Intelligent detection method for leakage of heating and ventilation pipeline

Abstract

The invention discloses a heating and ventilation pipeline leakage intelligent detection method, and belongs to the technical field of data processing, and the method comprises the steps: obtaining the monitoring data of a preset index of a to-be-detected heating and ventilation pipeline; setting the weight of other data according to the correlation between the other data in the monitoring data and the endpoint data, and fitting the endpoint data by using the other data according to the set weight to realize constraint optimization of the endpoint data; de-noising processing is performed on the monitoring data subjected to endpoint data constraint optimization; and intelligent detection of leakage of the heating and ventilation pipeline is realized based on the de-noised monitoring data. According to the scheme, the accuracy of leakage detection of the heating and ventilation pipeline can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent detection method for HVAC pipeline leakage. Background Art

[0002] Traditional HVAC pipeline leakage detection methods often rely on manual inspections or fixed sensors. This detection method has problems such as limited detection range, slow response speed, and high false alarm rate, and cannot meet the needs of remote, real-time, and accurate monitoring. With the development of artificial intelligence and Internet of Things technologies, intelligent detection technology has been widely used in the field of HVAC pipeline leakage monitoring. Intelligent detection technology obtains sensor data, processes the data, and intelligently determines whether there is a leak in the HVAC pipeline, which can meet the needs of remote, real-time, and accurate monitoring. However, when collecting pipeline monitoring data through sensors, the collected data often inevitably contains noise, and the presence of noise will cause problems such as false alarms or missed alarms, resulting in low detection accuracy. The existing denoising algorithms are usually affected by the endpoint effect, and the denoising effect is not ideal. Summary of the invention

[0003] The present invention provides an intelligent detection method for HVAC pipeline leakage, so as to solve the technical problem that the existing HVAC pipeline leakage detection method cannot effectively remove noise in data, resulting in low detection accuracy.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: An intelligent detection method for HVAC pipeline leakage, comprising: Obtain monitoring data of preset indicators of the HVAC pipeline to be tested; The weights of other data are set according to the correlation between other data and endpoint data in the monitoring data, and the endpoint data is fitted with other data according to the set weights to achieve constrained optimization of the endpoint data; De-noising the monitoring data that has completed the endpoint data constraint optimization; Intelligent detection of HVAC pipe leakage is achieved based on denoised monitoring data.

[0005] Furthermore, the weights of other data are set according to the correlation between other data in the monitoring data and the endpoint data, and the endpoint data is fitted using other data according to the set weights to achieve constrained optimization of the endpoint data, including: Perform interval division on the monitoring data, and divide the monitoring data into multiple data intervals; wherein the endpoint data in the monitoring data is separately divided into an interval, which is recorded as the endpoint data interval; Calculate the correlation between each data interval and the endpoint data interval; According to the correlation, similar data intervals of the endpoint data interval are screened out from all data intervals; Based on the correlation corresponding to each similar data interval, a weight is set for each similar data interval; According to the set weights of each similar data interval, the data in each similar data interval is used to fit the data in the endpoint data interval to complete the constrained optimization of the endpoint data.

[0006] Furthermore, the monitoring data is divided into multiple data intervals, including: Determine extreme values ​​in the monitoring data; wherein the extreme values ​​include maximum values ​​and minimum values; According to the distribution of extreme values ​​in the monitoring data, the monitoring data is divided into intervals, and the data between adjacent extreme values ​​are divided into the same interval, thereby dividing the monitoring data into multiple data intervals.

[0007] Furthermore, the calculation formula for calculating the correlation between each data interval and the endpoint data interval is: in, Indicates the endpoint data interval With The correlation between the data intervals; Indicates the endpoint data interval The average amplitude of the data points in ; Indicates The average amplitude of the data points in the data interval; Indicates the endpoint data interval The maximum amplitude of the data points in ; Indicates The maximum value of the amplitude of the data point in the data interval; Indicates the endpoint data interval The minimum amplitude of the data points in Indicates The minimum value of the amplitude of the data point in a data interval; the amplitude refers to the monitoring value of the preset indicator; Represents the endpoint data interval calculated by the dynamic time warping DTW algorithm With The DTW value between the data intervals, represents the linear normalization function; Represents the preset hyperparameter, whose value is greater than 0.

[0008] Furthermore, the method of screening out similar data intervals of the endpoint data interval from all data intervals according to the correlation includes: All data intervals are clustered by hierarchical clustering algorithm to obtain multiple clusters; Calculate the mean of the correlation corresponding to the data interval in each cluster; The data interval in the cluster with the largest mean is taken as the similar data interval of the endpoint data interval.

[0009] Furthermore, the formula for setting the weight of each similar data interval based on the correlation corresponding to each similar data interval is: in, Indicates The weights of similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Indicates the number of similar data intervals; Indicates The midpoint and endpoint data intervals of similar data intervals The Euclidean distance between the midpoints of .

[0010] Furthermore, according to the set weights of each similar data interval, the data in each similar data interval is used to fit the data in the endpoint data interval to complete the constrained optimization of the endpoint data, including: The similar data interval with the largest weight is taken as the reference interval, and the other intervals in all similar data intervals except the reference interval are taken as the fitting intervals, the fitting amplitude of each data point in the endpoint data interval is calculated, and the data in the endpoint data interval is fitted to complete the constrained optimization of the endpoint data; Among them, the calculation formula for the fitting amplitude of each data point in the endpoint data interval is: in, Indicates the endpoint data interval The The fitted amplitude of the data points; Indicates the first The amplitude of the data point; Indicates The weights of the fitting intervals; Indicates The first The amplitude of the data point; Indicates the number of similar data intervals; Represents the preset hyperparameter, whose value is greater than 0.

[0011] Furthermore, the preset indicator is the pressure value of the HVAC pipe to be detected.

[0012] Furthermore, the denoising process of the monitoring data after the endpoint data constraint optimization is completed includes: The complementary collective mode decomposition (CEEMD) algorithm is used to denoise the monitoring data after completing endpoint data constraint optimization.

[0013] Furthermore, intelligent detection of HVAC pipe leakage is achieved based on the denoised monitoring data, including: Compare the denoised monitoring data with the preset threshold; When the denoised monitoring data is less than the preset threshold, it is determined that the HVAC pipe is leaking.

[0014] The beneficial effects brought about by the technical solution provided by the present invention include at least: The present invention obtains the correlation of changes in different data segments in the original data by analyzing the original data, then sets the data weight according to the data correlation, and then performs constraint optimization on the endpoint data according to the data weight, and adds the data so that the endpoint data contains complete maximum and minimum points, so that when the envelope is fitted according to the extreme points, there will be no loss of data points, so that the data changes are more consistent with the changes in the original data, thereby overcoming the influence of the endpoint effect, effectively improving the denoising effect of the data, and then effectively improving the accuracy of HVAC pipeline leakage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 It is a schematic diagram of the execution flow of the intelligent detection method for HVAC pipeline leakage provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a process for constraining optimization of endpoint data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0018] This embodiment provides a method for intelligent detection of HVAC pipeline leakage. The execution process of the method is as follows: Figure 1 As shown, the following steps are included: S1, obtaining monitoring data of preset indicators of the HVAC pipeline to be tested; It should be noted that the main purpose of this embodiment is to analyze the monitoring data of the HVAC pipeline to determine whether the HVAC pipeline has leaked. To this end, this embodiment selects the pressure of the HVAC pipeline as the monitoring indicator, that is, to monitor the pressure at different points of the pipeline network. If an abnormal pressure drop is detected, it indicates that a leak may have occurred. Specifically, this embodiment arranges intelligent pressure sensors at different points on the HVAC pipeline to obtain pressure data at different points on the pipeline, and transmits the real-time collected pressure data to the data processing system, providing a data basis for subsequent data processing steps.

[0019] S2, setting weights of other data according to the correlation between other data in the monitoring data and the endpoint data, and fitting the endpoint data with other data according to the set weights to achieve constrained optimization of the endpoint data; Among them, it should be noted that when judging whether a pipeline leaks based on monitoring data, because the data obtained through sensor collection contains noise, and the noise will affect the accuracy of data recognition, it is necessary to first denoise the obtained data. A major challenge of data denoising is how to retain important features and information in the signal while removing noise. Excessive denoising may remove important signal details, while insufficient denoising cannot provide clean data. In addition, for real-time monitoring systems, the denoising algorithm needs to be fast enough and have a high enough execution efficiency so as not to delay the response time of signal processing and leak detection. In this regard, this embodiment selects the complementary set mode decomposition CEEMD algorithm as the denoising algorithm.

[0020] The complementary collective mode decomposition (CEEMD) algorithm is a time series analysis method for processing nonlinear and non-stationary signals. It is an improved version of the empirical mode decomposition (EMD) algorithm, which aims to overcome the modal aliasing problem that may be encountered in the decomposition process of EMD. In order to overcome modal aliasing, the CEEMD algorithm adds a set of different white noise signals to the original signal in each iteration. By performing EMD decomposition on a signal set containing random noise, a set of IMFs can be obtained. Then the corresponding IMFs of all noise auxiliary data sets are averaged to eliminate the influence of noise. However, when decomposing a signal, it is usually affected by the endpoint effect. The endpoint effect refers to the artificially introduced oscillation or inaccurate pseudo-component that may appear at the beginning and end of the signal, which is particularly prominent when the signal length is limited. This will cause the uncertainty of the edge area of ​​the signal to increase when the signal is decomposed, affecting the reliability of the entire signal analysis.

[0021] In order to overcome the impact of the endpoint effect, this embodiment analyzes the original data to obtain the correlation of changes in different data segments in the original data, and then sets the data weight according to the data correlation, and then performs constraint optimization on the endpoint data according to the data weight, thereby adding data. Specifically, in this embodiment, if Figure 2 As shown in Figure 1, the process of constrained optimization of endpoint data is as follows: S21, dividing the monitoring data into multiple data intervals; wherein the endpoint data in the monitoring data is separately divided into an interval, which is recorded as the endpoint data interval; Specifically, the interval division method is as follows: determine the extreme values ​​in the monitoring data; wherein the extreme values ​​include the maximum value and the minimum value; divide the monitoring data into intervals according to the distribution of the extreme values ​​in the monitoring data, divide the data between adjacent extreme values ​​into the same interval, and thus divide the monitoring data into multiple data intervals. Because the endpoint data only contains one extreme value point, it is a separate data interval, recorded as the endpoint data interval.

[0022] S22, calculating the correlation between each data interval and the endpoint data interval; Among them, the calculation formula for the correlation between each data interval and the endpoint data interval is: in, Indicates the endpoint data interval With The correlation between the data intervals; Indicates the endpoint data interval The average amplitude of the data points in ; Indicates The average amplitude of the data points in the data interval; Indicates the endpoint data interval The maximum amplitude of the data points in ; Indicates The maximum value of the amplitude of the data point in the data interval; Indicates the endpoint data interval The minimum amplitude of the data points in Indicates The minimum value of the amplitude of the data point in the data interval; the amplitude refers to the monitoring value of the preset indicator, that is, the pressure value; Represents the endpoint data interval calculated by the dynamic time warping DTW algorithm With The DTW value between the data intervals, represents the linear normalization function; It is a preset hyperparameter, which is a number greater than 0 such as 1, 0.1, or 0.01.

[0023] Indicates the endpoint data interval With The difference in the amplitude of the two data intervals is greater, the greater the difference in the average amplitude of the two data intervals, the greater the difference in the degree of data change in the two data intervals, and therefore the smaller the correlation; Indicates the endpoint data interval With The difference between the maximum values ​​of the data intervals, Indicates the endpoint data interval With The difference between the minimum values ​​of the two data intervals. The smaller the difference in extreme values, the more similar the ranges of the two data intervals are. The more similar the causes of data changes in the two data intervals are. At this time, the greater the correlation between the two data intervals. Therefore, the smaller the difference in extreme values ​​of the two intervals, the greater the correlation. Indicates the endpoint data interval With The DTW value between two data intervals indicates the difference between the two data intervals. The larger the value, the smaller the correlation.

[0024] S23, screening out similar data intervals of the endpoint data interval from all data intervals according to the correlation; Specifically, the screening method for similar data intervals of endpoint data intervals is: clustering all data intervals through a hierarchical clustering algorithm to obtain multiple clusters; wherein the number of clustering layers is set to 2; after the clustering is completed, calculating the mean of the correlation corresponding to the data intervals in each cluster; and taking the data interval in the cluster with the largest mean as the similar data interval of the endpoint data interval.

[0025] S24, setting a weight for each similar data interval based on the correlation corresponding to each similar data interval; Among them, it should be noted that when processing the endpoint data, because the endpoint data does not contain complete maximum points or minimum points, it will cause inaccurate data fitting when decomposing, resulting in errors when decomposing the data. Therefore, in order to avoid inaccurate denoising due to the incompleteness of the endpoint data, this embodiment fills the endpoint data by analyzing the correlation between the endpoint data and other data intervals, so that the endpoint data contains complete maximum points and minimum points, so that when the envelope is fitted according to the extreme points, there will be no loss of data points. To this end, this embodiment obtains the weight of the similar data interval based on the difference between the similar data interval of the endpoint data interval obtained above and the endpoint data interval. Among them, the calculation formula of the weight of the similar data interval is as follows: in, Indicates The weights of similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Represents the number of similar data intervals; Indicates The midpoint and endpoint data intervals of similar data intervals The Euclidean distance between the midpoints of .

[0026] Indicates the endpoint data interval With The larger the ratio of the correlation between similar data intervals in the entire data sequence, the more similar the data change in the data segment in the current similar data interval is to the endpoint data. This is because the closer the data in two data intervals are, the more similar the reasons for the changes in the data in the two data intervals may be. Therefore, the smaller the distance, the larger the weight should be.

[0027] S25, according to the set weights of the similar data intervals, the data in the endpoint data intervals are fitted using the data in the similar data intervals to complete the constrained optimization of the endpoint data.

[0028] It should be noted that, in the process of filling the endpoint data interval, it is necessary to constrain the endpoint data interval according to the weights of different similar data intervals, so that the filled data will not be distorted, and the data changes of the filled endpoint data interval are more consistent with the changes of the original data, so that when the data is decomposed by EMD, the data can be better fitted. In this regard, the process of constraining and optimizing the endpoint data in this embodiment is as follows: The similar data interval with the largest weight is taken as the reference interval, and the other intervals in all similar data intervals except the reference interval are taken as the fitting intervals, the fitting amplitude of each data point in the endpoint data interval is calculated, and the data in the endpoint data interval is fitted to complete the constrained optimization of the endpoint data; Among them, the calculation formula for the fitting amplitude of each data point in the endpoint data interval is: in, Indicates the endpoint data interval The The fitted amplitude of the data points; Indicates the first The amplitude of the data point, Indicates The weights of the fitting intervals; Indicates The first The amplitude of the data point; Represents the number of similar data intervals.

[0029] Indicates The weight of a fitting interval is multiplied by the The amplitude of the data point, because the larger the weight, the more similar the interval is to the endpoint data interval, so when fitting the data, the contribution of the data points in this interval should also be greater; Indicates the reference interval Data points and In the fitting interval The amplitude difference of the data points is the same as the amplitude difference of the data points. Because this embodiment takes the similar data interval with the greatest similarity as the reference interval, and then constrains the reference interval by the similarity between other similar data intervals and the endpoint data interval, so that when fitting the endpoint data interval, it can better conform to the variation law of the original data. Therefore, here we divide by This is to indicate the difference between the reference interval and the The smaller the difference between the data points corresponding to the first fitting interval, the The greater the contribution of a fitting interval, the greater the influence of the fitting interval on the fitting value when fitting the data.

[0030] S3, denoising the monitoring data after completing the endpoint data constraint optimization; Specifically, after obtaining the fitting value of each data point in the endpoint data interval according to the above calculation, this embodiment records the processed monitoring data as target data, and then uses the CEEMD algorithm to denoise the target data to obtain denoised data; wherein, the CEEMD algorithm is a prior art, so it will not be described in detail here.

[0031] S4, realizes intelligent detection of HVAC pipe leakage based on denoised monitoring data; Among them, the detection method adopted in this embodiment is: comparing the denoised monitoring data with a preset threshold; when the denoised monitoring data is less than the preset threshold, it is determined that the HVAC pipe is leaking.

[0032] Specifically, this embodiment sets the threshold When the monitored pipeline pressure data is less than When the pipeline leaks, it indicates that the pressure may drop and an alarm will be issued.

[0033] In summary, this embodiment provides an intelligent detection method for HVAC pipe leakage, which obtains the correlation of changes in different data segments in the original data by analyzing the original data, and then sets the data weight according to the data correlation, and then performs constraint optimization on the endpoint data according to the data weight, and adds the data so that the endpoint data contains complete maximum and minimum points, so that when the envelope is fitted according to the extreme points, there will be no loss of data points, so that the data changes are more consistent with the changes in the original data, thereby overcoming the influence of the endpoint effect, effectively improving the denoising effect of the data, and then effectively improving the accuracy of HVAC pipe leakage detection.

[0034] In addition, it should be noted that the present invention can be provided as a method, an apparatus or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0035] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0036] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0037] It should also be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0038] Finally, it should be noted that the above is a preferred embodiment of the present invention. It should be pointed out that although the preferred embodiment of the present invention has been described, for those skilled in the art, once the basic creative concept of the present invention is known, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A method for intelligent detection of HVAC pipeline leakage, characterized in that: include: Obtain monitoring data of preset indicators of the HVAC pipeline to be tested; The weights of other data are set according to the correlation between other data and endpoint data in the monitoring data. Based on the set weights, other data are used to fit the endpoint data to achieve constrained optimization of the endpoint data, including: Perform interval division on the monitoring data, and divide the monitoring data into multiple data intervals; wherein the endpoint data in the monitoring data is separately divided into an interval, which is recorded as the endpoint data interval; Calculate the correlation between each data interval and the endpoint data interval; According to the correlation, similar data intervals of the endpoint data interval are screened out from all data intervals; Based on the correlation corresponding to each similar data interval, a weight is set for each similar data interval; According to the set weights of each similar data interval, the data in each similar data interval is used to fit the data in the endpoint data interval to complete the constrained optimization of the endpoint data; De-noising the monitoring data that has completed the endpoint data constraint optimization; Intelligent detection of HVAC pipe leakage is achieved based on denoised monitoring data; The calculation formula for calculating the correlation between each data interval and the endpoint data interval is: in, Indicates the endpoint data interval With The correlation between the data intervals; Indicates the endpoint data interval The average amplitude of the data points in ; Indicates The average amplitude of the data points in the data interval; Indicates the endpoint data interval The maximum amplitude of the data points in ; Indicates The maximum value of the amplitude of the data point in the data interval; Indicates the endpoint data interval The minimum amplitude of the data points in the Indicates The minimum value of the amplitude of the data point in a data interval; the amplitude refers to the monitoring value of the preset indicator; Represents the endpoint data interval calculated by the dynamic time warping DTW algorithm With The DTW value between the data intervals, represents the linear normalization function; Represents the preset hyperparameter, whose value is greater than 0.

2. The intelligent detection method for HVAC pipeline leakage according to claim 1, characterized in that: The monitoring data is divided into multiple data intervals, including: Determine extreme values ​​in the monitoring data; wherein the extreme values ​​include maximum values ​​and minimum values; According to the distribution of extreme values ​​in the monitoring data, the monitoring data is divided into intervals, and the data between adjacent extreme values ​​are divided into the same interval, thereby dividing the monitoring data into multiple data intervals.

3. The intelligent detection method for HVAC pipeline leakage according to claim 1, characterized in that: The method of screening out similar data intervals of the endpoint data interval from all data intervals based on the correlation includes: All data intervals are clustered by hierarchical clustering algorithm to obtain multiple clusters; Calculate the mean of the correlation corresponding to the data interval in each cluster; The data interval in the cluster with the largest mean is taken as the similar data interval of the endpoint data interval.

4. The intelligent detection method for HVAC pipeline leakage according to claim 1, characterized in that: The formula for setting the weight of each similar data interval based on the correlation corresponding to each similar data interval is: in, Indicates The weights of similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Indicates the endpoint data interval With The correlation between similar data intervals; Represents the number of similar data intervals; Indicates The midpoint and endpoint data intervals of similar data intervals The Euclidean distance between the midpoints of .

5. The intelligent detection method for HVAC pipeline leakage according to claim 1, characterized in that: The method of fitting the data in the endpoint data interval using the data in each similar data interval according to the set weights of each similar data interval to complete the constrained optimization of the endpoint data includes: The similar data interval with the largest weight is taken as the reference interval, and the other intervals in all similar data intervals except the reference interval are taken as the fitting intervals, the fitting amplitude of each data point in the endpoint data interval is calculated, and the data in the endpoint data interval is fitted to complete the constrained optimization of the endpoint data; Among them, the calculation formula for the fitting amplitude of each data point in the endpoint data interval is: in, Indicates the endpoint data interval The The fitted amplitude of the data points; Indicates the first The amplitude of the data point; Indicates The weights of the fitting intervals; Indicates The first The amplitude of the data point; Represents the number of similar data intervals; Represents the preset hyperparameter, whose value is greater than 0.

6. The intelligent detection method for HVAC pipeline leakage according to claim 1, characterized in that: The preset indicator is the pressure value of the HVAC pipeline to be detected.

7. The intelligent detection method for HVAC pipeline leakage according to claim 6, characterized in that: The denoising process of the monitoring data after the endpoint data constraint optimization is completed includes: The complementary collective mode decomposition (CEEMD) algorithm is used to denoise the monitoring data after completing endpoint data constraint optimization.

8. The intelligent detection method for HVAC pipeline leakage according to claim 7, characterized in that: The intelligent detection of HVAC pipeline leakage based on the denoised monitoring data includes: Compare the denoised monitoring data with the preset threshold; When the denoised monitoring data is less than the preset threshold, it is determined that the HVAC pipe is leaking.

Images