CN117635030B - Chemical storage management method and system based on cloud computing - Google Patents
Chemical storage management method and system based on cloud computing Download PDFInfo
- Publication number
- CN117635030B CN117635030B CN202311673713.XA CN202311673713A CN117635030B CN 117635030 B CN117635030 B CN 117635030B CN 202311673713 A CN202311673713 A CN 202311673713A CN 117635030 B CN117635030 B CN 117635030B
- Authority
- CN
- China
- Prior art keywords
- time
- time period
- target
- loss
- natural
- 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
- 239000000126 substance Substances 0.000 title claims abstract description 168
- 238000003860 storage Methods 0.000 title claims abstract description 35
- 238000007726 management method Methods 0.000 title claims abstract description 29
- 230000008859 change Effects 0.000 claims abstract description 51
- 238000009826 distribution Methods 0.000 claims abstract description 39
- 230000009469 supplementation Effects 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 19
- 238000000034 method Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 6
- 238000013139 quantization Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000005070 sampling Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a chemical storage management method and system based on cloud computing, comprising the following steps: collecting chemical usage data; constructing a time-quality two-dimensional space coordinate system, and acquiring a time distribution curve; acquiring time segments, and constructing a natural loss time period sequence and an artificial loss time period sequence; presetting the size of a window scale, and acquiring a plurality of first data points; acquiring the use degree and the natural change degree of the time segment; acquiring a weight value of each data point in the target time period, and using degree or natural change degree; obtaining a clustering result of a target scale; acquiring a preference degree value of the target scale according to a clustering result of the target scale; acquiring an optimal window scale and a second data point; obtaining a change degree value of natural loss of chemicals and average consumption loss, and obtaining a chemical replenishment threshold; and performing chemical supplementation according to the chemical supplementation threshold value, so as to realize accurate chemical supplementation.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a chemical storage management method and system based on cloud computing.
Background
In order to facilitate the storage of chemicals, the chemicals are usually stored in a chemical storage cabinet, and the use state of each chemical is monitored in real time by arranging a sensor in the chemical storage cabinet by utilizing cloud computing technology, so that the chemicals are tracked and identified, and the real-time monitoring of the chemical inventory is facilitated, so that whether each chemical needs to be replenished or not is facilitated.
Conventional weight monitoring of chemicals in a chemical cabinet to determine whether the chemical needs to be replenished typically sets a fixed threshold, and the usage of the chemical may change over time, for example, the usage of a certain chemical is higher in one period and the usage of a certain chemical is lower in another period, and the fixed threshold cannot meet the usage change condition of the chemical in the different periods, resulting in premature or too late replenishment of the chemical. If the chemical is replenished too early, the chemical may deteriorate due to too long time, which may result in resource waste, and if the chemical is replenished too late, the production is interrupted or the work is delayed, so that the replenishment threshold of the chemical needs to be adaptively obtained according to the usage rate of the chemical.
Disclosure of Invention
In order to solve the problems, the invention provides a chemical storage management method and a chemical storage management system based on cloud computing.
The invention discloses a chemical storage management method and a chemical storage management system based on cloud computing, which adopt the following technical scheme:
one embodiment of the invention provides a chemical storage management method based on cloud computing, which comprises the following steps:
collecting chemical usage data, including a quality data sequence and a time data sequence; recording the data acquired last time as current data;
constructing a time-quality two-dimensional space coordinate system, converting chemical use data into data points, and acquiring a time distribution curve;
segmenting according to a time data sequence to obtain a plurality of time segments, wherein two endpoints in the time segments are data of the time data sequence; according to two endpoints of the time segment, a natural loss time period sequence and an artificial loss time period sequence are constructed, wherein the natural loss time period sequence and the artificial loss time period sequence respectively comprise a plurality of time segments;
presetting the size of a window scale, marking any time segment as a target time period, marking any window scale as a target scale, and carrying out local regression interpolation according to the target scale to obtain a plurality of first data points;
for any time segment in the human loss time period sequence of the target time period, acquiring the use degree of the target time period; for any time segment in a natural loss time period sequence of the target time period, acquiring the natural change degree of the target time period;
According to the time distribution curve of the target time period and the use degree or the natural change degree of the target time period, acquiring a weight value and the use degree or the natural change degree of each data point in the target time period;
constructing a sample coordinate system, wherein the abscissa in the sample coordinate system is quality data, the ordinate is the use degree or the natural change degree, and all first data points of the target scale are converted into sample data points; clustering all sample data points to obtain a clustering result of a target scale; acquiring a preference degree value of the target scale according to a clustering result of the target scale;
the window scale corresponding to the maximum value of the preference degree value is marked as an optimal window scale, and a plurality of second data points are obtained according to the local regression interpolation result of the optimal window scale;
acquiring a change degree value of the natural loss of the chemical according to the second data point and the natural loss time period sequence; obtaining average consumption according to the artificial consumption time period sequence; acquiring a chemical supplementation threshold according to the change degree value and the average consumption loss of the natural loss of the chemical;
chemical replenishment is performed according to a chemical replenishment threshold.
Further, the construction of the natural loss time period sequence and the artificial loss time period sequence comprises the following specific steps:
For any time segment, if any first time data exists in two endpoints of the time segment, screening the time segment to be a natural loss time segment, and constructing a natural loss time segment sequence; and if the two endpoints in the time segment are the first using time data, the former endpoint is the second using data, the time segment is screened out and marked as an artificial loss time period, and an artificial loss time period sequence is constructed.
Further, the local regression interpolation is performed according to the target scale to obtain a plurality of first data points, including the following specific steps:
and starting from the first data point in the target time period, performing polynomial fitting by using the data of the first data point under the target scale, wherein the data of the time to be fitted and interpolated is recorded as fitting data points, acquiring the ordinate values of the fitting data points under the target scale when the first data point is fitted, and so on, acquiring the ordinate values of the fitting data points of the second data point under the target scale, sequentially acquiring the ordinate values of the fitting data points obtained by fitting other data points in the target time period, replacing the fitting data points with the data points on the original time distribution curve, and obtaining a plurality of first data points without replacing the acquired original data points.
Further, the step of obtaining the usage degree of the target time period includes the following specific steps:
obtaining the consumption according to the first consumption data and the second consumption data corresponding to the two end point time in the target time period under the target scale; recording the ratio of the average value of the absolute values of the difference values of the longitudinal coordinates between the first data points of adjacent unit time under the target scale and the consumption as a first variation; calculating the absolute value of the difference of the abscissa between the artificial loss time period and the nearest endpoint of the previous artificial loss time period in any artificial loss time period in the artificial loss time period sequence, and recording the absolute value as the time interval between the two adjacent artificial loss time periods; the degree of use D1 of the nth artificial loss period in the artificial loss period sequence n The specific calculation method comprises the following steps:
;
wherein U is n A first variation representing an nth person being a loss period;a mean value representing a first variation of the all human loss time period; u (U) i A first variation representing an ith person's loss period; p (P) n Representing a time interval between an nth artificial loss time period and a previous artificial loss time period; p (P) i Representing a time interval between an ith artificial depletion period and a previous artificial depletion period; />A mean value representing the time intervals of all adjacent artificial loss time periods; exp () represents an exponential function based on a natural constant.
Further, the step of obtaining the natural variation degree of the target time period includes the following specific steps:
obtaining natural loss according to first quality data corresponding to two end point time in a target time period under a target scale; the ratio of the average value of the absolute values of the difference values of the longitudinal coordinates between the first data points of adjacent unit time under the target scale and the natural loss is recorded as a second variation; calculating the absolute value of the difference of the abscissa between the nearest end point of the natural loss time period and the previous natural loss time period in any natural loss time period in the natural loss time period sequence, and recording the absolute value as the time interval between the two adjacent natural loss time periods; wherein the natural change degree of the mth natural loss time period is recorded as D2 m The specific calculation method comprises the following steps:
;
wherein V is m A second variation representing an mth natural loss period;a mean value representing a second variation of all natural loss periods; v (V) j A second variation representing a j-th natural loss period; q (Q) m Representing the time interval between the mth natural loss period and the previous natural loss period; />A mean value representing the time intervals of all adjacent natural loss periods; q (Q) j Representing the time interval between the jth natural loss period and the previous natural loss period; exp () represents an exponential function based on a natural constant.
Further, the step of obtaining the weight value and the usage degree or the natural variation degree of each data point in the target time period includes the following specific steps:
the target time period is the s-th time segment, any one first data point of the target time period is recorded as a target data point,removing the target data points, marking the time distribution curve of the target time period as a first target time distribution curve, acquiring the time distribution curve of the target data period after the target data points are removed, and marking the time distribution curve as a second target time distribution curve; performing DTW matching on the first target time distribution curve and the second target time distribution curve to obtain a DTW distance, wherein the target data point is the first data point in the target time period, and the weight value epsilon of the first data point is the weight epsilon of the first data point sl The calculation method of (1) is as follows:
;
therein, dtw sl Dtw distance between the first target time profile and the second target time profile representing the first data point of the s-th time segment; exp () represents an exponential function based on a natural constant;
according to the weight value of the data point of the target data point, the using degree of the target data point is obtained, and when the target time period is the s-th time segment and is the artificial loss time period, the using degree D1 of the first data point is obtained sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) Representing a maximum value of the weight values of the data points in the s-th time segment; D1D 1 s Indicating the degree of use of the s-th time segment; max () represents the get maximum function;
according to the weight value of the target data point, acquiring the natural change degree of the target data point, wherein the target time period is the s-th time segment and is the natural loss time period, and the natural change degree D2 of the first data point sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) A maximum value of the weight values representing the data points of the s-th time segment; D2D 2 s Representing the natural degree of variation of the s-th time segment; max () represents the acquisition maximum function.
Further, the obtaining the preference degree value of the target scale according to the clustering result of the target scale includes the following specific steps:
performing DBSCAN clustering on all sample data points to obtain a clustering result of a target scale; and marking the sizes of the rest window scales as scales to be calculated, acquiring normalized mutual information values between the target scales and clustering results of any other scales to be calculated, obtaining a plurality of normalized mutual information values, and marking the average value of all the normalized mutual information values as the preference degree value of the target scales.
Further, the change degree value of the natural loss of the chemical is obtained according to the second data point and the natural loss time period sequence; the average consumption is obtained according to the artificial consumption time period sequence, and the method comprises the following specific steps:
for a plurality of natural loss time periods, combining any two natural loss time periods to obtain a plurality of time period combinations, recording any one of the obtained natural loss time periods as a first natural loss time period, obtaining all the plurality of time period combinations containing the first natural loss time period, recording the plurality of time period combinations as a plurality of first combinations, carrying out dtw matching on time distribution curves of the two natural loss time periods of any one of the first combinations, and carrying out exp (-dtw) quantization on the obtained dtw result to obtain the similarity degree of the first combinations, wherein dtw represents dtw distance between the time distribution curves of the two natural loss time periods of the first combinations, exp () represents an exponential function taking a natural constant as a base; obtaining the similarity degree of all the first combinations, calculating the average value of the similarity degree of all the first combinations, and recording the average value as the target value of the first natural loss time period; selecting the maximum value of target values of all natural loss time periods, taking the natural loss time period corresponding to the maximum value as a prediction time period, and acquiring the average value of the absolute values of the difference values of the longitudinal coordinates of adjacent unit time in the prediction time period as a change degree value delta of the natural loss of the chemical;
Acquiring the average time length of all the artificial loss time periods in the artificial loss time period sequence; acquiring the consumption of all artificial loss time periods in the artificial loss time period sequence, and for any one artificial loss time period, acquiring the ratio of the consumption of the artificial loss time period to the time length of the artificial loss time period, and recording the ratio as a first ratio; and obtaining the first ratio of other artificial loss time periods, obtaining the average value of a plurality of first ratios, and recording the average value as the average consumption amount sigma.
Further, the method for obtaining the chemical replenishment threshold comprises the following specific steps:
the chemical replenishment threshold F for the current data was calculated by:
;
wherein,representing the average time length of the artificial loss period; delta represents the change degree value of the natural loss of the chemical; sigma table represents the average usage loss for the artificial loss period; max (t) represents the maximum length of the human loss period; f represents a chemical hazard amount super parameter.
The invention also provides a chemical storage management system based on cloud computing, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the method when executing the computer program.
The technical scheme of the invention has the beneficial effects that: according to the invention, the chemical use data is collected and analyzed, the influence of natural loss and artificial use on the quality of the chemical is combined, and the data collected by adopting a local regression interpolation algorithm is subjected to self-adaptive interpolation, so that the self-adaptive quantification of the chemical replenishment quantity threshold value is realized, and the method is used for subsequent chemical replenishment. The method comprises the steps of carrying out sectional processing on a time-varying sequence of the quality of chemicals according to the quality data of the chemicals in a chemical cabinet and the data of experimenters during manual use, acquiring fitting data under different window scales, acquiring different time sections to quantify the influence on the quality of the chemicals, acquiring the use degree or natural variation degree of the data, further acquiring the optimal window scale of optimal local regression interpolation, acquiring the variation degree value of the natural loss of the chemicals and average consumption of the use according to the optimal window scale, acquiring a chemical supplementation threshold, avoiding the defect that the effect of the traditional local regression interpolation is influenced by the interpolation window scale, acquiring an accurate chemical supplementation threshold, and enabling the chemical cabinet to be more accurate for early warning in professionals, thereby avoiding unnecessary resource waste.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a chemical storage management method based on cloud computing according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of a chemical storage management method and system based on cloud computing according to the invention, which are provided by the invention, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a chemical storage management method and a chemical storage management system based on cloud computing.
Referring to fig. 1, a flowchart of steps of a chemical storage management method based on cloud computing according to an embodiment of the present invention is shown, the method includes the following steps:
and S001, collecting chemical use data.
It should be noted that, since some chemicals are often required to be stored under specific environmental conditions, natural losses are reduced. Therefore, chemicals are typically stored in chemical cabinets, which often have temperature and humidity control functions, and are adapted to chemicals requiring specific environmental conditions to reduce natural losses. In the embodiment, the chemical use data is collected and analyzed, the influence of natural loss and artificial use on the quality of the chemical is combined, and the data collected by adopting a local regression interpolation algorithm is subjected to self-adaptive interpolation so as to realize the self-adaptive quantification of the chemical replenishment quantity threshold value and be used for subsequent replenishment of the chemical. It is therefore necessary to arrange sensors to collect usage data of the chemicals.
Specifically, the RFID tag is adopted to rapidly identify chemicals, the RFID tag is set for all the chemicals by a professional to identify, and a quality sensor is arranged to monitor the quality of the chemicals, wherein the quality of the chemicals and time are recorded and recorded as first quality data and first time data. The sampling time is set to be acquired every 10 minutes, and the registration time, the return time and the corresponding chemical quality data of the chemicals used by the experimenters are acquired at the same time and are respectively recorded as first usage time data and second usage time data, and first usage amount data and second usage amount data, wherein the first usage time data and the first usage amount data are the registration time and the chemical quality data at the registration time, and the second usage time data and the second usage amount data are the return time and the chemical quality data at the return time. And arranging the first quality data, the first usage amount data and the second usage amount data according to a time sequence to obtain a chemical quality data sequence, wherein the first time data, the first usage time data and the second usage time data form a time data sequence. It should be noted that, in the process of using the chemical by the experimenter, if the set sampling time is reached but the experimenter does not return the chemical, the first quality data of the sampling time is set as the chemical quality data acquired last before the sampling time in the chemical quality data sequence. Wherein the last acquired data is recorded as current data.
S002, constructing a time-quality two-dimensional space coordinate system, constructing a natural loss time period sequence and an artificial loss time period sequence, and obtaining the use degree and the natural change degree of the target time period.
In the process of acquiring the chemical replenishment quantity threshold, the acquired chemical quality data are scattered due to the set sampling frequency and the influence of manual use, and larger data errors occur when the influence of natural loss and the influence of manual use of chemicals in the chemical cabinet is analyzed, so that the data in unit time are often filled in a data interpolation mode. Local regression interpolation is a way to interpolate by polynomial fitting of data points in each data and its local range, which algorithm has good adaptability to various complex relationships and nonlinear modes. However, in the process of local regression interpolation, because the effect of local regression interpolation is affected by the interpolation window scale, if the window scale is larger, local information of data of chemical quality changing along with time is distorted, and if the window scale is too small, the influence of randomness on the chemical quality changing along with time is larger, so that when the chemical quality changing data is filled by the local regression interpolation, an accurate window scale needs to be determined to obtain an accurate filling data result, and accurate quantization of a chemical supplementation threshold is performed according to the data acquired originally and the filling data. In this embodiment, the local regression interpolation is a local weighted linear regression interpolation method, and specific method steps thereof are known techniques, for example, methods described in journal articles entitled "PM 2.5 spatial interpolation method of local weighted linear regression model", by authors Lu Yueming and Wang Liang, etc.
It should be further described that when the storage conditions in the chemical cabinet are quite consistent with the storage conditions of each chemical, natural loss still occurs, and the natural loss and the artificial loss affect the quality data of the chemical differently along with the artificial loss of the chemical caused by the use of the chemical by the experimenter, so that in the process of quantifying the window scale, the quality data of the chemical in the chemical cabinet and the data of the experimenter during the artificial use need to be processed in a segmented manner according to the time-varying sequence of the quality of the chemical, and the influence on the quality of the chemical is quantified separately for different time segments. In the results of interpolation of the local regression values for different window sizes, it is expected that the results of the filled data do not affect the natural loss and the influence of the artificial use loss on the quality of chemicals, so in the embodiment, the influence of the natural loss and the artificial use loss on the quality of chemicals is obtained by segmenting different time, and the similarity of clustering results under different window sizes is obtained by clustering, so that the preference degree of the window size is obtained, and the optimal window is obtained. Wherein in acquiring the effects of natural loss and artificial use loss on chemical quality, it is necessary to acquire the changes over time of chemical quality data corresponding to different time segments, and the distribution positions of the respective time segments over time sequence.
Specifically, a time-mass two-dimensional space coordinate system is constructed, wherein the abscissa is time data, each 1 minute in the abscissa is a unit time, and the ordinate is mass data, and the acquired data are converted into data points in the two-dimensional space. In a two-dimensional space coordinate system, connecting data points of adjacent time in a time data sequence to obtain a time distribution curve, wherein the data points of each adjacent time form a time segment, and a plurality of time segments are obtained. For any time segment, if any first time data exists in two endpoints of the time segment, screening the time segment to be a natural loss time segment, and constructing a natural loss time segment sequence; and if the former end point is the first using time data and the latter end point is the second using data point in the two end points in the time segment, screening the time segment, marking the time segment as an artificial loss time segment, and constructing an artificial loss time segment sequence.
Further, any time segment is marked as a target time period, a plurality of window scale sizes are preset to perform local regression interpolation, wherein the window scale sizes preset in the embodiment are respectively marked as a first scale, a second scale, a third scale, a fourth scale, a fifth scale, a sixth scale and a seventh scale, the window scale sizes are the number of data points used in polynomial fitting, and it is noted that when the data in the target time period is operated, the boundary of the target time period may be exceeded, and if the number of data in a window is not smaller than the window scale size, the data number actually contained in the window is calculated in a subsequent calculation process. And for any scale, marking the scale as a target scale, starting from a first data point in a target time period, performing polynomial fitting by using data of the first data point at the target scale, marking the data of the time to be fitted and interpolated as fitting data points, acquiring the ordinate values of the fitting data points at the target scale when the first data point is fitted, and so on, acquiring the ordinate values of the fitting data points at the target scale of a second data point, and sequentially acquiring the ordinate values of the fitting data points obtained by fitting other data points in the target time period. It should be noted that, when a plurality of data points are used for fitting, a plurality of ordinate values are obtained at the same fitting time point, so in this embodiment, the determination of the ordinate values of the fitting time point is performed by adopting a manner of obtaining an average value of a plurality of data values of the same fitting time point, so as to obtain a fitting data point, and the fitting data point is replaced with a data point on an original time distribution curve, wherein the collected original data point is not replaced, so as to obtain a plurality of first data points.
Further, for the target scale, if the target time period is the artificial loss time period, the consumption is obtained according to the first consumption data and the second consumption data corresponding to the two end point time in the target time period. And recording the ratio of the average value of the absolute value of the difference value of the ordinate between the first data points of adjacent unit time under the target scale and the consumption as a first variation. And calculating the absolute value of the difference of the abscissa between the artificial loss time period and the nearest endpoint of the previous artificial loss time period in any artificial loss time period in the artificial loss time period sequence, and recording the absolute value as the time interval between the two adjacent artificial loss time periods. The time interval calculation is not performed for the first artificial loss period in the sequence of artificial loss periods, and the time interval is recorded as 0. The degree of use D1 of the nth artificial loss period in the artificial loss period sequence n The specific calculation method comprises the following steps:
;
wherein U is n A first variation representing an nth person being a loss period;a mean value representing a first variation of the all human loss time period; u (U) i A first variation representing an ith person's loss period; p (P) n Representing a time interval between an nth artificial loss time period and a previous artificial loss time period; p (P) i Representing a time interval between an ith artificial depletion period and a previous artificial depletion period; />A mean value representing the time intervals of all adjacent artificial loss time periods; exp () represents an exponential function based on a natural constant, and it should be noted that the exp (-x) model used in this embodiment is used only to represent the negative phaseThe results output by the relation and constraint model are in the interval of [0,1 ], and can be replaced by other models with the same purpose in implementation, and the embodiment only takes an exp (-x) model as an example for description, and is not particularly limited, wherein x represents the input of the model. Wherein->The first change amount of the artificial loss time period of the chemical in the historical data is larger than the first change amount of the artificial loss time period of the chemical in the historical data, so that the use amount of the chemical in the artificial loss time period of the n is larger, and the artificial use degree is higher; wherein->The smaller the value of the time interval representing the nth person's artificial loss period relative to the time interval of the artificial loss period of the chemical in the history data, the shorter the usage time interval of the chemical used by the nth person's artificial loss period, the more frequent the usage, and the higher the artificial usage degree.
Further, for the target scale, if the target time period is a natural loss time period. And obtaining the natural loss according to the first quality data corresponding to the two end point time in the target time period. And recording the ratio of the average value of the absolute value of the difference value of the ordinate between the first data points of adjacent unit time under the target scale and the natural loss as a second variation. In any one natural loss time period in the natural loss time period sequence, calculating the absolute value of the difference of the abscissa between the nearest end points of the natural loss time period and the previous natural loss time period, and recording the absolute value as the time interval between the two adjacent natural loss time periods, wherein the time interval calculation is not performed for the first natural loss time period in the natural loss time period sequence, and the time interval is recorded as 0. Wherein the natural change degree of the mth natural loss time period is recorded as D2 m The specific calculation method comprises the following steps:
;
wherein V is m A second variation representing an mth natural loss period;a mean value representing a second variation of all natural loss periods; v (V) j A second variation representing a j-th natural loss period; q (Q) m Representing the time interval between the mth natural loss period and the previous natural loss period; />A mean value representing the time intervals of all adjacent natural loss periods; q (Q) j Representing the time interval between the jth natural loss period and the previous natural loss period; the exp () represents an exponential function based on a natural constant, and it should be noted that the exp (-x) model used in this embodiment is only used to indicate that the negative correlation and the result output by the constraint model are within the interval of [0,1 ], and may be replaced by another model having the same purpose when implemented, and this embodiment is described only by taking the exp (-x) model as an example, and is not limited thereto. Wherein->The second change amount representing the mth natural loss period is larger in value than the second change amount of the natural loss period of the chemical in the history data, so that the larger the natural change amount of the chemical in the mth natural loss period is, the higher the natural change degree is; wherein->The smaller the value of the time interval representing the mth natural loss period relative to the time interval of the natural loss period of the chemical in the history data, the shorter the time interval of the mth natural loss period, the more frequent the natural loss, and the higher the natural change degree.
S003, acquiring a weight value and a use degree or a natural change degree of each data point in the target time period according to the time distribution curve of the target time period and the use degree or the natural change degree of the target time period; and obtaining a preference degree value of the target scale according to the clustering result of the target scale, obtaining a change degree value of the natural loss of the chemical and average consumption, and obtaining a chemical supplementation threshold.
After obtaining the usage degree and the natural variation degree corresponding to the different time segments, the usage degree and the natural variation degree of the data points in the different time segments need to be quantized, so as to obtain the usage degree and the natural variation degree of the data points in each unit time. And acquiring the using degree and the natural variation scale of the filled data points by using different filled data points obtained under different window scale sizes, and clustering according to the amplitudes, the using degree and the natural variation scale of the filled data points and the acquired original data points to obtain clustering results of different window scale sizes. And obtaining the similarity of clustering results under different window scales to obtain the window scale preference degree, further obtaining an optimal window, and further obtaining an optimal local regression interpolation result. And determining a time segment for acquiring the change characteristic of the natural loss according to the optimal local regression interpolation result, and adapting the chemical replenishment threshold value according to the use data of the chemical.
Specifically, any one time segment is recorded as a target time segment, wherein the target time segment is an s-th time segment, and the use degree and the natural change degree of a plurality of first data points in the target time segment are quantified according to the use degree and the natural change degree of the target time segment. And marking any one first data point of the target time period as a target data point, removing the target data point, marking the time distribution curve of the target time period as a first target time distribution curve, acquiring the time distribution curve of the target data period after the target data point is removed, and marking the time distribution curve as a second target time distribution curve. That is, the first target time profile is the time profile without removing the target data points, and the second target time profile is the time profile after removing the target data points.
Performing DTW matching on the first target time distribution curve and the second target time distribution curve to obtain a DTW distance, wherein the target data point is the first data point in the target time period, and the weight value epsilon of the first data point is the weight epsilon of the first data point sl The calculation method of (1) is as follows:
;
therein, dtw sl Dtw distance between the first target time profile and the second target time profile representing the first data point of the s-th time segment; the exp () represents an exponential function based on a natural constant, and it should be noted that the exp (-x) model used in this embodiment is only used to indicate that the negative correlation and the result output by the constraint model are within the interval of [0,1 ], and may be replaced by another model having the same purpose when implemented, and this embodiment is described only by taking the exp (-x) model as an example, and is not limited thereto. And if the time distribution curve difference before and after the target data point is removed is large, the change of the chemical quality of the target data point is important, and the weight value of the target data point is large.
Further, according to the weight value of the target data point, the usage degree of the target data point is obtained, and when the target time period is the s-th time segment and the time segment is the human loss time period, the target data point is the l data point in the target time period, and the usage degree D1 of the l data point is obtained sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) Representing a maximum value of the weight values of the data points in the s-th time segment; D1D 1 s Indicating the degree of use of the s-th time segment; max () represents the acquisition maximum function.
Further, according to the weight value of the target data point, the natural change degree of the target data point is obtained, and for the target time period which is the s-th time segment and is the natural loss time period, the target data point is the first data point in the target time period, and the natural change degree D2 of the first data point is obtained sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) A maximum value of the weight values representing the data points of the s-th time segment; D2D 2 s Representing the natural degree of variation of the s-th time segment; max () represents the acquisition maximum function.
Further, a sample coordinate system is constructed, wherein an abscissa in the sample coordinate system is quality data, an ordinate is a use degree or a natural change degree, for any window scale size, all first data points of the window scale size are converted into sample data points in the sample coordinate system, and it is to be noted that, for the first data points in different time segments, if the time segments are natural loss time segments, the ordinate values of the sample data points are natural change degrees, and if the time segments are artificial loss time segments, the ordinate values of the sample data points are use degrees. And performing DBSCAN clustering on all the sample data points to obtain a clustering result under the window scale. And marking the size of the window scale as a target scale, marking the sizes of the other window scales as scales to be calculated, acquiring normalized mutual information values between the target scale and clustering results of any other scales to be calculated, obtaining a plurality of normalized mutual information values, and marking the average value of all the normalized mutual information values as the preference degree value of the target scale. Similar operation is performed, the preference degree values under the sizes of other window dimensions are obtained, the maximum value of the preference degree values of different window dimensions is selected and recorded as the optimal window dimension, wherein the calculation of the normalized mutual information value is a known technology, and the embodiment is not repeated.
Further, the result of local regression interpolation corresponding to the optimal window scale is recorded as a second data point, and a plurality of second data points are replaced to the original time distribution curve. For a plurality of natural loss time periods, combining any two natural loss time periods to obtain a plurality of time period combinations, recording any one of the obtained natural loss time periods as a first natural loss time period, obtaining all the plurality of time period combinations containing the first natural loss time period as a plurality of first combinations, carrying out dtw matching on time distribution curves of the two natural loss time periods of any one of the first combinations, and carrying out exp (-dtw) quantization on the obtained dtw results to obtain the similarity degree of the first combinations, wherein dtw represents dtw distance between the time distribution curves of the two natural loss time periods of the first combinations, exp () represents an exponential function based on natural constants, and it is to be noted that exp (-x) models used in the embodiment are only used for representing negative correlation and the output result of the constraint model are in the [0,1 ] interval, and other models with the same purposes can be replaced when the embodiment is only described by taking exp (-x) models as examples, and the embodiment does not specifically describe the input model, wherein x represents the input model. And similarly, obtaining the similarity degree of all the first combinations, calculating the average value of the similarity degree of all the first combinations, and recording the average value as the target value of the first natural loss time period. And selecting the maximum value of the target values of all the natural loss time periods, taking the natural loss time period corresponding to the maximum value as a prediction time period, and obtaining the average value of the absolute values of the difference values of the longitudinal coordinates of adjacent unit time in the prediction time period as a change degree value delta of the natural loss of the chemical.
Further, obtaining the average time length of all the artificial loss time periods in the artificial loss time period sequence; and obtaining the consumption of all the artificial consumption time periods in the artificial consumption time period sequence, and for any one artificial consumption time period, obtaining the ratio of the consumption of the artificial consumption time period to the time length of the artificial consumption time period, and recording the ratio as a first ratio. Similarly, a first ratio of other artificial loss time periods is obtained, and the average value of a plurality of first ratios is obtained and is recorded as average consumption sigma.
Further, the method for calculating the chemical replenishment threshold F for the current data is as follows:
;
wherein,representing the average time length of the artificial loss period; delta represents the change degree value of the natural loss of the chemical; sigma table represents the average usage loss for the artificial loss period; max (t) represents the maximum length of the human loss period; f represents a chemical hazard superparameter, and the present example gives an empirical reference of fifteen percent of the chemical mass, depending on the implementation of the practitioner. Wherein->The average natural loss level at the time of artificial loss is shown, and σxmax (t) represents the amount of loss used for the longest time when the experimenter uses the chemical. The chemical hazard superparameter characterizes the minimum amount of chemical needed to be replenished, which, together with the artificial and natural losses, characterizes the threshold level of chemical need to be replenished.
S004, performing chemical storage management professional early warning according to the chemical supplementation threshold value of the current data.
Specifically, according to the acquired chemical replenishment threshold value of the current data, comparing the quality data of the chemical of the current data with the chemical replenishment threshold value of the current data, and if the quality data of the chemical of the current data is smaller than or equal to the chemical replenishment threshold value of the current data, carrying out early warning on a chemical storage management professional to prompt the chemical storage management professional to replenish the chemical.
The invention also provides a chemical storage management system based on cloud computing, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the steps S001-S004 are realized when the processor executes the computer program.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A cloud computing-based chemical storage management method, comprising the steps of:
Collecting chemical usage data, including a quality data sequence and a time data sequence; recording the data acquired last time as current data;
constructing a time-quality two-dimensional space coordinate system, converting chemical use data into data points, and acquiring a time distribution curve;
segmenting according to a time data sequence to obtain a plurality of time segments, wherein two endpoints in the time segments are data of the time data sequence; according to two endpoints of the time segment, a natural loss time period sequence and an artificial loss time period sequence are constructed, wherein the natural loss time period sequence and the artificial loss time period sequence respectively comprise a plurality of time segments;
presetting the size of a window scale, marking any time segment as a target time period, marking any window scale as a target scale, and carrying out local regression interpolation according to the target scale to obtain a plurality of first data points;
for any time segment in the human loss time period sequence of the target time period, acquiring the use degree of the target time period; for any time segment in a natural loss time period sequence of the target time period, acquiring the natural change degree of the target time period;
According to the time distribution curve of the target time period and the use degree or the natural change degree of the target time period, acquiring a weight value and the use degree or the natural change degree of each data point in the target time period;
constructing a sample coordinate system, wherein the abscissa in the sample coordinate system is quality data, the ordinate is the use degree or the natural change degree, and all first data points of the target scale are converted into sample data points; clustering all sample data points to obtain a clustering result of a target scale; acquiring a preference degree value of the target scale according to a clustering result of the target scale;
the window scale corresponding to the maximum value of the preference degree value is marked as an optimal window scale, and a plurality of second data points are obtained according to the local regression interpolation result of the optimal window scale;
acquiring a change degree value of the natural loss of the chemical according to the second data point and the natural loss time period sequence; obtaining average consumption according to the artificial consumption time period sequence; acquiring a chemical supplementation threshold according to the change degree value and the average consumption loss of the natural loss of the chemical;
chemical replenishment is performed according to a chemical replenishment threshold.
2. The chemical storage management method based on cloud computing according to claim 1, wherein the construction of the natural loss time period sequence and the artificial loss time period sequence comprises the following specific steps:
For any time segment, if any first time data exists in two endpoints of the time segment, screening the time segment to be a natural loss time segment, and constructing a natural loss time segment sequence; and if the two endpoints in the time segment are the first using time data, the former endpoint is the second using data, the time segment is screened out and marked as an artificial loss time period, and an artificial loss time period sequence is constructed.
3. The chemical storage management method based on cloud computing according to claim 1, wherein the obtaining a plurality of first data points by local regression interpolation according to a target scale comprises the following specific steps:
and starting from the first data point in the target time period, performing polynomial fitting by using the data of the first data point under the target scale, wherein the data of the time to be fitted and interpolated is recorded as fitting data points, acquiring the ordinate values of the fitting data points under the target scale when the first data point is fitted, and so on, acquiring the ordinate values of the fitting data points of the second data point under the target scale, sequentially acquiring the ordinate values of the fitting data points obtained by fitting other data points in the target time period, replacing the fitting data points with the data points on the original time distribution curve, and obtaining a plurality of first data points without replacing the acquired original data points.
4. The chemical storage management method based on cloud computing according to claim 1, wherein the obtaining the usage degree of the target time period comprises the following specific steps:
obtaining the consumption according to the first consumption data and the second consumption data corresponding to the two end point time in the target time period under the target scale; recording the ratio of the average value of the absolute values of the difference values of the longitudinal coordinates between the first data points of adjacent unit time under the target scale and the consumption as a first variation; calculating the absolute value of the difference of the abscissa between the artificial loss time period and the nearest endpoint of the previous artificial loss time period in any artificial loss time period in the artificial loss time period sequence, and recording the absolute value as the time interval between the two adjacent artificial loss time periods; the degree of use D1 of the nth artificial loss period in the artificial loss period sequence n The specific calculation method comprises the following steps:
;
wherein U is n A first variation representing an nth person being a loss period;a mean value representing a first variation of the all human loss time period; u (U) i A first variation representing an ith person's loss period; p (P) n Representing a time interval between an nth artificial loss time period and a previous artificial loss time period; p (P) i Representing a time interval between an ith artificial depletion period and a previous artificial depletion period; />A mean value representing the time intervals of all adjacent artificial loss time periods; exp () represents an exponential function based on a natural constant.
5. The chemical storage management method based on cloud computing according to claim 1, wherein the step of obtaining the natural variation degree of the target time period comprises the following specific steps:
obtaining natural loss according to first quality data corresponding to two end point time in a target time period under a target scale; the ratio of the average value of the absolute values of the difference values of the longitudinal coordinates between the first data points of adjacent unit time under the target scale and the natural loss is recorded as a second variation; calculating the absolute value of the difference of the abscissa between the nearest end point of the natural loss time period and the previous natural loss time period in any natural loss time period in the natural loss time period sequence, and recording the absolute value as the time interval between the two adjacent natural loss time periods; wherein the natural change degree of the mth natural loss time period is recorded as D2 m The specific calculation method comprises the following steps:
;
wherein V is m A second variation representing an mth natural loss period;a mean value representing a second variation of all natural loss periods; v (V) j A second variation representing a j-th natural loss period; q (Q) m Representing the mth natural loss periodThe time interval of the previous natural loss period; />A mean value representing the time intervals of all adjacent natural loss periods; q (Q) j Representing the time interval between the jth natural loss period and the previous natural loss period; exp () represents an exponential function based on a natural constant.
6. The method for managing chemical storage based on cloud computing according to claim 1, wherein the step of obtaining the weight value and the usage level or the natural variation level of each data point in the target time period comprises the following specific steps:
the target time period is the s-th time segment, any one first data point of the target time period is recorded as a target data point, the target data point is removed, the time distribution curve of the target time period is recorded as a first target time distribution curve, the time distribution curve of the target data period after the target data point is removed is obtained, and the second target time distribution curve is recorded; performing DTW matching on the first target time distribution curve and the second target time distribution curve to obtain a DTW distance, wherein the target data point is the first data point in the target time period, and the weight value epsilon of the first data point is the weight epsilon of the first data point sl The calculation method of (1) is as follows:
;
therein, dtw sl Dtw distance between the first target time profile and the second target time profile representing the first data point of the s-th time segment; exp () represents an exponential function based on a natural constant;
according to the weight value of the data point of the target data point, the using degree of the target data point is obtained, and when the target time period is the s-th time segment and is the artificial loss time period, the using degree D1 of the first data point is obtained sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) Representing a maximum value of the weight values of the data points in the s-th time segment; D1D 1 s Indicating the degree of use of the s-th time segment; max () represents the get maximum function;
according to the weight value of the target data point, acquiring the natural change degree of the target data point, wherein the target time period is the s-th time segment and is the natural loss time period, and the natural change degree D2 of the first data point sl The calculation method of (1) is as follows:
;
wherein ε sl A weight value representing a first data point of the s-th time segment; max (epsilon) s ) A maximum value of the weight values representing the data points of the s-th time segment; D2D 2 s Representing the natural degree of variation of the s-th time segment; max () represents the acquisition maximum function.
7. The chemical storage management method based on cloud computing according to claim 1, wherein the obtaining the preference degree value of the target scale according to the clustering result of the target scale comprises the following specific steps:
performing DBSCAN clustering on all sample data points to obtain a clustering result of a target scale; and marking the sizes of the rest window scales as scales to be calculated, acquiring normalized mutual information values between the target scales and clustering results of any other scales to be calculated, obtaining a plurality of normalized mutual information values, and marking the average value of all the normalized mutual information values as the preference degree value of the target scales.
8. The cloud computing-based chemical storage management method of claim 1, wherein the obtaining the change degree value of the natural loss of the chemical according to the second data point and the natural loss time period sequence; the average consumption is obtained according to the artificial consumption time period sequence, and the method comprises the following specific steps:
for a plurality of natural loss time periods, combining any two natural loss time periods to obtain a plurality of time period combinations, recording any one of the obtained natural loss time periods as a first natural loss time period, obtaining all the plurality of time period combinations containing the first natural loss time period, recording the plurality of time period combinations as a plurality of first combinations, carrying out dtw matching on time distribution curves of the two natural loss time periods of any one of the first combinations, and carrying out exp (-dtw) quantization on the obtained dtw result to obtain the similarity degree of the first combinations, wherein dtw represents dtw distance between the time distribution curves of the two natural loss time periods of the first combinations, exp () represents an exponential function taking a natural constant as a base; obtaining the similarity degree of all the first combinations, calculating the average value of the similarity degree of all the first combinations, and recording the average value as the target value of the first natural loss time period; selecting the maximum value of target values of all natural loss time periods, taking the natural loss time period corresponding to the maximum value as a prediction time period, and acquiring the average value of the absolute values of the difference values of the longitudinal coordinates of adjacent unit time in the prediction time period as a change degree value delta of the natural loss of the chemical;
Acquiring the average time length of all the artificial loss time periods in the artificial loss time period sequence; acquiring the consumption of all artificial loss time periods in the artificial loss time period sequence, and for any one artificial loss time period, acquiring the ratio of the consumption of the artificial loss time period to the time length of the artificial loss time period, and recording the ratio as a first ratio; and obtaining the first ratio of other artificial loss time periods, obtaining the average value of a plurality of first ratios, and recording the average value as the average consumption amount sigma.
9. The cloud computing-based chemical storage management method of claim 1, wherein said obtaining a chemical replenishment threshold comprises the specific steps of:
the chemical replenishment threshold F for the current data was calculated by:
;
wherein,representing the average time length of the artificial loss period; delta represents the change degree value of the natural loss of the chemical; sigma table represents the average usage loss for the artificial loss period; max (t) represents the maximum length of the human loss period; f represents a chemical hazard amount super parameter.
10. A cloud computing based chemical storage management system comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor, when executing the computer program, implements the steps of a cloud computing based chemical storage management method according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311673713.XA CN117635030B (en) | 2023-12-07 | 2023-12-07 | Chemical storage management method and system based on cloud computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311673713.XA CN117635030B (en) | 2023-12-07 | 2023-12-07 | Chemical storage management method and system based on cloud computing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117635030A CN117635030A (en) | 2024-03-01 |
| CN117635030B true CN117635030B (en) | 2024-04-02 |
Family
ID=90033771
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311673713.XA Active CN117635030B (en) | 2023-12-07 | 2023-12-07 | Chemical storage management method and system based on cloud computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117635030B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298402A (en) * | 2019-07-01 | 2019-10-01 | 国网内蒙古东部电力有限公司 | A kind of small target deteection performance optimization method |
| CN114519681A (en) * | 2021-12-31 | 2022-05-20 | 上海仙途智能科技有限公司 | Automatic calibration method and device, computer readable storage medium and terminal |
| US11487273B1 (en) * | 2021-04-30 | 2022-11-01 | Dalian University Of Technology | Distributed industrial energy operation optimization platform automatically constructing intelligent models and algorithms |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113554089B (en) * | 2021-07-22 | 2023-04-18 | 西安电子科技大学 | Image classification countermeasure sample defense method and system and data processing terminal |
-
2023
- 2023-12-07 CN CN202311673713.XA patent/CN117635030B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298402A (en) * | 2019-07-01 | 2019-10-01 | 国网内蒙古东部电力有限公司 | A kind of small target deteection performance optimization method |
| US11487273B1 (en) * | 2021-04-30 | 2022-11-01 | Dalian University Of Technology | Distributed industrial energy operation optimization platform automatically constructing intelligent models and algorithms |
| CN114519681A (en) * | 2021-12-31 | 2022-05-20 | 上海仙途智能科技有限公司 | Automatic calibration method and device, computer readable storage medium and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117635030A (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Davydenko et al. | Measuring forecasting accuracy: The case of judgmental adjustments to SKU-level demand forecasts | |
| CN112395046B (en) | Virtual machine migration planning and scheduling method, system and medium thereof | |
| CN114444231B (en) | Online self-adaptive prediction method, device, equipment and medium for residual life of mold | |
| CN111737078B (en) | Load type-based adaptive cloud server energy consumption measuring and calculating method, system and equipment | |
| WO2021007845A1 (en) | Cloud-edge collaborative forecasting system and method for aluminum oxide production indexes | |
| CN112907064B (en) | Electric quantity prediction method and device based on adaptive window, storage medium and terminal | |
| CN113537850A (en) | Storage optimization method and device, computer equipment and storage medium | |
| CN117635030B (en) | Chemical storage management method and system based on cloud computing | |
| CN114564345A (en) | Server abnormity detection method, device, equipment and storage medium | |
| CN116975503B (en) | Soil erosion information management method and system | |
| CN114330102A (en) | Rapid flood forecasting method and device based on rainfall similarity and intelligent model parameter adaptation | |
| CN110084814A (en) | Surface covering measurement method, device, computer equipment and storage medium | |
| CN115392591A (en) | Task processing method, device, equipment and storage medium | |
| CN112085926B (en) | River water pollution early warning method and system | |
| CN114388128A (en) | Method, apparatus, device and medium for detecting personal motion intelligent PAI index | |
| CN113111729B (en) | Training method, recognition method, system, device and medium for personnel recognition model | |
| CN110956675B (en) | Method and device for automatically generating technology maturity curve | |
| CN110322063B (en) | Power consumption simulation prediction method and storage medium | |
| CN111625525B (en) | Environment data repairing/filling method and system | |
| CN113312587A (en) | Sensor acquisition data missing value processing method based on ARIMA prediction and regression prediction | |
| CN112070162A (en) | Multi-class processing task training sample construction method, device and medium | |
| CN117239743B (en) | Electric energy meter electricity load acquisition method, device, equipment and medium | |
| CN117611031B (en) | Real-time monitoring method and system for logistics temperature of cold chain Internet of things | |
| CN111522828B (en) | User portrait tag value analysis method and device | |
| CN115351601B (en) | Tool wear monitoring method based on transfer learning |
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 |