An environmental data monitoring system for battery recycling and discharge crushing

Abstract

This invention relates to the field of battery recycling monitoring technology, specifically to an environmental data monitoring system for battery discharge and breakage during recycling. The system includes: a data acquisition module for obtaining sequences corresponding to each dimension based on environmental data from different dimensions within the battery breakage equipment; a sample space construction module for filtering dimensions to obtain the dimensions to be analyzed and constructing a sample space; a clustering analysis module for clustering data points within the sample space to obtain clusters; determining whether data points within each cluster are suspected anomalous data points; an anomaly degree acquisition module for clustering all suspected anomalous data points in time series to obtain time-series clusters and acquire the anomaly degree; and a monitoring module for acquiring the comprehensive anomaly degree and determining whether battery discharge and breakage is abnormal based on the comprehensive anomaly degree. This application can accurately monitor the discharge and breakage process of old batteries.

Description

Technical Field

[0001] This invention relates to the field of battery recycling monitoring technology, and more specifically to an environmental data monitoring system for battery recycling due to discharge and breakage. Background Technology

[0002] The recycling of used batteries is of great significance for both environmental protection and energy regeneration. Since used batteries may still contain residual charge, they typically need to be discharged and crushed to ultimately recover materials and valuable metals. Monitoring environmental data is essential during the battery recycling, discharging, and crushing process. This is crucial to prevent air pollution from waste gases generated during the crushing process, and also because thermal reactions can occur during discharge and crushing, especially with lithium batteries posing a risk of thermal runaway. Furthermore, fires caused by thermal explosions during battery crushing can ignite rapidly and spread pollutants quickly; therefore, monitoring environmental data within the battery crushing equipment is necessary.

[0003] Existing technologies for monitoring environmental data during the battery recycling, discharge, and crushing process typically employ threshold monitoring methods. These methods have limited monitoring dimensions, are insensitive to changes in equipment operating conditions, fail to detect risks of abnormal trends in environmental data, and have poor timeliness. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention aims to provide an environmental data monitoring system for the discharge and breakage of batteries used in recycling. The specific technical solution adopted is as follows:

[0005] One embodiment of the present invention provides an environmental data monitoring system for battery recycling discharge crushing, the system comprising:

[0006] The data acquisition module is used to collect environmental data of different dimensions inside the battery crushing equipment and preprocess it to obtain the sequence corresponding to each dimension.

[0007] The sample space construction module is used to calculate the sample feature degree of the corresponding dimension of a sequence based on the peak and trough points in the sequence; to filter the dimensions according to the sample feature degree to obtain the dimensions to be analyzed, and to construct the sample space.

[0008] The clustering analysis module is used to group the elements of the sequence at a certain time point in each dimension to be analyzed into a data point, and to cluster the data points in the sample space to obtain a cluster; based on the distance between a cluster and other clusters and the distance between the cluster and the zero point of the sample space, it is determined whether the data points in the cluster are suspected abnormal data points.

[0009] The anomaly degree acquisition module is used to cluster suspected anomaly data points based on the time interval between every two suspected anomaly data points to obtain time-series clusters; and to obtain the anomaly degree based on each time-series cluster.

[0010] The monitoring module is used to obtain the comprehensive degree of anomaly based on the degree of anomaly and the data of each dimension to be analyzed in the suspected abnormal data points, and to determine whether the battery discharge breakage is abnormal based on the comprehensive degree of anomaly.

[0011] Preferably, the sequence corresponding to each dimension is obtained, including:

[0012] The environmental data for each dimension is normalized, and the normalized environmental data for a dimension forms the sequence corresponding to that dimension.

[0013] Preferably, calculating the sample feature degree of a sequence in a corresponding dimension based on the peaks and troughs in the sequence includes:

[0014] Obtain the peaks and troughs in a sequence and arrange them into a fluctuation point sequence according to the time sequence; obtain the absolute value of the difference between the corresponding ordinate values ​​between every two adjacent fluctuation points in the fluctuation point sequence, and sum them to obtain the fluctuation change feature value; multiply the number of elements in the sequence, the fluctuation change feature value, and the difference between the maximum and minimum values ​​of the sequence to obtain the sample feature degree of the corresponding dimension of the sequence.

[0015] Preferably, the dimensions to be analyzed are selected by filtering the dimensions according to the degree of sample characteristics, and a sample space is constructed, including:

[0016] Obtain a preset number of dimensions with the highest degree of sample feature as the dimensions to be analyzed; construct the sample space using the dimensions to be analyzed.

[0017] Preferably, determining whether a data point within a cluster is a suspected outlier based on the distance between a cluster and other clusters, as well as the distance between the cluster and the zero point of the sample space, includes:

[0018] Obtain the distance between the cluster center of a cluster and the zero point of the sample space and normalize it to obtain the first distance; obtain the mean of the distances between the cluster center of this cluster and the cluster centers of all other clusters and normalize it to obtain the cluster distribution characteristics; the product of the first distance and the cluster distribution characteristics is the suspected anomaly degree of the cluster; based on the suspected anomaly degree of a cluster, determine whether the data points in the cluster are suspected anomalous data points.

[0019] Preferably, determining whether a data point within a cluster is a suspected anomalous data point based on the degree of suspected anomaly of the cluster includes:

[0020] If the suspected anomaly level of a cluster is greater than or equal to the first reference threshold, then the data points within that cluster are suspected anomaly data points.

[0021] Preferably, the degree of anomaly is obtained based on each time-series cluster, including:

[0022] The ratio of suspected outlier data points to the total number of data points is recorded as the outlier percentage. The time-series clusters are arranged in chronological order to obtain a time-series cluster arrangement sequence. Within this sequence, the mean time interval between the two suspected outlier data points with the closest time interval in each adjacent time-series cluster is calculated and normalized to obtain the inter-cluster distribution characteristic value. The mean time interval between each suspected outlier data point within a time-series cluster and the suspected outlier data point with the smallest time interval is calculated and recorded as the intra-cluster distribution characteristic value of that cluster. A negative correlation mapping is performed on the mean of the intra-cluster distribution characteristic values ​​of all time-series clusters using an exponential function with the natural constant base to obtain the cumulative intra-cluster distribution characteristic value. The cumulative intra-cluster distribution characteristic value is added to the inter-cluster distribution characteristic value and multiplied by the outlier percentage to obtain the degree of anomaly.

[0023] Preferably, the comprehensive anomaly level is obtained based on the degree of anomaly and the data of each dimension to be analyzed in the suspected anomaly data points, including:

[0024] Calculate and normalize the mean of the data corresponding to one of the dimensions to be analyzed among all suspected outlier data points, and denot it as the mean of the suspected outlier data for that dimension to be analyzed. Calculate the standard deviation of the mean of the suspected outlier data for all dimensions to be analyzed, and multiply it by the normalized degree of outlier to obtain the comprehensive degree of outlier.

[0025] Preferably, determining whether battery discharge breakage is abnormal based on the overall degree of abnormality includes:

[0026] When the overall abnormality level is greater than or equal to the second reference threshold, the battery discharge breakage is abnormal.

[0027] The embodiments of the present invention have at least the following beneficial effects: This application collects and analyzes environmental data from different dimensions within the battery crushing equipment, which can increase the amount of information and thus obtain more accurate analysis results; furthermore, it calculates the sample feature degree of each dimension for dimension screening, achieving dimensionality reduction while ensuring the importance of the selected dimension's environmental data for battery crushing and recycling monitoring, thereby ensuring the accuracy of monitoring; then, it performs cluster analysis on multiple data points composed of elements at each time step in the sequence corresponding to each dimension to be analyzed, determines whether the data points within the cluster are suspected abnormal data points based on the spatial distribution characteristics of the clusters, obtains the degree of abnormality based on the distribution characteristics of the suspected abnormal data points in the time domain, and obtains the comprehensive degree of abnormality by combining the characteristics of the data of each dimension to be analyzed in the suspected abnormal data points, thereby monitoring whether abnormalities occur in battery discharge crushing, capturing potential environmental anomalies early, improving timeliness, and obtaining more accurate environmental anomaly monitoring results within the battery crushing equipment during the battery discharge crushing process based on high timeliness. Attached Figure Description

[0028] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a system block diagram of an environmental data monitoring system for battery recycling and discharge-induced breakage, provided as an embodiment of the present invention. Detailed Implementation

[0030] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an environmental data monitoring system for discharge-induced breakdown of batteries according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0032] The following description, in conjunction with the accompanying drawings, details the specific scheme of the environmental data monitoring system for battery recycling discharge and crushing provided by the present invention.

[0033] Example: The main application scenario of this invention is as follows: During the battery recycling, discharging, and crushing process, it is necessary to monitor the environmental data within the battery crushing equipment and determine whether there are any abnormal environmental data that could cause abnormal gas concentrations or fire risks based on the monitored environmental data. Because the main temperature changes and gas releases occur during the crushing process, they are primarily monitored within the battery crushing equipment.

[0034] Please see Figure 1 The diagram illustrates a system block diagram of an environmental data monitoring system for battery recycling and discharge breakage according to an embodiment of the present invention. The system includes the following modules:

[0035] The data acquisition module is used to collect environmental data of different dimensions inside the battery crushing equipment and preprocess it to obtain the sequence corresponding to each dimension.

[0036] The discharge and crushing of old batteries is carried out inside the battery crushing equipment. Therefore, it is necessary to install temperature sensors, humidity sensors, and various gas concentration sensors inside the battery crushing equipment to collect environmental data from different dimensions. This allows for monitoring of the temperature and humidity of the equipment environment during the battery crushing process, as well as the concentration of various gases that may be generated during battery crushing, such as combustible gases H2 and CH4, and toxic gases HF, CO, and VOC.

[0037] To enable subsequent multidimensional data correlation analysis, all sensors are configured to monitor at the same frequency of 10Hz. The collected environmental data is then transmitted to a data processing terminal. Further, environmental data from the previous 10 seconds at each real-time monitoring point is acquired to monitor the situation at that point. The implementer can adjust the duration of the environmental data collection period based on actual conditions. Because different dimensions have different magnitudes, the collected values ​​for each dimension's environmental data need to be normalized. The specific normalization model is as follows:

[0038] x' represents the normalized data value, x max The maximum value in the current dimension of environmental data, i.e., the monitoring threshold, x min is the minimum value of the environmental data for the current dimension, i.e., the initial monitoring parameters of the equipment before crushing begins, such as exhaust gas concentration of 0, and temperature and humidity as actual air temperature and humidity. x is the data value before normalization. Therefore, the environmental data for each dimension is normalized, and the normalized environmental data for a dimension constitutes the sequence corresponding to that dimension.

[0039] The sample space construction module is used to calculate the sample feature degree of the corresponding dimension of a sequence based on the peak and trough points in the sequence; to filter the dimensions according to the sample feature degree to obtain the dimensions to be analyzed, and to construct the sample space.

[0040] Environmental anomalies are not solely caused by extreme changes in a single data point. For example, a temperature rise to a threshold indicates an abnormal fire risk, and an increase in exhaust gas concentration to a threshold indicates abnormal gas pollution. However, even before each individual environmental data point reaches its monitoring threshold, simultaneous changes in multiple environmental data dimensions also suggest potential environmental anomalies. For instance, if a change in temperature is accompanied by a change in gas concentration, the combined effect of exhaust gas and high temperature could lead to an explosion, potentially escalating into a fire. Similarly, if a change in gas concentration is accompanied by an increase in temperature, the rising temperature could cause the battery to break apart and release more exhaust gases. Furthermore, some battery materials (such as lithium metal and electrolyte) react violently with water, releasing hydrogen or heat. Therefore, abnormal changes in humidity also increase the likelihood of abnormal environmental monitoring values. Thus, simultaneous abnormal changes in multi-dimensional environmental data indicate a higher probability of environmental anomalies.

[0041] Therefore, this paper synthesizes multidimensional environmental data to construct a multidimensional sample space. Because with excessively high dimensionality, data points tend to be closer together, weakening the distance relationships between them and leading to inaccurate feature analysis results, dimensionality reduction is necessary to construct a multidimensional sample space. For sequences corresponding to dimensions with low fluctuation frequency and amplitude, there is less extractable feature information. Conversely, for datasets with high fluctuation frequency and amplitude, there is more extractable feature information, making them more suitable as sample spaces for subsequent analysis. Furthermore, the larger the difference between the maximum and minimum values ​​within a dimensional dataset, the more likely the data in that dimension is to exhibit large-scale anomalous changes, making it a more suitable indicator for anomaly feature analysis.

[0042] Furthermore, based on the peaks and troughs in a sequence, the sample feature degree of the corresponding dimension of the sequence is calculated. Specifically, the peaks and troughs in a sequence are obtained and arranged into a fluctuation point sequence in chronological order; the absolute value of the difference between the corresponding ordinate values ​​between every two adjacent fluctuation points in the fluctuation point sequence is obtained and summed to obtain the fluctuation change feature value; the number of elements in the sequence, the fluctuation change feature value, and the difference between the maximum and minimum values ​​of the sequence are multiplied to obtain the sample feature degree of the corresponding dimension of the sequence.

[0043] The specific calculation model is as follows:

[0044] Among them, R q n represents the degree of sample feature in the q-th dimension; pThis represents the number of peaks and troughs in the sequence of the q-th dimension, which is also the number of points in the fluctuation point sequence. The more peaks and troughs there are, the greater the frequency of environmental data fluctuations within that dimension, and the stronger the sample characteristics. Peaks and troughs are obtained using the AMPD peak finding algorithm; A i Let be the amplitude of the i-th fluctuation point in the fluctuation point sequence, that is, the vertical span from this fluctuation point to the next fluctuation point, expressed as the absolute value of the difference between the ordinates of two adjacent fluctuation points in the fluctuation point sequence. This represents the characteristic value of fluctuation change, summing the amplitudes of all fluctuations. The larger the cumulative sum of these amplitudes, the greater the overall level of fluctuation amplitude in the current dimension, and the stronger the sample characteristic. max Let x' be the maximum value in the normalized environmental data within the sequence corresponding to the q-th dimension. min x' is the minimum value in the normalized environmental data. max -x' min The greater the difference, the greater the difference between the maximum and minimum values ​​within the sequence corresponding to the current dimension, the more likely there is a large-scale abnormal change, and the stronger the sample feature.

[0045] Furthermore, a preset number of dimensions with the highest degree of sample feature intensity are obtained as the dimensions to be analyzed. Preferably, in this embodiment of the invention, the preset number reference value is set to 3, which can be determined by the implementer according to the actual situation. The sample space is constructed using the preset number of dimensions to be analyzed.

[0046] The clustering analysis module is used to group the elements of the sequence at a certain time point in each dimension to be analyzed into a data point, and to cluster the data points in the sample space to obtain a cluster; based on the distance between a cluster and other clusters and the distance between the cluster and the zero point of the sample space, it is determined whether the data points in the cluster are suspected abnormal data points.

[0047] The above constructs a sample space for analysis. Furthermore, a data point in the sample space is composed of elements at the corresponding time point in the sequence corresponding to each dimension to be analyzed. After obtaining the multidimensional sample space, multidimensional environmental monitoring values ​​can be jointly analyzed based on the multidimensional sample space to obtain environmental anomaly characteristics that may indicate risks. Further, the data points within the sample space are clustered using the k-means clustering algorithm. Clustering is based on Euclidean clustering between two data points, with the k value obtained through the elbow method. Initial cluster centers are randomly selected to obtain the k-means clustering results, thus yielding different clusters.

[0048] In the obtained k-means clustering results, if a cluster is farther away from the zero point of the sample space, the corresponding temperature, humidity, and exhaust gas concentration levels of that cluster are higher, and the more likely it is to be a cluster with anomalous characteristics. On the other hand, if a cluster is farther away from other clusters, it indicates a higher degree of outlierness, and the temperature, humidity, and exhaust gas concentration represented by that cluster deviate more from normal levels, making it more likely to be an anomalous cluster. The zero point of the sample space is the origin of the sample space; for example, in three-dimensional space, the zero point of the sample space is the origin of the three-dimensional Cartesian coordinate system.

[0049] Therefore, the distances between a cluster and other clusters, as well as the distance between the cluster and the zero point of the sample space, are used to determine whether data points within a cluster are suspected outliers. Specifically, the distance between the cluster center of a cluster and the zero point of the sample space is obtained and normalized to obtain the first distance; the mean of the distances between the cluster center of the cluster and the cluster centers of all other clusters is obtained and normalized to obtain the cluster distribution characteristics; the product of the first distance and the cluster distribution characteristics is the degree of suspected anomaly of the cluster.

[0050] The specific calculation model for the degree of suspected abnormality is as follows:

[0051] Among them, D c This represents the suspected anomaly level of the c-th cluster, where a0 represents the zero point of the sample space, ac represents the cluster center of the c-th cluster, and ||a0-a... c || represents the Euclidean distance from the cluster center of the c-th cluster to the zero point of the sample space, orm(||a0-a c ||) represents the first distance, indicating the distance between the c-th cluster and the zero point. The larger this distance is, the higher the degree of suspected anomaly of the data points within the cluster. n m n is the number of clusters. m -1 represents the number of clusters remaining after removing the c-th cluster, a i Let |a| represent the cluster center of the i-th cluster. i -a c || represents the Euclidean distance from the cluster center of the i-th cluster to the cluster center of the c-th cluster. This represents the cluster distribution characteristics, which is the normalized value of the average distance between all other clusters and the c-th cluster. The larger this value is, the higher the degree of outlier the cluster is, and the more the temperature, humidity, and exhaust gas concentration represented by the cluster deviate from the normal level. It is more likely to be an abnormal cluster. "norm" represents the normalization operation.

[0052] Furthermore, based on the degree of suspected anomaly of each cluster, it is determined whether the data points within each cluster are suspected anomalous data points, and a first reference threshold is set. Preferably, in this embodiment, the value of the first reference threshold is 0.25. Specifically, the implementer can adjust the first reference threshold based on scenario experiments to obtain a first reference threshold that conforms to the actual application scenario. If the degree of suspected anomaly of a cluster is greater than or equal to the first reference threshold, then the data points within that cluster are suspected anomalous data points.

[0053] This allows us to obtain all suspected abnormal data points.

[0054] The anomaly degree acquisition module is used to cluster suspected anomaly data points based on the time interval between every two suspected anomaly data points to obtain time-series clusters; and to obtain the anomaly degree based on each time-series cluster.

[0055] The cluster analysis described above identified suspected outlier data points. Further analysis of these suspected outlier data points is needed by combining them with time-series characteristics.

[0056] Excessive battery breakage or incomplete battery discharge within a short timeframe generates higher levels of heat and gas, increasing the risk of fire and abnormal emissions. Therefore, relying solely on the spatial distribution characteristics of these clusters for anomaly analysis has limitations. This study combines the temporal distribution characteristics of suspected anomaly data points to further determine the degree of anomaly in the dataset. Greater clustering of suspected anomaly data points in the temporal domain indicates the presence of a large number of broken batteries or incomplete battery discharge leading to heat or gas accumulation within a short timeframe, corresponding to a higher level of anomaly risk. Conversely, a relatively discrete and uniform temporal distribution of suspected anomaly data points indicates a lower risk of explosions, fires, or abnormal gas emissions.

[0057] Then, all suspected anomalous data points are clustered based on the time interval between any two suspected anomalous data points to obtain temporal clusters. That is, the time interval between suspected anomalous data points is used instead of Euclidean distance for clustering, and the degree of anomalousness of the dataset is determined by evaluating the clustering results. Based on the premise that a higher number of suspected anomalous data points indicates a higher risk of environmental anomalies, the following clustering methods are used: When the temporal clusters formed by suspected anomalous data points are densely distributed within each cluster and there are certain intervals between clusters, it indicates strong aggregation of suspected anomalous data points in the temporal domain, indicating a high risk of environmental anomalies. Conversely, when the temporal clusters formed by suspected anomalous data points are discretely distributed within each cluster and the intervals between clusters are small, it indicates weak aggregation of suspected anomalous data points in the temporal domain, making them less likely to form obvious clusters, corresponding to a lower risk of environmental anomalies.

[0058] Furthermore, the degree of anomaly is obtained based on each time-series cluster. Specifically, the ratio of the number of suspected anomalous data points to the total number of data points is obtained and denoted as the anomaly percentage. The time-series clusters are arranged in chronological order to obtain a time-series cluster arrangement sequence. In the time-series cluster arrangement sequence, the mean of the time interval between the two suspected anomalous data points with the closest time interval in each pair of adjacent time-series clusters is calculated and normalized to obtain the inter-cluster distribution characteristic value. The mean of the time interval between each suspected anomalous data point in a time-series cluster and the suspected anomalous data point with the smallest time interval is calculated and denoted as the intra-cluster distribution characteristic value of that time-series cluster. The mean of the intra-cluster distribution characteristic values ​​of all time-series clusters is negatively correlated using an exponential function with the natural constant as the base to obtain the cumulative intra-cluster distribution characteristic value. The cumulative intra-cluster distribution characteristic value is added to the inter-cluster distribution characteristic value and multiplied by the anomaly percentage to obtain the degree of anomaly.

[0059] The specific calculation model for the degree of anomaly is as follows:

[0060] Where E represents the degree of anomaly, and N is the total number of data points. t This represents the number of suspected outlier data points. This represents the percentage of suspected outlier data points, also known as the anomaly percentage. A higher percentage indicates more suspected outlier data points, a greater risk of environmental anomalies, and thus a higher degree of anomaly in the dataset. z n represents the number of temporal clusters after one-dimensional clustering. z -1 represents the number of time-series clustering intervals, Δh i The time interval between the i-th temporal cluster and the two suspected outlier data points with the closest time interval in the subsequent temporal cluster of the i-th temporal cluster arrangement sequence. This represents the mean time interval between time-series clusters, where tanh is a proportional normalization used to standardize the magnitude. This is the inter-cluster distribution characteristic value. The larger the value, the more regionally clustered the suspected anomalous data points are, and the stronger the clustering of anomalous clusters, the stronger the degree of anomalousness of the dataset. i This represents the intra-cluster distribution characteristic value of the i-th temporal cluster. n i Let s be the number of suspected outlier data points within the i-th time series cluster. j_near This represents the time interval between the j-th suspected outlier data point within the i-th time series cluster and the nearest suspected outlier data point (the suspected outlier data point with the shortest time interval). Represents the cumulative cluster distribution characteristic value; This represents the mean time interval between each suspected outlier data point and its nearest point within all time-series clusters. A smaller value indicates a tighter clustering of suspected outlier data points within the cluster distribution, stronger clustering, a higher degree of dataset anomaly, and a greater risk of environmental anomalies. exp is an exponential function with the natural constant e as its base. This allows us to determine the degree of anomaly in the currently collected data.

[0061] The monitoring module is used to obtain the comprehensive degree of anomaly based on the degree of anomaly and the data of each dimension to be analyzed in the suspected abnormal data points, and to determine whether the battery discharge breakage is abnormal based on the comprehensive degree of anomaly.

[0062] While single extreme outliers can also cause data points to become outliers in the sample space, these outliers are likely due to sensor errors or changes in equipment environmental parameters such as temperature switching. Furthermore, changes in a single environmental parameter do not possess the characteristics of explosion or fire risks. Therefore, this approach integrates the distribution of data values ​​across all dimensions to be analyzed to adjust and determine the final anomaly level of the dataset. The standard deviation of the mean of the data corresponding to each dimension to be analyzed is obtained for suspected outliers. A smaller standard deviation indicates that the outliers in the multidimensional sample space are likely composed of outliers from multiple dimensions, resulting in a relatively higher degree of anomaly in the multidimensional dataset. Conversely, a larger standard deviation across all dimensions suggests that the outliers in the multidimensional sample space are more likely to be composed of extreme outliers from a single dimension, resulting in a relatively lower degree of anomaly in the multidimensional dataset.

[0063] Therefore, the overall anomaly level is obtained based on the degree of anomaly and the data of each dimension to be analyzed in the suspected anomaly data points. Specifically, the mean of the data corresponding to one of the dimensions to be analyzed among all the suspected anomaly data points is calculated and normalized, and denoted as the mean of the suspected anomaly data for that dimension to be analyzed; the standard deviation of the mean of the suspected anomaly data for all dimensions to be analyzed is calculated, and multiplied by the normalized anomaly level to obtain the overall anomaly level.

[0064] The specific calculation model for the overall anomaly level is as follows:

[0065] Where E' represents the overall anomaly level, and Eg represents the normalized anomaly level value. This represents the result of mean normalization of the data corresponding to the first dimension to be analyzed among all suspected outlier data points. Similarly, and These represent the normalized mean values ​​of the data corresponding to the second and third dimensions to be analyzed among all suspected outlier data points, respectively. These mean values ​​reflect the deviation of the data value from the normal value in the current dimension. The standard deviation of the mean of the suspected abnormal data in the three dimensions to be analyzed is given. The smaller the standard deviation, the stronger the abnormality in all dimensions to be analyzed, which corresponds to the greater likelihood of anomalies in the monitoring environment. This excludes the possibility of a single extreme anomaly in one dimension causing a large degree of abnormality. The degree of abnormality obtained is composed of multiple anomalies, which indicates a higher level of environmental risk.

[0066] This allows us to obtain the overall anomaly level of the currently collected data. The higher the overall anomaly level, the greater the likelihood of an anomaly occurring. Finally, based on the overall anomaly level, we determine whether battery discharge breakage has occurred. Specifically, a second reference threshold is set. Preferably, in this embodiment of the invention, the value of the second reference threshold is 0.4. The accurate empirical threshold for specific scenarios can be adjusted based on the second reference threshold through usage experiments and debugging in the corresponding scenarios. When the overall anomaly level is greater than or equal to the second reference threshold, it is considered that the current real-time monitoring data indicates a certain environmental anomaly risk, and battery discharge breakage has occurred. At this time, the battery breakage process is stopped, and a real-time alarm is issued to remind the staff of the relevant monitored environmental hazards, and the batteries at the feed inlet are adjusted.

[0067] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0068] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An environmental data monitoring system for discharge crushing for battery recycling, characterized by, The system comprises: a data acquisition module, configured to acquire environmental data in different dimensions in a battery breaking device and perform preprocessing to obtain a sequence corresponding to each dimension; a sample space construction module, configured to calculate a sample feature degree of a dimension corresponding to a sequence based on wave peak points and wave valley points in the sequence, filter the dimensions according to the sample feature degrees to obtain dimensions to be analyzed, and construct a sample space; a clustering analysis module, configured to form a data point by elements at one time in a sequence corresponding to each dimension to be analyzed, and perform clustering on the data points in the sample space to obtain clusters; and determine whether the data points in a cluster are suspected abnormal data points according to distances between the cluster and other clusters and between the cluster and a zero point of the sample space; an abnormal degree acquisition module, configured to cluster the suspected abnormal data points based on time intervals between each two suspected abnormal data points to obtain time sequence clusters, and acquire an abnormal degree according to each time sequence cluster, including: obtaining a ratio of the suspected abnormal data points to the total data points, denoted as an abnormal proportion; arranging the time sequence clusters in a time sequence order to obtain a time sequence cluster arrangement sequence; calculating a mean value of a time interval between two suspected abnormal data points with the closest time interval in each two adjacent time sequence clusters in the time sequence cluster arrangement sequence, and performing normalization to obtain an inter-cluster distribution feature value; calculating a mean value of a time interval between each suspected abnormal data point and a suspected abnormal data point with the smallest time interval in a time sequence cluster, denoted as an intra-cluster distribution feature value of the time sequence cluster; performing a negative correlation mapping on a mean value of the intra-cluster distribution feature values of all the time sequence clusters by using an exponential function with a natural constant as a base to obtain a cumulative intra-cluster distribution feature value; adding the cumulative intra-cluster distribution feature value and the inter-cluster distribution feature value, and multiplying the sum by the abnormal proportion to obtain the abnormal degree; a monitoring module, configured to acquire a comprehensive abnormal degree according to the abnormal degree and data of each dimension to be analyzed in the suspected abnormal data points, and determine whether an abnormality occurs in battery discharge breaking based on the comprehensive abnormal degree.

2. The environmental data monitoring system for discharge crushing for battery recycling according to claim 1, characterized in that, The obtaining of the sequence corresponding to each dimension includes: performing normalization on the environmental data of each dimension, and forming a sequence corresponding to the dimension by using the normalized environmental data of the dimension.

3. The environmental data monitoring system for discharge crushing for battery recycling according to claim 1, characterized in that, The calculation of the sample feature degree of the dimension corresponding to the sequence based on the wave peak points and the wave valley points in the sequence includes: obtaining the wave peak points and the wave valley points in the sequence and forming a fluctuation point sequence in a time sequence order; obtaining an absolute value of a difference between corresponding ordinate values of each two adjacent fluctuation points in the fluctuation point sequence, and performing summation to obtain a fluctuation change feature value; multiplying the number of elements of the sequence, the fluctuation change feature value, and a difference between the maximum value and the minimum value of the sequence to obtain the sample feature degree of the dimension corresponding to the sequence.

4. The environmental data monitoring system for discharge crushing for battery recycling according to claim 1, characterized in that, The filtering of the dimensions according to the sample feature degrees to obtain the dimensions to be analyzed and the construction of the sample space include: obtaining a preset number of dimensions with the largest sample feature degrees as the dimensions to be analyzed; and constructing the sample space by using the dimensions to be analyzed.

5. The environmental data monitoring system for discharge crushing for battery recycling according to claim 1, characterized in that, The method comprises the following steps: The distance between the cluster center of a cluster and the zero point of the sample space is obtained and normalized to obtain a first distance; the mean distance between the cluster center of the cluster and the cluster centers of other clusters is obtained and normalized to obtain a cluster distribution feature; the product of the first distance and the cluster distribution feature is the suspected abnormality degree of the cluster; whether the data points in the cluster are suspected abnormal data points is determined based on the suspected abnormality degree of the cluster.

6. The environmental data monitoring system for discharge crushing for battery recycling according to claim 5, characterized in that, The method comprises the following steps: If the suspected abnormality degree of a cluster is greater than or equal to a first reference threshold, the data points in the cluster are suspected abnormal data points.

7. The environmental data monitoring system for discharge crushing for battery recycling according to claim 1, characterized in that, The method comprises the following steps: The mean value of the data corresponding to one of the dimensions to be analyzed in all suspected abnormal data points is calculated and normalized, and is recorded as the suspected abnormal data mean value of the dimension to be analyzed; the standard deviation of the suspected abnormal data mean values of all dimensions to be analyzed is calculated, and is multiplied by the normalized abnormality degree to obtain a comprehensive abnormality degree. 8.The environmental data monitoring system for discharging and crushing of battery recycling according to claim 1, wherein, The method comprises the following steps: When the comprehensive abnormality degree is greater than or equal to a second reference threshold, the battery discharge crushing is abnormal.

Images