CN117804971A - Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis - Google Patents
Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis Download PDFInfo
- Publication number
- CN117804971A CN117804971A CN202311855996.XA CN202311855996A CN117804971A CN 117804971 A CN117804971 A CN 117804971A CN 202311855996 A CN202311855996 A CN 202311855996A CN 117804971 A CN117804971 A CN 117804971A
- Authority
- CN
- China
- Prior art keywords
- data
- oil
- oil product
- temperature
- monitoring
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 125
- 238000004458 analytical method Methods 0.000 title claims abstract description 77
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000001514 detection method Methods 0.000 claims abstract description 54
- 230000008859 change Effects 0.000 claims abstract description 51
- 238000005457 optimization Methods 0.000 claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000007405 data analysis Methods 0.000 claims abstract description 18
- 230000007246 mechanism Effects 0.000 claims abstract description 15
- 230000002159 abnormal effect Effects 0.000 claims abstract description 14
- 230000001105 regulatory effect Effects 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims description 36
- 230000003044 adaptive effect Effects 0.000 claims description 23
- 238000010438 heat treatment Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 11
- 238000012417 linear regression Methods 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 11
- 238000004088 simulation Methods 0.000 claims description 10
- 238000003756 stirring Methods 0.000 claims description 10
- 230000007613 environmental effect Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 8
- 230000001276 controlling effect Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000001816 cooling Methods 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 6
- 230000001960 triggered effect Effects 0.000 claims description 6
- 238000000611 regression analysis Methods 0.000 claims description 5
- 230000000704 physical effect Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000003908 quality control method Methods 0.000 claims description 3
- 230000008054 signal transmission Effects 0.000 claims description 3
- 230000006866 deterioration Effects 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008844 regulatory mechanism Effects 0.000 description 1
- 230000007363 regulatory process Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Abstract
The invention discloses an oil product intelligent monitoring judging method and system based on self-adaptive trend analysis, and relates to the technical field of oil product monitoring; comprising S1: monitoring oil products in real time, and S2: judging the oil quality, S3: judging the early warning level, S4: parameter automatic adjustment and S5: data analysis and optimization; according to the invention, the real-time monitoring of the oil product is realized through the online oil product detection main device, so that the pollution and deterioration problems in the traditional offline detection mode are avoided; the monitored data is analyzed by adopting a self-adaptive algorithm and a trend analysis method, so that the judgment accuracy is improved, and the requirements of different oil types and running conditions are met; according to the analysis result and the change trend of the data, the early warning level of the oil product is determined, so that abnormal conditions of the oil product can be found in time and corresponding measures can be taken; and starting an automatic regulating mechanism to automatically regulate parameters according to the early warning level and the analysis result, and regulating according to different early warning levels and the analysis result so as to maintain the oil product within a normal range.
Description
Technical Field
The invention relates to the technical field of oil product monitoring, in particular to an oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis.
Background
The existing online oil detection mode mainly collects signals through an oil sensor, and detection is completed through the steps of signal conversion, noise reduction, amplification, data output and the like. However, the conventional off-line oil detection method has some problems, such as pollution and deterioration of the oil during the sampling and transportation processes, resulting in deviation of the off-line detection result from the actual situation.
In addition, because the viscosity index of the oil is influenced by factors such as specific oil type and detected operation condition, the traditional offline detection method cannot always obtain an accurate index capable of reflecting actual oil products, and particularly for equipment operated for a long time, real-time detection and timely maintenance are required.
Disclosure of Invention
The invention aims to provide an intelligent oil product monitoring and judging method and system based on self-adaptive trend analysis, which realize real-time monitoring and quality state judgment of oil products through on-line monitoring equipment and a judging model, determine early warning levels of the oil products according to analysis results and trend changes, automatically adjust parameters to maintain normal ranges, and utilize monitoring data to carry out deep analysis and correction.
The aim of the invention can be achieved by the following technical scheme:
the application provides an oil intelligent monitoring and judging method based on self-adaptive trend analysis, which comprises the following steps:
s1: monitoring the oil product in real time, and adopting an online oil product detection main device and online monitoring equipment to monitor the oil product in real time to obtain monitoring data;
s2: judging the quality of the oil product, establishing a judging model according to the data monitored in real time, processing and analyzing the monitored oil product data by using the judging model, analyzing by adopting a self-adaptive algorithm, and judging the quality state of the oil product to obtain a judging result;
s3: judging the early warning level, determining the early warning level of the oil product according to the judging result of the judging model and the change trend of the monitoring data, and carrying out early warning notification of the corresponding early warning level according to different abnormal conditions of the oil product;
s4: automatically adjusting parameters according to the early warning level and the judgment result of the judgment model, automatically adjusting the parameters through an automatic adjusting mechanism, and selecting an adjusting mode to adjust according to different early warning levels and analysis results;
s5: and (3) data analysis and optimization, namely performing deep data analysis and optimization by using the monitored data, verifying the accuracy of the data, correcting the data, eliminating abnormal values or errors, building a prediction model by using the monitored data, predicting the quality of oil products, and optimizing the prediction model by using historical data and algorithms.
Preferably, a set of contrast auxiliary device is additionally arranged in the main online oil product detection device, wherein the contrast auxiliary device comprises a heating part, a temperature control part, a detection part and a signal transmission and receiving part; the method comprises the steps of detecting a temperature index, a flow rate index and a viscosity index obtained by a temperature sensor, a flow rate sensor and a C1 viscosity sensor in a main device on line; the contrast auxiliary device is divided into a lower layer belonging to a heating part, a middle layer belonging to a detecting part and an upper layer belonging to a temperature control part; after the temperature indexes of the temperature sensor are received, the temperature is output from the main part of the system of the online detection device, and the heating element or the air cooling device is driven to maintain the temperature of the standard oil product at the same temperature; the middle detection part is connected with the stirring device through the temperature control part, the flow rate index of the flow rate sensor is output from the system main part of the online detection device, and the feedback product flow rate is simulated.
Preferably, the real-time monitoring of the oil product comprises monitoring temperature, flow rate and pressure indicators,
a temperature sensor is introduced to monitor the temperature change of the oil product in real time;
Introducing a flow velocity sensor to monitor the flow velocity change of the oil product in real time;
and a pressure sensor is introduced to monitor the pressure change of the oil product in real time.
Preferably, the method comprises establishing a judgment model according to the real-time monitored data, and processing and analyzing the oil product data, including
Preprocessing the monitored low-viscosity oil data, and extracting characteristics related to the oil state from the preprocessed data;
then, an adaptive algorithm is carried out by utilizing adaptive filtering, the characteristic vector is processed, and parameters are dynamically adjusted; the adaptive filtering calculation formula is as follows:
w(n+1)＝w(n)+μ*e(n)*x(n)；
wherein mu is a step size parameter, e (n) is an error signal, an input feature vector is x (n), an expected output is d (n), an estimated output of the system is y (n), and a weight w (n) of the filter is updated recursively;
then analyzing the processed data by a trend analysis method and adopting a regression analysis method of multiple linear regression analysis, and establishing a linear function model by utilizing dependent variables and independent variables;
wherein, multiple linear regression analysis method:
input feature vector x 1 ,x 2 ,...,x p Wherein p is the number of arguments;
the target value (or output) is y;
regression equation: y=β 0 +β 1 x 1 +β 2 x 2 +...+β p x p Wherein beta is 0 For the intercept, beta 1 ,β 2 ,...,β p Coefficients for each argument;
during multiple linear regression analysis, intercept and coefficients are found so that a regression equation is fitted with actual observed data, and future data is predicted or estimated by using the regression equation.
Preferably, the establishing a judging model to obtain a judging result, mapping the monitored real-time temperature data to the viscosity for classification by establishing a temperature-viscosity relation model, wherein the judging model comprises,
the first temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value is greater than or equal to a first preset temperature value and less than or equal to a second preset temperature value, the oil product is classified into a normal viscosity grade;
the second temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value exceeds a second preset temperature value, the oil product is classified into a high viscosity grade;
mapping the monitored real-time flow rate data to viscosity for classification by modeling the flow rate versus viscosity, wherein,
the first flow rate judgment, the monitored real-time flow rate data is compared with a pre-established flow rate and viscosity relation model, and if the flow rate value is greater than or equal to a first preset flow rate value and less than or equal to a second preset flow rate value, the oil product is classified as a normal viscosity grade;
Judging the second flow rate, comparing the monitored real-time flow rate data with a pre-established flow rate and viscosity relation model, and classifying the oil product into a high viscosity grade if the flow rate value exceeds a second preset flow rate value;
mapping the monitored real-time pressure value data to viscosity for classification by modeling the relationship of pressure to viscosity, wherein,
the first pressure judgment, wherein the monitored real-time pressure data is compared with a pre-established pressure and viscosity relation model, and if the pressure value is greater than or equal to a first preset pressure value and less than or equal to a second preset pressure value, the oil product is classified into a normal viscosity grade;
and judging the second pressure, comparing the monitored real-time pressure data with a pre-established pressure and viscosity relation model, and classifying the oil product into a high viscosity grade if the pressure value exceeds a second preset pressure value.
Preferably, the judging result of the judging model and the change trend of the monitoring data determine the early warning level of the oil product, including the first machine early warning, the second stage early warning and the third stage early warning,
the first-stage early warning is carried out, when the monitored real-time temperature data is smaller than a first preset temperature value or the real-time flow velocity value is smaller than the first preset flow velocity value or the real-time pressure value is smaller than or equal to the first preset pressure value, the first-stage early warning is judged to be of a low viscosity level, a first early warning signal is triggered, the temperature of the oil product is indicated to be lower than a preset temperature range, and the viscosity is lower;
The second-stage early warning is carried out, when the temperature value monitored in real time is larger than or equal to a first preset temperature value and smaller than or equal to a second preset temperature value, or the flow rate value is larger than or equal to the first preset flow rate value and smaller than or equal to the second preset flow rate value, or the pressure value is larger than or equal to the first preset pressure value and smaller than or equal to the second preset pressure value, the second early warning is judged to be normal, and the second early warning is started;
and when the temperature value monitored in real time is larger than a second preset temperature value, or the flow velocity value is larger than the second preset flow velocity value, or the pressure value is larger than the second preset pressure value, the third-stage early warning is judged to be of a high viscosity level, and a third early warning signal is triggered.
Preferably, in the oil quality control, the automatic adjustment mechanism dynamically adjusts parameters according to the values of the real-time monitoring data and the judging result, the dynamic adjustment mechanism comprises a first adjustment mode, a second adjustment mode and a third adjustment mode,
when the monitored real-time temperature data is lower than a first preset temperature value, the heating part receives the temperature index of the temperature sensor and outputs a temperature signal from the main part of the online detection system, and the heating element is driven to provide heat according to the temperature signal;
The monitored real-time temperature data is higher than a second preset temperature value, and after the temperature control part receives the temperature index of the temperature sensor, a temperature signal is output from the main part of the online detection system, and the temperature control part controls the working state of the air cooling equipment according to the signal;
the second adjusting mode is used for comparing the flow rate data monitored in real time with a first preset flow rate value or comparing the flow rate data monitored in real time with a second preset flow rate value, adjusting the rotating speed of the stirring device according to a signal output by a flow rate index of the flow rate sensor from the main part of the online detection system, and controlling the flow rate of the oil product by adjusting the rotating speed of the stirring device;
the third adjusting mode is to compare the pressure value data monitored in real time with a first preset pressure value or a second preset pressure value, and the pressure control device adjusts the pressure in the system by controlling the opening of the valve or adjusting the rotating speed of the pump according to the pressure index obtained by the pressure sensor.
Preferably, the pressure regulating process in the oil product detection process is recorded, the pressure of the oil product is regulated through the pressure control device in the oil product detection process, the operation and the parameter setting adopted in the regulation process are recorded,
Monitoring the change curves of all environmental variables and the physical properties of the oil products, and simultaneously closely monitoring the environmental variables related to oil product detection, and knowing the influence of environmental factors and the oil product properties on the pressure regulation process by recording the change curves;
based on analysis of the adjusting process and the change curve, a simulation adjusting model is established for describing the relation among various variables in the pressure adjusting process and predicting the change trend of the oil pressure under different parameter settings;
gradually generating an adjusting standard range and an optimal state range of the oil product, and gradually generating the adjusting standard range and the optimal state range of the oil product by continuously updating the simulation adjusting model, wherein the optimal state range of the oil product refers to a range in which the oil product achieves optimal performance and quality in a pressure adjusting process;
and applying the simulation regulation model to the pressure regulation of the same oil, applying the established simulation regulation model to the actual pressure regulation, calculating an error value, and correcting and optimizing the optimal state range of the oil according to the calculated error value and the actual observation result.
Preferably, the data analysis and optimization establishes a predictive model, collects a large amount of oil quality data, including monitoring data, cleans and processes the collected data for model training data sets, selects the most relevant features of the oil for modeling, uses the existing monitoring data sets for model training, and adjusts model parameters to obtain optimal performance,
According to the established prediction model, the actual monitoring data is used for predicting the oil quality, and the prediction result is compared with the required oil quality;
in the verification process, the predicted result is found to need to be corrected, more monitoring data are collected, the dynamic change of the oil product performance is obtained by analyzing the change trend of the data, the correction of the predicted model is assisted, and the predicted result is corrected and optimized by carrying out deep analysis and correction on the collected data;
and optimizing and adjusting the prediction model according to the newly collected data, adjusting parameters in the model, gradually optimizing the parameters, performing verification again after parameter adjustment and model optimization, and evaluating the corrected model.
Oil intelligent monitoring judging system based on self-adaptation trend analysis, include
The monitoring module is used for introducing an online oil product detection main device and online monitoring equipment to monitor the oil product in real time and timely acquire key data of the oil product;
the judging module is used for establishing a judging model according to the data acquired by the monitoring module, and processing and analyzing the monitored data through a self-adaptive algorithm and a trend analysis method so as to judge the state of the oil product;
The early warning module is used for determining early warning levels of the oil products according to the analysis result of the judging module and the change trend of the monitoring data, wherein the early warning levels are classified into different levels, and early warning notification is carried out according to the early warning levels;
the automatic adjusting module starts an automatic adjusting mechanism to adjust according to the early warning level of the early warning module and the analysis result of the judging module;
and the data analysis and optimization module is used for carrying out deep analysis and optimization according to the monitoring data, establishing a prediction model, predicting the quality of oil products aiming at different oil products, and then verifying and correcting.
The beneficial effects of the invention are as follows:
(1) The real-time monitoring of the oil products is realized through the online oil product detection main device and the online monitoring equipment, so that pollution and deterioration problems possibly existing in the traditional offline detection mode are avoided; the judgment model is utilized for processing and analyzing, and the self-adaptive algorithm and the trend analysis method are adopted for analyzing the monitored data to judge the quality state of the oil product, so that the judgment accuracy can be improved and the requirements of different oil product types and running conditions can be met;
(2) According to the analysis result and the change trend of the data, the early warning level of the oil product is determined, and corresponding early warning notification is carried out, so that abnormal conditions of the oil product can be found in time, corresponding measures can be taken, and the normal operation of equipment can be ensured; according to the early warning level and the analysis result, an automatic adjusting mechanism is started to automatically adjust parameters, and according to different early warning levels and analysis results, a proper adjusting mode is selected to adjust so as to maintain the oil product within a normal range;
(3) And (3) performing deep data analysis and optimization by using the monitoring data, verifying the accuracy of the data, correcting, eliminating abnormal values or errors, establishing a prediction model, performing optimization by using historical data and algorithms, and improving the accuracy of monitoring and judgment.
Drawings
For a better understanding and implementation, the technical solutions of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a step flowchart of an oil intelligent monitoring and judging method based on adaptive trend analysis provided in embodiment 1 of the present application;
fig. 2 is a schematic structural diagram of an oil intelligent monitoring and judging system based on adaptive trend analysis according to embodiment 1 of the present application.
Detailed Description
For further explanation of the technical means and effects adopted by the present invention for achieving the intended purpose, exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application. Rather, they are merely examples of methods and systems that are consistent with aspects of the present application, as detailed in the accompanying claims.
The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
The following detailed description of specific embodiments, features and effects according to the present invention is provided with reference to the accompanying drawings and preferred embodiments.
Example 1
Referring to fig. 1-2, the present embodiment provides an intelligent oil product monitoring and judging method and system based on adaptive trend analysis, which realizes real-time monitoring and quality state judgment of an oil product through an online monitoring device and a judging model, determines an early warning level of the oil product according to an analysis result and trend change, automatically adjusts parameters to maintain a normal range, and uses monitoring data to perform deep analysis and correction.
The invention provides an oil intelligent monitoring and judging method based on self-adaptive trend analysis, which comprises the following steps:
S1: monitoring the oil product in real time by adopting an online oil product detection main device and online monitoring equipment to obtain monitoring data, wherein the monitoring data can comprise key indexes such as temperature, flow rate, pressure and the like;
s2: judging the quality of the oil product, establishing a judging model according to the data monitored in real time, processing and analyzing the monitored oil product data by using the judging model, extracting key characteristics from the monitored data, and analyzing by adopting a self-adaptive algorithm to judge whether the quality state of the oil product is normal or not so as to obtain a judging result;
s3: judging early warning levels, determining early warning levels of oil products according to the judging result of the judging model and the change trend of the monitoring data, wherein the early warning levels can be divided into different levels, such as first-level early warning, second-level early warning and third-level early warning, and carrying out corresponding early warning notification according to conditions;
s4: automatically adjusting parameters according to the early warning level and the judgment result of the judgment model, automatically adjusting the parameters through an automatic adjusting mechanism, and selecting a proper adjusting mode to adjust according to different early warning levels and analysis results so as to ensure the stability of the oil quality and meet the requirements;
s5: and (3) data analysis and optimization, namely performing deep data analysis and optimization by using the monitoring data, verifying the accuracy of the data, correcting the data, eliminating abnormal values or errors, building a prediction model by using the corrected monitoring data, predicting the quality of oil products, and optimizing the prediction model by using historical data and algorithms, thereby improving the efficiency and quality of the production process.
Through the organic combination of the steps, the oil quality can be monitored in real time, judged in state, automatically regulated and analyzed and optimized in data, so that the production efficiency and the quality level are improved.
In this embodiment, a set of contrast auxiliary device is additionally installed in the online oil product detection main device, where the contrast auxiliary device includes a heating portion, a temperature control portion, a detection portion, and a signal transmission and reception portion; the method comprises the steps of detecting a temperature index, a flow rate index and a viscosity index obtained by a temperature sensor, a flow rate sensor and a C1 viscosity sensor in a main device on line; the contrast auxiliary device is divided into a lower layer belonging to a heating part, a middle layer belonging to a detecting part and an upper layer belonging to a temperature control part; after the temperature indexes of the temperature sensor are received, the temperature is output from the main part of the online detection system, and the heating element or the air cooling device is driven to maintain the temperature of the standard oil product at the same temperature; the middle detection part outputs the flow rate index of the flow rate sensor from the main part of the online detection system through the stirring device connected with the temperature control part, and simulates the fed-back product flow rate, wherein the oil flow rate is related to the rotation speed of the stirring device;
In addition, the detection part is internally provided with standard oil samples of the same type, wherein the detection part is immersed in the rotating device without filling up; in addition, the oil path structure in the detection part can be designed according to actual conditions, such as considering the conditions of pipeline pressure, or oil operation splashing and the like; in addition, the detection part can be externally connected with a C2 viscosity sensor at the beginning; and (3) comparing the data output by the C1 viscosity sensor and the C2 viscosity sensor in real time, drawing a viscosity index change curve of the standard oil product and the actual oil product along with temperature change, and judging the actual oil product condition according to the difference value of the viscosity index change curve and the viscosity index change curve of the standard oil product and the actual oil product.
In this embodiment, the real-time monitoring of the oil product includes monitoring temperature, flow rate and pressure indicators,
a temperature sensor is introduced to monitor the temperature of the oil product in real time, the temperature is one of important factors influencing the viscosity of the oil product, and the viscosity state of the oil product can be accurately judged by monitoring the temperature change;
the flow rate sensor is introduced to monitor the flow rate of the oil product in real time, the flow rate also has an influence on the viscosity of the oil product, and the fluidity and viscosity characteristics of the oil product can be better understood by monitoring the change of the flow rate;
pressure sensor is introduced to monitor the pressure of the oil in real time, the pressure change possibly affects the viscosity of the oil, and the viscosity condition of the oil can be more comprehensively known by monitoring the pressure change.
Through the real-time monitoring of temperature, flow rate and pressure index, can judge the viscosity state of oil more accurately, understand mobility and viscosity characteristic of oil to master the viscosity condition of oil comprehensively. This will help to optimize the production process, adjust the process parameters and ensure that the oil quality meets the requirements.
In this embodiment, the establishing a judgment model according to the data monitored in real time processes and analyzes the oil data, including
The monitored low-viscosity oil data are subjected to pretreatment operations such as cleaning, denoising, correction and the like, and characteristics related to the oil state are extracted from the pretreated data, wherein the characteristics comprise temperature change rate, flow velocity trend, pressure change amplitude and the like;
processing the feature vector by using an adaptive algorithm such as adaptive filtering, adaptive threshold value and the like, and dynamically adjusting parameters according to actual conditions so as to adapt to data changes and noise interference under different scenes; the adaptive filtering calculation formula is as follows:
w(n+1)＝w(n)+μ*e(n)*x(n)；
wherein mu is a step size parameter, e (n) is an error signal, an input feature vector is x (n), an expected output is d (n), an estimated output of the system is y (n), and a weight w (n) of the filter is updated recursively;
In the trend analysis method, a multiple linear regression analysis/multiple nonlinear regression analysis method is adopted to analyze the processed data, and a nonlinear/linear function model is established by utilizing a dependent variable and an independent variable, so that the square sum of the deviation average differences between the theoretical value and the observed value of the model is minimum, the long-term trend and the change mode of the data are helped to be captured, and the future state development is predicted;
wherein, multiple linear regression analysis method:
input feature vector x 1 ,x 2 ,...,x p Wherein p is the number of arguments;
the target value (or output) is y;
regression equation: y=β 0 +β 1 x 1 +β 2 x 2 +...+β p x p Wherein beta is 0 For the intercept, beta 1 ,β 2 ,...,β p Coefficients for each argument;
multiplex nonlinear regression analysis method:
input feature vector x 1 ,x 2 ,...,x p Wherein p is an auto-changeQuantity of the amount;
the target value (or output) is y;
regression equation: y=f (x 1 ,x 2 ,...,x p ) Where f represents a nonlinear function, different functional forms can be selected according to the actual problem;
in performing multiple linear regression or multiple nonlinear regression analysis, the goal is to find the best parameter estimates (i.e., intercept and coefficients) so that the regression equation fits best to the actual observed data, and future data is predicted or estimated using the regression equation.
In the judging module, the monitored low-viscosity oil data are preprocessed through feature extraction and the following data are assumed:
in the feature extraction process, preprocessing operations such as cleaning, denoising, correction and the like are performed on the data. Assuming that the following data were obtained after pretreatment:
next, in the adaptive algorithm method, the feature vector is processed using the adaptive filtering algorithm, assuming that we select the step size parameter μ to be 0.1 and set the initial weight w0 to be [0, 0], the weight w (n) of the filter can be recursively updated using the following formula:
w(n+1)＝w(n)+μ*e(n)*x(n)
where e (n) is the error signal, x (n) is the eigenvector, assuming the desired output d (n) is
[0.04,0.05,2000], the estimated output y (n) of the system is [0.03,0.04,1800], and iterative calculation is performed according to a formula to obtain updated weight w (n+1) as follows:
w(1)＝[0,0,0]+0.1*([0.04,0.05,2000]-[0.03,0.04,1800])*[0.02,2.5,10000]
＝[0,0,0]+0.1*[0.01,0.01,200]*[0.02,2.5,10000]
＝[0,0,0]+[0.000002,0.000025,0.2]
＝[0.000002,0.000025,0.2]
w(2)＝[0.000002,0.000025,0.2]+0.1*([0.04,0.05,2000]-[0.03,0.04,1800])*[0.03,3.0,11000]
＝[0.000002,0.000025,0.2]+0.1*[0.01,0.01,200]*[0.03,3.0,11000]
＝[0.000002,0.000025,0.2]+[0.000003,0.00003,2]
＝[0.000005,0.000055,2.2]
w(3)＝[0.000005,0.000055,2.2]+0.1*([0.04,0.05,2000]-[0.03,0.04,1800])*[0.01,2.8,10500]
＝[0.000005,0.000055,2.2]+0.1*[0.01,0.01,200]*[0.01,2.8,10500]
＝[0.000005,0.000055,2.2]+[0.000001,0.000028,2]
＝[0.000006,0.000083,4.2]
the final updated weight is w (3) = [0.000006,0.000083,4.2].
In the trend analysis method, a linear function model is established by using a multiple linear regression analysis method, assuming that the temperature change rate (x 1 ) Trend of flow velocity (x 2 ) As an independent variable, the pressure variation amplitude (y) is taken as a target value, and a regression equation can be expressed as:
y＝β 0 +β 1 x 1 +β 2 x 2
The parameter beta can be estimated by using least square method 0 、β 1 And beta 2 . Let us assume that we have the following data:
multiple linear regression analysis is carried out by utilizing the data, and the parameter estimation value of the regression equation is obtained as follows:
β 0 ＝700；β 1 ＝30000；β 2 ＝-2000
the final linear function model established is: y=700+30000 x 1 -2000x 2
By means of the model, future pressure change amplitude can be predicted according to given temperature change rate and flow velocity trend.
In this embodiment, the method includes creating a judgment model to obtain a judgment result, mapping the monitored real-time temperature data to viscosity for classification by creating a temperature-viscosity relationship model, wherein the judgment model includes,
the first temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value is greater than or equal to a first preset temperature value and less than or equal to a second preset temperature value, the oil product is classified into a normal viscosity grade;
the second temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value exceeds a second preset temperature value, the oil product is classified into a high viscosity grade;
mapping the monitored real-time flow rate data to viscosity for classification by modeling the flow rate versus viscosity, wherein,
The first flow rate judgment, the monitored real-time flow rate data is compared with a pre-established flow rate and viscosity relation model, and if the flow rate value is greater than or equal to a first preset flow rate value and less than or equal to a second preset flow rate value, the oil product is classified as a normal viscosity grade;
judging the second flow rate, comparing the monitored real-time flow rate data with a pre-established flow rate and viscosity relation model, and classifying the oil product into a high viscosity grade if the flow rate value exceeds a second preset flow rate value;
mapping the monitored real-time pressure value data to viscosity for classification by modeling the relationship of pressure to viscosity, wherein,
the first pressure judgment, wherein the monitored real-time pressure data is compared with a pre-established pressure and viscosity relation model, and if the pressure value is greater than or equal to a first preset pressure value and less than or equal to a second preset pressure value, the oil product is classified into a normal viscosity grade;
and judging the second pressure, comparing the monitored real-time pressure data with a pre-established pressure and viscosity relation model, and classifying the oil product into a high viscosity grade if the pressure value exceeds a second preset pressure value.
By comparing the relation models, the monitored real-time temperature, flow speed and pressure data can be mapped to the corresponding viscosity classifications, so that the judgment of the viscosity state of the oil product is realized, abnormal conditions can be found in time, the viscosity control is carried out, and the quality of the oil product is ensured to meet the requirements.
In the embodiment, the judgment result of the judgment model and the change trend of the monitoring data determine the early warning level of the oil product, including the first machine early warning, the second stage early warning and the third stage early warning,
the first-stage early warning is carried out, when the monitored real-time temperature data is smaller than a first preset temperature value or the real-time flow velocity value is smaller than the first preset flow velocity value or the real-time pressure value is smaller than or equal to the first preset pressure value, the low-viscosity level is judged, and a first early warning signal is triggered, so that the temperature of the oil product is lower than a preset temperature range, and the viscosity is lower;
a second-stage early warning, wherein when the temperature value monitored in real time is larger than or equal to a first preset temperature value and smaller than or equal to a second preset temperature value, or the flow rate value is larger than or equal to the first preset flow rate value and smaller than or equal to a second preset flow rate value, or the pressure value is larger than or equal to the first preset pressure value and smaller than or equal to the second preset pressure value, the second early warning is started to inform the detector that the oil product is normal;
and a third-stage early warning, wherein when the temperature value monitored in real time is larger than a second preset temperature value, or the flow rate value is larger than a second preset flow rate value, or the pressure value is larger than a second preset pressure value, the high viscosity level is judged, and a third early warning signal is triggered.
Through different early warning levels, early warning signals can be timely sent out according to the monitored real-time temperature, flow speed and pressure data so as to remind related personnel to pay attention to the oil state, necessary measures are taken to ensure that the oil quality meets the requirements, the occurrence or expansion of potential problems can be avoided, and the safety and stability of the production process are ensured.
In this embodiment, in the oil quality control, the automatic adjustment mechanism dynamically adjusts parameters according to the values of the real-time monitoring data and the determination result, the dynamic adjustment mechanism includes a first adjustment mode, a second adjustment mode and a third adjustment mode,
when the monitored real-time temperature data is lower than a first preset temperature value, the heating part receives a temperature index of the temperature sensor and outputs a temperature signal from the main part of the online detection system, and according to the signal, the heating element is driven to provide heat so that the temperature of the standard oil product is maintained within the same temperature range;
the monitored real-time temperature data is higher than a second preset temperature value, the temperature control part outputs a temperature signal from the main part of the online detection system after receiving the temperature index of the temperature sensor, and the temperature control part controls the working state of the air cooling equipment according to the signal so as to maintain the temperature of the standard oil product within a set range;
The second adjusting mode is used for comparing the flow rate data monitored in real time with a first preset flow rate value or comparing the flow rate data monitored in real time with a second preset flow rate value, adjusting the rotating speed of the stirring device according to a signal output by the flow rate index of the flow rate sensor from the main part of the online detection system, and controlling the flow rate of the oil product by adjusting the rotating speed of the stirring device so as to meet the required flow rate index;
and the third regulation mode is to compare the pressure value data monitored in real time with a first preset pressure value or a second preset pressure value, regulate the pressure control device according to the requirement according to the pressure index obtained by the pressure sensor, and regulate the pressure in the system by controlling the opening of the valve or regulating the rotation speed of the pump so as to reach the required pressure range.
Through different adjustment modes, the temperature, the flow speed and the pressure of the oil product can be automatically adjusted according to the real-time monitoring data, so that the oil product can be kept in a set range, the stability and the efficiency of the production process can be improved, and the quality of the oil product can be ensured to meet the requirements.
In this embodiment, the process of adjusting the pressure during the oil product detection process is recorded: in the oil product detection process, the pressure of the oil product is regulated through a pressure control device, and the operation and parameter setting adopted in the regulation process are recorded, including key information such as the flow of a regulating pump, the opening of a valve and the like.
And (3) monitoring the change curves of all environmental variables and the physical properties of the oil product, and simultaneously closely monitoring the environmental variables related to oil product detection, such as temperature, humidity, gas components and the like, and the physical properties of the oil product, such as viscosity, density and the like. By recording these curves, the impact of environmental factors and oil properties on the pressure regulation process can be understood.
Based on the analysis of the adjustment process and the change curve, a simulated adjustment model is established, which can be a mathematical model or a model based on a machine learning algorithm, for describing the relationship between the variables in the pressure adjustment process and predicting the change trend of the oil pressure under different parameter settings.
The adjusting standard range and the optimal state range of the oil product are gradually generated, and the adjusting standard range and the optimal state range of the oil product can be gradually generated by continuously updating the simulation adjusting model. The adjusting standard range refers to a reasonable range for adjusting the pressure of the oil product under the specific process requirement; the optimal state range of the oil product refers to the range in which the oil product achieves optimal performance and quality in the pressure adjusting process.
And applying the established simulation regulation model to the actual pressure regulation step in the pressure regulation of the same oil liquid type, so as to guide and assist the regulation operation. And observing whether the oil liquid subjected to adjustment is in the optimal state range of the oil product in the pressure adjustment process, and calculating an error value to evaluate the adjustment effect.
And correcting and optimizing the optimal state range of the oil product according to the calculated error value and the actual observation result, and further accurately controlling and adjusting other devices such as a pump and the like by continuously narrowing the range so as to ensure that the oil product can reach the optimal state more stably and accurately in the pressure adjusting process.
In this embodiment, the data analysis and optimization establishes a predictive model, collects a large amount of oil quality data, including monitoring data, cleans and processes the collected data, uses the data set for model training, selects the most relevant features of the oil for modeling, uses the existing monitoring data set for model training, and adjusts model parameters to obtain optimal performance,
according to the established prediction model, the existing monitoring data are used for predicting the oil quality, the prediction result is compared with the required oil quality, the accuracy and the reliability of the prediction are evaluated, and the accuracy and the deviation of the prediction result are evaluated by adopting methods such as statistical indexes, error analysis and the like;
the difference degree between the predicted value and the actual value is measured through an average absolute error and a root mean square error, and the average absolute error MAE calculation formula is as follows:
Mae= (1/n) ×Σ|predicted value-actual value|;
the root mean square error RMSE calculation formula is:
rmse=sqrt ((1/n) ×Σ (predicted value-actual value) 2 )；
The smaller the values of the mean absolute error and the root mean square error, the more accurate the prediction result.
The determining coefficient R2 represents the fitting degree of the model to the observed data, and the calculation formula is as follows:
r2=1- (Σ (predicted value-actual value) 2/Σ (actual value-average value) 2);
a value ranging from 0 to 1, with a value closer to 1 indicating a model that fits better.
For example, for data verification, the established prediction model is used for predicting the quality of the oil product of the existing monitoring data, and the prediction result is compared with the actual value to evaluate the accuracy and reliability of the prediction; the following monitoring data and corresponding actual oil quality are assumed:
according to a linear function model established before: y=700+30000 x 1 -2000x 2 The corresponding predicted oil quality can be calculated,
next, statistical indicators may be used to evaluate the accuracy and bias of the predicted outcome.
First calculate the Mean Absolute Error (MAE):
MAE＝(1/3)*(|88.4-85|+|94.6-90|+|82.2-80|)
＝(1/3)*(3.4+4.6+2.2)
＝10.2/3
≈3.4
root Mean Square Error (RMSE) is then calculated:
RMSE＝sqrt((1/3)*((88.4-85) 2 +(94.6-90) 2 +(82.2-80) 2 ))
＝sqrt((1/3)*(11.56+21.16+4.84))
＝sqrt(12.1867)
≈3.49
finally, the decision coefficient (R 2 ) Firstly, calculating an average value of actual oil quality:
average value of actual oil quality= (85+90+80)/3
＝255/3
＝85
R is then calculated 2 ：
R 2 ＝1-((88.4-85) 2 +(94.6-90) 2 +(82.2-80) 2 )/((85-85) 2 +(90-85) 2 +(80-85) 2 )
＝1-(10.96+20.16+4.84)/(0+25+25)
＝1-36.96/50
≈0.2688
Based on the calculation result, the Mean Absolute Error (MAE) was about 3.4, the Root Mean Square Error (RMSE) was about 3.49, and the coefficient (R 2 ) About 0.2688, smaller mean absolute and root mean square errors and closer to 0 determination coefficients indicate that there is some deviation between the predicted result and the actual value.
In the verification process, larger deviation or inaccuracy of the predicted result is found, more monitoring data are required to be corrected, more monitoring data are collected by increasing the frequency and the number of data acquisition, monitoring points are increased, sampling frequency is increased or monitoring time period is increased to obtain more data samples, continuous time series data are collected to track the evolution of oil quality along with time, dynamic change of oil performance is better known by analyzing the change trend of the data, correction of a prediction model is assisted, and the predicted result is corrected and optimized by further analyzing and correcting the collected data by using a proper statistical method, a machine learning algorithm or the prediction model.
According to the newly collected data, optimizing and adjusting the prediction model to improve the accuracy and reliability of prediction, adjusting parameters in the model according to the verification result and the actually observed deviation, gradually optimizing the parameters to enable the model to better fit the actual situation and provide a more accurate prediction result, and after parameter adjustment and model optimization, performing the verification step again to evaluate the corrected model. If deviations or inaccuracy are found to still exist, the data, optimization model and adjustment parameters can be collected continuously, and iterative verification and correction processes can be performed until the predicted effect is reached.
Oil intelligent monitoring judging system based on self-adaptation trend analysis, include
The monitoring module is used for introducing an online oil product detection main device and online monitoring equipment to monitor the oil product in real time and timely acquire key data of the oil product;
the judging module is used for establishing a judging model according to the data acquired by the monitoring module, and processing and analyzing the monitored data through a self-adaptive algorithm and a trend analysis method so as to judge the state of the oil product;
the early warning module is used for determining early warning levels of the oil products according to the analysis result of the judging module and the change trend of the monitoring data, wherein the early warning levels can be classified into different levels, and corresponding early warning notification is carried out according to the early warning levels;
the automatic adjusting module starts an automatic adjusting mechanism to adjust according to the early warning level of the early warning module and the analysis result of the judging module so as to ensure that the quality and performance of the oil product meet the requirements;
and the data analysis and optimization module is used for carrying out deep analysis and optimization according to the monitoring data, establishing a prediction model, predicting the quality of the oil products aiming at different oil products, and verifying and correcting the predicted quality of the oil products so as to improve the efficiency and quality of the production process.
The monitoring module is responsible for monitoring the oil product in real time, acquiring key data of the oil product through an online oil product detection main device and online monitoring equipment, transmitting the data of the oil product monitored in real time to the judging module, establishing a judging model by the judging module, and processing and analyzing the monitoring data by utilizing a self-adaptive algorithm and a trend analysis method so as to judge the state of the oil product;
The system comprises a judging module and an early warning module, wherein the analysis result of the judging module is transmitted to the early warning module, and the early warning module determines early warning levels of oil products according to the analysis result of the judging module and the change trend of monitoring data, and the early warning levels are divided according to set rules and threshold values so as to remind operators or trigger subsequent adjustment measures;
the automatic regulation module starts an automatic regulation mechanism to automatically regulate parameters according to the early warning level and the analysis result of the judging module, and the automatic regulation module selects a proper regulation mode according to different early warning levels and the analysis result so as to ensure the stability of the oil quality and meet the requirements;
the analysis result of the judgment module can also be used for the data analysis and optimization module, the data analysis and optimization module utilizes the monitoring data to conduct deep analysis and optimization, a prediction model is built, the quality of the oil product is predicted according to different oil products, and the data analysis and optimization module eliminates abnormal values or errors through verification and correction, so that the efficiency and quality of the production process are improved.
Through cooperation of all modules, real-time monitoring, state judgment, early warning notification, automatic adjustment and data analysis and optimization of the oil quality can be realized, so that the production efficiency is improved, the risk is reduced, and the oil quality is ensured to meet the requirements.
Example 2
The embodiment realizes the real-time monitoring and judgment of the oil quality based on the on-line sensor and the intelligent algorithm, solves the problems of pollution and deterioration of links such as oil sampling, transportation and the like, overcomes the limitations of the traditional off-line detection mode, and can timely discover and treat abnormal conditions of the oil by establishing a monitoring model, judging the quality in real time and informing the early warning, thereby ensuring the normal operation of equipment and the stability of the oil quality.
The embodiment provides an oil intelligent monitoring and judging method based on self-adaptive trend analysis, which comprises the following steps,
the on-line sensor is selected and calibrated, and a proper on-line sensor is selected and calibrated according to specific requirements of equipment and oil properties, and can measure a plurality of indexes of the oil, such as viscosity, moisture, temperature and the like;
the data acquisition and real-time transmission are carried out, the on-line sensor converts the acquired data into digital signals, and the data are transmitted to the central processing system through the real-time data transmission channel, so that the accuracy and the real-time performance of the data are ensured;
Preprocessing data and extracting features, wherein a central processing system preprocesses the transmitted data, including removing noise, filtering, detecting abnormal values and the like, and then extracting key features, such as the change rate of viscosity, the percentage of moisture and the like, from the preprocessed data;
establishing a monitoring model, and establishing an oil quality monitoring model by using a machine learning algorithm and historical data; through training the model, the model can learn and identify the mode and trend between the normal and abnormal oil states;
and inputting the data acquired in real time into a monitoring model for real-time quality judgment and early warning. If the model detects that the oil product is abnormal, triggering an early warning mechanism, and timely sending early warning information to related personnel through an early warning notification system;
and (3) automatically adjusting and optimizing, automatically adjusting equipment parameters or executing corresponding control strategies according to quality judgment and early warning results so as to stabilize the quality of oil products, and simultaneously, performing model optimization and parameter updating by utilizing real-time monitoring data so as to improve monitoring accuracy and adaptability.
The present invention is not limited to the above embodiments, but is capable of modification and variation in detail, and other modifications and variations can be made by those skilled in the art without departing from the scope of the present invention.
Claims (10)
1. An oil intelligent monitoring judging method based on self-adaptive trend analysis is characterized in that: the method comprises the following steps:
s1: monitoring the oil product in real time, and adopting an online oil product detection main device and online monitoring equipment to monitor the oil product in real time to obtain monitoring data;
s2: judging the quality of the oil product, establishing a judging model according to the data monitored in real time, processing and analyzing the monitored oil product data by utilizing the judging model, extracting key characteristics, analyzing by adopting a self-adaptive algorithm, and judging the quality state of the oil product to obtain a judging result;
s3: judging the early warning level, determining the early warning level of the oil product according to the judging result of the judging model and the change trend of the monitoring data, and carrying out early warning notification of the corresponding early warning level according to different abnormal conditions of the oil product;
s4: automatically adjusting parameters according to the early warning level and the judgment result of the judgment model, automatically adjusting the parameters through an automatic adjusting mechanism, and selecting an adjusting mode to adjust according to different early warning levels and analysis results;
s5: and (3) data analysis and optimization, namely performing deep data analysis and optimization by using the monitoring data, verifying the accuracy of the data, correcting the data, eliminating abnormal values or errors, finally establishing a prediction model by using the corrected monitoring data, predicting the quality of oil products, and optimizing the prediction model by using historical data and algorithms.
2. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: a set of contrast auxiliary device is additionally arranged in the online oil product detection main device, wherein the contrast auxiliary device comprises a heating part, a temperature control part, a detection part and a signal transmission and receiving part; the method comprises the steps of detecting a temperature index, a flow rate index and a viscosity index obtained by a temperature sensor, a flow rate sensor and a C1 viscosity sensor in a main device on line; the contrast auxiliary device is divided into a lower layer belonging to a heating part, a middle layer belonging to a detecting part and an upper layer belonging to a temperature control part; after the temperature indexes of the temperature sensor are received, the temperature is output from the main part of the system of the online detection device, and the heating element or the air cooling device is driven to maintain the temperature of the standard oil product at the same temperature; the middle detection part is connected with the stirring device through the temperature control part, the flow rate index of the flow rate sensor is output from the system main part of the online detection device, and the feedback product flow rate is simulated.
3. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: the real-time monitoring of the oil product includes monitoring temperature, flow rate and pressure indicators,
A temperature sensor is introduced to monitor the temperature change of the oil product in real time;
introducing a flow velocity sensor to monitor the flow velocity change of the oil product in real time;
and a pressure sensor is introduced to monitor the pressure change of the oil product in real time.
4. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: the method for establishing a judging model according to the data of the real-time monitoring to process and analyze the data of the oil product comprises the following steps of
Preprocessing the monitored low-viscosity oil data, and extracting characteristics related to the oil state from the preprocessed data;
then, an adaptive algorithm is carried out by utilizing adaptive filtering, the characteristic vector is processed, and parameters are dynamically adjusted; the adaptive filtering calculation formula is as follows:
w(n+1)＝w(n)+μ*e(n)*x(n)；
wherein mu is a step size parameter, e (n) is an error signal, an input feature vector is x (n), an expected output is d (n), an estimated output of the system is y (n), and a weight w (n) of the filter is updated recursively;
then analyzing the processed data by a trend analysis method and adopting a regression analysis method of multiple linear regression analysis, and establishing a linear function model by utilizing dependent variables and independent variables;
Wherein, multiple linear regression analysis method:
input feature vectors are x1, x2, xp, where p is the number of arguments;
the target value (or output) is y;
regression equation: y=β0+β1x1+β2x2+ + βpxp, where β0 is the intercept, β1, β2, # βp is the coefficient of each independent variable;
during multiple linear regression analysis, intercept and coefficients are found so that a regression equation is fitted with actual observed data, and future data is predicted or estimated by using the regression equation.
5. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: the method comprises establishing a judgment model to obtain a judgment result, mapping the monitored real-time temperature data to viscosity for classification by establishing a temperature-viscosity relation model, wherein the judgment model comprises,
the first temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value is greater than or equal to a first preset temperature value and less than or equal to a second preset temperature value, the oil product is classified into a normal viscosity grade;
the second temperature judgment is carried out, the monitored real-time temperature data is compared with a pre-established temperature and viscosity relation model, and if the temperature value exceeds a second preset temperature value, the oil product is classified into a high viscosity grade;
Mapping the monitored real-time flow rate data to viscosity for classification by modeling the flow rate versus viscosity, wherein,
the first flow rate judgment, the monitored real-time flow rate data is compared with a pre-established flow rate and viscosity relation model, and if the flow rate value is greater than or equal to a first preset flow rate value and less than or equal to a second preset flow rate value, the oil product is classified as a normal viscosity grade;
judging the second flow rate, comparing the monitored real-time flow rate data with a pre-established flow rate and viscosity relation model, and classifying the oil product into a high viscosity grade if the flow rate value exceeds a second preset flow rate value;
mapping the monitored real-time pressure value data to viscosity for classification by modeling the relationship of pressure to viscosity, wherein,
the first pressure judgment, wherein the monitored real-time pressure data is compared with a pre-established pressure and viscosity relation model, and if the pressure value is greater than or equal to a first preset pressure value and less than or equal to a second preset pressure value, the oil product is classified into a normal viscosity grade;
and judging the second pressure, comparing the monitored real-time pressure data with a pre-established pressure and viscosity relation model, and classifying the oil product into a high viscosity grade if the pressure value exceeds a second preset pressure value.
6. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: the judging result of the judging model and the change trend of the monitoring data determine the early warning level of the oil product, including the first machine early warning, the second stage early warning and the third stage early warning,
the first-stage early warning is carried out, when the monitored real-time temperature data is smaller than a first preset temperature value or the real-time flow velocity value is smaller than the first preset flow velocity value or the real-time pressure value is smaller than or equal to the first preset pressure value, the first-stage early warning is judged to be of a low viscosity level, a first early warning signal is triggered, the temperature of the oil product is indicated to be lower than a preset temperature range, and the viscosity is lower;
the second-stage early warning is carried out, when the temperature value monitored in real time is larger than or equal to a first preset temperature value and smaller than or equal to a second preset temperature value, or the flow rate value is larger than or equal to the first preset flow rate value and smaller than or equal to the second preset flow rate value, or the pressure value is larger than or equal to the first preset pressure value and smaller than or equal to the second preset pressure value, the second early warning is judged to be normal, and the second early warning is started;
and when the temperature value monitored in real time is larger than a second preset temperature value, or the flow velocity value is larger than the second preset flow velocity value, or the pressure value is larger than the second preset pressure value, the third-stage early warning is judged to be of a high viscosity level, and a third early warning signal is triggered.
7. The intelligent oil monitoring and judging system based on adaptive trend analysis according to claim 1, wherein the system is characterized in that: in the oil quality control, the automatic regulating mechanism dynamically adjusts parameters according to the values of real-time monitoring data and judging results, the dynamic regulating mechanism comprises a first regulating mode, a second regulating mode and a third regulating mode,
when the monitored real-time temperature data is lower than a first preset temperature value, the heating part receives the temperature index of the temperature sensor and outputs a temperature signal from the main part of the online detection system, and the heating element is driven to provide heat according to the temperature signal;
the monitored real-time temperature data is higher than a second preset temperature value, and after the temperature control part receives the temperature index of the temperature sensor, a temperature signal is output from the main part of the online detection system, and the temperature control part controls the working state of the air cooling equipment according to the signal;
the second adjusting mode is used for comparing the flow rate data monitored in real time with a first preset flow rate value or comparing the flow rate data monitored in real time with a second preset flow rate value, adjusting the rotating speed of the stirring device according to a signal output by a flow rate index of the flow rate sensor from the main part of the online detection system, and controlling the flow rate of the oil product by adjusting the rotating speed of the stirring device;
The third adjusting mode is to compare the pressure value data monitored in real time with a first preset pressure value or a second preset pressure value, and the pressure control device adjusts the pressure in the system by controlling the opening of the valve or adjusting the rotating speed of the pump according to the pressure index obtained by the pressure sensor.
8. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 2, wherein the method is characterized by comprising the following steps of: recording the process of adjusting the pressure in the oil product detection process, adjusting the pressure of the oil product through a pressure control device in the oil product detection process, recording the operation and parameter setting adopted in the adjustment process,
monitoring the change curves of all environmental variables and the physical properties of the oil products, and simultaneously closely monitoring the environmental variables related to oil product detection, and knowing the influence of environmental factors and the oil product properties on the pressure regulation process by recording the change curves;
based on analysis of the adjusting process and the change curve, a simulation adjusting model is established for describing the relation among various variables in the pressure adjusting process and predicting the change trend of the oil pressure under different parameter settings;
Gradually generating an adjusting standard range and an optimal state range of the oil product, and gradually generating the adjusting standard range and the optimal state range of the oil product by continuously updating the simulation adjusting model, wherein the optimal state range of the oil product refers to a range in which the oil product achieves optimal performance and quality in a pressure adjusting process;
and applying the simulation regulation model to the pressure regulation of the same oil, applying the established simulation regulation model to the actual pressure regulation, calculating an error value, and correcting and optimizing the optimal state range of the oil according to the calculated error value and the actual observation result.
9. The intelligent oil monitoring and judging method based on the adaptive trend analysis according to claim 1, wherein the method is characterized by comprising the following steps of: the data analysis and optimization, the establishment of a predictive model, the collection of a large amount of oil quality data, including monitoring data, the cleaning and processing of the collected data, the use of the data set for model training, the selection of the most relevant features of the oil for modeling, the use of the existing monitoring data set for model training, and the adjustment of model parameters to obtain the best performance,
according to the established prediction model, the actual monitoring data is used for predicting the oil quality, and the prediction result is compared with the required oil quality;
In the verification process, the predicted result is found to need to be corrected, more monitoring data are collected, the dynamic change of the oil product performance is obtained by analyzing the change trend of the data, the correction of the predicted model is assisted, and the predicted result is corrected and optimized by carrying out deep analysis and correction on the collected data;
and optimizing and adjusting the prediction model according to the newly collected data, adjusting parameters in the model, gradually optimizing the parameters, performing verification again after parameter adjustment and model optimization, and evaluating the corrected model.
10. Oil intelligent monitoring judging system based on self-adaptation trend analysis, its characterized in that: the system comprises a monitoring module, an online oil product detection main device and online monitoring equipment, wherein the monitoring module is used for monitoring oil products in real time and acquiring key data of the oil products in time;
the judging module is used for establishing a judging model according to the data acquired by the monitoring module, and processing and analyzing the monitored data through a self-adaptive algorithm and a trend analysis method so as to judge the state of the oil product;
the early warning module is used for determining early warning levels of the oil products according to the analysis result of the judging module and the change trend of the monitoring data, wherein the early warning levels are classified into different levels, and early warning notification is carried out according to the early warning levels;
The automatic adjusting module starts an automatic adjusting mechanism to adjust according to the early warning level of the early warning module and the analysis result of the judging module;
and the data analysis and optimization module is used for carrying out deep analysis and optimization according to the monitoring data, establishing a prediction model, predicting the quality of oil products aiming at different oil products, and then verifying and correcting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311855996.XA CN117804971A (en) | 2023-12-29 | 2023-12-29 | Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311855996.XA CN117804971A (en) | 2023-12-29 | 2023-12-29 | Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117804971A true CN117804971A (en) | 2024-04-02 |
Family
ID=90434524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311855996.XA Pending CN117804971A (en) | 2023-12-29 | 2023-12-29 | Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117804971A (en) |
-
2023
- 2023-12-29 CN CN202311855996.XA patent/CN117804971A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108681633B (en) | Condensate pump fault early warning method based on state parameters | |
| CN112527788A (en) | Method and device for detecting and cleaning abnormal value of transformer monitoring data | |
| CN111275288A (en) | XGboost-based multi-dimensional data anomaly detection method and device | |
| CN110378042B (en) | Wind turbine generator gearbox oil temperature anomaly detection method and system based on SCADA data | |
| CN106649755B (en) | Threshold value self-adaptive setting abnormity detection method for multi-dimensional real-time power transformation equipment data | |
| CN117113729B (en) | Digital twinning-based power equipment online state monitoring system | |
| JP7142257B2 (en) | Deterioration diagnosis system additional learning method | |
| CN112598144A (en) | CNN-LSTM burst fault early warning method based on correlation analysis | |
| CN117193222A (en) | Intelligent quality control system based on industrial Internet of things and big data and control method thereof | |
| CN114638435A (en) | Diesel engine security parameter prediction method based on data driving | |
| CN115827411A (en) | Online monitoring and operation and maintenance evaluation system and method for automation equipment | |
| CN113539382B (en) | Early warning positioning method and system for key technological parameters of dimethyl phosphite | |
| CN117639602A (en) | Self-adaptive motor running state adjusting method and system | |
| CN117804971A (en) | Oil product intelligent monitoring and judging method and system based on self-adaptive trend analysis | |
| CN117542169A (en) | Automatic equipment temperature abnormality early warning method based on big data analysis | |
| CN116602435A (en) | Machine learning-based method for analyzing moisture change of cut tobacco in cut tobacco making machine | |
| CN116127831A (en) | Soft measurement method for difficult-to-measure parameters of heavy gas turbine | |
| CN113325824B (en) | Regulating valve abnormity identification method and system based on threshold monitoring | |
| CN112651444B (en) | Self-learning-based non-stationary process anomaly detection method | |
| CN112560339B (en) | Method for predicting guide bearing bush temperature of hydroelectric generating set by utilizing machine learning | |
| CN111027719A (en) | Maintenance optimization method for multi-component system state opportunity | |
| CN117308275B (en) | Temperature difference-based pipeline connection abnormality detection method and system | |
| CN112396344A (en) | Chemical process reliability online evaluation method based on product quality | |
| CN115200296B (en) | Ice machine group control method, device, equipment and computer readable storage medium | |
| CN117273402B (en) | Energy-saving management system and method for glass deep processing production line based on Internet of Things technology |
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 |