An ion concentration analysis method and system

Abstract

The present application relates to the technical field of data processing, and discloses an ion concentration analysis method and system. The method includes: collecting spectral data at multiple monitoring points in the hydroxide ion chamber to obtain an original spectral data set for each monitoring point; respectively performing spectral amplitude, shape, and information difference measurement on the original spectral data set to obtain comprehensive spectral feature data for each monitoring point; inputting the comprehensive spectral feature data of each monitoring point into a preset partial least squares regression model for ion concentration prediction to obtain ion concentration prediction data for each monitoring point; inputting the ion concentration prediction data of each monitoring point into a preset integrated cost-sensitive deep neural network model for ion concentration anomaly detection to obtain a target ion concentration anomaly detection result. The present application improves the accuracy of ion concentration detection in the hydroxide ion chamber.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to an ion concentration analysis method and system. Background Art

[0002] Accurate ion concentration monitoring can greatly improve the operating performance and safety of the hydroxide ion chamber. However, traditional ion concentration monitoring methods often rely on offline chemical analysis techniques, which are not only time-consuming but also difficult to meet the requirements of rapid response because they cannot provide real-time data.

[0003] Although some existing online monitoring methods such as conductivity measurement and sensors have been widely used, these methods usually can only provide limited information and cannot capture more complex ion changes, such as the interaction of multiple ions and the dynamic change of concentration. That is, the accuracy rate of the existing technology is low. Summary of the Invention

[0004] This application provides an ion concentration analysis method and system, which are used to improve the accuracy rate of ion concentration detection in the hydroxide ion chamber.

[0005] In a first aspect, this application provides an ion concentration analysis method, and the ion concentration analysis method includes:

[0006] Collect spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set of each monitoring point;

[0007] Measure the spectral amplitude, shape, and information difference of the original spectral data set respectively to obtain the comprehensive spectral feature data of each monitoring point;

[0008] Input the comprehensive spectral feature data of each monitoring point into a preset partial least squares regression model for ion concentration prediction to obtain the ion concentration prediction data of each monitoring point;

[0009] Input the ion concentration prediction data of each monitoring point into a preset integrated cost-sensitive deep neural network model for ion concentration anomaly detection to obtain the target ion concentration anomaly detection result.

[0010] In a second aspect, this application provides an ion concentration analysis system, and the ion concentration analysis system includes:

[0011] A collection module, which is used to collect spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set of each monitoring point;

[0012] A measurement module, which is used to measure the spectral amplitude, shape, and information difference of the original spectral data set respectively to obtain the comprehensive spectral feature data of each monitoring point;

[0013] A prediction module, configured to input the comprehensive spectral feature data of each monitoring point into a preset partial least squares regression model for ion concentration prediction, so as to obtain the ion concentration prediction data of each monitoring point;

[0014] A monitoring module, configured to input the ion concentration prediction data of each monitoring point into a preset integrated cost-sensitive deep neural network model for ion concentration anomaly detection, so as to obtain the target ion concentration anomaly detection result.

[0015] The third aspect of the present application provides an ion concentration analysis device, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor calls the instructions in the memory, so that the ion concentration analysis device executes the above-mentioned ion concentration analysis method.

[0016] The fourth aspect of the present application provides a computer-readable storage medium, in which instructions are stored, and when it runs on a computer, it enables the computer to execute the above-mentioned ion concentration analysis method.

[0017] In the technical solution provided by the present application, by deploying a multi-channel spectrometer in the hydroxide ion chamber to collect spectral data in real time, the change of ion concentration can be continuously monitored, ensuring the timeliness and continuity of data. Through fine spectral amplitude, shape and information difference measurement, the small differences between each monitoring point can be accurately identified and analyzed, enhancing the accuracy of data and the meticulousness of analysis. Using the partial least squares regression model to predict the ion concentration from the comprehensive spectral feature data can effectively extract the factors highly related to the ion concentration from complex spectral information. The data structure is simplified through the dimensionality reduction technology, while retaining the most critical information for ion concentration prediction. It not only improves the accuracy of prediction, but also can dynamically adjust the prediction model based on historical and real-time data to adapt to possible environmental changes and equipment aging. By inputting the prediction data into the integrated cost-sensitive deep neural network model for ion concentration anomaly detection, the ability to identify potential risks is enhanced, and thus the accuracy of ion concentration detection in the hydroxide ion chamber is improved. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a schematic diagram of an embodiment of the ion concentration analysis method in the embodiments of the present application;

[0020] Figure 2 This is a schematic diagram of an embodiment of the ion concentration analysis system in the embodiments of the present application. Detailed implementation manners

[0021] The embodiments of the present application provide an ion concentration analysis method and system. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any deformation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0022] For ease of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 , an embodiment of the ion concentration analysis method in the embodiments of the present application includes:

[0023] Step S101: Collect spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set of each monitoring point;

[0024] It can be understood that the execution subject of the present application can be an ion concentration analysis system, or a terminal or a server. Specifically, it is not limited here. The embodiments of the present application will be described by taking the server as the execution subject as an example.

[0025] Specifically, multiple monitoring points are preset in the hydroxide ion chamber, and the setting of the points is based on the expected monitoring requirements and ion distribution characteristics. To ensure the accuracy and coverage of data collection, each monitoring point is equipped with a multi-channel spectrometer, which can operate according to the set wavelength range and acquisition frequency. When implementing data collection, the multi-channel spectrometer performs spectral scanning at each monitoring point according to the predetermined parameters to obtain the first spectral data set. To improve the data quality and ensure the accuracy of subsequent analysis, data cleaning is performed on the first spectral data set, including the removal of outliers and the interpolation of missing values, to obtain the second spectral data set. Data tags are marked on the second spectral data set, and these tags are based on factors such as the geographical location of the monitoring point and the acquisition time, so that each set of data can be accurately traced and analyzed. An effective dataset division is performed on the tagged spectral data set, and the data is assigned to the original spectral data sets corresponding to each monitoring point, so that the data of each monitoring point can be analyzed independently of other points, thereby providing more accurate input during model training and prediction and ensuring the efficiency and accuracy of the entire ion concentration analysis method.

[0026] Step S102: Respectively perform spectral amplitude, shape, and information difference measurements on the original spectral data sets to obtain the comprehensive spectral feature data of each monitoring point;

[0027] Specifically, curve fitting is respectively performed on the original spectral data sets to obtain the spectral curves of each monitoring point. The curve shows the basic graph of the spectral data and also reflects the spectral response of the monitoring point in a specific environment. Spectral amplitude comparison analysis is performed on the spectral curves to obtain the spectral amplitude differences between different monitoring points. The comparison results are obtained by comparing the peak heights and overall energy distributions of the spectral curves at each point, and the spectral amplitude differences are directly related to the changes in ion concentration. By comparing the spectral shapes, the differences in the peak shape symmetry, width, sharpness, etc. of the spectral curves at each monitoring point are analyzed. Shape comparison helps to understand the possible changes in the types and states of substances between different points. Spectral information measurement is performed on the spectral curves of each monitoring point, and the information entropy or other statistical measurement values of each spectral curve are calculated to quantify the information richness of the spectral data, obtaining the spectral information measurement differences between multiple monitoring points. The information measurement differences between different points help to determine which monitoring points may be affected by external interference or internal parameter changes. According to the spectral amplitude difference comparison results, spectral shape difference comparison results, and spectral information measurement difference results, spectral feature band extraction and fusion are respectively performed on the spectral curves of each monitoring point. By selecting the most representative and different bands in the spectral data of each monitoring point and then effectively fusing these bands, the comprehensive spectral feature data of each point is formed.

[0028] According to the results of the spectral amplitude difference comparison, the spectral curve of each monitoring point is analyzed, and the band reflecting the most significant change in spectral intensity is extracted to obtain the first spectral characteristic band. Identify those bands with large amplitude changes, which can often indicate changes in ion concentration or differences in chemical composition. Perform a second round of feature extraction based on the results of the spectral shape difference comparison. Analyze the characteristic differences in shape of each spectral curve, such as peak width, symmetry, and sharpness, and extract bands with significant shape differences. These bands are defined as the second spectral characteristic bands. The analysis of spectral shape helps identify changes caused by physical or chemical processes, which may be caused by specific ions in the monitoring environment. According to the difference in spectral information measurement, characteristic bands with high information content are extracted from each spectral curve to obtain the third spectral characteristic band. Use information entropy or other statistical methods to evaluate the information contribution of each band, and select bands with large information content as key features. This metric reflects the amount of information carried in the spectral data, which helps to mine and utilize those bands that may be most critical for ion concentration prediction. The first, second and third spectral characteristic bands are screened for similarity to identify the bands that appear repeatedly in different feature extraction steps, ensuring that the target spectral characteristic bands finally selected are representative and consistent from multiple angles. By comparing the overlap and similarity of each characteristic band, the target spectral characteristic band that best represents the characteristics of each monitoring point is selected. The characteristic weights of the target spectral characteristic bands of each spectral curve are calculated, and the weight data reflects the importance and contribution of each characteristic band in the overall analysis. The weights may be determined based on the effectiveness of the bands in the ion concentration prediction model, or their ability to distinguish different chemical components. According to the calculated characteristic weight data, the target spectral characteristic bands of each spectral curve are weighted and combined to form comprehensive spectral characteristic data for each monitoring point. The weighted combination retains the unique information of each characteristic band and enhances the accuracy and robustness of the model in predicting ion concentrations.

[0029] Step S103, inputting the comprehensive spectral characteristic data of each monitoring point into a preset partial least squares regression model to predict the ion concentration, and obtaining the predicted ion concentration data of each monitoring point;

[0030] Specifically, standard normal variate transformation is performed on the comprehensive spectral feature data of each monitoring point to convert the original spectral data into standardized spectral feature data, ensuring that the data has a unified scale and distribution before further analysis, thus effectively avoiding biases in the model during the training process and improving the accuracy of prediction. The standardized spectral feature data of each monitoring point is input into a pre-set partial least squares regression model. In this model, by performing matrix transformation processing on the input data, a spectral feature data matrix that can reveal the main change trends of the data is constructed. The transformation is achieved through the method of partial least squares regression, and its core lies in finding the most important components, that is, the latent variables of the model, by maximizing the covariance between the dependent variable and the independent variables. Calculate the scores of the spectral feature data matrix in multiple latent variable spaces, and accordingly construct a score matrix for each monitoring point. The score matrix reflects the main change characteristics and patterns of the spectral data of each monitoring point, providing key explanatory variables for the prediction of ion concentration. Through the score matrices of these latent variables, the main factors affecting the change of ion concentration are effectively captured. By obtaining the regression coefficients of the partial least squares regression model and combining with the score matrix obtained above, the predicted ion concentration data of each monitoring point is calculated. The regression coefficients provide a specific conversion method from the latent variable scores to the predicted ion concentration, ensuring the accuracy and reliability of the prediction. Through the comprehensive data processing and model application process, the ion concentration of each monitoring point is predicted, and the complex relationship between the spectral data and the ion concentration is obtained.

[0031] Step S104: Input the predicted ion concentration data of each monitoring point into a pre-set integrated cost-sensitive deep neural network model for ion concentration anomaly detection to obtain the target ion concentration anomaly detection result.

[0032] Specifically, the in-warehouse location information of each monitoring point is obtained. The abnormality of ion concentration may be closely related to the specific location of the monitoring point. After obtaining the location information, it is encoded and converted into an in-warehouse location code. The purpose of encoding is to digitize the location information so that the neural network model can process this information. The ion concentration change characteristics of the ion concentration prediction data of each monitoring point are calculated to determine the ion concentration change value of each monitoring point at different time points. These change values ​​can reveal potential trends or mutations, which are crucial for identifying anomalies. The ion concentration change values ​​are encoded to obtain the ion concentration change code so that they can be effectively processed by the deep neural network model. The in-warehouse location code and ion concentration change code of each monitoring point are input into the preset integrated cost-sensitive deep neural network model. The model includes an input layer composed of multiple base classifier models, which are designed based on cost-sensitive algorithms and can optimize the processing of different cost-sensitive problems, such as the trade-off between false positives and false negatives. After receiving the input data, each base classifier processes the data and gives the prediction result at the output layer. Through the processing of the integrated model, the position code in the bin and the ion concentration change code of each monitoring point are used to detect ion concentration anomalies. The model evaluates all input data and identifies data patterns that may indicate ion concentration anomalies based on the deep learning capabilities of the neural network.

[0033] Process and analyze the bin position code and multiple ion concentration change codes of each monitoring point. In the input layer, the bin position code and multiple ion concentration change codes of each monitoring point are vector-mapped to generate the ion concentration input vector of each monitoring point. Convert all important monitoring information into a format that the model can effectively process. Input each ion concentration input vector into multiple base classifier models configured in the model. After receiving the input vector, each base classifier independently calculates the abnormal probability of the ion concentration corresponding to the vector, through the complex calculation process inside the model, including weight adjustment, activation function processing and other standard operations of multi-layer neural networks. Each classifier responds to specific features and patterns in the data set to enhance the sensitivity and accuracy of the model to abnormal conditions. In order to integrate the outputs of multiple base classifiers and improve the reliability of decision-making, a hard voting mechanism is used to comprehensively evaluate the abnormal probabilities of each ion concentration. In the hard voting mechanism, each base classifier is regarded as one vote for the judgment of whether there is an abnormality, and the final decision on whether each monitoring point is abnormal is based on the majority vote principle. The voting mechanism simplifies the decision-making process, while taking advantage of the advantages of ensemble learning and reducing the possibility of misjudgment of a single model. The initial ion concentration anomaly detection results of all monitoring points are summarized to form the target ion concentration anomaly detection results of the entire hydroxyl ion warehouse. The result summary provides a warehouse-level anomaly monitoring view and can provide a comprehensive analysis and explanation of the abnormal status of different areas or monitoring points.

[0034] In the embodiments of the present application, by deploying a multi-channel spectrometer in the hydroxide ion chamber to collect spectral data in real time, the change of ion concentration can be continuously monitored, ensuring the timeliness and continuity of the data. Through fine spectral amplitude, shape, and information difference metrics, the minute differences between each monitoring point can be accurately identified and analyzed, enhancing the accuracy of the data and the meticulousness of the analysis. Using the partial least squares regression model to predict the ion concentration from the comprehensive spectral feature data can effectively extract the factors highly correlated with the ion concentration from the complex spectral information. The data structure is simplified through dimensionality reduction techniques while retaining the information most crucial for ion concentration prediction. This not only improves the accuracy of the prediction but also enables dynamic adjustment of the prediction model based on historical and real-time data to adapt to possible environmental changes and equipment aging. By inputting the prediction data into the integrated cost-sensitive deep neural network model for ion concentration anomaly detection, the ability to identify potential risks is enhanced, thereby improving the accuracy of ion concentration detection in the hydroxide ion chamber.

[0035] In a specific embodiment, the process of performing step S101 may specifically include the following steps:

[0036] (1) Preset multiple monitoring points in the hydroxide ion chamber and define the wavelength range and acquisition frequency of the multi-channel spectrometer;

[0037] (2) Through the multi-channel spectrometer, collect spectral data for the multiple monitoring points according to the wavelength range and acquisition frequency to obtain the first spectral data set;

[0038] (3) Remove outliers and interpolate missing values from the first spectral data set to obtain the second spectral data set;

[0039] (4) Mark data labels for the second spectral data set according to the multiple monitoring points to obtain the labeled spectral data set;

[0040] (5) Divide the labeled spectral data set to obtain the original spectral data set for each monitoring point.

[0041] Specifically, multiple monitoring points are preset in the hydroxide ion chamber. The selection of the points should be based on the uniformity and representativeness of the ion distribution in the chamber to ensure that the collected data can comprehensively reflect the ion concentration state throughout the chamber. Define the wavelength range and acquisition frequency of the multi-channel spectrometer. The setting of these parameters should take into account the characteristics of the monitoring target and the specific requirements of the monitoring environment. For example, the wavelength range should cover the characteristic spectral lines of key ion absorption or emission, and the acquisition frequency should be high enough to capture the rapid changes in ion concentration. Use the multi-channel spectrometer to collect spectral data at the monitoring points according to the set wavelength range and acquisition frequency. Collect the spectral data of each monitoring point to form the first spectral data set. Perform data cleaning on the first spectral data set, including removing outliers and interpolating missing values. Outlier removal is achieved through statistical analysis methods, such as setting thresholds or using standard deviations to judge, to exclude those non-representative or data points that may be caused by instrument errors. Missing value interpolation is achieved through various techniques such as linear interpolation, spline interpolation, or more complex machine learning methods, such as decision tree interpolation, etc., to fill in the blanks in the data set to ensure the integrity and coherence of the data, and obtain the second spectral data set. Perform data label marking on the second spectral data set. By annotating each data point with the monitoring point information, it provides the necessary background information for subsequent data analysis. The addition of labels facilitates tracking the source of each data point and also helps in grouping or classification processing in subsequent analysis. Divide the labeled spectral data set, separate the data by monitoring point, and form an independent original spectral data set for each monitoring point. By identifying the labels and assigning the data to the corresponding sets, each data set represents the spectral information of a specific monitoring point. In this way, it is ensured that when performing data analysis and model establishment, the data of each monitoring point is considered independently, thereby improving the accuracy and pertinence of the analysis.

[0042] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0043] (1) Perform curve fitting on the original spectral data set respectively to obtain the spectral curve of each monitoring point;

[0044] (2) Compare the spectral amplitudes of the spectral curves of each monitoring point to obtain the comparison result of the spectral amplitude differences between multiple monitoring points;

[0045] (3) Compare the spectral shapes of the spectral curves of each monitoring point to obtain the comparison result of the spectral shape differences between multiple monitoring points;

[0046] (4) Measure the spectral information of the spectral curves of each monitoring point to obtain the spectral information measurement differences between multiple monitoring points;

[0047] (5) According to the comparison results of spectral amplitude differences, spectral shape differences, and spectral information metrics, extract and fuse the spectral characteristic bands of the spectral curves at each monitoring point to obtain the comprehensive spectral characteristic data of each monitoring point.

[0048] Specifically, perform curve fitting on the original spectral data set respectively to obtain the spectral curves of each monitoring point. Convert the discrete data points into a continuous spectral curve through mathematical models such as polynomial fitting, Gaussian fitting, or Lorentz fitting. Curve fitting makes the data smoother, easier to analyze, and can more accurately depict the main characteristics and trends of the spectrum. Compare the spectral amplitudes of the spectral curves of each monitoring point. Quantitatively analyze the peak height and overall intensity of the curve. By comparing the spectral amplitudes of different monitoring points, the spatial differences in ion concentrations within the monitoring area can be effectively revealed. Compare the spectral shapes and observe the morphological characteristics such as the width, symmetry, and sharpness of the spectral curves at each monitoring point. Differences in spectral shapes often reflect differences in chemical compositions or the behavior of the same components under different environmental conditions. Conduct a differential analysis of spectral information metrics. Calculate metrics such as information entropy or other statistical indicators to measure the amount and quality of information contained in each spectral curve. This metric reflects the efficiency and differences in the spectral data of each monitoring point in expressing chemical information, and helps to identify which monitoring points' data may contain more chemical or physical information. According to the comparison results of spectral amplitude differences, spectral shape differences, and spectral information metrics, extract and fuse the spectral characteristic bands of the spectral curves at each monitoring point respectively. Identify the spectral regions that are significant in terms of amplitude, shape, or information metric, and extract the data from these regions for analysis and fusion. Feature extraction is achieved through signal processing techniques such as wavelet transform, Fourier transform, or machine learning-based feature selection methods. Fusing these characteristic bands requires considering their statistical importance and contribution to the prediction model, and using algorithms such as principal component analysis or linear discriminant analysis to optimize the feature combination, ultimately forming the comprehensive spectral characteristic data of each monitoring point.

[0049] In a specific embodiment, the process of performing the step of extracting and fusing the spectral characteristic bands of the spectral curves at each monitoring point according to the comparison results of spectral amplitude differences, spectral shape differences, and spectral information metrics to obtain the comprehensive spectral characteristic data of each monitoring point may specifically include the following steps:

[0050] (1) Extract the spectral characteristic bands of the spectral curves at each monitoring point respectively according to the comparison results of spectral amplitude differences to obtain multiple first spectral characteristic bands of each spectral curve;

[0051] (2)Extract the spectral characteristic bands of the spectral curve at each monitoring point according to the comparison results of the spectral shape differences, and obtain multiple second spectral characteristic bands of each spectral curve;

[0052] (3)Extract the spectral characteristic bands of the spectral curve at each monitoring point according to the spectral information metric differences, and obtain multiple third spectral characteristic bands of each spectral curve;

[0053] (4)Perform similarity screening on the multiple first spectral characteristic bands, multiple second spectral characteristic bands, and multiple third spectral characteristic bands of each spectral curve to obtain multiple target spectral characteristic bands of each spectral curve;

[0054] (5)Calculate the weights of the multiple target spectral characteristic bands of each spectral curve to obtain the characteristic weight data of each target spectral characteristic band;

[0055] (6)Perform band weighted combination on the multiple target spectral characteristic bands of each spectral curve according to the characteristic weight data to obtain the comprehensive spectral characteristic data of each monitoring point.

[0056] Specifically, according to the comparison results of spectral amplitude differences, spectral feature bands are extracted from the spectral curves of each monitoring point respectively to identify those regions with significant amplitude changes, which represent the presence and concentration changes of specific ions or chemical substances. The characteristic bands are determined by analyzing the peaks and overall energy distribution in the spectral curves to ensure that the extracted first spectral feature bands can accurately reflect the changes in concentration levels. According to the comparison results of spectral shape differences, a second round of extraction of spectral feature bands is performed on the spectral curves of each monitoring point to obtain multiple second spectral feature bands for each spectral curve. This stage focuses on the morphological features of the spectral curves, such as the width, symmetry, and sharpness of the peaks, which can reveal the molecular structure of the substance or the characteristics of its interactions. The spectral shape is analyzed to obtain important information related to chemical properties. According to the difference in spectral information metrics, a third round of extraction of feature bands is performed on the spectral curves of each monitoring point. Statistical methods or information theory are used to evaluate the information content in the spectral data. Methods such as calculating information entropy can help identify which bands contain the most information, and usually the third spectral feature bands are obtained. Similarity screening is performed on the first, second, and third spectral feature bands. By comparing the similarities of these bands, the feature bands that can represent the common features or significant differences of each monitoring point are screened out. The screening is based on algorithms such as cluster analysis or principal component analysis to ensure that the finally selected target spectral feature bands have high representativeness and distinguishability. Weight calculation is performed on the target spectral feature bands. The weights are determined according to the importance of each band in characterizing chemical properties and concentration changes. Weight calculation can use machine learning algorithms, such as the feature importance scores in support vector machines or random forests, which can effectively quantify the contribution of each band. According to the feature weight data, the target spectral feature bands of each spectral curve are weighted and combined. The combination process takes into account the individual effects of each band and also their interactions to form the comprehensive spectral feature data of each monitoring point.

[0057] In a specific embodiment, the process of performing step S103 may specifically include the following steps:

[0058] (1) Perform standard normal variable transformation on the comprehensive spectral feature data of each monitoring point to obtain the standardized spectral feature data of each monitoring point;

[0059] (2) Input the standardized spectral feature data of each monitoring point into a preset partial least squares regression model respectively, and perform data matrix transformation through the partial least squares regression model to obtain the spectral feature data matrix of each monitoring point;

[0060] (3) Calculate the scores of the spectral feature data matrix in multiple latent variable spaces and construct the score matrix of each monitoring point;

[0061] (4) Obtain the regression coefficients of the partial least squares regression model, and calculate the predicted ion concentration data for each monitoring point according to the score matrix and the regression coefficients respectively.

[0062] Specifically, perform standard normal variate transformation on the comprehensive spectral feature data of each monitoring point to make each feature data conform to the standard normal distribution. Standardization processing is helpful for subsequent statistical analysis and machine learning models. The performance of algorithms depends on the standardized form of the input data, which can reduce the influence between features with different dimensions and ensure that the influence of each feature in the model is balanced. Input the standardized spectral feature data of each monitoring point into the preset partial least squares regression model. Partial least squares regression is a statistical method for dealing with data with multicollinearity and is suitable for spectral data analysis. The partial least squares model converts the high-dimensional spectral features into a smaller number of latent variables through data matrix transformation. These latent variables are designed to capture the maximum information explaining the relationship between spectral data and target ion concentration. Calculate the scores of the spectral feature data matrix in multiple latent variable spaces. These scores reflect the performance of each monitoring point in each latent variable space of the partial least squares model, which is basically the projection of the original spectral data in the new latent variable space. The score matrix of each monitoring point shows the behavior and trend of the data in the model. Obtain the regression coefficients of the partial least squares regression model. These coefficients are obtained during the model training process and describe the specific contribution of each latent variable to the prediction of ion concentration. Use the regression coefficients to calculate the predicted ion concentration data for each monitoring point through the score matrix. Multiply the scores in the score matrix by the corresponding regression coefficients to obtain the predicted ion concentration values.

[0063] In a specific embodiment, the process of performing step S104 may specifically include the following steps:

[0064] (1) Obtain the in-warehouse position information of each monitoring point, and encode the in-warehouse position information to obtain the in-warehouse position encoding of each monitoring point;

[0065] (2) Calculate the ion concentration change characteristics for the predicted ion concentration data of each monitoring point respectively to obtain multiple ion concentration change values for each monitoring point;

[0066] (3) Encode the multiple ion concentration change values to obtain multiple ion concentration change encodings for each monitoring point;

[0067] (4) Input the in-warehouse position encoding and multiple ion concentration change encodings of each monitoring point into the preset integrated cost-sensitive deep neural network model. The preset integrated cost-sensitive deep neural network model includes an input layer, multiple base classifier models composed of cost-sensitive deep neural networks, and an output layer;

[0068] (5) Ion concentration anomaly detection is performed on the in - warehouse location encoding and multiple ion concentration change encodings of each monitoring point through an integrated cost - sensitive deep neural network model to obtain the target ion concentration anomaly detection result.

[0069] Specifically, obtain the in - warehouse location information of each monitoring point and perform effective encoding processing on this information. Location information encoding is to convert physical locations into a data format that the model can understand and process, usually using numerical encoding or one - hot encoding methods. Calculate the ion concentration change characteristics of the ion concentration prediction data for each monitoring point. Extract important statistical features from time - series data, such as change amplitude, trend, and periodicity, etc. These change values can be obtained by calculating the ion concentration difference between consecutive time points or using time - series analysis techniques. The ion concentration change characteristics reflect the dynamic change of ion concentration at each monitoring point over different time periods. Encode the obtained ion concentration change values and convert them into a format suitable for machine - learning model processing. Ion concentration change encoding can use methods similar to location encoding, such as numericalization or one - hot encoding, to ensure the uniqueness and identifiability of each change value. Through this encoding, the model can more accurately identify and learn the patterns and rules of ion concentration changes. After encoding, input the in - warehouse location encoding and ion concentration change encoding of each monitoring point into a pre - set integrated cost - sensitive deep neural network model. This model includes an input layer, a hidden layer composed of multiple base classifier models, and an output layer. Each base classifier is a cost - sensitive deep neural network that can be optimized according to the different costs of different types of errors (such as false alarms and missed alarms). Cost sensitivity is achieved by adjusting the loss function during network training or adjusting the weight distribution of the data set, making the model pay more attention to the error types with higher costs. When the input data is processed through these layers, the integrated cost - sensitive deep neural network model determines the final anomaly detection result by synthesizing the outputs of each base classifier, using techniques such as voting mechanisms or weighted averages. This integration method improves the robustness and accuracy of the model, synthesizes the judgments of multiple models, and reduces the risks that may be brought by relying on a single model. The model outputs the target ion concentration anomaly detection result for each monitoring point. These results are obtained based on the comprehensive analysis of the specific location of the monitoring point and the ion concentration changes, and can effectively indicate which monitoring points may have ion concentration anomalies.

[0070] In a specific embodiment, the process of performing step (5) of ion concentration anomaly detection on the in - warehouse location encoding and multiple ion concentration change encodings of each monitoring point through an integrated cost - sensitive deep neural network model to obtain the target ion concentration anomaly detection result can specifically include the following steps:

[0071] (1) In the input layer, vector mapping is performed on the in - silo position encoding and multiple ion concentration change encodings of each monitoring point to generate an ion concentration input vector for each monitoring point;

[0072] (2) The ion concentration input vectors of each monitoring point are respectively input into multiple base classifier models, and the ion concentration anomaly probabilities of the ion concentration input vectors are respectively calculated in the base classifier models;

[0073] (3) Using a hard voting mechanism, an abnormal comprehensive evaluation is carried out on the ion concentration anomaly probabilities to obtain the initial ion concentration anomaly detection results for each monitoring point;

[0074] (4) The initial ion concentration anomaly detection results of each monitoring point are summarized to obtain the target ion concentration anomaly detection results of the hydroxide ion silo.

[0075] Specifically, vector mapping is performed on the in-warehouse position encoding and ion concentration change encoding for each monitoring point. The categorical or sequential data is transformed into numerical vectors that can be effectively processed by machine learning algorithms. Usually, this transformation can be achieved through one-hot encoding to ensure the uniqueness and mutual exclusivity of each category, or through normalization methods to transform all features to the same scale, thereby avoiding biases caused by differences in feature dimensions during model training. After encoding, the ion concentration input vector for each monitoring point contains all the necessary information, which maps the characteristics of the position and ion concentration changes. The ion concentration input vector is input into multiple base classifier models. Each base classifier model is an independent deep neural network, and these networks are used to identify and predict abnormal ion concentrations. After receiving the input vector, each network will analyze the current data point according to the patterns learned during training and calculate the abnormal probability of the ion concentration. These calculations are based on the weights and biases of the network, and these parameters are optimized during network training through methods such as backpropagation and gradient descent to ensure that the model can accurately identify abnormal patterns. To improve the accuracy and reliability of decision-making, a hard voting mechanism is used to comprehensively evaluate the ion concentration abnormal probabilities of all base classifiers. The hard voting mechanism means that the output (abnormal or non-abnormal) of each classifier will be regarded as a vote, and the final classification decision is determined by majority voting. By combining the independent decisions of multiple models, the overall prediction performance is improved, especially when dealing with environmental data with uncertainty and variability, which can significantly reduce misjudgments that may occur in a single model. The initial ion concentration abnormal detection results for each monitoring point are summarized to obtain the target ion concentration abnormal detection results for the entire hydrogen and oxygen ion warehouse. Analyze the abnormal detection results of each monitoring point and synthesize this information to ensure that the safety status of the entire environment can be comprehensively evaluated. The comprehensive evaluation not only provides detailed information about potential problems at specific monitoring points but also reveals possible systematic risks or abnormal patterns in the entire warehouse.

[0076] The ion concentration analysis method in the embodiment of the present application is described above. Next, the ion concentration analysis system in the embodiment of the present application will be described. Please refer to Figure 2 , an embodiment of the ion concentration analysis system in the embodiment of the present application includes:

[0077] An acquisition module 201, configured to collect spectral data for multiple monitoring points in the hydrogen and oxygen ion warehouse to obtain the original spectral data set for each monitoring point;

[0078] A metric module 202, configured to respectively perform spectral amplitude, shape, and information difference metrics on the original spectral data set to obtain the comprehensive spectral feature data for each monitoring point;

[0079] A prediction module 203, configured to input the comprehensive spectral feature data of each monitoring point into a preset partial least squares regression model for ion concentration prediction, so as to obtain the ion concentration prediction data of each monitoring point;

[0080] A monitoring module 204, configured to input the ion concentration prediction data of each monitoring point into a preset integrated cost-sensitive deep neural network model for ion concentration anomaly detection, so as to obtain the target ion concentration anomaly detection result.

[0081] Through the collaborative cooperation of the above-mentioned various components, by deploying a multi-channel spectrometer in the hydroxide ion chamber to collect spectral data in real time, the change of ion concentration can be continuously monitored to ensure the timeliness and continuity of the data. Through the fine measurement of spectral amplitude, shape and information difference, the small differences between each monitoring point can be accurately identified and analyzed, enhancing the accuracy of the data and the meticulousness of the analysis. Using the partial least squares regression model to predict the ion concentration from the comprehensive spectral feature data can effectively extract the factors highly related to the ion concentration from the complex spectral information. The data structure is simplified through the dimensionality reduction technology while retaining the most critical information for ion concentration prediction. This not only improves the accuracy of the prediction, but also can dynamically adjust the prediction model based on historical and real-time data to adapt to possible environmental changes and equipment aging. By inputting the prediction data into the integrated cost-sensitive deep neural network model for ion concentration anomaly detection, the ability to identify potential risks is enhanced, thereby improving the accuracy of ion concentration detection in the hydroxide ion chamber.

[0082] This application also provides an ion concentration analysis device, which includes a memory and a processor. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, the processor executes the steps of the ion concentration analysis method in the above-mentioned various embodiments.

[0083] This application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions run on a computer, the computer executes the steps of the ion concentration analysis method.

[0084] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0085] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0086] As described above, the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of this application.

Claims

1. An ion concentration analysis method, characterized in that, The ion concentration analysis method includes: Collecting spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set for each monitoring point; Respectively performing spectral amplitude, shape, and information difference measurement on the original spectral data set to obtain the comprehensive spectral feature data for each monitoring point; Inputting the comprehensive spectral feature data of each monitoring point into a pre-set partial least squares regression model for ion concentration prediction to obtain the ion concentration prediction data for each monitoring point; Inputting the ion concentration prediction data of each monitoring point into a pre-set integrated cost-sensitive deep neural network model for ion concentration anomaly detection to obtain the target ion concentration anomaly detection result; specifically including: obtaining the in-chamber position information of each monitoring point, and encoding the in-chamber position information to obtain the in-chamber position code of each monitoring point; respectively calculating the ion concentration change characteristics of the ion concentration prediction data of each monitoring point to obtain multiple ion concentration change values of each monitoring point; encoding the multiple ion concentration change values to obtain multiple ion concentration change codes of each monitoring point; inputting the in-chamber position code and multiple ion concentration change codes of each monitoring point into a pre-set integrated cost-sensitive deep neural network model, and the pre-set integrated cost-sensitive deep neural network model includes an input layer, multiple base classifier models composed of cost-sensitive deep neural networks, and an output layer; in the input layer, performing vector mapping on the in-chamber position code and multiple ion concentration change codes of each monitoring point to generate the ion concentration input vector of each monitoring point; respectively inputting the ion concentration input vector of each monitoring point into the multiple base classifier models, and respectively calculating the ion concentration anomaly probability of the ion concentration input vector in the base classifier models; adopting a hard voting mechanism to perform anomaly comprehensive evaluation on the ion concentration anomaly probability to obtain the initial ion concentration anomaly detection result of each monitoring point; summarizing the initial ion concentration anomaly detection results of each monitoring point to obtain the target ion concentration anomaly detection result of the hydroxide ion chamber.

2. The ion concentration analysis method according to claim 1, wherein The collecting spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set for each monitoring point includes: Pre-setting multiple monitoring points in the hydroxide ion chamber, and defining the wavelength range and acquisition frequency of the multi-channel spectrometer; Using the multi-channel spectrometer to collect spectral data on the multiple monitoring points according to the wavelength range and the acquisition frequency to obtain the first spectral data set; Removing outliers and interpolating missing values from the first spectral data set to obtain the second spectral data set; Marking data labels for the second spectral data set according to the multiple monitoring points to obtain the labeled spectral data set; Dividing the labeled spectral data set to obtain the original spectral data set for each monitoring point.

3. The ion concentration analysis method according to claim 1, characterized in that The respectively performing spectral amplitude, shape, and information difference measurement on the original spectral data set to obtain the comprehensive spectral feature data for each monitoring point includes: Perform curve fitting on the original spectral data sets respectively to obtain spectral curves for each monitoring point; Perform spectral amplitude comparison on the spectral curves of each monitoring point to obtain the comparison result of spectral amplitude differences among the multiple monitoring points; Perform spectral shape comparison on the spectral curves of each monitoring point to obtain the comparison result of spectral shape differences among the multiple monitoring points; Perform spectral information measurement on the spectral curves of each monitoring point to obtain the spectral information measurement differences among the multiple monitoring points; According to the comparison result of spectral amplitude differences, the comparison result of spectral shape differences, and the spectral information measurement differences, perform spectral characteristic band extraction and fusion on the spectral curves of each monitoring point respectively to obtain the comprehensive spectral characteristic data for each monitoring point.

4. The ion concentration analysis method according to claim 3, wherein The step of performing spectral characteristic band extraction and fusion on the spectral curves of each monitoring point respectively according to the comparison result of spectral amplitude differences, the comparison result of spectral shape differences, and the spectral information measurement differences to obtain the comprehensive spectral characteristic data for each monitoring point includes: Perform spectral characteristic band extraction on the spectral curves of each monitoring point respectively according to the comparison result of spectral amplitude differences to obtain multiple first spectral characteristic bands for each spectral curve; Perform spectral characteristic band extraction on the spectral curves of each monitoring point respectively according to the comparison result of spectral shape differences to obtain multiple second spectral characteristic bands for each spectral curve; Perform spectral characteristic band extraction on the spectral curves of each monitoring point respectively according to the spectral information measurement differences to obtain multiple third spectral characteristic bands for each spectral curve; Perform similarity screening on the multiple first spectral characteristic bands, multiple second spectral characteristic bands, and multiple third spectral characteristic bands of each spectral curve to obtain multiple target spectral characteristic bands for each spectral curve; Calculate the feature weight data for each target spectral characteristic band by performing weight calculation on the multiple target spectral characteristic bands of each spectral curve; Perform band weighted combination on the multiple target spectral characteristic bands of each spectral curve according to the feature weight data to obtain the comprehensive spectral characteristic data for each monitoring point.

5. The ion concentration analysis method according to claim 1, characterized in that, The step of inputting the comprehensive spectral characteristic data of each monitoring point into a preset partial least squares regression model for ion concentration prediction to obtain the ion concentration prediction data for each monitoring point includes: Perform standard normal variate transformation on the comprehensive spectral characteristic data of each monitoring point to obtain the standardized spectral characteristic data for each monitoring point; Input the standardized spectral characteristic data of each monitoring point into the preset partial least squares regression model respectively, and perform data matrix transformation through the partial least squares regression model to obtain the spectral characteristic data matrix for each monitoring point; Calculate the scores of the spectral characteristic data matrix in multiple latent variable spaces, and construct the score matrix for each monitoring point; Obtain the regression coefficients of the partial least squares regression model, and calculate the ion concentration prediction data for each monitoring point respectively according to the score matrix and the regression coefficients.

6. An ion concentration analysis system, characterized in that, For performing the ion concentration analysis method described in any one of claims 1-5, the ion concentration analysis system includes: A collection module for collecting spectral data at multiple monitoring points in the hydroxide ion chamber to obtain the original spectral data set of each monitoring point; A metric module for respectively measuring the spectral amplitude, shape, and information difference of the original spectral data set to obtain the comprehensive spectral feature data of each monitoring point; A prediction module for inputting the comprehensive spectral feature data of each monitoring point into a preset partial least squares regression model for ion concentration prediction to obtain the ion concentration prediction data of each monitoring point; A monitoring module for inputting the ion concentration prediction data of each monitoring point into a preset integrated cost-sensitive deep neural network model for ion concentration anomaly detection to obtain the target ion concentration anomaly detection result.

7. An ion concentration analysis device, characterized in that, The ion concentration analysis device includes: a memory and at least one processor, and instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the ion concentration analysis device executes the ion concentration analysis method described in any one of claims 1-5.

8. A computer-readable storage medium having instructions stored thereon, characterized in that, When the instructions are executed by the processor, the ion concentration analysis method described in any one of claims 1-5 is implemented.

Images