CN117170422A - Intelligent regulation and control system for position and posture of stirring device - Google Patents
Intelligent regulation and control system for position and posture of stirring device Download PDFInfo
- Publication number
- CN117170422A CN117170422A CN202311446019.4A CN202311446019A CN117170422A CN 117170422 A CN117170422 A CN 117170422A CN 202311446019 A CN202311446019 A CN 202311446019A CN 117170422 A CN117170422 A CN 117170422A
- Authority
- CN
- China
- Prior art keywords
- period
- periodic
- target
- mode
- pressure
- 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.)
- Granted
Links
- 238000003756 stirring Methods 0.000 title claims abstract description 155
- 230000000737 periodic effect Effects 0.000 claims abstract description 187
- 239000007788 liquid Substances 0.000 claims abstract description 40
- 230000001105 regulatory effect Effects 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims description 30
- 239000013598 vector Substances 0.000 claims description 28
- 238000010606 normalization Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 4
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 101100272279 Beauveria bassiana Beas gene Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of stirring device control systems, and provides an intelligent regulation and control system for the position and the posture of a stirring device, which comprises the following components: distributing pressure sensors on blades of the stirring device, and collecting pressure data of each blade for a plurality of time periods; according to the periodic distribution in the pressure data of each period, a plurality of periodic modes and initial periodicity thereof are obtained, the first periodicity of each periodic mode is obtained according to the variation trend of the periodic modes, the stirring periodicity of each period is obtained, and the period to be regulated of each blade is obtained; according to the first similarity of the same time periods to be adjusted of different blades, obtaining stirring adjustment speed of the adjustment time periods, and obtaining second periodicity and second similarity of each adjusted time period; and judging the liquid mixing state in the device according to the second periodicity and the second similarity of the adjusted period, and completing the intelligent regulation and control of the stirring device. The invention aims to control the working state of the stirring device through pressure data change on different stirring blades.
Description
Technical Field
The invention relates to the field of stirring device control systems, in particular to an intelligent regulation and control system for the position and the posture of a stirring device.
Background
The stirring device is characterized in that the stirring blades are rotated to drive the liquid to be stirred in the device, and materials in the machine are uniformly mixed according to a certain circulating flow direction, so that the stirring of the liquid is completed; for liquid stirring, the aim of uniformly mixing the properties such as concentration, temperature and density is needed, so that the position and the gesture of an internal working element are required to be timely controlled by the stirring device, and the uniform stirring of the liquid is completed through intelligent regulation and control of the position and the gesture of the stirring device.
In the existing method, the stirring device is usually stirred for a fixed time, and the stirring is stopped when the stirring time reaches the fixed time, however, the types of the liquid in the stirring process are various, the type, the concentration and the density of the mixed liquid cannot be ensured to be consistent in each stirring process, even a large difference exists between the mixed liquids, and further the stirring effect cannot be ensured to be uniform due to the limitation of the fixed stirring time; therefore, the uniformity degree of mixed liquid in different areas in the stirring device is reflected according to the change of pressure data on the stirring blades in the stirring process, the stirring blades are controlled according to the uniformity degree, the intelligent regulation and control of the position and the posture of the stirring device are completed, and a good liquid mixing effect is achieved.
Disclosure of Invention
The invention provides an intelligent regulation and control system for the position and the posture of a stirring device, which solves the problem that the existing fixed stirring time can not ensure that uniform stirring effect can be achieved for each liquid, and adopts the following technical scheme:
the embodiment of the invention provides an intelligent regulation and control system for the position and the posture of a stirring device, which comprises the following components:
the blade sensor data acquisition module is used for distributing pressure sensors on blades of the stirring device and acquiring pressure data of each blade in a plurality of time periods;
blade stirring speed regulation and control module: according to the first periodicity of each periodic mode in the pressure data of each period of each blade, obtaining the stirring periodicity of each period of each blade, and obtaining the period to be regulated of each blade through threshold screening;
according to the first similarity of the same time periods to be adjusted of different blades, obtaining the stirring adjustment speed of the adjustment time periods, and obtaining the second periodicity and the second similarity of each adjusted time period;
and the stirring device control stopping module judges the liquid mixing state in the device according to the second periodicity and the second similarity of the adjusted time period, and completes the intelligent regulation and control of the stirring device.
Further, the specific method for obtaining the stirring periodicity of each blade in each period comprises the following steps:
according to the periodic distribution in the pressure data of each time period of each blade, a plurality of periodic modes and a first periodicity of each time period of each blade are obtained; taking the pressure data of any one blade in any period as target pressure data, marking a pressure curve corresponding to the target pressure data as a target pressure curve, and performing STL decomposition on the target pressure curve to obtain a periodic term curve; taking the maximum value of the first periodicity in a plurality of periodic modes in the target pressure data as the comprehensive periodicity of the target pressure data;
acquiring the ordinate ratio of each moment in the target pressure curve and the periodic term curve, wherein the ratio is obtained from a small value to a large value, and taking the product of the average value of the ratio at all the moments and the comprehensive periodicity as the stirring periodicity of the corresponding period of the target pressure data; the stirring periodicity per period of each blade is obtained.
Further, the specific obtaining method includes the steps of:
according to the periodic distribution in the pressure data of each period of each blade, a plurality of periodic modes, a mode pressure curve and initial periodicity of each period of each blade are obtained; taking pressure data of any one blade in any period as target pressure data, taking any one period mode in the target pressure data as a target period mode, calculating a first derivative of a mode pressure curve of the target period mode, obtaining a zero point, taking data points in the mode pressure curve corresponding to the zero point as characteristic points of the target period mode, taking coordinate data of each characteristic point as input of an SVD algorithm, outputting a plurality of characteristic vectors of the mode pressure curve, and recording the characteristic vectors as a plurality of characteristic vectors of the target period mode; acquiring a plurality of characteristic vectors of each periodic mode;
performing KM matching on feature vectors of any two periodic modes in the target pressure data, wherein a plurality of feature vectors of the same periodic mode are used as nodes on one side in the bipartite graph, the edge value between the nodes on the left side and the right side is represented by the similarity of the feature vectors, and the similarity of the feature vectors is obtained by a DTW distance; obtaining a plurality of matching pairs through KM matching, and taking the average value of the side values corresponding to the matching pairs as the trend similarity of the two periodic modes; acquiring trend similarity of any two periodic modes in the target pressure data;
according to the DTW distance of the mode pressure curves of any two periodic modes in the target pressure data, the overall similarity degree of the any two periodic modes is obtained, the average value of the overall similarity degree and the trend similarity degree is used as the comprehensive similarity degree of the any two periodic modes, the average value of the comprehensive similarity degree of each periodic mode and all other periodic modes in the target pressure data is calculated, linear normalization is carried out on all the average values, the obtained result is recorded as the gain coefficient of each periodic mode, and the product of the gain coefficient and the initial periodicity is used as the first periodicity of each periodic mode in the target pressure data.
Further, the method for obtaining the plurality of cycle modes, the mode pressure curves and the initial periodicity of each period of each blade comprises the following specific steps:
according to the periodic distribution in the pressure data of each period of each blade, a plurality of periodic modes of each period of each blade, a plurality of periodic pressure curves of each periodic mode and a plurality of mode pressure curves are obtained;
taking pressure data of any one blade in any period as target pressure data, taking any one periodic mode in the target pressure data as a target periodic mode, and initiating periodicity of the target periodic modeThe calculation method of (1) is as follows:
wherein,indicates the number of cycle pressure curves in the target cycle mode, +.>Representing all cycle pressures in target pressure dataTotal number of curves, ++>The cosine similarity mean value of any two sections of periodic pressure curves in the target periodic mode is represented by +.>Representing cosine similarity variance of any two sections of periodic pressure curves in target periodic mode, +.>Represents an exponential function based on natural constants, < ->To avoid over-parametrics with too small an exponential function result;
an initial periodicity of each periodic pattern in the pressure data for each period of each blade is obtained.
Further, the specific method for obtaining the plurality of periodic patterns, the plurality of periodic pressure curves and the plurality of pattern pressure curves of each periodic pattern of each blade is as follows:
taking pressure data of any one blade in any period as target pressure data, marking a pressure curve corresponding to the target pressure data as a target pressure curve, performing STL decomposition on the target pressure curve to obtain a periodic term curve, performing Fourier transformation on the periodic term curve to a frequency domain space, taking the reciprocal of frequency corresponding to each amplitude in the frequency domain space as one period, and obtaining a plurality of periods of the target pressure data;
dividing a target pressure curve by taking any one period as a target period to obtain a plurality of sections of local pressure curves, respectively calculating cosine similarity of each section of local pressure curve and all other local pressure curves, dividing two sections of local pressure curves corresponding to all cosine similarity larger than a preset first threshold value into the target period, and recording the two sections of local pressure curves as a plurality of sections of period pressure curves of the target period; dividing the periods into a plurality of sections of period pressure curves of each period one by one according to the sequence from small to large, and recording the period with the period pressure curves as a period mode;
taking any one periodic mode as a target periodic mode, acquiring cosine similarity of any two sections of periodic pressure curves in the target periodic mode, obtaining cosine similarity mean value of each section of periodic pressure curve and other sections of periodic pressure curves in the target periodic mode, and recording the periodic pressure curve with the maximum mean value as the mode pressure curve of the target periodic mode; and acquiring a plurality of periodic pressure curves and mode pressure curves of each periodic mode in the pressure data of each period of each blade.
Further, the specific method for obtaining the stirring regulation speed in the regulation period includes the following steps:
the first time interval of all the blades in the time interval is the time interval to be adjusted, and the time interval is recorded as an adjusting time interval; acquiring the stirring periodicity of the pressure data corresponding to each blade in the adjustment period and a periodic mode with the first greatest periodicity, taking the periodic mode with the first greatest periodicity in the pressure data corresponding to each blade as the periodic mode of each blade in the adjustment period, and calculating first similarity of the periodic modes of any two blades in the adjustment period, wherein the first similarity is obtained through the DTW distance between the mode pressure curves of the two periodic modes;
and obtaining the minimum value in all the first similarities, and taking the product of the difference value obtained by subtracting the minimum value from 1 and the maximum stirring speed of the stirring device as the stirring regulation speed of the regulation period.
Further, the method for obtaining the second periodicity and the second similarity of each adjusted time period includes the following specific steps:
recording each time period after the adjustment of the stirring speed as an adjusted time period, taking any one of the adjusted time periods as a target adjusted time period, acquiring pressure data of each blade in the target adjusted time period, acquiring the stirring periodicity of the pressure data of each blade in the target adjusted time period, and taking the average value of the stirring periodicity of all the blades in the target adjusted time period as the second periodicity of the target adjusted time period;
according to a calculation method of the first similarity of any two blades in the adjustment period, calculating the first similarity of any two blades in the target adjusted period, and taking the average value of all the first similarities as the second similarity of the target adjusted period; a second periodicity and a second similarity for each adjusted time period are obtained.
The beneficial effects of the invention are as follows: according to the invention, the pressure sensors are arranged on the blades in the stirring device, the condition of uniform stirring of the liquid near each blade is reflected by collecting the pressure data on different blades in real time, and meanwhile, the integral liquid mixing effect in the stirring device is obtained in a quantized manner, so that the intelligent regulation and control of the stirring device are realized by regulating the stirring speed; the method comprises the steps of screening to obtain a period to be regulated, in which pressure data in a single period of each blade is periodically distributed, through periodic variation of pressure data in each period of each blade, further obtaining a period to be regulated, in which all blades are in the period to be regulated, quantitatively obtaining stirring regulation speed according to the similarity of periodic patterns of different blades in the period to be regulated, and controlling the stirring device to stop according to the periodic variation and the similarity variation of the periodic patterns of different blades in the stirring device after the stirring speed is regulated; the device has the advantages that the problem that the traditional stirring device works for a fixed time, liquid in the device cannot be uniformly mixed, the situation that liquid is uniformly mixed near a single blade and the whole liquid is the target effect is avoided, the intelligent regulation and control instantaneity of the stirring device is improved, and the mixing effect of liquid in the stirring device is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a block diagram of an intelligent control system for position and posture of a stirring device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a block diagram of an intelligent control system for position and posture of a stirring device according to an embodiment of the invention is shown, where the system includes:
the blade sensor data acquisition module 101 is used for distributing pressure sensors on the blades of the stirring device and acquiring pressure data of each blade for a plurality of time periods.
The purpose of the embodiment is to analyze the liquid mixing state in the device through sensor data arranged on the blades in the stirring device, so as to realize intelligent regulation and control of the stirring device, so that a pressure sensor is firstly arranged on each blade in the stirring device, and the pressure data on the blades are acquired in real time; from the beginning of the stirring device, the embodiment analyzes every 20 minutes as a period, the pressure data is acquired by the pressure sensor on each blade in each period, the pressure data sampling time interval is set to be 10 seconds, and the real-time intelligent regulation and control of the stirring device is realized by taking every 20 minutes as a period through real-time acquisition, so that the pressure data which are expressed as time sequence sequences in a plurality of periods and each period can be obtained.
So far, pressure data are acquired in real time by arranging pressure sensors on the blades, and the pressure data of each blade in a plurality of time periods are obtained.
Blade stirring speed regulation module 102:
it should be noted that, the mixing liquid is just started in the stirring process, but the demarcation between different liquids is more obvious, especially the liquid with larger viscosity, and the stirring result is that the mixing liquid is fully mixed through the stirring process and is uniformly mixed; from the angle of the blades, when the pressure curve at each blade shows periodic variation and the pressure curves at different blades are similar in shape, a better stirring effect is achieved; when the stirring effect is good, the reason for the periodic variation of the pressure curve at the blade is: for sufficient stirring, the mixed liquid and the like in the stirring device often do not occupy the whole space and a certain space is reserved, so that the blades in the stirring device have only two states: in the unmixed liquid and in the mixed liquid, the blades are in continuous circular motion at a certain speed, so that the pressure curve at the blades shows periodic variation.
(1) According to the periodic distribution in the pressure data of each period, a plurality of periodic modes and initial periodicity thereof are obtained, the first periodicity of each periodic mode is obtained according to the change trend of the periodic modes, the stirring periodicity of each period is obtained, and the period to be adjusted of each blade is obtained through threshold screening.
It should be noted that, the pressure data of each period of each blade is a time sequence, the time sequence may be represented as a pressure curve, a periodic term curve is obtained by STL decomposition, according to the distribution of different periodic modes in the periodic term curve, the similarity between the variation trends of different periodic modes is combined, and further the stirring periodicity of the corresponding period is obtained quantitatively, the stirring periodicity of the corresponding period is obtained by the periodicity final quantization of a plurality of periodic modes, the periodic variation of the pressure data in the corresponding period of the blade can be reflected better, the period to be adjusted is obtained by threshold screening, that is, the stirring effect of the blade under the period is better, the pressure curve needs to be corrected further, and the part with higher reliability of the periodic term is reserved, so as to obtain the pressure correction curve.
It should be further noted that, since the periodicity is weaker immediately after stirring, the pressure data is decomposed by STL to often obtain multiple periods, and as the stirring proceeds, the periodicity of a certain period becomes stronger, and other periods become gradually weaker or even disappear, so that the stirring periodicity of the period corresponding to the blade can be obtained by comparing the period with other periods, and when the stirring periodicity is larger, the stirring stability at each blade is better; the stirring periodicity is larger, namely the corresponding frequency of the period is the largest, the corresponding frequencies of other periods are smaller than the corresponding frequency of the period, and the reaction on the pressure data is: the areas with larger similarity are obtained by dividing the pressure data in the period, and the number of the areas is large; meanwhile, if different periodic modes are similar, a gain coefficient needs to be given to each periodic mode, namely, the periodic modes are similar, the similarity of pressure curves of the different periodic modes is larger, and the periodicity of the pressure curves in the period is stronger through the gain coefficient, so that the subsequent processing is facilitated.
Specifically, taking pressure data of any period on any one blade as an example, the pressure data is a time sequence, and can be expressed in a curve form and recorded as a pressure curve; and performing STL decomposition on the pressure curve to obtain a periodic term curve, converting the periodic term curve into a frequency domain space through Fourier transformation, and taking the reciprocal of the frequency corresponding to each amplitude in the frequency domain space as one period to obtain a plurality of periods of the pressure data, wherein the STL decomposition and the Fourier transformation are known techniques, and the embodiment is not repeated.
Further, taking any period as an example, dividing the pressure curve through the period to obtain a plurality of sections of local pressure curves, and calculating cosine similarity with all other local pressure curves for each section of local pressure curve respectively; setting a preset first threshold value for judging the belonging relation between the local pressure curve and the period, wherein the preset first threshold value is described by adopting 0.7, dividing two sections of local pressure curves corresponding to all cosine similarities larger than the preset first threshold value into the period, and marking the period pressure curves as a plurality of sections of period pressure curves of the period, and simultaneously ensuring that the cosine similarities of any two sections of period pressure curves in the period are larger than the preset first threshold value; dividing the pressure curves one by one according to the sequence from small to large to obtain a plurality of sections of cycle pressure curves of each cycle, and recording the cycle with the cycle pressure curve as a cycle mode, wherein when the acquired cycle pressure curve of each cycle completely divides the pressure curve, the cycle division is stopped, and meanwhile, the acquired cycle pressure curve does not participate in the division of the pressure curve of other cycles; taking any one periodic mode as an example, acquiring cosine similarity of any two sections of periodic pressure curves in the periodic mode, acquiring cosine similarity mean value of each section of periodic pressure curve and other sections of periodic pressure curves in the same periodic mode, recording the periodic pressure curve with the largest mean value as the mode pressure curve of the periodic mode, and acquiring the mode pressure curve of each periodic mode of the pressure data according to the method.
Further, for the pressure data, the initial periodicity of any one of the periodic patternsThe calculation method of (1) is as follows:
wherein,indicates the number of periodic pressure curves in the periodic mode, +.>Indicating the total number of all cycle pressure curves resulting from the pressure curve, +.>Representing the cosine similarity mean value of any two sections of periodic pressure curves in the periodic mode, and (I)>Representing cosine similarity variance of any two sections of periodic pressure curves in the periodic mode, +.>Represents an exponential function based on natural constants, < ->To avoid hyper-parameters with too small an exponential function result, the present embodiment uses +.>Description is made; the greater the ratio of the number of periodic pressure curves of the periodic pattern to the total number, the greater the initial periodicity of the periodic pattern; the larger the average value of the similarity between the periodic pressure curves in the periodic mode is, the larger the initial periodicity is; the smaller the similarity variance, the larger the initial periodicity, this embodiment is achieved by +.>The inverse proportion relation and normalization processing are presented, and an implementer can set an inverse proportion function and a normalization function according to actual conditions; the initial periodicity of each periodic pattern in the pressure data is obtained as described above.
Further, taking any one periodic mode as an example, calculating a first derivative of a mode pressure curve of the periodic mode and obtaining a zero point, wherein the zero point of the first derivative is a change trend inflection point of the mode pressure curve, taking data points in the mode pressure curve corresponding to the zero point as characteristic points of the periodic mode, obtaining a plurality of characteristic points of the periodic mode, taking coordinate data of each characteristic point as input of an SVD algorithm, outputting to obtain a plurality of characteristic vectors of the mode pressure curve, and recording the characteristic vectors as a plurality of characteristic vectors of the periodic mode, wherein the SVD algorithm is singular value decomposition, and is a known technology, and the embodiment is not repeated; according to the method, a plurality of characteristic vectors of each periodic mode are obtained, for any two periodic modes, the plurality of characteristic vectors of one periodic mode are used as nodes on the left side of the bipartite graph, the plurality of characteristic vectors of the other periodic mode are used as nodes on the right side of the bipartite graph, the side values between the nodes on the left side and the right side are represented by the similarity of the characteristic vectors, and the calculation method of the similarity of the characteristic vectors is as follows: because the number of dimensions in the feature vectors of different periodic modes is different, the DTW distance of the feature vectors corresponding to the nodes at the left side and the right side is acquiredSimilarity as feature vector, wherein +.>DTW distance representing feature vector, +.>Represents an exponential function based on natural constants, < ->To avoid hyper-parameters with too small an exponential function result, the present embodiment uses +.>To describe, the present embodiment is by +.>The inverse proportion relation and normalization processing are presented, and an implementer can set an inverse proportion function and a normalization function according to actual conditions; after obtaining the edge values between the nodes at the left side and the right side, a plurality of matching pairs can be obtained through KM matching, namely, the successfully matched node pairs are obtained, and the average value of the edge values corresponding to the matching pairs is used as the trend similarity of the two periodic modes; and obtaining the trend similarity degree of any two periodic modes in the pressure data according to the method.
Further, taking any two periodic modes as an example, calculating the DTW distance of the mode pressure curves of the two periodic modes willAs an overall approximation of the two periodic patterns, wherein +.>DTW distance representing mode pressure curve, +.>Represents an exponential function based on natural constants, < ->To avoid too small an exponential function resultNumber, this embodiment uses +.>To describe, the present embodiment is by +.>The inverse proportion relation and normalization processing are presented, and an implementer can set an inverse proportion function and a normalization function according to actual conditions; after the overall similarity is obtained, taking the average value of the overall similarity and the trend similarity as the comprehensive similarity of the two periodic modes; according to the method, the comprehensive similarity degree of any two periodic modes is obtained, the average value of the comprehensive similarity degree of each periodic mode and all other periodic modes in the pressure data is calculated, linear normalization is carried out on all the average values, the obtained result is recorded as the gain coefficient of each periodic mode, and the product of the gain coefficient and the initial periodicity is used as the first periodicity of each periodic mode in the pressure data.
Further, for the pressure data, after the first periodicity of each periodic pattern is obtained, taking the maximum value of the first periodicity as the comprehensive periodicity of the pressure data; the method comprises the steps of obtaining the ratio of a pressure data value at each moment to corresponding periodic item data in a pressure curve corresponding to the pressure data, namely, the ordinate ratio of the pressure curve to the same moment in the periodic item curve, wherein the ratio is obtained by a small value to a large value, and taking the product of the average value of the ratio at all moments and the comprehensive periodicity as the stirring periodicity of the corresponding period of the pressure data, wherein the more obvious the periodic variation is, the larger the ratio is, and the less the comprehensive periodicity needs to be regulated; setting a preset second threshold value for judging the stirring effect of the corresponding period, wherein the preset second threshold value is described by adopting 0.8, and if the stirring periodicity is larger than the preset second threshold value, marking the period corresponding to the pressure data as a period to be regulated of the corresponding blade, wherein the period to be regulated indicates that the pressure data of the blade has obvious periodic variation in the period; according to the method, the pressure data of each time period of each blade are judged, and a plurality of time periods to be adjusted of each blade are finally obtained, and the fact that after the time periods to be adjusted appear in each blade due to the fact that the pressure data are collected in real time, each time period is the time period to be adjusted, and the follow-up processing process is based on the fact that each blade is in the time period to be adjusted and then processed.
So far, through the periodic variation analysis of the pressure data, a plurality of time periods to be adjusted of each blade are obtained.
(2) According to the first similarity of the same time periods to be adjusted of different blades, the stirring adjustment speed of the adjustment time periods is obtained, and the second periodicity and the second similarity of each adjusted time period are obtained.
It should be noted that, when the pressure curve at each blade is relatively strong in periodicity in the stirring process, the overall stirring degree is not necessarily good, at this time, it is possible that only the mixed liquid near each blade is uniformly mixed, but the mixed liquid at different blades is not necessarily uniformly mixed, so that the stirring speed is increased, the mixed liquid at different blades is subjected to flow exchange, and when the pressure curves at different blades are similar at a certain moment, the overall uniformity is good, namely, a good stirring effect is achieved; the smaller the similarity of different blade cycle modes, the more difficult the current stirring speed is to enable mixed liquid in different areas to circulate better, and the greater the speed is required to be adjusted, so that the mixed liquid in different areas can circulate and exchange.
Specifically, after the period to be regulated of each blade is obtained, obtaining a period in which the first all blades are in the period to be regulated, namely, obtaining pressure data according to time sequence in each period, obtaining a period in which the pressure data of the first all blades meet the requirement of the period to be regulated, wherein mixed liquid near each blade in the period is relatively uniform, and recording the period as a regulation period; acquiring a stirring periodicity, a comprehensive periodicity and a first largest periodicity mode of pressure data corresponding to each blade under a regulation period, taking the first largest periodicity mode of pressure data corresponding to each blade as a periodicity mode of each blade in the regulation period, calculating a first similarity of the periodicity modes of any two blades in the regulation period, and calculating a second similarity of the first similarityThe calculation method of the degree comprises the following steps: the periodic modes of the two blades in the adjusting period correspond to the mode pressure curves, and the DTW distance of the two mode pressure curves is calculated to beAs a first similarity of the two leaves in the adjustment period, wherein>DTW distance representing mode pressure curve, +.>Represents an exponential function based on natural constants, < ->To avoid hyper-parameters with too small an exponential function result, the present embodiment uses +.>To describe, the present embodiment is by +.>The inverse proportion relation and normalization processing are presented, and an implementer can set an inverse proportion function and a normalization function according to actual conditions; according to the method, the first similarity of any two blades in the adjustment period is obtained, the minimum value in all the first similarities is obtained, the product of the difference value obtained by subtracting the minimum value from 1 and the maximum stirring speed of the stirring device is used as the stirring adjustment speed of the adjustment period, the maximum stirring speed is the maximum angular speed of stirring of the stirring device, and the intrinsic parameter of the stirring device is not repeated in the embodiment; the smaller the minimum value of the first similarity, the larger the difference in the mixing uniformity effect of the liquids in the vicinity of the different blades, the more the stirring speed is required to uniformly mix the liquids.
Further, after the stirring regulation speed of the regulation period is obtained, the pressure data of the regulation period is already acquired, the regulation period is already ended, and after the regulation period is ended, the stirring speed of the stirring device is regulated to be the stirring regulation speed, and stirring is continued; meanwhile, the pressure sensor continuously collects pressure data, and each time period after the stirring speed is adjusted is recorded as an adjusted time period.
It should be further noted that, when the stirring speed starts to be increased, the periodicity may be reduced when the mixed liquids in different areas are subjected to flow exchange, and when the stirring effect is better, the periodicity of each blade and the similarity of the periodic patterns of different blades are both larger; when the whole stirring condition is good, the two values change slightly, otherwise, the two values change all the time, so that whether the mixed liquid reaches the ideal uniform mixing condition can be indicated according to the change of the binary group, and the two values in the binary group are necessarily larger.
Specifically, for any adjusted period, acquiring pressure data of each blade in the adjusted period, acquiring a stirring periodicity of the pressure data of each blade in the adjusted period, and taking an average value of the stirring periodicity of all the blades in the adjusted period as a second periodicity of the adjusted period; meanwhile, according to a calculation method of the first similarity of any two blades in the adjusted time period, calculating the first similarity for any two blades in the adjusted time period, and taking the average value of all the first similarities as the second similarity of the adjusted time period to obtain a binary group consisting of the second periodicity and the second similarity in the adjusted time period; and obtaining the second periodicity and the second similarity of each adjusted time period according to the method to obtain the corresponding binary group of each adjusted time period.
Therefore, the stirring speed of the stirring device is adjusted, and the intelligent regulation and control of the position and the posture of the blade in the stirring device are realized through the stirring speed adjustment.
The stirring device control stopping module 103 judges the liquid mixing state in the device according to the second periodicity and the second similarity of the adjusted period, and completes the intelligent regulation and control of the stirring device.
After obtaining the binary group corresponding to each adjusted time period, setting a preset third threshold value for judging the liquid mixing state in the device, wherein the preset third threshold value in the embodiment is described by adopting 0.9; and starting after the stirring speed is regulated after the regulating period, calculating the cosine similarity of the binary group for the second regulated period and the adjacent previous regulated period, stopping stirring if the cosine similarity is larger than a preset third threshold value, judging the stirring stopping of the regulated period one by one according to the method, stopping stirring after the regulating period when the first cosine similarity is larger than the preset third threshold value occurs, wherein the liquid mixing effect in the stirring device is better, the mixing is more uniform, and the intelligent regulation of the stirring device is finished.
So far, through laying the pressure sensor on the blade in the agitating unit, realize the real-time control to agitating unit, accomplished agitating unit's intelligent regulation and control.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (7)
1. An intelligent regulation and control system of agitating unit position and gesture, characterized in that, this system includes:
the blade sensor data acquisition module is used for distributing pressure sensors on blades of the stirring device and acquiring pressure data of each blade in a plurality of time periods;
blade stirring speed regulation and control module: according to the first periodicity of each periodic mode in the pressure data of each period of each blade, obtaining the stirring periodicity of each period of each blade, and obtaining the period to be regulated of each blade through threshold screening;
according to the first similarity of the same time periods to be adjusted of different blades, obtaining the stirring adjustment speed of the adjustment time periods, and obtaining the second periodicity and the second similarity of each adjusted time period;
and the stirring device control stopping module judges the liquid mixing state in the device according to the second periodicity and the second similarity of the adjusted time period, and completes the intelligent regulation and control of the stirring device.
2. The intelligent control system for the position and the posture of the stirring device according to claim 1, wherein the obtaining the stirring periodicity of each blade in each period comprises the following specific steps:
according to the periodic distribution in the pressure data of each time period of each blade, a plurality of periodic modes and a first periodicity of each time period of each blade are obtained; taking the pressure data of any one blade in any period as target pressure data, marking a pressure curve corresponding to the target pressure data as a target pressure curve, and performing STL decomposition on the target pressure curve to obtain a periodic term curve; taking the maximum value of the first periodicity in a plurality of periodic modes in the target pressure data as the comprehensive periodicity of the target pressure data;
acquiring the ordinate ratio of each moment in the target pressure curve and the periodic term curve, wherein the ratio is obtained from a small value to a large value, and taking the product of the average value of the ratio at all the moments and the comprehensive periodicity as the stirring periodicity of the corresponding period of the target pressure data; the stirring periodicity per period of each blade is obtained.
3. The intelligent control system for the position and the posture of the stirring device according to claim 2, wherein the specific acquisition method comprises the following steps of:
according to the periodic distribution in the pressure data of each period of each blade, a plurality of periodic modes, a mode pressure curve and initial periodicity of each period of each blade are obtained; taking pressure data of any one blade in any period as target pressure data, taking any one period mode in the target pressure data as a target period mode, calculating a first derivative of a mode pressure curve of the target period mode, obtaining a zero point, taking data points in the mode pressure curve corresponding to the zero point as characteristic points of the target period mode, taking coordinate data of each characteristic point as input of an SVD algorithm, outputting a plurality of characteristic vectors of the mode pressure curve, and recording the characteristic vectors as a plurality of characteristic vectors of the target period mode; acquiring a plurality of characteristic vectors of each periodic mode;
performing KM matching on feature vectors of any two periodic modes in the target pressure data, wherein a plurality of feature vectors of the same periodic mode are used as nodes on one side in the bipartite graph, the edge value between the nodes on the left side and the right side is represented by the similarity of the feature vectors, and the similarity of the feature vectors is obtained by a DTW distance; obtaining a plurality of matching pairs through KM matching, and taking the average value of the side values corresponding to the matching pairs as the trend similarity of the two periodic modes; acquiring trend similarity of any two periodic modes in the target pressure data;
according to the DTW distance of the mode pressure curves of any two periodic modes in the target pressure data, the overall similarity degree of the any two periodic modes is obtained, the average value of the overall similarity degree and the trend similarity degree is used as the comprehensive similarity degree of the any two periodic modes, the average value of the comprehensive similarity degree of each periodic mode and all other periodic modes in the target pressure data is calculated, linear normalization is carried out on all the average values, the obtained result is recorded as the gain coefficient of each periodic mode, and the product of the gain coefficient and the initial periodicity is used as the first periodicity of each periodic mode in the target pressure data.
4. The intelligent control system for the position and the posture of the stirring device according to claim 3, wherein the method for obtaining the plurality of periodic modes, the mode pressure curves and the initial periodicity of each period of each blade comprises the following specific steps:
according to the periodic distribution in the pressure data of each period of each blade, a plurality of periodic modes of each period of each blade, a plurality of periodic pressure curves of each periodic mode and a plurality of mode pressure curves are obtained;
taking pressure data of any one blade in any period as target pressure data, taking any one periodic mode in the target pressure data as a target periodic mode, and initiating periodicity of the target periodic modeThe calculation method of (1) is as follows:
wherein,indicates the number of cycle pressure curves in the target cycle mode, +.>Total number of all periodic pressure curves in the target pressure data, +.>The cosine similarity mean value of any two sections of periodic pressure curves in the target periodic mode is represented by +.>Representing cosine similarity variance of any two sections of periodic pressure curves in target periodic mode, +.>Represents an exponential function based on natural constants, < ->To avoid over-parametrics with too small an exponential function result;
an initial periodicity of each periodic pattern in the pressure data for each period of each blade is obtained.
5. The intelligent control system for the position and the posture of the stirring device according to claim 4, wherein the specific acquisition method is that the plurality of periodic patterns of each period of each blade, the plurality of periodic pressure curves and the plurality of mode pressure curves of each periodic pattern are:
taking pressure data of any one blade in any period as target pressure data, marking a pressure curve corresponding to the target pressure data as a target pressure curve, performing STL decomposition on the target pressure curve to obtain a periodic term curve, performing Fourier transformation on the periodic term curve to a frequency domain space, taking the reciprocal of frequency corresponding to each amplitude in the frequency domain space as one period, and obtaining a plurality of periods of the target pressure data;
dividing a target pressure curve by taking any one period as a target period to obtain a plurality of sections of local pressure curves, respectively calculating cosine similarity of each section of local pressure curve and all other local pressure curves, dividing two sections of local pressure curves corresponding to all cosine similarity larger than a preset first threshold value into the target period, and recording the two sections of local pressure curves as a plurality of sections of period pressure curves of the target period; dividing the periods into a plurality of sections of period pressure curves of each period one by one according to the sequence from small to large, and recording the period with the period pressure curves as a period mode;
taking any one periodic mode as a target periodic mode, acquiring cosine similarity of any two sections of periodic pressure curves in the target periodic mode, obtaining cosine similarity mean value of each section of periodic pressure curve and other sections of periodic pressure curves in the target periodic mode, and recording the periodic pressure curve with the maximum mean value as the mode pressure curve of the target periodic mode; and acquiring a plurality of periodic pressure curves and mode pressure curves of each periodic mode in the pressure data of each period of each blade.
6. The intelligent control system for the position and the posture of the stirring device according to claim 5, wherein the specific method for obtaining the stirring adjustment speed in the adjustment period comprises the following steps:
the first time interval of all the blades in the time interval is the time interval to be adjusted, and the time interval is recorded as an adjusting time interval; acquiring the stirring periodicity of the pressure data corresponding to each blade in the adjustment period and a periodic mode with the first greatest periodicity, taking the periodic mode with the first greatest periodicity in the pressure data corresponding to each blade as the periodic mode of each blade in the adjustment period, and calculating first similarity of the periodic modes of any two blades in the adjustment period, wherein the first similarity is obtained through the DTW distance between the mode pressure curves of the two periodic modes;
and obtaining the minimum value in all the first similarities, and taking the product of the difference value obtained by subtracting the minimum value from 1 and the maximum stirring speed of the stirring device as the stirring regulation speed of the regulation period.
7. The intelligent control system for the position and the posture of the stirring device according to claim 6, wherein the second periodicity and the second similarity of each adjusted time period are obtained, and the specific method comprises the following steps:
recording each time period after the adjustment of the stirring speed as an adjusted time period, taking any one of the adjusted time periods as a target adjusted time period, acquiring pressure data of each blade in the target adjusted time period, acquiring the stirring periodicity of the pressure data of each blade in the target adjusted time period, and taking the average value of the stirring periodicity of all the blades in the target adjusted time period as the second periodicity of the target adjusted time period;
according to a calculation method of the first similarity of any two blades in the adjustment period, calculating the first similarity of any two blades in the target adjusted period, and taking the average value of all the first similarities as the second similarity of the target adjusted period; a second periodicity and a second similarity for each adjusted time period are obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311446019.4A CN117170422B (en) | 2023-11-02 | 2023-11-02 | Intelligent regulation and control system for position and posture of stirring device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311446019.4A CN117170422B (en) | 2023-11-02 | 2023-11-02 | Intelligent regulation and control system for position and posture of stirring device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117170422A true CN117170422A (en) | 2023-12-05 |
| CN117170422B CN117170422B (en) | 2024-03-05 |
Family
ID=88937934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311446019.4A Active CN117170422B (en) | 2023-11-02 | 2023-11-02 | Intelligent regulation and control system for position and posture of stirring device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117170422B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120048619A1 (en) * | 2010-08-26 | 2012-03-01 | 1473706 Alberta Ltd. | System, method and apparatus for drilling agitator |
| JP2020144672A (en) * | 2019-03-07 | 2020-09-10 | 富士電機株式会社 | Information processing apparatus, information processing method, program, and abnormality diagnosis device |
| CN112708787A (en) * | 2021-03-26 | 2021-04-27 | 北京亿晟科技有限公司 | Mixing and stirring system based on lanthanum cerium rare earth processing production |
| US20210204563A1 (en) * | 2016-04-14 | 2021-07-08 | Nestec S.A. | Viscous semi-liquid food dispenser and method and system for controlling food characteristics |
| CN114098409A (en) * | 2020-08-31 | 2022-03-01 | 佛山市顺德区美的电热电器制造有限公司 | Control method of cooking apparatus, and computer-readable storage medium |
| CN116726774A (en) * | 2023-07-28 | 2023-09-12 | 江苏极易新材料有限公司 | Efficient and uniform mixing method for materials with different bulk densities |
-
2023
- 2023-11-02 CN CN202311446019.4A patent/CN117170422B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120048619A1 (en) * | 2010-08-26 | 2012-03-01 | 1473706 Alberta Ltd. | System, method and apparatus for drilling agitator |
| US20210204563A1 (en) * | 2016-04-14 | 2021-07-08 | Nestec S.A. | Viscous semi-liquid food dispenser and method and system for controlling food characteristics |
| JP2020144672A (en) * | 2019-03-07 | 2020-09-10 | 富士電機株式会社 | Information processing apparatus, information processing method, program, and abnormality diagnosis device |
| CN114098409A (en) * | 2020-08-31 | 2022-03-01 | 佛山市顺德区美的电热电器制造有限公司 | Control method of cooking apparatus, and computer-readable storage medium |
| CN112708787A (en) * | 2021-03-26 | 2021-04-27 | 北京亿晟科技有限公司 | Mixing and stirring system based on lanthanum cerium rare earth processing production |
| CN116726774A (en) * | 2023-07-28 | 2023-09-12 | 江苏极易新材料有限公司 | Efficient and uniform mixing method for materials with different bulk densities |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117170422B (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110826021B (en) | Robust identification and output estimation method for nonlinear industrial process | |
| WO2005019948A1 (en) | Computer-aided modeling and manufacture of products | |
| Jaramillo et al. | On-line estimation of the aerobic phase length for partial nitrification processes in SBR based on features extraction and SVM classification | |
| CN110263488B (en) | Industrial rubber compound Mooney viscosity soft measurement method based on integrated instant learning | |
| CN107924162B (en) | Optimal control device, optimal control method, recording medium, and optimal control system | |
| CN114693650A (en) | Intelligent control method of mixing machine based on machine vision | |
| CN117170422B (en) | Intelligent regulation and control system for position and posture of stirring device | |
| CN116205910B (en) | Injection molding temperature self-adaptive learning regulation and control system for power adapter | |
| Sheng et al. | Soft sensor design based on phase partition ensemble of LSSVR models for nonlinear batch processes | |
| CN117455320B (en) | QGMVPN model-based multi-kettle polymerization process melt quality index online prediction method | |
| CN116881635B (en) | Data management system for textile detergent equipment | |
| CN114781249A (en) | High-density clarification tank dosage prediction and control method based on multidimensional scoring model | |
| CN114403486A (en) | Intelligent control method of airflow type cut-tobacco drier based on local peak value coding circulation network | |
| CN117131314B (en) | Mixed material temperature monitoring and regulating system for asphalt pavement construction | |
| CN116991841A (en) | Data intelligent cleaning method for mixed wind data model | |
| CN111709585A (en) | Air conditioner load prediction method and device and storage medium | |
| CN108332362B (en) | Heating, ventilation and circulating water system energy-saving control system and method based on vectorization analysis | |
| Chen et al. | Tea leaves identification based on gray-level Co-occurrence matrix and K-nearest neighbors algorithm | |
| Liu et al. | Image quality evaluation based on SVM and improved grid search algorithm | |
| CN114386508A (en) | Gaussian process regression sequential design method based on sample point significance | |
| CN113743707A (en) | Product credibility calculation method based on uniform distribution | |
| CN114782427B (en) | Modified plastic mixing evaluation method based on data identification and artificial intelligence system | |
| CN117724431A (en) | Control method and control system for granularity of reaction materials in reaction kettle | |
| CN109491349B (en) | Batch running track and space adjusting method applied to continuous stirring reaction kettle | |
| CN117531438A (en) | Control method and system for improving stability of polyoxymethylene product |
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 |