CN111158338A - Chemical risk monitoring method based on principal component analysis - Google Patents
Chemical risk monitoring method based on principal component analysis Download PDFInfo
- Publication number
- CN111158338A CN111158338A CN201911318171.8A CN201911318171A CN111158338A CN 111158338 A CN111158338 A CN 111158338A CN 201911318171 A CN201911318171 A CN 201911318171A CN 111158338 A CN111158338 A CN 111158338A
- Authority
- CN
- China
- Prior art keywords
- principal component
- chemical
- component analysis
- data
- risk 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 27
- 238000000513 principal component analysis Methods 0.000 title claims abstract description 24
- 239000000126 substance Substances 0.000 title claims abstract description 23
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000005070 sampling Methods 0.000 claims abstract description 9
- 238000004519 manufacturing process Methods 0.000 claims abstract description 8
- 238000012824 chemical production Methods 0.000 claims abstract description 7
- 230000005856 abnormality Effects 0.000 claims abstract description 3
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 239000011159 matrix material Substances 0.000 claims description 16
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 4
- 239000013598 vector Substances 0.000 description 3
- 238000001311 chemical methods and process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- General Factory Administration (AREA)
Abstract
The invention provides a chemical risk monitoring method based on principal component analysis, which comprises the steps of establishing an analysis algorithm model in an off-line mode and monitoring risks in an on-line mode; the off-line establishment Analysis algorithm model is used for establishing a Principal Component Analysis (PCA) model after data preprocessing is carried out by sampling a plurality of index data which normally run in the chemical production process, analyzing the online real-time data of the chemical production by utilizing the established PCA model in the online risk monitoring, judging whether abnormality exists in the production process and giving an early warning in time. The invention solves the chemical risk early warning problem under mass data, carries out comprehensive diagnosis and analysis on high-dimensional data and effectively realizes the early warning of chemical safety production.
Description
Technical Field
The invention belongs to the technical field of chemical risk monitoring, and particularly relates to a chemical risk monitoring method based on principal component analysis.
Background
Along with the enlargement of the scale of the chemical process and the complication of the flow, the probability of various chemical accidents is increased day by day. At present, a DCS is generally adopted to monitor each single index in the chemical production process in real time, and the chemical risk is judged and alarmed and early warned by judging whether the index exceeds a threshold value. However, accurate and effective diagnosis of chemical risks and faults under massive data still has great difficulty. The DCS monitors single indexes respectively, and the DCS is various in quantity and has no relation with each other; in the face of a large-scale chemical process, accurate and effective diagnosis and early warning of chemical risks and faults under mass data cannot be realized by single threshold alarm of the DCS.
Disclosure of Invention
The invention aims to provide a chemical risk monitoring method based on principal component analysis, which solves the chemical risk early warning problem under mass data, carries out comprehensive diagnosis and analysis on high-dimensional data and effectively realizes chemical safety production early warning.
The invention provides the following technical scheme:
a chemical risk monitoring method based on principal component analysis comprises the steps of establishing an analysis algorithm model in an off-line mode and monitoring risks in an on-line mode; the off-line establishment Analysis algorithm model is used for establishing a plurality of running index models (PCA) after data preprocessing by sampling a plurality of index data which normally run in the chemical production process, analyzing the on-line real-time data of the chemical production by utilizing the established principal Component Analysis model in the on-line risk monitoring, judging whether abnormality exists in the production process and giving early warning in time.
Preferably, the off-line establishing of the analysis algorithm model includes the following steps:
s11: obtaining sample data of a plurality of sensors through multiple sampling to form a matrix;
s12: carrying out standardization processing on the data;
s13: performing eigenvalue decomposition on the matrix to obtain eigenvalues of different sizes and corresponding eigenvectors;
s14: and arranging according to the magnitude of the characteristic values to obtain a load matrix and a score matrix, and finally obtaining a principal component model constructed by the matrix.
Preferably, the online risk monitoring comprises the steps of:
s21: calculating SPE and T2 statistics and control limits of real-time data;
s22: and comparing the statistic and the control limit result, judging whether the production process is abnormal or not according to the control limit, and if so, early warning.
Preferably, the sampling is performed by selecting a plurality of sensors which are mutually associated to construct multi-dimensional data, and establishing a principal component analysis model to extract comprehensive information in the multi-dimensional data for analysis.
Preferably, the online risk monitoring part selects a plurality of indexes which are the same as those in offline modeling to calculate SPE and T2 statistics and control limits.
The invention has the beneficial effects that: in the invention, a plurality of sensors which are mutually associated are selected to construct multi-dimensional data, a principal component analysis model is established to extract comprehensive information in the multi-dimensional data for analysis, and compared with a DCS system for monitoring single data, noise interference is eliminated, and the result is more accurate and reliable; main information is extracted from the mass data, abnormal data in the mass data are detected through online monitoring, and equipment faults are more effectively positioned, so that time for enterprise personnel to troubleshoot risk faults can be greatly shortened, and possible accident potential hazards can be eliminated in time; on the other hand, the method is also helpful for providing real-time and relatively accurate risk data and improving the auxiliary decision-making capability of fire fighters.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic diagram of an off-line partial setup process of the present invention;
FIG. 2 is a schematic of the on-line partial analysis process of the present invention.
Detailed Description
As shown in fig. 1 and 2, a chemical risk monitoring method based on principal component analysis includes two steps, an offline part and an online part, and analyzes sample data and real-time data respectively. And the off-line part establishes a principal component analysis algorithm model based on the normal data sample, and the on-line part performs on-line chemical risk analysis on the real-time data by using the established model.
Step 1: establishing a principal component analysis algorithm model (off-line part):
1-1, obtaining m times of samples of n sensors through sampling to form an X e to Rm×nA matrix of (a);
1-2, carrying out standardization processing on data;
1-3, carrying out eigenvalue decomposition on the covariance matrix of X to obtain eigenvalues with different sizes and corresponding eigenvectors;
1-4, arranging according to the magnitude of the characteristic values to obtain a load matrix P (formed by characteristic vectors) and a score matrix T (principal component variables);
and finally obtaining a principal component model of the X structure: t ═ XP.
The first A pivot variables are selected, and the model of X can be expressed as
Step 2: on-line chemical risk monitoring (on-line part)
2-1. calculating SPE and T of real-time data2Statistics and control limits:
(1)T2the statistic is defined as the standard square sum of all score vectors, representing the magnitude of the decentration in the principal component analysis model at each sampling time, and for the statistic at the ith time, the statistic is defined as:
wherein Λ is a characteristic value diagonal matrix, A is the number of selected principal elements, x is a row vector of each variable at a certain time,
t representing a confidence of a2Controlling limit:
(2) the SPE statistic is used to observe changes in the part not interpreted by the pivot, defined as:
wherein delta2Control limit representing confidence level a:
wherein c isaIs a standard normal distribution with a confidence limit of a,
λjis the jth eigenvalue of the covariance matrix.
2-2. comparison of results, when SPE > δ2And is
And (4) indicating that the production process is abnormal, and immediately warning.
According to the method, main information is extracted from mass data, abnormal data in the mass data are detected through online monitoring, and equipment faults are more effectively positioned, so that time for enterprise personnel to troubleshoot risk faults can be greatly shortened, and possible accident potential hazards can be eliminated in time; on the other hand, the method is also helpful for providing real-time and relatively accurate risk data and improving the auxiliary decision-making capability of fire fighters.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. A chemical risk monitoring method based on principal component analysis is characterized by comprising the steps of establishing an analysis algorithm model in an off-line mode and monitoring risks in an on-line mode; the method comprises the steps of establishing an analysis algorithm model in an off-line mode, preprocessing data by sampling a plurality of index data which normally run in the chemical production process, establishing a principal component analysis model, analyzing online real-time data of the chemical production by using the established principal component analysis model in the online risk monitoring process, judging whether abnormality exists in the production process, and giving early warning in time.
2. The chemical risk monitoring method based on principal component analysis according to claim 1, wherein the off-line establishment of the analysis algorithm model comprises the following steps:
s11: obtaining sample data of a plurality of sensors through multiple sampling to form a matrix X;
s12: carrying out standardization processing on the data;
s13: performing eigenvalue decomposition on the matrix to obtain eigenvalues of different sizes and corresponding eigenvectors;
s14: and arranging according to the magnitude of the characteristic values to obtain a load matrix and a score matrix, and finally obtaining a principal component model constructed for the matrix X.
3. The chemical risk monitoring method based on principal component analysis according to claim 1, wherein the online risk monitoring comprises the following steps:
s21: calculating SPE and T2 statistics and control limits of real-time data;
s22: and comparing the statistic and the control limit result, judging whether the production process is abnormal or not according to the control limit, and if so, early warning.
4. The chemical risk monitoring method based on principal component analysis as claimed in claim 1, wherein a plurality of interrelated sensors are selected through multiple sampling to construct multi-dimensional data, and a principal component analysis model is established to extract comprehensive information therein for analysis.
5. The chemical risk monitoring method based on principal component analysis of claim 3, wherein the online risk monitoring comprises selecting a plurality of indexes same as those in offline modeling to calculate SPE and T2 statistics and control limits.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911318171.8A CN111158338A (en) | 2019-12-19 | 2019-12-19 | Chemical risk monitoring method based on principal component analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911318171.8A CN111158338A (en) | 2019-12-19 | 2019-12-19 | Chemical risk monitoring method based on principal component analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111158338A true CN111158338A (en) | 2020-05-15 |
Family
ID=70557456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911318171.8A Pending CN111158338A (en) | 2019-12-19 | 2019-12-19 | Chemical risk monitoring method based on principal component analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111158338A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723725A (en) * | 2020-05-25 | 2021-11-30 | 中国石油化工股份有限公司 | Risk early warning method and device for operation process of chemical device and terminal equipment |
| CN114167826A (en) * | 2021-11-26 | 2022-03-11 | 华中科技大学 | Mixed multivariable monitoring method for casting production process |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105700518A (en) * | 2016-03-10 | 2016-06-22 | 华中科技大学 | Fault diagnosis method during industrial process |
-
2019
- 2019-12-19 CN CN201911318171.8A patent/CN111158338A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105700518A (en) * | 2016-03-10 | 2016-06-22 | 华中科技大学 | Fault diagnosis method during industrial process |
Non-Patent Citations (1)
| Title |
|---|
| 王中伟: "基于对数变换和最大信息系数PCA的过程监测", 《科学技术与工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723725A (en) * | 2020-05-25 | 2021-11-30 | 中国石油化工股份有限公司 | Risk early warning method and device for operation process of chemical device and terminal equipment |
| CN114167826A (en) * | 2021-11-26 | 2022-03-11 | 华中科技大学 | Mixed multivariable monitoring method for casting production process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109459993B (en) | Online adaptive fault monitoring and diagnosing method for process industrial process | |
| CN100489870C (en) | Method and multidimensional system for statistical control of a process | |
| CN110687895B (en) | Chemical process fault detection method based on self-adaptive kernel principal component analysis | |
| CN106845826B (en) | PCA-Cpk-based cold continuous rolling production line service quality state evaluation method | |
| CN115951606A (en) | Intelligent factory production environment early warning processing method | |
| CN116665401A (en) | Accident prevention alarm system for chemical production | |
| CN110175682A (en) | A kind of optimization core pivot element analysis fault monitoring method based on Chaos-Particle Swarm Optimization | |
| CN104616209B (en) | A kind of power cable connector information fusion evaluation method based on on-line monitoring | |
| CN116112283A (en) | CNN-LSTM-based power system network security situation prediction method and system | |
| CN111158338A (en) | Chemical risk monitoring method based on principal component analysis | |
| CN117850375B (en) | Multi-dimensional monitoring system of production line | |
| CN108171271B (en) | Early warning method and system for equipment degradation | |
| CN117975372B (en) | Construction site safety detection system and method based on YOLOv and transducer encoder | |
| CN105809257B (en) | Trigger type state maintenance method based on power communication network | |
| CN117391458B (en) | Safety production risk detection and early warning method and system based on data analysis | |
| CN112131516A (en) | Anomaly detection method based on feature weight mixed naive Bayes model | |
| CN117768235A (en) | Real-time flow monitoring alarm system based on Internet of things | |
| CN116542956A (en) | Automatic detection method and system for fabric components and readable storage medium | |
| CN115187026A (en) | Industrial risk monitoring method and system and readable storage medium | |
| CN117521084B (en) | Active safety early warning method for complex system | |
| CN118713991B (en) | Fault prediction method and system based on observable data | |
| CN104808652A (en) | Chemical engineering device operating state scoring method | |
| CN114154673A (en) | Motor fault diagnosis and prediction method based on big data | |
| CN118972235A (en) | Power communication network fault diagnosis method, system and device based on deep learning | |
| CN112578740A (en) | Fault diagnosis and processing method and system in industrial production process |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200515 |