CN114662977A - Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment - Google Patents
Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment Download PDFInfo
- Publication number
- CN114662977A CN114662977A CN202210376318.4A CN202210376318A CN114662977A CN 114662977 A CN114662977 A CN 114662977A CN 202210376318 A CN202210376318 A CN 202210376318A CN 114662977 A CN114662977 A CN 114662977A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- service data
- anomaly
- abnormal
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides an anomaly detection method, an anomaly detection system and electronic equipment for the motion state of an offshore drilling platform, wherein the anomaly detection method is used for carrying out anomaly detection on the motion state of the offshore drilling platform according to multi-dimensional service data, an anomaly detection model can be used for carrying out anomaly detection processing on a time sequence characteristic and second target service data, the obtained anomaly service data can accurately reflect the abnormal condition of the motion state of the offshore drilling platform, the detection process is convenient, accurate and effective, and the technical problem that the prior art cannot carry out convenient, accurate and effective detection on the anomaly of the motion state of the offshore drilling platform is solved.
Description
Technical Field
The invention relates to the technical field of oil and gas exploration, in particular to an abnormal detection method, system and electronic equipment for the motion state of an offshore drilling platform.
Background
The drilling platform is a sharp tool for offshore operation in the oil industry, and related exploration and exploitation are developed based on the offshore drilling platform along with the discovery of offshore oil and gas resources in China. Offshore oil and gas resource development operation has the characteristics of high investment, high technology, high risk and the like, the operation space is limited, personnel and equipment are intensively operated at high strength, and due to the fact that the size of the drilling platform is large, the inertia is large, the operation is limited, the offshore operation environment is complex, emergency working conditions frequently occur in the operation process of the offshore drilling platform, and once safety accidents are easily caused by improper operation, the safety of the offshore drilling platform and the safety of the periphery can be endangered. Therefore, it is very urgent to analyze, identify, prevent and control the abnormality and risk in the offshore platform drilling operation to ensure the operation safety.
At present, most of abnormal detection aiming at the offshore drilling platform is regularly developed structure detection, or after abnormal working conditions occur, manual overhaul is carried out, and the work is complicated and the timeliness is low. The offshore drilling platform is an engineering machine which operates for a long time, judges the operation state of the offshore drilling platform in real time, is related to the structural state of the offshore drilling platform, and is influenced by external environment and operation conditions, so that the operation process of the offshore drilling platform needs to be monitored by more strict online and long-term data. At present, most of digital monitoring on the operation process of the offshore drilling platform still stays in the collection and storage of signals or parameters, collected data are rarely analyzed and mined, the inherent value of the data is not fully utilized, and less research is performed on a complete system in the aspects of abnormity monitoring, evaluation and early warning in the operation process of the platform. At present, anomaly detection or risk assessment for offshore drilling platforms at home and abroad is mostly carried out based on business knowledge and simulation technology, a set of complete risk analysis and assessment system is already provided, but the methods are mostly developed based on specific business problems and small data sets, and the law of how mass data collected and accumulated by the platforms reflect the operating state of the platforms for a long time is not systematically and deeply excavated at present. Along with the rapid development of global economy, modern industrial processes are gradually developed towards large-scale and integrated directions, the acquisition of platform state data is completed by arranging sensors at key positions of a drilling platform, and abnormal operation states in the operation process of the platform can be found in time by performing training modeling and online early warning based on an abnormal detection technology of big data of the drilling platform, so that the platform can be found, intervened and controlled in time before a fault occurs, and the effects that business knowledge and a simulation technology cannot achieve are achieved.
The offshore platform motion state data have the characteristics of multiple sources, high dimensionality, high sampling frequency and the like, for an operation engineer of a platform, compared with the abnormality detection of each variable, the operation engineer is more concerned about the state of the whole equipment, in the face of the high-dimensionality data, the cost of training and maintaining an independent abnormality detection model for each variable is very high, usually, an event on the platform can cause the abnormality of multiple dimensionality variables, the abnormality detected at a single-variable level needs to be defined based on business knowledge, the abnormality of all the variables is processed, and therefore whether the platform is abnormal or not is determined, namely whether the platform is abnormal or not can not be accurately reflected by the single-variable abnormality.
Comprehensively, how to conveniently, accurately and effectively detect the abnormity of the motion state of the offshore drilling platform becomes a problem which needs to be solved urgently at present.
Disclosure of Invention
In view of this, the present invention provides a method, a system and an electronic device for detecting an anomaly of a motion state of an offshore drilling platform, so as to solve the technical problem that the anomaly of the motion state of the offshore drilling platform cannot be detected conveniently, accurately and effectively in the prior art.
In a first aspect, an embodiment of the present invention provides an anomaly detection method for a motion state of an offshore drilling platform, including:
obtaining multi-dimensional service data of the offshore drilling platform;
preprocessing the multidimensional service data to obtain processed service data;
extracting time sequence characteristics of first target service data in the processed service data, wherein the first target service data is service data selected by a user;
and performing anomaly detection processing on the time sequence characteristics and second target business data in the processed business data by adopting an anomaly detection model to obtain abnormal business data in the business data and further obtain an anomaly detection result of the motion state of the offshore drilling platform, wherein the second target business data is the business data except the first target business data in the processed business data.
Further, preprocessing the multidimensional service data to obtain processed service data, including:
unifying the sampling frequency of the multidimensional service data to obtain the multidimensional service data with unified sampling frequency;
and performing correlation analysis on the multidimensional service data with uniform sampling frequency, and performing dimensionality filtration according to a correlation analysis result to obtain the multidimensional service data with filtered dimensionality, thereby obtaining the processed service data.
Further, extracting a time sequence feature of the first target service data in the processed service data includes:
dividing the first target service data of each dimension into a plurality of subsequences by adopting a sliding window with a fixed length;
calculating the median and standard deviation of each subsequence in the plurality of subsequences;
and calculating the time sequence characteristics according to the median and the standard deviation of the adjacent subsequences.
Further, the anomaly detection model is a depth automatic coding Gaussian mixture model adopting unsupervised anomaly detection.
In a second aspect, an embodiment of the present invention further provides an anomaly detection system for a motion state of an offshore drilling platform, where the system includes: the system comprises a data management module, an abnormality analysis module and a model management module;
the data management module is used for managing data information of each project, parameter information of each anomaly detection model and anomaly data labeling information;
the anomaly analysis module is integrated with various anomaly detection models, and is used for performing anomaly detection processing on the service data under each project according to the anomaly detection method for the motion state of the offshore drilling platform in any one of the first aspect, and giving an alarm according to an anomaly detection result;
and the model management module is used for updating and training each abnormal detection model and managing the version of each abnormal detection model.
Further, the data management module includes: a project information database, a business data import state database, a data field information database, a model parameter information database and an abnormal data information database;
the project information database is used for managing project information of each project;
the business data import state database is used for recording import state information of the business data corresponding to each project;
the data field information database is used for maintaining data structure information of the service data corresponding to each project;
the model information database is used for recording the evaluation index results of the abnormal detection models of all versions and the version information of the abnormal detection models of all versions;
the model parameter information database is used for recording model parameters of the abnormal detection models of all versions;
and the abnormal data information database is used for maintaining abnormal detection result data and artificially marked abnormal data so as to store the abnormal detection result data and use the artificially marked abnormal data for updating training of an abnormal detection model.
Further, the anomaly analysis module comprises: an abnormality analysis unit of the existing project and an abnormality analysis unit of the newly-built project;
the abnormity analysis unit of the existing project is used for acquiring the project selected by the user, adopting an abnormity detection model selected by the user to carry out abnormity detection processing on the business data triggered by the user according to the business data import instruction and the model selection instruction triggered by the user, and alarming according to an abnormity detection result;
the anomaly analysis unit of the newly-built project is used for acquiring the project newly-built by the user, acquiring a model field table created by the user for the newly-built project, further adopting an anomaly detection model selected by the user to perform anomaly detection processing on the service data triggered by the user according to a service data import instruction and a model selection instruction triggered by the user, and giving an alarm according to an anomaly detection result.
Further, the model management module includes: the system comprises a model training unit, a model evaluation unit and a model updating unit;
the model training unit is used for acquiring model parameters input by a user, and training an anomaly detection model to be updated based on the model parameters, historical business data and artificial labeling anomaly data to obtain the trained anomaly detection model to be updated;
the model evaluation unit is used for evaluating the trained abnormal detection model to be updated and the abnormal detection model to be updated to obtain an evaluation result;
and the model updating unit is used for determining whether the trained abnormal detection model to be updated is better than the abnormal detection model to be updated according to the evaluation result, and if the trained abnormal detection model to be updated is better than the abnormal detection model to be updated, updating the version of the abnormal detection model to be updated.
Further, the model parameters include at least: the number of iterations and the learning rate.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method according to any one of the above first aspects when executing the computer program.
In an embodiment of the present invention, a method for detecting an anomaly of a motion state of an offshore drilling platform is provided, including: obtaining multi-dimensional service data of an offshore drilling platform; preprocessing the multidimensional service data to obtain processed service data; extracting time sequence characteristics of first target service data in the processed service data, wherein the first target service data is service data selected by a user; and performing anomaly detection processing on the second target service data in the sequence characteristics and the processed service data by adopting an anomaly detection model to obtain the abnormal service data in the service data and further obtain an anomaly detection result of the motion state of the offshore drilling platform, wherein the second target service data is the service data except the first target service data in the processed service data. According to the anomaly detection method, the anomaly detection is carried out on the motion state of the offshore drilling platform according to the multidimensional service data, the anomaly detection model can carry out anomaly detection processing on the time sequence characteristic and the second target service data, the obtained anomaly service data can accurately reflect the anomaly condition of the motion state of the offshore drilling platform, the detection process is convenient, accurate and effective, and the technical problem that the anomaly of the motion state of the offshore drilling platform cannot be detected conveniently, accurately and effectively in the prior art is solved.
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, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for detecting an anomaly in a motion state of an offshore drilling platform according to an embodiment of the present invention;
FIG. 2 is a characteristic variable curve diagram after the time series characteristic analysis provided by the embodiment of the present invention;
FIG. 3 is a diagram illustrating an anomaly detection result of an anomaly detection model according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an anomaly detection system for a motion state of an offshore drilling platform according to an embodiment of the present invention;
FIG. 5 is a flow chart of the operation of the anomaly detection system for the motion state of the offshore drilling platform according to the embodiment of the invention;
fig. 6 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood that the described embodiments are some, but not all, embodiments of the present invention. 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 detected abnormality at the univariate level, the abnormality of all variables needs to be defined and processed based on business knowledge, so as to determine whether the offshore drilling platform is abnormal, that is, the univariate abnormality cannot accurately reflect whether the offshore drilling platform is abnormal.
Based on the above, the anomaly detection method provided by the invention is used for carrying out anomaly detection on the motion state of the offshore drilling platform according to the multi-dimensional service data, the anomaly detection model can carry out anomaly detection processing on the time sequence characteristic and the second target service data, the obtained anomaly service data can accurately reflect the anomaly condition of the motion state of the offshore drilling platform, and the detection process is convenient, accurate and effective.
For the convenience of understanding the embodiment, a detailed description will be given to an anomaly detection method for a motion state of an offshore drilling platform disclosed in the embodiment of the invention.
The first embodiment is as follows:
according to an embodiment of the invention, an embodiment of a method for detecting an anomaly in a motion state of an offshore drilling platform is provided. It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.
Fig. 1 is a flowchart of an anomaly detection method for a motion state of an offshore drilling platform according to an embodiment of the present invention, as shown in fig. 1, the method comprising the steps of:
step S102, obtaining multi-dimensional service data of the offshore drilling platform;
specifically, the multidimensional service data may be data in a preset time range (historical time range), and the dimensionality of the multidimensional service data may be determined according to different items, for example, if one item needs to analyze stress data and motion data, the multidimensional service data includes: stress data and motion data, wherein the motion data may include: the latitude, longitude, speed, angular velocity, yaw angle, altitude and other parameters, and the stress data may be stress data detected by sensors arranged at different positions of the hull.
Step S104, preprocessing the multidimensional service data to obtain processed service data;
step S106, extracting the time sequence characteristics of first target business data in the processed business data, wherein the first target business data is the business data selected by the user;
the purpose of extracting the time sequence characteristics is to distinguish abnormal service data from normal service data more obviously, so that the abnormal service data in the service data can be detected more accurately by the abnormal detection model subsequently.
And S108, performing anomaly detection processing on the second target service data in the sequence characteristics and the processed service data by using an anomaly detection model to obtain abnormal service data in the service data, and further obtain an anomaly detection result of the motion state of the offshore drilling platform, wherein the second target service data is the service data except the first target service data in the processed service data.
In an embodiment of the present invention, a method for detecting an anomaly of a motion state of an offshore drilling platform is provided, including: obtaining multi-dimensional service data of an offshore drilling platform; preprocessing the multidimensional service data to obtain processed service data; extracting time sequence characteristics of first target service data in the processed service data, wherein the first target service data is service data selected by a user; and performing anomaly detection processing on the second target service data in the sequence characteristics and the processed service data by adopting an anomaly detection model to obtain the abnormal service data in the service data and further obtain an anomaly detection result of the motion state of the offshore drilling platform, wherein the second target service data is the service data except the first target service data in the processed service data. According to the anomaly detection method, the anomaly detection is carried out on the motion state of the offshore drilling platform according to the multidimensional service data, the anomaly detection model can carry out anomaly detection processing on the time sequence characteristic and the second target service data, the obtained anomaly service data can accurately reflect the anomaly condition of the motion state of the offshore drilling platform, the detection process is convenient, accurate and effective, and the technical problem that the anomaly of the motion state of the offshore drilling platform cannot be detected conveniently, accurately and effectively in the prior art is solved.
The foregoing description briefly introduces the method for detecting an abnormal motion state of an offshore drilling platform according to the present invention, and the detailed description thereof will be described in detail.
In an optional embodiment of the present invention, the preprocessing is performed on the multidimensional service data to obtain processed service data, and the method specifically includes the following steps:
(1) unifying the sampling frequency of the multidimensional service data to obtain the multidimensional service data with unified sampling frequency;
the following description takes multidimensional service data as stress data and motion data as an example:
because the data sampling frequencies of the motion data and the stress data are different, the sampling frequencies of the motion data and the stress data need to be unified so as to perform data characteristic analysis on the motion data and the stress data in the following.
(2) And performing correlation analysis on the multidimensional service data with uniform sampling frequency, and performing dimensionality filtration according to a correlation analysis result to obtain the multidimensional service data with filtered dimensionality, so as to obtain the processed service data.
Specifically, because the dimensionality of the data is high, through correlation analysis, it is found that the correlation among some dimensionalities is high, and when the high-dimensionality data is subjected to subsequent processing, consumed resources are large, and the processing of the data of the anomaly detection model is not facilitated.
The correlation analysis may be to calculate the correlation between the dimensions by some distance calculation method, for example, euclidean distance, cosine distance, etc.
In an optional embodiment of the present invention, the extracting a time sequence feature of the first target service data in the processed service data specifically includes the following steps:
1) dividing first target service data of each dimension into a plurality of subsequences by adopting a sliding window with a fixed length;
under the condition that the multidimensional service data are stress data and motion data, the first target service data can be the stress data, and the extracted time sequence features are features extracted from the distribution and stability of the stress data in the time dimension.
For example, the stress data detected by each stress sensor is a data dimension, and then the stress data detected by each stress sensor is divided into a plurality of subsequences by adopting a sliding window with a fixed length.
For example, the data amount of the stress data detected by one stress sensor is 1000, and if the sliding window with the fixed length is 100, the stress data can be divided into 10 subsequences.
2) Calculating the median and standard deviation of each subsequence in the plurality of subsequences;
3) and calculating the time sequence characteristics according to the median and the standard deviation of the adjacent subsequences.
Specifically, the greater the difference, the greater the probability that two adjacent subsequences are abnormal, as shown in fig. 2 (characteristic variable graph after time series characteristic analysis), so the result of comparing the median and standard deviation of the adjacent subsequences can be used as the time series characteristic of the dimension data, that is, the difference between the median and standard deviation of the adjacent subsequences is used as the time series characteristicAnd (5) carrying out characterization. The set of timing characteristics may be expressed as Q ═ xmd，xsd]Wherein x ismd＝medianright-medianleft；xsd＝stdright-stdleftThe time sequence characteristic can obviously distinguish abnormal service data in different forms from normal service data, so that the abnormality in the data can be more accurately detected in the detection process of the subsequent model.
In an alternative embodiment of the present invention, the anomaly detection model is a depth automatic coding gaussian mixture model using unsupervised anomaly detection.
Specifically, the training process of the anomaly detection model is as follows:
acquiring historical multi-dimensional normal service data as a training data set, for example, acquiring historical normal stress data and historical normal motion data as the training data set, preprocessing the normal stress data and the normal motion data according to the processes (1) and (2), extracting time sequence characteristics of the processed normal stress data after preprocessing to obtain time sequence characteristics, further taking the time sequence characteristics and the processed normal motion data as input data of an original anomaly detection model (comprising a compression network and an estimation network), generating low-dimensional characteristic information and reconstruction errors after the input data passes through the compression network, splicing the low-dimensional characteristic information and the reconstruction errors to be used as input of the estimation network to obtain weight probabilities of each Gaussian distribution, and bringing the obtained weight probabilities into a preset maximum likelihood function, and obtaining a maximum likelihood function value, and after multiple iterations, enabling the maximum likelihood function value to be larger and larger (the larger the maximum likelihood function value is, the larger the possibility that the data is abnormal is), and after the iterations are repeated, determining that the training of the abnormal detection model is finished when the maximum likelihood function values obtained by the two iterations are basically unchanged, and obtaining a likelihood function threshold value in a training data set (normal data set). In the subsequent application process, the anomaly detection model carries out anomaly detection processing on the data to obtain a maximum likelihood function value, the obtained maximum likelihood function value is compared with a likelihood function threshold determined in the training process, if the maximum likelihood function value is larger than the likelihood function threshold, the service data detected at the moment is determined to be the abnormal service data, and then an anomaly detection result is obtained.
It should be noted that: although a deep automatic coding gaussian mixture model (DAGMM) focuses on multi-dimensional anomaly detection, time dependency among samples cannot be well focused, but for a multi-element time sequence, historical data is helpful for reconstructing a current state, so before constructing a DAGMM network, multi-dimensional time sequence feature analysis is firstly carried out, time sequence features under different anomaly forms are extracted, so that the model can acquire the time dependency among the samples, and therefore the time sequence features of processed normal stress data need to be extracted, and then the time sequence features are input into the network.
The anomaly detection model takes a maximum likelihood function consisting of a minimized reconstruction error and a likelihood function of a negative estimation network as an optimization direction. Fig. 3 is a schematic diagram showing an abnormality detection result of the abnormality detection model.
Example two:
the embodiment of the invention also provides an anomaly detection system for the motion state of the offshore drilling platform, which is specifically introduced below.
Fig. 4 is a schematic diagram of an anomaly detection system for a motion state of an offshore drilling platform according to an embodiment of the present invention, as shown in fig. 4, the system mainly comprises: data management module 10, anomaly analysis module 20, and model management module 30, wherein:
the data management module is used for managing data information of each project, parameter information of each anomaly detection model and anomaly data labeling information;
the anomaly analysis module is integrated with various anomaly detection models and is used for performing anomaly detection processing on the service data under each project according to the anomaly detection method for the motion state of the offshore drilling platform in the first embodiment and giving an alarm according to an anomaly detection result;
and the model management module is used for carrying out updating training on each abnormal detection model and managing the version of each abnormal detection model.
Specifically, the user can manage, model and analyze the data of various structures in each project. The anomaly detection system for the motion state of the offshore drilling platform adopts a project management mode, so that anomaly modeling and analysis of different services and different table field structure service data are supported.
The anomaly detection system for the motion state of the offshore drilling platform disclosed by the invention combines the actual service data of the offshore drilling platform with an artificial intelligence anomaly detection algorithm, and realizes multiple functions of data management of different service scenes, automatic training of an anomaly detection model, model management, detection and early warning of the abnormal operation state of the offshore drilling platform and the like, thereby providing a reliable scientific basis for the operation and maintenance of the offshore drilling platform.
In an optional embodiment of the invention, the data management module comprises: a project information database, a business data import state database, a data field information database, a model parameter information database and an abnormal data information database;
a project information database for managing project information of each project; wherein the project information at least includes: basic information of the project and data dimension information of the project; for example, basic information of an item includes: the method comprises the following steps of creating time information of a project, name information of the project, creator information of the project and the like, wherein the data dimension information of the project comprises the following steps: and only the stress data dimension comprises the stress data dimension, the motion data dimension and the like.
The business data import state database is used for recording the import state information of the business data corresponding to each project; wherein the import status information may include one or more of the following: status information of successful import, status information of failed import, reason information of failure, etc.
The data field information database is used for maintaining data structure information of the service data corresponding to each project; for example, the data structure information may include: data table structures such as attribute fields of data and data types of the attribute fields.
The model information database is used for recording the evaluation index results of the abnormal detection models of all versions and the version information of the abnormal detection models of all versions;
the model parameter information database is used for recording model parameters of the abnormal detection models of all versions; wherein the model parameters include at least: the number of iterations and the learning rate.
And the abnormal data information database is used for maintaining the abnormal detection result data and the artificially marked abnormal data so as to store the abnormal detection result data and use the artificially marked abnormal data for updating and training the abnormal detection model.
In an optional embodiment of the invention, the anomaly analysis module comprises: an abnormality analysis unit of the existing project and an abnormality analysis unit of the newly-built project;
the system comprises an existing project abnormity analysis unit, a model selection unit and a fault detection unit, wherein the existing project abnormity analysis unit is used for acquiring a project selected by a user, adopting an abnormity detection model selected by the user to perform abnormity detection processing on service data triggered by the user according to a service data import instruction and a model selection instruction triggered by the user, and giving an alarm according to an abnormity detection result;
and the anomaly analysis unit of the newly-built project is used for acquiring the newly-built project of the user, acquiring a model field table created by the user for the newly-built project, further adopting an anomaly detection model selected by the user to perform anomaly detection processing on the service data triggered by the user according to a service data import instruction and a model selection instruction triggered by the user, and giving an alarm according to an anomaly detection result.
In addition, still include: and the manual labeling unit is used for continuously labeling and correcting abnormal data on the basis of the abnormal detection result according to the cognition of the relevant personnel after the abnormal detection result is obtained and the relevant personnel check the abnormal detection result.
The specific process is shown in fig. 5.
In an alternative embodiment of the invention, the model management module comprises: the system comprises a model training unit, a model evaluation unit and a model updating unit;
the model training unit is used for acquiring model parameters input by a user, training the anomaly detection model to be updated based on the model parameters, historical business data and artificial labeling anomaly data, and obtaining the trained anomaly detection model to be updated;
the model evaluation unit is used for evaluating the trained abnormal detection model to be updated and the trained abnormal detection model to be updated to obtain an evaluation result;
and the model updating unit is used for determining whether the trained abnormal detection model to be updated is superior to the abnormal detection model to be updated or not according to the evaluation result, and if the trained abnormal detection model to be updated is superior to the abnormal detection model to be updated, performing version updating on the abnormal detection model to be updated.
Specifically, model updates and different version management can be performed in each project of the system. When the anomaly detection precision of the newly-added data cannot be met with the lapse of time, the model needs to be retrained and updated based on the new historical data; in addition, key parameters (e.g., iteration number, learning rate, etc.) of the model can be adjusted at a model training interface in the project. If the re-trained model evaluation index is superior to the model of the previous version, the model is updated to the latest version, and the model is stored, otherwise, the model is rolled back to the previous version without updating. The relevant process is illustrated with reference to fig. 5.
Aiming at the problem of abnormal detection of unmarked high-dimensional time sequence motion data of the offshore drilling platform, the invention combines high-dimensional time sequence characteristic analysis with a friendly high-dimensional data-oriented depth automatic coding Gaussian mixture network, captures high-dimensional time sequence data characteristics with time dependence and reconstruction information, and performs abnormal identification and alarm of the overall motion state of the platform.
The modeling idea of the invention is suitable for the anomaly detection problem of high-dimensional time sequence data. On one hand, high-dimensional time sequence characteristic analysis is carried out, high-dimensional time sequence data characteristics with time dependency are obtained, and abnormal data and normal data in various different forms are more obviously distinguished; on the other hand, the DAGMM is used for anomaly detection modeling, the processes of capturing low-dimensional space mapping and density estimation are combined together, and compared with an anomaly detection method in which dimension reduction and density estimation are respectively carried out in machine learning, the anomaly key information can be captured more accurately; different from a method for carrying out abnormity detection on some key univariate indexes commonly used in the industry, the model evaluates the integral motion state of the platform, and provides good reference for abnormity identification of ocean engineering equipment operation and formulation of operation and maintenance strategies; the system for detecting the abnormal motion state of the offshore drilling platform realizes the full flow of data storage, model analysis and alarm monitoring of the offshore drilling platform, is not only limited to the collection and storage of signals or parameters, and provides a support reference for the risk abnormal prevention and control and operation and maintenance of the offshore drilling platform.
As shown in fig. 6, an electronic device 600 provided in an embodiment of the present application includes: a processor 601, a memory 602 and a bus, wherein the memory 602 stores machine readable instructions executable by the processor 601, when the electronic device is operated, the processor 601 and the memory 602 communicate with each other through the bus, and the processor 601 executes the machine readable instructions to execute the steps of the method for detecting the abnormality of the offshore drilling platform motion state.
Specifically, the memory 602 and the processor 601 can be general-purpose memory and processor, which are not limited in particular, and the processor 601 can execute the above-mentioned method for detecting the abnormality of the offshore drilling platform motion state when executing the computer program stored in the memory 602.
The processor 601 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 601. The Processor 601 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, etc. as is well known in the art. The storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602 and completes the steps of the method in combination with the hardware thereof.
In response to the method for detecting an anomaly in a motion state of an offshore drilling platform, embodiments of the present application further provide a computer-readable storage medium storing machine executable instructions, which when invoked and executed by a processor, cause the processor to perform the steps of the method for detecting an anomaly in a motion state of an offshore drilling platform.
The device for detecting the abnormal motion state of the offshore drilling platform provided by the embodiment of the application can be specific hardware on equipment or software or firmware installed on the equipment. The device provided by the embodiment of the present application has the same implementation principle and technical effect as the foregoing method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the foregoing method embodiments where no part of the device embodiments is mentioned. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the foregoing systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
For another example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solutions of the present application, or portions of the technical solutions that substantially contribute to the prior art, may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling an electronic device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the vehicle marking method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus once an item is defined in one figure, it need not be further defined and explained in subsequent figures, and moreover, the terms "first", "second", "third", etc. are used merely to distinguish one description from another and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the scope of the embodiments of the present application. Are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for detecting an abnormality of a motion state of an offshore drilling platform, comprising:
obtaining multi-dimensional service data of the offshore drilling platform;
preprocessing the multidimensional service data to obtain processed service data;
extracting time sequence characteristics of first target service data in the processed service data, wherein the first target service data is service data selected by a user;
and performing anomaly detection processing on the time sequence characteristics and second target business data in the processed business data by adopting an anomaly detection model to obtain abnormal business data in the business data and further obtain an anomaly detection result of the motion state of the offshore drilling platform, wherein the second target business data is the business data except the first target business data in the processed business data.
2. The method of claim 1, wherein preprocessing the multidimensional service data to obtain processed service data comprises:
unifying the sampling frequency of the multidimensional service data to obtain the multidimensional service data with unified sampling frequency;
and performing correlation analysis on the multidimensional service data with uniform sampling frequency, and performing dimensionality filtration according to a correlation analysis result to obtain the multidimensional service data with filtered dimensionality, so as to obtain the processed service data.
3. The method of claim 1, wherein extracting the timing characteristics of the first target service data in the processed service data comprises:
dividing the first target service data of each dimension into a plurality of subsequences by adopting a sliding window with a fixed length;
calculating the median and standard deviation of each subsequence in the plurality of subsequences;
and calculating the time sequence characteristics according to the median and the standard deviation of the adjacent subsequences.
4. The method of claim 1, wherein the anomaly detection model is a depth-automatically coded Gaussian mixture model with unsupervised anomaly detection.
5. An anomaly detection system for a state of motion of an offshore drilling rig, the system comprising: the system comprises a data management module, an abnormality analysis module and a model management module;
the data management module is used for managing data information of each project, parameter information of each anomaly detection model and anomaly data labeling information;
the anomaly analysis module is integrated with various anomaly detection models and is used for carrying out anomaly detection processing on the service data under each project according to the anomaly detection method of the offshore drilling platform motion state in any one of claims 1 to 4 and giving an alarm according to an anomaly detection result;
and the model management module is used for updating and training each abnormal detection model and managing the version of each abnormal detection model.
6. The system of claim 5, wherein the data management module comprises: a project information database, a business data import state database, a data field information database, a model parameter information database and an abnormal data information database;
the project information database is used for managing project information of each project;
the business data import state database is used for recording import state information of the business data corresponding to each project;
the data field information database is used for maintaining data structure information of the service data corresponding to each project;
the model information database is used for recording the evaluation index results of the abnormal detection models of all versions and the version information of the abnormal detection models of all versions;
the model parameter information database is used for recording model parameters of the abnormal detection models of all versions;
and the abnormal data information database is used for maintaining abnormal detection result data and artificially marked abnormal data so as to store the abnormal detection result data and use the artificially marked abnormal data for updating training of an abnormal detection model.
7. The system of claim 5, wherein the anomaly analysis module comprises: an abnormality analysis unit of the existing project and an abnormality analysis unit of the newly-built project;
the abnormity analysis unit of the existing project is used for acquiring the project selected by the user, adopting an abnormity detection model selected by the user to carry out abnormity detection processing on the business data triggered by the user according to the business data import instruction and the model selection instruction triggered by the user, and alarming according to an abnormity detection result;
the anomaly analysis unit of the newly-built project is used for acquiring the project newly-built by the user, acquiring a model field table created by the user for the newly-built project, further adopting an anomaly detection model selected by the user to perform anomaly detection processing on the service data triggered by the user according to a service data import instruction and a model selection instruction triggered by the user, and giving an alarm according to an anomaly detection result.
8. The system of claim 5, wherein the model management module comprises: the system comprises a model training unit, a model evaluation unit and a model updating unit;
the model training unit is used for acquiring model parameters input by a user, and training an anomaly detection model to be updated based on the model parameters, historical business data and artificial labeling anomaly data to obtain the trained anomaly detection model to be updated;
the model evaluation unit is used for evaluating the trained abnormal detection model to be updated and the abnormal detection model to be updated to obtain an evaluation result;
and the model updating unit is used for determining whether the trained abnormal detection model to be updated is better than the abnormal detection model to be updated according to the evaluation result, and if the trained abnormal detection model to be updated is better than the abnormal detection model to be updated, updating the version of the abnormal detection model to be updated.
9. The system of claim 8, wherein the model parameters include at least: the number of iterations and the learning rate.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any of the preceding claims 1 to 4 are implemented when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210376318.4A CN114662977A (en) | 2022-04-11 | 2022-04-11 | Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210376318.4A CN114662977A (en) | 2022-04-11 | 2022-04-11 | Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114662977A true CN114662977A (en) | 2022-06-24 |
Family
ID=82035760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210376318.4A Pending CN114662977A (en) | 2022-04-11 | 2022-04-11 | Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114662977A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115406488A (en) * | 2022-09-13 | 2022-11-29 | 北京千尧新能源科技开发有限公司 | Offshore operation platform, boarding corridor bridge safety early warning method and related equipment |
-
2022
- 2022-04-11 CN CN202210376318.4A patent/CN114662977A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115406488A (en) * | 2022-09-13 | 2022-11-29 | 北京千尧新能源科技开发有限公司 | Offshore operation platform, boarding corridor bridge safety early warning method and related equipment |
| CN115406488B (en) * | 2022-09-13 | 2023-10-10 | 北京千尧新能源科技开发有限公司 | Offshore operation platform, boarding corridor safety pre-warning method and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111506478A (en) | Method for realizing alarm management control based on artificial intelligence | |
| US20120025997A1 (en) | System and method for failure prediction for rod pump artificial lift systems | |
| CN112414694B (en) | Equipment multistage abnormal state identification method and device based on multivariate state estimation technology | |
| CN112101554B (en) | Abnormality detection method and apparatus, device, and computer-readable storage medium | |
| CN112036042B (en) | Power equipment abnormality detection method and system based on variational modal decomposition | |
| EP1279104A1 (en) | Generalized lensing angular similarity operator | |
| CN112766429B (en) | Method, device, computer equipment and medium for anomaly detection | |
| CN111027615A (en) | Middleware fault early warning method and system based on machine learning | |
| CN111881594B (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN114662977A (en) | Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment | |
| CN112882898A (en) | Anomaly detection method, system, device and medium based on big data log analysis | |
| CN116665421A (en) | Early warning processing method and device for mechanical equipment and computer readable storage medium | |
| CN117240522A (en) | Vulnerability intelligent mining method based on attack event model | |
| CN116862109A (en) | Regional carbon emission situation awareness early warning method | |
| CN116910526A (en) | Model training method, device, communication equipment and readable storage medium | |
| JP7095443B2 (en) | Anomaly detection program, anomaly detection method and anomaly detection device | |
| CN113218683A (en) | Petroleum underground electric casing cutter fault identification method based on vibration signals | |
| CN114548259B (en) | PISA fault identification method based on Semi-supervised Semi-KNN model | |
| CN115293379B (en) | Knowledge graph-based on-orbit spacecraft equipment anomaly detection method | |
| CN116861204B (en) | Intelligent manufacturing equipment data management system based on digital twinning | |
| CN117433591B (en) | Automatic equipment monitoring system based on multi-sensor data | |
| CN117520040B (en) | Micro-service fault root cause determining method, electronic equipment and storage medium | |
| CN117520964B (en) | Motor pre-diagnosis method based on artificial intelligence | |
| CN116010383B (en) | Deep sea submersible result information collection and management method | |
| Zhao et al. | Burst detection in district metering areas using flow subsequences clustering–reconstruction analysis |
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 |