Temperature sensor arrangement method and system for digital electrolytic cell

Abstract

The invention relates to the technical field of sensor deployment, in particular to a temperature sensor arrangement method and system for a digital electrolytic cell. The method comprises the following steps: acquiring structural data of the electrolytic cell; analyzing a topological structure of the electrolytic cell structure data, and performing three-dimensional digitalization on the electrolytic cell structure data based on the topological structure to construct an electrolytic cell digital model; extracting an electrolytic cell anode region, an electrolytic cell cathode region and an electrolyte flow channel of the electrolytic cell digital model; and analyzing the micro-precipitated phase interface regions of the electrolytic cell anode region and the electrolytic cell cathode region to determine sensor deployment modes of the electrolytic cell anode region and the electrolytic cell cathode region so as to obtain sensor deployment data of the electrode region. According to the method, digital modeling, region division and flow characteristic analysis are combined with virtual simulation, so that the problems of lack of scientific basis, poor pertinence and insufficient verification of traditional electrolytic cell temperature sensor arrangement are effectively solved, and accurate and efficient sensor optimization deployment is realized.

Description

Technical Field

[0001] This invention relates to the field of sensor deployment technology, and in particular to a method and system for arranging temperature sensors in a digital electrolytic cell. Background Technology

[0002] Early electrolytic cell temperature monitoring relied primarily on a few fixed-point thermocouples or thermistors, which struggled to comprehensively reflect the complex temperature distribution within the electrolytic cell, resulting in limited monitoring data and hindering precise control of the electrolysis process. With advancements in sensor technology, multi-point array temperature sensors emerged, enabling simultaneous temperature data acquisition across different areas of the electrolytic cell and improving spatial resolution. However, their layout design lacked systematicity, often leading to data redundancy or monitoring blind spots. In the era of digitalization and intelligentization, combined with a three-dimensional digital model of the electrolytic cell and based on computational fluid dynamics (CFD) and thermodynamic simulation analysis, temperature sensor placement methods have gradually become more scientific and optimized. Digital modeling of the polarization regions, flow channels, and key components within the electrolytic cell has enabled the design of targeted sensor deployment schemes. However, current arrangements are mostly empirical or rely on single-point measurements, neglecting the functional differences between the anode, cathode, and flow channels. Furthermore, the complex electrolyte flow makes it difficult for a single sensor to capture temperature changes in different flow channel sections, resulting in low accuracy and coverage of temperature sensor deployment. Summary of the Invention

[0003] Therefore, it is necessary to provide a method and system for arranging temperature sensors in a digital electrolytic cell to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for arranging temperature sensors in a digital electrolytic cell is provided, the method comprising the following steps:

[0005] Step S1: Obtain electrolytic cell structure data; analyze the topology of the electrolytic cell structure data, and perform three-dimensional digitization of the electrolytic cell structure data based on the topology to construct a digital model of the electrolytic cell;

[0006] Step S2: Extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-deposition phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region;

[0007] Step S3: Analyze the flow path of the electrolyte channel and segment the flow path to generate a near-mouth channel and a middle channel; analyze the eddy current backflow characteristics of the near-mouth channel and deploy the first channel sensor in the near-mouth channel to obtain the first channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle channel and deploy the second channel sensor in the middle channel to obtain the second channel sensor deployment data.

[0008] Step S4: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the digital model of the electrolytic cell to perform virtual temperature sensor deployment and monitoring and verification, so as to perform the temperature sensor layout optimization operation of the digital electrolytic cell.

[0009] This invention, through focused analysis of the micro-precipitated phase interface regions in the anode and cathode regions and the segmentation of the flow channel (near-mouth section and middle section), enables precise deployment of sensors in localized high-thermal-sensitivity or thermally disturbed areas, improving the spatial resolution and monitoring sensitivity of sensor data. Utilizing fluid dynamics characteristics (such as eddy current backflow and pulsation characteristics) to guide the placement of temperature sensors in the flow channel sections helps to accurately capture thermal anomalies in fluid disturbance areas, effectively reducing blind spots and redundant placement. Integrating sensor deployment data into the three-dimensional digital model of the electrolytic cell allows for simulation and verification of temperature response characteristics in a virtual environment, reducing physical testing costs and improving the feasibility and robustness of the deployment scheme. Based on the optimized temperature sensor arrangement, dynamic sensing and precise control of the temperature field during electrolysis can be achieved, helping to maintain optimal process parameters and improve product quality and energy efficiency. Precise sensor deployment lays a data foundation for subsequent intelligent tasks such as temperature field evolution analysis, anomaly detection, and predictive maintenance, enhancing the intelligence and safety of the electrolysis system. Therefore, this invention effectively solves the problems of lack of scientific basis, poor targeting and insufficient verification in the traditional electrolytic cell temperature sensor placement by combining digital modeling, region division and flow characteristic analysis with virtual simulation, and achieves accurate and efficient sensor optimization deployment.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Obtain electrolytic cell structure data;

[0012] Step S12: Identify the structural units of the electrolytic cell structure data, extract the boundaries of key components such as cathode, anode, cell shell, and diaphragm, and generate structural unit distribution data; construct a structural connection diagram based on the structural unit distribution data, perform logical mapping on the connection methods between each structural unit, and generate topological structure description data.

[0013] Step S13: Use the topological description data to perform three-dimensional spatial coordinate mapping, convert the position and relative relationship of the structural unit into three-dimensional configuration parameters, and generate spatial configuration data;

[0014] Step S14: Perform three-dimensional modeling on the spatial configuration data to generate three-dimensional digital configuration data of the electrolytic cell.

[0015] This invention identifies the boundaries of key components of an electrolytic cell (such as cathode, anode, cell shell, and diaphragm) and extracts the distribution of structural units, giving the original structural data a clear logical hierarchy and providing a structured foundation for subsequent analysis and simulation. Based on the structural connection diagram, topological description data is constructed, realizing the logical mapping of connection methods between structural units, effectively supporting the digital expression, rule abstraction, and analytical modeling of the internal structural relationships of the electrolytic cell. By performing three-dimensional coordinate mapping on the positional relationships of structural units and generating spatial configuration data, the model reconstruction not only reflects the real spatial layout but also ensures the integrity and consistency of structural relationships. The constructed three-dimensional digital configuration data can be directly used in various application scenarios such as CFD simulation analysis, sensor deployment simulation, and operating condition optimization analysis, possessing high adaptability and engineering operability. The structural logic-driven spatial configuration modeling method avoids repetitive construction and shape / position deviations in traditional manual three-dimensional modeling, significantly improving modeling efficiency and data consistency.

[0016] Preferably, step S2, which extracts the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer, includes:

[0017] Identify the boundary features of the anode components in the digital model of the electrolytic cell, and extract the structural data of the anode area and mark it as the anode area of ​​the electrolytic cell when the following conditions are met: the length of the anode plate is 800mm to 1500mm, the width is 400mm to 600mm, the thickness is not less than 10mm, the spacing between the anode plates is maintained in the range of 20mm to 50mm, and the anode connection structure is a continuous linear structure or an array arrangement.

[0018] When identifying the boundary morphology of the cathode region in the digital model of the electrolytic cell, and meeting the following geometric and material parameters, the cathode region structural data is extracted and marked as the cathode region of the electrolytic cell: cathode plate length is 850mm~1550mm, width is 450mm~650mm, plate thickness is not less than 12mm, cathode plate spacing is controlled between 25mm and 60mm, and the material property is conductivity greater than 4.5×10 7 Metallic materials with S / m;

[0019] Based on the spatial path identification results between the anode and cathode regions of the electrolytic cell, electrolyte flow channel structure data was extracted and labeled as electrolyte flow channels: the width of the main electrolyte channel is 100mm to 300mm, the width of the branch channels is not less than 30mm, the overall length of the flow channel is between 1.5m and 3.5m, and the rate of change of curvature is not greater than 0.2m. -1 Furthermore, the flow channel cross-section is a regular rectangle or ellipse.

[0020] This invention, based on the geometric dimensions (such as length, width, thickness, and spacing) and arrangement characteristics of the anode and cathode plates, can reliably identify the anode and cathode regions in an electrolytic cell, effectively avoiding structural misjudgments caused by model complexity or morphological similarity. By setting constraints on size ranges, arrangement methods, and material conductivity in the extraction rules, the extraction results more closely match the actual industrial structural standards of electrolytic cells, possessing high practical value adaptable to various engineering specifications. By limiting parameters such as channel curvature, width, length, and cross-sectional shape, interference from non-target structures can be eliminated, ensuring that the identified channels possess good fluid continuity and engineering flowability, providing an accurate spatial basis for subsequent flow field simulation and sensor deployment. The integration of multiple parameter conditions for combined judgment (such as geometry + material + connection method) upgrades structure extraction from "morphological recognition" to "structural semantic recognition," which is beneficial for improving the intelligent analysis capabilities of the electrolytic cell digital model. The highly accurate extraction of the anode, cathode, and electrolyte flow channels will serve as key foundational areas for subsequent digital operations such as temperature field distribution simulation, electric field analysis, and sensor deployment, significantly improving the rigor and applicability of the overall electrolysis system modeling.

[0021] Preferably, step S2 involves analyzing the micro-deposition phase interface regions in the anode and cathode regions of the electrolytic cell to determine the sensor deployment patterns in the anode and cathode regions of the electrolytic cell, including:

[0022] Extract microstructure images of the anode and cathode regions of the electrolytic cell;

[0023] Interface feature enhancement is performed on microstructure image data to generate interface-enhanced images;

[0024] Edge recognition and phase interface segmentation are performed on the interface enhancement image to obtain micro-extruded phase interface contour data;

[0025] Calculate the morphological change gradient of the micro-precipitated phase interface contour data and generate an interface morphological gradient distribution map.

[0026] Based on the interface morphology gradient distribution map, the concentrated and sparse regions of precipitates are identified, and a precipitate distribution density map is generated.

[0027] Sensitive regions are extracted based on the precipitate distribution density map to identify active micro-precipitate regions in the anode and cathode regions;

[0028] Based on the active region of the micro-precipitated phase, sensor sensing optimization analysis is performed to determine the sensor deployment mode in the anode and cathode regions;

[0029] Generate sensor deployment data for the anode area and sensor deployment data for the cathode area based on the sensor deployment mode;

[0030] By integrating the sensor deployment data of the anode region and the sensor deployment data of the cathode region, the sensor deployment data of the electrode region is obtained.

[0031] This invention achieves high-precision segmentation of micro-precipitated phase interfaces by employing interface feature enhancement and edge recognition technologies, effectively capturing interface details and subtle morphological changes, thus avoiding identification errors caused by interface ambiguity in traditional methods. By generating morphological gradient distribution maps and distribution density maps, the spatial distribution and activity level of the precipitated phases can be objectively quantified, providing a reliable quantitative basis for subsequent sensitive area identification. Sensitive area extraction and optimization analysis based on the active regions of the precipitated phases ensure that sensor deployment focuses on the microscopic regions with the most significant structural changes and performance fluctuations, significantly improving the sensitivity and response efficiency of sensor monitoring. Combined with the differences in microscopic precipitation characteristics of each region, targeted sensor deployment patterns are formulated to meet the monitoring needs of different electrode regions during electrolysis, improving the overall targeting and practicality of the monitoring system. Integrating sensor deployment data from the anode and cathode regions forms a complete electrode region sensor layout scheme, facilitating unified management and subsequent intelligent analysis within the digital model. Precise monitoring layout of the active micro-precipitated phase regions helps to promptly capture local electrode anomalies and performance degradation, improving the operational safety and maintenance efficiency of the electrolyzer.

[0032] Preferably, step S3, analyzing the flow path of the electrolyte channel and segmenting the flow path of the electrolyte channel, includes:

[0033] Extract the electrolyte flow channel geometry data from the electrolytic cell structure data;

[0034] The starting point of the electrolyte flow channel is identified based on the electrolyte flow channel geometry data, and an initial flow path reference line is constructed.

[0035] Flow simulation was performed on the flow path of the electrolyte channel to obtain the electrolyte velocity field and pressure distribution field;

[0036] Based on the electrolyte velocity field and pressure distribution field, the main flow path is extracted to generate the main flow path data.

[0037] The main flow path data is divided into equally spaced sections along its length, and key path nodes for changes in flow velocity are extracted.

[0038] Based on the flow path reference line, the near-entry segment path range of key path nodes is identified, and the near-entry segment flow channel is generated.

[0039] The remaining channels in the electrolyte flow path are selected based on the near-mouth flow channel and marked as the middle channel.

[0040] This invention accurately constructs a reference line for the flow path of the electrolyte channel based on the geometric features of structural data and flow simulation results, ensuring a high degree of consistency between fluid dynamic characteristics and spatial geometry. By extracting key nodes of flow velocity changes, the distribution location of flow characteristic changes is precisely grasped, providing a scientific basis for subsequent channel segmentation and local flow anomaly analysis. Using an equidistant division combined with a near-entry path range identification method, two different flow characteristic regions—the near-entry channel and the middle channel—are effectively divided, facilitating targeted fluid dynamics optimization and sensor deployment. By closely integrating velocity and pressure field data with the channel structure, the analysis of flow characteristics and geometric distribution is achieved, providing a precise spatial basis for flow anomaly detection and sensor placement. The channel segmentation results lay the foundation for in-depth analysis of subsequent eddy current recirculation and eddy current pulsation characteristics, promoting refined research on fluid behavior inside the electrolyzer and optimization of process parameters. The scientific segmentation method reduces subjectivity and experience dependence in channel segmentation, improves the objectivity and repeatability of channel analysis, and is conducive to the systematization and standardization of sensor deployment.

[0041] Preferably, step S3, which involves analyzing the eddy current recirculation characteristics of the near-mouth section flow channel and deploying the first flow channel sensor in the near-mouth section flow channel, includes:

[0042] The structural parameter data of the near-mouth section flow channel are divided into cross-sectional partitions to generate local flow region data of the near-mouth section.

[0043] Fluid particle trajectory simulation is performed based on local flow region data near the inlet section to generate eddy flow return path data.

[0044] The disturbance frequency of the return flow core area is analyzed using eddy current return flow path data to generate return flow disturbance characteristic data;

[0045] Based on the backflow disturbance characteristic data, the first type of sensor deployment sensitive sections are identified, and sensor sensitive deployment site data are generated;

[0046] Structural interference assessment is performed on sensor sensitive deployment site data to eliminate areas with excessive interference and generate candidate sensor deployment site data.

[0047] Based on the sensor candidate deployment point data and the flow direction alignment principle, point calibration is performed to generate the first flow channel sensor deployment data.

[0048] This invention accurately depicts the eddy current recirculation path through local flow region partitioning and particle trajectory simulation, revealing in-depth complex fluid disturbance behavior within the near-mouth section of the flow channel. Based on the disturbance frequency data of the eddy current recirculation path, the dynamic characteristics of the recirculation core area are scientifically quantified, providing quantitative indicators for subsequent monitoring and control. Sensitive sections are identified using disturbance characteristic data, ensuring that sensor placement is focused on key locations where fluid disturbances are most significant and signal responses are strongest, improving monitoring sensitivity and accuracy. High-interference areas are eliminated through structural interference assessment, avoiding false alarms caused by sensor complexity and improving the stability and data quality of the monitoring system. Point calibration is performed based on flow direction alignment principles, achieving coordinated optimization of sensor placement and fluid movement direction, improving the efficiency and completeness of sensor capture of flow information. The overall sensor deployment scheme achieves efficient perception of eddy current recirculation characteristics, assisting in electrolytic cell flow field monitoring, fault diagnosis, and operational optimization, improving system safety and production efficiency.

[0049] Preferably, step S3, analyzing the vortex pulsation characteristics of the mid-section flow channel and deploying the second flow channel sensor in the mid-section flow channel, includes:

[0050] The geometric parameters of the middle channel are mapped to the channel region to generate spatial grid data of the middle channel.

[0051] Based on the spatial grid data of the mid-section flow channel, velocity and pressure field variation data are extracted to generate flow state data of the mid-section flow channel.

[0052] The instantaneous velocity disturbance is calculated based on the flow state data of the middle channel, and vortex pulsation spectrum data is generated.

[0053] Extract the dominant frequency of the vortex pulsation spectrum data, and use the dominant frequency to identify the high response zone of the middle section of the flow channel to generate sensor pulsation response region data;

[0054] Spatial distribution equalization optimization is performed on the sensor pulse response region data to generate a second type of deployable point set data for sensors.

[0055] Based on the deployable point set data of the second type of sensor, deployment sites are selected according to the principle of minimum flow interference to generate deployment data for the second flow channel sensor.

[0056] This invention achieves high-resolution capture of flow field details by extracting velocity and pressure field changes based on spatial grid data, enhancing the dynamic analysis capability of flow states. By calculating instantaneous velocity disturbances and performing spectral analysis, it scientifically identifies the dominant frequency of pulsations and their dynamic response, improving the depth of understanding of flow disturbance characteristics. Based on dominant frequency information, it identifies key response regions, providing a clearly defined spatial range for sensor deployment, improving the targeting and sensitivity of monitoring. It ensures a uniform distribution of sensor deployment points, avoiding blind spots and redundancy, and improving the coverage and data representativeness of the monitoring system. Deployment sites are selected based on the principle of minimum flow interference, ensuring that sensor installation does not significantly affect the natural flow state of the flow field, maintaining measurement accuracy. The overall sensor deployment scheme enables effective perception of the vortex pulsation characteristics in the mid-channel flow, supporting fault warning, flow control, and process optimization, improving the safety and efficiency of electrolyzer operation.

[0057] Preferably, step S4 includes the following steps:

[0058] Step S41: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the electrolytic cell digital model for spatial arrangement mapping, and generate three-dimensional deployment candidate point cloud data;

[0059] Step S42: Initialize the thermal field simulation of the 3D deployment candidate point cloud data to generate initial temperature distribution estimation data; perform virtual temperature sensor deployment simulation based on the initial temperature distribution estimation data to generate virtual temperature sensor deployment scheme data;

[0060] Step S43: Perform temperature response simulation tests on the electrolytic cell based on the virtual temperature sensor deployment scheme data to generate simulation monitoring response data; conduct a comparative analysis of the virtual and real arrangement differences on the simulation monitoring response data to generate arrangement deviation evaluation results;

[0061] Step S44: Based on the layout deviation assessment results, perform virtual deployment adjustment and optimization to generate optimized digital temperature sensor layout data, and execute the temperature sensor layout optimization operation of the digital electrolytic cell.

[0062] This invention generates a 3D point cloud from sensor deployment data, accurately reflecting the spatial distribution characteristics of sensors within the electrolytic cell and improving the spatial rationality of the deployment scheme. By utilizing thermal field simulation initialization and temperature distribution estimation, it scientifically predicts temperature changes around the sensors, ensuring the coverage of key temperature areas by the deployment scheme. Through a combination of virtual sensor deployment simulation and actual temperature response simulation testing, the response effect of the deployment scheme is accurately evaluated, ensuring that the simulated scheme meets actual operating conditions. A comparative analysis of deployment deviations quantifies the differences between virtual and real deployments, guiding adjustments and improving the accuracy and monitoring effect of sensor layout. Dynamic adjustments are implemented based on simulation feedback results, forming a closed-loop optimization mechanism to achieve intelligent, dynamic, and efficient temperature sensor deployment. The optimized deployment scheme ensures that key temperature change areas are effectively monitored, improving the real-time performance and accuracy of overall temperature monitoring, which is beneficial for the safe operation and process control of the electrolytic cell.

[0063] Preferably, step S43, which involves performing a temperature response simulation test on the electrolyzer based on the virtual temperature sensor deployment scheme data, includes:

[0064] Based on the data from the virtual temperature sensor deployment scheme, simulation monitoring nodes are set up within a range of 20mm to 50mm directly above the anode plate of the electrolytic cell, 30mm to 60mm on both sides of the cathode plate, and 10mm to 30mm on both sides of the central axis of the electrolyte flow channel. The total number of simulation nodes is no less than 30, and the spacing is no more than 100mm, forming a complete temperature field monitoring network.

[0065] The initial simulation conditions were set as follows: ambient temperature was set to 25℃±2℃, initial electrolytic liquid temperature was set to 50℃~70℃, and boundary heat transfer coefficient was set to 30~80W / (m²). 2 •K), and gradually sample the temperature sensing response data over a simulation period of 120 minutes;

[0066] During the simulation, constant current operating conditions were simulated, and the dynamic temperature response data of the anode region, cathode region, and flow channel region were recorded. When the temperature rise rate exceeded 2.5℃ / min or the local temperature exceeded 90℃, it was recorded as a high-temperature risk point. Specifically, the constant current operating conditions were set with a current density range of 150~300A / m. 2 ;

[0067] The dynamic temperature response change data of the anode region, cathode region and flow channel region are integrated into the simulation monitoring response data.

[0068] This invention forms a comprehensive temperature monitoring network covering the anode, cathode, and core flow channel areas by rationally setting the location and spacing of simulation nodes, ensuring the comprehensiveness and continuity of temperature data. By combining the ambient temperature fluctuation range, the initial electrolyte temperature, and boundary heat transfer characteristics, it accurately simulates the actual operating environment of the electrolyzer, enhancing the reliability and applicability of the simulation results. Step-by-step sampling within a 120-minute simulation cycle meticulously records the temperature change over time, helping to promptly identify temperature anomalies and high-risk points. Dynamic monitoring based on the temperature rise rate and local temperature threshold effectively locates potential high-temperature risk points, providing data support for preventing local overheating and ensuring the safe operation of the electrolyzer. This applies to temperatures ranging from 150 to 300 A / m. 2 The current density range simulates real operating conditions, ensuring the relevance and practicality of the simulation monitoring response data. It integrates dynamic temperature response data from the anode, cathode, and flow channel regions to form comprehensive simulation monitoring response data, guiding adjustments to the layout scheme and improving the response sensitivity and coverage of the temperature sensing system.

[0069] This specification provides a temperature sensor arrangement system for a digital electrolytic cell, used to execute the above-described temperature sensor arrangement method for a digital electrolytic cell. The temperature sensor arrangement system for the digital electrolytic cell includes:

[0070] The digitization module is used to acquire electrolytic cell structural data; analyze the topological structure of the electrolytic cell structural data; and perform three-dimensional digitization of the electrolytic cell structural data based on the topological structure to construct a digital model of the electrolytic cell.

[0071] The polarization region deployment module is used to extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-precipitated phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region;

[0072] The flow channel deployment module is used to analyze the flow path of the electrolyte flow channel and segment the flow path to generate a near-mouth flow channel and a middle flow channel; analyze the eddy current backflow characteristics of the near-mouth flow channel and deploy the first flow channel sensor in the near-mouth flow channel to obtain the first flow channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle flow channel and deploy the second flow channel sensor in the middle flow channel to obtain the second flow channel sensor deployment data.

[0073] The monitoring and verification module is used to input the sensor deployment data of the electrode area, the sensor deployment data of the first flow channel, and the sensor deployment data of the second flow channel into the digital model of the electrolyzer to perform virtual temperature sensor deployment and monitoring and verification, so as to perform temperature sensor layout optimization work in the digital electrolyzer.

[0074] The beneficial effects of this invention lie in its accurate construction of a digital model of the electrolyzer through topological analysis and three-dimensional digital processing, providing a solid data foundation for subsequent sensor deployment. The anode region, cathode region, and electrolyte flow channel are extracted, and based on the analysis of the micro-precipitated phase interface region, the sensor deployment mode is precisely determined, enabling accurate monitoring of the polarization region. The electrolyte flow channel is subdivided into near-mouth and middle sections, and different types of sensors are deployed based on eddy current backflow and eddy current pulsation characteristics, respectively, ensuring effective capture of fluid dynamic changes. Through virtual temperature sensor deployment and monitoring verification, combined with the digital model, the layout scheme is dynamically adjusted, significantly improving the coverage efficiency and monitoring accuracy of the sensor network. Real-time and accurate monitoring of temperature changes in key areas of the electrolyzer is achieved, promoting safe operation and process optimization. The modules cooperate with each other while remaining relatively independent, facilitating system function upgrades and customized development to meet the application needs of different electrolyzers. Therefore, this invention, through digital modeling, region division, and flow characteristic analysis combined with virtual simulation, effectively solves the problems of lack of scientific basis, poor targeting, and insufficient verification in traditional electrolyzer temperature sensor deployment, achieving precise and efficient optimized sensor deployment. Attached Figure Description

[0075] Figure 1 A schematic diagram illustrating the steps of a method for arranging temperature sensors in a digital electrolytic cell;

[0076] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1.

[0077] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S4.

[0078] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0079] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0080] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0081] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0082] To achieve the above objectives, please refer to Figures 1 to 3 A method for arranging temperature sensors in a digital electrolytic cell, the method comprising the following steps:

[0083] Step S1: Obtain electrolytic cell structure data; analyze the topology of the electrolytic cell structure data, and perform three-dimensional digitization of the electrolytic cell structure data based on the topology to construct a digital model of the electrolytic cell;

[0084] Step S2: Extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-deposition phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region;

[0085] Step S3: Analyze the flow path of the electrolyte channel and segment the flow path to generate a near-mouth channel and a middle channel; analyze the eddy current backflow characteristics of the near-mouth channel and deploy the first channel sensor in the near-mouth channel to obtain the first channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle channel and deploy the second channel sensor in the middle channel to obtain the second channel sensor deployment data.

[0086] Step S4: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the digital model of the electrolytic cell to perform virtual temperature sensor deployment and monitoring and verification, so as to perform the temperature sensor layout optimization operation of the digital electrolytic cell.

[0087] This invention, through focused analysis of the micro-precipitated phase interface regions in the anode and cathode regions and the segmentation of the flow channel (near-mouth section and middle section), enables precise deployment of sensors in localized high-thermal-sensitivity or thermally disturbed areas, improving the spatial resolution and monitoring sensitivity of sensor data. Utilizing fluid dynamics characteristics (such as eddy current backflow and pulsation characteristics) to guide the placement of temperature sensors in the flow channel sections helps to accurately capture thermal anomalies in fluid disturbance areas, effectively reducing blind spots and redundant placement. Integrating sensor deployment data into the three-dimensional digital model of the electrolytic cell allows for simulation and verification of temperature response characteristics in a virtual environment, reducing physical testing costs and improving the feasibility and robustness of the deployment scheme. Based on the optimized temperature sensor arrangement, dynamic sensing and precise control of the temperature field during electrolysis can be achieved, helping to maintain optimal process parameters and improve product quality and energy efficiency. Precise sensor deployment lays a data foundation for subsequent intelligent tasks such as temperature field evolution analysis, anomaly detection, and predictive maintenance, enhancing the intelligence and safety of the electrolysis system. Therefore, this invention effectively solves the problems of lack of scientific basis, poor targeting and insufficient verification in the traditional electrolytic cell temperature sensor placement by combining digital modeling, region division and flow characteristic analysis with virtual simulation, and achieves accurate and efficient sensor optimization deployment.

[0088] In this embodiment of the invention, reference Figure 1 The diagram shown is a flowchart illustrating the steps of a temperature sensor arrangement method for a digital electrolytic cell according to the present invention. In this example, the temperature sensor arrangement method for the digital electrolytic cell includes the following steps:

[0089] Step S1: Obtain electrolytic cell structure data; analyze the topology of the electrolytic cell structure data, and perform three-dimensional digitization of the electrolytic cell structure data based on the topology to construct a digital model of the electrolytic cell;

[0090] Step S2: Extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-deposition phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region;

[0091] Step S3: Analyze the flow path of the electrolyte channel and segment the flow path to generate a near-mouth channel and a middle channel; analyze the eddy current backflow characteristics of the near-mouth channel and deploy the first channel sensor in the near-mouth channel to obtain the first channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle channel and deploy the second channel sensor in the middle channel to obtain the second channel sensor deployment data.

[0092] Step S4: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the digital model of the electrolytic cell to perform virtual temperature sensor deployment and monitoring and verification, so as to perform the temperature sensor layout optimization operation of the digital electrolytic cell.

[0093] In this embodiment of the invention, structural data of the electrolytic cell is acquired. This data originates from 3D laser scanning, structural design drawings, or digital modeling documents, and includes information such as the length, width, height, electrode installation positions, and electrolyte channel structure of the electrolytic cell. Next, topological analysis is performed on the electrolytic cell structural data, including identifying the connection relationships, spatial hierarchy, and structural unit divisions between key internal components. Examples include the anode region, cathode region, electrolyte inlet and outlet, support structure, and inner wall geometry. Subsequently, based on the topological information, a 3D digital model is created from the structural data. Using CAD / BIM modeling, the electrolytic cell is restored to a real-scale 3D geometric model, generating a "digital model of the electrolytic cell" with spatial layering logic. This model contains the positional information and spatial attributes of each structural unit and supports subsequent embedded deployment simulation operations. The anode region, cathode region, and the flow channels holding the electrolyte are extracted from the digital model of the electrolytic cell. The focus is on the "micro-precipitation phase interface" at the junction of the electrode and electrolyte, i.e., the region where metal or gas precipitation is frequent, typically within 10 cm of the lower edge of the anode and the upper edge of the cathode. Microscopic interface characteristic analysis was conducted in this area. Historical operating data and thermal field distribution simulations identified key monitoring locations characterized by drastic local temperature changes, concentrated precipitation, and unstable hotspots. Temperature sensor points were installed symmetrically in the anode and cathode areas, spaced 5 to 8 centimeters apart. The output was "Electrode Area Sensor Deployment Data," which included the point number, specific coordinates (based on the center of the electrolyzer), area classification (anode or cathode), and sensing target type (temperature anomaly detection or precipitation hotspot tracking). The area approximately 20 centimeters before the electrolyte inlet was defined as the "near-inlet flow channel." This area is prone to backflow and local vortices, affecting the uniformity of temperature distribution. Typical backflow vortex centers were identified through fluid dynamics simulations (such as CFD simulations), and first-type temperature sensors were arranged at equal intervals around these centers along the inlet wall. Typically, 3 to 5 points were set, spaced 5 centimeters apart, with the sensors attached to the tank wall or suspended in the middle of the flow channel. The central region of the electrolyzer (from 20 cm after the inlet to approximately 20 cm before the outlet) is defined as the "mid-section flow channel." This region commonly experiences eddy current pulsations and temperature disturbances, requiring intensive monitoring of dynamic changes. Areas with significant flow velocity fluctuations are identified through high-frequency disturbance simulation and historical anomaly records, and a second type of temperature sensor is deployed in these areas. This type of sensor deployment is more meticulous, with a recommended spacing of 3 cm, and multiple points are placed in each cross-section to form a planar monitoring network. The deployment results are output as "first flow channel sensor deployment data" and "second flow channel sensor deployment data," containing point coordinates, functional type, response time requirements, and location level (wall-attached / floating in the tank), respectively. The aforementioned three types of sensor deployment data (electrode area, first flow channel, second flow channel) are imported into the electrolyzer's digital model and virtually deployed using simulation tools. Each sensor point is mapped to a specific structural location in the 3D model, and a monitoring thermal field distribution response relationship is established.The thermal simulation results are used to verify whether each placement point can cover: areas of drastic temperature changes; areas of abnormal thermal coupling; and points with high-frequency response points. If some areas are not effectively covered, the sensor layout is automatically adjusted or manually optimized. After the simulation verification is completed, a "virtual temperature monitoring model of the electrolytic cell" is generated. This model can be used for subsequent layout drawing output and integration with the real-time temperature monitoring system.

[0094] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S1 includes:

[0095] Step S11: Obtain electrolytic cell structure data;

[0096] Step S12: Identify the structural units of the electrolytic cell structure data, extract the boundaries of key components such as cathode, anode, cell shell, and diaphragm, and generate structural unit distribution data; construct a structural connection diagram based on the structural unit distribution data, perform logical mapping on the connection methods between each structural unit, and generate topological structure description data.

[0097] Step S13: Use the topological description data to perform three-dimensional spatial coordinate mapping, convert the position and relative relationship of the structural unit into three-dimensional configuration parameters, and generate spatial configuration data;

[0098] Step S14: Perform three-dimensional modeling on the spatial configuration data to generate three-dimensional digital configuration data of the electrolytic cell.

[0099] In this embodiment of the invention, structural parameter information, including length, width, height, plate thickness, electrode arrangement, and electrolyte volume, is extracted by reading the design drawings, two-dimensional structural diagrams, construction specifications, and original CAD files of the electrolytic cell. The electrolytic cell is then scanned using a laser scanning or structured light imaging system to extract surface geometric contours and features such as holes, protrusions, and diaphragms, generating preliminary three-dimensional point cloud data or a mesh structure. This ultimately forms "electrolytic cell structural data," containing component dimensions, positions, shapes, and connection boundaries, providing a foundation for subsequent boundary identification and spatial modeling. The main functional component boundaries of the electrolytic cell are identified and extracted from the structural data: Cathode: identifying the conductive material units at the bottom and sides, extracting their surface areas and slot positions; Anode: extracting the boundaries of the suspended electrode plates in the middle or top; Shell: identifying the overall structural shell boundaries, including the surrounding walls and base structure; Diaphragm: analyzing the thin-layer component between the electrodes, serving as an ion isolation layer. The extracted boundary data is labeled, numbered, and partitioned to form structural unit distribution data. This data includes the shape, spatial position, boundary range, functional type, and coordinate attributes of each structural unit. Based on the structural unit distribution data, a structural connection diagram is established to represent the physical connection methods and spatial adjacency relationships between each unit. For example, the anode and cathode are separated by an electrolyte region; the tank shell is in close contact with the cathode, serving as a fixed support for the negative electrode; the diaphragm runs through the anode and cathode, corresponding to the conduction path. The final output is "topology description data," used to describe the spatial connectivity and structural combination relationships between components. After obtaining the topology description data, a three-dimensional spatial mapping operation is performed: based on the boundary range of each structural unit, its center coordinate point is defined, generating an initial spatial position set; relationships such as "contact," "interval," and "embedding" in the structural connection diagram are converted into numerical parameters such as spatial spacing, included angles, and normal vectors; based on the above spatial positions and relative relationships, the positioning parameters of each structural unit in three-dimensional space are generated, including: three-dimensional coordinate position (x, y, z); rotation angle (orientation around each axis); relative offset; and local connection surface direction. Finally, "spatial configuration data" is formed, providing complete coordinate constraints and component orientations for the next step of modeling. Using modeling software (such as SolidWorks, Rhino, and Revit) or custom modeling scripts, corresponding 3D geometric models are constructed according to the shape, size, and spatial position of each component. All components are assembled into a unified 3D electrolytic cell structure based on their actual connection relationships. Detailed modeling is performed on each component, including chamfering of edges and corners, opening locations, insulation layer thickness, and mounting interface shapes, ensuring the model possesses the same technological characteristics as the actual object. Simulation preview, overlap detection, and spatial collision analysis are used to verify whether there are dimensional conflicts or mismatches between structural components. If conflicts are found, they are corrected by fine-tuning spatial parameters.The final output is a standardized "three-dimensional digital configuration data of electrolytic cell", including a complete structural model, component attribute information and spatial relationship structure, which supports subsequent simulation analysis, manufacturing planning and sensor placement.

[0100] Preferably, step S2, which extracts the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer, includes:

[0101] Identify the boundary features of the anode components in the digital model of the electrolytic cell, and extract the structural data of the anode area and mark it as the anode area of ​​the electrolytic cell when the following conditions are met: the length of the anode plate is 800mm to 1500mm, the width is 400mm to 600mm, the thickness is not less than 10mm, the spacing between the anode plates is maintained in the range of 20mm to 50mm, and the anode connection structure is a continuous linear structure or an array arrangement.

[0102] When identifying the boundary morphology of the cathode region in the digital model of the electrolytic cell, and meeting the following geometric and material parameters, the cathode region structural data is extracted and marked as the cathode region of the electrolytic cell: cathode plate length is 850mm~1550mm, width is 450mm~650mm, plate thickness is not less than 12mm, cathode plate spacing is controlled between 25mm and 60mm, and the material property is conductivity greater than 4.5×10 7 Metallic materials with S / m;

[0103] Based on the spatial path identification results between the anode and cathode regions of the electrolytic cell, electrolyte flow channel structure data was extracted and labeled as electrolyte flow channels: the width of the main electrolyte channel is 100mm to 300mm, the width of the branch channels is not less than 30mm, the overall length of the flow channel is between 1.5m and 3.5m, and the rate of change of curvature is not greater than 0.2m. -1 Furthermore, the flow channel cross-section is a regular rectangle or ellipse.

[0104] In this embodiment of the invention, by extracting the anode region, cathode region, and electrolyte flow channel from the digital model of the electrolytic cell, the boundary of the digital model is first identified to extract plate-shaped components as candidate anode components. These are then screened based on their geometric parameters: when the component's length is between 800 mm and 1500 mm, its width is between 400 mm and 600 mm, its thickness is not less than 10 mm, its plate spacing is between 20 mm and 50 mm, and its arrangement is continuous linear or array-like, the structure is determined to be the anode region of the electrolytic cell and spatially labeled. Subsequently, plate-shaped components located relative to the anode region in the model are identified. Components with a length between 850 mm and 1550 mm, a width between 450 mm and 650 mm, and a thickness not less than 10 mm are selected. When the plate spacing is controlled between 25 mm and 60 mm and the material conductivity is higher than 45 million Siemens per meter, it is marked as the cathode area of ​​the electrolytic cell. Between the anode and cathode areas, the spatial channel structure is further identified. When the main channel width is 100 mm to 300 mm, the branch channel width is not less than 30 mm, the overall length is between 1.5 m and 3.5 m, the curvature change rate is not greater than 0.2 m per meter, and the cross-section is a regular rectangle or ellipse, the area is marked as the electrolyte flow channel. The above extraction process combines structural boundary analysis, geometric dimension comparison and material property judgment, and performs spatial position marking and functional identification of all identified areas in the three-dimensional model, thereby completing the high-precision extraction of the functional areas of the anode area, cathode area and electrolyte flow channel of the electrolytic cell.

[0105] Preferably, step S2 involves analyzing the micro-deposition phase interface regions in the anode and cathode regions of the electrolytic cell to determine the sensor deployment patterns in the anode and cathode regions of the electrolytic cell, including:

[0106] Extract microstructure images of the anode and cathode regions of the electrolytic cell;

[0107] Interface feature enhancement is performed on microstructure image data to generate interface-enhanced images;

[0108] Edge recognition and phase interface segmentation are performed on the interface enhancement image to obtain micro-extruded phase interface contour data;

[0109] Calculate the morphological change gradient of the micro-precipitated phase interface contour data and generate an interface morphological gradient distribution map.

[0110] Based on the interface morphology gradient distribution map, the concentrated and sparse regions of precipitates are identified, and a precipitate distribution density map is generated.

[0111] Sensitive regions are extracted based on the precipitate distribution density map to identify active micro-precipitate regions in the anode and cathode regions;

[0112] Based on the active region of the micro-precipitated phase, sensor sensing optimization analysis is performed to determine the sensor deployment mode in the anode and cathode regions;

[0113] Generate sensor deployment data for the anode area and sensor deployment data for the cathode area based on the sensor deployment mode;

[0114] By integrating the sensor deployment data of the anode region and the sensor deployment data of the cathode region, the sensor deployment data of the electrode region is obtained.

[0115] In this embodiment of the invention, a high-resolution industrial camera (e.g., a 50-megapixel CMOS camera) is used to capture images of the microstructure region near the electrolyte interface between the anode and cathode regions of the electrolytic cell, obtaining feature images of the electrode surface. The image coverage is preferably an area extending inward from the electrode edge by approximately 10 millimeters, and the shooting angle is set to an orthogonal direction to ensure geometric accuracy. The extracted microstructure image must have a minimum resolution of 1 micrometer per pixel for subsequent detail enhancement and boundary detection processing. Feature enhancement processing is performed on the microstructure image, including: grayscale enhancement (enhancing contrast); local edge sharpening; high-frequency noise filtering; and image histogram equalization. The processing goal is to highlight the metal precipitation edges and the heterogeneous areas of the electrode surface, making the microprecipitated phase interface more prominent. The output is an interface-enhanced image, suitable for subsequent edge extraction and segmentation operations. Based on the enhanced image data, the edges of the precipitated phase region are identified. The edge region is continuously segmented using region growing or active contour models to extract the boundary contours of micro-precipitates, generating standardized "micro-precipitate interface contour data." This data includes descriptive information such as contour coordinates, region contour area, and boundary roughness. Using this data, gradient analysis is performed on the geometric variations of the contours, including local curvature changes, boundary fluctuation frequency, and concavity / convexity depth gradients. A "gradient distribution map of the interface morphology" is generated in a two-dimensional image coordinate system using a grid discretization method (e.g., 0.5 mm as the grid unit). This map clearly reflects the degree of drastic change in the interface morphology of the precipitation region and is the basis for determining the intensity of precipitation. Combined with the gradient distribution map, the frequency of precipitation occurrence per unit area (e.g., per square millimeter) within the region is statistically analyzed to determine the density of the precipitation region. A "precipitate distribution density map" is output, where regions with a density higher than a set threshold (e.g., more than 5 precipitation units per square millimeter) are marked as "precipitate concentration feature regions," and regions with a density lower than the set value are marked as "precipitate sparse regions." Cluster analysis of the geometric connectivity, positional stability, and distribution area of ​​dense precipitation zones identifies sensitive regions with long-term active precipitation trends within the anode and cathode regions. Generally, 3 to 5 active micro-precipitated phase regions can be identified in each electrode region, with an area of ​​not less than 10 square millimeters, and a strip-like or patchy distribution. Based on the location of these active regions, a placement simulation analysis is performed, considering the following factors: vertical distance from the interface center (preferably within the range of 2–4 mm); historical temperature fluctuation data of the region; thermal field overlap between proposed deployment points; and structural accessibility and maintenance availability. The optimal placement locations are selected, typically establishing 1–2 temperature sensing points in each active region, ultimately forming a "sensor deployment pattern."Based on the optimized results, "Anode Area Sensor Deployment Data" and "Cathode Area Sensor Deployment Data" are generated separately. The data includes: sensor number; three-dimensional spatial coordinates (based on the center of the tank); the electrode region (anode or cathode); the monitored active area number; and the expected response frequency and temperature measurement frequency parameters. The two types of data are integrated and output as complete "Electrode Area Sensor Deployment Data" for subsequent simulation verification and physical deployment.

[0116] Preferably, step S3, analyzing the flow path of the electrolyte channel and segmenting the flow path of the electrolyte channel, includes:

[0117] Extract the electrolyte flow channel geometry data from the electrolytic cell structure data;

[0118] The starting point of the electrolyte flow channel is identified based on the electrolyte flow channel geometry data, and an initial flow path reference line is constructed.

[0119] Flow simulation was performed on the flow path of the electrolyte channel to obtain the electrolyte velocity field and pressure distribution field;

[0120] Based on the electrolyte velocity field and pressure distribution field, the main flow path is extracted to generate the main flow path data.

[0121] The main flow path data is divided into equally spaced sections along its length, and key path nodes for changes in flow velocity are extracted.

[0122] Based on the flow path reference line, the near-entry segment path range of key path nodes is identified, and the near-entry segment flow channel is generated.

[0123] The remaining channels in the electrolyte flow path are selected based on the near-mouth flow channel and marked as the middle channel.

[0124] In this embodiment of the invention, surface element screening is performed on the closed fluid region of the model based on the three-dimensional geometric information of the structure already constructed in the digital model of the electrolyzer. Structural blocks with the material attribute label "electrolyte channel" are identified, and the set of surfaces with physical channel connections to the electrodes is extracted using a three-dimensional geometric Boolean operation method. Continuous path connection verification is performed on the three-dimensional surfaces of the channel space to confirm the channel's closure, and its vertex coordinate set and three-dimensional spatial volume constraint boundary are extracted. The final obtained channel geometric structure data includes: channel inlet surface, channel outlet surface, channel main cavity surface mesh data and its three-dimensional spatial dimensions, wherein the inlet and outlet surfaces are defined by criteria of an area greater than 100 square millimeters and an edge-to-tank wall fit greater than 95%. In the extracted electrolyte flow channel mesh model, the directionality of the normal vector of the cross-sectional surface element located at the lower end near the inlet pipe connection is determined. The set of surfaces whose normal vectors are less than 15 degrees from the Z-axis of the electrolyte tank bottom coordinate system, combined with the minimum spatial Z-coordinate value of the surface element center point, is used as the inlet reference center point. Starting from this point, a linear trajectory is projected upwards. An initial path segment is constructed in 3D space using a step-by-step nearest neighbor connectivity method (each step size is 2 mm) until it crosses a terminal surface element (with an angle less than 15 degrees to the Z-axis, corresponding to the maximum Z-axis coordinate) aligned with the liquid surface normal direction. This path is represented by a single continuous polyline in 3D space, serving as the initial flow path reference line. A computational domain mesh is established using the extracted 3D flow channel geometry data, employing tetrahedral elements with a side length limited to no more than 0.8 mm. The boundary layer region is configured with three layers of refinement (thicknesses of 0.1, 0.2, and 0.3 mm). For simulating boundary conditions, a steady flow input with a velocity of 0.1 m / s is applied at the inlet, the outlet is set to a zero-pressure atmospheric flow boundary, and the wall condition is set to a no-slip boundary. The electrolyte fluid properties are set as follows: dynamic viscosity 0.0012 Pa·s, density 1320 kg / m³. The finite volume method was used to numerically solve the flow field, with convergence criteria set as velocity residuals less than 1e-5 and pressure residuals less than 1e-4. After simulation, the three-dimensional velocity vector field and corresponding hydrostatic distribution field inside the entire electrolyte channel were obtained. The numerical format was a structured array, supporting path extraction and projection operations. In the velocity field data, the inlet point was selected to iteratively trace the main fluid flow path along the velocity direction. Specifically, starting from the inlet point, the path was extended along the velocity vector direction at the current position, with a step size of 0.5 mm, tracing the next position until the outlet position was reached. The tracing process required that the velocity magnitude at the current position was not less than 40% of the maximum velocity; otherwise, the current path segment was discarded. After tracing, the path data was stored in the form of a spatial polyline, and curvature analysis was performed on the entire path segment to ensure that the turning angle between segments did not exceed 25 degrees, thus guaranteeing the continuity of the main path.The final output main path line contains at least 150 nodes, with an average spacing of less than 1 mm between nodes, forming the main path data for subsequent segmentation analysis. The total length of the main path line data is statistically analyzed; if the calculated result is 153 mm, then each 10 mm segment is defined as a unit with 16 nodes (including the starting point). Within each unit, the velocity magnitude of the current node in the velocity field is compared with the velocity magnitudes of its two adjacent nodes. If a velocity change exceeding 25% occurs (e.g., the current point's velocity is 0.08 m / s, and the adjacent point's is 0.06 m / s or less), it is marked as a critical node for velocity change. Critical nodes must record spatial coordinates, local velocity values, curvature values, and pressure gradients for subsequent segmentation identification. Starting from the entry point, the path is advanced along the reference path, with the first 30 mm segment serving as the starting interval for determining the near-entry segment. Based on the location of critical path nodes, if at least two velocity abrupt change points exist within this range (the abrupt change threshold remains at 25%), then this 30 mm path segment is defined as the near-mouth flow channel, and the start and end node numbers, total length, and average pressure gradient value are recorded. If the number of velocity abrupt change points is insufficient, the range is expanded to 40 mm and the judgment is repeated until the condition of at least two abrupt change points is met. The final output near-mouth flow channel is labeled with path segment numbers, and its corresponding coordinate data is used for deployment analysis. The remaining main path segments, excluding the near-mouth segment, are numbered and marked, with the starting node of the path segment being the next node after the end point of the near-mouth segment, and the ending node being the end point of the main path. For this path segment, its spatial start and end coordinates, corresponding average velocity, pressure gradient, and wall distance parameters are recorded. All such path segments are collectively referred to as the mid-section flow channel and assigned a unified identification number for subsequent sensor placement point analysis. The length of the mid-section flow channel is typically over 100 mm, containing at least 10 velocity evaluation nodes to ensure sufficient data support for eddy current pulsation characteristic evaluation.

[0125] Preferably, step S3, which involves analyzing the eddy current recirculation characteristics of the near-mouth section flow channel and deploying the first flow channel sensor in the near-mouth section flow channel, includes:

[0126] The structural parameter data of the near-mouth section flow channel are divided into cross-sectional partitions to generate local flow region data of the near-mouth section.

[0127] Fluid particle trajectory simulation is performed based on local flow region data near the inlet section to generate eddy flow return path data.

[0128] The disturbance frequency of the return flow core area is analyzed using eddy current return flow path data to generate return flow disturbance characteristic data;

[0129] Based on the backflow disturbance characteristic data, the first type of sensor deployment sensitive sections are identified, and sensor sensitive deployment site data are generated;

[0130] Structural interference assessment is performed on sensor sensitive deployment site data to eliminate areas with excessive interference and generate candidate sensor deployment site data.

[0131] Based on the sensor candidate deployment point data and the flow direction alignment principle, point calibration is performed to generate the first flow channel sensor deployment data.

[0132] In this embodiment of the invention, a spatial region extending 30 mm downstream from the inlet starting point is selected from the obtained three-dimensional structural model of the near-mouth section flow channel. This section of the model is vertically sectioned along the axial direction of the flow channel, with a section spacing of 5 mm per layer, resulting in a total of 6 sections. Each section is further divided into 9 sector-shaped sub-regions, with each sector spanning 40 degrees and numbered A1 to A9. For each sector, structural parameters such as area, average normal angle, shortest distance to the wall, and local geometric curvature are recorded, forming a "near-mouth section local flow region dataset," containing a total of 6 × 9 = 54 independent sub-region structural units. The Lagrange particle tracking method is used to simulate the fluid particle trajectories of these 54 local regions. 300 tracer particles are released into each sub-region, with a particle diameter of 10 micrometers, a density equal to that of the electrolyte (1320 kg / m³), an initial velocity equal to the average flow velocity of that region, and a simulation time range of 2 seconds. Based on the Dynamic Mesh and Discrete PhaseModel modules in Fluent software, three-dimensional path data of particles were recorded during the simulation period. The path data was indexed by time, recording the XYZ coordinate changes of the particles at 0.01-second intervals, and the output format was a trajectory vector sequence. If a particle's path moved more than 5 mm backward in the original region, it was considered a recirculation path, and its spatial path was extracted to form an eddy current recirculation path dataset, which ultimately included attributes such as the recirculation path label, start and end coordinates, maximum recirculation radius, and residence time. Statistical analysis was performed on particle paths exhibiting time-periodic perturbations in the recirculation path data. The spatial position changes of all recirculation paths in the X, Y, and Z directions over time were extracted, and the main perturbation frequencies were extracted using the Fast Fourier Transform (FFT) method. If a dominant frequency peak existed between 1 and 3 Hz, it indicated that there was significant perturbation recirculation behavior in the region to which the path belonged. For each sub-region, the number of particle paths and their peak frequency amplitude that meet the criteria are statistically analyzed. The average perturbation frequency, maximum perturbation amplitude, and average particle residence time of that sub-region are calculated and combined to form "backflow perturbation characteristic data." Perturbation frequency is expressed in Hertz, residence time in seconds, and amplitude in millimeters. High-perturbation sub-regions meeting the requirements are selected based on an average perturbation frequency greater than 1.2 Hertz, an average particle residence time greater than 0.5 seconds, and a maximum perturbation amplitude greater than 3 millimeters. The spatial locations of these sub-regions are used as initial sensitive deployment sites, recording their corresponding local section number, sector number, spatial center coordinates, and corresponding perturbation parameters. A preliminary deployment point is generated at the center point of each identified sub-region, forming a "sensor sensitive deployment site dataset," with a total of no more than 12 deployment points. For each sensor sensitive deployment point, the shortest distance between it and the inner wall of the flow channel is analyzed. If the shortest distance is less than 5 millimeters, it is determined to be physical interference with the equipment.Further assess the placement risk of the deployment point in areas with drastic changes in flow channel curvature. Areas with curvature greater than 0.4 (in 1 / mm) are also marked as high-risk. After eliminating these two types of high-interference deployment points, the remaining deployment points are retained, forming the "sensor candidate deployment point data." This step ensures that sensor placement does not interfere with the normal flow channel structure and flow path. For each retained candidate sensor deployment point, its orientation is adjusted according to the mainstream direction vector at the center of its cross-section. The sensor probe orientation must have an angle of less than 15 degrees with the local velocity direction; if this requirement is not met, the deployment point is adjusted to the center of the nearest adjacent sector within the same cross-section. After calibration, record the spatial coordinates, placement angle (with the Z-axis as a reference), cross-section number, and final probe orientation vector for each point. Finally, the data is compiled into the "First Flow Channel Sensor Deployment Dataset," used for sensor mapping in the digital electrolyzer model. Each data point has a clear location and orientation, and the number is controlled between 4 and 8.

[0133] Of particular importance, fluid particle trajectory simulation based on local flow region data near the inlet also includes:

[0134] Extract the boundary layer velocity gradient from the local flow region data near the mouth section;

[0135] Fluid particle release initialization is performed based on the boundary layer velocity gradient to generate initial particle release position data.

[0136] Calculate the time step of the initial position data of the particle release to generate the instantaneous position trajectory data of the particle;

[0137] Vortex structure capture is performed based on particle instantaneous position trajectory data to generate local particle vortex aggregation feature data;

[0138] Trajectory curves are fitted to local particle vortex aggregation characteristic data to generate vortex path fitting data for the near-mouth section.

[0139] Based on the fitting data of the vortex path in the near-mouth section, the recirculation direction and path connectivity are analyzed, and finally the vortex recirculation path data is generated.

[0140] In this embodiment of the invention, the boundary layer region within 10 mm around the channel wall is first selected as the velocity gradient analysis range from the acquired local flow region data near the inlet section. This region is then meshed using the finite volume method, and the change in velocity vector along the normal direction is extracted using a fifth-order weighted intrinsically non-oscillating (WENO) scheme. The gradient value of the velocity along the wall normal direction in each mesh cell is calculated and output as a two-dimensional matrix of boundary layer velocity gradient dataset. Based on the obtained boundary layer velocity gradient data, regions with gradient amplitudes between 500 s⁻¹ and 2000 s⁻¹ are selected as high shear stress release regions, and the center of these regions is used as the initial point for particle release. The number of released particles is no less than 300, uniformly distributed along the channel cross-section. The initial particle release height is limited to 2 mm to 5 mm from the bottom of the channel to ensure the simulation of a real boundary layer disturbance source. The released particles are time-stepped, and the position change of the particles over time is calculated with a time step of 2 milliseconds. Particle motion is guided by a velocity field, obtained from previous flow simulations. The position of each particle is updated at each time step, and this update forms a point cloud dataset of instantaneous particle trajectories. The total simulation time is 30 seconds. Based on the instantaneous particle trajectory point cloud data, the Q-criterion method is used to identify regions in the trajectory containing both positive rotation tensors and negative strain tensors, thereby capturing the vortex core region. The Q-value threshold is set to 1×10⁻⁶. 5 s -2 The system extracts highly continuous and entangled particle vortex aggregation regions as local particle vortex aggregation features. All identified feature regions are output in three-dimensional coordinate form. For the identified vortex aggregation feature data, a cubic B-spline curve fitting method is used to fit the particle motion trajectory. Each curve consists of 10 control points, with the fitting accuracy controlled to a maximum deviation of less than 0.2 mm, generating a set of continuous smooth trajectory curves representing the particle vortex structure, constituting the near-mouth vortex path fitting data. Spatial path analysis is performed on the fitted vortex path, calculating the angle between the path tangent vector direction and the mainstream direction; when the angle is greater than 120°, it is marked as a typical return path, and the topological connectivity between these paths is analyzed. Using Euclidean distance-based connectivity analysis, trajectories with a connection spacing of less than 5 mm and convergent path directions (angle less than 15°) are marked as a continuous return unit. Finally, the vortex return path data is integrated, including the path coordinate sequence, direction identifier, and topological connectivity matrix.

[0141] Preferably, step S3, analyzing the vortex pulsation characteristics of the mid-section flow channel and deploying the second flow channel sensor in the mid-section flow channel, includes:

[0142] The geometric parameters of the middle channel are mapped to the channel region to generate spatial grid data of the middle channel.

[0143] Based on the spatial grid data of the mid-section flow channel, velocity and pressure field variation data are extracted to generate flow state data of the mid-section flow channel.

[0144] The instantaneous velocity disturbance is calculated based on the flow state data of the middle channel, and vortex pulsation spectrum data is generated.

[0145] Extract the dominant frequency of the vortex pulsation spectrum data, and use the dominant frequency to identify the high response zone of the middle section of the flow channel to generate sensor pulsation response region data;

[0146] Spatial distribution equalization optimization is performed on the sensor pulse response region data to generate a second type of deployable point set data for sensors.

[0147] Based on the deployable point set data of the second type of sensor, deployment sites are selected according to the principle of minimum flow interference to generate deployment data for the second flow channel sensor.

[0148] In this embodiment of the invention, a mid-section of the flow channel, ranging from 30 mm to 90 mm downstream of the near-end of the main electrolyte flow path, is selected. This section of the flow channel is divided into regions using a structured mesh generation method, with hexahedral elements used for volume discretization. The element side length is set to 2 mm, generating approximately 50,000 to 60,000 mesh elements within the three-dimensional structure. During the mesh generation process, CFD preprocessing software (such as ICEM CFD) is used to perform boundary fitting, ensuring that each element fits the inner wall structure of the flow channel. Furthermore, the element size in high-curvature regions is appropriately reduced to a side length of 1 mm to improve the resolution of these regions. This spatial mesh data provides the discretization basis for subsequent flow simulations. The mid-section flow channel mesh data is imported into computational fluid dynamics simulation software (such as ANSYS Fluent). The electrolyte flow velocity inlet boundary condition is set to 0.6 m / s, the outlet pressure is set to atmospheric pressure (101325 Pa), the electrolyte density is set to 1320 kg / m³, and the viscosity is set to 1.1 × 10⁻³ Pa·s. The SST k-ω turbulence model was used for both steady and unsteady solutions. 100 time steps were recorded per second, with each step size of 0.01 seconds, for a total simulation duration of 3 seconds. The three-dimensional velocity vector and pressure scalar value of each grid node in the mid-section flow channel at each time step were output, forming a complete "flow state data" matrix containing dimensions such as time, position, velocity, and pressure. For the velocity-time series at the center point of each grid cell, velocity perturbation analysis was performed. First, the difference between the instantaneous velocity and the average velocity over 3 seconds was calculated. Then, using this difference sequence as input, a Fast Fourier Transform (FFT) was performed to obtain its spectral distribution. The highest frequency value of the amplitude in the spectral data was extracted, which is the dominant frequency of the velocity perturbation at that point. Simultaneously, the energy spectral density within the frequency range was recorded. The spectral data of all grid nodes in the entire mid-section flow channel were compiled into "vortex pulsation spectral data," with each node containing information such as dominant frequency (in Hertz), maximum perturbation amplitude (mm / s), and spectral energy density value (normalized units). The aforementioned spectral data was filtered by dominant frequency. A threshold was set for nodes with a dominant frequency higher than 2 Hz, a perturbation amplitude exceeding 30 mm / s, and energy density ranking in the top 10% as "high-response points." Spatial region clustering was performed on these high-response points, with a spatial aggregation distance of 5 mm and a minimum number of points of 20. Each cluster formed an independent high-pulsation response region. Its geometric center, bounding box size, and perturbation characteristic values ​​were extracted and combined to generate "sensor pulsation response region data." Within each high-response region, a cross-section parallel to the perturbation direction was selected based on its maximum perturbation direction. Equidistant point sampling was performed on the cross-section region, with 4 to 6 points sampled per region, ensuring a minimum distance of 6 mm between points. Then, a three-dimensional global spatial analysis was performed on the point set across all response regions. The distance between any two points was detected; if it was less than 5 mm, points with lower perturbation amplitudes were discarded. This process achieved both minimum distance constraints between points and balanced response distribution.The final result is a set of balanced points covering the entire high-response area, forming the "second-type sensor deployment point set data," with approximately 20 points. For each deployment point, the local velocity vector direction of its grid cell is queried, and the angle between it and the mainstream electrolyte direction (inlet velocity direction) is calculated. If the angle exceeds 25 degrees, the sensor placement at that point is considered to cause disturbance to the local flow and is therefore discarded. Further structural evaluation of the area where the point is located is performed; areas with a distance of less than 3 mm from the wall or a wall curvature radius of less than 4 mm are also marked as high-interference areas. Points that meet the flow alignment principle, are moderately far from the wall, and have a smooth structure are retained as the final deployment points. The three-dimensional spatial coordinates, reference flow vector, disturbance amplitude, and dominant frequency value are output as additional information to form the "second-channel sensor deployment data," used for precise point placement in a digital electrolyzer environment; the number of deployments is generally controlled between 6 and 10.

[0149] Of particular importance, optimizing the spatial distribution of sensor pulsation response region data also includes:

[0150] The sensor pulsation response region data is processed to extract the region boundary, generating pulsation response region boundary data;

[0151] Internal meshing is performed based on the boundary data of the pulsating response region to generate the initial response region mesh data;

[0152] Statistical analysis of local pulsation intensity is performed on the grid data of the initial response region to generate grid pulsation intensity distribution data;

[0153] Calculate the spatial uniformity of grid pulsation intensity distribution data;

[0154] Based on spatial balance, low-balance regions are identified in the sensor pulsation response region data, and unbalanced response region data is generated.

[0155] Perform center point filtering on the data of the unbalanced response area to generate second-category candidate deployment point data for sensors;

[0156] Based on the candidate deployment point data of the second type of sensor, geometric spacing is optimized and filtered to finally generate the deployment point set data of the second type of sensor.

[0157] In this embodiment of the invention, the sensor pulsation response region data is segmented into two-dimensional regions. The Sobel edge detection operator is used to calculate the pixel gradient in the pulsation response data map to extract edge positions. When the gradient amplitude exceeds a set threshold of 10, it is marked as a boundary point. Subsequently, connected region identification is performed on the boundary points, and the output is a sequence of polygon boundary coordinates, which constitutes the boundary data of the pulsation response region. Based on the extracted boundary polygon coordinate data, the Delaunay triangulation method is used to divide the region within the boundary into non-overlapping grids. The maximum side length of each triangular unit is controlled to not exceed 20 mm. The output of the triangulation result is the initial response region grid data, with a total number of grids of not less than 150, and all grids contain center coordinates and vertex coordinates. The calculated vortex pulsation spectrum data of the middle channel is mapped to the initial grid. According to the spectrum amplitude range covered by the center point of each grid, the average pulsation intensity value is calculated. The statistical window size is set to 5 mm × 5 mm, the spectrum amplitude unit is Hertz, and each grid is assigned an average spectrum amplitude, which constitutes a grid pulsation intensity distribution dataset. Discrete metric calculation is performed on the entire grid pulsation intensity distribution dataset. The average spectral amplitude of all grids is statistically analyzed using the standard deviation method. First, the average intensity of all grids is calculated. Then, the sum of the squares of the differences between each grid value and the average is averaged, and the square root of the result is taken as the spatial uniformity of the pulsation in the entire region. A uniformity higher than 10 Hz is considered to indicate a significant imbalance. Based on the uniformity results, areas with local deviations greater than 20% of the average are selected from the grid pulsation intensity distribution map. Using an 8-adjacency principle-based region growing model, grids with consecutive deviations greater than the threshold are clustered into imbalanced response areas. Each imbalanced area must contain at least 5 consecutive grids, and their corresponding coordinate ranges and numbers are output, forming the imbalanced response area data. A center point selection operation is performed on the grids within the identified imbalanced response areas, selecting the geometric center coordinates of each area as candidate deployment points for the second type of sensors. If the interior of a region is polygonal, the center coordinates of the smallest circumcircle of the polygon are used as the geometric center point, and all center points constitute the candidate deployment point data for the second type of sensors. Spatial geometric spacing analysis is performed on all candidate deployment points. The Euclidean distance between any two points is calculated. For point pairs with a distance of less than 25 mm, the point with the higher pulsation intensity value is retained, and the other is discarded. This operation is repeated until the minimum spacing between all candidate points is greater than or equal to 25 mm. Finally, the second type of sensor deployment point set data is generated, which serves as the basis for subsequent deployment map layout.

[0158] As an example of the present invention, reference is made to... Figure 3 As shown, step S4 in this example includes:

[0159] Step S41: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the electrolytic cell digital model for spatial arrangement mapping, and generate three-dimensional deployment candidate point cloud data;

[0160] Step S42: Initialize the thermal field simulation of the 3D deployment candidate point cloud data to generate initial temperature distribution estimation data; perform virtual temperature sensor deployment simulation based on the initial temperature distribution estimation data to generate virtual temperature sensor deployment scheme data;

[0161] Step S43: Perform temperature response simulation tests on the electrolytic cell based on the virtual temperature sensor deployment scheme data to generate simulation monitoring response data; conduct a comparative analysis of the virtual and real arrangement differences on the simulation monitoring response data to generate arrangement deviation evaluation results;

[0162] Step S44: Based on the layout deviation assessment results, perform virtual deployment adjustment and optimization to generate optimized digital temperature sensor layout data, and execute the temperature sensor layout optimization operation of the digital electrolytic cell.

[0163] In this embodiment of the invention, the three-dimensional coordinate information, flow direction vector, and electrode surface normal information of the anode and cathode deployment points in the electrode area sensor deployment data are imported into the electrolytic cell digital model (constructed using STEP or Parasolid format). Next, the sensor deployment point data of the first and second flow channels are imported, including their spatial location, corresponding flow disturbance characteristic values, and local velocity vectors. Under a unified coordinate system, the three types of points are globally reprojected, and then integrated into the three-dimensional structure of the electrolytic cell using a CAD modeling platform (such as Siemens NX or SolidWorks) to establish a spatial mapping correspondence. Each point is marked with a sphere, with a uniform radius of 2 mm, and cross-projection between points is not allowed. A unified "deployment candidate point cloud model" is generated through Boolean operations, and then the point cloud is exported as a PLY or OBJ format to form three-dimensional deployment candidate point cloud data, containing approximately 40 to 60 points. The deployment point cloud model is imported into CFD heat conduction simulation software (such as ANSYS Icepak or COMSOL Multiphysics), and the electrolytic cell digital model is set as the simulation object. The boundary conditions are set as follows: the heat flux input density on the anode surface is set to 12,000 watts per square meter, the heat input density on the cathode surface is set to 8,500 watts per square meter, the initial temperature of the electrolyte inside the flow channel is set to 45 degrees Celsius, the electrolyte heat transfer coefficient is 0.6 W / m·K, and the tank wall temperature is set to a constant 40 degrees Celsius. The total simulation time is set to 100 seconds, with a time step of 1 second. After the simulation is executed, the temperature response curve of each candidate deployment point under the thermal field is obtained, and the temperature value and rate of change of each point at 50 seconds are extracted to construct the initial temperature distribution estimation data. Based on this data, through the multi-point distribution maximum response selection mechanism, points with larger temperature differences and faster temperature responses are selected first, redundant points with temperature differences below 0.5 degrees Celsius are eliminated, and no more than 30 points are retained as the first batch of proposed deployment points to form the virtual temperature sensor deployment scheme data. The virtual temperature sensor deployment scheme is imported into the simulation platform, and the selected points are re-set as "sensor nodes". Next, a heat conduction simulation is performed. Under the same initial conditions as in step S42, the temperature change curve of each sensor node over time is recorded, with the simulation sampling frequency set to twice per second. Simultaneously, reference measurement point data is selected from the actual deployment data (temperature compensation can be performed using high-precision thermal field partitioning interpolation technology), and the simulation results are compared with the reference real temperature data to assess the error. For each sensor deployment point, the deviation between its virtual response and the reference temperature is calculated, with the deviation standard set to no more than 1 degree Celsius; those exceeding this are marked as outliers. The error data from all sensor points are used to form a deviation matrix, and thermal response pattern clustering is performed to identify the deviation. The deployment deviation assessment results are output, including the number of outliers, maximum error value, mean error, and error distribution trend chart. Based on the points with excessively high errors marked in the deployment deviation assessment results, point adjustment operations are performed.If the deviation of a deployment point exceeds 2 degrees Celsius, the area where it is located is determined to be a heat-sensitive area, and the deployment density needs to be adjusted. Two to three virtual detection points should be redeployed within a 5 mm radius around the affected point. For areas with smaller deviations and redundant points, retain the point with the best response and discard the rest. During the point adjustment process, the redundancy index should also be considered to ensure that there are staggered coverage areas (at least two deployment points with different thermal characteristics every 10 mm). After adjustment, an updated deployment point set is generated, and thermal simulation is performed again for verification. If the error of all points is less than 1 degree Celsius, the deployment is considered effective. Finally, optimized digital temperature sensor deployment data is generated, including the three-dimensional coordinates of all points, local temperature response curves, deployment area numbers, and sensor types (labeled as anode area, cathode area, near-mouth section, and middle section). This data serves as the final deployment scheme output and is used for actual deployment planning.

[0164] Preferably, step S43, which involves performing a temperature response simulation test on the electrolyzer based on the virtual temperature sensor deployment scheme data, includes:

[0165] Based on the data from the virtual temperature sensor deployment scheme, simulation monitoring nodes are set up within a range of 20mm to 50mm directly above the anode plate of the electrolytic cell, 30mm to 60mm on both sides of the cathode plate, and 10mm to 30mm on both sides of the central axis of the electrolyte flow channel. The total number of simulation nodes is no less than 30, and the spacing is no more than 100mm, forming a complete temperature field monitoring network.

[0166] The initial simulation conditions were set as follows: ambient temperature was set to 25℃±2℃, initial electrolytic liquid temperature was set to 50℃~70℃, and boundary heat transfer coefficient was set to 30~80W / (m²). 2 •K), and gradually sample the temperature sensing response data over a simulation period of 120 minutes;

[0167] During the simulation, constant current operating conditions were simulated, and the dynamic temperature response data of the anode region, cathode region, and flow channel region were recorded. When the temperature rise rate exceeded 2.5℃ / min or the local temperature exceeded 90℃, it was recorded as a high-temperature risk point. Specifically, the constant current operating conditions were set with a current density range of 150~300A / m. 2 ;

[0168] The dynamic temperature response change data of the anode region, cathode region and flow channel region are integrated into the simulation monitoring response data.

[0169] In this embodiment of the invention, simulation monitoring nodes are uniformly arranged along the parallel direction of the anode plate surface within a height range of 20 mm to 50 mm directly above the anode plate of the electrolytic cell. The node spacing is no greater than 100 mm to ensure sufficient node density to capture the temperature gradient in the anode area. The number of nodes is controlled to be around 12, and the node coordinates accurately map the geometric contour of the anode plate surface. Monitoring nodes are set within a range of 30 mm to 60 mm perpendicular to the cathode plate plane on both sides of the cathode plate. Considering the complex structure of the cathode plate, the number of nodes is set to about 10, with a spacing of no more than 100 mm, to uniformly cover the heat-sensitive area of ​​the cathode region. Monitoring nodes are arranged within a range of 10 mm to 30 mm on both sides of the central axis of the electrolyte flow channel to cover the temperature changes of the main electrolyte flow path. The number of nodes is about 8, ensuring a reasonable node distribution along the flow direction and laterally. The total number of nodes arranged in the above three areas is no less than 30, forming a complete and spatially balanced three-dimensional temperature monitoring network to meet the simulation accuracy and coverage requirements. The ambient temperature is set to 25 degrees Celsius, allowing a fluctuation range of ±2℃ to simulate the natural temperature changes in the actual production workshop environment. The initial temperature of the electrolyte in the electrolytic cell is set within the range of 50℃ to 70℃, reflecting the dynamic range of the initial temperature state under actual electrolysis conditions. The boundary heat transfer coefficient is taken as 30 to 80 W / (m²·Kelvin). 2 The simulation, taking into account the heat transfer characteristics between the tank surface and the ambient air, as well as the differences in heat dissipation at different locations, was conducted. The total simulation duration was set to 120 minutes, employing a stepwise time sampling method, collecting temperature data every minute to achieve real-time tracking of the dynamic temperature field response. Constant current operation was simulated during the simulation, with the current density set to 150 to 300 amperes per square meter (A / m²). 2 Within the specified range, this conforms to the actual operating conditions of industrial electrolytic cells. Temperature changes over time are continuously recorded in the anode, cathode, and flow channel regions, with a focus on the rate of temperature rise and local temperature peaks. Monitoring nodes with a temperature rise rate exceeding 2.5 degrees Celsius per minute or a local temperature exceeding 90 degrees Celsius are defined as high-temperature risk points and highlighted as potential safety hazards. The dynamic temperature response data of the anode, cathode, and flow channel regions are integrated to form a spatiotemporally continuous temperature response matrix. The output data includes node numbers, spatial coordinates, sampling time points, and corresponding temperature values, stored in a standard format (such as CSV or HDF5) suitable for subsequent error analysis and layout optimization.

[0170] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0171] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for arranging temperature sensors in a digital electrolytic cell, characterized in that, Includes the following steps: Step S1: Obtain electrolytic cell structure data; analyze the topology of the electrolytic cell structure data, and perform three-dimensional digitization of the electrolytic cell structure data based on the topology to construct a digital model of the electrolytic cell; Step S2: Extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-deposition phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region; Step S3: Analyze the flow path of the electrolyte channel and segment the flow path to generate a near-mouth channel and a middle channel; analyze the eddy current backflow characteristics of the near-mouth channel and deploy the first channel sensor in the near-mouth channel to obtain the first channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle channel and deploy the second channel sensor in the middle channel to obtain the second channel sensor deployment data. Step S4: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the digital model of the electrolytic cell to perform virtual temperature sensor deployment and monitoring and verification, so as to perform the temperature sensor layout optimization operation of the digital electrolytic cell.

2. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain electrolytic cell structure data; Step S12: Identify the structural units of the electrolytic cell structure data, extract the boundaries of key components such as cathode, anode, cell shell, and diaphragm, and generate structural unit distribution data; construct a structural connection diagram based on the structural unit distribution data, perform logical mapping on the connection methods between each structural unit, and generate topological structure description data. Step S13: Use the topological description data to perform three-dimensional spatial coordinate mapping, convert the position and relative relationship of the structural unit into three-dimensional configuration parameters, and generate spatial configuration data; Step S14: Perform three-dimensional modeling on the spatial configuration data to generate three-dimensional digital configuration data of the electrolytic cell.

3. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S2 extracts the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer, including: Identify the boundary features of the anode components in the digital model of the electrolytic cell, and extract the structural data of the anode area and mark it as the anode area of ​​the electrolytic cell when the following conditions are met: the length of the anode plate is 800mm to 1500mm, the width is 400mm to 600mm, the thickness is not less than 10mm, the spacing between the anode plates is maintained in the range of 20mm to 50mm, and the anode connection structure is a continuous linear structure or an array arrangement. When identifying the boundary morphology of the cathode region in the digital model of the electrolytic cell, and meeting the following geometric and material parameters, the cathode region structural data is extracted and marked as the cathode region of the electrolytic cell: cathode plate length is 850mm~1550mm, width is 450mm~650mm, plate thickness is not less than 12mm, cathode plate spacing is controlled between 25mm and 60mm, and the material property is conductivity greater than 4.5×10 7 Metallic materials with S / m; Based on the spatial path identification results between the anode and cathode regions of the electrolytic cell, electrolyte flow channel structure data was extracted and labeled as electrolyte flow channels: the width of the main electrolyte channel is 100mm to 300mm, the width of the branch channels is not less than 30mm, the overall length of the flow channel is between 1.5m and 3.5m, and the rate of change of curvature is not greater than 0.2m. -1 Furthermore, the flow channel cross-section is a regular rectangle or ellipse.

4. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S2 involves analyzing the micro-deposition phase interface regions in the anode and cathode regions of the electrolytic cell to determine the sensor deployment patterns in these regions. Extract microstructure images of the anode and cathode regions of the electrolytic cell; Interface feature enhancement is performed on microstructure image data to generate interface-enhanced images; Edge recognition and phase interface segmentation are performed on the interface enhancement image to obtain micro-extruded phase interface contour data; Calculate the morphological change gradient of the micro-precipitated phase interface contour data and generate an interface morphological gradient distribution map. Based on the interface morphology gradient distribution map, the concentrated and sparse regions of precipitates are identified, and a precipitate distribution density map is generated. Sensitive regions are extracted based on the precipitate distribution density map to identify active micro-precipitate regions in the anode and cathode regions; Based on the active region of the micro-precipitated phase, sensor sensing optimization analysis is performed to determine the sensor deployment mode in the anode and cathode regions; Generate sensor deployment data for the anode area and sensor deployment data for the cathode area based on the sensor deployment mode; By integrating the sensor deployment data of the anode region and the sensor deployment data of the cathode region, the sensor deployment data of the electrode region is obtained.

5. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S3, which involves analyzing the flow path of the electrolyte channel and segmenting the flow path, includes: Extract the electrolyte flow channel geometry data from the electrolytic cell structure data; The starting point of the electrolyte flow channel is identified based on the electrolyte flow channel geometry data, and an initial flow path reference line is constructed. Flow simulation was performed on the flow path of the electrolyte channel to obtain the electrolyte velocity field and pressure distribution field; Based on the electrolyte velocity field and pressure distribution field, the main flow path is extracted to generate the main flow path data. The main flow path data is divided into equally spaced sections along its length, and key path nodes for changes in flow velocity are extracted. Based on the flow path reference line, the near-entry segment path range of key path nodes is identified, and the near-entry segment flow channel is generated. The remaining channels in the electrolyte flow path are selected based on the near-mouth flow channel and marked as the middle channel.

6. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S3, which involves analyzing the vortex backflow characteristics of the near-mouth section flow channel and deploying the first flow channel sensor in the near-mouth section flow channel, includes: The structural parameter data of the near-mouth section flow channel are divided into cross-sectional partitions to generate local flow region data of the near-mouth section. Fluid particle trajectory simulation is performed based on local flow region data near the inlet section to generate eddy flow return path data. The disturbance frequency of the return flow core area is analyzed using eddy current return flow path data to generate return flow disturbance characteristic data; Based on the backflow disturbance characteristic data, the first type of sensor deployment sensitive sections are identified, and sensor sensitive deployment site data are generated; Structural interference assessment is performed on sensor sensitive deployment site data to eliminate areas with excessive interference and generate candidate sensor deployment site data. Based on the sensor candidate deployment point data and the flow direction alignment principle, point calibration is performed to generate the first flow channel sensor deployment data.

7. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S3, which involves analyzing the vortex pulsation characteristics of the mid-section flow channel and deploying a second flow channel sensor in the mid-section flow channel, includes: The geometric parameters of the middle channel are mapped to the channel region to generate spatial grid data of the middle channel. Based on the spatial grid data of the mid-section flow channel, velocity and pressure field variation data are extracted to generate flow state data of the mid-section flow channel. The instantaneous velocity disturbance is calculated based on the flow state data of the middle channel, and vortex pulsation spectrum data is generated. Extract the dominant frequency of the vortex pulsation spectrum data, and use the dominant frequency to identify the high response zone of the middle section of the flow channel to generate sensor pulsation response region data; Spatial distribution equalization optimization is performed on the sensor pulse response region data to generate a second type of deployable point set data for sensors. Based on the deployable point set data of the second type of sensor, deployment sites are selected according to the principle of minimum flow interference to generate deployment data for the second flow channel sensor.

8. The method for arranging temperature sensors in a digital electrolytic cell according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Input the electrode area sensor deployment data, the first flow channel sensor deployment data, and the second flow channel sensor deployment data into the electrolytic cell digital model for spatial arrangement mapping, and generate three-dimensional deployment candidate point cloud data; Step S42: Initialize the thermal field simulation of the 3D deployment candidate point cloud data to generate initial temperature distribution estimation data; perform virtual temperature sensor deployment simulation based on the initial temperature distribution estimation data to generate virtual temperature sensor deployment scheme data; Step S43: Perform temperature response simulation tests on the electrolytic cell based on the virtual temperature sensor deployment scheme data to generate simulation monitoring response data; conduct a comparative analysis of the virtual and real arrangement differences on the simulation monitoring response data to generate arrangement deviation evaluation results; Step S44: Based on the layout deviation assessment results, perform virtual deployment adjustment and optimization to generate optimized digital temperature sensor layout data, and execute the temperature sensor layout optimization operation of the digital electrolytic cell.

9. The method for arranging temperature sensors in a digital electrolytic cell according to claim 8, characterized in that, Step S43, which involves performing a temperature response simulation test on the electrolyzer based on the data from the virtual temperature sensor deployment scheme, includes: Based on the data from the virtual temperature sensor deployment scheme, simulation monitoring nodes are set up within a range of 20mm to 50mm directly above the anode plate of the electrolytic cell, 30mm to 60mm on both sides of the cathode plate, and 10mm to 30mm on both sides of the central axis of the electrolyte flow channel. The total number of simulation nodes is no less than 30, and the spacing is no more than 100mm, forming a complete temperature field monitoring network. The initial simulation conditions were set as follows: ambient temperature was set to 25℃±2℃, initial electrolytic liquid temperature was set to 50℃~70℃, and boundary heat transfer coefficient was set to 30~80W / (m²). 2 •K), and gradually sample the temperature sensing response data over a simulation period of 120 minutes; During the simulation, constant current operating conditions were simulated, and the dynamic temperature response data of the anode region, cathode region, and flow channel region were recorded. When the temperature rise rate exceeded 2.5℃ / min or the local temperature exceeded 90℃, it was recorded as a high-temperature risk point. Specifically, the constant current operating conditions were set with a current density range of 150~300A / m. 2 ; The dynamic temperature response change data of the anode region, cathode region and flow channel region are integrated into the simulation monitoring response data.

10. A temperature sensor arrangement system for a digital electrolytic cell, characterized in that, For performing the temperature sensor arrangement method of the digital electrolytic cell as described in claim 1, the temperature sensor arrangement system of the digital electrolytic cell includes: The digitization module is used to acquire electrolytic cell structural data; analyze the topological structure of the electrolytic cell structural data; and perform three-dimensional digitization of the electrolytic cell structural data based on the topological structure to construct a digital model of the electrolytic cell. The polarization region deployment module is used to extract the anode region, cathode region, and electrolyte flow channel of the electrolyzer from the digital model of the electrolyzer; analyze the micro-precipitated phase interface region of the anode region and cathode region of the electrolyzer to determine the sensor deployment mode of the anode region and cathode region of the electrolyzer, thereby obtaining the sensor deployment data of the electrode region; The flow channel deployment module is used to analyze the flow path of the electrolyte flow channel and segment the flow path to generate a near-mouth flow channel and a middle flow channel; analyze the eddy current backflow characteristics of the near-mouth flow channel and deploy the first flow channel sensor in the near-mouth flow channel to obtain the first flow channel sensor deployment data; analyze the eddy current pulsation characteristics of the middle flow channel and deploy the second flow channel sensor in the middle flow channel to obtain the second flow channel sensor deployment data. The monitoring and verification module is used to input the sensor deployment data of the electrode area, the sensor deployment data of the first flow channel, and the sensor deployment data of the second flow channel into the digital model of the electrolyzer to perform virtual temperature sensor deployment and monitoring and verification, so as to perform temperature sensor layout optimization work in the digital electrolyzer.

Images