Sewage Fault Diagnosis Method and Device Based on Adaptive Dynamic Slow Feature Analysis

Abstract

The present invention provides a sewage fault diagnosis method and device based on adaptive dynamic slow feature analysis. The method includes: classifying the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data; inputting the multiple clusters of online classification data into a pre-constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; determining the corresponding online dominant statistics and online inferior statistics according to the online dominant slow features and online inferior slow features respectively; comparing the size of the online dominant statistics with the dominant adaptive control limit, and comparing the size of the online inferior statistics and the inferior adaptive control limit, and determining that the system has a fault based on the corresponding statistic exceeding the corresponding adaptive control limit. Otherwise, it is determined that the operation state of the sewage treatment process is normal and there is no fault. The present invention realizes timely and accurate fault detection and reduces the losses caused by faults.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and particularly to a sewage fault diagnosis method and device based on adaptive dynamic slow feature analysis. Background Art

[0002] At present, sewage treatment plants are gradually moving towards automated and intelligent control. With the popularization and application of industrial Internet of Things and intelligent sensor technologies, sewage treatment plants can collect a large amount of sewage treatment process data, which often contains important operation information of the sewage treatment plant. With the help of technologies such as big data analysis and artificial intelligence, data-driven sewage treatment process monitoring and fault diagnosis have become a research hotspot.

[0003] Commonly used fault diagnosis methods for sewage treatment processes based on statistics mainly include: principal component analysis, canonical correlation analysis, independent component analysis, etc. However, most of the current algorithms are static monitoring algorithms, which are difficult to fully extract and process the time-varying dynamic characteristics in the sewage treatment process. In addition, the sewage treatment process has multi-stage characteristics, with different biochemical reactions and dominant parameters in different stages, resulting in significant strong dynamic characteristics of the sewage treatment process. Methods for extracting dynamic features based on time series data mainly include methods based on neural networks, etc.

[0004] However, existing research has not fully considered the stage characteristics and dynamic characteristics of sewage, resulting in often poor diagnostic effects for mild sludge bulking faults. Summary of the Invention

[0005] The present invention provides a sewage fault diagnosis method and device based on adaptive dynamic slow feature analysis, to solve the defect in the prior art that the stage characteristics and dynamic characteristics of sewage are not fully considered, resulting in poor diagnostic effects for mild sludge bulking faults, and realizes timely and accurate fault judgment.

[0006] The present invention provides a sewage fault diagnosis method based on adaptive dynamic slow feature analysis, including: classifying the online acquired sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data; inputting the multiple clusters of online classification data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features; determining the corresponding online dominant statistic and online inferior statistic according to the online dominant slow features and online inferior slow features respectively; comparing the size of the online dominant statistic with the dominant adaptive control limit, and comparing the size of the online inferior statistic and the inferior adaptive control limit, and determining that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determining that no fault exists and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature and the fixed control limit determined based on the corresponding offline slow feature.

[0007] According to the sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, inputting the multiple clusters of online classification data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model includes: inputting the multiple clusters of online classification data into a previously constructed slow feature analysis model, and respectively performing data augmentation on the classification data within each cluster by using time window delay to obtain corresponding augmented data; performing whitening processing on the augmented data to obtain whitened data; performing singular value decomposition on the whitened data to obtain a corresponding transformation matrix, and determining a corresponding slow feature according to the transformation matrix; dividing the slow features based on a preset speed threshold, taking the slow features not greater than the preset speed threshold as online dominant slow features, and taking the slow features greater than the preset speed threshold as inferior dominant slow features.

[0008] According to the sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, the online acquired sewage treatment data includes input variables collected from sewage treatment, and the input variables include dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration and alkalinity; before dividing the slow features based on a preset speed threshold, it includes: screening the slow features and removing the slow features whose speed is faster than the corresponding slow features of the input variables.

[0009] A sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, before inputting multi-cluster online classification data into a pre-constructed slow feature analysis model, includes: obtaining sewage treatment offline data; classifying the sewage treatment offline data based on the affinity propagation clustering algorithm to obtain multi-cluster offline classification data; constructing a slow feature analysis model based on the multi-cluster offline classification data, and using the slow feature analysis model to extract corresponding offline dominant slow features and offline inferior slow features; determining corresponding offline dominant statistics and offline inferior statistics according to the offline dominant slow features and offline inferior slow features, and combining with the chi-square distribution to determine corresponding dominant fixed control limits and inferior fixed control limits.

[0010] A sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, before comparing the magnitudes of the online dominant statistics and the dominant adaptive control limits, and comparing the magnitudes of the online inferior statistics and the inferior adaptive control limits, includes: determining the mean value of the offline dominant slow features based on the offline dominant slow features; determining the dominant difference norm based on the online dominant slow features and the mean value of the offline dominant slow features; determining the dominant fluctuation value of the online dominant slow features relative to the offline dominant slow features according to the dominant difference norm, the online dominant slow features, and the mean value of the offline dominant slow features; obtaining the dominant adaptive control limit according to the dominant fluctuation value and the dominant fixed control limit; determining the mean value of the offline inferior slow features based on the offline inferior slow features; determining the inferior difference norm based on the online inferior slow features and the mean value of the offline inferior slow features; determining the inferior fluctuation value of the online inferior slow features relative to the offline inferior slow features according to the inferior difference norm, the online inferior slow features, and the mean value of the offline inferior slow features; obtaining the inferior adaptive control limit according to the inferior fluctuation value and the inferior fixed control limit.

[0011] A sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, obtaining sewage treatment offline data includes: collecting sewage treatment process data, where the sewage treatment process data includes input variables and output variables, the input variables include dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration, and alkalinity, and the output variables include the concentration of suspended solids in water, biodegradable organic matter, particulate inert organic carbon, active heterotrophic bacteria, active autotrophic bacteria, degradable particulate matter, soluble biodegradable organic nitrogen, total suspended solids; collecting sludge bulking fault data, and adjusting the heterotrophic bacteria decay μH value and the heterotrophic bacteria growth bH value of the benchmark simulation model No. 1 BSM1 to produce fault sample data; obtaining sewage treatment offline data according to the sewage treatment process data and the fault sample data, and performing standardization processing on the sewage treatment offline data.

[0012] A sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention, based on the affinity propagation clustering algorithm, classifies the sewage treatment data obtained online to obtain multi-cluster online classification data, including: obtaining the similarity between any two sample points in the sewage treatment data obtained online by using the negative Euclidean distance according to the sewage treatment data obtained online; initializing the responsibility matrix and the availability matrix between any two sample points; wherein, the responsibility matrix is used to represent the suitability of considering a second sample point as a cluster center for any first sample point, and the availability matrix is used to represent the possibility of the first sample point considering the second sample point as a cluster center; for any first sample point, update the responsibility matrix according to the availability matrix between the first sample point and other sample points, the similarity between the first sample point and other sample points, and the similarity between the first sample point and the second sample point, and update the availability matrix based on the responsibility matrix of other sample points and the second sample point; use a preset damping coefficient to alternately iterate and update the responsibility matrix and the availability matrix, and end the update based on meeting the convergence condition to obtain the cluster center, and allocate the sample points in the sewage treatment data obtained online to the corresponding clusters according to the cluster center to obtain multi-cluster online classification data.

[0013] The present invention also provides a sewage fault diagnosis device based on adaptive dynamic slow feature analysis, including: a clustering module that classifies the sewage treatment data obtained online based on the affinity propagation clustering algorithm to obtain multi-cluster online classification data; a feature extraction module that inputs the multi-cluster online classification data into a pre-constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features; a statistics module that respectively determines the corresponding online dominant statistic and online inferior statistic according to the online dominant slow features and online inferior slow features; a fault judgment module that compares the size of the online dominant statistic with the dominant adaptive control limit, and compares the size of the online inferior statistic and the inferior adaptive control limit, and determines that there is a fault based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determines that there is no fault and the sewage treatment process is operating normally; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and the fixed control limit determined based on the corresponding offline slow feature.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, it implements the sewage fault diagnosis method based on adaptive dynamic slow feature analysis as described in any one of the above.

[0015] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the sewage fault diagnosis method based on adaptive dynamic slow feature analysis as described in any one of the above.

[0016] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the sewage fault diagnosis method based on adaptive dynamic slow feature analysis as described in any one of the above.

[0017] The sewage fault diagnosis method and device based on adaptive dynamic slow feature analysis provided by the present invention classify the online obtained sewage treatment data through the affinity propagation clustering algorithm, so as to directly determine a suitable number of clusters automatically based on the similarity between data without specifying the number of clusters, improve the flexibility when dealing with complex data, and adapt to different data distributions and structures; utilize the slow feature analysis model to extract slow features from the classified data, and divide the slow features into dominant slow features that are relatively stable during normal operation and inferior slow features that are prone to appear when disturbed or faulty, so as to fully extract the stage and dynamic characteristics contained in the sewage treatment data. On the basis of monitoring the sewage treatment process based on the long-term dynamics of slow features, the deviation degrees of the dominant slow features and the inferior slow features are quantified respectively through statistics, more comprehensively reflecting the changes in the state of the sewage treatment process, and by comparing the size of the statistics with the adaptive control limit, it is timely to judge whether there is a fault in the sewage treatment process, thereby improving the accuracy of fault determination, improving the sensitivity of fault diagnosis, and effectively monitoring the sewage treatment process in real time, preventing the situation that the water quality of the sewage treatment effluent exceeds the standard due to faults such as sludge bulking, and reducing the losses caused by faults. Description of the Drawings

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

[0019] Figure 1 It is a schematic flowchart of the sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention;

[0020] Figure 2 It is a change diagram of the dominant slow feature statistics during the online test process provided by the present invention;

[0021] Figure 3 It is a change diagram of the inferior slow feature statistics during the online test process provided by the present invention;

[0022] Figure 4 It is a schematic structural diagram of a sewage fault diagnosis device based on adaptive dynamic slow feature analysis provided by the present invention;

[0023] Figure 5 It is a schematic structural diagram of an electronic device provided by the present invention. Specific embodiments

[0024] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] Figure 1 It is a schematic flowchart of a sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the present invention. As Figure 1 shown, the method includes:

[0026] S11, classify the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data;

[0027] S12, input the multiple clusters of online classification data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features;

[0028] S13, respectively determine the corresponding online dominant statistic and online inferior statistic according to the online dominant slow feature and online inferior slow feature;

[0029] S14, compare the size of the online dominant statistic with the dominant adaptive control limit, and compare the size of the online inferior statistic and the inferior adaptive control limit, and determine that there is a fault based on the corresponding statistic exceeding the corresponding adaptive control limit. Otherwise, determine that there is no fault and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature and the fixed control limit determined based on the corresponding offline slow feature.

[0030] It should be noted that the step numbers "S1N" in this specification do not represent the sequence of the sewage fault diagnosis method based on adaptive dynamic slow feature analysis. The following specifically combines Figure 2-3 to describe the sewage fault diagnosis method based on adaptive dynamic slow feature analysis of the present invention.

[0031] Step S11: Classify the sewage treatment data obtained online based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data.

[0032] In this embodiment, classifying the sewage treatment data obtained online based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data includes:

[0033] S111: According to the sewage treatment data obtained online, use the negative Euclidean distance to obtain the similarity between any two sample points in the sewage treatment data obtained online.

[0034] It should be added that for any two sample points in the sewage treatment data obtained online and the similarity , is expressed as:

[0035]

[0036] S112: Initialize the responsibility matrix and the availability matrix between any two sample points; among them, the responsibility matrix is used to characterize the suitability of the second sample point as a cluster center for the first sample point, and the availability matrix is used to characterize the possibility of the first sample point considering the second sample point as a cluster center.

[0037] In this embodiment, the responsibility matrix between sample points and is expressed as r(i,k), and the availability matrix between sample points and is expressed as a(i,k).

[0038] S113: For any first sample point, update the responsibility matrix according to the availability matrix between the first sample point and other sample points, the similarity between the first sample point and other sample points, and the similarity between the first sample point and the second sample point, and update the availability matrix based on the responsibility matrix between any other sample point and the second sample point.

[0039] It should be added that the updated responsibility matrix is expressed as:

[0040]

[0041] Among them, represents the similarity between the first sample point and the second sample point , represents the availability matrix between the first sample point and other sample points , represents the first sample point and other sample points The similarity between

[0042] The updated availability matrix, expressed as:

[0043]

[0044] Where Represents the second sample point Of the responsibility matrix, which can be based on other sample points And the second sample point Of the responsibility matrix and the maximum value between 0; Represents other sample points And the second sample point Of the responsibility matrix.

[0045] S114, using the preset damping coefficient, alternately iteratively updates the responsibility matrix and the availability matrix, and based on meeting the convergence condition, ends the update, obtains the cluster center, and according to the cluster center, assigns the sample points in the online obtained sewage treatment data to the corresponding clusters to obtain multi-cluster online classification data.

[0046] In this embodiment, the responsibility matrix and the availability matrix updated by alternately iteratively using the preset damping coefficient are respectively expressed as:

[0047]

[0048] Where Represents the responsibility matrix between the first sample point And the second sample point After the m-th iteration; Represents the preset damping coefficient, which can be set based on prior experience or actual design requirements; Represents the first sample point obtained after the m-th iteration And the second sample point Between the responsibility matrix; Represents the first sample point updated after the m-th iteration And the second sample point Between the availability matrix; Represents the first sample point obtained after the m-th iteration And the second sample point Between the availability matrix.

[0049] It should be noted that the convergence condition can be set according to actual design requirements, such as matrix convergence or reaching a predetermined number of iterations. Through the affinity propagation AP clustering algorithm, the cluster center can be determined through the similarity between data points and the transmitted information to divide the collected data into different clusters to obtain multi-cluster online classification data, expressed as: , where represents the i-th sewage treatment stage, and B represents the number of stages of sewage treatment data stratification, that is, the number of clusters.

[0050] Step S12: Input multi-cluster online classification data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model. Among them, the slow feature analysis model is constructed based on sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features.

[0051] In this embodiment, inputting multi-cluster online classification data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model includes:

[0052] S121: Input multi-cluster online classification data into a previously constructed slow feature analysis model, and for the classification data within each cluster, use time window delay to perform data augmentation respectively to obtain corresponding augmented data.

[0053] It should be noted that for the data within each cluster, in order to better describe its process dynamic characteristics, the time window delay method can be used to augment all the data at the current moment and within the subsequent d moments to obtain augmented data , which is expressed as:

[0054]

[0055] S122: Perform whitening processing on the augmented data to obtain whitened data.

[0056] It should be noted that performing whitening processing on the augmented data includes: according to the augmented matrix , obtain the outer product of the augmented matrix ; perform whitening processing on the outer product of the augmented matrix, that is, calculate , where U is the eigenvector matrix of is a diagonal matrix, and the diagonal elements are the singular values of . Perform whitening processing on the augmented data through the whitening matrix Q to obtain whitened data , where represents the identity matrix of

[0057] S123: Perform singular value decomposition on the whitened data to obtain the corresponding transformation matrix, and determine the corresponding slow features according to the transformation matrix.

[0058] It should be noted that the whitened data is subjected to singular value decomposition to obtain the corresponding transformation matrix, including: performing singular value decomposition on the covariance matrix of the first-order difference of the whitened data, that is , where P is the eigenvector matrix of , is the eigenvalue of ; according to the singular value decomposition result, the transformation matrix e is obtained, where X represents the amplification matrix and W .

[0059] S124, based on a preset speed threshold, the slow features are divided. The slow features not greater than the preset speed threshold are used as online dominant slow features, and the dominant slow feature transformation matrix is W d , and the slow features greater than the preset speed threshold are used as inferior dominant slow features, and the inferior slow feature transformation matrix is W e .

[0060] It should be noted that the dominant slow features are represented as , and the inferior slow features are represented as . In addition, the dominant slow features represent the relatively stable and important features during the normal operation of sewage treatment, while the inferior slow features are more likely to be disturbed or show abnormalities first during a failure. By dividing the slow features into dominant slow features and inferior slow features, more targeted analysis can be carried out on different aspects of sewage treatment, which helps to more accurately locate the source and nature of the failure in the follow-up.

[0061] In addition, after obtaining the online dominant slow features and online inferior slow features, it also includes: using the transformation matrix corresponding to the online dominant slow features as the online dominant transformation matrix, and using the transformation matrix corresponding to the online inferior slow features as the online inferior transformation matrix; determining the number of online dominant slow features and online inferior slow features, which is specifically expressed as:

[0062]

[0063] where represents the number of inferior slow features; M represents the number of dominant slow features; represents the speed of change of the transformation of each column of slow feature data to be obtained; represents the speed of change of each column of the original amplified data; m represents the dimension of the original data; d represents the dimension of the sample to be amplified, represents the total dimension of the sample after amplification.

[0064] In an alternative embodiment, the sewage treatment data obtained online includes the input variables collected during sewage treatment, and the input variables include dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration, and alkalinity. Accordingly, before partitioning the slow features based on a preset speed threshold, it includes: screening the slow features and removing the slow features whose speed is faster than the corresponding slow features of the input variables.

[0065] In an alternative embodiment, before inputting the multi-cluster online classification data into the previously constructed slow feature analysis model, it includes: obtaining the sewage treatment offline data; classifying the sewage treatment offline data based on the affinity propagation clustering algorithm to obtain multi-cluster offline classification data; constructing a slow feature analysis model based on the multi-cluster offline classification data, and using the slow feature analysis model to extract the corresponding offline dominant slow features and offline inferior slow features; determining the corresponding offline dominant statistics and offline inferior statistics according to the offline dominant slow features and offline inferior slow features, and combining the chi-square distribution to determine the corresponding dominant fixed control limit and inferior fixed control limit.

[0066] It should be noted that for classifying the sewage treatment offline data based on the affinity propagation clustering algorithm, reference can be made to step S11; for extracting the corresponding offline dominant slow features and offline inferior slow features using the slow feature analysis model, reference can be made to step S12; no repeated elaboration will be made here.

[0067] Specifically, obtaining the sewage treatment offline data includes: collecting the sewage treatment process data, where the sewage treatment process data includes input variables and output variables, the input variables include dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration, and alkalinity, and the output variables include the concentration of suspended solids in water, biodegradable organic matter, particulate inert organic carbon, active heterotrophic bacteria, active autotrophic bacteria, degradable particulate matter, soluble biodegradable organic nitrogen, total suspended solids; collecting the sludge bulking fault data, and producing fault sample data by adjusting the heterotrophic bacteria decay μH value and heterotrophic bacteria growth bH value of the benchmark simulation model 1 BSM1; obtaining the sewage treatment offline data according to the sewage treatment process data and the fault sample data, and performing standardization processing on the sewage treatment offline data.

[0068] It should be noted that the data of the input variables and output variables can be obtained based on the BSM1 model. In addition, the standardized data is expressed as , where , represents the mean of the original data, represents the variance of the original data.

[0069] In addition, according to the slow characteristics of offline advantages and the slow characteristics of offline disadvantages, the corresponding offline advantage statistics and offline disadvantage statistics are determined respectively, and combined with the chi-square distribution, the corresponding fixed control limits for advantages and disadvantages are determined, including: for the slow characteristics of advantages and the slow characteristics of disadvantages, Hotelling T 2 statistics are constructed respectively to obtain the offline advantage statistics and the offline disadvantage statistics; based on the preset components, the offline advantage statistics and the offline disadvantage statistics, the fixed control limits for advantages and the fixed control limits for disadvantages are obtained.

[0070] It should be added that the offline advantage statistics are expressed as , and the offline disadvantage statistics are respectively expressed as , measures the static variation within the subspace spanned by the slowest main slow characteristics and is used as the statistic for the slow characteristics of advantages; while measures the static variation of the remaining fastest slow characteristics and is used as the statistic for the slow characteristics of disadvantages. In addition, according to , assuming that is an independent Gaussian distribution, then the statistic follows a distribution with degrees of freedom of , the statistic follows a distribution with degrees of freedom of , that is:

[0071]

[0072] Furthermore, the fixed control limit for advantages is expressed as:

[0073]

[0074] where represents the fixed control limit for advantages; represents the quantile corresponding to the confidence level in the chi-square distribution, represents the confidence level; represents and the proportionality coefficient, which combines the information of the slow characteristics of advantages and the Hotelling's T² statistic of the overall sample and is used to determine a scaling factor in the fixed control limit for advantages; represents and the proportionality coefficient, which, together with and , jointly determines the specific value of the control limit when determining the fixed control limit of the Hotelling's T² statistic; Represents the overall average level of the sample in terms of Hotelling's T² statistic, which is an important reference value for constructing fixed control limits; Represents the dominant statistic; n represents the sample size; Represents the variance of the sample; generally .

[0075] The inferior fixed control limit, expressed as:

[0076]

[0077] Among them, Represents the superior fixed control limit; Represents the quantile corresponding to the confidence level in the chi-square distribution, Represents the confidence level; Represents and The proportionality coefficient of, combines the information of the superior slow feature and the overall Hotelling's T² statistic of the sample, and is used to determine a scaling factor in the superior fixed control limit; Represents and The proportionality coefficient of, together with and jointly determines the specific value of the control limit when determining the fixed control limit of Hotelling's T² statistic; Represents the overall average level of the sample in terms of Hotelling's T² statistic, which is an important reference value for constructing fixed control limits; Represents the dominant statistic; n represents the sample size; Represents the variance of the sample, generally .

[0078] Step S13, according to the online superior slow feature and the online inferior slow feature, determine the corresponding online superior statistic and online inferior statistic respectively. It should be noted that the online superior statistic and the online inferior statistic can refer to the above-mentioned offline superior statistic and offline inferior statistic, and will not be repeated here. In addition, through Hotelling T 2 statistic, quantify the deviation degree of each superior slow feature and inferior slow feature, considering the mutual correlation between multiple features, so as to more comprehensively reflect the change of the sewage treatment state and more sensitively detect the abnormal situation in the sewage treatment process.

[0079] Step S14: Compare the magnitudes of the online advantage statistic and the advantage adaptive control limit, and compare the magnitudes of the online disadvantage statistic and the disadvantage adaptive control limit. Based on the corresponding statistic exceeding the corresponding adaptive control limit, determine that a fault exists; otherwise, determine that no fault exists and the operation state of the sewage treatment process is normal. Among them, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and the fixed control limit determined based on the corresponding offline slow feature.

[0080] It should be noted that by comparing the magnitudes of the statistic and the adaptive control limit, a clear judgment criterion is provided to reduce the uncertainty of subjective judgment, thereby accurately determining whether a fault exists in the system. If the statistic exceeds the control limit, it indicates that the current state of the sewage treatment process has deviated from the normal range and the possibility of a fault is relatively high, improving the accuracy of fault determination.

[0081] In an optional embodiment, before comparing the magnitudes of the online advantage statistic and the advantage adaptive control limit, and comparing the magnitudes of the online disadvantage statistic and the disadvantage adaptive control limit, it includes: determining the mean value of the offline advantage slow feature based on the offline advantage slow feature; determining the advantage difference norm based on the online advantage slow feature and the mean value of the offline advantage slow feature; determining the advantage fluctuation value of the online advantage slow feature relative to the offline advantage slow feature according to the advantage difference norm, the online advantage slow feature, and the mean value of the offline advantage slow feature; and obtaining the advantage adaptive control limit according to the advantage fluctuation value and the advantage fixed control limit.

[0082] Similarly, before comparing the magnitudes of the online advantage statistic and the advantage adaptive control limit, and comparing the magnitudes of the online disadvantage statistic and the disadvantage adaptive control limit, it also includes: determining the mean value of the offline disadvantage slow feature based on the offline disadvantage slow feature; determining the disadvantage difference norm based on the online disadvantage slow feature and the mean value of the offline disadvantage slow feature; determining the disadvantage fluctuation value of the online disadvantage slow feature relative to the offline disadvantage slow feature according to the disadvantage difference norm, the online disadvantage slow feature, and the mean value of the offline disadvantage slow feature; and obtaining the disadvantage adaptive control limit according to the disadvantage fluctuation value and the disadvantage fixed control limit.

[0083] It is worth noting that by considering the deviation of the average value of the online slow feature and the offline slow feature as a weight and combining the fixed control limit calculated in the offline process, the control limit is adaptively adjusted to better reflect the difference between the current state and the normal state of the sewage treatment process, enabling the fault detection to better adapt to different working conditions and environments.

[0084] It should be added that the difference norm is expressed as:

[0085]

[0086] where represents the corresponding online slow feature; represents the corresponding offline slow feature mean; represents the difference norm; represents the corresponding offline slow feature; t represents the sampling time, t ∈ 1, …, m.

[0087] In addition, at the t-th sampling, the fluctuation value of the online slow feature relative to the corresponding offline slow feature is expressed as:

[0088]

[0089] When a mutation fault occurs, , then , represents the th variable, so . Similarly, when the online data is all normal and less than the mean of the normal data, . Therefore, the adaptive control limit can be set according to the adaptive coefficient .

[0090] It should be noted that if the finally calculated is the dominant adaptive control limit, the slow feature in the above formula is the corresponding dominant slow feature, otherwise it is the corresponding inferior slow feature.

[0091] In addition, the dominant adaptive control limit and the inferior adaptive control limit are respectively expressed as:

[0092]

[0093] In an alternative embodiment, the method further includes: evaluating the performance of fault detection using metrics such as accuracy, fault detection rate, and false alarm rate, and the accuracy, fault detection rate, and false alarm rate metrics are respectively expressed as:

[0094]

[0095] where TP (True Positives) represents true positives, that is, the number of positive examples correctly judged as positive or negative; FP (False Positives) represents false positives, that is, the number of negative examples misjudged as positive; FN (False Negatives) represents false negatives, that is, the number of positive examples misjudged as negative; TN (True Negatives) represents true negatives, that is, the number of negative examples correctly judged as negative.

[0096] In an alternative embodiment, in order to better verify the effect of the present invention in industrial process fault detection, a dataset obtained from the simulation platform of the Benchmark Simulation Model No. 1 (BSM1) developed by the International Water Association is selected to verify the effect of the present invention.

[0097] Sludge bulking is a common type of fault in the sewage treatment process. When sludge bulking occurs, the sedimentation and compression characteristics of the sludge deteriorate, the suspended solids increase, the sludge loss is serious, and the biochemical system may malfunction, directly threatening the normal operation of the sewage treatment system. Toxic shock fault is caused by the discharge of a large amount of toxic sewage, resulting in a reduction in the microbial activity of the activated sludge and even the failure of the sewage biochemical reaction system. Retention fault occurs when the normal growth of heterotrophic organisms decreases. In this experiment, different degrees of retention faults can be generated by adjusting the heterotrophic bacteria growth rate μH and the heterotrophic decay coefficient bH parameters. Table 1 describes the sludge bulking fault among the 6 types of fault types involved in the experiment of this study. Table 2 provides the sewage process parameters of some sewage treatment process variables.

[0098] During the implementation process, 14 days of offline data were collected based on the BSM1 simulation platform as training data, and a total of 1344 samples were collected. The online data used was the data of BSM1 operating in the normal state for 2 days under three different working conditions (DRY, RAIN, STORM), and then continuously operating for 7 days under the fault conditions. A total of 864 samples were collected for each fault type, and the occurrence time of the fault was between the 192nd and 864th moments.

[0099] Table 1 Fault data types generated by BSM1

[0100]

[0101] Table 2 Variable names collected by BSM1

[0102]

[0103] The stage division results based on the AP clustering algorithm under three different working conditions are shown in Table 3. The stage division results basically correspond to the two anoxic tanks and three aerobic tanks in the biochemical reaction tank, which also illustrates the effectiveness of the stage division results.

[0104] Table 3 Stage division results of the sewage treatment plant

[0105]

[0106] The statistics of the superior and inferior slow features are respectively compared with the adaptive control limits. When the statistic exceeds the corresponding adaptive control limit, it indicates that there is a fault in the process; otherwise, the operating state of the process is considered to be in the normal state.

[0107] Figure 2 and Figure 3 are the monitoring results of the sludge bulking fault in the sewage treatment plant based on the multi-level adaptive dynamic slow feature analysis model proposed by the present invention, where Figure 2is the fault detection result in the dominant slow feature subspace, Figure 3 is the fault detection result in the inferior slow feature. From Figure 2 and Figure 3 In the experimental results, during the online test phase, the sewage treatment plant operated normally in the first two days, and then the statistic exceeded the adaptive threshold, indicating that a fault occurred in the sewage treatment process. It can be seen from the figure that the model developed by the present invention based on multi-level adaptive dynamic slow feature analysis has relatively superior performance.

[0108] In summary, the embodiment of the present invention classifies the online obtained sewage treatment data through the affinity propagation clustering algorithm, so as to directly determine the appropriate number of clusters automatically based on the similarity between the data without specifying the number of clusters, improving the flexibility in processing complex data to adapt to different data distributions and structures; using the slow feature analysis model to extract slow features from the classified data, and dividing the slow features into dominant slow features that are relatively stable during normal operation and inferior slow features that are prone to appear when disturbed or faulty, so as to fully extract the stage and dynamic characteristics contained in the sewage treatment data. On the basis of long-term dynamic monitoring of the sewage treatment process based on slow features, the deviation degrees of the dominant slow features and the inferior slow features are quantified respectively through statistics, more comprehensively reflecting the changes in the state of the sewage treatment process, and by comparing the size of the statistic with the adaptive control limit, it is timely judged whether there is a fault in the sewage treatment process, thereby improving the accuracy of fault determination, improving the sensitivity of fault diagnosis, and effectively monitoring the sewage treatment process in real time, preventing the situation that the effluent quality of sewage treatment exceeds the standard due to faults such as sludge bulking, and reducing the losses caused by faults.

[0109] Next, the sewage fault diagnosis device based on adaptive dynamic slow feature analysis provided by the present invention will be described. The sewage fault diagnosis device based on adaptive dynamic slow feature analysis described below can be mutually referred to the sewage fault diagnosis method based on adaptive dynamic slow feature analysis described above.

[0110] Figure 4 FIG. shows a schematic structural diagram of a sewage fault diagnosis device based on adaptive dynamic slow feature analysis. The device includes:

[0111] A clustering module 41 that classifies the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classified data;

[0112] A feature extraction module 42 that inputs the multiple clusters of online classified data into a pre-constructed slow feature analysis model to obtain online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features;

[0113] A statistical module 43 determines corresponding online advantage statistics and online disadvantage statistics according to the online advantage slow features and the online disadvantage slow features respectively.

[0114] A fault judgment module 44 compares the magnitude of the online advantage statistics with the advantage adaptive control limit, and compares the magnitude of the online disadvantage statistics and the disadvantage adaptive control limit, and determines that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determines that no fault exists and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and a fixed control limit determined based on the corresponding offline slow feature.

[0115] In this embodiment, the clustering module 41 includes: a similarity determination unit that obtains the similarity between any two sample points in the online acquired sewage treatment data by using the negative Euclidean distance according to the online acquired sewage treatment data; an initialization unit that initializes the responsibility matrix and the availability matrix between any two sample points; wherein, the responsibility matrix is used to represent the suitability of considering the second sample point as the cluster center for the first sample point, and the availability matrix is used to represent the possibility of the first sample point considering the second sample point as the cluster center; a matrix update unit that updates the responsibility matrix according to the availability matrix between the first sample point and other sample points, the similarity between the first sample point and other sample points, and the similarity between the first sample point and the second sample point for any first sample point, and updates the availability matrix based on the responsibility matrix between any other sample point and the second sample point; a clustering unit that alternately iteratively updates the responsibility matrix and the availability matrix by using a preset damping coefficient, and ends the update based on meeting the convergence condition to obtain the cluster center, and assigns the sample points in the online acquired sewage treatment data to the corresponding clusters according to the cluster center to obtain multi-cluster online classification data.

[0116] The feature extraction module 42 includes: a data augmentation unit that inputs the multi-cluster online classification data into a previously constructed slow feature analysis model to perform data augmentation on the classification data within each cluster by using time window delay respectively to obtain corresponding augmented data; a whitening processing unit that performs whitening processing on the augmented data to obtain whitened data; a feature extraction unit that performs singular value decomposition on the whitened data to obtain a corresponding transformation matrix, and determines the corresponding slow feature according to the transformation matrix; a feature division unit that divides the slow features based on a preset speed threshold, takes the slow features not greater than the preset speed threshold as online advantage slow features, and takes the slow features greater than the preset speed threshold as disadvantage advantage slow features.

[0117] In an alternative embodiment, the feature extraction module 42 further includes: a matrix partitioning unit, which, after obtaining the online dominant slow features and the online inferior slow features, uses the transformation matrix corresponding to the online dominant slow features as the online dominant transformation matrix and the transformation matrix corresponding to the online inferior slow features as the online inferior transformation matrix; and a quantity determination unit, which determines the quantities of the online dominant slow features and the online inferior slow features.

[0118] In an alternative embodiment, the feature extraction module 42 further includes: a feature screening unit, which screens the slow features before partitioning the slow features based on a preset speed threshold, and eliminates the slow features whose speeds are faster than the slow features corresponding to the input variables.

[0119] In an alternative embodiment, the apparatus further includes: an offline data acquisition module, which acquires sewage treatment offline data before inputting the multi-cluster online classification data into a previously constructed slow feature analysis model; a clustering module, which classifies the sewage treatment offline data based on the affinity propagation clustering algorithm to obtain multi-cluster offline classification data; a feature extraction module, which constructs a slow feature analysis model based on the multi-cluster offline classification data and uses the slow feature analysis model to extract corresponding offline dominant slow features and offline inferior slow features; and a statistics module, which respectively determines corresponding offline dominant statistics and offline inferior statistics based on the offline dominant slow features and the offline inferior slow features, and determines corresponding dominant fixed control limits and inferior fixed control limits in combination with the chi-square distribution.

[0120] Specifically, the offline data acquisition module includes: a first data acquisition unit, which acquires sewage treatment process data, where the sewage treatment process data includes input variables and output variables, and the input variables include at least one of dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration, and alkalinity, and the output variables include at least one of the concentration of suspended solids in water, biodegradable organic matter, particulate inert organic carbon, active heterotrophic bacteria, active autotrophic bacteria, degradable particulate matter, soluble biodegradable organic nitrogen, and total suspended solids; a second data acquisition unit, which acquires sludge bulking fault data and adjusts the heterotrophic bacteria decay μH value and the heterotrophic bacteria growth bH value of the benchmark simulation model No. 1 BSM1 to produce fault sample data; and a standardization processing unit, which obtains the sewage treatment offline data based on the sewage treatment process data and the fault sample data and performs standardization processing on the sewage treatment offline data.

[0121] In addition, the statistics module includes: a statistics unit, which respectively constructs Hotelling T 2 statistics for the dominant slow features and the inferior slow features to obtain offline dominant statistics and offline inferior statistics; and a control limit determination unit, which obtains dominant fixed control limits and inferior fixed control limits based on preset components and the offline dominant statistics and the offline inferior statistics.

[0122] In an alternative embodiment, the apparatus further includes: a mean determination module configured to determine an offline dominant slow feature mean based on an offline dominant slow feature before comparing the magnitude of an online dominant statistic with a dominant adaptive control limit and comparing the magnitude of an online inferior statistic with an inferior adaptive control limit; a norm determination module configured to determine a dominant difference norm based on the online dominant slow feature and the offline dominant slow feature mean; a fluctuation determination module configured to determine a dominant fluctuation value of the online dominant slow feature relative to the offline dominant slow feature according to the dominant difference norm, the online dominant slow feature, and the offline dominant slow feature mean; and a control limit adjustment module configured to obtain a dominant adaptive control limit according to the dominant fluctuation value and a dominant fixed control limit.

[0123] Similarly, the apparatus further includes: a mean determination module configured to determine an offline inferior slow feature mean based on an offline inferior slow feature before comparing the magnitude of an online dominant statistic with a dominant adaptive control limit and comparing the magnitude of an online inferior statistic with an inferior adaptive control limit; a norm determination module configured to determine an inferior difference norm based on the online inferior slow feature and the offline inferior slow feature mean; a fluctuation determination module configured to determine an inferior fluctuation value of the online inferior slow feature relative to the offline inferior slow feature according to the inferior difference norm, the online inferior slow feature, and the offline inferior slow feature mean; and a control limit adjustment module configured to obtain an inferior adaptive control limit according to the inferior fluctuation value and an inferior fixed control limit.

[0124] In an alternative embodiment, the apparatus further includes: an evaluation module configured to evaluate the performance of fault detection by using indicators such as accuracy rate, fault detection rate, and false alarm rate.

[0125] In summary, in the embodiment of the present invention, the clustering module uses the affinity propagation clustering algorithm to classify the online obtained sewage treatment data, so as to directly and automatically determine an appropriate number of clusters based on the similarity between the data without specifying the number of clusters, improving the flexibility in processing complex data to adapt to different data distributions and structures; the feature extraction module uses the slow feature analysis model to extract slow features from the classified data, and divides the slow features into dominant slow features that are relatively stable during normal operation and inferior slow features that are likely to appear when disturbed or faulty, so as to fully extract the periodic and dynamic characteristics contained in the sewage treatment data. On the basis of long-term dynamic monitoring of the sewage treatment process based on slow features, the statistical module quantifies the deviation degrees of the dominant slow features and the inferior slow features respectively, more comprehensively reflecting the changes in the state of the sewage treatment process, and the fault judgment module compares the magnitudes of the statistics with the adaptive control limits to timely determine whether there is a fault in the sewage treatment process, thereby improving the accuracy of fault determination, enhancing the sensitivity of fault diagnosis, effectively performing real-time monitoring on the sewage treatment process, preventing the situation that the effluent quality of sewage treatment exceeds the standard due to faults such as sludge bulking, and reducing the losses caused by faults.

[0126] Figure 5 An entity structure schematic diagram of an electronic device is exemplified, as Figure 5 shown. The electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communication bus 540. Among them, the processor 510, the communications interface 520, and the memory 530 complete mutual communication through the communication bus 540. The processor 510 may call the logical instructions in the memory 530 to execute a sewage fault diagnosis method based on adaptive dynamic slow feature analysis. The method includes: classifying the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classification data; inputting the multiple clusters of online classification data into a pre-constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features; determining the corresponding online dominant statistic and online inferior statistic respectively according to the online dominant slow features and online inferior slow features; comparing the size of the online dominant statistic with the dominant adaptive control limit, and comparing the size of the online inferior statistic and the inferior adaptive control limit, and determining that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determining that no fault exists and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and the fixed control limit determined based on the corresponding offline slow feature.

[0127] In addition, when the logical instructions in the above-mentioned memory 530 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The 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 the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0128] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the above-mentioned various methods. The method includes: classifying the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classified data; inputting the multiple clusters of online classified data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features; determining the corresponding online dominant statistic and online inferior statistic respectively according to the online dominant slow features and online inferior slow features; comparing the size of the online dominant statistic with the dominant adaptive control limit, and comparing the size of the online inferior statistic and the inferior adaptive control limit, and determining that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determining that no fault exists and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and a fixed control limit determined based on the corresponding offline slow feature.

[0129] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the sewage fault diagnosis method based on adaptive dynamic slow feature analysis provided by the above-mentioned various methods. The method includes: classifying the online obtained sewage treatment data based on the affinity propagation clustering algorithm to obtain multiple clusters of online classified data; inputting the multiple clusters of online classified data into a previously constructed slow feature analysis model to obtain the online dominant slow features and online inferior slow features output by the slow feature analysis model; wherein, the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline dominant slow features and offline inferior slow features; determining the corresponding online dominant statistic and online inferior statistic respectively according to the online dominant slow features and online inferior slow features; comparing the size of the online dominant statistic with the dominant adaptive control limit, and comparing the size of the online inferior statistic and the inferior adaptive control limit, and determining that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determining that no fault exists and the operation state of the sewage treatment process is normal; wherein, the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature, and a fixed control limit determined based on the corresponding offline slow feature.

[0130] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention 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 equivalently replace some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A sewage fault diagnosis method based on adaptive dynamic slow feature analysis, characterized in that: include: Based on the neighbor propagation clustering algorithm, the sewage treatment data obtained online are classified to obtain multiple clusters of online classification data; Inputting the multi-cluster online classification data into the previously constructed slow feature analysis model to obtain the online superior slow features and the online inferior slow features output by the slow feature analysis model; wherein the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline superior slow features and offline inferior slow features; According to the online advantage slow feature and the online disadvantage slow feature, respectively determine the corresponding online advantage statistics and online disadvantage statistics; Compare the sizes of the online advantage statistic and the advantage adaptive control limit, and compare the sizes of the online disadvantage statistic and the disadvantage adaptive control limit, and determine that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, determine that there is no fault and the sewage treatment process is operating normally; wherein the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature and the fixed control limit determined based on the corresponding offline slow feature.

2. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 1 is characterized in that: Inputting the multi-cluster online classification data into the previously constructed slow feature analysis model to obtain online superior slow features and online inferior slow features output by the slow feature analysis model includes: Inputting the multi-cluster online classification data into the previously constructed slow feature analysis model, so as to perform data amplification using time window delay for the classification data in each cluster, respectively, to obtain corresponding amplified data; Performing whitening processing on the amplified data to obtain whitened data; Performing singular value decomposition on the whitened data to obtain a corresponding transformation matrix, and determining a corresponding slow feature according to the transformation matrix; Based on a preset speed threshold, the slow features are divided, and the slow features not greater than the preset speed threshold are taken as online dominant slow features, and the slow features greater than the preset speed threshold are taken as inferior dominant slow features.

3. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 2 is characterized in that: The sewage treatment data obtained online includes input variables collected during sewage treatment, and the input variables include dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration and alkalinity; Before the slow feature is divided based on the preset speed threshold, the method includes: The slow features are screened to eliminate slow features that are faster than the slow features corresponding to the input variables.

4. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 1 is characterized in that: Before inputting the multi-cluster online classification data into the previously constructed slow feature analysis model, the process includes: Obtain offline data of sewage treatment; Based on the neighbor propagation clustering algorithm, the sewage treatment offline data is classified to obtain multiple clusters of offline classified data; Based on the multi-cluster offline classification data, a slow feature analysis model is constructed, and the corresponding offline advantage slow features and offline disadvantage slow features are extracted using the slow feature analysis model; According to the offline advantage slow feature and the offline disadvantage slow feature, the corresponding offline advantage statistics and offline disadvantage statistics are determined respectively, and in combination with the chi-square distribution, the corresponding advantage fixed control limits and disadvantage fixed control limits are determined.

5. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 4 is characterized in that: Before comparing the online superiority statistic with the superiority adaptive control limit, and comparing the online inferiority statistic with the inferiority adaptive control limit, the method includes: Based on the offline dominant slow feature, determining the offline dominant slow feature mean; Determining a dominant difference norm based on the online dominant slow feature and the mean of the offline dominant slow feature; Determine, according to the advantage difference norm, the online advantage slow feature and the mean of the offline advantage slow feature, the advantage fluctuation value of the online advantage slow feature relative to the offline advantage slow feature; Obtaining a dominant adaptive control limit according to the dominant fluctuation value and the dominant fixed control limit; Based on the offline disadvantage slow feature, determining the offline disadvantage slow feature mean; Determining a disadvantage difference norm based on the online disadvantage slow feature and the offline disadvantage slow feature mean; Determining a disadvantage fluctuation value of the online disadvantage slow feature relative to the offline disadvantage slow feature according to the disadvantage difference norm, the online disadvantage slow feature and the mean of the offline disadvantage slow feature; According to the inferior fluctuation value and the inferior fixed control limit, the inferior adaptive control limit is obtained.

6. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 4 is characterized in that: The obtaining of offline sewage treatment data includes: Collecting sewage treatment process data, the sewage treatment process data including input variables and output variables, the input variables including dissolved oxygen, nitrate and nitrite, ammonia nitrogen, ion concentration and alkalinity, the output variables including the concentration of suspended solids in water, biodegradable organic matter, particulate inert organic carbon, active heterotrophic bacteria, active autotrophic bacteria, degradable particulate matter, soluble biodegradable organic nitrogen, and total suspended solids; Collect sludge bulking failure data, and produce failure sample data by adjusting the heterotrophic bacteria decay μH value and heterotrophic bacteria growth bH value of the benchmark simulation model No. 1 BSM1; According to the sewage treatment process data and the fault sample data, sewage treatment offline data is obtained, and the sewage treatment offline data is standardized.

7. The sewage fault diagnosis method based on adaptive dynamic slow feature analysis according to claim 1 is characterized in that: The sewage treatment data obtained online is classified based on the nearest neighbor propagation clustering algorithm to obtain multiple clusters of online classified data, including: According to the sewage treatment data obtained online, using negative Euclidean distance to obtain the similarity between any two sample points in the sewage treatment data obtained online; Initialize the responsibility matrix and availability matrix between any two sample points; wherein the responsibility matrix is ​​used to characterize the suitability of considering the second sample point as the cluster center for any first sample point, and the availability matrix is ​​used to characterize the possibility that the first sample point considers the second sample point as the cluster center; For any of the first sample points, the responsibility matrix is ​​updated according to the availability matrix between the first sample point and other sample points, the similarity between the first sample point and the other sample points, and the similarity between the first sample point and the second sample point, and the availability matrix is ​​updated based on the responsibility matrix between the other sample points and the second sample point; Using the preset damping coefficient, the responsibility matrix and the availability matrix are updated alternately and iteratively, and based on meeting the convergence condition, the update is terminated to obtain the cluster center, and according to the cluster center, the sample points in the sewage treatment data obtained online are allocated to the corresponding clusters to obtain multi-cluster online classification data.

8. A sewage fault diagnosis device based on adaptive dynamic slow feature analysis, characterized in that: include: The clustering module classifies the sewage treatment data obtained online based on the nearest neighbor propagation clustering algorithm to obtain multiple clusters of online classified data; A feature extraction module, inputting the multi-cluster online classification data into a previously constructed slow feature analysis model, and obtaining online superior slow features and online inferior slow features output by the slow feature analysis model; wherein the slow feature analysis model is constructed based on the sewage treatment offline data and its corresponding offline superior slow features and offline inferior slow features; A statistical module, which determines corresponding online advantage statistics and online disadvantage statistics according to the online advantage slow feature and the online disadvantage slow feature; A fault judgment module compares the size of the online advantage statistic with the advantage adaptive control limit, and compares the size of the online disadvantage statistic with the disadvantage adaptive control limit, and determines that a fault exists based on the corresponding statistic exceeding the corresponding adaptive control limit, otherwise, it is determined that there is no fault and the sewage treatment process is operating normally; wherein the adaptive control limit is obtained based on the corresponding online slow feature, the corresponding offline slow feature and the fixed control limit determined based on the corresponding offline slow feature.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the sewage fault diagnosis method based on adaptive dynamic slow feature analysis as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the sewage fault diagnosis method based on adaptive dynamic slow feature analysis as described in any one of claims 1 to 7 is implemented.

Images