Purification system real-time monitoring and repair early warning method and system
By clustering and optimizing distance calculations of air quality monitoring data, the problem of large amount of calculation and low accuracy of the air quality monitoring system under high accuracy requirements is solved, and the accuracy and reliability of air quality monitoring are improved.
Patent Information
- Application Number
- CN202411463302.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-21
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-10-21
AI Technical Summary
Under the requirements of high accuracy, the increase in the frequency and dimension of the air monitoring data acquisition data has led to a large amount of calculations, and there are problems of repeated redundant calculations and low accuracy during detection of the LOF algorithm.
By clustering the air quality monitoring data, the state dispersion and trend characteristic values of each cluster class are determined, the distance calculation between cluster classes is optimized, and the LOF algorithm is improved for abnormal detection.
It improves the accuracy and robustness of outlier data point detection, and improves the accuracy and reliability of air quality monitoring.
Smart Images

Figure CN118981735A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of air purification processing, and in particular to a purification system real-time monitoring and repair warning method and system. Background Art
[0002] Factory purification projects require that factories must more strictly consider the impact of the production environment in their overall planning and design. For example, pharmaceutical workshops, electronic factories, food factories and other purification workshops must comply with clean production requirements to ensure product quality. Therefore, configuring an air purification system in the factory to purify the production environment is an important part of ensuring stable product quality. During the operation of the air purification system, it is necessary to collect a variety of air monitoring data at each moment, and use the LOF (Local Outlier Factor) algorithm to detect outliers in these air monitoring data to detect abnormal air quality and issue repair warnings for the air purification system.
[0003] When the monitoring accuracy requirement is high, it is often necessary to increase the collection frequency and data dimension of air monitoring data. At this time, a huge amount of air monitoring data will be generated. When using the LOF algorithm to detect outliers, it is necessary to traverse a large amount of data and there are a lot of repeated and redundant calculations, which will lead to a large amount of calculation. In order to reduce the amount of calculation, the air monitoring data can be clustered first, and then each cluster is used as a benchmark for outlier detection. However, due to the influence of the distribution characteristics of cluster data, when the distance between the center points of two clusters is used as the distance measurement feature, it is often impossible to characterize the true distance between the clusters as a whole, so that outliers cannot be accurately detected, resulting in low accuracy of air quality monitoring. Summary of the invention
[0004] The object of the present invention is to provide a purification system real-time monitoring and repair warning method and system, which are used to solve the problem of low accuracy of existing air quality monitoring.
[0005] In order to solve the above technical problems, the present invention provides a purification system real-time monitoring and repair warning method, comprising the following steps: Acquire air quality monitoring data at each set time, wherein the air quality monitoring data includes at least two types of air quality parameters; Constructing data points corresponding to the air quality monitoring data, clustering all data points to obtain at least two clusters and their cluster centers, and determining the state discreteness corresponding to each cluster according to the number of data points in each cluster, location distribution characteristics, and adjacent conditions at a set time; Project the data points in any two clusters onto a straight line passing through the class center points of the two clusters to determine the projection points of the data points in any two clusters, and determine the trend characteristic value between any two clusters according to the position distribution of the projection points, the position relationship between the class center points of any two clusters and their projection points, and the distribution of the surrounding neighboring data points of each data point in each cluster; Determine the Euclidean distance between any two clusters, improve the Euclidean distance according to the trend characteristic value between any two clusters and the state discreteness corresponding to any two clusters, determine the improved distance value between any two clusters, and perform anomaly detection on the air quality monitoring data based on the improved distance value to determine the air quality state.
[0006] Furthermore, the state discreteness corresponding to each cluster is determined, including: Determine the distance between each data point in each cluster and the class center point of the cluster and the distance between any two data points in each cluster; According to the set time corresponding to each data point in each cluster, determine the number of time-series adjacent data points corresponding to each data point in each cluster; The state discreteness corresponding to each cluster is determined based on the number of data points in each cluster, the discreteness of the distance from each data point to the class center point of the cluster, the maximum value of the distance between any two data points, and the number of time-series adjacent data points corresponding to each data point.
[0007] Furthermore, the state discreteness corresponding to each cluster is determined, and the corresponding calculation formula is: ;in, Indicates The state dispersion corresponding to each cluster; Indicates The first Data points to The distance between the center points of the clusters; Indicates Each data point in the cluster to the The mean distance between the cluster centers; Indicates The first The number of temporally adjacent data points corresponding to a data point; Indicates The maximum value of the distance between any two data points in the clusters; Indicates The set consisting of the distances between any two data points in a cluster; Indicates The number of data points in each cluster; represents a normalized function; e represents a natural constant.
[0008] Furthermore, the trend characteristic values between any two clusters are determined, including: The projection points corresponding to any two clusters and located between the projection points of the class center points of the two clusters are determined as the target projection points corresponding to any two clusters; Determine the center position of the projection point of the class center points of any two clusters, and determine the distance from each target projection point corresponding to any two clusters to the center position, so as to obtain a first distance of each target projection point corresponding to any two clusters; Determine the distance from the class center point of the cluster to which the data point corresponding to each target projection point corresponding to any two clusters belongs to the center position, thereby obtaining a second distance of each target projection point corresponding to any two clusters; According to the distribution of neighboring data points around each pixel in each cluster, the local data point density of each target projection point corresponding to any two clusters is determined; According to the first distance, the second distance and the local data point density of each target projection point corresponding to any two clusters, and the proportion of the number of target projection points corresponding to any two clusters in all projection points, the trend characteristic value between any two clusters is determined.
[0009] Furthermore, the trend characteristic value between any two clusters is determined, and the corresponding calculation formula is: ;in, Indicates Clusters and The trend characteristic values between clusters; Indicates Clusters and The cluster class corresponds to The first distance of the target projection point; Indicates Clusters and The cluster class corresponds to The second distance of the target projection point; Indicates Clusters and The cluster class corresponds to The local data point density of each target projection point; Indicates Clusters and The total number of target projection points corresponding to each cluster; Indicates Clusters and The proportion of target projection points corresponding to each cluster in all projection points; Indicates Clusters and The number of all projection points corresponding to the clusters; e represents a natural constant.
[0010] Furthermore, the improved distance value between any two clusters is determined, and the corresponding calculation formula is: ;in, Indicates Clusters and Improved distance value between clusters; Indicates Clusters and The Euclidean distance between the cluster centers; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates Clusters and Normalized result of negative correlation of eigenvalues between clusters.
[0011] Furthermore, based on the improved distance value between any two clusters, the LOF algorithm is used to perform anomaly detection on the air quality monitoring data.
[0012] Further, constructing data points corresponding to the air quality monitoring data includes: Each type of air quality parameter contained in the air quality monitoring data is determined as a coordinate value on a coordinate axis corresponding to a sample space, and sample points determined in the sample space by each coordinate value corresponding to the air quality monitoring data are determined as data points corresponding to the air quality monitoring data.
[0013] Furthermore, based on the silhouette coefficient, The optimal number of clusters in the algorithm, and using The algorithm clusters all data points to obtain the optimal number of clusters.
[0014] To solve the above technical problems, the present invention also provides a purification system real-time monitoring and repair warning system, including a processor and a memory, wherein the processor is used to process computer instructions stored in the memory to implement the steps of a purification system real-time monitoring and repair warning method as described in any of the above items.
[0015] The present invention has the following beneficial effects: the present invention constructs data points corresponding to the air quality monitoring data at each set time, and clusters all the data points to obtain multiple clusters, each of which can be regarded as an air quality state. By analyzing the discrete conditions between the data points in each cluster, the state discreteness corresponding to each cluster is determined. At the same time, combined with the distribution characteristics of the data points of any two clusters, the trend characteristics of the data points of any two clusters are quantified to measure the possibility that the two clusters represent the same air quality state, thereby obtaining the trend characteristic value between any two clusters. According to the state discreteness corresponding to each cluster and the trend characteristic value between two clusters, the distance between different clusters is optimized, thereby determining a more accurate distance between different clusters, effectively improving the detection accuracy and robustness of outlier data points, and thus improving the accuracy and reliability of air quality monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0017] Figure 1 The figure is a flow chart of a method for real-time monitoring and repair warning of a purification system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. In addition, all parameters or indicators in the formulas involved in this article are normalized values that eliminate the influence of dimensions.
[0020] Purification system real-time monitoring and repair warning method embodiment: In order to solve the problem of low accuracy of existing air quality monitoring, this embodiment provides a purification system real-time monitoring and repair warning method, the corresponding flow chart of the method is as follows: Figure 1 As shown, the following steps are included: Step S1: Acquire air quality monitoring data at each set time, wherein the air quality monitoring data includes at least two types of air quality parameters.
[0021] During the operation of the purification system, in order to better monitor the quality of the purified air, it is necessary to collect air quality monitoring data that reflects the air quality. Air quality monitoring data includes various gas components such as oxygen and carbon dioxide, as well as various types of air quality parameters such as temperature and humidity. When collecting air quality monitoring data, various types of air quality parameters are synchronously collected according to the set sampling frequency, and each collection moment is taken as a set moment, so that various types of air quality parameters at each set moment in the most recent time period in the past can be obtained in real time.
[0022] Step S2: construct data points corresponding to the air quality monitoring data, cluster all data points to obtain at least two clusters and their cluster centers, and determine the state discreteness corresponding to each cluster according to the number of data points in each cluster, location distribution characteristics and adjacent conditions at a set time.
[0023] For the air quality monitoring data at each set time, that is, various types of air quality parameters, corresponding data points are constructed, that is, each type of air quality parameter is used as the coordinate value on the corresponding coordinate axis of the sample space, and all sample points determined by the coordinate values in the sample space are determined as corresponding data points.
[0024] After obtaining each data point, due to the large number of data points and high dimensionality, if the LOF algorithm is directly used to detect outlier data points, the amount of calculation will be too high. However, considering that the same type of data points often correspond to a certain air quality state, in order to reduce the amount of calculation to a large extent and monitor the air quality state more reasonably, all data points can be clustered first, and data point outlier detection can be performed on each cluster that represents a certain air quality state, thereby ultimately realizing the monitoring of the purified air quality.
[0025] Based on the above analysis, the Euclidean distance between any two data points in the sample space is determined, and it is used as the distance measurement feature between any two data points. All data points in the sample space are clustered to obtain multiple clusters and the center point of each cluster. When clustering, this embodiment adopts The (k-means clustering) algorithm clusters all data points in the sample space. The specific process is: determine the The optimal number of clusters in the algorithm is The specific determination steps belong to the prior art and will not be described here. Based on the determined optimal number of clusters , for all data points in the sample space Clustering, thus obtaining Clusters and There are multiple data points in each cluster, and each cluster represents a certain air quality state. It should be understood that since the subsequent detection of outlier data points is based on clusters, too few clusters may lead to inaccurate detection results. Therefore, the number of clusters is limited here. To limit the minimum value, if the contour coefficient is obtained When the value is less than or equal to the set cluster number threshold, The value is set to the set cluster quantity threshold. The specific value of the set cluster quantity threshold can be reasonably set according to needs. In this embodiment, the value of the set cluster quantity threshold is set to 20.
[0026] When each cluster is used as the object in the LOF algorithm, the conventional idea is to use the center point of each cluster as the standard, and use the Euclidean distance between the cluster centers as the distance between clusters, and then perform the LOF algorithm. However, in practical applications, when the Euclidean distance between the cluster centers of two clusters is used as the distance measurement feature, it is often unable to characterize the actual differences between the clusters as a whole. For example, different air quality states correspond to different distributions of data points. The corresponding air quality monitoring data in the cluster may be relatively more chaotic, that is, it does not tend to one of the air quality states mentioned above, but because it is relatively discrete, most of the data points are forced to be distributed together. For these clusters, when calculating the distance measurement features between clusters, adjustments need to be made to improve the detection accuracy and robustness of outlier clusters, thereby improving the accuracy of air quality monitoring.
[0027] Based on the above analysis, the discrete distribution degree of the data points in each cluster is analyzed to characterize the aggregation degree of the current cluster. At the same time, based on the above-mentioned fact that each cluster should correspond to a certain control quality state, and the state should often last for a period of time, the state discreteness of the distribution of data points in each cluster is determined, so as to obtain the state discreteness corresponding to each cluster. The corresponding calculation formula is: ;in, Indicates The state dispersion corresponding to each cluster; Indicates The first Data points to The distance between the center points of the clusters; Indicates Each data point in the cluster to the The mean distance between the cluster centers; Indicates The first The number of temporally adjacent data points corresponding to a data point; Indicates The maximum value of the distance between any two data points in the clusters; Indicates The set consisting of the distances between any two data points in a cluster; Indicates The number of data points in each cluster; represents a normalized function; e represents a natural constant.
[0028] In the above-mentioned In the calculation formula of the state discreteness corresponding to the cluster class, by calculating the Each data point in the cluster to The difference between the distance of the class center point of a cluster and the mean distance can represent the The overall difference degree of the distance of the data points in the clusters is higher. The higher the overall difference degree of the distance, the more discrete the distribution of the data points within the cluster, and the greater the state dispersion of the corresponding cluster. The number of temporal adjacent data points corresponding to each data point in the cluster is the number of temporal adjacent data points in the first The total number of consecutive data points at the set time corresponding to each data point in the cluster is calculated, and the exponential function is used to perform negative correlation mapping on the number of adjacent data points in the time series. Since the position of the data point in the sample space and the distance between the data points are essentially determined by the numerical value of the multi-dimensional air quality parameters themselves, The more time-series adjacent data points in a cluster, the longer the duration of the change in the numerical value of the data point is, and the closer the corresponding cluster is to a certain air quality state. Correspondingly, for data points with a small number of time-series adjacent data points in the cluster, we are more inclined to believe that these data points are some discrete abnormal data points. Therefore, when judging whether the cluster mainly represents a certain air quality state, we prefer to give a higher weight to the distance difference determined by these discrete abnormal data points and thus the degree of discreteness. By determining the The farthest distance between two data points in a cluster. The farther the distance is, the more discrete the overall distribution of the data points within the cluster is, which also indicates that the distribution range of the data points constituting the cluster is larger. Under the same cluster conditions, The wider the distribution of a cluster, the worse its overall aggregation becomes, that is, the less it resembles a cluster characterized by a specific air quality state. The state dispersion corresponding to each cluster The larger the value of , the worse the degree to which the multiple data points constituting the cluster show a similar air quality state, and the greater the degree of dispersion between its internal data points. Therefore, when the cluster participates in the subsequent LOF algorithm calculation, it is more likely that the distance between the cluster and other clusters is often greater than the actual distance when calculating the distance.
[0029] Step S3: Project the data points in any two clusters onto the straight line passing through the class center points of the two clusters to determine the projection points of the data points in any two clusters, and determine the trend characteristic value between any two clusters based on the position distribution of the projection points, the positional relationship between the class center points of any two clusters and their projection points, and the distribution of the surrounding neighboring data points of each data point in each cluster.
[0030] In the LOF algorithm, although each cluster can be calculated as a whole, due to the surrounding discrete points, when clustering is performed in order to consider the silhouette coefficient, it is usually affected by the surrounding outliers, resulting in a dense area being divided into multiple clusters. However, for these clusters, at this time, only the distance is quantified based on the center represented by the mean of the data points in the cluster, which will cause the distance value to be too large or too small, and thus cannot accurately represent the true distance between clusters.
[0031] Therefore, for any two clusters, a straight line is drawn through the class center points of the two clusters, and a perpendicular line is drawn from each data point in the cluster to the straight line. The foot point of the perpendicular line is the projection point, so as to obtain the projection point of each data point in the two clusters. Then, the projection points between the projection points of the class center points of the two clusters are determined as the target projection points corresponding to the two clusters. The center position of the projection point of the class center points of the two clusters is determined, and the distance from each target projection point corresponding to the two clusters to the center position is determined, so as to obtain the first distance of each target projection point corresponding to the two clusters. The distance from the class center point of the cluster to which the data point corresponding to each target projection point corresponding to the two clusters belongs to the center position is determined, so as to obtain the second distance of each target projection point corresponding to the cluster. With the data point corresponding to each target projection point corresponding to the two clusters as the center point, a circular local neighborhood is determined, and the ratio of the number of data points in the local neighborhood to the area of the local neighborhood is used as the local data point density, so as to obtain the local data point density of each target projection point corresponding to the two clusters. The radius of the local neighborhood can be reasonably set according to actual needs. In this embodiment, the radius of the local neighborhood is set to 1 / 20 of the maximum value of the distance between any two data points in the cluster where the data point corresponding to the target projection point is located. For example, when the cluster where the data point corresponding to the target projection point is located is When there are clusters, the radius of the local neighborhood is On this basis, according to the first distance, second distance and local data point density of each target projection point corresponding to any two clusters, and the proportion of the number of target projection points corresponding to any two clusters in all projection points, the trend characteristic value between any two clusters is determined, and the corresponding calculation formula is: ;in, Indicates Clusters and The trend characteristic values between clusters; Indicates Clusters and The cluster class corresponds to The first distance of the target projection point; Indicates Clusters and The cluster class corresponds to The second distance of the target projection point; Indicates Clusters and The cluster class corresponds to The local data point density of each target projection point; Indicates Clusters and The total number of target projection points corresponding to each cluster; Indicates Clusters and The proportion of target projection points corresponding to each cluster in all projection points; Indicates Clusters and The number of all projection points corresponding to the clusters; e represents a natural constant.
[0032] In the above-mentioned Clusters and In the calculation formula of the trend characteristic value between clusters, by calculating the ratio of the first distance and the second distance of each target projection point, when the ratio is smaller, that is, the target projection point is closer to the center position of the projection point corresponding to the class center point of the two clusters, it means that the trend characteristics in the two clusters are higher, and the air quality states corresponding to the two clusters are closer. At the same time, by calculating the local data point density of each target projection point, and using the local data point density as the weight of the above ratio part, when the local data point density is larger, the weight of the target projection point in calculating the distance relative to the center position is higher, and vice versa, when the local data point density is smaller, the corresponding weight is smaller. Considering that the local data point density may correspond to an outlier data point when it is small, the exponential function is used to map the local data point density. Since the exponential function changes faster when the independent variable is small, it can ensure that the smaller the local data point density, the smaller the corresponding weight is, and the weight corresponding to the smaller local data point density can be weakened to a greater extent. By calculating the proportion of the target projection points corresponding to the two clusters in all projection points, the higher the proportion, the stronger the trend characteristics between the two clusters, and the more likely it is that the same air quality state is divided into multiple parts due to outliers.
[0033] Step S4: Determine the Euclidean distance between any two clusters, improve the Euclidean distance according to the trend characteristic value between any two clusters and the state discreteness corresponding to any two clusters, determine the improved distance value between any two clusters, and based on the improved distance value, perform anomaly detection on the air quality monitoring data to determine the air quality state.
[0034] Based on the state discreteness corresponding to each cluster determined in the above steps, and the trend characteristic values corresponding to any two clusters, the Euclidean distance between the two clusters can be adjusted to obtain a more accurate distance characteristic representation between the two clusters. When the state discreteness corresponding to a cluster is higher, we think that the distance between the cluster and the other clusters should be farther than the Euclidean distance obtained based on the cluster center point. At the same time, when the trend characteristic value between the two clusters is higher, it is believed that even if the state discreteness obtained above is higher, there are certain trend characteristics between the two clusters. However, due to the limitation of the number of clusters and some outliers, they are finally divided into multiple clusters. Therefore, the Euclidean distance between the two clusters needs to be optimized.
[0035] In order to optimize the Euclidean distance between any two clusters, the state dispersion corresponding to each cluster is normalized by positive correlation, so that the value range of the positive correlation normalization result of the state dispersion is , the greater the state dispersion, the closer the value of the positive correlation normalization result of the state dispersion is to 1. At the same time, the negative correlation normalization of the trend characteristic value between any two clusters is performed, so that the value range of the negative correlation normalization result of the trend characteristic value is At this time, the larger the trend eigenvalue is, the closer the value of the negative correlation normalized result of the trend eigenvalue is to -1.
[0036] According to the negative correlation normalization result of the trend characteristic value between any two clusters and the positive correlation normalization result of the state discreteness corresponding to any two clusters, the Euclidean distance between the class centers of any two clusters is improved to obtain a more accurate improved distance value between any two clusters. The corresponding calculation formula is: ;in, Indicates Clusters and Improved distance value between clusters; Indicates Clusters and The Euclidean distance between the cluster centers; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates Clusters and Normalized result of negative correlation of eigenvalues between clusters.
[0037] In the above-mentioned Clusters and In the calculation formula of the improved distance value between clusters, Clusters and The average value of the positive correlation normalized results of the state dispersion corresponding to the clusters is added to the negative correlation normalized results of the trend eigenvalues corresponding to the two clusters to obtain the hyperparameter , the hyperparameter The value range is , and using this hyperparameter The Euclidean distance between the center points of the two clusters is optimized, and the range of the improved distance value between the two clusters is .
[0038] Based on the improved distance value obtained after optimization between any two clusters determined in the above manner, the LOF algorithm is used to detect anomalies in the air quality monitoring data. That is, based on the improved distance value obtained after optimization between any two clusters, the LOF algorithm is used to detect outlier clusters. When no outlier clusters are detected, the air quality state is determined to be normal, which means that the purification system is working normally. When an outlier cluster is detected, the air quality monitoring data corresponding to the data points contained in the outlier cluster is used as abnormal monitoring data, and the air quality state is determined to be abnormal. At this time, it means that the purification system is very likely to have a fault, and a repair warning is issued to the purification system.
[0039] Purification system real-time monitoring and repair warning system implementation example: In order to solve the problem of low accuracy of existing air quality monitoring, this embodiment also provides a purification system real-time monitoring and repair warning system, which includes a processor and a memory, and the processor is used to process computer instructions stored in the memory to implement the steps of a purification system real-time monitoring and repair warning method. Since the system is essentially a soft system, its core function is to implement a purification system real-time monitoring and repair warning method. Since the method has been described in detail in the above-mentioned embodiment of a purification system real-time monitoring and repair warning method, the system will not be described here.
[0040] The present invention constructs data points corresponding to the air quality monitoring data at each set time, and clusters these data points to obtain clusters. Then, taking each cluster as an object, according to the discrete distribution characteristics of the data points within each cluster, the distribution trend characteristics of the data points between clusters are quantified, and the distances between different clusters are optimized accordingly, thereby improving the detection accuracy and robustness of outlier data, and further improving the accuracy and reliability of air quality monitoring.
[0041] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A purification system real-time monitoring and repair warning method, characterized in that: The following steps are involved: Acquire air quality monitoring data at each set time, wherein the air quality monitoring data includes at least two types of air quality parameters; Constructing data points corresponding to the air quality monitoring data, clustering all data points to obtain at least two clusters and their cluster centers, and determining the state discreteness corresponding to each cluster according to the number of data points in each cluster, location distribution characteristics, and adjacent conditions at a set time; Project the data points in any two clusters onto a straight line passing through the class center points of the two clusters to determine the projection points of the data points in any two clusters, and determine the trend characteristic value between any two clusters according to the position distribution of the projection points, the position relationship between the class center points of any two clusters and their projection points, and the distribution of the surrounding neighboring data points of each data point in each cluster; Determine the Euclidean distance between any two clusters, improve the Euclidean distance according to the trend characteristic value between any two clusters and the state discreteness corresponding to any two clusters, determine the improved distance value between any two clusters, and based on the improved distance value, perform anomaly detection on the air quality monitoring data to determine the air quality state.
2. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Determine the state discreteness corresponding to each cluster, including: Determine the distance between each data point in each cluster and the class center point of the cluster and the distance between any two data points in each cluster; According to the set time corresponding to each data point in each cluster, determine the number of time-series adjacent data points corresponding to each data point in each cluster; The state discreteness corresponding to each cluster is determined based on the number of data points in each cluster, the discreteness of the distance from each data point to the class center point of the cluster, the maximum value of the distance between any two data points, and the number of time-series adjacent data points corresponding to each data point.
3. A purification system real-time monitoring and repair warning method according to claim 2, characterized in that: Determine the state discreteness corresponding to each cluster class, and the corresponding calculation formula is: ;in, Indicates The state dispersion corresponding to each cluster; Indicates The first Data points to The distance between the center points of the clusters; Indicates Each data point in the cluster to the The mean distance between the center points of the clusters; Indicates The first The number of temporally adjacent data points corresponding to a data point; Indicates The maximum value of the distance between any two data points in the clusters; Indicates The set consisting of the distances between any two data points in a cluster; Indicates The number of data points in each cluster; represents a normalized function; e represents a natural constant.
4. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Determine the trend characteristic values between any two clusters, including: The projection points corresponding to any two clusters and located between the projection points of the class center points of the two clusters are determined as the target projection points corresponding to any two clusters; Determine the center position of the projection point of the class center points of any two clusters, and determine the distance from each target projection point corresponding to any two clusters to the center position, so as to obtain a first distance of each target projection point corresponding to any two clusters; Determine the distance from the class center point of the cluster to which the data point corresponding to each target projection point corresponding to any two clusters belongs to the center position, thereby obtaining a second distance of each target projection point corresponding to any two clusters; According to the distribution of neighboring data points around each pixel in each cluster, the local data point density of each target projection point corresponding to any two clusters is determined; According to the first distance, the second distance and the local data point density of each target projection point corresponding to any two clusters, and the proportion of the number of target projection points corresponding to any two clusters in all projection points, the trend characteristic value between any two clusters is determined.
5. A purification system real-time monitoring and repair warning method according to claim 4, characterized in that: Determine the trend characteristic value between any two clusters. The corresponding calculation formula is: ;in, Indicates Clusters and The trend characteristic values between clusters; Indicates Clusters and The cluster class corresponds to The first distance of the target projection point; Indicates Clusters and The cluster class corresponds to The second distance of the target projection point; Indicates Clusters and The cluster class corresponds to The local data point density of each target projection point; Indicates Clusters and The total number of target projection points corresponding to each cluster; Indicates Clusters and The proportion of target projection points corresponding to each cluster in all projection points; Indicates Clusters and The number of all projection points corresponding to the clusters; e represents a natural constant.
6. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Determine the improved distance value between any two clusters. The corresponding calculation formula is: ;in, Indicates Clusters and Improved distance value between clusters; Indicates Clusters and The Euclidean distance between the cluster centers; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates The positive correlation normalization result of the state discreteness corresponding to each cluster; Indicates Clusters and Normalized result of negative correlation of eigenvalues between clusters.
7. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Based on the improved distance value between any two clusters, the LOF algorithm is used to detect anomalies in air quality monitoring data.
8. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Constructing data points corresponding to the air quality monitoring data, including: Each type of air quality parameter contained in the air quality monitoring data is determined as a coordinate value on a coordinate axis corresponding to a sample space, and a sample point determined in the sample space by each coordinate value corresponding to the air quality monitoring data is determined as a data point corresponding to the air quality monitoring data.
9. A purification system real-time monitoring and repair warning method according to claim 1, characterized in that: Determined based on the silhouette coefficient The optimal number of clusters in the algorithm, and using The algorithm clusters all data points to obtain the optimal number of clusters.
10. A purification system real-time monitoring and repair warning system, characterized in that: It comprises a processor and a memory, wherein the processor is used to process computer instructions stored in the memory to implement the steps of a purification system real-time monitoring and repair warning method as described in any one of claims 1 to 9.
Citation Information
Patent Citations
Intelligent cable life prediction method and system based on data driving
CN116361679A
Anesthesia depth monitoring method and system based on artificial intelligence
CN118000683A
Intelligent analysis method for monitoring data of disinfection supply center
CN118051796A
MEA-BP neural network-based WSN abnormality detection method
WO2018126984A2
Cited By
Diagnostic control method for energy-saving operation of industrial air conditioning system
CN119374209A
A diagnostic control method for energy-saving operation of industrial air-conditioning systems
CN119374209B
