CN112036727A - Method for positioning risk degree of gas pipeline - Google Patents
Method for positioning risk degree of gas pipeline Download PDFInfo
- Publication number
- CN112036727A CN112036727A CN202010861904.9A CN202010861904A CN112036727A CN 112036727 A CN112036727 A CN 112036727A CN 202010861904 A CN202010861904 A CN 202010861904A CN 112036727 A CN112036727 A CN 112036727A
- Authority
- CN
- China
- Prior art keywords
- risk
- data
- degree
- positioning model
- degree positioning
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000003066 decision tree Methods 0.000 claims description 6
- 230000001364 causal effect Effects 0.000 claims description 4
- 238000004451 qualitative analysis Methods 0.000 claims description 4
- 238000012098 association analyses Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 230000004807 localization Effects 0.000 claims 2
- 238000004364 calculation method Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012047 cause and effect analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention relates to a method for positioning the risk degree of a gas pipeline, which comprises the following steps: qualitatively identifying and collecting a risk index set, and preprocessing to obtain a risk data set; constructing a risk identification degree positioning model according to the risk data set; and carrying out risk degree positioning on the gas pipeline data by utilizing the risk identification degree positioning model. By adopting the scheme, the qualitative and quantitative modes are simultaneously applied, and the data-driven concept is applied to the risk degree positioning and the model establishing process, so that the scientificity and the automation degree of the risk degree positioning process are improved, and the labor cost of risk identification is reduced.
Description
Technical Field
The invention relates to the technical field of town gas Internet of things application, in particular to a method for positioning risk degree of a gas pipeline.
Background
The existing urban gas pipe network risk degree positioning is mainly based on an index weight superposition method, the construction and selection of a risk index set are usually formed in a form discussed by experts, and the risk degree positioning is determined by utilizing long-term experience of the experts. In the risk evaluation process, index weight distribution as the hyper-parameters of the risk model is also determined manually through the experience of experts, and the accuracy of weight distribution is also controversial, so most of the risk models lack a feedback mechanism of practical application effect, the models are difficult to continuously optimize in practical use, and finally obtained risk degree positioning is not accurate.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a method for positioning the risk degree of a gas pipeline.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
a method for positioning risk degree of a gas pipeline comprises the following steps:
step S1: qualitatively identifying and collecting a risk index set, and preprocessing to obtain a risk data set;
step S2: constructing a risk identification degree positioning model according to the risk data set;
step S3: and carrying out risk degree positioning on the gas pipeline data by utilizing the risk identification degree positioning model.
Further, in order to better implement the present invention, the step S1 specifically includes the following steps:
step S1-1: preliminarily identifying a risk index set by a qualitative analysis method, carrying out causal association analysis and importance analysis on each event in the preliminarily identified risk index set, and outputting the preliminary risk index set;
step S1-2: calculating the correlation coefficient of the events on the preliminarily determined risk index set by using the data expression of each event, filtering two events with strong correlation according to the analyzed importance, reserving one event with larger importance, and outputting the refined risk index set;
step S1-3: and collecting risk degree data in the risk case library according to the refined risk index set, and preprocessing the collected risk degree data to obtain a risk data set.
Further, in order to better implement the present invention, the step of preprocessing the collected risk degree data includes:
and converting the events in the risk degree data into a numerical form, wherein the classification attribute is converted into a numerical attribute, and the continuous numerical value is converted into a discrete numerical value.
Further, in order to better implement the present invention, the step S2 specifically includes the following steps:
and modeling by a data mining method according to the characteristics of the data attributes in the risk data set, and learning a risk identification degree positioning model by using a classification regression tree or an artificial neural network learning method to obtain the risk identification degree positioning model.
Furthermore, in order to better implement the present invention, the step of learning the risk identification degree location model by using the learning method of the classification regression tree to obtain the risk identification degree location model includes:
and determining the splitting index of the risk identification degree positioning model by using the information entropy or the Gini non-existence degree, constructing a decision tree by using recursion, and obtaining the risk identification degree positioning model expressed by the decision tree when all the splitting indexes are split or the information gain ratio of the splitting indexes is smaller than a preset threshold value.
As another possible implementation manner, the step of learning the risk identification degree location model by using the learning method of the artificial neural network to obtain the risk identification degree location model includes:
and updating the weight parameters of the neurons in the risk identification degree positioning model by using a gradient descent method until the error of the weight change meets the requirement of an error limit, and obtaining the weight parameters of each layer of the risk identification degree positioning model so as to obtain the risk identification degree positioning model expressed by the neural network.
Compared with the prior art, the invention has the beneficial effects that:
by adopting the scheme, the qualitative and quantitative modes are simultaneously applied, and the data-driven concept is applied to the risk degree positioning and the model establishing process, so that the scientificity and the automation degree of the risk degree positioning process are improved, and the labor cost of risk identification is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a flow chart of a risk level location method of the present invention;
FIG. 2 is a schematic diagram of a fault tree according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a minimal cut set fault tree of a fault tree according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
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. Also, in the description of the present invention, the terms "first", "second", and the like are used for distinguishing between descriptions and not necessarily for describing a relative importance or implying any actual relationship or order between such entities or operations.
Example 1:
the invention is realized by the following technical scheme, as shown in figure 1, a method for positioning the risk degree of a gas pipeline comprises the following steps:
step S100: and qualitatively identifying and collecting a risk index set, and preprocessing to obtain a risk data set.
The set of risk indicators is initially identified by a qualitative analysis method, which may be summarized empirically or analyzed by a logistic analysis method. The scheme uses a logic analysis method, namely a fault tree analysis method, which is a deductive failure analysis method from top to bottom, and then utilizes a Boolean algebra algorithm to analyze an unexpected event.
In detail, a pipeline failure is taken as a top event, the reason causing the pipeline failure is analyzed, and the reason is taken as a lower-level event (bottom event); and performing the same cause analysis on all lower-level events, then taking the cause as the lower-level events, and so on until all events cannot continue to perform the splitting of the cause analysis, and then completing a fault tree in the whole splitting process. According to the formed fault tree, causal association analysis can be performed on each event in the fault tree, namely causal association between event nodes in the fault tree is analyzed, for example, a lower-level event is a reason of an upper-level event, and the upper-level event is an effect of the lower-level event. Importance analysis can also be performed, for example, according to the fault tree, the influence weight of all events in the risk indicator set on the top event can be obtained, and the influence weight is the importance and is obtained in a numerical manner, and a larger numerical value indicates more importance.
And selecting a minimum cut set of the fault tree by using a Boolean algebra algorithm, wherein the minimum cut set forms a risk index set.
In detail, the main purpose of qualitative analysis of fault treeTo figure out how likely a device will fail (top event), i.e. to analyze which factors will cause the device to fail. A cut set is a collection of several bottom events that cause a device failure, i.e., a cut set represents a possibility of a device failure. If the bottom event set of a fault tree is { x }1,x2,...xnWhen there is a subset of { x }i1,xi2,...,xilK, and have
When x is satisfiedi1＝xi2＝...＝xilWhen equal to 1, let Φ (x) equal to 1, that is, when all bottom events contained in the subset occur, the top event must occur, and the subset is a cut set, and the number of cut sets is K.
The minimal cut set is the set containing the least number of necessary bottom events, and the complete set of all minimal cut sets represents all faults given, so the significance of the minimal cut set is that it gives the factors that must be repaired in the faulty equipment. In finding the minimal cut set, the intermediate events at the lowest level can be started for a given fault tree, and the AND gate structure function can be used if the intermediate events are logically linked to the gate bottom event
If the intermediate event is a logical OR gate to bottom event correlation, then an OR gate function may be used
And (5) sequentially moving upwards until a top event, and finishing the operation. In the obtained calculation results, if the same bottom event occurs, the Boolean algebra algorithm is used for simplification.
Such as the fault tree shown in fig. 2, can result in:
T＝(x1+x2+x3)(x3+x4)(x1+x4)
＝(x1x3+x1x4+x2x3+x2x4+x3+x3x4)(x1+x4)
＝x1x3+x1x4+x1x2x3+x1x2x4+x1x3x4+x2x3x4+x2x4+x3x4
thus, 8 cut sets are obtained, and the above formula is simplified by using a Boolean algebra algorithm to obtain a minimum cut set:
T＝x1x3+x1x4+x2x4+x3x4
that is, the fault tree has 4 minimal cut sets, x respectively1x3、x1x4、x2x4、x3x4At the same time, an equivalent fault tree with the minimal cut set can be obtained, as shown in fig. 3. The event contained in the set obtained by performing union on each minimal cut set is the set of the preliminary risk indicators.
The event node generally includes five major categories of three-party destruction, natural disaster, corrosion protection, essential defect and operation and maintenance, and event subdivision is performed on each major category. The next level of the top event is one or more of these five categories, and the next level is a subdivision if the pipeline fails.
And after the event cause and effect analysis and the importance analysis are carried out on the risk index set formed by the fault tree, a preliminary risk index set can be output. And calculating a correlation coefficient between the events on the preliminarily determined risk index set by using the data representation of the events. The data representation of each event is generally observed through data visualization to form a preliminary judgment of data relevance, then the data is converted into a numerical value form, and then the numerical value is utilized to calculate the correlation coefficient among the events.
Correlation coefficient calculation description:
assuming that the set of preliminary risk indicators outputted this time is (a, b, c, d, e, f), where each event represents a factor affecting the risk of the pipeline, during the pipeline production operation, it is necessary to pay attention to collecting the data in 6 aspects, and assuming that the following data are obtained after a period of time (t 1-tn):
correlation coefficient calculation formula:
where X, Y represents any event, Cov (X, Y) represents the covariance of event X, Y, d (X) represents the variance of event X, and d (Y) represents the variance of event Y. Combining the events a, b, c, d, e and f in pairs into X, Y, the correlation coefficient between the events can be calculated.
It should be noted that, because the calculation of the correlation coefficient is to be performed, the collected data must be in a numerical form to be brought into the formula, and when the events existing in the set of the risk indicator set can be directly expressed in a numerical form, the directly collected numerical values are used as the values of the events. For example, if event a represents the wall thickness of the pipe, a1 is obtained as a wall thickness value; if the event b represents the weather, the numerical value cannot be directly obtained, and therefore the acquired data needs to be assigned in advance, for example, the weather is assigned to 1 for sunny days, 2 for cloudy days, 3 for gust rain, and the like, and the data can be brought into a formula to calculate the correlation coefficient after being converted into a numerical value form.
And each collected data corresponds to a risk degree result, and the risk degree can include high risk, medium risk, low risk and the like.
And then, combining the analyzed importance, filtering the two events with strong relevance, reserving one event with higher importance, and outputting a refined risk index set. For example, the closer the calculated correlation coefficient is to 1, the stronger the correlation between two events, and only one of the events needs to be left to filter the other events with strong correlation. When the threshold is set to 0.7, the correlation coefficient is greater than 0.7, which indicates that the correlation is strong, one of the events needs to be filtered, and the remaining one of the two events with strong correlation is the one with relatively high importance.
And according to the refined risk index set, collecting risk degree data in a risk case base, wherein the risk case base is equivalent to a historical database, and mapping the collected risk degree data on the risk index set according to an event mode existing in the risk index set to form a risk data set.
When risk degree data are collected, the collected risk degree data need to be preprocessed, and events in the collected risk degree data are digitally converted to form a risk data set. Like the foregoing manner of processing the preliminary risk index set into the refined risk index set, for example, when an event in the risk degree data is a pipe wall thickness, a numerical value may be directly obtained, and when an event is weather, the weather needs to be converted into a numerical value in the form of an assignment, for example, the weather is a clear assignment of 1, a cloudy assignment of 2, a cloudy-sunny assignment of 3, and the like. This results in a risk data set that is then used to subsequently build a risk identification degree model.
Step S200: and constructing a risk identification degree positioning model according to the risk data set.
Modeling by a data mining method according to the characteristics of the data attributes in the risk data set, such as the data of each event and the corresponding result, and learning the risk identification degree positioning model by using a classification regression tree or an artificial neural network learning method to obtain the risk identification degree positioning model embedded with the risk degree positioning rule.
And when a learning method of the classification regression tree is adopted, determining the risk identification degree positioning model split index by using the information entropy or the degree of uncertainty of the kini. The process of determining the split indicator, such as if entropy is used, is:
firstly, calculating the information entropy H before the splitting of the current sample, namely the information entropy H of a risk data set, calculating the information entropy weighted sum H1 of the total split sample when each index event is split according to the index event, wherein H1-H is the information gain of the splitting scheme, traversing and comparing the splitting schemes with the maximum information gain in all index splitting schemes, the corresponding event is the event selected by the current splitting, and H1 is the H selected by the next splitting. And then, a decision tree is constructed by utilizing recursion, and when all events are split or the information gain ratio of the split events is smaller than a threshold value, a risk identification degree positioning model expressed by the decision tree is obtained.
When the learning method of the artificial neural network is adopted, the weight parameters of the neurons in the risk identification degree positioning model are updated by using a gradient descent method until the error of the weight change meets the requirement of an error limit, and the weight parameters of each layer of the risk identification degree positioning model are obtained. And if the change value or the change percentage of the weight is smaller than the threshold value, stopping iteration, and taking the weight parameter obtained at the moment as a final weight parameter so as to obtain a risk identification degree positioning model expressed by the neural network.
Step S300: and carrying out risk degree positioning on the input gas pipeline data by utilizing the risk identification degree positioning model.
After the risk identification degree positioning model is built, the newly acquired data can be directly input into the risk identification degree positioning model, and the risk degree of the gas pipeline can be positioned according to the input data.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A method for positioning risk degree of a gas pipeline is characterized by comprising the following steps: the method comprises the following steps:
step S1: qualitatively identifying and collecting a risk index set, and preprocessing to obtain a risk data set;
step S2: constructing a risk identification degree positioning model according to the risk data set;
step S3: and carrying out risk degree positioning on the gas pipeline data by utilizing the risk identification degree positioning model.
2. The method for locating the risk level of a gas pipeline according to claim 1, wherein: the step S1 specifically includes the following steps:
step S1-1: preliminarily identifying a risk index set by a qualitative analysis method, carrying out causal association analysis and importance analysis on each event in the preliminarily identified risk index set, and outputting the preliminary risk index set;
step S1-2: calculating the correlation coefficient of the events on the preliminarily determined risk index set by using the data expression of each event, filtering two events with strong correlation according to the analyzed importance, reserving one event with larger importance, and outputting the refined risk index set;
step S1-3: and collecting risk degree data in the risk case library according to the refined risk index set, and preprocessing the collected risk degree data to obtain a risk data set.
3. A gas pipeline risk degree localization method as claimed in claim 2, characterized in that: the step of preprocessing the collected risk level data comprises:
events in the risk degree data are converted into numerical form.
4. A gas pipeline risk degree localization method as claimed in claim 2, characterized in that: the step S2 specifically includes the following steps:
and modeling by a data mining method according to the characteristics of the data attributes in the risk data set, and learning a risk identification degree positioning model by using a classification regression tree or an artificial neural network learning method to obtain the risk identification degree positioning model.
5. The method for locating the risk level of a gas pipeline according to claim 4, wherein: the step of learning the risk identification degree positioning model by using the learning method of the classification regression tree to obtain the risk identification degree positioning model comprises the following steps:
and determining the splitting index of the risk identification degree positioning model by using the information entropy or the Gini non-existence degree, constructing a decision tree by using recursion, and obtaining the risk identification degree positioning model expressed by the decision tree when all the splitting indexes are split or the information gain ratio of the splitting indexes is smaller than a preset threshold value.
6. The method for locating the risk level of a gas pipeline according to claim 4, wherein: the step of learning the risk identification degree positioning model by using the learning method of the artificial neural network to obtain the risk identification degree positioning model comprises the following steps:
and updating the weight parameters of the neurons in the risk identification degree positioning model by using a gradient descent method until the error of the weight change meets the requirement of an error limit, and obtaining the weight parameters of each layer of the risk identification degree positioning model so as to obtain the risk identification degree positioning model expressed by the neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010861904.9A CN112036727A (en) | 2020-08-25 | 2020-08-25 | Method for positioning risk degree of gas pipeline |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010861904.9A CN112036727A (en) | 2020-08-25 | 2020-08-25 | Method for positioning risk degree of gas pipeline |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112036727A true CN112036727A (en) | 2020-12-04 |
Family
ID=73581243
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010861904.9A Pending CN112036727A (en) | 2020-08-25 | 2020-08-25 | Method for positioning risk degree of gas pipeline |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112036727A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113516850A (en) * | 2021-09-14 | 2021-10-19 | 成都千嘉科技有限公司 | Pipeline traffic flow data acquisition method based on space syntactic analysis |
-
2020
- 2020-08-25 CN CN202010861904.9A patent/CN112036727A/en active Pending
Non-Patent Citations (4)
| Title |
|---|
| 周大伟;: "基于BP神经网络的油气管道风险分析评估方法", 管道技术与设备, no. 04, 15 July 2020 (2020-07-15) * |
| 朱继伟;曲娜;王嘉豪;: "基于神经网络的火灾报警系统失效风险分析", 科技展望, no. 15, 10 August 2015 (2015-08-10), pages 123 * |
| 王晓睿;李玉霞;薛亚东;: "基于粗糙集和神经网络的深基坑工程风险评价", 人民黄河, no. 08, 20 August 2011 (2011-08-20) * |
| 赵振宇, 刘伊生, 杨华春: "故障树法引入工程项目风险管理研究", 现代电力, no. 02, 30 April 2002 (2002-04-30) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113516850A (en) * | 2021-09-14 | 2021-10-19 | 成都千嘉科技有限公司 | Pipeline traffic flow data acquisition method based on space syntactic analysis |
| CN113516850B (en) * | 2021-09-14 | 2021-11-12 | 成都千嘉科技有限公司 | Pipeline traffic flow data acquisition method based on space syntactic analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109766583B (en) | Aircraft engine life prediction method based on unlabeled, unbalanced and initial value uncertain data | |
| CN110046743B (en) | Public building energy consumption prediction method and system based on GA-ANN | |
| CN101093559B (en) | Method for constructing expert system based on knowledge discovery | |
| CN110555230B (en) | Rotary machine residual life prediction method based on integrated GMDH framework | |
| CN112131212A (en) | Hybrid cloud scene-oriented time sequence data anomaly prediction method based on ensemble learning technology | |
| CN104820716A (en) | Equipment reliability evaluation method based on data mining | |
| CN111401599B (en) | Water level prediction method based on similarity search and LSTM neural network | |
| CN111598352A (en) | Concrete beam type bridge comprehensive evaluation method based on Bayesian network | |
| CN112101431A (en) | Electronic equipment fault diagnosis system | |
| CN112508053A (en) | Intelligent diagnosis method, device, equipment and medium based on integrated learning framework | |
| CN112487058A (en) | Numerical control machine tool fault monitoring and diagnosing system based on data mining | |
| CN116150897A (en) | Machine tool spindle performance evaluation method and system based on digital twin | |
| CN112756759B (en) | Spot welding robot workstation fault judgment method | |
| Zhu et al. | A novel decision-making model for selecting a construction project delivery system | |
| CN111178585A (en) | Fault reporting amount prediction method based on multi-algorithm model fusion | |
| CN109522962B (en) | Chemical plant safety quantitative evaluation method | |
| CN114548494A (en) | Visual cost data prediction intelligent analysis system | |
| CN114862032B (en) | XGBoost-LSTM-based power grid load prediction method and device | |
| CN115718472A (en) | Fault scanning and diagnosing method for hydroelectric generating set | |
| CN116739376A (en) | Highway pavement preventive maintenance decision method based on data mining | |
| CN111898673A (en) | Dissolved oxygen content prediction method based on EMD and LSTM | |
| CN112036727A (en) | Method for positioning risk degree of gas pipeline | |
| CN113408659A (en) | Building energy consumption integrated analysis method based on data mining | |
| CN114779739A (en) | Fault monitoring method for industrial process under cloud edge end cooperation based on probability map model | |
| CN115481841A (en) | Material demand prediction method based on feature extraction and improved random forest |
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 |