CN-122022291-A - Intelligent management method and system for cement factory equipment management

CN122022291ACN 122022291 ACN122022291 ACN 122022291ACN-122022291-A

Abstract

The embodiment of the invention provides an intelligent management method and system for cement factory equipment management, wherein the method comprises the steps of collecting technological parameters and carrying out preprocessing steps to determine data labels of the technological parameters, obtaining equipment parameters of a timestamp corresponding to steady-state data, further determining equipment health state residual errors, establishing a regression model, outputting health deviation residual errors and outputting health conditions of real-time parameters, obtaining front-order data slices in a specific time window before an abnormal working condition data corresponding period, extracting data features, storing the data features into a mode feature library, carrying out matching to the obtained equipment parameters and the mode feature library to determine similarity scores, triggering corresponding hidden danger alarms, constructing a multi-objective optimization model of equipment corresponding to the hidden danger alarms, solving total expected cost of each maintenance route in the model, and further outputting maintenance decision reports.

Inventors

YUAN YIBIN
Mao Chenkai
WANG XU
XIE CONGCONG
LUO LIANG
LI BO
ZOU JIABIN
WEI CAN
LI XIANTAO
PAN WEIDONG
ZHENG GUANGDONG
ZHOU JIANJUN
ZHANG GUOMAO
WANG ZHECHENG

Assignees

邦业(杭州)智能技术有限公司
西南水泥有限公司

Dates

Publication Date: 20260512
Application Date: 20260119

Claims (10)

1. An intelligent management method for cement plant equipment management, the method comprising: acquiring process parameters and performing a preprocessing step, comparing rule labels of a preset rule base, and determining data labels of the process parameters, wherein the data labels comprise steady-state data and unsteady-state data, and the unsteady-state data further comprise abnormal working condition data; Acquiring equipment parameters of the timestamp corresponding to the steady-state data, further determining residual errors of the health state of the equipment, and establishing an equipment regression model based on the steady-state data and the residual errors, wherein the equipment regression model outputs health deviation residual errors corresponding to the real-time parameters of the equipment, and further outputs the health condition of the real-time parameters; acquiring a preamble data slice in a specific time window before a corresponding time period of the abnormal working condition data, extracting data features of the preamble data slice, storing the data features into a mode feature library, continuously matching acquired equipment parameters with the mode feature library, determining a similarity score based on a comparison result, and triggering a corresponding hidden danger alarm; And constructing a multi-target optimization model of the equipment corresponding to the hidden danger alarm, wherein the input of the multi-target optimization model comprises fault information, task data, resource data and economic cost, and the total expected cost of each maintenance route in the model is solved, so that a maintenance decision report is output.
2. The method according to claim 1, wherein the method further comprises: and establishing a regression model by taking the steady-state data as an independent variable and the equipment parameters as the dependent variable, wherein the calculation formula of the regression model comprises the following steps: Y=f(X)+ε, Wherein Y is a device parameter, f (X) is an argument X, an expected normal value of the device is determined, and epsilon represents a residual error.
3. The method according to claim 2, wherein the method further comprises: calculating a residual sequence of an actual value of the equipment parameter and a predicted value of the equipment parameter in the historical steady-state data, and calculating a standard deviation of the historical residual sequence; And comparing the standard deviation with a preset healthy central line, and determining a residual error control range of the corresponding parameter based on a difference value of the standard deviation and the preset healthy central line.
4. The method of claim 1, wherein the data features comprise: Time domain features, frequency domain features and time-frequency domain features; the storing the data features into the pattern feature library comprises the following steps: and determining feature extraction tendency based on the fault type corresponding to the abnormal working condition data, determining key features in the time domain features, the frequency domain features and the time-frequency domain features based on the feature extraction tendency, and storing the key features into a mode feature library.
5. The method according to claim 1, wherein the method further comprises: acquiring fault information of equipment corresponding to the hidden danger alarm, and further acquiring task data, resource data and economic cost of the current time; presetting selectable maintenance routes of each fault mode of the equipment, and calculating the total expected cost of each maintenance route under corresponding constraint, wherein the total expected cost comprises risk cost, maintenance direct cost and shutdown cost; A corresponding structured decision report is generated based on the overall expectations.
6. An intelligent management system for cement plant equipment management, which is characterized by comprising; The acquisition module is used for acquiring the process parameters and carrying out a preprocessing step, comparing rule labels of a preset rule base, and determining data labels of the process parameters, wherein the data labels comprise steady-state data and unsteady-state data, and the unsteady-state data further comprise abnormal working condition data; The regression module is used for acquiring the equipment parameters of the time stamp corresponding to the steady-state data, further determining the residual error of the health state of the equipment, and establishing an equipment regression model based on the steady-state data and the residual error, wherein the equipment regression model outputs the health deviation residual error corresponding to the real-time parameters of the equipment, and further outputs the health condition of the real-time parameters; The matching module is used for acquiring a preceding data slice in a specific time window before a period corresponding to the abnormal working condition data, extracting data characteristics of the preceding data slice, storing the data characteristics into a mode characteristic library, continuously matching the acquired equipment parameters with the mode characteristic library, determining a similarity score based on a comparison result, and triggering a corresponding hidden danger alarm; the decision module is used for constructing a multi-objective optimization model of the equipment corresponding to the hidden danger alarm, wherein the input of the multi-objective optimization model comprises fault information, task data, resource data and economic cost, the total expected cost of each maintenance route in the model is solved, and then a maintenance decision report is output.
7. The system of claim 6, wherein the system further comprises: And the trend module is used for determining feature extraction trend based on the fault type corresponding to the abnormal working condition data, determining key features in time domain features, frequency domain features and time-frequency domain features based on the feature extraction trend, and storing the key features into a mode feature library.
8. The system of claim 6, wherein the system further comprises: The maintenance module is used for acquiring fault information of equipment corresponding to the hidden danger alarm, and further acquiring task data, resource data and economic cost at the current time; the cost module is used for presetting selectable maintenance routes of each fault mode of the equipment, and calculating the total expected cost of each maintenance route under corresponding constraint, wherein the total expected cost comprises risk cost, maintenance direct cost and shutdown cost; And a reporting module for generating a corresponding structured decision report based on the total expectations.
9. An electronic device includes a processor and a memory; the processor is connected with the memory; The memory is used for storing executable program codes; The processor runs a program corresponding to executable program code stored in the memory by reading the executable program code for performing the method according to any one of claims 1-5.
10. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1-5.

Description

Intelligent management method and system for cement factory equipment management Technical Field The invention relates to the technical field of equipment management, in particular to an intelligent management method and system for equipment management of a cement factory. Background At present, the equipment management of a cement factory generally adopts a traditional manual leading mode, and core management links such as equipment file storage and update, point inspection task arrangement and process recording, operation data measurement statistics and analysis and the like all depend on manual operation, and the recording form is paper standing accounts or electronic forms. The deployment and collaboration of daily equipment management work are mainly carried out in an off-line mode such as pre-shift notification, telephone communication, weChat message transmission and the like, the discovery of equipment hidden danger is highly dependent on the field experience and responsibility consciousness of staff, and the whole management mode is mainly maintained afterwards. However, in the above prior art, there may be workflow too much off-line, resulting in a lack of standardized on-line architecture, and associated decision execution is difficult to manage. And during data processing, the quality is insufficient, and manual work is relied on, so that an 'information island' is formed, and decision execution cannot be supported. Disclosure of Invention Aiming at the problems existing in the prior art, the embodiment of the invention provides an intelligent management method and system for cement factory equipment management. The embodiment of the invention provides an intelligent management method for cement factory equipment management, which comprises the following steps: acquiring process parameters and performing a preprocessing step, comparing rule labels of a preset rule base, and determining data labels of the process parameters, wherein the data labels comprise steady-state data and unsteady-state data, and the unsteady-state data further comprise abnormal working condition data; Acquiring equipment parameters of the timestamp corresponding to the steady-state data, further determining residual errors of the health state of the equipment, and establishing an equipment regression model based on the steady-state data and the residual errors, wherein the equipment regression model outputs health deviation residual errors corresponding to the real-time parameters of the equipment, and further outputs the health condition of the real-time parameters; Acquiring a preamble data slice in a specific time window before a corresponding time period of the abnormal working condition data, extracting data features of the preamble data slice, storing the data features into a mode feature library, continuously matching the acquired equipment parameters with the mode feature library, determining a similarity score based on a comparison result, and triggering a corresponding hidden danger alarm; And constructing a multi-target optimization model of the equipment corresponding to the hidden danger alarm, wherein the input of the multi-target optimization model comprises fault information, task data, resource data and economic cost, and the total expected cost of each maintenance route in the model is solved, so that a maintenance decision report is output. In one embodiment, the method further comprises: and establishing a regression model by taking the steady-state data as an independent variable and the equipment parameters as the dependent variable, wherein the calculation formula of the regression model comprises the following steps: Y=f(X)+ε, Wherein Y is a device parameter, f (X) is an argument X, an expected normal value of the device is determined, and epsilon represents a residual error. In one embodiment, the method further comprises: calculating a residual sequence of an actual value of the equipment parameter and a predicted value of the equipment parameter in the historical steady-state data, and calculating a standard deviation of the historical residual sequence; And comparing the standard deviation with a preset healthy central line, and determining a residual error control range of the corresponding parameter based on a difference value of the standard deviation and the preset healthy central line. In one embodiment, the data features include: Time domain features, frequency domain features and time-frequency domain features; the storing the data features into the pattern feature library comprises the following steps: and determining feature extraction tendency based on the fault type corresponding to the abnormal working condition data, determining key features in the time domain features, the frequency domain features and the time-frequency domain features based on the feature extraction tendency, and storing the key features into a mode feature library. In one embodiment, the method further comprise