A machine learning method and its application for early warning of electrolyzer failure
A fault warning and machine learning technology, applied in the field of clustering and prediction of machine learning, can solve problems such as equipment loss and operator paralysis, and achieve the effect of avoiding production accidents, excellent model fitting effect, and abnormal discrimination effect.
Active Publication Date: 2021-07-30
上海新增鼎数据科技有限公司
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, if a single operation data occasionally exceeds the threshold, it does not necessarily cause damage to the equipment. Too many false alarms can easily paralyze the operator. Correlation, before the failure occurs, it is of great significance to predict the upcoming problems of the equipment based on the abnormality of the operating parameters of the equipment
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0058] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the specific implementation steps will be described in detail below.
[0059] Such as figure 1 As shown, the application of the GMM model in the early warning of electrolyzer failure is mainly realized through the following steps:
[0060] Step 1, data preparation, the data source used for modeling and analysis needs to be obtained through several steps:
[0061] Step 1.1, select the detection point, such as figure 2 As shown, according to experience, determine the relevant detection points that have an impact on the operation of the electrolytic cell. The selected detection points include the pressure difference between the positive and negative chambers of the electrolytic cell, the voltage difference between the front and rear ends of the electrolytic cell, the anode circulation flow, the cathode circulation flow, the supplementary brine flow, ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
This paper provides a machine learning method for early warning of electrolyzer failure, which is used to establish a predictive model for electrolyzer failure. The main process includes: extracting detection point sequence data, data preprocessing, training data sets into the GMM clustering model, Define anomaly discrimination rules, optimize discrimination parameters, improve the GMM clustering model, evaluate the fitting effect of the training model, and provide an application of machine learning methods for early warning of electrolyzer failures. The main process includes: extracting new sequence data of detection points , data preprocessing, time series forecasting, training model early warning and fault judgment. The invention can effectively reduce the paralysis of operators caused by traditional conditional value alarms, replace experienced operators in judging faults, and avoid judgment errors caused by human factors.
Description
technical field [0001] The invention relates to the technical field of clustering and prediction methods of machine learning, in particular to a machine learning method and application thereof for early warning of operating parameter failures of electrolyzer equipment, and is suitable for electrolyzer equipment whose operating parameters can be automatically collected and transmitted. Background technique [0002] At present, in the maintenance of production equipment, most enterprises still stay in the preventive maintenance of equipment, which consumes a lot of manpower and material resources, and once the problem is found, the problem has already occurred, causing chain parking and bringing great losses to production. However, in the production of modern chemical enterprises, the automation of production data measurement has been realized. The production and consumption data can be transmitted to the DCS system through sensing equipment, and the operation data (flow, press...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/12
CPCG06N3/126G06F18/23G06F18/214
Inventor 沈佳杰王彦婷邱振鲁陈宜川韩彩亮
Owner 上海新增鼎数据科技有限公司