CN118152969B - Temperature monitoring method and system based on multiple sensors - Google Patents
Temperature monitoring method and system based on multiple sensors Download PDFInfo
- Publication number
- CN118152969B CN118152969B CN202410571834.1A CN202410571834A CN118152969B CN 118152969 B CN118152969 B CN 118152969B CN 202410571834 A CN202410571834 A CN 202410571834A CN 118152969 B CN118152969 B CN 118152969B
- Authority
- CN
- China
- Prior art keywords
- temperature
- sensor node
- sensor
- data
- trend
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000002159 abnormal effect Effects 0.000 claims abstract description 143
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 76
- 238000001514 detection method Methods 0.000 claims abstract description 43
- 230000015654 memory Effects 0.000 claims abstract description 15
- 238000001816 cooling Methods 0.000 claims abstract description 14
- 238000004458 analytical method Methods 0.000 claims description 32
- 230000000630 rising effect Effects 0.000 claims description 23
- 238000002955 isolation Methods 0.000 claims description 21
- 230000005856 abnormality Effects 0.000 claims description 19
- 230000007774 longterm Effects 0.000 claims description 12
- 230000000737 periodic effect Effects 0.000 claims description 12
- 230000002068 genetic effect Effects 0.000 claims description 11
- 230000035772 mutation Effects 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 15
- 230000007787 long-term memory Effects 0.000 description 6
- 230000006403 short-term memory Effects 0.000 description 6
- 238000007405 data analysis Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention relates to the technical field of data processing, and discloses a temperature monitoring method and system based on multiple sensors, which are used for improving the accuracy of temperature monitoring based on the multiple sensors. The method comprises the following steps: acquiring temperature data of a monitoring area to obtain a temperature data set; performing anomaly detection on the temperature data set to obtain an anomaly detection report; inputting the abnormal mode and the temperature change data into a long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend; respectively carrying out change trend data matching on each sensor node to obtain a change trend data set; extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node, and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level; and (3) identifying a high-temperature area of the node group label of each sensor node by using an Euclidean distance method and a clustering algorithm to obtain a target high-temperature area, and carrying out air exhaust cooling treatment on the target high-temperature area.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a temperature monitoring method and system based on multiple sensors.
Background
In the current temperature monitoring field, common background technologies include sensor networks, data analysis algorithms, and temperature control systems. The sensor network collects the environmental temperature data in real time through the sensor nodes deployed in the monitoring area, and transmits the data to the data center or the control center for processing and analysis. The data analysis algorithm is responsible for processing, analyzing and anomaly detecting the acquired temperature data so as to discover and cope with temperature anomalies in time. And the temperature control system controls and adjusts the temperature of the monitored area by adjusting the working states of the air conditioner, the fan and other equipment according to the monitored temperature data, so that the temperature is ensured to be in a proper range.
However, the prior art still has some disadvantages. Firstly, the traditional temperature monitoring system generally only adopts a single sensor for monitoring, has limited coverage range and low accuracy, and is difficult to meet the real-time monitoring requirement in a complex environment. Secondly, the traditional anomaly detection algorithm has limited capability of identifying complex temperature change modes, and is easy to miss report or misreport, thereby affecting the accuracy and reliability of the monitoring result. In addition, the existing temperature control system is poor in effect when processing a high-temperature area, often has the problems of insufficient control precision, high energy consumption and the like, and cannot effectively meet the temperature management requirement in a high-temperature environment.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide a temperature monitoring method and system based on multiple sensors, which are used for improving the accuracy of temperature monitoring based on multiple sensors.
The invention provides a temperature monitoring method based on multiple sensors, which comprises the following steps: acquiring temperature data of a monitoring area through a preset sensor network to obtain a temperature data set;
performing anomaly detection on the temperature data set through a preset isolation forest algorithm to obtain an anomaly detection report, wherein the anomaly detection report comprises: abnormal mode and temperature change data;
Inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend;
Based on the temperature change trend, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node;
Extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node, and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level, wherein the node group label specifically comprises the following steps: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
and respectively identifying a high-temperature region of the node group label of each sensor node by using an Euclidean distance method and a clustering algorithm to obtain a target high-temperature region, and carrying out air exhaust cooling treatment on the target high-temperature region.
In the invention, before the step of collecting the temperature data of the monitoring area through the preset sensor network, the method further comprises the following steps:
carrying out sensor layout analysis on the monitoring area through a genetic algorithm to obtain layout parameters, wherein the layout parameters comprise a plurality of sensor nodes;
respectively carrying out sensor parameter matching on each sensor node to obtain the sensor parameters of each sensor;
and configuring a sensor network based on the layout parameters and the sensor parameters of each sensor, and acquiring the temperature data of the monitoring area through the sensor network to obtain a temperature data set.
In the invention, the abnormality detection is performed on the temperature dataset through a preset isolation forest algorithm to obtain an abnormality detection report, wherein the abnormality detection report comprises: an abnormal mode and temperature change data step comprising:
Extracting change characteristics of the temperature data set through the isolated forest algorithm to obtain temperature change characteristics;
Classifying the temperature change characteristics to obtain temperature mutation characteristics, long-term trend change characteristics and periodic change characteristics;
Setting the number of trees and subsampled capacity of the isolated forest algorithm based on the temperature abrupt change feature, the long-term trend change feature, and the periodic change feature;
carrying out isolation characteristic anomaly score analysis on the temperature mutation characteristic, the long-term trend change characteristic and the periodic change characteristic through an isolation forest algorithm after parameter setting to obtain an anomaly score set;
Based on a preset score threshold, carrying out abnormal condition identification on the temperature data set through the abnormal score set to obtain an abnormal detection report, wherein the abnormal detection report comprises: abnormal pattern and temperature change data.
In the present invention, the step of inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend and obtain the temperature change trend includes:
Data grouping is carried out on the abnormal mode and the temperature change data to obtain a plurality of groups of abnormal data;
respectively carrying out historical temperature average analysis on each group of abnormal data to obtain a historical temperature average corresponding to each group of abnormal data;
extracting the occurrence frequency of the abnormal event from each group of abnormal data respectively to obtain the occurrence frequency of the abnormal event corresponding to each group of abnormal data;
Fusing the historical temperature average value corresponding to each group of abnormal data and the abnormal event occurrence frequency corresponding to each group of abnormal data into abnormal temperature occurrence frequency;
inputting the abnormal temperature occurrence frequency into the long-short-period memory network to perform abnormal time interval matching to obtain a plurality of abnormal time intervals;
respectively predicting the temperature trend of each abnormal time interval to obtain the temperature change trend of each abnormal time interval;
And carrying out temperature trend fusion analysis on the temperature change trend of each abnormal time interval to obtain the temperature change trend.
In the invention, based on the temperature change trend, the change trend data is respectively matched with each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend step of each sensor node, and the method comprises the following steps:
Carrying out time sequence synchronization on each sensor node through the temperature change trend, and extracting an identifier of each sensor node;
extracting key features of each sensor node based on the identifier of each sensor node, wherein the key features include: temperature peak, minimum temperature, and average temperature;
and based on the key characteristics of each sensor node and the temperature change trend, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node.
In the invention, the high temperature region identification is carried out on the node group labels of each sensor node through the Euclidean distance method and the clustering algorithm to obtain a target high temperature region, and the air exhaust cooling treatment step is carried out on the target high temperature region, which comprises the following steps:
grouping a plurality of sensor nodes through each sensor node to obtain a plurality of sensor groups;
Carrying out Euclidean distance calculation on each sensor group through the Euclidean distance method to obtain Euclidean distance sets;
Clustering the Euclidean distance set into sensor group areas through the clustering algorithm to obtain clustering areas;
calibrating the high-temperature region of the clustering region to obtain a target high-temperature region;
Collecting the current temperature of the target high-temperature area to obtain current temperature data;
and matching the current temperature data with a target exhaust force, and controlling a preset exhaust module to exhaust and cool the target high-temperature region based on the target exhaust force.
The invention also provides a temperature monitoring system based on multiple sensors, which comprises:
the acquisition module is used for acquiring temperature data of the monitoring area through a preset sensor network to obtain a temperature data set;
the detection module is used for carrying out anomaly detection on the temperature data set through a preset isolation forest algorithm to obtain an anomaly detection report, wherein the anomaly detection report comprises the following components: abnormal mode and temperature change data;
The input module is used for inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend;
The matching module is used for respectively matching the change trend data of each sensor node based on the temperature change trend to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node;
The extracting module is used for extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level, and specifically comprises the following steps: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
The identification module is used for respectively carrying out high-temperature area identification on the node group labels of each sensor node through a Euclidean distance method and a clustering algorithm to obtain a target high-temperature area, and carrying out air exhaust cooling treatment on the target high-temperature area.
According to the technical scheme provided by the invention, the sensor layout analysis is carried out on the monitoring area through the genetic algorithm, so that the layout parameters are obtained, the layout position of the sensor can be optimized, and the monitoring coverage rate and accuracy are improved. And secondly, the temperature data set is subjected to change feature extraction and abnormality detection through an isolation forest algorithm, so that the temperature abnormality mode can be timely and accurately identified, and the potential temperature abnormality situation can be quickly found. And then, the long-term and short-term memory network is utilized to conduct trend prediction on the abnormal mode and the temperature change data, so that the trend of temperature change is understood, and timely decision and adjustment are made. Further, by matching and analyzing the temperature change trend of the sensor node, the rule and trend of the temperature change can be identified, and the temperature change condition of the monitoring area can be better known. On the basis, the high Wen Cancha sample of each sensor node is extracted and converted into a node group label with an alarm level, so that the area possibly having high temperature problem can be accurately identified, and the quick processing of corresponding measures can be helped. Finally, the high-temperature area is identified for the node group labels of the sensor nodes through the Euclidean distance method and the clustering algorithm, and the target high-temperature area is subjected to air exhaust and cooling treatment, so that the temperature of the high-temperature area can be effectively reduced, and the safety and stability of equipment and the environment are ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a temperature monitoring method based on multiple sensors in an embodiment of the invention.
Fig. 2 is a schematic diagram of a temperature monitoring system based on multiple sensors according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
For ease of understanding, the following describes a specific flow of an embodiment of the present invention, referring to fig. 1, fig. 1 is a flow chart of a multi-sensor-based temperature monitoring method according to an embodiment of the present invention, as shown in fig. 1, including the following steps:
S101, acquiring temperature data of a monitoring area through a preset sensor network to obtain a temperature data set;
s102, carrying out anomaly detection on a temperature data set through a preset isolation forest algorithm to obtain an anomaly detection report, wherein the anomaly detection report comprises the following steps: abnormal mode and temperature change data;
S103, inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend;
S104, based on the temperature change trend, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node;
S105, extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node, and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level, wherein the method specifically comprises the following steps: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
S106, respectively identifying the high-temperature areas of the node group labels of each sensor node through a Euclidean distance method and a clustering algorithm to obtain target high-temperature areas, and carrying out air exhaust cooling treatment on the target high-temperature areas.
It should be noted that, through a preset sensor network, a plurality of temperature sensors are disposed in a monitoring area, for example, in a factory workshop. The sensors can collect temperature data of various positions of the workshop in real time to form a temperature data set. And further, using a preset isolation forest algorithm to detect abnormality of the temperature data set. This algorithm can effectively identify abnormal patterns and provide temperature change data. For example, if a sensor at a corner of a plant detects an abnormal temperature rise, the isolated forest algorithm can identify this abnormal pattern and provide temperature change data associated therewith.
And then, inputting the abnormal modes and the temperature change data into a preset long-short-period memory network to predict the temperature trend. The long-term and short-term memory network can learn the historical data of the temperature and forecast the future temperature change trend. For example, if the temperature of a certain workshop area has long been in an ascending trend, the long-short-term memory network can predict the temperature change of the area in the future by learning historical data. Based on the temperature change trends, the change trend data of each sensor node can be matched to obtain a change trend data set. This means that the trend of the temperature change of each sensor node can be analyzed so as to more accurately understand the temperature change condition of each location. For example, if a certain sensor node detects a significant temperature rise within one hour, this anomaly can be understood from this trend data.
Next, a high Wen Cancha sample in the trend of temperature change of each sensor node is extracted and converted into a node group label with alarm level. This means that it is possible to judge which areas have a high temperature abnormality according to the condition of temperature change and give a corresponding alarm level. For example, if a trend of temperature change detected by a certain sensor node shows a high Wen Cancha, it may be marked as an abnormality of a high alarm level. And finally, carrying out high-temperature region identification on the node group labels of each sensor node by using a Euclidean distance method and a clustering algorithm to obtain a target temperature region, and carrying out exhaust cooling treatment on the target temperature region. This means that it is possible to determine which areas are areas of high temperature based on the labels of the sensor nodes and take corresponding measures to reduce the temperature. For example, if the clustering algorithm determines that a region is a target temperature region, the temperature of the region may be reduced by an exhaust system to ensure that the temperature within the plant remains within a safe range.
By executing the steps, the sensor layout analysis is carried out on the monitoring area through the genetic algorithm, so that the layout parameters are obtained, the layout position of the sensor can be optimized, and the monitoring coverage rate and accuracy are improved. And secondly, the temperature data set is subjected to change feature extraction and abnormality detection through an isolation forest algorithm, so that the temperature abnormality mode can be timely and accurately identified, and the potential temperature abnormality situation can be quickly found. And then, the long-term and short-term memory network is utilized to conduct trend prediction on the abnormal mode and the temperature change data, so that the trend of temperature change is understood, and timely decision and adjustment are made. Further, by matching and analyzing the temperature change trend of the sensor node, the rule and trend of the temperature change can be identified, and the temperature change condition of the monitoring area can be better known. On the basis, the high Wen Cancha sample of each sensor node is extracted and converted into a node group label with an alarm level, so that the area possibly having high temperature problem can be accurately identified, and the quick processing of corresponding measures can be helped. Finally, the high-temperature area is identified for the node group labels of the sensor nodes through the Euclidean distance method and the clustering algorithm, and the target high-temperature area is subjected to air exhaust and cooling treatment, so that the temperature of the high-temperature area can be effectively reduced, and the safety and stability of equipment and the environment are ensured.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Carrying out sensor layout analysis on the monitoring area through a genetic algorithm to obtain layout parameters, wherein the layout parameters comprise a plurality of sensor nodes;
(2) Respectively carrying out sensor parameter matching on each sensor node to obtain the sensor parameters of each sensor;
(3) And configuring a sensor network based on the layout parameters and the sensor parameters of each sensor, and acquiring temperature data of a monitoring area through the sensor network to obtain a temperature data set.
Specifically, a genetic algorithm is used to perform sensor layout analysis on the monitored area. The genetic algorithm is an optimization method, and an optimal solution can be found according to the characteristics of specific problems. In this case, it is desirable to maximize coverage of the monitored area by the sensor layout and ensure comprehensive acquisition of temperature data. An optimal sensor layout scheme can be designed through a genetic algorithm, and the optimal sensor layout scheme comprises the number and the positions of sensor nodes. For example, assuming a large warehouse that needs to monitor temperature, genetic algorithms can be used to determine the optimal sensor locations, ensuring that each area can be covered, e.g., sensor nodes are placed at each corner and center location.
Next, for each sensor node, sensor parameter matching is performed to obtain a sensor parameter for each sensor. These parameters may include sensitivity of the sensor, sampling rate, accuracy, etc. By matching the sensor parameters, each sensor can be ensured to have consistent performance when acquiring temperature data, thereby improving the accuracy and reliability of monitoring. For example, for sensor nodes at different locations in a warehouse, it may be necessary to adjust their sensitivity and sampling rate to accommodate temperature variations in different areas. Such as increasing the sampling rate in areas of greater temperature fluctuation to more timely capture temperature changes.
And finally, configuring a sensor network based on the layout parameters and the sensor parameters of each sensor, and acquiring temperature data of a monitoring area through the sensor network to obtain a temperature data set. This means that the parameters of each sensor node can be set according to the previously determined sensor layout scheme and connected into a network to collect temperature data in the monitored area in real time. For example, the optimal positions for placing 10 sensor nodes inside a warehouse are determined according to a genetic algorithm, and parameters of the sensors are adjusted according to the characteristics of each position. Then, by configuring the sensor nodes and establishing a network, the temperatures at different locations within the warehouse can be monitored in real time and a complete temperature data set is formed for subsequent temperature monitoring and analysis.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Extracting change characteristics of the temperature dataset through an isolation forest algorithm to obtain temperature change characteristics;
(2) Classifying the temperature change characteristics to obtain temperature mutation characteristics, long-term trend change characteristics and periodic change characteristics;
(3) Setting the number of trees and subsampled capacity of an isolated forest algorithm based on the temperature mutation characteristics, the long-term trend change characteristics and the periodic change characteristics;
(4) Carrying out isolation characteristic abnormal score analysis on the temperature mutation characteristic, the long-term trend change characteristic and the periodic change characteristic through an isolation forest algorithm after parameter setting to obtain an abnormal score set;
(5) Based on a preset score threshold, carrying out abnormal condition identification on the temperature data set through an abnormal score set to obtain an abnormal detection report, wherein the abnormal detection report comprises: abnormal pattern and temperature change data.
The temperature change feature is extracted from the temperature data set by using an isolated forest algorithm. The isolated forest algorithm is an unsupervised learning algorithm, and can effectively identify abnormal values and abnormal modes. In this scenario, the temperature dataset is input into an isolated forest algorithm, which analyzes the dataset to find out the change characteristics thereof, such as the fluctuation, rising or falling trend of the temperature, etc. For example, assuming that the isolation forest algorithm finds that the temperature recorded by a certain sensor node has significantly fluctuated within a certain period of time while monitoring the plant temperature, this is a temperature change feature, which may suggest that an abnormal situation exists in the area.
And then, classifying the temperature change characteristics to obtain temperature mutation characteristics, long-term trend change characteristics and periodic change characteristics. These classifications may help to more accurately understand the pattern of changes in the temperature dataset and provide more information for subsequent anomaly detection. For example, if a sudden temperature rise or drop is identified in the temperature dataset, then this is a temperature abrupt feature; if the temperature of a certain area shows a continuous rising or falling trend, the temperature is a long-term trend change characteristic; if the temperature data exhibits periodic fluctuations, this is a periodically varying feature.
Based on these features, the number of trees and subsampled volumes of the isolated forest algorithm are set to better detect anomalies in the temperature dataset. The setting of these parameters can affect the performance and accuracy of the algorithm and thus need to be adjusted on a case-by-case basis. For example, the isolated forest algorithm may be set to contain 100 trees, each tree having a subsampled capacity of 50. Such an arrangement may improve the robustness and accuracy of the algorithm, enabling it to better identify abnormal situations.
And then, carrying out isolation characteristic anomaly score analysis on the temperature mutation characteristic, the long-term trend change characteristic and the periodic change characteristic through an isolation forest algorithm after parameter setting to obtain an anomaly score set. This means that the anomaly can be identified by an algorithm that analyzes the different features in the temperature dataset and gives a corresponding anomaly score. For example, if a sudden drastic change in temperature occurs in an area and a high anomaly score is obtained in the isolated forest algorithm, this may be an anomaly that requires further observation and processing.
And finally, based on a preset score threshold, carrying out abnormal condition identification on the temperature data set through an abnormal score set to obtain an abnormal detection report, wherein the abnormal detection report comprises an abnormal mode and temperature change data. This means that it is possible to determine which cases belong to anomalies according to the set threshold value and to generate a corresponding report. For example, if the anomaly score of a region exceeds a preset threshold, then the region may be identified as having anomalies and a corresponding anomaly detection report may be generated, including anomaly patterns and temperature change data, for subsequent processing and analysis.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Data grouping is carried out on the abnormal mode and the temperature change data to obtain a plurality of groups of abnormal data;
(2) Respectively carrying out historical temperature average analysis on each group of abnormal data to obtain a historical temperature average corresponding to each group of abnormal data;
(3) Extracting the occurrence frequency of the abnormal event from each group of abnormal data respectively to obtain the occurrence frequency of the abnormal event corresponding to each group of abnormal data;
(4) Fusing the historical temperature average value corresponding to each group of abnormal data and the abnormal event occurrence frequency corresponding to each group of abnormal data into abnormal temperature occurrence frequency;
(5) Inputting the occurrence frequency of the abnormal temperature into a long-period and short-period memory network to perform abnormal time interval matching, so as to obtain a plurality of abnormal time intervals;
(6) Respectively predicting the temperature trend of each abnormal time interval to obtain the temperature change trend of each abnormal time interval;
(7) And carrying out temperature trend fusion analysis on the temperature change trend of each abnormal time interval to obtain the temperature change trend.
Specifically, the abnormal mode and the temperature change data are subjected to data grouping to obtain a plurality of groups of abnormal data. This means that the abnormal situation in the temperature data set is classified according to a specific criterion, and a plurality of abnormal data sets are formed. For example, if a rapid rise in temperature is found in three consecutive hours while monitoring the plant temperature, the temperature data in this period can be classified into a set of abnormal data. And then, respectively carrying out historical temperature average analysis on each group of abnormal data to obtain a historical temperature average corresponding to each group of abnormal data. This means that the background and trend of the anomaly can be known by analyzing the historical temperature averages of the anomaly data set. For example, if the historical temperature average for an abnormal data set is relatively high, this may mean that there is a normal rising temperature trend in the region.
And then, respectively extracting the occurrence frequency of the abnormal event from each group of abnormal data to obtain the occurrence frequency of the abnormal event corresponding to each group of abnormal data. This means that the frequency of occurrence of the abnormal data set can be analyzed to judge the frequency of occurrence of the abnormal situation. For example, if a certain anomaly data set occurs multiple times in the past week, it may mean that there is a persistent anomaly in that region. Next, the historical temperature average value corresponding to each set of abnormal data and the occurrence frequency of the abnormal event are fused into the occurrence frequency of the abnormal temperature. This means that the severity of the abnormal situation can be estimated by taking the historical temperature average and the occurrence frequency of the abnormal event into consideration. For example, if the historical temperature average of an abnormal data set is high and the occurrence frequency of abnormal events is high, it may mean that there is a serious abnormality in the area.
And then, inputting the occurrence frequency of the abnormal temperature into a long-short-period memory network to perform abnormal time interval matching, so as to obtain a plurality of abnormal time intervals. This means that the long and short term memory network can be used to match and identify the abnormal time intervals to determine the duration and interval of the abnormal situation. For example, if the long and short term memory network identifies that an abnormal time interval has lasted 3 days in the past week, then the duration of the abnormal condition may be determined to be 3 days. And then, respectively carrying out temperature trend prediction on each abnormal time interval to obtain the temperature change trend of each abnormal time interval. This means that the trend of temperature change in the abnormal time interval can be analyzed by using the predictive model, so that the trend of the abnormal situation can be known. For example, if the predictive model shows a constant rise in temperature over some abnormal time interval, this may mean that the abnormal situation is being exacerbated.
And finally, carrying out temperature trend fusion analysis on the temperature change trend of each abnormal time interval to obtain the temperature change trend. This means that the temperature variation trend in each abnormal time interval can be comprehensively considered, thereby obtaining the overall temperature variation trend. For example, if the temperature variation trend over a plurality of abnormal time intervals shows a continuous rise in temperature, it may mean that the overall temperature trend is also rising.
In a specific embodiment, the process of executing step S104 may specifically include the following steps:
(1) Carrying out time sequence synchronization on each sensor node through the temperature change trend, and extracting an identifier of each sensor node;
(2) Extracting key features of each sensor node based on the identifier of each sensor node, wherein the key features include: temperature peak, minimum temperature, and average temperature;
(3) And based on the key characteristics and the temperature change trend of each sensor node, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node.
Specifically, each sensor node is time-sequence synchronized through the temperature change trend, and the identifier of each sensor node is extracted. This means that the time relationship between the sensor nodes can be determined from the temperature data recorded by them and a unique identifier generated for each sensor node for subsequent data processing and analysis. For example, assume that while monitoring the temperature of a plant floor, three sensor nodes each record temperature data at different locations. By timing synchronizing these data, the time relationship between them can be determined and an identifier, such as "S1", "S2" and "S3", can be generated for each sensor node.
Next, based on the identifier of each sensor node, key features of each sensor node are extracted, including temperature peaks, lowest temperatures, and average temperatures. This means that the most important features of each sensor node can be extracted by analysing the temperature data recorded by it for subsequent data matching and analysis. For example, for the three sensor nodes described above, their temperature peaks, lowest temperatures, and average temperatures may be extracted, respectively. Assume that the temperature peak recorded by sensor node "S1" is 30 ℃, the lowest temperature is 20 ℃, and the average temperature is 25 ℃.
And then, based on key characteristics and temperature change trend of each sensor node, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node. This means that the sensor nodes can be analyzed for their temperature change pattern from their characteristics and their temperature change trend and recorded in the data set. For example, for the sensor node "S1", the temperature variation trend thereof may be analyzed, and the temperature thereof is found to exhibit a gradually rising trend in the operating period. For the sensor node "S2", a situation in which the temperature fluctuation is large may occur, indicating that there may be an abnormal situation at this location. Therefore, the state and the change condition of the temperature monitoring area can be further known according to the change trend data of each sensor node.
In one embodiment, the process of executing step S105 specifically includes: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node;
the high Wen Cancha samples of each sensor node are respectively converted into node group labels with alarm levels based on the alarm level data of each sensor node.
Specifically, trend classification is performed on the temperature variation trend of each sensor node, so as to know the temperature rising trend of each sensor node. This means that the trend type of the temperature change thereof, such as rising, falling or fluctuating, can be determined by analyzing the temperature data recorded by each sensor node. By this step, it is possible to identify which sensor nodes have a temperature rise. For example, assuming that the temperature of a plant is monitored, by analyzing the sensor node data, it is found that the temperature recorded by sensor node A is continuously rising, the temperature recorded by sensor node B is kept stable, and the temperature recorded by sensor node C is decreasing. Therefore, it can be judged that there is a tendency for the sensor node a to rise in temperature. Next, a high Wen Cancha analysis is performed on the temperature rising trend of each sensor node through a sliding window algorithm, so as to obtain a high Wen Cancha sample of each sensor node. This means that abnormally high temperature data points can be detected by setting a sliding window during the period of temperature rise, thereby identifying a high Wen Cancha sample. For example, for a trend of temperature rise recorded by sensor node a, a sliding window algorithm may be used to slide a window in the time series data, detecting abnormally high temperature data points within the window, and thus determining a high Wen Cancha sample. Next, residual value conversion is performed on the high Wen Cancha samples of each sensor node separately to obtain a high Wen Cancha value set for each sensor node. This means that the detected high Wen Cancha samples can be converted into numerical form for subsequent data analysis and processing. For example, for sensor node a, a series of high Wen Cancha samples were detected, which were converted to numerical form, forming a high Wen Cancha numerical set. And then, respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node. This means that the high Wen Cancha values can be mapped to different alarm levels for subsequent exception handling and management, depending on their size. For example, the alarm level data of each sensor node can be determined by dividing the sensor node into three levels, namely a low level, a medium level and a high level according to the magnitude of a high Wen Cancha value, wherein the three levels respectively represent abnormal conditions with different degrees. Finally, based on the alarm level data of each sensor node, respectively converting the high Wen Cancha samples of each sensor node into node group labels with alarm levels. This means that each sensor node can be categorized into different node groups according to its alarm level for subsequent anomaly identification and processing. For example, for sensor node a, if its alarm level is high, its high Wen Cancha samples are marked as node group labels with high level alarms.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Grouping a plurality of sensor nodes through each sensor node to obtain a plurality of sensor groups;
(2) Carrying out Euclidean distance calculation on each sensor group through an Euclidean distance method to obtain Euclidean distance sets;
(3) Clustering the Euclidean distance set by using a clustering algorithm to obtain a clustering area;
(4) Calibrating the high-temperature region of the clustering region to obtain a target high-temperature region;
(5) Acquiring the current temperature of a target high-temperature area to obtain current temperature data;
(6) And matching the target exhaust force based on the current temperature data, and controlling a preset exhaust module to exhaust and cool the target high-temperature region based on the target exhaust force.
The plurality of sensor groups are obtained by grouping a plurality of sensor nodes for each sensor node. This means that the sensor nodes can be grouped according to their positional relationship and interactions in order to better understand the temperature change situation of the monitored area. For example, assuming that a plurality of temperature sensors are installed inside one large warehouse, the sensors cover different areas, they may be divided into different sensor groups such as a front area, a middle area, and a rear area according to their positional relationship. Then, euclidean distance calculation is performed on each sensor group by using the Euclidean distance method, and Euclidean distance sets are obtained. This means that the euclidean distance between sensor nodes inside the sensor group can be calculated from the distance between them in order to determine the similarity and the difference between the sensor groups. For example, for a sensor group in a front region, the Euclidean distance between each pair of sensor nodes can be calculated to obtain a Euclidean distance set, so as to describe the distance relation between each sensor node in the front region. And then, carrying out sensor group region clustering on the Euclidean distance set through a clustering algorithm to obtain a clustering region. This means that the sensor groups can be categorized into different cluster areas based on their similarity and difference between them in order to better understand the overall temperature distribution of the monitored area. For example, by performing cluster analysis on the groups of sensors in the front, middle and rear regions, they can be categorized into different cluster regions, respectively, for subsequent high temperature region identification and processing. And then, calibrating the high-temperature region of the clustering region to obtain a target temperature region. This means that it is possible to determine which areas have high temperature problems according to the temperature conditions of the various clustered areas within the monitored area, thereby determining the target temperature area. For example, according to the temperature distribution condition of the clustered regions, it is possible to determine that some regions exist at abnormally high temperatures, and mark them as target high temperature regions. And then, carrying out current temperature acquisition on the target high-temperature area to obtain current temperature data. This means that the temperature conditions of the target temperature zone can be monitored in real time for subsequent processing and adjustment. For example, a temperature sensor may be installed to monitor the temperature of the target temperature region in real time, and obtain current temperature data. And finally, matching the target exhaust strength based on the current temperature data, and performing exhaust cooling treatment on the target high-temperature area by controlling a preset exhaust module. This means that the air exhaust force of the air exhaust module can be adjusted according to the real-time temperature condition of the target temperature area so as to reduce the temperature of the target temperature area. For example, if the temperature of the target temperature area exceeds a preset threshold, the air exhausting force of the air exhausting module can be increased to accelerate the hot air to be exhausted, so that the temperature of the target temperature area is reduced.
The embodiment of the invention also provides a temperature monitoring system based on multiple sensors, as shown in fig. 2, which specifically comprises:
The acquisition module 201 is configured to acquire temperature data of a monitoring area through a preset sensor network, and obtain a temperature dataset;
The detection module 202 is configured to perform anomaly detection on the temperature dataset through a preset isolated forest algorithm to obtain an anomaly detection report, where the anomaly detection report includes: abnormal mode and temperature change data;
The input module 203 is configured to input the abnormal mode and the temperature change data into a preset long-short-period memory network to perform temperature trend prediction, so as to obtain a temperature change trend;
the matching module 204 is configured to match the change trend data of each sensor node based on the temperature change trend, so as to obtain a change trend data set, where the change trend data set includes a temperature change trend of each sensor node;
The extracting module 205 is configured to extract a high Wen Cancha sample of each sensor node in the temperature variation trend of each sensor node, and convert the high Wen Cancha sample of each sensor node into a node group label with an alarm level, which specifically includes: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
and the identification module 206 is configured to identify the high-temperature area of the node group label of each sensor node by using a euclidean distance method and a clustering algorithm, obtain a target high-temperature area, and perform exhaust cooling treatment on the target high-temperature area.
Through the cooperative work of the modules, the sensor layout analysis is carried out on the monitoring area through a genetic algorithm, so that the layout parameters are obtained, the layout positions of the sensors can be optimized, and the monitoring coverage rate and accuracy are improved. And secondly, the temperature data set is subjected to change feature extraction and abnormality detection through an isolation forest algorithm, so that the temperature abnormality mode can be timely and accurately identified, and the potential temperature abnormality situation can be quickly found. And then, the long-term and short-term memory network is utilized to conduct trend prediction on the abnormal mode and the temperature change data, so that the trend of temperature change is understood, and timely decision and adjustment are made. Further, by matching and analyzing the temperature change trend of the sensor node, the rule and trend of the temperature change can be identified, and the temperature change condition of the monitoring area can be better known. On the basis, the high Wen Cancha sample of each sensor node is extracted and converted into a node group label with an alarm level, so that the area possibly having high temperature problem can be accurately identified, and the quick processing of corresponding measures can be helped. Finally, the high-temperature area is identified for the node group labels of the sensor nodes through the Euclidean distance method and the clustering algorithm, and the target high-temperature area is subjected to air exhaust and cooling treatment, so that the temperature of the high-temperature area can be effectively reduced, and the safety and stability of equipment and the environment are ensured.
The above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the scope of the claims.
Claims (7)
1. A multi-sensor based temperature monitoring method, comprising:
Acquiring temperature data of a monitoring area through a preset sensor network to obtain a temperature data set;
performing anomaly detection on the temperature data set through a preset isolation forest algorithm to obtain an anomaly detection report, wherein the anomaly detection report comprises: abnormal mode and temperature change data;
Inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend;
Based on the temperature change trend, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node;
Extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node, and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level, wherein the node group label specifically comprises the following steps: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
and respectively identifying a high-temperature region of the node group label of each sensor node by using an Euclidean distance method and a clustering algorithm to obtain a target high-temperature region, and carrying out air exhaust cooling treatment on the target high-temperature region.
2. The multi-sensor based temperature monitoring method according to claim 1, further comprising, before the step of collecting temperature data of a monitored area through a preset sensor network:
carrying out sensor layout analysis on the monitoring area through a genetic algorithm to obtain layout parameters, wherein the layout parameters comprise a plurality of sensor nodes;
respectively carrying out sensor parameter matching on each sensor node to obtain the sensor parameters of each sensor;
and configuring a sensor network based on the layout parameters and the sensor parameters of each sensor, and acquiring the temperature data of the monitoring area through the sensor network to obtain a temperature data set.
3. The multi-sensor based temperature monitoring method according to claim 1, wherein the abnormality detection is performed on the temperature data set by a preset isolation forest algorithm to obtain an abnormality detection report, and the abnormality detection report includes: an abnormal mode and temperature change data step comprising:
Extracting change characteristics of the temperature data set through the isolated forest algorithm to obtain temperature change characteristics;
Classifying the temperature change characteristics to obtain temperature mutation characteristics, long-term trend change characteristics and periodic change characteristics;
Setting the number of trees and subsampled capacity of the isolated forest algorithm based on the temperature abrupt change feature, the long-term trend change feature, and the periodic change feature;
carrying out isolation characteristic anomaly score analysis on the temperature mutation characteristic, the long-term trend change characteristic and the periodic change characteristic through an isolation forest algorithm after parameter setting to obtain an anomaly score set;
Based on a preset score threshold, carrying out abnormal condition identification on the temperature data set through the abnormal score set to obtain an abnormal detection report, wherein the abnormal detection report comprises: abnormal pattern and temperature change data.
4. The multi-sensor-based temperature monitoring method according to claim 1, wherein the step of inputting the abnormal pattern and the temperature change data into a preset long-short-term memory network to perform temperature trend prediction to obtain a temperature change trend comprises the steps of:
Data grouping is carried out on the abnormal mode and the temperature change data to obtain a plurality of groups of abnormal data;
respectively carrying out historical temperature average analysis on each group of abnormal data to obtain a historical temperature average corresponding to each group of abnormal data;
extracting the occurrence frequency of the abnormal event from each group of abnormal data respectively to obtain the occurrence frequency of the abnormal event corresponding to each group of abnormal data;
Fusing the historical temperature average value corresponding to each group of abnormal data and the abnormal event occurrence frequency corresponding to each group of abnormal data into abnormal temperature occurrence frequency;
inputting the abnormal temperature occurrence frequency into the long-short-period memory network to perform abnormal time interval matching to obtain a plurality of abnormal time intervals;
respectively predicting the temperature trend of each abnormal time interval to obtain the temperature change trend of each abnormal time interval;
And carrying out temperature trend fusion analysis on the temperature change trend of each abnormal time interval to obtain the temperature change trend.
5. The multi-sensor-based temperature monitoring method according to claim 1, wherein the step of matching the change trend data of each sensor node based on the temperature change trend to obtain a change trend data set, wherein the change trend data set includes a temperature change trend step of each sensor node, includes:
Carrying out time sequence synchronization on each sensor node through the temperature change trend, and extracting an identifier of each sensor node;
extracting key features of each sensor node based on the identifier of each sensor node, wherein the key features include: temperature peak, minimum temperature, and average temperature;
and based on the key characteristics of each sensor node and the temperature change trend, respectively carrying out change trend data matching on each sensor node to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node.
6. The multi-sensor-based temperature monitoring method according to claim 1, wherein the step of identifying the high-temperature area of the node group label of each sensor node by the euclidean distance method and the clustering algorithm to obtain a target high-temperature area, and performing the air exhaust cooling treatment on the target high-temperature area comprises the following steps:
grouping a plurality of sensor nodes through each sensor node to obtain a plurality of sensor groups;
Carrying out Euclidean distance calculation on each sensor group through the Euclidean distance method to obtain Euclidean distance sets;
Clustering the Euclidean distance set into sensor group areas through the clustering algorithm to obtain clustering areas;
calibrating the high-temperature region of the clustering region to obtain a target high-temperature region;
Collecting the current temperature of the target high-temperature area to obtain current temperature data;
and matching the current temperature data with a target exhaust force, and controlling a preset exhaust module to exhaust and cool the target high-temperature region based on the target exhaust force.
7. A multi-sensor based temperature monitoring system for performing the multi-sensor based temperature monitoring method of any one of claims 1 to 6, comprising:
the acquisition module is used for acquiring temperature data of the monitoring area through a preset sensor network to obtain a temperature data set;
the detection module is used for carrying out anomaly detection on the temperature data set through a preset isolation forest algorithm to obtain an anomaly detection report, wherein the anomaly detection report comprises the following components: abnormal mode and temperature change data;
The input module is used for inputting the abnormal mode and the temperature change data into a preset long-short-period memory network to predict the temperature trend, so as to obtain the temperature change trend;
The matching module is used for respectively matching the change trend data of each sensor node based on the temperature change trend to obtain a change trend data set, wherein the change trend data set comprises the temperature change trend of each sensor node;
The extracting module is used for extracting a high Wen Cancha sample of each sensor node in the temperature change trend of each sensor node and respectively converting the high Wen Cancha sample of each sensor node into a node group label with an alarm level, and specifically comprises the following steps: trend classification is carried out on the temperature change trend of each sensor node respectively to obtain the temperature rising trend of each sensor node; carrying out high Wen Cancha analysis on the temperature rising trend of each sensor node through a sliding window algorithm to obtain a high Wen Cancha sample of each sensor node; residual numerical conversion is carried out on the high Wen Cancha samples of each sensor node respectively, so that a high Wen Cancha numerical set of each sensor node is obtained; respectively carrying out alarm level mapping on the high Wen Cancha numerical value sets of each sensor node to obtain alarm level data of each sensor node; respectively converting a high Wen Cancha sample of each sensor node into a node group label with an alarm level based on alarm level data of each sensor node;
The identification module is used for respectively carrying out high-temperature area identification on the node group labels of each sensor node through a Euclidean distance method and a clustering algorithm to obtain a target high-temperature area, and carrying out air exhaust cooling treatment on the target high-temperature area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410571834.1A CN118152969B (en) | 2024-05-10 | 2024-05-10 | Temperature monitoring method and system based on multiple sensors |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410571834.1A CN118152969B (en) | 2024-05-10 | 2024-05-10 | Temperature monitoring method and system based on multiple sensors |
Publications (2)
Publication Number | Publication Date |
---|---|
CN118152969A CN118152969A (en) | 2024-06-07 |
CN118152969B true CN118152969B (en) | 2024-07-05 |
Family
ID=91290933
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410571834.1A Active CN118152969B (en) | 2024-05-10 | 2024-05-10 | Temperature monitoring method and system based on multiple sensors |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN118152969B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117308275A (en) * | 2023-11-28 | 2023-12-29 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
CN117633459A (en) * | 2023-11-29 | 2024-03-01 | 国网山西省电力公司太原供电公司 | High-precision distributed optical fiber sensing on-line monitoring device |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108777873B (en) * | 2018-06-04 | 2021-03-02 | 江南大学 | Wireless sensor network abnormal data detection method based on weighted mixed isolated forest |
-
2024
- 2024-05-10 CN CN202410571834.1A patent/CN118152969B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117308275A (en) * | 2023-11-28 | 2023-12-29 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
CN117633459A (en) * | 2023-11-29 | 2024-03-01 | 国网山西省电力公司太原供电公司 | High-precision distributed optical fiber sensing on-line monitoring device |
Also Published As
Publication number | Publication date |
---|---|
CN118152969A (en) | 2024-06-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110895526A (en) | Method for correcting data abnormity in atmosphere monitoring system | |
CN103812577A (en) | Method for automatically identifying and learning abnormal radio signal type | |
US20120271587A1 (en) | Equipment status monitoring method, monitoring system, and monitoring program | |
CN111898669B (en) | Abnormal event early warning system of direct-current submerged arc furnace based on machine learning | |
CN116148679B (en) | Battery health state prediction method and related device | |
CN117290802B (en) | Host power supply operation monitoring method based on data processing | |
CN113344133B (en) | Method and system for detecting abnormal fluctuation of time sequence behaviors | |
CN106936517A (en) | A kind of automatic recognition system and its method of abnormal radio signal | |
CN116823227A (en) | Intelligent equipment management system and method based on Internet of things | |
CN112734977B (en) | Equipment risk early warning system and algorithm based on Internet of things | |
CN117082105B (en) | Environment-friendly intelligent hospital facility monitoring system and method | |
CN113359637A (en) | Data quality guarantee system and method based on station house operation environment and equipment operation state | |
CN111181971A (en) | System for automatically detecting industrial network attack | |
CN116909339B (en) | Intelligent household safety early warning method and system based on artificial intelligence | |
CN117354171B (en) | Platform health condition early warning method and system based on Internet of things platform | |
KR20190082715A (en) | Data classification method based on correlation, and a computer-readable storege medium having program to perform the same | |
CN117318614A (en) | Photovoltaic inverter fault prediction method | |
CN118152969B (en) | Temperature monitoring method and system based on multiple sensors | |
CN117086699B (en) | Stainless steel member cutting temperature measurement method and system | |
JP7158624B2 (en) | Abnormality detection device | |
CN117494009A (en) | Electrical equipment state evaluation method based on insulating material pyrolysis analysis and cloud platform | |
CN112418281A (en) | Fire detection sensor data anomaly detection method and system | |
CN114429148A (en) | Power equipment state detection method based on multi-source data fusion | |
CN117970168B (en) | High-efficiency processing method for monitoring data of dual-power conversion device | |
Kim et al. | Fault detection method using inverse distance weight-based local outlier factor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |