CN-122020055-A - Intelligent management system for energy efficiency of air separation equipment based on data analysis

CN122020055ACN 122020055 ACN122020055 ACN 122020055ACN-122020055-A

Abstract

The invention relates to the technical field of energy efficiency management of air separation equipment, in particular to an energy efficiency intelligent management system of the air separation equipment based on data analysis, which comprises a data acquisition and quality calibration module, a working condition normalization energy efficiency baseline module and a mechanism-data fusion digital twin module, wherein the data acquisition and quality calibration module is used for carrying out time alignment, deletion and exception handling on instrument data, correcting key measurement point drift based on quantile alignment of gold window data, and the working condition normalization energy efficiency baseline module is used for converting the output of multiple products into equivalent output and correcting the inlet air temperature and the ambient pressure to obtain a normalization energy efficiency index.

Inventors

LI ZHONG
Zhi Fuwei
XU XIAOLING
XU KAIXUAN
CUI YAN

Assignees

浙江锦华空分设备有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. The intelligent management system for the energy efficiency of the air separation equipment based on data analysis is characterized by comprising the following components: The data acquisition and quality calibration module is used for carrying out time alignment, deletion and exception processing on the instrument data and correcting the drift of the key measurement points based on quantile alignment of the golden window data; the working condition normalization energy efficiency baseline module is used for converting the yield of multiple products into equivalent yield and correcting the inlet air temperature and the environmental pressure to obtain a normalization energy efficiency index; The mechanism-data fusion digital twin module is used for superposing the simplified thermodynamic mechanism model output and the data residual error model to form energy efficiency prediction, determining an energy conservation error upper bound according to gold window statistics and applying monotonicity constraint; The component-level energy consumption attribution module is used for calculating a sensitivity matrix by the twin and solving the sensitivity matrix by sparse regularization to obtain energy consumption increment contribution and confidence coefficient; A robust suggestion generation module for solving and outputting maintenance priorities within boundaries only for operating variables consistent with high contribution and high confidence attribution, using the contribution and confidence as inverse constraints; and the verification self-calibration module is used for comparing verification benefits with neighborhood matching under the same working condition and executing attribution consistency verification after maintenance so as to trigger updating of the baseline, twin and suggested parameters.
2. The intelligent management system for energy efficiency of air separation equipment based on data analysis as set forth in claim 1, wherein the data collected by the data collection and quality calibration module comprises at least two of compressor power or total power of the device, intake air temperature, relative humidity, ambient pressure, temperature difference of key heat exchange ends, tower pressure difference, oxygen product flow and product purity.
3. The intelligent management system of air separation equipment energy efficiency based on data analysis of claim 2, wherein the golden window data is a data set of device acceptance, performance test time period and historical optimal stable operation time period, and the quantile alignment is used for enabling a measurement point to be corrected to be statistically consistent with the golden window at a preset quantile point so as to obtain drift correction.
4. The intelligent management system of air separation plant energy efficiency based on data analysis of claim 3, wherein the equivalent yield is obtained by converting the oxygen product yield by a conversion factor, and the conversion factor is determined according to device performance test data, historical steady operation data and digital twin energy consumption sharing results.
5. The intelligent management system for air separation equipment energy efficiency based on data analysis of claim 4, wherein the normalized energy efficiency index of the working condition satisfies the following conditions: ; Wherein, the Is a normalized energy efficiency index; Representing the energy consumption corresponding to the unit equivalent yield as a specific power index; Is the temperature of the intake air; Is the ambient pressure; And (3) with The air inlet temperature and the ambient pressure under the reference working condition are obtained; And (3) with The temperature correction coefficient and the pressure correction coefficient are respectively.
6. The intelligent management system for energy efficiency of air separation plant based on data analysis according to claim 5, wherein the intelligent management system is characterized in that And (3) with And carrying out robust regression identification through gold window data, and carrying out online correction on the coefficient by the verification self-calibration module based on the comparison result of the same working condition so as to inhibit seasonal drift.
7. The intelligent management system for energy efficiency of air separation equipment based on data analysis as set forth in claim 6, wherein said simplified thermodynamic mechanism model is at least covered with compression, heat exchange and rectification separation processes in said mechanism-data fusion digital twin module, said data residual model is trained with historical operating data for compensating for mechanism model unmodeled errors.
8. The intelligent management system of air separation plant energy efficiency based on data analysis of claim 7, wherein the upper energy conservation error bound is determined by a statistical bit value of an energy balance residual error in golden window data, and the monotonicity constraint is used for limiting the change direction of the normalized energy efficiency index to conform to a preset physical rule when at least one working condition variable is changed.
9. The intelligent management system of energy efficiency of space division equipment based on data analysis according to claim 8, wherein the component-level energy consumption attribution module obtains a sensitivity matrix through computing a bias derivative or numerical disturbance of the twin model, and adopts sparse regularized optimization to solve energy consumption increment contribution of an output Top-k component, and the confidence is obtained through resampling evaluation and Bayesian posterior evaluation.
10. The intelligent management system of air separation equipment energy efficiency based on data analysis of claim 9, wherein the verification self-calibration module constructs a comparison sample set before and after recommended execution and outputs an energy-saving benefit confidence interval through the same-condition neighborhood matching, and performs consistency verification on an attribution result and a maintenance object after a maintenance action occurs, and when the consistency does not meet a preset threshold, the digital twin module and the attribution module are triggered to update parameters or output an alarm.

Description

Intelligent management system for energy efficiency of air separation equipment based on data analysis Technical Field The invention relates to the technical field of energy efficiency management of space division equipment, in particular to an energy efficiency intelligent management system of space division equipment based on data analysis. Background The air separation device is used as typical high-energy-consumption complete equipment in the process industry, the operation energy efficiency is influenced by multi-loop coupling such as compression, heat exchange, rectification separation and pretreatment, the existing energy efficiency management is dependent on-line monitoring and trend analysis of power, yield and key temperature and pressure parameters, the prompt and post analysis of abnormal energy consumption are realized by setting a threshold value, single-point alarming or rule judgment based on experience, the change of indexes such as ton oxygen consumption and the like can be reflected by the existing scheme to a certain extent, the obvious deviation or sudden faults are found by operators, and the air separation device has the advantages of low deployment cost and realization of mature path. However, the air separation energy efficiency degradation is usually represented by slow variable drift caused by fouling of a heat exchanger, molecular sieve performance attenuation, valve leakage, entropy efficiency reduction of a compressor and the like, and ton oxygen consumption is often only slightly increased and is easily covered by environmental temperature, air inlet condition and output structure change, so that a simple threshold value or single-point alarm coverage is insufficient, and the early detection omission problem of normal and real degradation exists; In addition, after the energy efficiency deviation occurs, the existing system generally lacks working condition normalization evaluation and component level energy consumption attribution capability, and is difficult to detect how much energy consumption is lost due to a certain component, so that maintenance decision is dependent on experience and inaccurate action, and systematic burdens such as invalid or excessive maintenance, increased shutdown loss, alarm flooding, long-term energy consumption drifting and the like are easily caused. Disclosure of Invention The invention aims to solve the problem that in the prior art, the energy efficiency degradation is difficult to identify in early stage due to the lack of working condition normalization evaluation and component level energy consumption quantitative attribution of the existing air separation energy efficiency management system under complex working conditions, and provides an air separation equipment energy efficiency intelligent management system based on data analysis. In order to achieve the purpose, the technical scheme adopted by the invention is that the intelligent management system for the energy efficiency of the space division equipment based on data analysis comprises the following components: The data acquisition and quality calibration module is used for carrying out time alignment, deletion and exception processing on the instrument data and correcting the drift of the key measurement points based on quantile alignment of the golden window data; the working condition normalization energy efficiency baseline module is used for converting the yield of multiple products into equivalent yield and correcting the inlet air temperature and the environmental pressure to obtain a normalization energy efficiency index; The mechanism-data fusion digital twin module is used for superposing the simplified thermodynamic mechanism model output and the data residual error model to form energy efficiency prediction, determining an energy conservation error upper bound according to gold window statistics and applying monotonicity constraint; The component-level energy consumption attribution module is used for calculating a sensitivity matrix by the twin and solving the sensitivity matrix by sparse regularization to obtain energy consumption increment contribution and confidence coefficient; A robust suggestion generation module for solving and outputting maintenance priorities within boundaries only for operating variables consistent with high contribution and high confidence attribution, using the contribution and confidence as inverse constraints; and the verification self-calibration module is used for comparing verification benefits with neighborhood matching under the same working condition and executing attribution consistency verification after maintenance so as to trigger updating of the baseline, twin and suggested parameters. Specifically, the data collected by the data collection and quality calibration module at least comprises at least two types of compressor power or total power of the device, air inlet temperature, relative humidity, ambient pressure, key heat e