A method and system for intelligent monitoring of chloride ion content based on multi-source data correction

Abstract

This invention discloses an intelligent monitoring method and system for chloride ion content based on multi-source data correction, belonging to the field of intelligent chloride ion content monitoring technology. The method includes: acquiring chloride ion sensing data and electrolysis process parameters during copper foil electrolysis production to construct a basic monitoring data set; identifying stable operating intervals based on the basic monitoring data set and generating a micro-perturbation control sequence to be applied to the electrolysis process to create a perturbation state; extracting the time interval and preceding / following neighborhoods of the perturbation effect to construct a perturbation response data sequence; segmenting the perturbation response data sequence to extract the response change amplitude, change rate, and recovery characteristics to form a perturbation response feature set; performing consistency analysis based on the response characteristics to identify and correct abnormal response intervals, obtaining a correction sequence; performing continuity screening and fluctuation convergence processing on the correction sequence to output the target chloride ion content value. This invention achieves dynamic and accurate monitoring of chloride ion content.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for chloride ion content, specifically to an intelligent monitoring method and system for chloride ion content based on multi-source data correction. Background Technology

[0002] In continuous industrial production processes such as copper foil electrolysis, chloride ions are a key component of the electrolyte system, and their concentration changes directly affect the stability of the electrolytic reaction and the consistency of product quality. With the development of industrial automation and online monitoring technologies, the acquisition of chloride ion data in the production process has gradually shifted from offline detection to online sensor monitoring, forming a multi-source data system centered on sensor detection data and process parameter data. Simultaneously, the electrolysis process is characterized by strong coupling, multi-parameter synergistic changes, and continuous dynamic operation, resulting in data exhibiting high-frequency sampling, strong time-series dependence, and complex fluctuation characteristics, providing a data foundation for real-time analysis and process control of chloride ion content.

[0003] However, existing chloride ion content monitoring methods mostly rely on static data processing or simple filtering methods, lacking a comprehensive processing mechanism for stable operating condition identification, disturbance response characteristic modeling, and multi-source data consistency correction. This makes it difficult to effectively distinguish between true concentration changes and abnormal sampling errors when there are process micro-disturbances or fluctuations in operating status. Furthermore, the ability to identify and correct abnormal response ranges is insufficient in complex electrolysis environments, resulting in problems with insufficient monitoring accuracy and dynamic adaptability. Therefore, it is necessary to provide an intelligent chloride ion content monitoring method and system based on multi-source data correction to solve the above-mentioned problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides an intelligent monitoring method and system for chloride ion content based on multi-source data correction. This technical solution solves the problem mentioned in the background that existing big data storage servers have different system or hardware performance, and the resource utilization rate of the same system or hardware is different. When the automated operation and maintenance system configures and manages them, configuration failures or mismatches may occur, resulting in reduced operation and maintenance efficiency of the automated operation and maintenance system or waste of resources.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A smart monitoring method for chloride ion content based on multi-source data correction includes:

[0007] Acquire chloride ion sensing data during the copper foil electrolysis production process, and acquire electrolysis process parameter data within the corresponding time window to construct a basic monitoring data set;

[0008] Based on the basic monitoring data set, the stable operating range of the current electrolysis process is determined, and a micro-perturbation control sequence is generated within the stable operating range. The micro-perturbation control sequence is then applied to the electrolysis process to obtain the perturbation effect state.

[0009] Based on the disturbance effect state, the action time interval corresponding to the micro-disturbance control sequence is determined, and chloride ion sensing detection data fragments of the action time interval and its adjacent intervals are extracted to construct the disturbance response data sequence.

[0010] Based on the disturbance response data sequence, segmentation processing is performed to extract the response change amplitude, response change rate and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set;

[0011] Based on the perturbation response feature set, response consistency analysis is performed on chloride ion sensing detection data to identify abnormal response intervals that do not match the perturbation response features. The abnormal response intervals are then replaced and corrected to obtain a correction sequence.

[0012] The calibration sequence is subjected to continuity screening and fluctuation suppression processing to determine the target chloride ion content value, and the chloride ion content monitoring results are output.

[0013] In an optional embodiment, the step of determining the stable operating range of the current electrolysis process based on the basic monitoring data set, generating a micro-perturbation control sequence within the stable operating range, and applying the micro-perturbation control sequence to the electrolysis process to obtain the perturbation effect state specifically includes:

[0014] Acquire chloride ion sensing data collected by a chloride ion sensor installed in the electrolytic cell during the copper foil electrolysis production process. The chloride ion sensing data is a numerical sequence characterizing the change of chloride ion concentration in the electrolyte over time.

[0015] Acquire electrolysis process parameter data within the time window corresponding to the chloride ion sensing detection data, wherein the electrolysis process parameter data includes electrolysis current parameters, solution flow rate parameters, and electrolyte temperature parameters;

[0016] Chloride ion sensing data and electrolysis process parameter data are time-aligned, and chloride ion sensing data and electrolysis process parameter data at the same time point are combined to construct a basic monitoring data set;

[0017] Chloride ion sensing detection data in the basic monitoring dataset are sorted by time to obtain a sequence of detection data arranged continuously by time.

[0018] Based on the detection data sequence, the numerical change between adjacent data points is calculated within a preset time window to obtain the change sequence;

[0019] Based on the change series, the difference between the maximum and minimum change within each time window is calculated to obtain the fluctuation amplitude series;

[0020] Based on the fluctuation amplitude sequence, time windows with fluctuation amplitudes below a preset fluctuation amplitude threshold are marked as candidate stable windows, thus obtaining a set of candidate stable windows;

[0021] Based on the candidate stable window set, temporally adjacent candidate stable windows are merged to obtain multiple continuous time intervals;

[0022] Based on continuous time intervals, the duration of each continuous time interval is filtered, and time intervals whose duration meets the preset conditions are retained to obtain stable operating intervals;

[0023] Based on the stable operating range, the stable operating range is divided into multiple equally spaced time nodes to obtain the time node sequence of disturbance effect;

[0024] Based on the time node sequence of the disturbance effect, a time-varying micro-disturbance control sequence is generated, wherein the micro-disturbance control sequence is a time-series control signal that corresponds one-to-one with the time node sequence of the disturbance effect.

[0025] By applying a micro-perturbation control sequence to the electrolysis process, the dynamic response state of the electrolysis process under the micro-perturbation is obtained, which is taken as the perturbation state.

[0026] In an optional embodiment, the step of determining the action time interval corresponding to the micro-perturbation control sequence based on the perturbation action state, and extracting chloride ion sensing detection data fragments from the action time interval and its adjacent intervals to construct a perturbation response data sequence specifically includes:

[0027] Based on the disturbance effect state, the overall application process of the micro-disturbance control sequence on the time axis is analyzed to determine the continuous time range in which the micro-disturbance control sequence has an actual impact on the electrolysis process, which is taken as the corresponding action time interval of the micro-disturbance control sequence;

[0028] The action time interval is located and marked in the chloride ion sensing detection data sequence, and the current chloride ion sensing detection data segment corresponding to the action time interval is obtained.

[0029] Based on the start and end times of the time interval, the time intervals are extended forward by a preset time length to obtain the preceding and following time intervals.

[0030] Extract chloride ion sensing data fragments corresponding to the action time interval and the preceding and following time intervals;

[0031] The chloride ion sensing data segments corresponding to each time interval are sequentially spliced ​​together to obtain a chloride ion data set with a continuous time structure.

[0032] Based on a continuous-time structure of chloride ion data set, a perturbation response data sequence is constructed.

[0033] In an optional embodiment, the segmentation processing based on the disturbance response data sequence to extract the response change amplitude, response change rate, and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set, specifically includes:

[0034] The disturbance response data sequence is processed by time-order traversal to obtain the response value sequence corresponding to each sampling point;

[0035] Based on the response numerical sequence, the difference between adjacent sampling points is calculated to obtain the response change difference sequence;

[0036] Based on the response change difference sequence, the interval of continuous positive difference is identified as the response rise interval, and the corresponding start and end positions are recorded.

[0037] Based on the response change difference sequence, the interval of consecutive negative difference is identified as the response decrease interval, and the corresponding start and end positions are recorded.

[0038] Based on the connection between the rising and falling intervals of the response, the response change interval corresponding to the disturbance application stage is determined;

[0039] Based on the response change range, the maximum and minimum response values ​​within the corresponding range are extracted, and the difference between the two is calculated to obtain the response change amplitude.

[0040] Based on the response change interval, the ratio of the response change amplitude to the corresponding time length within that interval is calculated to obtain the response change rate.

[0041] Based on the response numerical sequence after the end of the response change interval, the interval that regresses to the corresponding baseline response level before the disturbance is applied is identified, and the recovery process interval is obtained.

[0042] Based on the recovery process interval, the recovery start value and recovery end value are extracted, and the recovery duration and recovery change difference are calculated to obtain the recovery process characteristics;

[0043] A set of disturbance response characteristics is formed by combining the response change amplitude, response change rate, and recovery process characteristics.

[0044] In an optional embodiment, the step of performing response consistency analysis on chloride ion sensing detection data based on the perturbation response feature set, identifying abnormal response intervals that do not match the perturbation response features, and performing replacement correction processing on the abnormal response intervals to obtain a correction sequence specifically includes:

[0045] The chloride ion sensing data is processed by time sequence analysis to obtain a sequence of response values ​​arranged continuously in time.

[0046] Based on the response numerical sequence, the change difference between adjacent sampling points is calculated to obtain the response change difference sequence;

[0047] Based on the response change difference sequence, time intervals in which the change amplitude exceeds a preset fluctuation amplitude threshold are identified as candidate response intervals;

[0048] Based on the candidate response intervals, the rate of change of response within each interval is calculated to obtain the candidate response rate sequence;

[0049] The candidate response intervals and candidate response rate sequences are matched and analyzed with the response change amplitude and response change rate in the disturbance response feature set. Combined with the changes in electrolysis process parameters at the corresponding time points, the consistency of the abnormal response intervals is determined, and the abnormal response intervals are preliminarily identified.

[0050] Based on the initially determined abnormal response interval, candidate recovery process features within the corresponding interval are extracted, and the recovery process features are matched and analyzed with the recovery process features in the disturbance response feature set. The abnormal response interval is then screened and confirmed to obtain the target abnormal response interval.

[0051] Based on the target abnormal response interval, the target abnormal response data segment within the corresponding time range is extracted from the chloride ion sensing detection data;

[0052] Based on the adjacent normal response data segments before and after the abnormal data segment, the abnormal data segment is numerically replaced to obtain the correction sequence.

[0053] In an optional embodiment, the step of performing continuity screening and fluctuation convergence processing on the calibration sequence to determine the target chloride ion content value and outputting the chloride ion content monitoring result specifically includes:

[0054] The time sequence integrity of the calibrated chloride ion measurement sequence is verified to obtain the integrity verification sequence.

[0055] Data completion and noise reduction are performed based on the integrity verification sequence to obtain the preprocessed measurement sequence;

[0056] The preprocessed measurement sequence is continuously screened to identify continuous data intervals where the variation amplitude between adjacent sampling points meets the preset fluctuation convergence threshold and the electrolysis process parameters at the corresponding time points are in a stable state of change, thus obtaining the effective chloride ion measurement data sequence;

[0057] Fluctuation identification is performed on the effective chloride ion measurement data sequence, and sampling points whose amplitude changes exceed the preset fluctuation convergence threshold are marked as abnormal sampling points;

[0058] Amplitude compression is performed on abnormal sampling points, and local extrema are smoothed and replaced to obtain a smooth measurement sequence;

[0059] The target chloride ion content value was determined by extracting interval stable values ​​from the smoothed measurement sequence.

[0060] Output the target chloride ion content value to obtain the chloride ion content monitoring results of the copper foil electrolysis process.

[0061] Furthermore, a smart chloride ion content monitoring system based on multi-source data correction is proposed to realize the smart chloride ion content monitoring method described above, including:

[0062] The data acquisition module is used to acquire chloride ion sensing detection data and electrolysis process parameter data within the corresponding time window during the copper foil electrolysis production process, and to construct a basic monitoring data set.

[0063] The stability control module is used to determine the stable operating range of the electrolysis process based on the basic monitoring data set, and generate a micro-disturbance control sequence within the stable operating range and apply it to the electrolysis process to obtain the disturbance effect state.

[0064] The disturbance response construction module is used to determine the action time interval corresponding to the micro-disturbance control sequence according to the disturbance action state, and extract chloride ion sensing detection data fragments of the action time interval and its adjacent time intervals to construct the disturbance response data sequence.

[0065] The response feature extraction module is used to segment the disturbance response data sequence, extract the response change amplitude, response change rate and recovery process features, and form a disturbance response feature set.

[0066] The data correction module is used to perform response consistency analysis on chloride ion sensing detection data based on the perturbation response feature set, identify abnormal response intervals and perform replacement correction processing to obtain a correction sequence.

[0067] The result generation module is used to perform continuity screening and fluctuation convergence processing on the calibration sequence, determine the target chloride ion content value, and output the chloride ion content monitoring results.

[0068] In an optional embodiment, the stability control module includes:

[0069] A data alignment unit is used to perform time alignment processing on chloride ion sensing detection data and electrolysis process parameter data to generate a basic monitoring data set.

[0070] A fluctuation calculation unit is used to sort the chloride ion sensing detection data in the basic monitoring data set by time and calculate the change and fluctuation amplitude between adjacent sampling points.

[0071] A stable window filtering unit is used to identify candidate stable windows based on fluctuation amplitude, and to merge the candidate stable windows to generate a set of candidate stable windows.

[0072] A stable interval determination unit is used to filter the candidate stable window set by duration and determine a stable operating interval that meets preset conditions.

[0073] A disturbance node generation unit is used to generate an equally spaced time node sequence based on a stable operating interval, forming a disturbance effect time node sequence.

[0074] A micro-perturbation generation unit is used to generate a micro-perturbation control sequence based on the perturbation action time node sequence and apply the micro-perturbation control sequence to the electrolysis process.

[0075] In an optional embodiment, the disturbance response building module includes:

[0076] The effective range determination unit is used to determine the effective time range based on the actual effective time range of the micro-disturbance control sequence analyzed by the disturbance effect state.

[0077] A data positioning unit is used to locate the data segment corresponding to the time interval in the chloride ion sensing detection data.

[0078] A time extension unit is used to extend the time interval forward by a preset time length to obtain the preceding time interval, and to extend it backward to obtain the following time interval.

[0079] The data extraction unit is used to extract chloride ion sensing detection data segments corresponding to the action time interval and the preceding and following time intervals.

[0080] A sequence construction unit is used to splice data segments from each time interval to construct a continuous time structured disturbance response data sequence.

[0081] In an optional embodiment, the data correction module includes:

[0082] The difference analysis unit is used to perform time-series analysis on chloride ion sensing detection data and calculate the change difference between adjacent sampling points.

[0083] A candidate interval identification unit is used to identify candidate response intervals whose change amplitude exceeds a preset fluctuation amplitude threshold based on the change difference.

[0084] The feature matching unit is used to perform matching analysis on the response change rate and recovery process features of the candidate response interval with the disturbance response feature set to determine the abnormal response interval;

[0085] A secondary confirmation unit is used to perform secondary screening and confirmation of the abnormal response interval based on the characteristics of the recovery process to obtain the target abnormal response interval.

[0086] An anomaly interception unit is used to intercept the abnormal data segment corresponding to the target abnormal response interval;

[0087] A data replacement unit is used to perform numerical replacement processing based on adjacent normal response data segments before and after the abnormal data segment to generate a correction sequence.

[0088] Compared with the prior art, the beneficial effects of the present invention are:

[0089] This proposal presents an intelligent monitoring method and system for chloride ion content based on multi-source data correction. By constructing a stable operating range identification mechanism and a micro-disturbance control sequence, and introducing a disturbance response characteristic modeling and consistency analysis process, it achieves dynamic correction and abnormal response range correction of chloride ion sensing data. Simultaneously, by combining multi-source process parameter collaborative analysis, it improves the ability to distinguish between data fluctuations and sampling anomalies under complex electrolysis conditions. Furthermore, by continuously screening and fluctuation convergence processing, it outputs the target chloride ion content value, thereby improving the accuracy, stability, and dynamic adaptability of chloride ion content monitoring. Attached Figure Description

[0090] Figure 1 This is a flowchart of an intelligent monitoring method for chloride ion content based on multi-source data correction proposed in this invention;

[0091] Figure 2 This is a flowchart illustrating the stable region and micro-perturbation generation process in this invention;

[0092] Figure 3 This is a flowchart of the disturbance response modeling and correction process in this invention;

[0093] Figure 4 This is a system framework diagram of an intelligent monitoring system for chloride ion content based on multi-source data correction proposed in this invention. Detailed Implementation

[0094] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0095] Reference Figure 1 - Figure 4 As shown in the figure, an intelligent monitoring method for chloride ion content based on multi-source data correction in an embodiment of the present invention includes:

[0096] Acquire chloride ion sensing data during the copper foil electrolysis production process, and acquire electrolysis process parameter data within the corresponding time window to construct a basic monitoring data set;

[0097] Based on the basic monitoring data set, the stable operating range of the current electrolysis process is determined, and a micro-perturbation control sequence is generated within the stable operating range. The micro-perturbation control sequence is then applied to the electrolysis process to obtain the perturbation effect state.

[0098] Based on the disturbance effect state, the action time interval corresponding to the micro-disturbance control sequence is determined, and chloride ion sensing detection data fragments of the action time interval and its adjacent intervals are extracted to construct the disturbance response data sequence.

[0099] Based on the disturbance response data sequence, segmentation processing is performed to extract the response change amplitude, response change rate and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set;

[0100] Based on the perturbation response feature set, response consistency analysis is performed on chloride ion sensing detection data to identify abnormal response intervals that do not match the perturbation response features. The abnormal response intervals are then replaced and corrected to obtain a correction sequence.

[0101] The calibration sequence is subjected to continuity screening and fluctuation suppression processing to determine the target chloride ion content value, and the chloride ion content monitoring results are output.

[0102] Furthermore, based on the basic monitoring data set, the stable operating range of the current electrolysis process is determined, and a micro-perturbation control sequence is generated within the stable operating range. This micro-perturbation control sequence is applied to the electrolysis process to obtain the perturbation effect state, specifically including:

[0103] Chloride ion sensing data collected by a chloride ion sensor installed in the electrolytic cell during the copper foil electrolysis production process is obtained. The chloride ion sensing data is a numerical sequence characterizing the change of chloride ion concentration in the electrolyte over time.

[0104] Acquire electrolysis process parameter data within the time window corresponding to the chloride ion sensing detection data. The electrolysis process parameter data includes electrolysis current parameters, solution flow rate parameters, and electrolyte temperature parameters.

[0105] Chloride ion sensing data and electrolysis process parameter data are time-aligned, and chloride ion sensing data and electrolysis process parameter data at the same time point are combined to construct a basic monitoring data set;

[0106] Chloride ion sensing detection data in the basic monitoring dataset are sorted by time to obtain a sequence of detection data arranged continuously by time.

[0107] Based on the detection data sequence, the numerical change between adjacent data points is calculated within a preset time window to obtain the change sequence;

[0108] Based on the change series, the difference between the maximum and minimum change within each time window is calculated to obtain the fluctuation amplitude series;

[0109] Specifically, in the copper foil electrolysis production process, chloride ion sensing data is collected by a chloride ion sensor installed in the electrolytic cell. This data is a numerical sequence characterizing the change of chloride ion concentration in the electrolyte over time. For example, during continuous electrolysis operation, concentration measurements are continuously output at a fixed sampling period, set between 1 and 5 seconds, to match the relatively slow dynamic characteristics of chloride ion concentration changes in the electrolyte. This ensures that the sampling frequency can cover the gradual concentration changes caused by adjustments in the electrolysis current or fluctuations in solution flow rate. For example, when the electrolysis current changes from 10... When kA changes within a small adjustment range, the chloride ion concentration does not exhibit an instantaneous jump, but rather shows a continuous response curve of gradual increase or decrease, thus ensuring that the sensor data can accurately reflect the dynamic evolution of the electrolysis process. Simultaneously, electrolysis process parameter data within the time window corresponding to this detection data is acquired. This process parameter data includes electrolysis current parameters, solution flow rate parameters, and electrolyte temperature parameters. The electrolysis current parameter characterizes the change in the driving force of the electrolysis reaction; for example, fine-tuning the current directly affects the ion migration rate. The solution flow rate parameter characterizes the electrolyte circulation and mass transfer intensity; for example, changes in flow rate affect local... Concentration uniformity and electrolyte temperature parameters are used to characterize changes in the system's thermal stability; for example, temperature increases affect the ion diffusion coefficient. Chloride ion concentration data and process parameter data are aligned point-by-point according to timestamps to form a one-to-one correspondence of multidimensional data combinations at the same time point, thereby constructing a basic monitoring data set. After the set is formed, the chloride ion concentration data is rearranged in chronological order to eliminate time misalignment caused by sensor sampling delay, data transmission delay, or local disorder, so that the overall data restores a strictly monotonically increasing structure on the time axis, thus ensuring that subsequent analysis is based on the true time continuity.

[0110] After completing the time consistency processing, the continuous detection data sequence is segmented using a fixed-length sliding time window. The window length is determined based on the main period statistical characteristics of chloride ion concentration changes during stable operation. This main period is obtained by performing autocorrelation function analysis on historical stable operation data and extracting the main peak period. This ensures that the window scale corresponds to the dominant fluctuation rhythm of concentration changes in the electrolytic system. For example, in the continuous production process of copper foil electrolysis, when the current control system maintains electrolytic balance through periodic fine-tuning, the chloride ion concentration typically exhibits repetitive, slow fluctuations within a time scale of approximately 20 to 40 seconds. Therefore, this time scale is used as the basis for window segmentation, ensuring that each window fully covers a typical process disturbance-response process. Within each time window, the concentration values ​​between adjacent sampling points are differentially calculated to obtain a change sequence. This change is expressed in a directional differential form to characterize the rising or falling trend of concentration. For example, a positive value indicates a gradual increase in concentration, and a negative value indicates a gradual decrease in concentration. To avoid missing trend direction information due to using only absolute values, the difference between the maximum and minimum changes within each time window is calculated based on the change sequence to obtain a fluctuation amplitude sequence. The fluctuation amplitude characterizes the dispersion and intensity of concentration changes within that window and is sensitive to local anomalies. Time windows with fluctuation amplitudes below a preset fluctuation amplitude threshold are marked as candidate stable windows. This preset fluctuation amplitude threshold characterizes the maximum allowable fluctuation boundary of the electrolysis process under stable operating conditions. This threshold is obtained by statistically extracting historical stable operating data. Specifically, data from unstable operating conditions such as current switching, chemical dosing, and start-up / shutdown phases are first removed. Then, the mean and standard deviation of the fluctuation amplitude in the remaining stable segments are statistically analyzed. The statistical mean combined with the upper limit of the standard deviation forms a stability judgment boundary, ensuring that the threshold truly reflects the natural fluctuation range under stable operating conditions. The resulting set of candidate stable windows is used for further screening and merging analysis of subsequent stable operating intervals.

[0111] Based on the fluctuation amplitude sequence, time windows with fluctuation amplitudes below a preset fluctuation amplitude threshold are marked as candidate stable windows, thus obtaining a set of candidate stable windows;

[0112] Based on the candidate stable window set, temporally adjacent candidate stable windows are merged to obtain multiple continuous time intervals;

[0113] Based on continuous time intervals, the duration of each continuous time interval is filtered, and time intervals whose duration meets the preset conditions are retained to obtain stable operating intervals;

[0114] Based on the stable operating range, the stable operating range is divided into multiple equally spaced time nodes to obtain the time node sequence of disturbance effect;

[0115] Based on the time node sequence of disturbance effect, a time-varying micro-disturbance control sequence is generated. The micro-disturbance control sequence is a time-series control signal that corresponds one-to-one with the time node sequence of disturbance effect.

[0116] By applying a micro-perturbation control sequence to the electrolysis process, the dynamic response state of the electrolysis process under the micro-perturbation is obtained, which is taken as the perturbation state.

[0117] Specifically, after constructing the fluctuation amplitude sequence, the fluctuation amplitude corresponding to each time window is compared with the preset fluctuation amplitude threshold. The preset fluctuation amplitude threshold is used to characterize the maximum fluctuation boundary allowed when the electrolysis process is in a stable operating state. When the fluctuation amplitude of a certain time window is lower than the threshold, the time window is marked as a candidate stable window, and thus a set of candidate stable windows is obtained. For example, if the concentration in a certain stage of the continuous sampling sequence always fluctuates slightly around 0.35 g / L without abrupt change, then the window corresponding to that stage is included in the candidate set.

[0118] After obtaining the candidate stable window set, candidate stable windows that are temporally adjacent and have an interval less than a preset time gap are merged. A candidate stable window refers to a set of time segments that meet the low fluctuation condition. The time gap characterizes the maximum allowable break time scale. For example, in actual electrolysis operation, short-term control adjustments may cause the stable window to be split into multiple short intervals. Merging can restore it to a continuous time interval, thus obtaining multiple continuous time intervals. The duration of each continuous time interval is then filtered, obtained by accumulating the window coverage time. For example, an interval composed of multiple consecutive 10-second windows... Longer stable segments can be accumulated, and only the intervals whose duration meets the minimum duration characteristic obtained from the historical stable operation stage statistics are retained, thus obtaining the stable operation interval. Within the stable operation interval, the time node sequence is divided in an equally spaced manner, so that each time node represents the isochronous sampling position in the stable operation process, thus forming a disturbance effect time node sequence. Based on this node sequence, a micro-disturbance control sequence is generated. This control sequence is a time-sequential control signal that corresponds one-to-one with the time node. It forms joint control by introducing small-amplitude alternating disturbances in electrolytic current, solution flow rate, and electrolyte temperature. For example, at some nodes, the current is superimposed with a change of ±0.1 kA, the flow rate is superimposed with a change of ±2%, and the temperature is superimposed with a change of ±0.2℃, so that different process parameters form a coupled disturbance structure in the time dimension, thereby stimulating an observable dynamic response under a stable background. This dynamic response state is used as the disturbance effect state. For example, when the current increases slightly, the concentration shows a slow decreasing trend, and when the flow rate increases slightly, a short-term dilution effect occurs, thus forming an analyzable response trajectory.

[0119] Furthermore, based on the disturbance state, the action time interval corresponding to the micro-disturbance control sequence is determined, and chloride ion sensing detection data fragments of the action time interval and its adjacent intervals are extracted to construct the disturbance response data sequence, specifically including:

[0120] Based on the disturbance effect state, the overall application process of the micro-disturbance control sequence on the time axis is analyzed to determine the continuous time range in which the micro-disturbance control sequence has an actual impact on the electrolysis process, which is taken as the corresponding action time interval of the micro-disturbance control sequence;

[0121] The action time interval is located and marked in the chloride ion sensing detection data sequence, and the current chloride ion sensing detection data segment corresponding to the action time interval is obtained.

[0122] Based on the start and end times of the time interval, the time intervals are extended forward by a preset time length to obtain the preceding and following time intervals.

[0123] Extract chloride ion sensing data fragments corresponding to the action time interval and the preceding and following time intervals;

[0124] The chloride ion sensing data segments corresponding to each time interval are sequentially spliced ​​together to obtain a chloride ion data set with a continuous time structure.

[0125] Based on a continuous-time structure of chloride ion data set, a perturbation response data sequence is constructed.

[0126] Specifically, the overall application process of the micro-perturbation control sequence on the time axis is analyzed based on the perturbation state. The perturbation state characterizes the dynamic response of the electrolysis system under the combined micro-perturbation effects of multiple process parameters such as current, solution flow rate, and electrolyte temperature. The overall application process of the micro-perturbation control sequence on the time axis describes the continuous input relationship of the control signals at each time point. By performing time-series consistency analysis on this continuous input relationship, the continuous time range from the initial application of the perturbation input to the complete stable change phase of the chloride ion concentration response is identified. This determines the corresponding action time interval of the micro-perturbation control sequence. For example, in a certain stable operating phase, the current is ±0.1 kA added to a base of 10 kA. After the kA and flow rate are varied by ±2% on the baseline value, the chloride ion concentration begins to fluctuate synchronously after about 5 seconds, and gradually enters a new stable fluctuation range after about 40 seconds. The entire interval from 5 seconds to 40 seconds is taken as the action time interval, which is used to characterize the effective coupling process between the disturbance input and the concentration response. In this process, the action time interval does not refer to a single peak response moment, but covers the complete dynamic evolution process from the initial response lag, gradual enhancement to the tendency to stabilize, so that subsequent analysis can reflect the complete response link.

[0127] After determining the effective time interval, this time interval is located and marked in the chloride ion sensing detection data sequence. The chloride ion sensing detection data sequence is a sequence of concentration values ​​continuously recorded according to the sampling time interval, for example, forming a continuous concentration point series with a 1-second sampling period. The effective time interval is accurately mapped in the data sequence through timestamp matching, thereby extracting the current chloride ion sensing detection data segment corresponding to the effective time interval. Based on the start and end time points of the effective time interval, a preset time length is extended forward and backward, respectively, to obtain the preceding and following time intervals. The preset time length is used to supplement the baseline state before the disturbance and the recovery state after the disturbance. Its value is determined based on the average time span of the concentration transitioning from a stable state to the disturbance response and then back to stability in historical stable operation data. For example, in a typical electrolysis process, this transition time is about 10 to 30 seconds. The forward and backward extended intervals are used to fully cover this transition process, so that the response analysis not only includes the disturbance itself, but also covers the natural transition behavior of the system before and after the disturbance. Furthermore, the chloride ion sensing data segments corresponding to the action time interval, the preceding time interval, and the following time interval are extracted separately and spliced ​​in chronological order to maintain a strict temporal continuity structure. For example, the preceding time interval typically shows a concentration stable at a baseline level of approximately 0.35 g / L, while the concentration in the action time interval gradually deviates from the baseline and fluctuates between 0.33 g / L and 0.37 g / L. The following time interval gradually returns to a stable state of approximately 0.35 g / L. This splicing process forms a complete "stability-perturbation-recovery" continuous change curve, thereby avoiding the loss of boundary information caused by only extracting the perturbation interval. This allows the perturbation response data sequence to fully characterize the overall dynamic evolution of chloride ion concentration under micro-perturbation and provides a unified temporal structure basis for subsequent analysis of response amplitude, rate, and consistency.

[0128] Furthermore, based on the disturbance response data sequence, segmentation processing is performed to extract the response change amplitude, response change rate, and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set, specifically including:

[0129] The disturbance response data sequence is processed by time-order traversal to obtain the response value sequence corresponding to each sampling point;

[0130] Based on the response numerical sequence, the difference between adjacent sampling points is calculated to obtain the response change difference sequence;

[0131] Based on the response change difference sequence, the interval of continuous positive difference is identified as the response rise interval, and the corresponding start and end positions are recorded.

[0132] Based on the response change difference sequence, the interval of consecutive negative difference is identified as the response decrease interval, and the corresponding start and end positions are recorded.

[0133] Based on the connection between the rising and falling intervals of the response, the response change interval corresponding to the disturbance application stage is determined;

[0134] Based on the response change range, the maximum and minimum response values ​​within the corresponding range are extracted, and the difference between the two is calculated to obtain the response change amplitude.

[0135] Based on the response change interval, the ratio of the response change amplitude to the corresponding time length within that interval is calculated to obtain the response change rate.

[0136] Based on the response numerical sequence after the end of the response change interval, the interval that regresses to the corresponding baseline response level before the disturbance is applied is identified, and the recovery process interval is obtained.

[0137] Based on the recovery process interval, the recovery start value and recovery end value are extracted, and the recovery duration and recovery change difference are calculated to obtain the recovery process characteristics;

[0138] A set of disturbance response characteristics is formed by combining the response change amplitude, response change rate, and recovery process characteristics.

[0139] Specifically, the disturbance response data sequence is processed by time-sequential traversal. This sequence is a continuous concentration change sequence obtained by splicing together the affected time interval and its preceding and following intervals. The response values ​​at each time point are obtained through point-by-point sampling. For example, at a 1-second sampling period, a set of concentration values ​​arranged in ascending order over time is formed. This traversal process is used to reconstruct the dynamic change process of chloride ion concentration point-by-point along the time axis, ensuring strict temporal continuity for subsequent differential calculations. Based on this, differential calculations are performed on the response values ​​between adjacent sampling points to obtain a response change difference sequence. The difference is used to characterize the increasing or decreasing trend of chloride ion concentration at adjacent times. A positive difference indicates an increasing concentration trend, while a negative difference indicates a decreasing concentration trend. Based on this difference sequence, continuity is determined. Intervals with consecutive positive difference values ​​are identified as increasing response intervals, and their corresponding start and end positions are recorded. For example, in a certain disturbance phase, the concentration gradually increases from 0.34 g / L to 0.37 g / L. At g / L, multiple consecutive positive difference points constitute an upward interval. Similarly, consecutive negative difference intervals are identified as response downward intervals. For example, a continuous negative difference segment is formed during the process of the concentration falling from 0.37 g / L to 0.35 g / L, and the start and end positions of the interval are recorded simultaneously. The upward and downward intervals are used to characterize the response enhancement process and recovery and decline process caused by the disturbance, respectively.

[0140] After obtaining the response rise and fall intervals, the response change interval corresponding to the perturbation application stage is determined based on their connection relationship on the time axis. This connection relationship characterizes whether there is a continuous transition between the rise and fall stages, i.e., whether the two intervals form a continuous response link without breaks in time. For example, under a micro-perturbation, the concentration may first rise rapidly and then slowly fall back; this complete continuous interval constitutes the perturbation change interval. The maximum and minimum response values ​​are extracted within this interval, for example, a maximum of 0.37 g / L and a minimum of 0.34 g / L. The difference between these two values ​​is used to calculate the response change amplitude, which characterizes the maximum deviation of the concentration from the baseline state under the perturbation. Furthermore, this amplitude is divided by the corresponding time length to obtain the response change rate, which characterizes the intensity of concentration change per unit time. Subsequently, a backtracking analysis is performed based on the response value sequence after the response change interval ends to identify continuous intervals where the concentration returns to the baseline response level before the perturbation application, for example, recovering to 0.35 g / L. The time period of stable chloride ion concentration (g / L) is defined as the recovery process interval. The recovery process interval is used to characterize the dynamic process of the system returning to steady state from the disturbance state. The recovery start value and recovery end value are extracted, and the recovery duration and recovery change difference are calculated to obtain the recovery process characteristics. Finally, the response change amplitude, response change rate and recovery process characteristics are combined to form a disturbance response feature set, which is used to characterize the complete dynamic characteristics of chloride ion concentration from response, evolution to recovery under micro-perturbation, so that subsequent consistency analysis has a unified feature description basis.

[0141] Furthermore, based on the perturbation response feature set, response consistency analysis is performed on the chloride ion sensing detection data to identify abnormal response intervals that do not match the perturbation response features. These abnormal response intervals are then replaced and corrected to obtain a correction sequence, specifically including:

[0142] The chloride ion sensing data is processed by time sequence analysis to obtain a sequence of response values ​​arranged continuously in time.

[0143] Based on the response numerical sequence, the change difference between adjacent sampling points is calculated to obtain the response change difference sequence;

[0144] Based on the response change difference sequence, time intervals in which the change amplitude exceeds a preset fluctuation amplitude threshold are identified as candidate response intervals;

[0145] Based on the candidate response intervals, the rate of change of response within each interval is calculated to obtain the candidate response rate sequence;

[0146] The candidate response intervals and candidate response rate sequences are matched and analyzed with the response change amplitude and response change rate in the disturbance response feature set. Combined with the changes in electrolysis process parameters at the corresponding time points, the consistency of the abnormal response intervals is determined, and the abnormal response intervals are preliminarily identified.

[0147] Specifically, the chloride ion sensing data undergoes time-series analysis. This data is a sequence of concentration measurements continuously collected according to sampling timestamps. By sorting the sampling points chronologically, the data forms a strictly increasing sequence of response values ​​on the time axis. For example, with a sampling period of 1 second, concentration values ​​at t=1 s, t=2 s, and t=3 s can be generated sequentially, ensuring that subsequent change analysis is based on a continuous time structure. Furthermore, the concentration values ​​between adjacent sampling points are calculated by subtracting the concentration at the previous time from the current concentration, resulting in a response change difference sequence. This difference simultaneously characterizes the direction and magnitude of concentration change. For instance, when the concentration increases from 0.35 g / L to 0.352 g / L, the difference is +0.002 g / L, while when it decreases from 0.352 g / L to 0.349 g / L, the difference is -0.003 g / L. The difference sequence of g / L can fully reflect the dynamic evolution trend of concentration over time. Further, based on the response change difference sequence, statistical analysis is performed on the change amplitude within each consecutive time period. When the absolute value of the difference continuously exceeds a preset fluctuation amplitude threshold within a certain time period, that time period is identified as a candidate response interval. This preset fluctuation amplitude threshold is consistent with the threshold used in the aforementioned stable operation interval identification process, used to uniformly define the boundary between "normal stable fluctuation" and "significant response change." For example, under stable operation conditions, concentration fluctuations typically do not exceed ±0.003 g / L, but when changes exceeding ±0.01 g / L occur continuously within a certain time period, the interval is determined to have deviated from the stable fluctuation range and is thus marked as a candidate response interval.

[0148] After obtaining the candidate response intervals, the rate of change of response within each interval is calculated. This rate of change is obtained as the ratio of the maximum response difference within the interval to the corresponding duration, characterizing the drastic change in concentration per unit time. For example, if the concentration changes from 0.34 g / L to 0.37 g / L within 5 seconds, the rate of change can be expressed as a change of approximately 0.006 g / L per unit time, thus forming a candidate response rate sequence. Based on this, the candidate response intervals and their corresponding candidate response rate sequences are matched with the response amplitude and rate of change in the disturbance response feature set. This matching analysis is conducted using a dimensional comparison approach. First, the amplitude of the change is matched to determine whether the amplitude of the candidate interval falls within the range statistically obtained from the disturbance response feature set, for example, a change amplitude of 0.02 g / L to 0.04 g / L. The range of g / L is considered the normal perturbation response range, and intervals below or above this range are judged as amplitude mismatches. Subsequently, a consistency judgment is made on the rate of change, determining whether the rate of change in the candidate interval falls within the corresponding rate of change range. For example, a change per unit time within the range of 0.002 g / L to 0.005 g / L is considered a normal response evolution rate, and candidate intervals that significantly deviate from this range are judged as rate mismatches. Simultaneously, changes in electrolysis process parameters are introduced as auxiliary consistency constraints, comparing the changes in electrolytic current, solution flow rate, and electrolyte temperature at the corresponding time points of the candidate interval with the perturbation patterns recorded in the perturbation response feature set. For example, under micro-perturbation conditions, the current may only exist at ±0.1 g / L. With kA-level fine-tuning and small changes in flow rate and temperature, if a sudden change in current or an abnormal jump in flow occurs at the corresponding time point in the candidate interval, it indicates that the response is not caused by a micro-perturbation, thus reducing its matching consistency. By comprehensively evaluating the matching results of change amplitude, change rate, and process parameter consistency, such as by using weighted scoring or rule-based judgment, candidate intervals that do not meet the consistency conditions are initially identified as abnormal response intervals for further screening and correction.

[0149] Based on the initially determined abnormal response interval, candidate recovery process features within the corresponding interval are extracted, and the recovery process features are matched and analyzed with the recovery process features in the disturbance response feature set. The abnormal response interval is then screened and confirmed to obtain the target abnormal response interval.

[0150] Based on the target abnormal response interval, the target abnormal response data segment within the corresponding time range is extracted from the chloride ion sensing detection data;

[0151] Based on the adjacent normal response data segments before and after the abnormal data segment, the abnormal data segment is numerically replaced to obtain the correction sequence.

[0152] Specifically, after obtaining the initially determined abnormal response intervals, the concentration change process within each abnormal response interval is further analyzed to extract candidate recovery process features. These recovery process features characterize the dynamic behavior of chloride ion concentration returning to the baseline state before the disturbance is applied after experiencing abnormal fluctuations. These features include the concentration value at the start of recovery, the concentration value at the end of recovery, the recovery duration, and the overall magnitude and trend of concentration change during the recovery process. For example, if the concentration rapidly drops from 0.38 g / L to 0.35 g / L within an abnormal interval and stabilizes within approximately 20 seconds, then the recovery start value can be extracted as 0.38 g / L, the recovery end value as 0.35 g / L, the recovery duration as 20 seconds, and the recovery change magnitude as 0.03. Based on characteristic parameters such as g / L, the extracted candidate recovery process features are matched with the recovery process features in the disturbance response feature set. The matching analysis is carried out by comparing each dimension, namely, whether the recovery duration is within the corresponding recovery time distribution range in the disturbance response feature set, whether the deviation between the recovery end value and the baseline concentration before the disturbance is within a reasonable fluctuation range, and whether the trend of the recovery process maintains continuous monotonic regression characteristics. For example, if the recovery time is significantly shorter than the actual concentration change inertia or significantly longer than the normal regression process, or if the concentration after recovery still deviates significantly from the baseline level, it indicates that the recovery behavior of this interval is inconsistent with the disturbance response law, and thus the abnormal response interval is confirmed in the secondary screening to obtain the target abnormal response interval. Conversely, if the recovery process is consistent with the recovery behavior recorded in the disturbance response feature set in terms of time scale and numerical regression characteristics, it indicates that the interval belongs to the normal disturbance response process and is eliminated in the secondary screening to avoid misjudgment.

[0153] After screening and confirming the target abnormal response interval, based on the start and end positions of this interval on the time axis, a target abnormal response data segment within the corresponding time range is extracted from the chloride ion sensing detection data sequence. This data segment is a set of concentration values ​​within a continuous time range, used for subsequent correction processing. Subsequently, the abnormal data segment is numerically replaced based on the adjacent normal response data segments before and after it. The adjacent normal response data segments refer to the concentration data intervals located before and after the abnormal interval that have been determined to be normal disturbance responses or stable operating states through consistency checks. By utilizing the numerical distribution characteristics of these normal intervals, the abnormal interval is reconstructed to eliminate the impact of abnormal fluctuations on the continuity of the overall sequence. For example, the local mean of the preceding and following normal intervals can be used as the replacement benchmark to smoothly fill the abnormal interval, or linear interpolation can be performed based on the changing trends of the preceding and following intervals to ensure that the replaced data maintains a continuous transition relationship with the neighboring domain in terms of numerical values. For example, when the concentration of the previous normal interval stabilizes at 0.35 g / L and the subsequent normal interval also returns to 0.35 g / L, the original peak value of 0.38 g / L in the abnormal interval can be replaced with a value around 0.35 g / L. The small fluctuation sequence of g / L is used to avoid the interference of mutations on the overall trend. The above replacement process yields a corrected sequence with good continuity and no abnormal mutations, providing a reliable data basis for the subsequent extraction of stable concentration values.

[0154] Furthermore, the calibration sequence undergoes continuity screening and fluctuation convergence processing to determine the target chloride ion content value, and the chloride ion content monitoring results are output, specifically including:

[0155] The time sequence integrity of the calibrated chloride ion measurement sequence is verified to obtain the integrity verification sequence.

[0156] Data completion and noise reduction are performed based on the integrity verification sequence to obtain the preprocessed measurement sequence;

[0157] The preprocessed measurement sequence is continuously screened to identify continuous data intervals where the variation amplitude between adjacent sampling points meets the preset fluctuation convergence threshold and the electrolysis process parameters at the corresponding time points are in a stable state of change, thus obtaining the effective chloride ion measurement data sequence;

[0158] Fluctuation identification is performed on the effective chloride ion measurement data sequence, and sampling points whose amplitude changes exceed the preset fluctuation convergence threshold are marked as abnormal sampling points;

[0159] Amplitude compression is performed on abnormal sampling points, and local extrema are smoothed and replaced to obtain a smooth measurement sequence;

[0160] The target chloride ion content value was determined by extracting interval stable values ​​from the smoothed measurement sequence.

[0161] Output the target chloride ion content value to obtain the chloride ion content monitoring results of the copper foil electrolysis process.

[0162] Specifically, the time sequence integrity of the corrected chloride ion measurement sequence is verified. This corrected chloride ion measurement sequence is a continuous concentration value sequence corrected by replacing abnormal response intervals. The sequence is arranged sequentially with timestamps as indices at fixed sampling intervals (e.g., 1-second or 2-second sampling intervals). By comparing adjacent timestamps for continuity, the system identifies missing sampling points, duplicate sampling points, or time-order discrepancies. For example, if t=101 s is missing between t=100 s and t=101 s... If the data corresponds to 's', a time-series breakpoint is identified at that time position. Interpolation is then performed to fill in the breakpoint, restoring the temporal continuity of the sequence and forming an integrity verification sequence. This process ensures that subsequent concentration analysis is based on a continuous time axis, thus avoiding trend misjudgment caused by sampling interruptions. Based on this, the integrity verification sequence undergoes data completion and noise reduction processing. Data completion is used to recover missing or abnormally broken data. For example, when sampling is interrupted for a short period, linear interpolation or moving average is used to estimate the missing value based on the trend of normal sampling points before and after, ensuring the sequence maintains a continuous evolution relationship on the time axis. Noise reduction is used to suppress high-frequency spikes caused by sensor jitter, instantaneous disturbances in the electrolytic cell, and electromagnetic interference. For example, sliding window averaging or local median filtering is used to weaken single-point abnormal fluctuations, ensuring that local abrupt changes do not affect the overall trend identification, thus obtaining a preprocessed measurement sequence.

[0163] After obtaining the preprocessed measurement sequence, a continuous screening process is performed, which involves identifying continuous intervals on the time axis where the variation amplitude of adjacent sampling points is consistently small and shows a convergence trend. In this process, a preset fluctuation convergence threshold is introduced. This threshold is used to uniformly measure whether the concentration has entered the final stable convergence state. Its value is derived from the statistical analysis of the residual distribution of data after correction of historical stable operation stages, for example, it is determined based on the upper bound quantile value of the fluctuation residual in the stable segment, thereby ensuring that the screening results correspond to the real stable operating conditions. At the same time, it is required that the electrolysis process parameters at the corresponding time points are in a stable change state, that is, the electrolysis current, solution flow rate and electrolyte temperature do not have abrupt changes, but only have small continuous adjustments. This yields the effective chloride ion measurement data sequence. For example, if the concentration fluctuates slightly around 0.35 g / L for a long time in the stable production stage without significant drift, then this interval is retained as the effective data interval. Further fluctuation identification is performed on the effective chloride ion measurement data sequence. Sampling points with amplitude changes exceeding a preset fluctuation convergence threshold are marked as abnormal sampling points, such as peaks where the concentration suddenly rises from 0.35 g / L to 0.38 g / L and then quickly falls back. These points represent residual local unstable disturbances. Subsequently, amplitude compression processing is performed on abnormal sampling points, that is, abnormal values ​​are pulled back to the mean of normal sampling in the neighborhood, so that their amplitude is limited to the range of normal local fluctuations. Local extreme values ​​are smoothed and replaced, for example, by using a neighborhood weighted average to replace peaks, thereby weakening the impact of abnormal fluctuations and finally obtaining a smooth measurement sequence. Based on the smooth measurement sequence, interval stable values ​​are extracted. By identifying time intervals that continuously meet the convergence conditions, such as intervals where the fluctuation is always below the convergence threshold for a continuous period of time, the concentration values ​​in this interval are statistically processed, for example, the mean or median value is taken as a stable representative value to reduce the impact of single-point fluctuations, thereby determining the target chloride ion content value. Finally, the target chloride ion content value is output to form the chloride ion content monitoring result of the copper foil electrolysis process.

[0164] Furthermore, a smart chloride ion content monitoring system based on multi-source data correction is proposed to realize the smart chloride ion content monitoring method described above, including:

[0165] The data acquisition module is used to acquire chloride ion sensing data and electrolysis process parameter data within the corresponding time window during the copper foil electrolysis production process, and to construct a basic monitoring data set.

[0166] The stability control module is used to determine the stable operating range of the electrolysis process based on the basic monitoring data set, and generate a micro-disturbance control sequence within the stable operating range and apply it to the electrolysis process to obtain the disturbance effect state.

[0167] The disturbance response construction module is used to determine the action time interval corresponding to the micro-disturbance control sequence based on the disturbance action state, and extract chloride ion sensing detection data fragments of the action time interval and its adjacent time intervals to construct the disturbance response data sequence.

[0168] The response feature extraction module is used to segment the disturbance response data sequence, extract the response change amplitude, response change rate and recovery process features, and form a disturbance response feature set.

[0169] The data correction module is used to perform response consistency analysis on chloride ion sensing detection data based on the perturbation response feature set, identify abnormal response intervals and perform replacement correction processing to obtain a correction sequence.

[0170] The results generation module is used to perform continuity screening and fluctuation convergence processing on the calibration sequence, determine the target chloride ion content value, and output the chloride ion content monitoring results.

[0171] Furthermore, the stability control module includes:

[0172] The data alignment unit is used to perform time alignment processing on chloride ion sensing detection data and electrolysis process parameter data to generate a basic monitoring data set.

[0173] The fluctuation calculation unit is used to sort the chloride ion sensing detection data in the basic monitoring dataset by time and calculate the change and fluctuation amplitude between adjacent sampling points.

[0174] The stable window filtering unit is used to identify candidate stable windows based on the fluctuation amplitude, and to merge the candidate stable windows to generate a set of candidate stable windows.

[0175] The stable interval determination unit is used to filter the candidate stable window set for a continuous duration and determine the stable operating interval that meets the preset conditions.

[0176] The disturbance node generation unit is used to generate an equally spaced time node sequence based on the stable operating interval, forming a disturbance effect time node sequence.

[0177] The micro-perturbation generation unit is used to generate a micro-perturbation control sequence based on the perturbation action time node sequence and apply the micro-perturbation control sequence to the electrolysis process.

[0178] Furthermore, the disturbance response building blocks include:

[0179] The action interval determination unit is used to analyze the actual action time range of the micro-disturbance control sequence based on the disturbance action state and determine the action time interval.

[0180] The data positioning unit is used to locate the data segment corresponding to the time interval in the chloride ion sensing detection data.

[0181] The time extension unit is used to extend the time interval forward by a preset time length to obtain the preceding time interval, and to extend it backward to obtain the following time interval.

[0182] The data extraction unit is used to extract chloride ion sensing detection data segments corresponding to the action time interval and the preceding and following time intervals.

[0183] Sequence building unit: The sequence building unit is used to splice data segments from different time intervals to construct a continuous time structured disturbance response data sequence.

[0184] Furthermore, the data correction module includes:

[0185] The difference analysis unit is used to perform time-series analysis on chloride ion sensing detection data and calculate the difference in change between adjacent sampling points.

[0186] The candidate interval identification unit is used to identify candidate response intervals whose change amplitude exceeds a preset fluctuation amplitude threshold based on the change difference.

[0187] The feature matching unit is used to match and analyze the response change rate and recovery process characteristics of the candidate response interval with the disturbance response feature set to determine the abnormal response interval.

[0188] The secondary confirmation unit is used to perform secondary screening and confirmation of the abnormal response interval based on the characteristics of the recovery process to obtain the target abnormal response interval.

[0189] The exception interception unit is used to intercept the exception data segment corresponding to the target exception response interval;

[0190] The data replacement unit is used to perform numerical replacement processing based on the adjacent normal response data segments before and after the abnormal data segment to generate a correction sequence.

[0191] In summary, the advantages of this invention are as follows: by collaboratively analyzing chloride ion sensing data and electrolysis process parameters during copper foil electrolysis production, a basic monitoring data set is constructed. Based on the stable operating range identification and micro-disturbance control sequence generation mechanism, a disturbance response feature extraction and consistency correction method is introduced to identify and correct abnormal response ranges in the chloride ion detection data. Simultaneously, by combining continuous screening and fluctuation convergence processing, the corrected data is optimized for output, thereby improving the accuracy, stability, and adaptability to complex operating conditions of the chloride ion content monitoring results.

[0192] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for intelligent monitoring of chloride ion content based on multi-source data correction, characterized in that, include: Acquire chloride ion sensing data during the copper foil electrolysis production process, and acquire electrolysis process parameter data within the corresponding time window to construct a basic monitoring data set; Based on the basic monitoring data set, the stable operating range of the current electrolysis process is determined, and a micro-perturbation control sequence is generated within the stable operating range. The micro-perturbation control sequence is then applied to the electrolysis process to obtain the perturbation effect state. Based on the disturbance effect state, the action time interval corresponding to the micro-disturbance control sequence is determined, and chloride ion sensing detection data fragments of the action time interval and its adjacent intervals are extracted to construct the disturbance response data sequence. Based on the disturbance response data sequence, segmentation processing is performed to extract the response change amplitude, response change rate and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set; Based on the perturbation response feature set, response consistency analysis is performed on chloride ion sensing detection data to identify abnormal response intervals that do not match the perturbation response features. The abnormal response intervals are then replaced and corrected to obtain a correction sequence. The calibration sequence is subjected to continuity screening and fluctuation suppression processing to determine the target chloride ion content value, and the chloride ion content monitoring results are output. 2.The method of claim 1, wherein, The process involves determining the stable operating range of the current electrolysis process based on a set of basic monitoring data, generating a micro-perturbation control sequence within the stable operating range, and applying the micro-perturbation control sequence to the electrolysis process to obtain the perturbation effect state. Specifically, this includes: Acquire chloride ion sensing data collected by a chloride ion sensor installed in the electrolytic cell during the copper foil electrolysis production process. The chloride ion sensing data is a numerical sequence characterizing the change of chloride ion concentration in the electrolyte over time. Acquire electrolysis process parameter data within the time window corresponding to the chloride ion sensing detection data, wherein the electrolysis process parameter data includes electrolysis current parameters, solution flow rate parameters, and electrolyte temperature parameters; Chloride ion sensing data and electrolysis process parameter data are time-aligned, and chloride ion sensing data and electrolysis process parameter data at the same time point are combined to construct a basic monitoring data set; Chloride ion sensing detection data in the basic monitoring dataset are sorted by time to obtain a sequence of detection data arranged continuously by time. Based on the detection data sequence, the numerical change between adjacent data points is calculated within a preset time window to obtain the change sequence; Based on the change series, the difference between the maximum and minimum change within each time window is calculated to obtain the fluctuation amplitude series; Based on the fluctuation amplitude sequence, time windows with fluctuation amplitudes below a preset fluctuation amplitude threshold are marked as candidate stable windows, thus obtaining a set of candidate stable windows; Based on the candidate stable window set, temporally adjacent candidate stable windows are merged to obtain multiple continuous time intervals; Based on continuous time intervals, the duration of each continuous time interval is filtered, and time intervals whose duration meets the preset conditions are retained to obtain stable operating intervals; Based on the stable operating range, the stable operating range is divided into multiple equally spaced time nodes to obtain the time node sequence of disturbance effect; Based on the time node sequence of the disturbance effect, a time-varying micro-disturbance control sequence is generated, wherein the micro-disturbance control sequence is a time-series control signal that corresponds one-to-one with the time node sequence of the disturbance effect. By applying a micro-perturbation control sequence to the electrolysis process, the dynamic response state of the electrolysis process under the micro-perturbation is obtained, which is taken as the perturbation state. 3.The method of claim 1, wherein, Based on the disturbance state, the action time interval corresponding to the micro-disturbance control sequence is determined, and chloride ion sensing detection data fragments of the action time interval and its adjacent intervals are extracted to construct a disturbance response data sequence, specifically including: Based on the disturbance effect state, the overall application process of the micro-disturbance control sequence on the time axis is analyzed to determine the continuous time range in which the micro-disturbance control sequence has an actual impact on the electrolysis process, which is taken as the corresponding action time interval of the micro-disturbance control sequence; The action time interval is located and marked in the chloride ion sensing detection data sequence, and the current chloride ion sensing detection data segment corresponding to the action time interval is obtained. Based on the start and end times of the time interval, the time intervals are extended forward by a preset time length to obtain the preceding and following time intervals. Extract chloride ion sensing data fragments corresponding to the action time interval and the preceding and following time intervals; The chloride ion sensing data segments corresponding to each time interval are sequentially spliced ​​together to obtain a chloride ion data set with a continuous time structure. Based on a continuous-time structure of chloride ion data set, a perturbation response data sequence is constructed.

4. The intelligent monitoring method for chloride ion content based on multi-source data correction according to claim 1, characterized in that, The segmentation processing based on the disturbance response data sequence extracts the response change amplitude, response change rate, and recovery process features corresponding to the disturbance application stage, forming a disturbance response feature set, specifically including: The disturbance response data sequence is processed by time-order traversal to obtain the response value sequence corresponding to each sampling point; Based on the response numerical sequence, the difference between adjacent sampling points is calculated to obtain the response change difference sequence; Based on the response change difference sequence, the interval of continuous positive difference is identified as the response rise interval, and the corresponding start and end positions are recorded. Based on the response change difference sequence, the interval of consecutive negative difference is identified as the response decrease interval, and the corresponding start and end positions are recorded. Based on the connection between the rising and falling intervals of the response, the response change interval corresponding to the disturbance application stage is determined; Based on the response change range, the maximum and minimum response values ​​within the corresponding range are extracted, and the difference between the two is calculated to obtain the response change amplitude. Based on the response change interval, the ratio of the response change amplitude to the corresponding time length within that interval is calculated to obtain the response change rate. Based on the response numerical sequence after the end of the response change interval, the interval that regresses to the corresponding baseline response level before the disturbance is applied is identified, and the recovery process interval is obtained. Based on the recovery process interval, the recovery start value and recovery end value are extracted, and the recovery duration and recovery change difference are calculated to obtain the recovery process characteristics; A set of disturbance response characteristics is formed by combining the response change amplitude, response change rate, and recovery process characteristics.

5. The intelligent monitoring method for chloride ion content based on multi-source data correction according to claim 1, characterized in that, The method involves performing response consistency analysis on chloride ion sensing data based on a set of perturbation response features, identifying abnormal response intervals that do not match the perturbation response features, and then performing replacement and correction processing on these abnormal response intervals to obtain a correction sequence. Specifically, this includes: The chloride ion sensing data is processed by time sequence analysis to obtain a sequence of response values ​​arranged continuously in time. Based on the response numerical sequence, the change difference between adjacent sampling points is calculated to obtain the response change difference sequence; Based on the response change difference sequence, time intervals in which the change amplitude exceeds a preset fluctuation amplitude threshold are identified as candidate response intervals; Based on the candidate response intervals, the rate of change of response within each interval is calculated to obtain the candidate response rate sequence; The candidate response intervals and candidate response rate sequences are matched and analyzed with the response change amplitude and response change rate in the disturbance response feature set. Combined with the changes in electrolysis process parameters at the corresponding time points, the consistency of the abnormal response intervals is determined, and the abnormal response intervals are preliminarily identified. Based on the initially determined abnormal response interval, candidate recovery process features within the corresponding interval are extracted, and the recovery process features are matched and analyzed with the recovery process features in the disturbance response feature set. The abnormal response interval is then screened and confirmed to obtain the target abnormal response interval. Based on the target abnormal response interval, the target abnormal response data segment within the corresponding time range is extracted from the chloride ion sensing detection data; Based on the adjacent normal response data segments before and after the abnormal data segment, the abnormal data segment is numerically replaced to obtain the correction sequence.

6. The intelligent monitoring method for chloride ion content based on multi-source data correction according to claim 1, characterized in that, The process of performing continuity screening and fluctuation convergence processing on the calibration sequence to determine the target chloride ion content value and outputting the chloride ion content monitoring results specifically includes: The time sequence integrity of the calibrated chloride ion measurement sequence is verified to obtain the integrity verification sequence. Data completion and noise reduction are performed based on the integrity verification sequence to obtain the preprocessed measurement sequence; The preprocessed measurement sequence is continuously screened to identify continuous data intervals where the variation amplitude between adjacent sampling points meets the preset fluctuation convergence threshold and the electrolysis process parameters at the corresponding time points are in a stable state of change, thus obtaining the effective chloride ion measurement data sequence; Fluctuation identification is performed on the effective chloride ion measurement data sequence, and sampling points whose amplitude changes exceed the preset fluctuation convergence threshold are marked as abnormal sampling points; Amplitude compression is performed on abnormal sampling points, and local extrema are smoothed and replaced to obtain a smooth measurement sequence; The target chloride ion content value was determined by extracting interval stable values ​​from the smoothed measurement sequence. Output the target chloride ion content value to obtain the chloride ion content monitoring results of the copper foil electrolysis process.

7. A smart chloride ion content monitoring system based on multi-source data correction, used to implement the smart chloride ion content monitoring method as described in any one of claims 1-6, characterized in that, include: The data acquisition module is used to acquire chloride ion sensing detection data and electrolysis process parameter data within the corresponding time window during the copper foil electrolysis production process, and to construct a basic monitoring data set. The stability control module is used to determine the stable operating range of the electrolysis process based on the basic monitoring data set, and generate a micro-disturbance control sequence within the stable operating range and apply it to the electrolysis process to obtain the disturbance effect state. The disturbance response construction module is used to determine the action time interval corresponding to the micro-disturbance control sequence according to the disturbance action state, and extract chloride ion sensing detection data fragments of the action time interval and its adjacent time intervals to construct the disturbance response data sequence. The response feature extraction module is used to segment the disturbance response data sequence, extract the response change amplitude, response change rate and recovery process features, and form a disturbance response feature set. The data correction module is used to perform response consistency analysis on chloride ion sensing detection data based on the perturbation response feature set, identify abnormal response intervals and perform replacement correction processing to obtain a correction sequence. The result generation module is used to perform continuity screening and fluctuation convergence processing on the calibration sequence, determine the target chloride ion content value, and output the chloride ion content monitoring results.

8. The intelligent monitoring system for chloride ion content based on multi-source data correction according to claim 7, characterized in that, The stability control module includes: A data alignment unit is used to perform time alignment processing on chloride ion sensing detection data and electrolysis process parameter data to generate a basic monitoring data set. A fluctuation calculation unit is used to sort the chloride ion sensing detection data in the basic monitoring data set by time and calculate the change and fluctuation amplitude between adjacent sampling points. A stable window filtering unit is used to identify candidate stable windows based on fluctuation amplitude, and to merge the candidate stable windows to generate a set of candidate stable windows. A stable interval determination unit is used to filter the candidate stable window set by duration and determine a stable operating interval that meets preset conditions. A disturbance node generation unit is used to generate an equally spaced time node sequence based on a stable operating interval, forming a disturbance effect time node sequence. A micro-perturbation generation unit is used to generate a micro-perturbation control sequence based on the perturbation action time node sequence and apply the micro-perturbation control sequence to the electrolysis process.

9. The intelligent monitoring system for chloride ion content based on multi-source data correction according to claim 7, characterized in that, The disturbance response construction module includes: The effective range determination unit is used to determine the effective time range based on the actual effective time range of the micro-disturbance control sequence analyzed by the disturbance effect state. A data positioning unit is used to locate the data segment corresponding to the time interval in the chloride ion sensing detection data. A time extension unit is used to extend the time interval forward by a preset time length to obtain the preceding time interval, and to extend it backward to obtain the following time interval. The data extraction unit is used to extract chloride ion sensing detection data segments corresponding to the action time interval and the preceding and following time intervals. A sequence construction unit is used to splice data segments from each time interval to construct a continuous time structured disturbance response data sequence.

10. The intelligent monitoring system for chloride ion content based on multi-source data correction according to claim 7, characterized in that, The data correction module includes: The difference analysis unit is used to perform time-series analysis on chloride ion sensing detection data and calculate the change difference between adjacent sampling points. A candidate interval identification unit is used to identify candidate response intervals whose change amplitude exceeds a preset fluctuation amplitude threshold based on the change difference. The feature matching unit is used to perform matching analysis on the response change rate and recovery process features of the candidate response interval with the disturbance response feature set to determine the abnormal response interval; A secondary confirmation unit is used to perform secondary screening and confirmation of the abnormal response interval based on the characteristics of the recovery process to obtain the target abnormal response interval. An anomaly interception unit is used to intercept the abnormal data segment corresponding to the target abnormal response interval; A data replacement unit is used to perform numerical replacement processing based on adjacent normal response data segments before and after the abnormal data segment to generate a correction sequence.

Images