to a translation unit;
the translation unit 43 is configured to translate the entire filtering bandwidth, so that the value of the first data point is within a range of a maximum bandwidth value and a minimum bandwidth value corresponding to the translated filtering bandwidth, and send a signal to the filtering unit and a fourth setting unit, where the fourth setting unit updates the maximum initial bandwidth value and the minimum initial bandwidth value according to the signal of the translation unit;
the filtering unit 44 receives the signal of the judging unit, and filters the value of the first data point into a bandwidth target value, wherein the bandwidth target value is any value within a range of an initial bandwidth maximum value and an initial bandwidth minimum value; receiving a signal of a translation unit, and filtering the value of the first data point into a translated bandwidth target value, wherein the translated bandwidth target value is within a bandwidth maximum value and a bandwidth minimum value range corresponding to the translated filtering bandwidth;
the second counting unit 45 counts the data points filtered by the filtering unit, and sends the next data point as the first data point to the third judging unit.
In an optional embodiment, the bandwidth cycle setting module 10 includes:
the first judging unit 11 is used for judging whether the model output time sequence has periodic fluctuation in a steady state or not, and sending a signal to the first setting unit if the model output time sequence does not have the periodic fluctuation; if the periodic fluctuation exists, sending a signal to a second setting unit;
a first setting unit 12 that sets a bandwidth period, which is a set number of data points;
and a second setting unit 13, configured to determine a bandwidth period according to the periodic fluctuation, where the bandwidth period is a number of data points of one fluctuation period of the periodic fluctuation.
In an alternative embodiment, the filtering bandwidth obtaining module 30 comprises a fixed bandwidth setting part 31 or a bandwidth adapting part 32,
the fixed bandwidth setting unit 31 randomly selects continuous steady-state data points output by the online control model with a bandwidth period not less than the bandwidth period over time, obtains a maximum value and a minimum value of the continuous steady-state data points, and a difference between the maximum value and the minimum value is a filtering bandwidth;
wherein the bandwidth adaptation part 32 includes:
a third setting unit 321, configured to set a queue length of the filtering bandwidth queue to m, where m is greater than a data point n of the bandwidth period;
a search unit 322, which searches for the first data point of the model output time series in the steady state;
a building unit 323, adding the first data point to the filter bandwidth queue;
a second determining unit 324, configured to determine whether a data point subsequent to the first data point in the model output time sequence is also in a steady state, and send a signal to a constructing unit when the data point subsequent to the first data point is also in the steady state, where the constructing unit adds the subsequent data point to the filter bandwidth queue; when the latter data point of the first data point is not in a stable state, sending a signal to an emptying unit and a searching unit, wherein the emptying unit empties the filtering bandwidth queue, and the searching unit searches the first data point in the stable state again;
a first counting unit 325, which counts the queue length of the filtering bandwidth queue constructed by the construction unit, and sends a signal to the bandwidth obtaining unit when the queue length reaches m;
the bandwidth obtaining unit 326 receives the signal of the first counting unit, and sends a signal to the emptying unit and the searching unit by taking the difference between the maximum value and the minimum value in the filter bandwidth queues built by the building unit as the filter bandwidth, wherein the emptying unit empties the filter bandwidth queue with the queue length of m in the building unit, and the searching unit re-searches the first data point in a steady state in the data points of the model output time sequence after the filter bandwidth queue with the queue length of m is added.
In a specific embodiment of the invention, a passivation filtering process is added between the temperature field calculation result and the process formulation model, and passivation filtering processing is performed on the temperature field calculation result (referred to as "calculation result" for short) required by process formulation. The calculation results include the casting slab nodal temperatures, solid phase ratios, solidification ends (the distance from the position of the solidification end of the casting slab to the meniscus of the mold), reduction start positions and reduction end positions, and the like.
The fluctuation of the temperature field calculation result is within a certain bandwidth and is regularly found, that is, the fluctuation in a steady state is regular fluctuation in a certain range and has periodicity, the calculation result data in the steady state is selected, the maximum value Bmax and the minimum value Bmin which are larger than the data point n in the whole period are counted, the bandwidth parameter of the calculation result under the model condition is obtained, and the solidification end change of the calculation result is taken as an example:
in steps S1 to S3, as shown in fig. 5, under the condition that parameters such as the pull rate are not changed, the time series change of the coagulation end calculated by the dynamic depression model has a periodic rule and fluctuates in a range, there are 21 data points in a whole change period in fig. 5, that is, n =21, as long as the number of consecutive data points is greater than 21 points, the maximum value Bmax and the minimum value Bmin are taken to obtain the filter bandwidth at this time, where the maximum value Bmax of the bandwidth is 14.34m, the minimum value Bmin of the bandwidth is 14.2, and the filter bandwidth is 14.34-14.2= 4.14m. Meanwhile, if the bandwidth target value Aim is the average value rule of the maximum value and the minimum value, the solidification end filtering value is Aim = (14.2 + 14.34)/2 =14.27m no matter how the solidification end calculation result oscillates at the moment, and the filtering value is always kept unchanged.
In step S4, in the steady state, taking the freezing end as an example, as shown in fig. 6a, the freezing end is within the filtering bandwidth, and as shown in fig. 6b, the freezing end filtering value is always the bandwidth target value Aim. Therefore, the stability of the calculation result under the steady state condition is ensured, the stability accords with the actual condition, and the fluctuation noise caused by the temperature field calculation model processing is safely and completely removed.
In an unsteady state, the filter bandwidth is kept unchanged (i.e., the difference between the maximum value and the minimum value of the bandwidth is unchanged, and the passivation range is unchanged), once the calculation result exceeds the existing filter bandwidth range, the filter bandwidth is moved, if the calculation result is greater than the maximum value of the filter bandwidth, the whole filter bandwidth is moved to the maximum value of the filter bandwidth as the calculation result, as shown in fig. 7, taking the freezing end filtering as an example, if the freezing end calculated in the present period is greater than the filter bandwidth in the previous period, the whole filter bandwidth is moved so that the maximum value of the filter bandwidth is the freezing end calculated in the present period, and the filter value is the bandwidth target value Aim. And if the calculation result is smaller than the minimum value of the bandwidth, the bandwidth is integrally moved to the minimum value of the bandwidth to serve as the calculation result, and in both cases, the bandwidth target value Aim is taken as a filtering value.
Or in the unsteady state, the bandwidth is kept unchanged (i.e. the difference between the maximum value and the minimum value of the bandwidth is unchanged, and the passivation range is unchanged), once the calculation result exceeds the existing filter bandwidth, the filter bandwidth is moved to a bandwidth target value as the calculation result, and the calculation result is used as a filter value, as shown in fig. 8, taking the freezing end filtering as an example, if the freezing end calculated in the present period is greater than the bandwidth of the previous period, the bandwidth is moved as a whole, so that the bandwidth target value Aim is the freezing end calculated in the present period, and the filter value is the bandwidth target value Aim.
The filtering method can ensure the stable process under the steady state condition, can ensure the instantaneity of change under the unsteady state condition, and accords with the actual condition.
Aiming at the same variable, a fixed bandwidth is only suitable for a certain value range, and passivation filtering processing in the variable full value range can be realized through online bandwidth self-adaption. The specific method comprises the following steps: in the online operation of the model, a data queue of the calculation result of the temperature field is established, the number of the data queue is larger than the data point n in the whole period under the steady state condition, the real-time bandwidth parameter can be obtained by counting the maximum value Bmax and the minimum value Bmin of the calculation result in the data queue in real time, thereby realizing the online statistics of the bandwidth parameter, realizing the online bandwidth changing function, meeting the online bandwidth self-adapting capability and realizing the bandwidth self-adapting method, which comprises the following steps:
step 1: establishing an output time sequence (such as a solid phase rate time sequence and a solidification tail end position time sequence) of the online controllable model, wherein the length of the sequence is set to be m, and m is greater than the number of data points in the whole period (such as the number of data points in the period with a periodic change rule in a steady state) n;
step 2: collecting continuous steady state output data: under the online condition, each moment is compared with the previous moment, and if the moment is in a stable state, the output result at the moment is added into an output time sequence; if the temperature field is in an unsteady state (namely the parameters used by the temperature field calculation model at the moment are different from those used by the temperature field calculation model at the previous moment), clearing the output time sequence;
and 3, step 3: judging that the length of the output time sequence is m, representing that the number of the collected continuous stable data points reaches m and is greater than the number n of the data points in the whole period, and entering a step 4;
and 4, step 4: solving a maximum value and a minimum value in the output time sequence, wherein the difference value between the maximum value and the minimum value is the latest filtering bandwidth, and the updated bandwidth parameter is the latest bandwidth parameter and is used for processing subsequent solid phase rate filtering; while emptying the output time series queue.
In one embodiment of the invention, the filtering effect of the invention was verified using 180 × 1600mm slabs, wherein the model calculation period was 3s, the slab thickness in the direction of drawing was 20mm, the calculation steel type was selected as Q235, and the secondary cooling process was forced cooling.
The time series of the frozen end was subjected to passivation filtering, and the results were as follows:
fig. 9 is a graph of the temperature field before and after filtering at steady state, and it can be seen from fig. 9 that the temperature field briquetting process resulted in a fluctuation of the solidification end in the range of 14.2m to 14.34m at steady state (stable parameters such as pull rate), and that the periodicity was significant, the data point n for one whole period was 21, the filter bandwidth was 0.14m, and the bandwidth target Aim was the average of the maximum and minimum values of the bandwidth, and the obtained results of the solidification end passivation filter remained stable as shown in fig. 9.
The passivation filtering result of the unsteady state (for example, the pulling rate is changed) is shown in fig. 10, the passivation filtering result of the solidification end has no oscillation, and can reflect the normal solidification change trend in time without hysteresis.
Fig. 11 to 13 show the processing results over a long period of time, and it is apparent from the graphs that after the passivation filtering processing, the solidification end can be kept stable in a steady state (with constant pull rate), and the solidification end can be truly reflected in an unsteady state (with change in pull rate), without delay of change, and the change rate of the pull rate does not affect the filtering processing result.
The passivation filtering treatment is carried out on the solid fraction time sequence, and the results are as follows:
the central solid phase rate of the casting blank at the position of the rolling roller is generally a main parameter for determining the reduction position and the reduction amount, the central solid phase rate of the casting blank at different reduction positions obtained by a calculation model of a casting blank temperature field is a fluctuation trend, fig. 14 shows a coordinate graph of the solid phase rate of the casting blank at two different positions (which can be any two positions, such as the center of the casting blank and the thickness of the casting blank 1/4) along with the change of time, if the reduction process is established by adopting the solid phase rate in fig. 14, the reduction position and the reduction amount are obviously a continuous process, and a reduction control system and equipment are always in a continuous adjustment process, which is obviously not beneficial to the control system and equipment.
Fig. 15 is a graph of a comparison coordinate before filtering and after passivation filtering of a casting blank solid state rate time series at one position in a short time, and fig. 16 is a graph of a comparison coordinate before filtering and after passivation filtering of a casting blank solid state rate time series at one position in a long time, wherein the filtering bandwidth is 0.031, and the number n of data points in a bandwidth period is 21; fig. 17 is a comparison graph of the solid fraction time series of the casting blank before filtering and after passivation filtering at another position, wherein the filtering bandwidth is 0.015, and the number n of data points of the bandwidth cycle is 21. From fig. 15 to 17, it can be seen that after the passivation filtering process, the solid phase ratios at two positions can be kept stable in a steady state (the pulling rate is not changed), while the solid phase ratio change can be truly reflected in an unsteady state (the pulling rate is changed), without delay of the change, and the change rate of the pulling rate does not affect the filtering process result.
As can be seen from fig. 13, for the coagulated end, the filtering requirement can be satisfied with a fixed bandwidth (0.14 m bandwidth) in different value ranges. As can be seen from fig. 15 and 16, the number n of data points in the bandwidth period is the same for the solid phase ratios at two different positions, and is 21, but the most suitable filter bandwidths are different due to different ranges of the solid phase ratio value ranges, and it can also be found from fig. 16 that the filtering of the solid phase ratio at one position also shows different ranges of the solid phase ratio value ranges, and the optimal filter bandwidths are different.
In the above embodiments of the present invention, the bandwidth period is set as the number of data points, but the present invention is not limited to this, and the bandwidth period may be set time length or time length of one fluctuation period in which the model output time series fluctuates periodically.
While the foregoing disclosure shows illustrative embodiments of the invention, it should be noted that various changes and modifications could be made herein without departing from the scope of the invention as defined by the appended claims. The functions, steps and/or actions of the method claims in accordance with the inventive embodiments described herein need not be performed in any particular order. Furthermore, although elements of the invention may be described or claimed in the singular, the plural is contemplated unless limitation to a single element is explicitly stated.
Claims (10)
1. A filtering method for realizing the process stability of an online control model is characterized by comprising the following steps:
setting a bandwidth period, wherein the bandwidth period is a set number of data points;
collecting data points output by the online control model along with time;
taking a steady-state data point output by an online control model with a bandwidth period not less than time to obtain the maximum value and the minimum value of the steady-state data point, wherein the difference between the maximum value and the minimum value is a filtering bandwidth;
filtering each data point of the online control model by using the filtering bandwidth, including:
setting the maximum value and the minimum value of the initial bandwidth as any value under the condition that the difference between the maximum value and the minimum value of the initial bandwidth is the filtering bandwidth;
judging whether the value of a first data point output by the online control model along with time is within the range of the maximum value of the initial bandwidth and the minimum value of the initial bandwidth;
if the first data point is in the range, filtering the value of the first data point into a bandwidth target value, wherein the bandwidth target value is any value in the range of the initial bandwidth maximum value and the initial bandwidth minimum value;
if the first data point is not in the range, integrally translating the filtering bandwidth, so that the numerical value of the first data point is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth, and filtering the numerical value of the first data point into a translated bandwidth target value, wherein the translated bandwidth target value is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth;
and repeating the steps, and sequentially filtering the data points output by the online control model along with the time.
2. The filtering method for realizing the process stability of the online control model according to claim 1, wherein the bandwidth target value is set as a bandwidth minimum value, or a bandwidth maximum value, or an average value of the bandwidth maximum value and the bandwidth minimum value, or a golden section point of the bandwidth minimum value and the bandwidth maximum value.
3. The filtering method for realizing the process stability of the online control model according to claim 1, wherein the method for setting the bandwidth period comprises the following steps:
judging whether the model output time sequence has periodic fluctuation in a steady state;
if no periodic fluctuation exists, setting a bandwidth period, wherein the bandwidth period is the number of data points with a set quantity;
and if the periodic fluctuation exists, determining a bandwidth period according to the periodic fluctuation, wherein the bandwidth period is the number of data points of one fluctuation period of the periodic fluctuation.
4. The filtering method for realizing the process stability of the online control model according to claim 1, further comprising:
acquiring a parameter time sequence of parameters for a temperature field model along with time, wherein the parameters for the temperature field model comprise a pulling speed, a tundish temperature, a crystallizer water quantity, a temperature difference and a secondary cooling water quantity, the parameters for the temperature field model are taken as a stable state when not changed, and the parameters for the temperature field model are taken as an unstable state when changed;
and acquiring a model output time sequence of an online control model along with time corresponding to the parameter time sequence, wherein the online control model comprises one or two of a dynamic secondary cooling model and a dynamic reduction model, and the model output comprises one or more of solid phase rate, surface temperature, shell thickness and solidification tail end position at different positions of a casting blank.
5. The filtering method for realizing the process stability of the online control model according to claim 1, wherein the filtering bandwidth is adaptively changed according to a model output time sequence of the online model output along with time, and the method comprises the following steps:
setting the queue length of a filtering bandwidth queue as m, wherein m is greater than the number n of data points in a bandwidth period;
searching a first data point of the model output time sequence in a steady state, and adding the first data point into the filtering bandwidth queue;
judging whether a data point subsequent to the first data point in the model output time series is also in a steady state;
when the latter data point of the first data point is also in a steady state, adding the latter data point into the filter bandwidth queue;
when the next data point of the first data point is not in a steady state, emptying the filtering bandwidth queue, and searching the first data point in the steady state again;
repeating the steps until the queue length of the filter bandwidth queue reaches m, and taking the difference value between the maximum value and the minimum value in the filter bandwidth queue as the filter bandwidth;
and clearing the filter bandwidth queue with the queue length reaching m, returning to the step of searching the first data point in the steady state, and updating the filter bandwidth.
6. The filtering method for realizing on-line control model process stability according to claim 1, wherein the bandwidth period is a set time length or a time length of one fluctuation period of periodic fluctuation of the model output time series.
7. The utility model provides a realize filtering system of online control model process stability which characterized in that includes:
the bandwidth period setting module is used for setting a bandwidth period, wherein the bandwidth period is a set number of data points;
the acquisition module acquires data points output by the online control model along with time;
the filtering bandwidth obtaining module is used for obtaining a steady-state data point output by the online control model with the bandwidth period not less than the time to obtain the maximum value and the minimum value of the steady-state data point, and the difference between the maximum value and the minimum value is the filtering bandwidth;
a filtering module for filtering each data point of the online control model using the filtering bandwidth,
wherein the filtering module comprises:
a fourth setting unit, setting the maximum value of the initial bandwidth and the minimum value of the initial bandwidth as any value under the condition that the difference between the maximum value of the initial bandwidth and the minimum value of the initial bandwidth is the filtering bandwidth;
the third judging unit is used for judging whether the value of the first data point output by the online control model along with the time is in the range of the maximum value of the initial bandwidth and the minimum value of the initial bandwidth; if the first data point is in the range, sending a signal to a filtering unit, and if the first data point is not in the range, sending a signal to a translation unit;
the translation unit is used for translating the whole filtering bandwidth to ensure that the numerical value of the first data point is in the range of the maximum bandwidth value and the minimum bandwidth value corresponding to the translated filtering bandwidth, and sending signals to the filtering unit and the fourth setting unit, and the fourth setting unit is used for updating the maximum initial bandwidth value and the minimum initial bandwidth value according to the signals of the translation unit;
the filtering unit is used for receiving the signal of the judging unit and filtering the numerical value of the first data point into a bandwidth target value, wherein the bandwidth target value is any value in the range of the initial bandwidth maximum value and the initial bandwidth minimum value; receiving a signal of a translation unit, and filtering the value of the first data point into a translated bandwidth target value, wherein the translated bandwidth target value is within a bandwidth maximum value and a bandwidth minimum value range corresponding to the translated filtering bandwidth;
and the second counting unit counts the data points filtered by the filtering unit and sends the next data point serving as the first data point to the third judging unit.
8. The filtering system according to claim 7, wherein the bandwidth target value is set to a bandwidth minimum, or a bandwidth maximum, or an average of a bandwidth maximum and a bandwidth minimum, or a golden section point of a bandwidth minimum and a bandwidth maximum.
9. The filtering system according to claim 7, wherein the bandwidth period setting module comprises:
the first judging unit is used for judging whether the model output time sequence has periodic fluctuation in a steady state or not, and sending a signal to the first setting unit if the model output time sequence does not have the periodic fluctuation; if the periodic fluctuation exists, sending a signal to a second setting unit;
the device comprises a first setting unit, a second setting unit and a control unit, wherein the first setting unit sets a bandwidth period which is a set number of data points;
and the second setting unit is used for determining a bandwidth period according to the periodic fluctuation, wherein the bandwidth period is the number of data points of one fluctuation period of the periodic fluctuation.
10. Filtering system according to claim 7,
the filtering bandwidth obtaining module includes a fixed bandwidth setting part or a bandwidth adaptation part,
the fixed bandwidth setting part randomly selects continuous steady-state data points output by the online control model with the bandwidth period not less than the bandwidth period along with time to obtain the maximum value and the minimum value of the continuous steady-state data points, the difference between the maximum value and the minimum value is a filtering bandwidth, and the filtering module adopts the filtering bandwidth obtained by the fixed bandwidth setting part to filter each data point output by the online control model along with the time with the fixed bandwidth;
wherein the bandwidth adaptation part includes:
a third setting unit, setting the queue length of the filter bandwidth queue as m, wherein m is greater than the number n of data points in the bandwidth period;
the searching unit is used for searching a first data point of the model output time series in a steady state;
the construction unit is used for adding the first data point into the filtering bandwidth queue;
the second judging unit is used for judging whether a data point behind the first data point in the model output time sequence is also in a stable state or not, and sending a signal to the constructing unit when the data point behind the first data point is also in the stable state, wherein the constructing unit also adds the data point behind the first data point into the filtering bandwidth queue; when the latter data point of the first data point is not in a stable state, sending a signal to an emptying unit and a searching unit, wherein the emptying unit empties the filtering bandwidth queue, and the searching unit searches the first data point in the stable state again;
the first counting unit is used for counting the queue length of the filtering bandwidth queue constructed by the construction unit, and sending a signal to the bandwidth obtaining unit when the queue length reaches m;
and the bandwidth obtaining unit is used for sending a signal to the emptying unit and the searching unit by taking the difference value between the maximum value and the minimum value in the filtering bandwidth queues built by the building unit as the filtering bandwidth after receiving the signal of the first counting unit, wherein the emptying unit empties the filtering bandwidth queue with the queue length reaching m in the building unit, and the searching unit re-searches the first data point in a steady state in the data points of the model output time sequence after the filtering bandwidth queue with the queue length reaching m is added.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910027096.3A CN109902335B (en) | 2019-01-11 | 2019-01-11 | Filtering method and system for realizing on-line control model process stability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910027096.3A CN109902335B (en) | 2019-01-11 | 2019-01-11 | Filtering method and system for realizing on-line control model process stability |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109902335A CN109902335A (en) | 2019-06-18 |
| CN109902335B true CN109902335B (en) | 2023-02-28 |
Family
ID=66943618
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910027096.3A Active CN109902335B (en) | 2019-01-11 | 2019-01-11 | Filtering method and system for realizing on-line control model process stability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109902335B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112686539A (en) * | 2020-12-29 | 2021-04-20 | 华润电力技术研究院有限公司 | Performance calculation method, device and equipment of thermodynamic system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102353957A (en) * | 2011-09-15 | 2012-02-15 | 哈尔滨工程大学 | Method for processing multi-beam bathymetric data based on variable-bandwidth filter |
| EP2474873A1 (en) * | 2011-01-11 | 2012-07-11 | Siemens Aktiengesellschaft | Method and device for filtering a signal and regulating device for a process |
| CN109067381A (en) * | 2018-07-05 | 2018-12-21 | 无锡北微传感科技有限公司 | A kind of Real-Time Filtering system and method for MEMS gyroscope random noise |
-
2019
- 2019-01-11 CN CN201910027096.3A patent/CN109902335B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2474873A1 (en) * | 2011-01-11 | 2012-07-11 | Siemens Aktiengesellschaft | Method and device for filtering a signal and regulating device for a process |
| CN102353957A (en) * | 2011-09-15 | 2012-02-15 | 哈尔滨工程大学 | Method for processing multi-beam bathymetric data based on variable-bandwidth filter |
| CN109067381A (en) * | 2018-07-05 | 2018-12-21 | 无锡北微传感科技有限公司 | A kind of Real-Time Filtering system and method for MEMS gyroscope random noise |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109902335A (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902335B (en) | Filtering method and system for realizing on-line control model process stability | |
| CN109996087B (en) | Code rate self-adaption method and device for live video broadcast based on finite state machine | |
| CN107423433B (en) | Data sampling rate control method and device | |
| EP2081326B1 (en) | Statistical processing apparatus capable of reducing storage space for storing statistical occurence frequency data and a processing method therefor | |
| CN1050550C (en) | Method for the continuous casting of high-carbon steels | |
| CN110312131B (en) | Content self-adaptive online video coding method based on deep learning | |
| JP2005227972A (en) | Traffic situation prediction device and method | |
| JP3591422B2 (en) | Level control method and level control device for continuous casting machine | |
| CN114985697B (en) | Crystallizer liquid level fluctuation monitoring method and system | |
| CN115277055B (en) | Method and device for guaranteeing data interaction safety of Internet of things and electronic equipment | |
| EP0819033B1 (en) | Device for early detection of run-out in continuous casting | |
| CN110930065A (en) | Casting blank quality determination method and system | |
| CN117872984A (en) | Performance evaluation method of tobacco shred drying process control system based on data driving | |
| CN113890764A (en) | Time synchronization system with prediction function and safety monitoring method and device thereof | |
| Vergados et al. | A fuzzy controller for rate adaptation in MPEG-DASH clients | |
| CN102207497A (en) | Method for forecasting slag inclusion situations of surfaces of continuous casting slabs before rolling | |
| CN116542153A (en) | Reverse design method of metal material processing technological parameters based on reinforcement learning | |
| CN113134587B (en) | Method for judging water gap blockage and dissolution loss through stopper rod opening degree variation trend | |
| CN114925313A (en) | Self-adaptive method and system based on distributed link tracking dynamic sampling rate | |
| CN112747742A (en) | Terminal position self-adaptive updating method based on Kalman filtering | |
| CN114264786B (en) | Casting blank quality evaluation method and system based on continuous casting tundish immersion nozzle falling-off nodulation | |
| CN112077273B (en) | Plate blank quality judgment and evaluation system and method with variable pulling speed | |
| CN111273625B (en) | Feedforward calculation method and device for dynamic matrix control | |
| CN117311300B (en) | Method and device for dynamically adjusting sampling frequency of distributed control system and electronic equipment | |
| CN116237484B (en) | Judgment method for crystallizer immersed nozzle nodulation and molten steel bias flow |
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 |