CN111090939A - Early warning method and system for abnormal working condition of petrochemical device - Google Patents
Early warning method and system for abnormal working condition of petrochemical device Download PDFInfo
- Publication number
- CN111090939A CN111090939A CN201911302436.5A CN201911302436A CN111090939A CN 111090939 A CN111090939 A CN 111090939A CN 201911302436 A CN201911302436 A CN 201911302436A CN 111090939 A CN111090939 A CN 111090939A
- Authority
- CN
- China
- Prior art keywords
- data
- real
- related parameters
- alarm
- result data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The embodiment of the invention discloses an early warning method and a system for abnormal working conditions of a petrochemical device, relating to the technical field of petrochemical, wherein the early warning method comprises the following steps: providing a basic prediction model, process related parameters and real-time process data of petrochemical process data; obtaining prediction result data according to the process related parameters and the basic prediction model; judging whether the numerical value variation trends of the prediction result data and the real-time process data are consistent or not; if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, further judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value; and if the difference value between the prediction result data and the real-time process data is larger than the first alarm threshold value, alarming. The invention can automatically identify the abnormal working condition of the petrochemical process and can carry out early warning.
Description
Technical Field
The embodiment of the invention relates to the technical field of petrochemical industry, in particular to an early warning method and system for abnormal working conditions of a petrochemical device.
Background
Domestic petrochemical enterprises face huge challenges of excess capacity and aggravated market competition, bear the pressure of intensive process manufacturing and extremely complex production process and high-safety environment requirements, and along with continuous improvement of automation degree of production equipment and continuous reduction of personnel, the intelligent production device is especially important for timely prediction and early warning of abnormal working conditions of the device and realization of intelligent production.
At present, data processing and application of petrochemical enterprises are mostly limited to statistics and query, and potential values contained in data are far from being mined.
In fact, the petrochemical production process is a complex physicochemical change process, factors influencing the process are multiple, the mechanism is very complex, linear means description and accurate equation complete expression are difficult to use, and a simple mathematical model is difficult to meet the requirement of the petrochemical industry on high reliability of a model operation result. Therefore, how to explore, exploit, refine and model the large data pool in the petrochemical industry and obtain valuable information from the large data pool has great significance for safe production, optimized management, quality improvement and efficiency improvement.
Disclosure of Invention
Therefore, the embodiment of the invention provides an early warning method and system for the abnormal working condition of a petrochemical device, which can identify the abnormal working condition of a petrochemical process and can perform early warning.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
the embodiment of the first aspect of the invention discloses a method for early warning abnormal working conditions of a petrochemical device, which comprises the following steps: providing a basic prediction model, process related parameters and real-time process data of petrochemical process data; obtaining prediction result data according to the process related parameters and the basic prediction model; judging whether the numerical value variation trends of the prediction result data and the real-time process data are consistent or not; if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value; and if the difference value between the prediction result data and the real-time process data is larger than the first alarm threshold value, alarming.
Further, the determining whether the difference between the predicted result data and the real-time process data is greater than a first alarm threshold specifically includes: and windowing the deviation value by adopting a data-based time sequence prediction method, taking the last data in each window as a target variable, then performing time sequence prediction by using a support vector machine, and judging whether the predicted deviation and the actual deviation obtained according to the basic prediction model are greater than the first alarm threshold value.
Further, still include: when the process related parameters are changed, adjusting the first alarm threshold value according to the changed process related parameters to obtain a second alarm threshold value; and judging whether to alarm or not according to the second alarm threshold value.
Further, the obtaining of the second alarm threshold according to the first alarm preset specifically includes: and adjusting the first alarm threshold value according to the changed process related parameters and the Euclidean distance of the process related parameters before the change to obtain a second alarm threshold value.
Further, the alarming comprises: providing a fault tree; selectively alarming according to the prediction result data and the real-time process data based on the fault tree.
The embodiment of the second aspect of the invention discloses an early warning system for the abnormal working condition of a petrochemical device, which comprises: the system comprises a providing module, a prediction module and a processing module, wherein the providing module is used for providing a basic prediction model of petrochemical process data, process related parameters and real-time process data; the alarm module is used for giving an alarm; the control processing module is used for obtaining prediction result data according to the process related parameters and the basic prediction model and judging whether the numerical variation trends of the prediction result data and the real-time process data are consistent; if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value; and if the difference value between the prediction result data and the real-time process data is larger than the first alarm threshold value, alarming through the alarm module.
Further, the control processing module is specifically configured to perform windowing on the deviation value by using a data-based time sequence prediction method, use the last data in each window as a target variable, perform time sequence prediction by using a support vector machine, and determine whether a predicted deviation and an actual deviation obtained according to the basic prediction model are greater than the first alarm threshold.
The monitoring module is used for monitoring whether the process related parameters are changed; and the control processing module is further used for adjusting the first alarm threshold value according to the changed process related parameters to obtain a second alarm threshold value after the process related parameters are changed, and judging whether the alarm module needs to give an alarm according to the second alarm threshold value.
Further, the control processing module is further configured to adjust the first alarm threshold according to the changed process-related parameter and the euclidean distance of the process-related parameter before the change, so as to obtain a second alarm threshold.
Further, the providing module is further used for providing a fault tree, and the control processing module is further used for selectively alarming according to the prediction result data and the real-time process data based on the fault tree.
The invention has the following advantages:
forecasting is carried out through a given basic forecasting model and relevant process parameters to obtain forecasting result data, trend comparison and difference judgment are carried out on the forecasting result data and real-time process data, and then alarming is carried out selectively. In addition, whether process related parameters of petrochemical process equipment are changed or not can be monitored, and an alarm threshold value is adjusted according to the changed process related parameters, so that false alarm is reduced. The invention can automatically identify the abnormal working condition of the petrochemical process and can carry out early warning.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
The structures, ratios, sizes, and the like shown in the present specification are only used for matching with the contents disclosed in the specification, so that those skilled in the art can understand and read the present invention, and do not limit the conditions for implementing the present invention, so that the present invention has no technical significance, and any structural modifications, changes in the ratio relationship, or adjustments of the sizes, without affecting the functions and purposes of the present invention, should still fall within the scope covered by the contents disclosed in the present invention.
FIG. 1 is a flowchart illustrating a petrochemical device abnormal condition warning method according to an embodiment of the present invention;
FIG. 2 is a schematic hyperplane view of an embodiment of the present invention supporting vector regression;
FIG. 3 is a schematic diagram of the tremor algorithm in an embodiment of the present invention;
fig. 4 is a block diagram illustrating a structure of an early warning device for an abnormal operating condition of a petrochemical device according to an embodiment of the present invention.
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "connected" and "connected" are to be interpreted broadly, e.g., as being fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
FIG. 1 is a flowchart illustrating an early warning method for abnormal operating conditions of a petrochemical device according to an embodiment of the present disclosure. As shown in fig. 1, the method for warning abnormal operating conditions of a petrochemical device according to the embodiment of the present invention includes:
s1: and providing a basic prediction model of petrochemical process data, process related parameters and real-time process data.
Specifically, the basic prediction model of the petrochemical process data may be trained in an off-line manner and provided to the corresponding terminal device for data prediction.
As one example of the present invention, a worker collects historical data of a petrochemical process plant, performs data validation cleaning and checks for missing repairs. The data verification, cleaning and leakage detection and defect supplement comprise time matching, invalid point elimination, adaptation of low-frequency data to high-frequency data rules and data loss supplement in a large-span time period. For example, historical data of a plurality of parameters of the petrochemical process device 2017 and 2018 year are extracted according to the frequency per hour, or requirements are provided for users, the users derive data from a DCS or other systems, whether the historical data have values at each time point and whether the state of each time point data is 'Good' are checked, if missing data are inserted according to an interpolation rule, if part of the values are not logical (judged by a graph), more detailed data need to be checked, confirmed and modified.
And finally determining relevant parameters of the process through whole-process mechanism influence analysis, multi-data historical trend comparison analysis and correlation analysis. For example, the target parameter is the bed differential pressure of a certain tower, the process related parameters such as inlet flow, tower bottom liquid level, tower top reflux amount and the like are selected firstly from process mechanism analysis, other most probably related bit numbers are selected by utilizing correlation analysis, probably related parameters are found out by utilizing multi-parameter trend graph comparison, and the process related parameters of the model are formed by combining the parameters and then are provided for the terminal equipment.
And (3) establishing a dependent variable and an independent variable of the model based on the process related parameters, establishing an initial prediction model by combining historical training data by means such as mathematical regression and neural network, importing historical test data for test comparison, continuously regressing to correct the weight of the neuron, and finally obtaining the basic prediction model after the model meets the precision requirement.
Real-time process data can be obtained through the self-system of the petrochemical device.
S2: and obtaining prediction result data according to the process related parameters and the basic prediction model, namely inputting the process related parameters into the basic prediction model, and outputting the prediction result data by the basic prediction model.
S3: and judging whether the numerical value change trends of the prediction result data and the real-time process data are consistent. Wherein, the numerical value variation trend is the variation trend of numerical value increase and decrease. For example, at the time from t to t +1, if the numerical value of the prediction result data is increased and the numerical value of the real-time process data is increased, the numerical change trends of the prediction result data and the real-time process data are judged to be consistent; and if the numerical value of the prediction result data is increased and the numerical value of the real-time process data is decreased, judging that the numerical value change trends of the prediction result data and the real-time process data are inconsistent.
S4: and if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is greater than a preset first alarm threshold value.
S5: and if the difference value between the prediction result data and the real-time process data is larger than a first alarm threshold value, alarming.
In an embodiment of the present invention, in step S4, the determining whether the difference between the predicted result data and the real-time process data is greater than a first alarm threshold specifically includes: and windowing the deviation value by adopting a data-based time sequence prediction method, taking the last data in each window as a target variable, then performing time sequence prediction by using a support vector machine, and judging whether the predicted deviation and the actual deviation obtained according to the basic prediction model are greater than a first alarm threshold value.
Specifically, a set first algorithm (herein referred to as a shadow algorithm) is used to determine whether to alarm. The shadow algorithm uses SVR (support vector regression) of RBF (Radial Basis Function) network in terms of modeling. The working principle of the model is that a function related to a target variable is fitted by collecting data information related to various attributes of a certain main body, accurate fitting and further prediction are realized, a regression algorithm is very suitable, the linear regression algorithm is not only long-lived in prediction analysis, but also most easy to explain function fitting, and SVR regression is selected after verification because the relation among petrochemical process parameters does not exist linearly.
FIG. 2 is a schematic hyperplane view of an embodiment of the present invention supporting vector regression. As shown in fig. 2, the RBF neural network is a three-layer neural network including an input layer, a hidden layer, and an output layer. The transformation from the input space to the hidden layer space is nonlinear, while the transformation from the hidden layer space to the output layer space is linear, and meanwhile, a mechanism layer is added into a single hidden layer by considering the complexity in process production. So as to adjust the error learning and overfitting in the learning process at any time.
The shadow algorithm is similar to the shadow tracking algorithm in the game in the aspect of realizing the concept of predicting alarm, the time factor is considered, the deviation is tracked, a time sequence prediction method based on data is adopted for the deviation, the deviation value is divided into windows, the last data in each window is used as a target variable, then the SVM is used for completing time sequence prediction, and whether the predicted deviation and the actual deviation obtained by the basic prediction model are larger than a first alarm threshold value or not is judged. And if the alarm is greater than the first alarm threshold, alarming, for example, through the integration of a short message interface and a mail interface, and transmitting alarm messages and reasons to terminals of related workers according to configured message transmission rules.
In one embodiment of the present invention, the alarming of step S5 includes: providing a fault tree; and selectively alarming according to the prediction result data and the real-time process data based on the fault tree.
In one example of the present invention, the condition setting of the fault tree includes:
when the model output value is greater than 0.8, outputting information: when the pressure difference reaches the scaling critical value, scaling phenomenon occurs;
when the model output value is greater than 0.85, the output information is as follows: severe scaling, requiring close observation and preparing maintenance consumables;
when the model output value is greater than 0.9, the output information is as follows: if the scale formation is serious, please clean the wastewater or adjust the treatment capacity.
Thus, when the output value is 0.82, the alarm message displays: when the pressure difference reaches the scaling critical value, the scaling phenomenon occurs.
In an embodiment of the present invention, after step S2, the method further includes: when the process related parameters are changed, adjusting the first alarm threshold value according to the changed process related parameters to obtain a second alarm threshold value; and judging whether to alarm or not according to the second alarm threshold value. And adjusting the first alarm threshold value according to the changed process related parameters and the Euclidean distance of the process related parameters before the change to obtain a second alarm threshold value.
FIG. 3 is a schematic diagram of the tremor algorithm in an embodiment of the present invention. As shown in fig. 3, the present invention uses a second algorithm (referred to herein as the tremor algorithm) to adjust the alarm threshold. The tremor algorithm is used for some model points which run more smoothly (such as temperature difference control), and the principle is distance-based outlier detection. Euclidean distance is a statistical method of data used to compute the distance between two high-dimensional vectors, and euclidean distance is a commonly used definition of distance, which refers to the true distance between two points in an m-dimensional space, or the natural length of a vector (i.e., the distance of the point from the origin). The euclidean distance in two and three dimensions is the actual distance between two points.
The early warning method for the abnormal working condition of the petrochemical device can specifically perform the following early warning:
1. monitoring and early warning the activity of the catalyst of the fixed bed reactor: evaluating the activity of the cracking and refining catalyst; monitoring the inactivation rate of the cracking and refining catalyst on line; predicting a future catalyst deactivation rate; the method has the advantages that the abnormal conditions of the inactivation rates of the cracking and refining catalysts are warned in time, and the production operation is effectively guided; guiding and regulating the temperature of a cracking and refining catalyst bed layer; predicting the service life of the cracking and refining catalyst, and accurately setting the regeneration and replacement period of the catalyst and the like; for production optimization, such as: optimizing raw material properties (different feeding proportions), optimizing catalyst deactivation rate (life cycle), optimizing conversion rate, optimizing processing load and the like, and the method helps to obviously improve economic benefits.
2. Monitoring and early warning the pressure drop of a reactor bed layer: monitoring the reactor bed layer differential pressure in real time on line; under normal conditions, the device can be used as a soft measuring instrument; predicting the variation trend of the differential pressure of the reactor bed; the condition that the differential pressure of the bed layer has abnormal change is early warned in time, and the production operation is effectively guided; is one of the important indexes for predicting the operation period of the device and the regeneration and replacement period of the catalyst; this model (in conjunction with the catalyst deactivation model) was used for production optimization.
3. Monitoring and early warning of temperature rise of a catalyst bed layer: monitoring the temperature rise condition of each catalyst bed layer in real time on line; the temperature rise of an abnormal catalyst bed layer is early warned, and the phenomenon of temperature runaway or overtemperature is prevented; guiding to balance and adjust the reaction load among different catalyst beds of the reactor; the gradient temperature rise of the bed layer is balanced, and the deactivation rates of different bed layers of the same type of catalyst are controlled to be synchronous.
4. Monitoring radial fluid distribution and early warning abnormity of a hydrogenation reactor: monitoring the distribution uniformity of the fluid of each bed layer on line in real time; and early warning the phenomenon of temperature difference abnormity.
5. Monitoring and early warning of coking of a furnace tube of the heating furnace: monitoring the load distribution balance condition of each branch of the heating furnace; early warning is carried out on the abnormal condition of load distribution; and (4) monitoring and evaluating the running condition of each furnace tube on line, and early warning the occurrence of the coking condition of the furnace tubes in advance.
6. Monitoring and early warning temperature deviation of each branch of the multi-branch condenser: monitoring the balanced distribution condition of cooling load of each branch of the condenser; preventing liquid impact; the phenomenon of abnormal temperature difference of the branch is early warned in time, and the salt deposition and corrosion rates are reduced; in northern areas, the anti-freezing and anti-condensation function can be achieved in winter.
7. Liquid level sensing abnormity early warning: monitoring the liquid level sensing system behaviors related to liquid level indication, control and alarm linkage of the tower and the container in real time on line; for the condition of abnormal liquid level change, early warning of 'too fast liquid level rise' or 'too fast liquid level fall' is sent out in advance, and operation guidance prompt is assisted; the early warning of abnormal liquid level sensing is timely sent out when a certain set of liquid level sensing system is abnormal, so that operation fluctuation, self-protection misoperation and even safety accidents caused by false liquid level are prevented; manual inspection of the liquid level is replaced, and the labor intensity of operators is reduced; increase HSE levels.
8. High-pressure heat exchanger scale deposit (salt) early warning: the online real-time monitoring trades the scale deposit phenomenon characteristic change situation, to having the operating mode of scale deposit phenomenon, just can discover and send "there is salt deposition phenomenon" early warning in the initial stage of the incident model, through discovering early, taking measures early, effectively reduces the production loss that consequently causes, reduces operation consumption and maintenance cost.
9. Early warning of internal leakage of the high-pressure heat exchanger: the characteristic value is monitored on line in real time through the model, the initial leakage stage of the high-pressure heat exchanger is timely found, and the potential fault is early warned in advance.
The early warning method for the abnormal working condition of the petrochemical device has the advantages of simple and practical modeling and short domestication time; the monitoring and early warning are accurate, and the robustness is good; modular design and strong generalization performance; the system can be operated on the existing informatization platform of an enterprise, and the system does not need to be built to increase large investment for the informatization foundation platform; by effectively monitoring and accurately predicting the production working condition on line, production optimization can be carried out according to the life cycle of the asset, and the production intellectualization of the device is realized; by accurately identifying and timely early warning abnormal working conditions, management, technology and operators can be helped to sense and scientifically decide the abnormal working conditions of the device in an early stage, time is saved for effective treatment, treatment cost is reduced, and key technology support is provided for intelligent and quick early warning response of device production, so that the daily work burden of related personnel is obviously reduced, the occurrence of various emergencies is effectively prevented and controlled, and the safe, stable and long-period operation of the device is ensured; four capabilities are obviously improved: comprehensive perception, optimization cooperation, prediction early warning and scientific decision-making capability; the method has better expansion performance, and with the lapse of operation time and the increase of demand, the types and the number of Agents are continuously increased, and finally, a system with full functions and full area coverage can be formed, so that the conversion from local intelligence to ubiquitous intelligence is realized.
Fig. 4 is a block diagram illustrating a structure of an early warning device for an abnormal operating condition of a petrochemical device according to an embodiment of the present invention. As shown in fig. 4, the warning device for the abnormal operating condition of the petrochemical device according to the embodiment of the present invention includes: a module 100, an alarm module 200 and a control processing module 300 are provided.
The providing module 100 is configured to provide a basic prediction model of petrochemical process data, process-related parameters, and real-time process data. The control processing module 300 is configured to obtain prediction result data according to the process-related parameters and the basic prediction model, and determine whether the numerical variation trends of the prediction result data and the real-time process data are consistent; if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value; and if the difference value between the prediction result data and the real-time process data is greater than a first alarm threshold value, alarming through an alarm module 200.
In an embodiment of the present invention, the control processing module 300 is specifically configured to perform windowing on the deviation value by using a data-based time sequence prediction method, take the last data in each window as a target variable, perform time sequence prediction by using a support vector machine, and determine whether a predicted deviation and an actual deviation obtained according to a basic prediction model are greater than a first alarm threshold.
In an embodiment of the present invention, the warning system for abnormal operating conditions of a petrochemical plant further includes a monitoring module, and the monitoring module is configured to monitor whether process-related parameters change. The control processing module 300 is further configured to, after the process related parameter is changed, adjust the first alarm threshold according to the changed process related parameter to obtain a second alarm threshold, and determine whether an alarm needs to be performed through the alarm module 200 according to the second alarm threshold.
In an embodiment of the present invention, the control processing module 300 is further configured to adjust the first alarm threshold according to the euclidean distance between the changed process-related parameter and the process-related parameter before the change, so as to obtain the second alarm threshold.
In one embodiment of the present invention, the providing module 100 is further configured to provide a fault tree. The control processing module 300 is also configured to selectively alert based on the fault tree based on the predicted outcome data and the real-time process data.
It should be noted that, a specific implementation manner of the warning system for the abnormal operating condition of the petrochemical device in the embodiment of the present invention is similar to a specific implementation manner of the warning method for the abnormal operating condition of the petrochemical device in the embodiment of the present invention, and specific reference is specifically made to the description of the warning method for the abnormal operating condition of the petrochemical device, and no further description is given for reducing redundancy.
In the description herein, references to the description of "one embodiment" and "an example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (10)
1. The early warning method for the abnormal working condition of the petrochemical device is characterized by comprising the following steps of:
providing a basic prediction model, process related parameters and real-time process data of petrochemical process data;
obtaining prediction result data according to the process related parameters and the basic prediction model;
judging whether the numerical value variation trends of the prediction result data and the real-time process data are consistent or not;
if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value;
and if the difference value between the prediction result data and the real-time process data is larger than the first alarm threshold value, alarming.
2. The method for warning of abnormal operating conditions of a petrochemical plant as defined in claim 1, wherein the determining whether the difference between the predicted result data and the real-time process data is greater than a first warning threshold specifically comprises:
and windowing the deviation value by adopting a data-based time sequence prediction method, taking the last data in each window as a target variable, then performing time sequence prediction by using a support vector machine, and judging whether the predicted deviation and the actual deviation obtained according to the basic prediction model are greater than the first alarm threshold value.
3. The method for warning the abnormal operating condition of the petrochemical device according to claim 1 or 2, further comprising:
when the process related parameters are changed, adjusting the first alarm threshold value according to the changed process related parameters to obtain a second alarm threshold value;
and judging whether to alarm or not according to the second alarm threshold value.
4. The method for warning of abnormal operating conditions of a petrochemical device according to claim 3, wherein obtaining a second warning threshold according to the first warning threshold specifically comprises:
and adjusting the first alarm threshold value according to the changed process related parameters and the Euclidean distance of the process related parameters before the change to obtain a second alarm threshold value.
5. The method for warning of abnormal operating conditions of a petrochemical device as defined in claim 1, wherein the alarming comprises:
providing a fault tree;
selectively alarming according to the prediction result data and the real-time process data based on the fault tree.
6. The utility model provides an early warning system of petrochemical device abnormal conditions which characterized in that includes:
the system comprises a providing module, a prediction module and a processing module, wherein the providing module is used for providing a basic prediction model of petrochemical process data, process related parameters and real-time process data;
the alarm module is used for giving an alarm;
the control processing module is used for obtaining prediction result data according to the process related parameters and the basic prediction model and judging whether the numerical variation trends of the prediction result data and the real-time process data are consistent; if the numerical value change trends of the prediction result data and the real-time process data are inconsistent, judging whether the difference value between the prediction result data and the real-time process data is larger than a preset first alarm threshold value; and if the difference value between the prediction result data and the real-time process data is larger than the first alarm threshold value, alarming through the alarm module.
7. The early warning system for the abnormal operating conditions of the petrochemical device, according to claim 6, wherein the control processing module is specifically configured to perform windowing on the deviation value by using a data-based time sequence prediction method, take the last data in each window as a target variable, perform time sequence prediction by using a support vector machine, and determine whether the predicted deviation and the actual deviation obtained according to the basic prediction model are greater than the first warning threshold.
8. The warning system for the abnormal operating condition of the petrochemical device according to claim 6 or 7, further comprising:
the monitoring module is used for monitoring whether the process related parameters are changed;
and the control processing module is further used for adjusting the first alarm threshold value according to the changed process related parameters to obtain a second alarm threshold value after the process related parameters are changed, and judging whether the alarm module needs to give an alarm according to the second alarm threshold value.
9. The warning system for the abnormal operating conditions of the petrochemical device as claimed in claim 8, wherein the control processing module is further configured to adjust the first warning threshold to obtain a second warning threshold according to the euclidean distance between the changed process-related parameters and the process-related parameters before the change.
10. The warning system for the abnormal operating conditions of the petrochemical device as claimed in claim 6, wherein the providing module is further configured to provide a fault tree, and the control processing module is further configured to selectively alarm based on the fault tree according to the predicted result data and the real-time process data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911302436.5A CN111090939B (en) | 2019-12-17 | 2019-12-17 | Early warning method and system for abnormal working condition of petrochemical device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911302436.5A CN111090939B (en) | 2019-12-17 | 2019-12-17 | Early warning method and system for abnormal working condition of petrochemical device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111090939A true CN111090939A (en) | 2020-05-01 |
| CN111090939B CN111090939B (en) | 2023-08-15 |
Family
ID=70395655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911302436.5A Active CN111090939B (en) | 2019-12-17 | 2019-12-17 | Early warning method and system for abnormal working condition of petrochemical device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111090939B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069744A (en) * | 2020-09-03 | 2020-12-11 | 中冶赛迪重庆信息技术有限公司 | Heating furnace operation parameter recommendation system and method based on data mining |
| CN112132316A (en) * | 2020-08-19 | 2020-12-25 | 张家口卷烟厂有限责任公司 | System and method for monitoring abnormality of on-line equipment in silk making link |
| CN113778776A (en) * | 2020-06-23 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Method and device for early warning task abnormity and storage medium |
| CN114841396A (en) * | 2022-03-16 | 2022-08-02 | 广东石油化工学院 | Method for predicting metamorphic trend and warning catastrophe risk in petrochemical production process |
| CN116823175A (en) * | 2023-07-10 | 2023-09-29 | 深圳市昭行云科技有限公司 | Intelligent operation and maintenance method and system for petrochemical production informatization system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070220368A1 (en) * | 2006-02-14 | 2007-09-20 | Jaw Link C | Data-centric monitoring method |
| CN103824130A (en) * | 2014-02-27 | 2014-05-28 | 武汉理工大学 | Grain condition forecasting and early warning method and system based on SVM |
| CN104196506A (en) * | 2014-08-01 | 2014-12-10 | 中国石油大学(北京) | Injection and production parameter joint debugging method, device and system for SAGD single well set |
| CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
| CN110298455A (en) * | 2019-06-28 | 2019-10-01 | 西安因联信息科技有限公司 | A kind of mechanical equipment fault intelligent early-warning method based on multivariable estimation prediction |
-
2019
- 2019-12-17 CN CN201911302436.5A patent/CN111090939B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070220368A1 (en) * | 2006-02-14 | 2007-09-20 | Jaw Link C | Data-centric monitoring method |
| CN103824130A (en) * | 2014-02-27 | 2014-05-28 | 武汉理工大学 | Grain condition forecasting and early warning method and system based on SVM |
| CN104196506A (en) * | 2014-08-01 | 2014-12-10 | 中国石油大学(北京) | Injection and production parameter joint debugging method, device and system for SAGD single well set |
| CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
| CN110298455A (en) * | 2019-06-28 | 2019-10-01 | 西安因联信息科技有限公司 | A kind of mechanical equipment fault intelligent early-warning method based on multivariable estimation prediction |
Non-Patent Citations (2)
| Title |
|---|
| 曹晓红;陆鹏飞;周业永;叶峰;: "混合建模技术在石化装置异常工况预警中的应用", 计算机与应用化学 * |
| 朱晓东;王杰;: "基于分层模糊***的石油钻井参数预测模型", 石油学报 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113778776A (en) * | 2020-06-23 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Method and device for early warning task abnormity and storage medium |
| CN112132316A (en) * | 2020-08-19 | 2020-12-25 | 张家口卷烟厂有限责任公司 | System and method for monitoring abnormality of on-line equipment in silk making link |
| CN112069744A (en) * | 2020-09-03 | 2020-12-11 | 中冶赛迪重庆信息技术有限公司 | Heating furnace operation parameter recommendation system and method based on data mining |
| CN112069744B (en) * | 2020-09-03 | 2023-03-31 | 中冶赛迪信息技术(重庆)有限公司 | Heating furnace operation parameter recommendation system and method based on data mining |
| CN114841396A (en) * | 2022-03-16 | 2022-08-02 | 广东石油化工学院 | Method for predicting metamorphic trend and warning catastrophe risk in petrochemical production process |
| CN114841396B (en) * | 2022-03-16 | 2023-02-17 | 广东石油化工学院 | Method for predicting metamorphic trend and warning catastrophe risk in petrochemical production process |
| CN116823175A (en) * | 2023-07-10 | 2023-09-29 | 深圳市昭行云科技有限公司 | Intelligent operation and maintenance method and system for petrochemical production informatization system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111090939B (en) | 2023-08-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111090939B (en) | Early warning method and system for abnormal working condition of petrochemical device | |
| US7213174B2 (en) | Provision of process related information | |
| US6925338B2 (en) | Fiducial technique for estimating and using degradation levels in a process plant | |
| JP2010506257A (en) | Multivariate monitoring and diagnosis of process variable data | |
| CN105264448A (en) | Real-time chemical process monitoring, assessment and decision-making assistance method | |
| KR20190062739A (en) | Method, algorithm and device for Data analytics for predictive maintenance using multiple sensors | |
| US20210377141A1 (en) | Identification of facility state and operating mode in a particular event context | |
| CN115039047A (en) | Industrial plant monitoring | |
| CN113326585A (en) | Energy efficiency abnormity early warning method and device for gas-fired boiler and computer equipment | |
| CN115407712A (en) | Intelligent maintenance system for hydraulic station of steel mill and working process | |
| CN112740133A (en) | System and method for monitoring the technical state of a technical installation | |
| CN115186754A (en) | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model | |
| US11644390B2 (en) | Contextual data modeling and dynamic process intervention for industrial plants | |
| US20230376024A1 (en) | Device and Method for Identifying Anomalies in an Industrial System for Implementing a Production Process | |
| CN118089287B (en) | Water chiller energy efficiency optimizing system based on intelligent algorithm | |
| Arroyo et al. | Supporting plant disturbance analysis by dynamic causal digraphs and propagation look-up tables | |
| Ming et al. | Intelligent Monitoring and Diagnosis of CCWS Heat Exchanger For Nuclear Power Plant | |
| CN116949452B (en) | Glacial acetic acid, nitric acid and phosphoric acid ratio control system of aluminum etching solution | |
| US20240119342A1 (en) | General reinforcement learning framework for process monitoring and anomaly/ fault detection | |
| CN116292520A (en) | Method for predicting hydraulic equipment faults based on multi-element parameter matrix | |
| CN118297224A (en) | Multi-equipment linkage fault prediction method, medium and system for refrigeration machine room | |
| CN116127392A (en) | Valve fault early warning method and system based on health state | |
| Basheer et al. | Maintenance planning optimization through equipment performance prediction using machine learning based on inline instrument datasets—a surface condenser case study | |
| CN117892869A (en) | Coal chemical industry intelligent carbon emission monitoring system and method of deep neural network model | |
| CN118092154A (en) | Substation water use control system and method based on neural network self-adaptive PID control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |