CN-121976248-A - Ultrapure water electrolysis intelligent monitoring system

CN121976248ACN 121976248 ACN121976248 ACN 121976248ACN-121976248-A

Abstract

The invention relates to the technical field of pure water electrolysis analysis control, in particular to an ultrapure water electrolysis intelligent monitoring system which comprises an electrolysis sensing module, an image analysis module, a uniformity judgment module and an electrolysis regulation and control execution module. The method comprises the steps of periodically collecting surface visual images of catalytic layers on two sides of a proton exchange membrane, monitoring gas purity, combining image meshing segmentation and feature extraction, identifying an abnormal visual area, intelligently judging the type of an abnormal root cause based on spatial distribution, time evolution and gas purity data of the abnormal area, automatically generating a side-specific regulation and control instruction according to a judgment result, and realizing full-flow intelligent monitoring from early abnormal identification, root cause diagnosis to self-adaptive regulation and control. The invention overcomes the defect that the traditional method is difficult to identify the microscopic degradation of the catalyst and locate the fault source, and remarkably improves the operation safety, reliability and maintenance efficiency of the PEM electrolytic hydrogen production system.

Inventors

LIU TINGTING
LV FENG
YANG WENWEI
XU HAIXU

Assignees

率氢技术(北京)股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (10)

1. An ultrapure water electrolysis intelligent monitoring system, characterized by comprising: The system comprises an electrolysis sensing module, a proton exchange membrane, a hydrogen concentration monitoring module and a hydrogen concentration monitoring module, wherein the electrolysis sensing module is used for periodically collecting surface visual image groups of catalytic layers on two sides of the proton exchange membrane and monitoring the oxygen concentration in hydrogen and the hydrogen concentration in oxygen on line, and the surface visual image groups comprise an anode side surface visual image and a cathode side surface visual image; The image analysis module is used for receiving the surface visual image group, dividing the catalytic layer region, and extracting the image characteristics of each divided catalytic region respectively to identify the abnormal visual region of the surface of the membrane electrode; the uniformity judging module is used for calculating and generating respective macroscopic uniform characterization indexes of catalyst coatings at two sides of the membrane electrode based on the geometric distribution dispersion of the abnormal visual area so as to judge whether the corresponding side catalytic layer is abnormal or not, and analyzing the spatial correspondence relation and time evolution correlation of the abnormal visual areas at two sides so as to judge an abnormal root cause type, wherein the abnormal root cause type comprises a single-side local type, a double-side systematic type or a membrane assembly penetrability type; the electrolysis regulation and control execution module receives the judging result of the abnormal root type and generates a corresponding side specific regulation and control instruction, wherein if the abnormal root is an anode side local problem, a regulating signal for anode water circulation flow and temperature is generated, if the abnormal root is a cathode side local problem, a regulating signal for cathode air flow pressure or humidity is generated, and if the abnormal root is a double-side systematic type or membrane assembly penetrability type problem, a load reduction operation, safety early warning and/or immediate maintenance instruction is generated.
2. The ultrapure water electrolysis intelligent monitoring system according to claim 1, wherein the anode side surface visual image is used for determining a state of an oxygen evolution catalyst layer, and the cathode side surface visual image is used for determining a state of a hydrogen evolution catalyst layer.
3. The ultrapure water electrolysis intelligent monitoring system according to claim 1, wherein the image analysis module divides the anode side surface visual image and the cathode side surface visual image according to n x n regular grids respectively to obtain a plurality of catalytic areas; The catalytic areas comprise anode catalytic areas and cathode catalytic areas, and the anode catalytic areas and the cathode catalytic areas are in one-to-one correspondence.
4. The ultrapure water electrolysis intelligent monitoring system according to claim 3, wherein the image analysis module calculates the area average brightness and the area brightness standard deviation of each of the catalytic areas, respectively; And identifying the abnormal visual zone based on the zone average luminance and zone luminance standard deviation for each of the catalytic zones.
5. The intelligent monitoring system for the electrolysis of ultrapure water according to claim 4, wherein the uniformity determination module calculates the abnormal area occupation ratio of each of the catalytic areas on both sides of the catalytic layer according to the abnormal visual areas identified by the image analysis module, and performs statistical analysis on the abnormal area occupation ratios of all the catalytic areas to calculate the macroscopic uniformity characterization index.
6. The ultrapure water electrolysis intelligent monitoring system according to claim 5, wherein the uniformity determination module determines whether or not there is an abnormality in the corresponding side catalytic layer based on the respective macroscopic uniformity characterization index of the both side catalytic coatings, And responding to a judging result that the macroscopic uniformity characterization index is smaller than or equal to a preset uniformity index, and judging that the corresponding side catalytic layer is normal by the uniformity judging module.
7. The ultrapure water electrolysis intelligent monitoring system according to claim 6, wherein the uniformity determination module establishes a spatial coordinate mapping relationship of the corresponding catalytic areas between the anode side surface visual image and the cathode side surface visual image, and determines a spatial correspondence relationship of the abnormal visual areas of the two catalytic layers based on the spatial coordinate mapping relationship; And acquiring abnormal area ratio data of the relevance abnormal region in a plurality of continuous detection periods, and analyzing the change trend of the abnormal area ratio data along with time to determine the time evolution relevance of the abnormal visual regions of the two catalytic layers.
8. The intelligent monitoring system for ultrapure water electrolysis according to claim 7, wherein the uniformity determination module determines according to the existence state of the abnormal visual area on the corresponding spatial positions of the anode side and the cathode side, and marks the corresponding positions as the relevance abnormal area if the abnormal visual areas occur simultaneously; And determining the type of the anomaly root cause based on the time variation trend of the anomaly area ratio data of the relevance anomaly area in a plurality of continuous detection periods; And if no relevance abnormal area is found or the change trend of the relevance abnormal area is asynchronous, judging that the abnormal root type is a single-side local type.
9. The ultrapure water electrolysis intelligent monitoring system according to claim 8, wherein the uniformity determination module is used for taking the oxygen concentration in the hydrogen, the hydrogen concentration in the oxygen and the abnormal area ratio data to further determine the abnormal root cause type; if the gas purity is in the deviation allowable range and the space overlapping area ratio is lower than a preset overlapping ratio threshold value, judging that the abnormal root cause type is a bilateral systematic type; And if the gas purity monitoring data exceeds the deviation allowable range and/or the space overlapping area ratio is higher than the preset overlapping ratio threshold value, judging that the abnormal source type is a membrane module penetrability type.
10. The ultrapure water electrolysis intelligent monitoring system according to claim 9, wherein the electrolysis regulation execution module generates a regulation instruction corresponding to the side pertinence according to the determination result of the abnormal root type, and comprises: Responding to the judging result to be of a single-side local type and corresponding to the anode side, and generating control signals of a flow regulating valve and a temperature regulating device of the anode water circulation system; responding to the judging result to be of a single-side local type and corresponding to the cathode side, and generating control signals of a back pressure regulating valve and a humidifying and dehumidifying device of the cathode side gas treatment unit; Responding to the judging result to be of a bilateral systematic type, generating a load reduction control signal of the direct-current power supply of the electrolytic tank, a control signal for carrying out balanced adjustment on the flow and the temperature of the anode and cathode feeding systems, and triggering a system parameter optimization alarm; And responding to the judging result to be the membrane assembly penetrability type, generating an emergency stop signal for the direct current power supply of the electrolytic tank, and generating an instant maintenance instruction for checking or replacing the membrane electrode.

Description

Ultrapure water electrolysis intelligent monitoring system Technical Field The invention relates to the technical field of pure water electrolysis analysis control, in particular to an ultrapure water electrolysis intelligent monitoring system. Background The Proton Exchange Membrane (PEM) pure water electrolysis hydrogen production technology is regarded as an important way for green hydrogen mass production because of the advantages of high efficiency, quick start and stop, high product purity, good adaptability to fluctuating renewable energy sources and the like. The core of the technology is that the physical and chemical states of the catalyst layers (anode oxygen evolution and cathode hydrogen evolution) on the two sides of the membrane electrode assembly, such as thickness uniformity, surface morphology integrity and active site distribution, directly determine the hydrogen production efficiency, energy consumption level and long-term operation life of the electrolytic cell. Currently, monitoring of PEM electrolyzer operating conditions in the industry generally relies on measurement of macroscopic electrothermal parameters of the system and purity of the outlet gas and threshold alarms. Although the method can catch serious performance decline or sudden faults, the microscopic and progressive physical degradation process of the catalyst layer is difficult to identify and position early, so the technical scheme which can surpass macroscopic parameter monitoring, realize in-situ visual evaluation of the physical state of the catalyst layer and can diagnose abnormal sources intelligently is urgently needed in the field, so that the operation reliability and safety of the PEM electrolytic hydrogen production system are improved. Disclosure of Invention Therefore, the invention provides an ultrapure water electrolysis intelligent monitoring system, which is used for solving the problems that the operation monitoring method of the proton exchange membrane pure water electrolytic tank in the prior art depends on macroscopic electric heating parameters and outlet gas purity monitoring, and is difficult to identify early stage of the physical degradation process of the catalyst coating on two sides of a membrane electrode, so that the service life of the proton exchange membrane is reduced. In order to achieve the above object, the present invention provides an ultrapure water electrolysis intelligent monitoring system, comprising: The system comprises an electrolysis sensing module, a proton exchange membrane, a hydrogen concentration monitoring module and a hydrogen concentration monitoring module, wherein the electrolysis sensing module is used for periodically collecting surface visual image groups of catalytic layers on two sides of the proton exchange membrane and monitoring the oxygen concentration in hydrogen and the hydrogen concentration in oxygen on line, and the surface visual image groups comprise an anode side surface visual image and a cathode side surface visual image; The image analysis module is used for receiving the surface visual image group, dividing the catalytic layer region, and extracting the image characteristics of each divided catalytic region respectively to identify the abnormal visual region of the surface of the membrane electrode; the uniformity judging module is used for calculating and generating respective macroscopic uniform characterization indexes of catalyst coatings at two sides of the membrane electrode based on the geometric distribution dispersion of the abnormal visual area so as to judge whether the corresponding side catalytic layer is abnormal or not, and analyzing the spatial correspondence relation and time evolution correlation of the abnormal visual areas at two sides so as to judge an abnormal root cause type, wherein the abnormal root cause type comprises a single-side local type, a double-side systematic type or a membrane assembly penetrability type; the electrolysis regulation and control execution module receives the judging result of the abnormal root type and generates a corresponding side specific regulation and control instruction, wherein if the abnormal root is an anode side local problem, a regulating signal for anode water circulation flow and temperature is generated, if the abnormal root is a cathode side local problem, a regulating signal for cathode air flow pressure or humidity is generated, and if the abnormal root is a double-side systematic type or membrane assembly penetrability type problem, a load reduction operation, safety early warning and/or immediate maintenance instruction is generated. As a preferable technical scheme of the ultrapure water electrolysis intelligent monitoring system, the visual image of the surface of the anode is used for determining the state of the oxygen evolution catalyst layer, and the visual image of the surface of the cathode is used for determining the state of the hydrogen evolution catalyst layer