CN117167903B - Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment - Google Patents
Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment Download PDFInfo
- Publication number
- CN117167903B CN117167903B CN202311454671.0A CN202311454671A CN117167903B CN 117167903 B CN117167903 B CN 117167903B CN 202311454671 A CN202311454671 A CN 202311454671A CN 117167903 B CN117167903 B CN 117167903B
- Authority
- CN
- China
- Prior art keywords
- data
- sequence
- exhaust gas
- window
- temperature
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 21
- 238000009423 ventilation Methods 0.000 title claims abstract description 17
- 238000010438 heat treatment Methods 0.000 title claims abstract description 16
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 12
- 238000012544 monitoring process Methods 0.000 claims abstract description 69
- 230000035772 mutation Effects 0.000 claims abstract description 48
- 230000008859 change Effects 0.000 claims abstract description 39
- 238000004378 air conditioning Methods 0.000 claims abstract description 22
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 42
- 238000006073 displacement reaction Methods 0.000 claims description 27
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 230000005856 abnormality Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000011002 quantification Methods 0.000 description 2
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of air conditioning equipment detection, and provides a heating ventilation equipment foreign matter fault detection method based on artificial intelligence, which comprises the following steps: collecting exhaust gas quantity data, temperature data, humidity data and pressure data of air conditioning equipment to obtain a plurality of exhaust gas quantity sequences; obtaining a plurality of extreme points and interpolation sections for each exhaust gas quantity sequence; obtaining a plurality of transverse windows of each exhaust sequence; obtaining mutation degree and exhaust gas quantity change characteristics of each transverse window; obtaining a cubic spline curve formed by boundary conditions and interpolation functions; and (3) carrying out EMD decomposition and reconstruction according to a cubic spline curve to obtain denoising exhaust gas quantity data of each exhaust gas quantity sequence, acquiring denoising temperature data, denoising humidity data and denoising pressure data of the same monitoring period, and carrying out foreign matter fault detection of equipment according to the correlation of the denoising exhaust gas quantity data and denoising data of other dimensions. The invention aims to solve the problem that the multi-dimensional data anomaly detection of heating and ventilation equipment is interfered by noise data.
Description
Technical Field
The invention relates to the technical field of air conditioning equipment detection, in particular to a warm ventilation equipment foreign matter fault detection method based on artificial intelligence.
Background
The heating ventilation equipment comprises a plurality of types, the air conditioner is heating ventilation equipment, and for the air conditioner, the fault caused by the foreign matter is usually that the foreign matter blocks the air outlet, so that the exhaust gas quantity of the air conditioner is abnormal, and meanwhile, the abnormality of data of a plurality of dimensions such as the temperature, the moderate degree and the pressure of the air conditioner can be caused, so that the abnormality detection of the exhaust gas quantity is carried out by analyzing the correlation between the exhaust gas quantity and the time sequence data of each dimension, and the fault detection of the foreign matter of the air conditioner is realized.
Noise data exists in each dimension data of the time sequence of the air conditioning equipment, and the EMD decomposition and reconstruction can remove the noise data, so that the accuracy of a foreign matter fault detection result of the air conditioning equipment is ensured; however, the time sequence data with single dimension has fluctuation and mutation, the generation of abnormal data is continuous, and noise data is randomly generated, so that when three spline curve fitting is used in the EMD decomposition process, the fitting results of upper and lower envelopes are affected, the envelopes cannot accurately fit the local fluctuation change of the time sequence data, fitting and overfitting phenomena are generated, the denoising result of the time sequence data with each dimension is inaccurate, and the accuracy of the foreign matter fault detection result of the air conditioning equipment is reduced.
Disclosure of Invention
The invention provides an artificial intelligence-based foreign matter fault detection method for heating and ventilation equipment, which aims to solve the problem that the multi-dimensional data anomaly detection of the conventional heating and ventilation equipment is interfered by noise data, and adopts the following specific technical scheme:
the embodiment of the invention provides an artificial intelligence-based foreign matter fault detection method for heating and ventilation equipment, which comprises the following steps of:
collecting exhaust gas data, temperature data, humidity data and pressure data of air conditioning equipment, and obtaining a plurality of exhaust gas sequences according to the exhaust gas data of each monitoring period;
obtaining a plurality of extreme points and interpolation sections for each exhaust gas quantity sequence; obtaining a plurality of transverse windows of each exhaust sequence according to the slope and distribution of extreme points in the exhaust sequence and other dimension data and the interpolation section;
according to the exhaust amount data and the slope of the data points in the transverse windows, the mutation degree and the exhaust amount change characteristic of each transverse window are obtained; obtaining a cubic spline curve formed by boundary conditions of each transverse window and interpolation functions according to the displacement change characteristics and the slope of the extreme points;
and (3) carrying out EMD decomposition and reconstruction according to a cubic spline curve to obtain denoising exhaust gas quantity data of each exhaust gas quantity sequence, acquiring denoising temperature data, denoising humidity data and denoising pressure data of the same monitoring period, and carrying out foreign matter fault detection of equipment according to the correlation of the denoising exhaust gas quantity data and denoising data of other dimensions.
Further, the specific method for obtaining a plurality of extreme points and interpolation segments for each exhaust gas amount sequence includes:
for any one exhaust gas amount sequence, acquiring a data curve of the exhaust gas amount sequence; and acquiring a plurality of extreme points of the data curve, wherein the curve between adjacent extreme points is used as an interpolation section, and a plurality of interpolation sections are obtained for the displacement sequence.
Further, the method for obtaining the plurality of transverse windows of each exhaust gas amount sequence comprises the following specific steps:
obtaining a plurality of reserved extreme points and noise extreme points of each exhaust volume sequence according to the slope and distribution of the extreme points in the exhaust volume sequence and other dimension data; for any one exhaust gas amount sequence, acquiring the length of each interpolation section, and recording 2 times of the maximum value in all lengths as the minimum width of a window of the exhaust gas amount sequence;
starting from the first data point of the exhaust sequence, framing the data points with the minimum width of the window, and if the last data point framed is a reserved extreme point, forming a transverse window by all the data points framed; if the last data point selected by the frame is not the reserved extreme point, continuing to select the frame to the rear frame until the frame selects one reserved extreme point, and taking all the data points selected by the frame as a transverse window;
after the first transverse window is obtained, starting from the next data point of the transverse window, continuously obtaining the transverse window until all data points in the exhaust volume sequence are divided into corresponding transverse windows, and obtaining a plurality of transverse windows of the exhaust volume sequence.
Further, the specific method for obtaining the plurality of reserved extremum points and noise extremum points of each exhaust gas quantity sequence includes:
for any exhaust gas amount sequence, acquiring the slope of each extreme point; acquiring a monitoring period corresponding to the exhaust gas amount sequence, and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, extreme points in each sequence and the slope of each extreme point;
for the temperature sequence of the monitoring period, acquiring cosine similarity between the temperature sequence and the exhaust gas sequence, and respectively acquiring an extreme point curve of the exhaust gas sequence and an extreme point curve of the temperature sequence if the cosine similarity is more than or equal to 0; if the cosine similarity is smaller than 0, acquiring an extreme point curve of the exhaust gas quantity sequence, acquiring coordinate points of each extreme point in the temperature sequence, which are symmetrical about a transverse axis, connecting adjacent coordinate points, and recording the acquired curve as the extreme point curve of the temperature sequence;
performing DTW matching on the two extreme point curves to obtain extreme points matched with each extreme point in the exhaust gas sequence in the temperature sequence, and marking the extreme points as temperature matched extreme points of each extreme point in the exhaust gas sequence;
for any extreme point in the exhaust gas quantity sequence, obtaining the product of the slope of the extreme point and the slope of the temperature matching extreme point, and recording the product as the temperature correlation of the extreme point; acquiring the temperature correlation of the extreme point adjacent to the previous extreme point and the temperature correlation of the extreme point adjacent to the next extreme point in the exhaust gas quantity sequence, and respectively marking the temperature correlation as the left temperature correlation and the right temperature correlation of the extreme point;
if the signs of the temperature correlations are the same as those of the left temperature correlations and those of the right temperature correlations, no noise exists between the extreme point and the temperature dimension; if the sign of the temperature correlation is different from the sign of the left temperature correlation or the sign of the right temperature correlation, the extreme point is marked as a noise extreme point;
noise judgment is carried out on the extreme point in the humidity dimension and the pressure dimension respectively, and if no noise exists in the three dimensions, the extreme point is marked as a reserved extreme point; if any one dimension is judged to be a noise extreme point, the extreme point is marked as a noise extreme point; and carrying out noise judgment on all extreme points of the exhaust gas quantity sequence to obtain a plurality of reserved extreme points and noise extreme points.
Further, the mutation degree and the exhaust gas amount change characteristic of each transverse window are obtained by the specific method that:
for any one exhaust gas amount sequenceA lateral window, obtain->Slope of each data point within the lateral windows; get->A number of data points with a slope greater than 0 and a number of data points with a slope less than 0 within a lateral window, adding the sum of the maximum of the two numbers plus the number of data points with a slope equal to 0 to +.>Transverse windowsThe ratio of the number of intraoral data points is denoted +.>A variation factor of the lateral windows; the exhaust gas amount sequence is +.>Mutation degree of the individual transverse windows->The calculation method of (1) is as follows:
wherein,indicate->The variation factor of the individual lateral windows->Indicate->The number of data points in a single transverse window,indicate->The (th) of the lateral windows>Data point exhaust data +.>Indicate->Displacement data of the first data point in the lateral window, +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Representing absolute value;
and obtaining the variation characteristic of the exhaust gas quantity of each transverse window according to the mutation degree of the transverse window in the exhaust gas quantity sequence.
Further, the specific method for obtaining the exhaust gas quantity change characteristic of each transverse window includes the following steps:
for any one exhaust sequence, obtaining the mutation degree of each transverse window in the exhaust sequence, carrying out linear normalization on all mutation degrees, marking the obtained result as a mutation factor of each transverse window, setting two exhaust change characteristics, namely a transverse window with stable data and a transverse window with abrupt data; marking a transverse window with the mutation factor larger than the mutation threshold value as a transverse window of data mutation, and marking the transverse window as a mutation window; marking a transverse window with the mutation factor smaller than or equal to the mutation threshold value as a transverse window with stable data, and marking the transverse window as a stable window; and obtaining the exhaust gas quantity change characteristics of each transverse window in the exhaust gas quantity sequence to obtain a plurality of stable windows and abrupt change windows.
Further, the method for obtaining the cubic spline curve formed by the boundary condition of each transverse window and the interpolation function comprises the following specific steps:
for the stable window, setting the first derivative and the second derivative of two data points of the boundary of the stable window to be 0, and obtaining the boundary condition of the stable window; for a non-stationary window, setting the second derivative of two data points of the boundary of the non-stationary window to 0, wherein the first derivative is the slope and is not changed, so as to obtain the boundary condition of the abrupt window;
and obtaining an interpolation function of each transverse window according to the boundary condition of the transverse window and forming a cubic spline curve of each displacement sequence.
Further, the method for obtaining the interpolation function of each transverse window and forming the cubic spline curve of each displacement sequence comprises the following specific steps:
for any exhaust volume sequence, the interpolation function adopts a cubic equation, the interpolation function of each transverse window needs to meet the boundary condition at the end point, and a cubic spline curve of the exhaust volume sequence is obtained through cubic spline fitting and connection of interpolation functions of different transverse windows.
Further, the method for obtaining the denoising exhaust gas amount data of each exhaust gas amount sequence and obtaining denoising temperature data, denoising humidity data and denoising pressure data of the same monitoring period comprises the following specific steps:
for the exhaust gas quantity sequence of the latest monitoring period, carrying out EMD (empirical mode decomposition) on a data curve of the exhaust gas quantity sequence according to the acquired cubic spline curve to obtain a plurality of IMF components, reconstructing the IMF components, and recording the reconstructed time sequence data as denoising exhaust gas quantity data; and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, and acquiring denoising temperature data, denoising humidity data and denoising pressure data.
Further, the method for detecting the foreign matter fault of the equipment according to the correlation between the denoising exhaust gas amount data and the denoising data in other dimensions comprises the following specific steps:
acquiring an exhaust gas amount sequence and denoising exhaust gas amount data of each monitoring period, and each temperature sequence and denoising temperature data, calculating a clearman correlation coefficient for the denoising exhaust gas amount data and denoising temperature data of the same period, and recording the correlation degree of the exhaust gas amount data and the temperature data of each monitoring period;
calculating a mean value of all the correlation degrees except the latest monitoring period, and taking the maximum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold value if the mean value is smaller than 0; if the average value is greater than or equal to 0, taking the minimum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold;
acquiring humidity sequence and denoising humidity data of each monitoring period, and pressure sequence and denoising pressure data of each monitoring period, and obtaining a humidity correlation threshold, a pressure correlation threshold and the correlation degree of the exhaust gas quantity data of the latest monitoring period with the humidity data and the correlation degree of the pressure data respectively;
for any dimension of temperature, humidity and pressure, if the correlation threshold value of the dimension is obtained by the minimum value, and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is greater than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment; if the correlation threshold value of the dimension is obtained by the maximum value and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is smaller than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment;
and carrying out abnormal judgment on the correlation degree of the exhaust amount data in the latest time period and the data in each dimension, wherein the abnormality occurs in the judgment of one dimension, and the air conditioning equipment has foreign matter faults.
The beneficial effects of the invention are as follows: according to the invention, through the data correlation relationship between the air displacement data of the air conditioner and other dimension data in time sequence, the foreign matter fault detection in the air outlet of the air conditioner is carried out, in the correlation analysis process, the time sequence data of each dimension is denoised through EMD decomposition and reconstruction, so that the quantification of the correlation relationship of different dimensions is not influenced due to the existence of noise data in the analysis process, and the accuracy of the foreign matter fault detection is further improved; in the EMD decomposition process, the boundary conditions are corrected, so that the cubic spline curve formed by the interpolation function cannot be over-fitted and under-fitted, smooth connection between data points is ensured, meanwhile, the noise extreme points are highlighted, the accuracy of the denoising result of the time sequence data of each dimension is further improved, and the accuracy of the quantification of the correlation relationship is ensured.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for detecting a foreign matter fault of a heating ventilation device based on artificial intelligence according to an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. 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.
Referring to fig. 1, a flowchart of a method for detecting foreign matter faults of an artificial intelligence-based heating and ventilation device according to an embodiment of the present invention is shown, the method includes the following steps:
and S001, collecting exhaust gas quantity data, temperature data, humidity data and pressure data of the air conditioning equipment, and obtaining a plurality of exhaust gas quantity sequences according to the exhaust gas quantity data of each monitoring period.
The purpose of this embodiment is to detect a foreign matter fault of an air conditioner by using the air displacement data and other dimension data of the air conditioner, so that the air displacement data and the dimension data need to be acquired first; in the embodiment, every 30 minutes after the air conditioner starts working is taken as a monitoring period, if the air conditioner stops running or is stopped until the last monitoring period at the current moment is less than 30 minutes, the actually obtained period is taken as the monitoring period, and no completion is needed; according to the embodiment, data of each dimension are obtained through various sensors, sampling frequencies of all the sensors are consistent, the data are collected every 30 seconds, indoor temperature data and humidity data of air conditioner work are obtained through a temperature sensor and a humidity sensor, pressure data of air conditioning equipment are obtained through a pressure sensor, and exhaust volume data of every 30 seconds in an air outlet of the air conditioner are obtained through a flow sensor.
Further, the exhaust gas amount data of each monitoring period is arranged according to the acquisition sequence, so that an exhaust gas amount sequence of each monitoring period can be obtained, and a plurality of exhaust gas amount sequences including an exhaust gas amount sequence of the latest monitoring period corresponding to the current moment are obtained.
Thus, the air displacement data, the temperature data, the humidity data and the pressure data of the air conditioning equipment and a plurality of air displacement sequences are obtained.
Step S002, obtaining a plurality of extreme points and interpolation segments for each exhaust gas amount sequence; and obtaining a plurality of transverse windows of each exhaust sequence according to the slope and the distribution of the extreme points in the exhaust sequence and other dimension data and the interpolation section.
It should be noted that, the data generated by the air conditioning equipment due to the influence of the foreign matter has an abnormality, and the abnormality includes abrupt change, transient state, etc. of the exhaust amount data; in the process of denoising the exhaust gas amount data through EMD decomposition and reconstruction, the original data is needed to be fit into an upper envelope line and a lower envelope line for subsequent analysis; a cubic spline curve is used for obtaining the interpolation function, and in an environment where the exhaust volume data is abnormal, the cubic spline curve is limited by the principle of the interpolation function because the cubic spline curve is obtained, so that the phenomenon of over-fitting or under-fitting of the cubic spline curve can occur; therefore, the distribution of data points in the exhaust gas volume sequence needs to be analyzed, a transverse window is obtained through the quantized interpolation section, and a foundation is provided for the acquisition of the exhaust gas volume change characteristics of the subsequent transverse window and the final cubic spline curve.
It should be further noted that, there are several extremum points in the exhaust gas amount sequence, where the extremum points may cause local data abnormality, and it is necessary to obtain interpolation segments according to the extremum points first, and then divide subsequent transverse windows; if the interpolation function is quantized directly from the entire displacement sequence, local data anomalies are ignored, resulting in inaccuracy of the cubic spline curve fitting.
Specifically, for any one exhaust gas volume sequence, a data curve of the exhaust gas volume sequence is obtained, namely, the sequence value of each exhaust gas volume data in the exhaust gas volume sequence is taken as an abscissa, and the data value is taken as an ordinate to obtain the data curve; acquiring a plurality of extreme points of a data curve, wherein the curve between adjacent extreme points is used as an interpolation section, and then a plurality of interpolation sections are obtained for the exhaust gas quantity sequence, and the extreme points are acquired as a known technology, so that the embodiment is not repeated; and obtaining a plurality of interpolation segments of each exhaust gas quantity sequence according to the method.
It should be further noted that, after the extremum points and the interpolation segments in the data curve of the exhaust gas sequence are obtained, the data in the same monitoring period in different dimensions will show correlation, that is, if no noise influence exists, the number of the extremum points in different dimensions should be the same, and the overall variation trend is the same, through the correlation distribution, the extremum points possibly caused by noise in the data curve of the exhaust gas sequence are removed according to the slope distribution of the data points, and then a transverse window is constructed according to the remaining extremum points, and the subsequent exhaust gas variation characteristics are obtained through the transverse window and the boundary condition is determined; in the process of obtaining the transverse window, the minimum width of the transverse window is determined through the interpolation section, and the transverse window is obtained by combining extreme point distribution.
Specifically, for any one exhaust gas amount sequence, acquiring the slope of each extreme point; acquiring a monitoring period corresponding to the exhaust gas amount sequence, and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, extreme points in each sequence and the slope of each extreme point according to the method; taking the temperature sequence of the monitoring period as an example, acquiring cosine similarity of the temperature sequence and the exhaust gas sequence, if the cosine similarity is greater than or equal to 0, indicating that the temperature sequence and the exhaust gas sequence are positively correlated in the monitoring period, respectively acquiring an extreme point curve of the exhaust gas sequence and an extreme point curve of the temperature sequence, namely extracting extreme points in the data curve, and connecting adjacent extreme points; if the cosine similarity is smaller than 0, the temperature sequence and the exhaust gas quantity sequence are indicated to be in negative correlation in the monitoring period, an extreme point curve of the exhaust gas quantity sequence is obtained, coordinate points of each extreme point in the temperature sequence, which are symmetrical about a transverse axis, are obtained, adjacent coordinate points are connected, the obtained curve is recorded as the extreme point curve of the temperature sequence (the change trend of the extreme point curve is consistent by taking the opposite number of the negative correlation, and the DTW matching is convenient).
Further, performing DTW matching on the two extreme point curves to obtain extreme points matched with each extreme point in the exhaust gas amount sequence in the temperature sequence, and marking the extreme points as temperature matched extreme points of each extreme point in the exhaust gas amount sequence; for any extreme point in the exhaust gas quantity sequence, obtaining the product of the slope of the extreme point and the slope of the temperature matching extreme point, and recording the product as the temperature correlation of the extreme point; acquiring the temperature correlation of the extreme point adjacent to the previous extreme point and the temperature correlation of the extreme point adjacent to the next extreme point in the exhaust gas quantity sequence according to the method, and respectively marking the temperature correlation as the left temperature correlation and the right temperature correlation of the extreme point; analyzing the signs (signs) of the temperature correlation, the left temperature correlation and the right temperature correlation of the extreme point, and if the signs of the temperature correlation are the same as the signs of the left temperature correlation and the signs of the right temperature correlation, no noise exists between the extreme point and the temperature dimension; if the sign of the temperature correlation is different from the sign of the left temperature correlation or the sign of the right temperature correlation, the extreme point is marked as a noise extreme point; according to the method, noise judgment is carried out on the extreme point in the humidity dimension and the pressure dimension respectively, and if no noise exists in all three dimensions, the extreme point is marked as a reserved extreme point; if any one dimension is judged to be a noise extreme point, the extreme point is marked as a noise extreme point; noise judgment is carried out on all extreme points of the exhaust gas quantity sequence according to the method, so that a plurality of reserved extreme points and noise extreme points are obtained; it should be noted that, for the first extreme point or the last extreme point in the exhaust gas amount sequence, only the temperature correlation of one adjacent extreme point is acquired to perform noise judgment.
Further, for the displacement sequence, the length of each interpolation segment is obtained, and 2 times of the maximum value in all the lengths is recorded as the window minimum width of the displacement sequence (the units of the interpolation segment length and the window minimum width are actually the number of data points); starting from the first data point of the exhaust sequence, framing the data points with the minimum width of the window, and if the last data point framed is a reserved extreme point, forming a transverse window by all the data points framed; if the last data point selected by the frame is not the reserved extreme point, continuing to select the frame to the rear frame until the frame selects one reserved extreme point, and taking all the data points selected by the frame as a transverse window; after obtaining the first transverse window, starting from the next data point of the transverse window, namely the next data point of the data point (reserved extreme point) of the frame selection stop, continuously obtaining the transverse window according to the method until all data points in the exhaust sequence are divided into corresponding transverse windows, and obtaining a plurality of transverse windows of the exhaust sequence; if the remaining data points of the exhaust gas amount sequence are insufficient to form a transverse window, the data points are used as only one window, the boundary condition of the window is obtained according to a conventional method, an interpolation function is obtained, and the subsequent exhaust gas amount change characteristic analysis is not carried out on the window.
Further, several lateral windows of each displacement sequence are obtained according to the above method.
So far, through extreme points and interpolation sections in the exhaust gas quantity sequence, a plurality of transverse windows are obtained and are used for analyzing the change characteristics of the exhaust gas quantity and obtaining boundary conditions.
Step S003, according to the exhaust data and the slope of the data points in the transverse windows, the mutation degree and the exhaust change characteristic of each transverse window are obtained; and obtaining a cubic spline curve formed by boundary conditions of each transverse window and interpolation functions according to the displacement change characteristics and the slope of the extreme points.
After the transverse windows are obtained, quantifying the mutation degree of each transverse window according to the distribution and the slope of data points in the transverse windows and the exhaust amount data of data points in adjacent transverse windows, and quantifying the exhaust amount change characteristics, namely whether the transverse windows are stable windows or abrupt windows according to the mutation degree; and then, according to the displacement change characteristics, performing lateral window self-adaptive boundary condition acquisition to generate a cubic spline curve of the displacement sequence.
Specifically, for the first exhaust gas amount in any exhaust gas amount sequenceThe slope calculation of each data point in the transverse window is calculated as the prior art, and the embodiment is not repeated, the slope of the first data point in the exhaust volume sequence is set as the slope of the second data point, and the slopes of other data points are normally calculated; acquiring the number of data points with the slope larger than 0 and the number of data points with the slope smaller than 0 in the transverse window, and recording the ratio of the maximum value of the two numbers to the sum of the number of data points with the slope equal to 0 and the number of data points in the transverse window as a change factor of the transverse window; then +.>Mutation degree of the individual transverse windows->The calculation method of (1) is as follows:
wherein,indicate->The variation factor of the individual lateral windows->Indicate->The number of data points in a single transverse window,indicate->The (th) of the lateral windows>The displacement data of the data point, namely the displacement data of the last data point in the transverse window; />Indicate->Displacement data of the first data point in the lateral window, +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Representing absolute value; by passing throughAccumulating differences of adjacent data points in a transverse window, wherein the larger the accumulated value is, the more frequent the change in the transverse window is, the larger the corresponding mutation degree is, and meanwhile, the larger the difference is, the larger the change between the transverse window and the adjacent window is, the larger the mutation degree is; the change factor is obtained based on the slope, the larger the change factor is, the more the slope is towards the same sign, the more the change trend is single, and the smaller the mutation degree is.
Further, the mutation degree of each transverse window in the exhaust sequence is obtained according to the method, all mutation degrees are subjected to linear normalization, the obtained result is recorded as a mutation factor of each transverse window, a mutation threshold is preset, the mutation threshold is described by adopting 0.7, two exhaust change characteristics are set, and the two mutation characteristics are respectively a transverse window with stable data and a transverse window with abrupt data; marking a transverse window with the mutation factor larger than the mutation threshold value as a transverse window of data mutation, and marking the transverse window as a mutation window; marking a transverse window with the mutation factor smaller than or equal to the mutation threshold value as a transverse window with stable data, and marking the transverse window as a stable window; and obtaining the exhaust gas quantity change characteristics of each transverse window in the exhaust gas quantity sequence to obtain a plurality of stable windows and abrupt change windows.
Further, for the stable window, setting the first derivative and the second derivative of two data points of the boundary of the stable window to be 0, so that the curve of the stable window has no slope and curvature change at the head end and the tail end of the interpolation section, and obtaining the boundary condition of the stable window; for a non-stationary window, setting the second derivative of two data points of the boundary of the non-stationary window to be 0, wherein the first derivative is the slope, and the slope is not changed, so that the end points of the transverse window can be smoothly connected, bending change can not occur, overfitting and under fitting are avoided, and the boundary condition of the abrupt window is obtained.
Furthermore, in the embodiment of the interpolation function, a cubic equation is adopted, the interpolation function of each transverse window needs to meet the boundary condition at the end point, and a cubic spline curve of the displacement sequence is obtained through cubic spline fitting and connection of interpolation functions of different transverse windows; a cubic spline curve for each displacement sequence was obtained as described above.
So far, through carrying out mutation degree analysis on each transverse window in the exhaust gas quantity sequence, the exhaust gas quantity change characteristic is obtained, the boundary condition of each transverse window is determined according to the exhaust gas quantity change characteristic, and an interpolation function is obtained, so that a cubic spline curve formed by the interpolation function can avoid over fitting and under fitting, the change of the exhaust gas quantity sequence at an extreme point is smoother, the noise extreme point is highlighted, and the denoising effect of subsequent decomposition reconstruction is further improved.
Step S004, according to the cubic spline curve, through EMD decomposition and reconstruction, denoising exhaust gas quantity data of each exhaust gas quantity sequence are obtained, denoising temperature data, denoising humidity data and denoising pressure data in the same monitoring period are obtained, and according to the correlation of the denoising exhaust gas quantity data and denoising data in other dimensions, foreign matter fault detection of equipment is carried out.
After obtaining a cubic spline curve of an exhaust sequence, carrying out EMD (empirical mode decomposition) on a data curve of the exhaust sequence according to the obtained cubic spline curve for the exhaust sequence of the latest one, namely the exhaust sequence of the latest one monitoring period, so as to obtain a plurality of IMF (inertial measurement unit) components, and carrying out self-adaptive denoising on the exhaust sequence through the obtained cubic spline curve when the IMF components are reconstructed, wherein the reconstructed time sequence data is recorded as denoising exhaust data; and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, and acquiring denoising temperature data, denoising humidity data and denoising pressure data according to the method, wherein denoising exhaust gas amount data, denoising temperature data, denoising humidity data and denoising pressure data are all time sequence data sequences.
Further, acquiring an exhaust gas amount sequence and denoising exhaust gas amount data of each monitoring period and each temperature sequence and denoising temperature data according to the method, calculating a clearman correlation coefficient for denoising exhaust gas amount data and denoising temperature data of the same period, and recording the correlation degree of the exhaust gas amount data and the temperature data of each monitoring period; calculating a mean value of all the correlation degrees except the latest monitoring period, and taking the maximum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold value if the mean value is smaller than 0; if the average value is greater than or equal to 0, taking the minimum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold; acquiring humidity sequences and denoising humidity data of each monitoring period, and pressure sequences and denoising pressure data of each monitoring period according to the method, and obtaining a humidity related threshold value, a pressure related threshold value and the correlation degree of the exhaust gas quantity data of the latest monitoring period with the humidity data and the correlation degree of the pressure data respectively; for any dimension of temperature, humidity and pressure, if the correlation threshold value of the dimension is obtained by the minimum value and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is greater than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment; if the correlation threshold value of the dimension is obtained by the maximum value and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is smaller than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment; according to the method, the degree of correlation between the latest exhaust gas amount data in one time period and the data in each dimension is judged abnormally, and if abnormality occurs in the judgment of one dimension, the condition that the air conditioning equipment has a foreign matter fault is indicated, the exhaust port is possibly blocked by the foreign matter, and the air conditioning equipment needs to be repaired timely.
Thus, the foreign matter fault detection of the air conditioner in the heating ventilation equipment is completed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (4)
1. The method for detecting the foreign matter faults of the heating ventilation equipment based on the artificial intelligence is characterized by comprising the following steps of:
collecting exhaust gas data, temperature data, humidity data and pressure data of air conditioning equipment, and obtaining a plurality of exhaust gas sequences according to the exhaust gas data of each monitoring period;
obtaining a plurality of extreme points and interpolation sections for each exhaust gas quantity sequence; obtaining a plurality of transverse windows of each exhaust sequence according to the slope and distribution of extreme points in the exhaust sequence and other dimension data and the interpolation section;
according to the exhaust amount data and the slope of the data points in the transverse windows, the mutation degree and the exhaust amount change characteristic of each transverse window are obtained; obtaining a cubic spline curve formed by boundary conditions of each transverse window and interpolation functions according to the displacement change characteristics and the slope of the extreme points;
according to the three spline curves, carrying out EMD decomposition and reconstruction to obtain denoising exhaust gas quantity data of each exhaust gas quantity sequence, obtaining denoising temperature data, denoising humidity data and denoising pressure data of the same monitoring period, and carrying out foreign matter fault detection of equipment according to the correlation of the denoising exhaust gas quantity data and denoising data of other dimensions;
the method for obtaining the plurality of transverse windows of each exhaust gas amount sequence comprises the following specific steps:
obtaining a plurality of reserved extreme points and noise extreme points of each exhaust volume sequence according to the slope and distribution of the extreme points in the exhaust volume sequence and other dimension data; for any one exhaust gas amount sequence, acquiring the length of each interpolation section, and recording 2 times of the maximum value in all lengths as the minimum width of a window of the exhaust gas amount sequence;
starting from the first data point of the exhaust sequence, framing the data points with the minimum width of the window, and if the last data point framed is a reserved extreme point, forming a transverse window by all the data points framed; if the last data point selected by the frame is not the reserved extreme point, continuing to select the frame to the rear frame until the frame selects one reserved extreme point, and taking all the data points selected by the frame as a transverse window;
after the first transverse window is obtained, starting from the next data point of the transverse window, continuously obtaining the transverse window until all data points in the exhaust sequence are divided into corresponding transverse windows, and obtaining a plurality of transverse windows of the exhaust sequence;
the specific method for obtaining the plurality of reserved extreme points and the noise extreme points of each exhaust gas quantity sequence comprises the following steps:
for any exhaust gas amount sequence, acquiring the slope of each extreme point; acquiring a monitoring period corresponding to the exhaust gas amount sequence, and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, extreme points in each sequence and the slope of each extreme point;
for the temperature sequence of the monitoring period, acquiring cosine similarity between the temperature sequence and the exhaust gas sequence, and respectively acquiring an extreme point curve of the exhaust gas sequence and an extreme point curve of the temperature sequence if the cosine similarity is more than or equal to 0; if the cosine similarity is smaller than 0, acquiring an extreme point curve of the exhaust gas quantity sequence, acquiring coordinate points of each extreme point in the temperature sequence, which are symmetrical about a transverse axis, connecting adjacent coordinate points, and recording the acquired curve as the extreme point curve of the temperature sequence;
performing DTW matching on the two extreme point curves to obtain extreme points matched with each extreme point in the exhaust gas sequence in the temperature sequence, and marking the extreme points as temperature matched extreme points of each extreme point in the exhaust gas sequence;
for any extreme point in the exhaust gas quantity sequence, obtaining the product of the slope of the extreme point and the slope of the temperature matching extreme point, and recording the product as the temperature correlation of the extreme point; acquiring the temperature correlation of the extreme point adjacent to the previous extreme point and the temperature correlation of the extreme point adjacent to the next extreme point in the exhaust gas quantity sequence, and respectively marking the temperature correlation as the left temperature correlation and the right temperature correlation of the extreme point;
if the signs of the temperature correlations are the same as those of the left temperature correlations and those of the right temperature correlations, no noise exists between the extreme point and the temperature dimension; if the sign of the temperature correlation is different from the sign of the left temperature correlation or the sign of the right temperature correlation, the extreme point is marked as a noise extreme point;
noise judgment is carried out on the extreme point in the humidity dimension and the pressure dimension respectively, and if no noise exists in the three dimensions, the extreme point is marked as a reserved extreme point; if any one dimension is judged to be a noise extreme point, the extreme point is marked as a noise extreme point; noise judgment is carried out on all extreme points of the exhaust gas quantity sequence, so that a plurality of reserved extreme points and noise extreme points are obtained;
the mutation degree and the exhaust gas amount change characteristics of each transverse window are obtained by the following specific methods:
for any one exhaust gas amount sequenceA lateral window, obtain->Slope of each data point within the lateral windows; get->A number of data points with a slope greater than 0 and a number of data points with a slope less than 0 within a lateral window, adding the sum of the maximum of the two numbers plus the number of data points with a slope equal to 0 to +.>The ratio of the number of data points in the lateral window is denoted by +.>A variation factor of the lateral windows; the exhaust gas amount sequence is +.>Mutation degree of the individual transverse windows->The calculation method of (1) is as follows:
wherein,indicate->The variation factor of the individual lateral windows->Indicate->The number of data points in the individual transverse windows, +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Indicate->Displacement data of the first data point in the lateral window, +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Indicate->The (th) of the lateral windows>Data point exhaust data +.>Representing absolute value;
obtaining the variation characteristic of the exhaust gas quantity of each transverse window according to the mutation degree of the transverse window in the exhaust gas quantity sequence;
the specific method for obtaining the exhaust gas quantity change characteristic of each transverse window comprises the following steps:
for any one exhaust sequence, obtaining the mutation degree of each transverse window in the exhaust sequence, carrying out linear normalization on all mutation degrees, marking the obtained result as a mutation factor of each transverse window, setting two exhaust change characteristics, namely a transverse window with stable data and a transverse window with abrupt data; marking a transverse window with the mutation factor larger than the mutation threshold value as a transverse window of data mutation, and marking the transverse window as a mutation window; marking a transverse window with the mutation factor smaller than or equal to the mutation threshold value as a transverse window with stable data, and marking the transverse window as a stable window; obtaining an exhaust gas quantity change characteristic for each transverse window in the exhaust gas quantity sequence to obtain a plurality of stable windows and abrupt change windows;
the method for obtaining the cubic spline curve formed by the boundary condition of each transverse window and the interpolation function comprises the following specific steps:
for the stable window, setting the first derivative and the second derivative of two data points of the boundary of the stable window to be 0, and obtaining the boundary condition of the stable window; for a non-stationary window, setting the second derivative of two data points of the boundary of the non-stationary window to 0, wherein the first derivative is the slope and is not changed, so as to obtain the boundary condition of the abrupt window;
obtaining an interpolation function of each transverse window according to the boundary condition of the transverse window and forming a cubic spline curve of each displacement sequence;
the method for obtaining the interpolation function of each transverse window and forming the cubic spline curve of each displacement sequence comprises the following specific steps:
for any exhaust volume sequence, the interpolation function adopts a cubic equation, the interpolation function of each transverse window needs to meet the boundary condition at the end point, and a cubic spline curve of the exhaust volume sequence is obtained through cubic spline fitting and connection of interpolation functions of different transverse windows.
2. The method for detecting the foreign matter fault of the heating and ventilation equipment based on the artificial intelligence according to claim 1, wherein the steps of obtaining a plurality of extreme points and interpolation segments for each exhaust gas quantity sequence comprise the following specific steps:
for any one exhaust gas amount sequence, acquiring a data curve of the exhaust gas amount sequence; and acquiring a plurality of extreme points of the data curve, wherein the curve between adjacent extreme points is used as an interpolation section, and a plurality of interpolation sections are obtained for the displacement sequence.
3. The method for detecting the foreign matter fault of the heating and ventilation equipment based on the artificial intelligence according to claim 1, wherein the specific method for obtaining the denoising exhaust gas quantity data of each exhaust gas quantity sequence and obtaining the denoising temperature data, the denoising humidity data and the denoising pressure data of the same monitoring period comprises the following steps:
for the exhaust gas quantity sequence of the latest monitoring period, carrying out EMD (empirical mode decomposition) on a data curve of the exhaust gas quantity sequence according to the acquired cubic spline curve to obtain a plurality of IMF components, reconstructing the IMF components, and recording the reconstructed time sequence data as denoising exhaust gas quantity data; and acquiring a temperature sequence, a humidity sequence and a pressure sequence of the monitoring period, and acquiring denoising temperature data, denoising humidity data and denoising pressure data.
4. The method for detecting the foreign matter fault of the heating and ventilation equipment based on the artificial intelligence according to claim 3, wherein the method for detecting the foreign matter fault of the equipment according to the correlation between the denoising exhaust gas amount data and the denoising data of other dimensions comprises the following specific steps:
acquiring an exhaust gas amount sequence and denoising exhaust gas amount data of each monitoring period, and each temperature sequence and denoising temperature data, calculating a clearman correlation coefficient for the denoising exhaust gas amount data and denoising temperature data of the same period, and recording the correlation degree of the exhaust gas amount data and the temperature data of each monitoring period;
calculating a mean value of all the correlation degrees except the latest monitoring period, and taking the maximum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold value if the mean value is smaller than 0; if the average value is greater than or equal to 0, taking the minimum value of all the correlation degrees except the latest monitoring period as a temperature correlation threshold;
acquiring humidity sequence and denoising humidity data of each monitoring period, and pressure sequence and denoising pressure data of each monitoring period, and obtaining a humidity correlation threshold, a pressure correlation threshold and the correlation degree of the exhaust gas quantity data of the latest monitoring period with the humidity data and the correlation degree of the pressure data respectively;
for any dimension of temperature, humidity and pressure, if the correlation threshold value of the dimension is obtained by the minimum value, and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is greater than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment; if the correlation threshold value of the dimension is obtained by the maximum value and the correlation degree of the exhaust gas quantity data of the latest monitoring period and the dimension data is smaller than the correlation threshold value of the dimension, the exhaust gas quantity data of the latest monitoring period is abnormal, and foreign matter faults exist in the air conditioning equipment;
and carrying out abnormal judgment on the correlation degree of the exhaust amount data in the latest time period and the data in each dimension, wherein the abnormality occurs in the judgment of one dimension, and the air conditioning equipment has foreign matter faults.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311454671.0A CN117167903B (en) | 2023-11-03 | 2023-11-03 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311454671.0A CN117167903B (en) | 2023-11-03 | 2023-11-03 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117167903A CN117167903A (en) | 2023-12-05 |
| CN117167903B true CN117167903B (en) | 2024-01-30 |
Family
ID=88945444
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311454671.0A Active CN117167903B (en) | 2023-11-03 | 2023-11-03 | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117167903B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117387162B (en) * | 2023-12-11 | 2024-03-08 | 湖南康家环保科技有限公司 | Device and method for removing indoor formaldehyde and TVOC |
| CN117408497B (en) * | 2023-12-14 | 2024-04-05 | 山东锦都食品有限公司 | Food processing material management and control method and system |
| CN117688311B (en) * | 2024-02-04 | 2024-05-31 | 深圳市纯水一号水处理科技有限公司 | Real-time monitoring method and system for advanced treatment of refractory salt-containing wastewater |
| CN118010939B (en) * | 2024-04-10 | 2024-06-18 | 山东华检检测有限公司 | Intelligent formaldehyde detection method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226649A (en) * | 2013-03-25 | 2013-07-31 | 西安交通大学 | Ensemble noise-reconstructed EMD (empirical mode decomposition) method for early and compound faults of machinery |
| CN112269367A (en) * | 2020-10-13 | 2021-01-26 | 西安建筑科技大学 | Fault detection method and system based on EEMD-combined neural network air processing system |
| CN114234361A (en) * | 2021-12-14 | 2022-03-25 | 北京工业大学 | Central air-conditioning sensor fault detection method based on double noise reduction and fuzzy indexes |
| CN115628522A (en) * | 2022-10-31 | 2023-01-20 | 华南理工大学 | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM |
| CN116304581A (en) * | 2023-05-10 | 2023-06-23 | 佛山市钒音科技有限公司 | Intelligent electric control system for air conditioner |
| CN116955938A (en) * | 2023-09-20 | 2023-10-27 | 苏州新耀环保科技有限公司 | Dry-type waste gas treatment equipment monitoring method and system based on data analysis |
-
2023
- 2023-11-03 CN CN202311454671.0A patent/CN117167903B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226649A (en) * | 2013-03-25 | 2013-07-31 | 西安交通大学 | Ensemble noise-reconstructed EMD (empirical mode decomposition) method for early and compound faults of machinery |
| CN112269367A (en) * | 2020-10-13 | 2021-01-26 | 西安建筑科技大学 | Fault detection method and system based on EEMD-combined neural network air processing system |
| CN114234361A (en) * | 2021-12-14 | 2022-03-25 | 北京工业大学 | Central air-conditioning sensor fault detection method based on double noise reduction and fuzzy indexes |
| CN115628522A (en) * | 2022-10-31 | 2023-01-20 | 华南理工大学 | Market central air-conditioning load prediction method, system and medium based on EMD-PSO-LSTM |
| CN116304581A (en) * | 2023-05-10 | 2023-06-23 | 佛山市钒音科技有限公司 | Intelligent electric control system for air conditioner |
| CN116955938A (en) * | 2023-09-20 | 2023-10-27 | 苏州新耀环保科技有限公司 | Dry-type waste gas treatment equipment monitoring method and system based on data analysis |
Non-Patent Citations (5)
| Title |
|---|
| A self-adaptive data analysis for fault diagnosis of an automotive air- conditioner blower;Wu J D;Expert Systems with Applications;第38卷(第1期);545-552 * |
| 中央空调传感器双重降噪模糊故障检测方法;高学金;电子测量与仪器学报;第36卷(第8期);77-88 * |
| 基于VMD和SVM的家用空调外机故障识别研究;刘芝庭;王宇华;郑文炜;;佛山科学技术学院学报(自然科学版)(第04期);39-48 * |
| 小波脊线解调与两次EMD分解相结合的故障识别方法及应用研究;姜万录;仪器仪表学报;第34卷(第5期);1132-1138 * |
| 徐文皙.经验模态分解端点延拓算法研究与波谱数据处理应用.2021, * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117167903A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117167903B (en) | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment | |
| CN116111727B (en) | Comprehensive distribution box abnormity monitoring method based on dynamic temperature threshold | |
| CN117034043B (en) | Intelligent building comprehensive energy consumption monitoring method and system based on multi-energy Internet of things | |
| CN117238058B (en) | Starter monitoring method for automobile based on data analysis | |
| CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN116992322B (en) | Smart city data center management system | |
| CN117195007B (en) | Heat exchanger performance prediction method and system | |
| CN113342610B (en) | Time sequence data anomaly detection method and device, electronic equipment and storage medium | |
| CN116451029A (en) | Dehumidifier working state early warning method | |
| CN116308963A (en) | Government affair data analysis method and system | |
| CN117540238A (en) | Data security management method for industrial digital information acquisition device | |
| CN116861313B (en) | Kalman filtering working condition identification method and system based on vibration energy trend | |
| CN115081331A (en) | Wind turbine generator running state abnormity detection method based on state parameter reconstruction error | |
| CN117092980B (en) | Electrical fault detection control system based on big data | |
| CN112416661B (en) | Multi-index time sequence anomaly detection method and device based on compressed sensing | |
| CN116576553B (en) | Data optimization acquisition method and system for air conditioner | |
| CN117093947A (en) | Power generation diesel engine operation abnormity monitoring method and system | |
| CN112202630A (en) | Network quality abnormity detection method and device based on unsupervised model | |
| CN109163894B (en) | Running-in state identification method based on friction temperature signal | |
| CN116561927A (en) | Digital twin-driven small sample rotary machine residual life prediction method and system | |
| CN114925731A (en) | Method for detecting abnormal value of monitoring data of flexible inclinometer | |
| CN112700039A (en) | Steady state detection and extraction method for load operation data of thermal power plant | |
| CN117560015B (en) | Efficient processing system for air conditioner installation engineering data | |
| CN117312617B (en) | Real-time sewage treatment method and system based on sewage data monitoring | |
| CN117870993B (en) | Multi-sensor-based return air leakage detection method and system for rotary dehumidifier |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| 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 |