CN117422345B

CN117422345B - Oil-residue separation quality assessment method and system

Info

Publication number: CN117422345B
Application number: CN202311733137.3A
Authority: CN
Inventors: 张道泉; 葛传迎; 陈文超
Original assignee: Taian Jinguanhong Food Technology Co ltd
Current assignee: Taian Jinguanhong Food Technology Co ltd
Priority date: 2023-12-18
Filing date: 2023-12-18
Publication date: 2024-03-12
Anticipated expiration: 2043-12-18
Also published as: CN117422345A

Abstract

The invention relates to the field of electronic data processing, in particular to an oil-residue separation quality evaluation method and system, wherein the method acquires oil content time sequence data, temperature time sequence data and acid-base value time sequence data at a feed inlet in a preset historical time period before the current moment; dividing the oil content time sequence data into N data point intervals; calculating the fluctuation correlation of the multi-source time sequence data at the current time by using the temperature time sequence data and the acid-base value time sequence data; acquiring a target data point interval corresponding to the current grease content, and calculating a data fluctuation optimization factor at the current time according to the current grease content, the target data point interval and the multisource time sequence data fluctuation correlation degree; and optimizing the current oil content by utilizing a data fluctuation optimization factor so as to calculate and obtain the oil-slag separation quality index at the current moment according to the optimized current oil content and the oil content at the discharge port after oil-slag separation, thereby improving the accuracy of oil-slag separation quality assessment.

Description

Oil-residue separation quality assessment method and system

Technical Field

The invention relates to the field of electronic data processing, in particular to an oil-residue separation quality evaluation method and system.

Background

In the production process of edible oil, a centrifuge is generally used for separating oil from residues, so as to remove the oil residues in the oil. In the process of oil and slag separation by using the centrifugal machine, equipment damage caused by long-term use of the centrifugal machine can cause deterioration of oil and slag separation effect, so that quality of oil and slag removed is deteriorated, and therefore, in order to ensure quality of oil and slag removed, the oil and slag separation effect of the oil and slag separation process is monitored and evaluated in real time, so that equipment adjustment is performed on the centrifugal machine according to the oil and slag separation effect.

In the prior art, the oil and slag separation effect is evaluated, the oil content at the feed inlet and the discharge outlet is obtained respectively in the oil and slag separation process, and the corresponding oil and slag separation effect is obtained according to the oil content difference between the feed inlet and the discharge outlet. However, when the raw materials of the grease to be separated are different, fluctuation in the content of the grease at the feed port is caused, and further, the evaluation accuracy of the oil-residue separation effect becomes low, for example: when the oil content at the feed inlet is reduced and the oil content at the discharge outlet is unchanged, the index corresponding to the estimated oil separation effect is increased, the oil separation effect is better as the index is increased,

Therefore, how to improve the evaluation accuracy of the oil-residue separation effect becomes a problem to be solved.

Disclosure of Invention

In view of the above, the embodiment of the invention provides an oil-residue separation quality evaluation method to solve the problem of how to improve the evaluation accuracy of the oil-residue separation effect.

In a first aspect, an embodiment of the present invention provides an oil-residue separation quality evaluation method, including:

in the grease production process, according to the grease content, the temperature and the acid-base value at the feeding port at each sampling moment, obtaining grease content time sequence data, temperature time sequence data and acid-base value time sequence data at the feeding port in a preset historical time period before the current moment;

acquiring nearest neighbor data points of any data point in the grease content time sequence data, forming data point pairs by the data points and the nearest neighbor data points corresponding to the data points, obtaining all data point pairs formed in the grease content time sequence data, and dividing the grease content time sequence data into N data point intervals according to all the data point pairs, wherein N is more than 1;

acquiring the current grease content, the current temperature and the current pH value of grease to be separated at the feed inlet at the current time, correspondingly acquiring a real-time grease content sequence in which the current grease content is positioned according to the grease content time sequence data, the temperature time sequence data and the pH value time sequence data, respectively, and calculating the fluctuation correlation degree of multi-source time sequence data at the current time according to the real-time grease content sequence, the real-time temperature sequence and the real-time pH value sequence in which the current pH value is positioned;

Acquiring a target data point interval corresponding to the current grease content in the N data point intervals, and calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation correlation degree, the current temperature and the current pH value;

and optimizing the current grease content by using the data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after the grease to be separated at the feed inlet is subjected to the oil-residue separation, and calculating to obtain the oil-residue separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet.

Preferably, the acquiring the nearest neighbor data point of any data point in the grease content time series data includes:

taking the data point as a starting point, acquiring a grease content sub-sequence with a preset length from the grease content time sequence data, and calculating the absolute value of a data difference value between other data points and the data point aiming at any other data point except the data point in the grease content sub-sequence;

And traversing all other data points except the data points in the oil-containing subsequence, correspondingly obtaining all data difference absolute values, obtaining a minimum data difference absolute value from all data difference absolute values, and taking the other data points corresponding to the minimum data difference absolute value as nearest neighbor data points of the data points.

Preferably, the partitioning the oil content time series data into N data point intervals according to all data point pairs includes:

counting the number of data point pairs formed by data points at two sides of the data point aiming at any data point in the oil content time sequence data;

and acquiring the number of data point pairs corresponding to each data point in the oil content time sequence data, forming a data point pair number sequence, acquiring a minimum value in the data point pair number sequence, determining the number of N-1 data point pairs according to the minimum value, taking the data points respectively corresponding to the number of the N-1 data point pairs as dividing points, and dividing the oil content time sequence data into N data point sections.

Preferably, the acquiring the real-time grease content sequence in which the current grease content is located, the real-time temperature sequence in which the current temperature is located, and the real-time acid-base value sequence in which the current acid-base value is located according to the grease content time sequence data, the temperature time sequence data, and the acid-base value time sequence data respectively correspond to each other, includes:

Taking the current oil content as the last data point in the real-time oil content sequence, and extracting a preset number of data points before the current moment from the oil content time sequence data to form the real-time oil content sequence;

taking the current temperature as the last data point in the real-time temperature sequence, and extracting a preset number of data points before the current moment from the temperature time sequence data to form the real-time temperature sequence;

and taking the current pH value as the last data point in the real-time pH value sequence, and extracting a preset number of data points before the current moment from the pH time sequence data to form the real-time pH value sequence.

Preferably, the calculating the multi-source time sequence data fluctuation correlation under the current time according to the real-time grease content sequence, the real-time temperature sequence and the real-time acid-base number sequence includes:

respectively carrying out normalization processing on the real-time temperature sequence and the real-time acid-base number sequence to correspondingly obtain a normalized real-time temperature sequence and a normalized real-time acid-base number sequence;

calculating a first similarity between the normalized real-time temperature sequence and the real-time oil content sequence by using a DTW algorithm, and calculating a second similarity between the real-time oil content sequence and the normalized real-time acid-base value sequence by using the DTW algorithm;

Calculating an addition result between the first similarity and the second similarity, normalizing the addition result to obtain a normalized addition result, and taking a difference value between a preset value and the normalized addition result as a multisource time sequence data fluctuation correlation degree at the current time.

Preferably, the acquiring the target data point interval corresponding to the current grease content in the N data point intervals includes:

and acquiring nearest neighbor data points of the current grease content from the grease content time sequence data as target data points, and taking a data point interval where the target data points are located as a target data point interval.

Preferably, the calculating the data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation correlation degree, the current temperature and the current acid-base number includes:

the current temperature and the current acid-base value form current two-dimensional data, the temperature and the acid-base value respectively corresponding to each sampling time in the temperature time sequence data and the acid-base value time sequence data form historical two-dimensional data, a historical two-dimensional data set is formed, and the historical two-dimensional data set and the current two-dimensional data are subjected to outlier detection by using a COF algorithm to obtain a connectivity outlier factor of the current two-dimensional data;

Acquiring a first multiplication result between a connectivity outlier factor of the current two-dimensional data and the multi-source time sequence data fluctuation correlation degree;

performing outlier detection on the oil content time sequence data and the current oil content by utilizing the COF algorithm to obtain a connectivity outlier factor of the current oil content, calculating an oil content average value of all historical oil contents contained in the target data point interval, and calculating a first difference value between the current oil content and the oil content average value;

obtaining a second difference value between a constant 1 and the fluctuation correlation of the multi-source time sequence data, and obtaining a second multiplication result between the first difference value, the second difference value and a connectivity outlier factor of the current grease content;

and normalizing the addition result between the first multiplication result and the second multiplication result, wherein the normalization result obtained correspondingly is used as the data fluctuation optimization factor at the current time.

Preferably, the optimizing the current grease content by using the data fluctuation optimization factor to obtain the optimized current grease content includes:

and acquiring the historical grease content at the feed inlet at the moment before the current moment, calculating the difference value between the historical grease content and the current grease content, acquiring the product of the difference value and the data fluctuation optimization factor, and taking the addition result between the historical grease content and the product as the optimized current grease content.

Preferably, the calculating to obtain the oil-slag separation quality index at the current time according to the optimized current oil content and the oil content at the discharge port includes:

calculating the oil content difference between the optimized current oil content and the oil content at the discharge port, calculating the ratio between the oil content difference and the optimized current oil content, and taking the ratio as an oil-residue separation quality index at the current time.

In a second aspect, an embodiment of the present invention provides an oil-residue separation quality evaluation system, including:

the data sequence acquisition module is used for acquiring time sequence data of the oil content, the temperature time sequence data and the acid-base value time sequence data of the feed inlet in a preset historical time period before the current moment according to the oil content, the temperature and the acid-base value of the feed inlet at each sampling moment in the oil production process;

the data sequence segmentation module is used for acquiring the nearest neighbor data point of any data point in the grease content time sequence data, forming data point pairs by the data point and the nearest neighbor data point corresponding to the data point, obtaining all data point pairs formed in the grease content time sequence data, and dividing the grease content time sequence data into N data point intervals according to all the data point pairs, wherein N is more than 1;

The data fluctuation analysis module is used for acquiring the current grease content, the current temperature and the current pH value of the grease to be separated at the feed inlet at the current moment, correspondingly acquiring a real-time grease content sequence in which the current grease content is positioned according to the grease content time sequence data, the temperature time sequence data and the pH value time sequence data, and calculating the fluctuation correlation degree of the multi-source time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time pH value sequence in which the current pH value is positioned;

the optimization factor acquisition module is used for acquiring a target data point interval corresponding to the current grease content in the N data point intervals, and calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation correlation degree, the current temperature and the current pH value;

the separation quality evaluation module is used for optimizing the current grease content by utilizing the data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after the grease is subjected to the grease residue separation, and calculating to obtain the grease residue separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet.

Compared with the prior art, the embodiment of the invention has the beneficial effects that:

according to the invention, in the grease production process, according to the grease content, the temperature and the acid-base value at the feeding port at each sampling moment, the grease content time sequence data, the temperature time sequence data and the acid-base value time sequence data at the feeding port in a preset historical time period before the current moment are obtained; acquiring nearest neighbor data points of any data point in the grease content time sequence data, forming data point pairs by the data points and the corresponding nearest neighbor data points, obtaining all data point pairs formed in the grease content time sequence data, and dividing the grease content time sequence data into N data point intervals according to all the data point pairs, wherein N is more than 1; acquiring the current grease content, the current temperature and the current pH value of grease to be separated at a feeding port at the current moment, correspondingly acquiring a real-time grease content sequence in which the current grease content is positioned, a real-time temperature sequence in which the current temperature is positioned and a real-time pH value sequence in which the current pH value is positioned according to the grease content time sequence data, the temperature time sequence data and the pH value time sequence data, and calculating the fluctuation correlation degree of multi-source time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time pH value sequence; acquiring a target data point interval corresponding to the current grease content in the N data point intervals, and calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the fluctuation correlation degree of multi-source time sequence data, the current temperature and the current pH value; and optimizing the current grease content by utilizing a data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after the grease is separated from the grease at the feed inlet, and calculating to obtain the grease separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet. The method comprises the steps of acquiring historical monitoring time sequence data in a historical time period between current moments, carrying out fluctuation analysis on the grease content at a lower feed inlet at the current moment, so as to obtain a data fluctuation optimization factor at the current moment, carrying out data optimization on the grease content at the lower feed inlet at the current moment by utilizing the data fluctuation optimization factor, further enabling the estimated grease separation effect to be more accurate according to the grease content difference before and after grease separation, and eliminating the estimated fluctuation of grease separation quality caused by the grease content fluctuation at the feed inlet.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present 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 a method for evaluating the quality of oil-residue separation according to an embodiment of the present invention;

fig. 2 is a block diagram of an oil-residue separation quality evaluation system according to a second embodiment of the present invention.

Detailed Description

Embodiments of the present disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present disclosure and are not to be construed as limiting the present disclosure.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.

The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations. In order to illustrate the technical scheme of the invention, the following description is made by specific examples.

The invention aims at the following scenes: in the oil-residue separation process of oil production, the operation parameters of the centrifugal machine are adjusted in real time through the evaluation of the oil-residue separation effect after oil-residue separation, so that the self-adaptive effect of the operation parameters of the centrifugal machine is achieved.

Referring to fig. 1, a method flowchart of an oil-residue separation quality evaluation method according to an embodiment of the present invention, as shown in fig. 1, may include:

step S101, in the grease production process, according to the grease content, the temperature and the acid-base value at the lower feed inlet at each sampling moment, obtaining grease content time sequence data, temperature time sequence data and acid-base value time sequence data at the feed inlet in a preset historical time period before the current moment.

Specifically, in the grease production process, various production data in the grease production process can be collected through various sensors, instruments and other collection equipment. In the process of collecting various production data in the grease production process by using the collecting device, the sampling frequency of the collecting device is set, for example, the collecting is performed once for 2 seconds.

In the embodiment of the invention, in the oil-residue separation process, a grease content sensor is arranged in a grease container to be separated at a feed inlet so as to collect the grease content in the grease container to be separated by using the grease content sensor, wherein the unit is thatAnd the temperature sensor and the pH sensor are utilized to collect the temperature and the pH value (pH value) in the grease container to be separated at the feed inlet, so that the grease content, the temperature and the pH value at the feed inlet can be obtained at each sampling time.

Based on the current moment, acquiring the grease content, the temperature and the acid-base value at each sampling moment before the current moment, and respectively forming grease content time sequence data by the historical grease content at all sampling moments, temperature time sequence data by the historical temperature at all sampling moments and acid-base value time sequence data by the acid-base values at all sampling moments in a preset historical time period, wherein the preset historical time period can be 1 minute, 2 minutes and the like. The time sequence data is obtained by arranging the collected data according to the time sequence.

Step S102, the nearest neighbor data point of any data point in the grease content time sequence data is obtained, the data point and the corresponding nearest neighbor data point form data point pairs, all the data point pairs formed in the grease content time sequence data are obtained, and the grease content time sequence data are divided into N data point intervals according to all the data point pairs.

Specifically, in order to perform fluctuation analysis on the oil content at the feed inlet at the current moment, it is necessary to analyze a data fluctuation interval to which the oil content at the feed inlet at the current moment belongs, so after historical monitoring data of the oil content at the feed inlet before the current moment is acquired, that is, after the oil content time series data, the temperature time series data and the acid-base time series data in a preset historical time period are acquired, fluctuation analysis on the oil content time series data can be performed on the oil content time series data, so that the oil content time series data is divided into a plurality of time series mode intervals, and the time series mode intervals are used for subsequently analyzing to which time series mode interval the data fluctuation interval to which the oil content at the feed inlet belongs at the current moment corresponds. Therefore, the oil content time series data is divided into N data point intervals, and N is more than 1, and the specific dividing method is as follows:

(1) And acquiring the nearest neighbor data point of any data point in the grease content time sequence data, and combining the data point and the nearest neighbor data point corresponding to the data point into a data point pair to obtain all the data point pairs combined in the grease content time sequence data.

Specifically, firstly, for any data point in the oil content time sequence data, taking the data point as a starting point, acquiring an oil content sub-sequence with a preset length from the oil content time sequence data, and for any other data point except the data point in the oil content sub-sequence, calculating the absolute value of a data difference value between the other data point and the data point.

For example: for data points in time series data of grease contentData point +.>For the starting point, a fat content subsequence of preset length l=11 is obtained in the fat content time sequence data, the fat content subsequence comprising the data points ∈ ->To data point->Calculating data points +.>Divide data point with oil content subsequence +.>The absolute value of the data difference between each data point is also called the absolute value of the oil content difference.

And then traversing all other data points except the data points in the oil-containing subsequence, correspondingly obtaining all data difference absolute values, obtaining a minimum data difference absolute value from all data difference absolute values, and taking the other data points corresponding to the minimum data difference absolute value as nearest neighbor data points of the data points.

For example: by using the method for calculating the absolute value of the data difference value, the data point can be obtainedObtaining the minimum data difference absolute value in all data difference absolute values, and dividing the oil containing quantum sequence corresponding to the minimum data difference absolute value by data point->Data points other than the data point +.>Is the nearest neighbor data point of (2), and the data point is +.>Data points->Forms a data point pair.

It is worth noting that assume data pointsNearest neighbor data point of (1) is data point->Data point->The nearest neighbor data point of (a) is data point +.>Without the need for data points->The nearest neighbor data are acquired, so that when a certain data point in the grease content time sequence data is used as the nearest neighbor data of other data points, analysis of the nearest neighbor data does not need to be carried out on the certain data point, and each data point in the grease content time sequence data can only have a unique nearest neighbor data point when the data pair is composed. For example, if the fat content time series data includes 8 data points, the 8 data points can be paired two by two, and only 4 pairs can be obtained.

(2) And dividing the grease content time sequence data into N data point intervals according to all data point pairs.

Specifically, for any data point in the oil content time sequence data, counting the number of data point pairs formed by data points at two sides of the data point; and acquiring the number of data point pairs corresponding to each data point in the oil content time sequence data, forming a data point pair number sequence, acquiring a minimum value in the data point pair number sequence, determining the number of N-1 data point pairs according to the minimum value, taking the data points respectively corresponding to the number of the N-1 data point pairs as dividing points, and dividing the oil content time sequence data into N data point sections.

For example: and (3) taking each data point in the grease content time sequence data as an abscissa, taking the number of data point pairs as an ordinate, constructing a data point pair number change curve according to the number of data point pairs of each data point in the grease content time sequence data, acquiring minimum values of the data point pair change curve, arranging the acquired minimum values in a sequence from small to large to obtain a minimum value sequence, taking the first N-1 minimum values in the minimum value sequence as target minimum values, confirming the number of the data point pairs corresponding to each target minimum value, confirming the target data point corresponding to the data point pair, taking the target data point as a dividing point, and dividing the grease content time sequence data into N data point sections according to N-1 dividing points.

Step S103, obtaining the current grease content, the current temperature and the current pH value of grease to be separated at a feed inlet at the current moment, correspondingly obtaining a real-time grease content sequence in which the current grease content is located, a real-time temperature sequence in which the current temperature is located and a real-time pH value sequence in which the current pH value is located according to the grease content time sequence data, the temperature time sequence data and the pH value time sequence data, and calculating the fluctuation correlation degree of the multi-source time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time pH value sequence.

Specifically, in the fluctuation measurement process, there is a change in the oil content monitoring value caused by the oil change, the oil change is the fluctuation of the temperature and the acid-base value of the produced oil to be separated from oil residues in the oil production process, if the oil change occurs in the oil production process and the monitored oil content is also fluctuated correspondingly, the fluctuation of the oil change needs to be distinguished from the simple oil content fluctuation, so that the fluctuation degree of the oil fluctuation change can be used for carrying out fluctuation optimization on the oil content for the oil content fluctuation change caused by the oil change, thus obtaining the current oil content, the current temperature and the current acid-base value of the oil to be separated at the feed inlet at the current moment, and combining the oil content time sequence data, the temperature time sequence data and the acid-base value time sequence data in the preset history time period before the current oil content at the current moment is optimized, and carrying out similarity analysis on the current oil content, the current temperature and the current acid-base value fluctuation of the oil to be separated at the feed inlet at the current moment so as to confirm the influence degree of the temperature and the acid-base value on the oil content, and the specific similarity analysis is carried out as follows:

(1) Taking the current oil content as the last data point in the real-time oil content sequence, and extracting a preset number of data points before the current moment from the oil content time sequence data to form the real-time oil content sequence; taking the current temperature as the last data point in the real-time temperature sequence, and extracting a preset number of data points before the current moment from the temperature time sequence data to form the real-time temperature sequence; and taking the current pH value as the last data point in the real-time pH value sequence, and extracting a preset number of data points before the current moment from the pH time sequence data to form the real-time pH value sequence.

Specifically, assuming that the current time is t, the corresponding sampling time corresponding to the most data point of the oil content time sequence data is t-1, and the preset number is 11, the oil content corresponding to the sampling time t-1 to t-11 and the current oil content corresponding to the current time t are obtainedComposing real-time grease content sequence at current time>. Similarly, based on the temperature time sequence data, the temperatures corresponding to the sampling moments t-1 to t-11 are calculated And the current temperature corresponding to the current time t +.>Composing a real-time temperature sequence at the current time instant +.>The method comprises the steps of carrying out a first treatment on the surface of the Based on the time sequence data of the pH values, the pH values corresponding to the sampling time t-1 to t-11 and the current pH value corresponding to the current time t are calculated>Constitute the real-time acid-base number sequence +.>。

(2) And calculating the fluctuation correlation degree of the multisource time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time acid-base value sequence.

Specifically, the real-time temperature sequence and the real-time acid-base number sequence are respectively normalized, and a normalized real-time temperature sequence and a normalized real-time acid-base number sequence are correspondingly obtained; calculating a first similarity between the normalized real-time temperature sequence and the real-time oil content sequence by using a DTW algorithm, and calculating a second similarity between the real-time oil content sequence and the normalized real-time acid-base value sequence by using the DTW algorithm; calculating an addition result between the first similarity and the second similarity, normalizing the addition result to obtain a normalized addition result, and taking a difference value between a preset value and the normalized addition result as a multisource time sequence data fluctuation correlation degree at the current time.

The calculation expression of the fluctuation relevance of the multisource time sequence data at the current moment is as follows:

wherein,for the multisource time sequence data fluctuation correlation at the current time t,/degree of correlation>As a linear normalization function>For the first similarity between the normalized real-time temperature sequence and the real-time fat content sequence,/I->And 1 is a preset value for the second similarity between the real-time oil content sequence and the normalized real-time acid-base value sequence.

The first similarity between the normalized real-time temperature sequence and the real-time oil content sequence is obtained by calculating through a DTW algorithm, and is used for detecting whether the change amplitude corresponding to the real-time temperature sequence and the change amplitude corresponding to the real-time oil content sequence are similar or not, the larger the first similarity is, the more the change amplitude between the two is the same, the smaller the influence of the change of the corresponding temperature on the change of the oil content is, the second similarity between the real-time oil content sequence and the normalized real-time acid-base value sequence is obtained by calculating through the DTW algorithm, the more the change amplitude corresponding to the real-time acid-base value sequence and the change amplitude corresponding to the real-time oil content sequence are similar or not, the larger the second similarity is, the smaller the change amplitude between the two is, the change of the corresponding acid-base value is the smaller the change influence of the oil content is, therefore, the sum of the first similarity and the second similarity is combined to serve as an evaluation reference for evaluating the influence of the temperature and the acid-base value on the oil content, and the sum of the first similarity is the larger, the influence of the temperature and the acid-base value on the oil content is low, and the corresponding multi-source time sequence data is low in fluctuation.

Step S104, a target data point interval corresponding to the current grease content in the N data point intervals is obtained, and a data fluctuation optimization factor at the current time is calculated according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation correlation degree, the current temperature and the current pH value.

Specifically, after obtaining the correlation degree (multisource time sequence data fluctuation correlation degree) between the current grease content at the feed inlet at the current moment, the current temperature and the current acid-base number, the fluctuation optimization factor of the real-time data point can be measured according to the obtained multisource time sequence data fluctuation correlation degree and the time sequence modal interval corresponding to the current grease content at the feed inlet at the current moment, specifically:

(1) And acquiring a target data point interval corresponding to the current grease content in the N data point intervals.

Specifically, the nearest neighbor data point of the current grease content is obtained from the grease content time sequence data to serve as a target data point, and a data point interval where the target data point is located is taken as a target data point interval. The method comprises the following steps: and (3) reconstructing the current grease content and grease content time sequence data into new time sequence data, then acquiring all data point pairs in the new time sequence data by utilizing the method for acquiring all data point pairs formed in the grease content time sequence data in the step S102, determining nearest neighbor data points of the current grease content as target nearest neighbor data points, and determining the data point interval where the target nearest neighbor data points are located as target data point intervals according to the N data point intervals divided in the step S102.

(2) And calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation correlation degree, the current temperature and the current pH value.

Specifically, the current temperature and the current acid-base value form current two-dimensional data, the temperature and the acid-base value respectively corresponding to each sampling time in the temperature time sequence data and the acid-base value time sequence data form historical two-dimensional data, a historical two-dimensional data set is formed, and the historical two-dimensional data set and the current two-dimensional data are subjected to outlier detection by using a COF algorithm to obtain connectivity outliers of the current two-dimensional data; and obtaining a first multiplication result between the connectivity outlier factor of the current two-dimensional data and the fluctuation correlation degree of the multi-source time sequence data.

It should be noted that, the COF algorithm is in the prior art, and is not described in detail in the embodiment of the present invention.

Performing outlier detection on the oil content time sequence data and the current oil content by utilizing the COF algorithm to obtain a connectivity outlier factor of the current oil content, calculating an oil content average value of all oil contents contained in the target data point interval, and calculating a first difference value between the current oil content and the oil content average value; and obtaining a second difference value between a constant 1 and the fluctuation correlation degree of the multi-source time sequence data, and obtaining a second multiplication result between the first difference value, the second difference value and the connectivity outlier factor of the current grease content.

The calculation expression of the data fluctuation optimization factor at the current time is as follows:

wherein,represents the data fluctuation optimization factor at the current time t, < ->A linear normalization function is represented and,representing the current two-dimensional data consisting of the current temperature and the current pH value, < >>A connectivity outlier representing the current two-dimensional data,/>indicating the current fat content at the feed inlet at the current moment t +.>Indicating the current fat content->Total amount of grease content contained in the corresponding target data point interval, +.>Representing the a-th data point, i.e. a-th fat content, in the target data point interval +.>And a connectivity outlier factor representing the current grease content at the feed inlet at the current moment.

It should be noted that, connectivity outlier factor of current two-dimensional dataCharacterizing the fluctuation difference of the current temperature and the current pH value at the current time compared with the temperature and the pH value in the preset historical time period, wherein the larger the fluctuation difference is, the +. >The larger the value is, the data fluctuation optimization factor +_for the current time t is>The larger, the multisource time sequence data fluctuation correlation degree at the current time t is utilized at the same time>Limiting the data fluctuation optimizing factor at the current time to improve the data fluctuation optimizing factor +.>Accuracy of (2); />For representing the difference between the current grease content at the feed inlet at the current time t and the data points in the same time sequence modal interval, the larger the difference is, the data fluctuation optimization factor corresponding to the current time t is->The bigger the->For characterizing the fluctuation difference of the current fat content at the current time compared with the fat content in the preset history time period, the larger the fluctuation difference is, +.>The larger the value is, the data fluctuation optimization factor +_for the current time t is>The larger.

Step S105, optimizing the current grease content by utilizing a data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed port at the current moment at the discharge port after the grease is separated from the grease at the feed port, and calculating to obtain the grease separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge port.

Specifically, the data fluctuation optimization factor at the current moment is acquired Thereafter, by the data fluctuation optimization factor +.>For the current grease content->Optimizing, wherein the optimizing process is to reduce fluctuation of data points, and specifically comprises the following steps: acquiring the content of the historical grease at the feeding port at the moment before the current moment, and calculating the content of the historical greaseAnd obtaining a product of the difference value and the data fluctuation optimization factor by the difference value between the quantity and the current grease content, and taking the addition result of the historical grease content and the product as the optimized current grease content.

Wherein, the optimization formula is:

wherein,representing the optimized current grease content, +.>Representing the historical fat content at the feed inlet at sampling instant t-1 +.>Indicating the current grease content at the feed inlet at the current time t.

Further, after the current oil content at the current time is optimized, the oil content at the discharge port after oil-residue separation is performed on the oil to be separated at the feed port at the current time is obtained, which is worth noting that the oil to be separated at the feed port is usually discharged from the discharge port after a period of oil-residue separation, so that the oil content at the discharge port after oil-residue separation is performed on the oil to be separated at the feed port at the current time can be collected according to the oil-residue separation time, for example, the oil-residue separation time is 5S, and the oil content after oil-residue separation is performed on the oil to be separated at the discharge port at the 10 th S at the feed port can be collected at the 15 th S.

After the oil content at the discharge port is obtained, the oil-residue separation quality can be evaluated by combining the oil content at the discharge port, and the method specifically comprises the following steps: calculating the oil content difference between the optimized current oil content and the oil content at the discharge port, calculating the ratio between the oil content difference and the optimized current oil content, and taking the ratio as an oil-residue separation quality index at the current time.

The calculation expression of the oil-slag separation quality index is as follows:

wherein,is the oil-slag separation quality index at the current time, < > in the current time>The oil content of the oil to be separated at the feeding port at the current moment is the oil content of the oil to be separated at the discharging port after oil-residue separation.

In summary, in the grease production process, according to the grease content, the temperature and the acid-base value at the feed inlet at each sampling time, the grease content time sequence data, the temperature time sequence data and the acid-base value time sequence data at the feed inlet in the preset historical time period before the current time are obtained; acquiring nearest neighbor data points of any data point in the grease content time sequence data, forming data point pairs by the data points and the corresponding nearest neighbor data points, obtaining all data point pairs formed in the grease content time sequence data, and dividing the grease content time sequence data into N data point intervals according to all the data point pairs, wherein N is more than 1; acquiring the current grease content, the current temperature and the current pH value of grease to be separated at a feeding port at the current moment, correspondingly acquiring a real-time grease content sequence in which the current grease content is positioned, a real-time temperature sequence in which the current temperature is positioned and a real-time pH value sequence in which the current pH value is positioned according to the grease content time sequence data, the temperature time sequence data and the pH value time sequence data, and calculating the fluctuation correlation degree of multi-source time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time pH value sequence; acquiring a target data point interval corresponding to the current grease content in the N data point intervals, and calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the fluctuation correlation degree of multi-source time sequence data, the current temperature and the current pH value; and optimizing the current grease content by utilizing a data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after the grease is separated from the grease at the feed inlet, and calculating to obtain the grease separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet. The method comprises the steps of acquiring historical monitoring time sequence data in a historical time period between current moments, carrying out fluctuation analysis on the grease content at a lower feed inlet at the current moment, so as to obtain a data fluctuation optimization factor at the current moment, carrying out data optimization on the grease content at the lower feed inlet at the current moment by utilizing the data fluctuation optimization factor, further enabling the estimated grease separation effect to be more accurate according to the grease content difference before and after grease separation, and eliminating the estimated fluctuation of grease separation quality caused by the grease content fluctuation at the feed inlet.

Based on the same inventive concept as the above-described oil-residue separation quality evaluation method, fig. 2 shows a block diagram of an oil-residue separation quality evaluation system according to the second embodiment of the present invention, and for convenience of explanation, only the portions related to the embodiments of the present invention are shown.

Referring to fig. 2, the oil-residue separation quality evaluation system includes:

the data sequence acquisition 21 is used for acquiring time sequence data of the oil content, the temperature time sequence data and the acid-base value time sequence data of the feed inlet in a preset historical time period before the current moment according to the oil content, the temperature and the acid-base value of the feed inlet at each sampling moment in the oil production process;

the data sequence segmentation module 22 is configured to obtain a nearest neighbor data point of any one of the grease content time series data, and form a data point pair from the data point and the nearest neighbor data point corresponding to the data point, obtain all data point pairs formed in the grease content time series data, and divide the grease content time series data into N data point intervals according to all the data point pairs, where N >1;

the data fluctuation analysis module 23 is configured to obtain a current grease content, a current temperature, and a current ph of grease to be separated at the feed inlet at the current time, correspondingly obtain a real-time grease content sequence in which the current grease content is located, a real-time temperature sequence in which the current temperature is located, and a real-time ph sequence in which the current ph is located according to the grease content time sequence, the temperature time sequence, and the ph time sequence, and calculate a multi-source time sequence data fluctuation correlation at the current time according to the real-time grease content sequence, the real-time temperature sequence, and the real-time ph time sequence;

An optimization factor obtaining module 24, configured to obtain a target data point interval corresponding to the current grease content in the N data point intervals, and calculate a data fluctuation optimization factor at the current time according to a difference between the current grease content and a historical grease content included in the target data point interval, the multi-source time sequence data fluctuation correlation degree, the current temperature and the current acid-base number;

the separation quality evaluation module 25 is configured to optimize the current grease content by using the data fluctuation optimization factor to obtain an optimized current grease content, obtain the grease content of the to-be-separated grease at the feed inlet at the current time at the discharge outlet after the to-be-separated grease is subjected to the oil-residue separation, and calculate to obtain the oil-residue separation quality index at the current time according to the optimized current grease content and the grease content at the discharge outlet.

Optionally, the data sequence segmentation module 22 includes:

the difference analysis unit is used for acquiring a grease content sub-sequence with a preset length from the grease content time sequence data by taking the data point as a starting point, and calculating absolute values of data difference values between the other data points and the data points aiming at any other data point except the data point in the grease content sub-sequence;

And the neighbor confirmation unit is used for traversing all other data points except the data points in the oil-containing subsequence, correspondingly obtaining all data difference absolute values, acquiring a minimum data difference absolute value from all the data difference absolute values, and taking the other data points corresponding to the minimum data difference absolute value as nearest neighbor data points of the data points.

Optionally, the data sequence segmentation module 22 includes:

the quantity counting unit is used for counting the quantity of data point pairs formed by the data points at two sides of the data point aiming at any data point in the grease content time sequence data;

the data dividing unit is used for acquiring the number of data point pairs corresponding to each data point in the grease content time sequence data, forming a data point pair number sequence, acquiring a minimum value in the data point pair number sequence, determining the number of N-1 data point pairs according to the minimum value, taking the data points respectively corresponding to the number of the N-1 data point pairs as dividing points, and dividing the grease content time sequence data into N data point sections.

Optionally, the data fluctuation analysis module 23 includes:

the first sequence acquisition unit is used for taking the current oil content as the last data point in the real-time oil content sequence, and extracting a preset number of data points before the current moment from the oil content time sequence data to form the real-time oil content sequence;

The second sequence acquisition unit is used for taking the current temperature as the last data point in the real-time temperature sequence, and extracting a preset number of data points before the current moment from the temperature time sequence data to form the real-time temperature sequence;

and the third sequence acquisition unit is used for taking the current pH value as the last data point in the real-time pH value sequence, and extracting a preset number of data points before the current moment from the pH value time sequence data to form the real-time pH value sequence.

Optionally, the data fluctuation analysis module 23 includes:

the normalization unit is used for respectively carrying out normalization processing on the real-time temperature sequence and the real-time acid-base number sequence, and correspondingly obtaining a normalized real-time temperature sequence and a normalized real-time acid-base number sequence;

the similarity calculation unit is used for calculating a first similarity between the normalized real-time temperature sequence and the real-time oil content sequence by using a DTW algorithm, and calculating a second similarity between the real-time oil content sequence and the normalized real-time acid-base value sequence by using the DTW algorithm;

the fluctuation analysis unit is used for calculating an addition result between the first similarity and the second similarity, normalizing the addition result to obtain a normalized addition result, and taking a difference value between a preset value and the normalized addition result as the multisource time sequence data fluctuation correlation degree at the current time.

Optionally, the obtaining, by the optimization factor obtaining module 24, a target data point interval corresponding to the current fat content in the preset number of data point intervals includes:

Optionally, the optimization factor acquisition module 24 includes:

the first detection unit is used for forming current two-dimensional data from the current temperature and the current acid-base value, forming historical two-dimensional data from the temperature time sequence data and the acid-base value corresponding to each sampling time in the acid-base value time sequence data respectively, forming a historical two-dimensional data set, and performing outlier detection on the historical two-dimensional data set and the current two-dimensional data by using a COF algorithm to obtain a connectivity outlier factor of the current two-dimensional data;

the first multiplication unit is used for obtaining a first multiplication result between the connectivity outlier factor of the current two-dimensional data and the multi-source time sequence data fluctuation correlation degree;

the second detection unit is used for detecting outliers of the oil content time sequence data and the current oil content by utilizing the COF algorithm to obtain a connectivity outlier factor of the current oil content, calculating an oil content average value of all historical oil contents contained in the target data point interval, and calculating a first difference value between the current oil content and the oil content average value;

The second multiplying unit is used for obtaining a second difference value between a constant 1 and the fluctuation correlation of the multi-source time sequence data and obtaining a second multiplying result between the first difference value, the second difference value and a connectivity outlier factor of the current grease content;

and the data processing unit is used for carrying out normalization processing on the addition result between the first multiplication result and the second multiplication result, and the corresponding normalization processing result is used as the data fluctuation optimization factor at the current time.

Optionally, the optimizing the current grease content by using the data fluctuation optimizing factor in the separation quality evaluation module 25, to obtain the optimized current grease content, which includes:

Optionally, in the separation quality evaluation module 25, according to the optimized current grease content and the optimized grease content at the discharge port, the calculating obtains an oil-residue separation quality index at the current time, including:

It should be noted that, because the content of information interaction and execution process between the modules and units is based on the same concept as the method embodiment of the present invention, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein.

The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.

Claims

1. The oil-slag separation quality evaluation method is characterized by comprising the following steps of:

optimizing the current grease content by using the data fluctuation optimization factor to obtain an optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after oil-residue separation, and calculating to obtain the oil-residue separation quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet;

the acquiring the nearest neighbor data point of any data point in the grease content time sequence data comprises the following steps:

Traversing all other data points except the data points in the oil-containing subsequence, correspondingly obtaining all data difference absolute values, obtaining a minimum data difference absolute value from all data difference absolute values, and taking the other data points corresponding to the minimum data difference absolute value as nearest neighbor data points of the data points;

the step of dividing the grease content time sequence data into N data point intervals according to all data point pairs comprises the following steps:

acquiring the number of data point pairs corresponding to each data point in the oil content time sequence data, forming a data point pair number sequence, acquiring a minimum value in the data point pair number sequence, determining the number of N-1 data point pairs according to the minimum value, taking the data points respectively corresponding to the number of the N-1 data point pairs as dividing points, and dividing the oil content time sequence data into N data point sections;

the calculating the fluctuation relativity of the multisource time sequence data at the current moment according to the real-time grease content sequence, the real-time temperature sequence and the real-time acid-base number sequence comprises the following steps:

calculating an addition result between the first similarity and the second similarity, normalizing the addition result to obtain a normalized addition result, and taking a difference value between a preset value and the normalized addition result as a multisource time sequence data fluctuation correlation degree at the current time;

the calculating a data fluctuation optimization factor at the current time according to the difference between the current grease content and the historical grease content contained in the target data point interval, the multisource time sequence data fluctuation relativity, the current temperature and the current acid-base number, comprises the following steps:

performing outlier detection on the oil content time sequence data and the current oil content by utilizing the COF algorithm to obtain a connectivity outlier factor of the current oil content, calculating an oil content average value of all oil contents contained in the target data point interval, and calculating a first difference value between the current oil content and the oil content average value;

normalizing the addition result between the first multiplication result and the second multiplication result, wherein the normalization result obtained correspondingly is used as a data fluctuation optimization factor at the current time;

the step of optimizing the current grease content by using the data fluctuation optimization factor to obtain the optimized current grease content, comprising the following steps:

acquiring the historical oil content at the feed inlet at the moment before the current moment, calculating the difference value between the historical oil content and the current oil content, acquiring the product of the difference value and the data fluctuation optimization factor, and taking the addition result between the historical oil content and the product as the optimized current oil content;

The oil-slag separation quality index at the current time is calculated according to the optimized current oil content and the oil content at the discharge port, and the oil-slag separation quality index comprises the following components:

2. The oil-residue separation quality evaluation method according to claim 1, wherein the obtaining the real-time oil content sequence in which the current oil content is located, the real-time temperature sequence in which the current temperature is located, and the real-time acid-base value sequence in which the current acid-base value is located according to the oil content time sequence data, the temperature time sequence data, and the acid-base value time sequence data, respectively, includes:

3. The method for evaluating the quality of oil and sludge separation according to claim 1, wherein the obtaining a target data point interval corresponding to the current oil content in the N data point intervals includes:

4. An oil and slag separation quality assessment system, characterized in that the oil and slag separation quality assessment system comprises:

The separation quality evaluation module is used for optimizing the current grease content by utilizing the data fluctuation optimization factor to obtain the optimized current grease content, obtaining the grease content of the to-be-separated grease at the feed inlet at the current moment at the discharge outlet after the grease is subjected to the grease-to-be-separated at the feed inlet at the current moment, and calculating to obtain the grease-to-be-separated quality index at the current moment according to the optimized current grease content and the grease content at the discharge outlet;

the data sequence segmentation module comprises:

the neighbor confirmation unit is used for traversing all other data points except the data points in the oil-containing subsequence, correspondingly obtaining all data difference absolute values, obtaining a minimum data difference absolute value from all data difference absolute values, and taking the other data points corresponding to the minimum data difference absolute value as nearest neighbor data points of the data points;

The data sequence segmentation module comprises:

the data dividing unit is used for acquiring the number of data point pairs corresponding to each data point in the oil content time sequence data, forming a data point pair number sequence, acquiring a minimum value in the data point pair number sequence, determining the number of N-1 data point pairs according to the minimum value, taking the data points respectively corresponding to the number of the N-1 data point pairs as dividing points, and dividing the oil content time sequence data into N data point sections;

the data fluctuation analysis module comprises:

The fluctuation analysis unit is used for calculating an addition result between the first similarity and the second similarity, normalizing the addition result to obtain a normalized addition result, and taking a difference value between a preset value and the normalized addition result as a multisource time sequence data fluctuation correlation degree at the current time;

the optimization factor acquisition module comprises:

the data processing unit is used for carrying out normalization processing on the addition result between the first multiplication result and the second multiplication result, and the corresponding normalization processing result is used as the data fluctuation optimization factor at the current time;

optimizing the current grease content by utilizing the data fluctuation optimization factor in a separation quality evaluation module to obtain the optimized current grease content, wherein the method comprises the following steps:

and calculating an oil-residue separation quality index at the current time according to the optimized current oil content and the oil content at the discharge port in a separation quality evaluation module, wherein the oil-residue separation quality index comprises the following components: