An air conditioner anomaly diagnosis method and device based on air conditioner terminal monitoring data

Abstract

The present invention discloses an air conditioner anomaly diagnosis method and device based on air conditioner terminal monitoring data, comprising the following steps: data monitoring and fusion; preprocessing data; identifying the building air conditioner system model; estimating the air conditioner operating power; diagnosing the air conditioner operating status; and also discloses a corresponding air conditioner anomaly diagnosis device and storage medium based on air conditioner terminal monitoring data. The present invention diagnoses air conditioner anomalies based on meteorological data and real-time air conditioner monitoring data, achieving the diagnosis of air conditioner abnormal states with low cost and high reliability, timely discovering air conditioner abnormal states, and improving the air conditioner operation quality.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy conservation, and particularly to an air conditioner anomaly diagnosis method and device based on air conditioner terminal monitoring data. Background Art

[0002] Among the energy consumption of buildings, air conditioners account for the largest proportion, generally about 40% - 50% of the total building energy consumption, and the air conditioner energy consumption in shopping malls and comprehensive buildings may be as high as more than 60%. As a "major energy consumer" in the power industry, the air conditioner power consumption in large and medium-sized cities in China accounts for about 60% of the summer peak load. In 2021, the domestic sales volume of air conditioners in China was 57.16 million units, and in 2022, the domestic sales volume of air conditioners in China increased to 84.70 million units, maintaining a continuous and rapid growth trend. Under normal circumstances, the energy consumption of the air conditioner system depends on the building insulation performance, air conditioner type, and energy efficiency level. However, in the case of air conditioner system failures or abnormal operating states, the energy consumption may increase sharply, which not only wastes energy, affects the service life of the air conditioner, but also fails to meet the comfort requirements of the occupants. Therefore, timely and accurately diagnosing the abnormal state (including faults) of the air conditioner system has always been a problem explored and solved by many researchers and engineers.

[0003] Currently, the mainstream research and practice methods are all based on collecting data collected by various sensors installed in each link of the air conditioner system for fault diagnosis. For example, split fixed-frequency air conditioners generally include an evaporator, a condenser, a thermostatic expansion valve, and a scroll compressor. Available sensors include valve opening sensors, air pressure sensors, temperature and humidity sensors, flow rate sensors, etc. However, sensors sometimes also have faults, including false alarms, drifts, low precision, or complete damage, etc. Sensor faults are often difficult to detect in a timely manner. If a sensor fails and cannot be detected in a timely manner, it will seriously affect the accuracy and credibility of the diagnosis results. Therefore, this data collection method is costly, and the more sensors there are, the more likely they are to fail, and the higher the probability of misjudgment of the diagnosis system. The lack of reliability problems makes it difficult to promote and apply this technical route.

[0004] In fact, various abnormal states during the operation of air conditioners are usually reflected in the real-time power data of the air conditioners. The air conditioner steward terminal device based on the Internet of Things can collect the operation status data of the building air conditioning system in real time at a relatively low cost and with high reliability, including the temperature and humidity in the room, the mode of the air conditioner, and the temperature, wind speed, and power set by the user. In addition, conventional meteorological data such as weather type, outdoor temperature, and wind force are also very easy to obtain, with almost no cost. There is currently no research report on how to make full use of the large amount of sample data of the above-mentioned air conditioner operation status and outdoor weather to achieve rapid and accurate diagnosis of air conditioner abnormal states. If the above-mentioned easily obtainable data can be fused to achieve a high-precision diagnosis of air conditioner abnormal states, it will be of great significance for the normal operation of the building air conditioning system, avoiding high energy consumption of the air conditioner caused by faults, improving the comfort and service life of the air conditioner. Summary of the Invention

[0005] The present invention proposes an air conditioner abnormal diagnosis method and device based on air conditioner terminal monitoring data to solve the above technical problems.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] One aspect of the present invention provides an air conditioner abnormal diagnosis method based on air conditioner terminal monitoring data, including the following steps:

[0008] Step S1, data monitoring and fusion: Collect the real-time monitoring data of the air conditioner in the room where the corresponding air conditioner is located through the air conditioner terminal provided in each air conditioner in the building air conditioning system, and at the same time collect meteorological data, and fuse the meteorological data and the real-time monitoring data of the air conditioner to form an initial data set;

[0009] Step S102, preprocess the data: Preprocess the initial data set to obtain a preprocessed sample set;

[0010] Step S103, identify the building air conditioning system model: Construct an ETP model of the building air conditioning system, and calculate the equivalent thermal resistance and equivalent heat capacity of the building air conditioning system;

[0011] Step S104, estimate the air conditioner operation power: Obtain the estimated value and measured value of the air conditioner load power according to the ETP model of the building air conditioning system and the preprocessed sample set;

[0012] Step S105, diagnose the air conditioner operation status: Judge whether the air conditioner operation status is abnormal according to the correlation coefficient between the estimated value and measured value of the air conditioner load power.

[0013] Preferably, in the step S1, the air-conditioning terminal monitors and collects the air-conditioning monitoring data in the room where the corresponding air conditioner is located in real time, including indoor temperature, indoor humidity, air-conditioning mode, set temperature of the air conditioner, air-conditioning wind speed, and air-conditioning power. The meteorological data includes weather type, outdoor temperature, and wind force.

[0014] Preferably, the initial data set X obtained in the step S1 = [X1, X2, …, X i , …, X q T , where q is the dimension of the data type, X i ∈ R 1×p contains samples at p moments, where p, q, and i are all positive integers, and T represents matrix transpose;

[0015] In the step S102, the "preprocessing of the initial data set" includes the following steps:

[0016] Step S1021, calculate the time-domain mean matrix of the data set X where is the time-domain mean of the i-th type of data, and calculate the standard deviation σ i of the i-th type of data:

[0017]

[0018] to form the standard deviation vector σ = [σ1, σ2, …, σ i , …, σ q T ,

[0019] If the data set X j = [X 1j , X 2j , …, X ij , …, X qj T at the j-th moment satisfies Equation (2), then the data set X j at the j-th moment is deleted as abnormal data:

[0020]

[0021] where |*| ρ represents the L ρ -norm of *, ρ is 0, 1, 2, or ∞, j is a positive integer, and K is a set positive integer;

[0022] ​​​Step S1022: Check the time-domain coherence of various types of data after removing abnormal data. If the continuous missing exceeds m time periods or the continuous startup duration is less than h hours, then all the sample data in the continuous startup time period of the air conditioner in these two cases, where the continuous missing exceeds m time periods or the air conditioner continuously starts up for less than h hours, are removed from the sample set to obtain the refined data sample set, where both m and h are set positive integers.

[0023] Step S1023: For the refined data sample set, use the interpolation method to fill in the sporadic missing data to obtain the preprocessed sample set.

[0024] Preferably, in step S1023, the interpolation method is linear interpolation, mean interpolation, polynomial interpolation, or spline interpolation.

[0025] Preferably, in step S103, the "construct the ETP model of the building air-conditioning system and calculate the equivalent thermal resistance and equivalent heat capacity of the building air-conditioning system" includes the following steps:

[0026] Step S1031: Simulate the thermodynamics characteristics of the building air-conditioning system and preliminarily construct the ETP model of the building air-conditioning system:

[0027]

[0028] where, T in is the indoor temperature, T out is the outdoor temperature, R is the equivalent thermal resistance, C is the equivalent heat capacity, P is the air-conditioning load power, and η is the air-conditioning energy efficiency ratio;

[0029] Step S1032: Extract the indoor temperature T in , outdoor temperature T out , and air-conditioning power P in the preprocessed sample set during the time period from TM0 to TM1, and construct the sample data matrix N J is the number of sampling time points of the Jth group of sample data, TM0 is the initial operation time of the air conditioner, and TM1 is the node time of the sample data set;

[0030] Step S1033: Use each group of sample data matrices Y J , and according to the ETP model of the building air-conditioning system, identify the ETP parameters: equivalent thermal resistance R and equivalent heat capacity C for each air-conditioning startup period, and store them in the ETP data set V of the building air-conditioning system;

[0031] Step S1034: Classify the ETP data set V of the building air-conditioning system, select the category with the largest data volume, and calculate the central value of all ETP parameters in this category, and use it as the equivalent thermal resistance R and equivalent heat capacity C of the building air-conditioning system.

[0032] Preferably, in the step S104, the step of "obtaining the estimated value and measured value of the air-conditioning load power according to the ETP model of the building air-conditioning system and the preprocessing sample set" includes the following steps:

[0033] Step S1041: After the moment TM1, according to the real-time monitoring data of the air conditioner collected by the air-conditioning terminal and the meteorological data, extract the indoor temperature T from the preprocessing sample set obtained in step S102 in , the outdoor temperature T out , and substitute them together with the equivalent thermal resistance R and equivalent heat capacity C obtained in step S1034 into the ETP model of the building air-conditioning system to obtain the estimated value of the air-conditioning load power during the Jth consecutive startup period N J is the number of sampling time points of the Jth group of sample data, and J is a positive integer;

[0034] Step S1042: Extract the measured value of the air-conditioning load power during the Jth consecutive startup period after the moment TM1 from the preprocessing sample set obtained in step S102

[0035] Preferably, in the step S105, the step of "judging whether the air-conditioning operation state is abnormal according to the correlation coefficient between the estimated value and measured value of the air-conditioning load power" includes the following steps:

[0036] Step S1051: According to the latest L groups of estimated values P of the air-conditioning load power obtained in step S104 estimate and the measured values P of the air-conditioning load power real , calculate the correlation coefficient γ between the estimated value P of the air-conditioning load power of the Jth group estimate and the measured value P of the air-conditioning load power real ; J ;

[0037]

[0038] Among them, both J and L are positive integers and 1 ≤ J ≤ L, P estimate,i is the i-th item of P estimate , P real,i is the i-th item of P real , is the mean value of P estimate , is the mean value of P real ;

[0039] Step S1052: Among the L correlation coefficients γ J , if more than α% are lower than the threshold δ, it is judged that the air-conditioning operation state is abnormal and a warning is issued; otherwise, it is judged that the air-conditioning operation state is normal.

[0040] Another aspect of the present invention provides an air conditioning abnormality diagnosis device based on air conditioning terminal monitoring data, including a data monitoring fusion module, a preprocessing module, a model identification module, a power estimation module and an air conditioning diagnosis module;

[0041] The data monitoring and fusion module is used to collect the air conditioning monitoring data in the room where the corresponding air conditioner is located through the air conditioning terminal provided by each air conditioner in real time, collect meteorological data, and fuse the meteorological data and the air conditioning monitoring data to form an initial data set;

[0042] The preprocessing module is used to preprocess the initial data set to obtain a preprocessed sample set;

[0043] The model identification module is used to construct the ETP model of the building air conditioning system and calculate the equivalent thermal resistance and equivalent heat capacity of the building air conditioning system;

[0044] The power estimation module is used to obtain the air conditioning load power estimation value and the air conditioning load power measured value according to the ETP model of the building air conditioning system and the pre-processed sample set;

[0045] The air conditioning diagnosis module is used to determine whether the air conditioning operation state is abnormal according to the correlation coefficient between the air conditioning load power estimation value and the air conditioning load power actual measurement value.

[0046] Another aspect of the present invention provides a storage medium storing a program, which, when executed by a processor, implements the above-mentioned air-conditioning abnormality diagnosis method based on air-conditioning terminal monitoring data.

[0047] Compared with the prior art, the present invention performs air conditioning abnormality diagnosis based on the air conditioning housekeeper monitoring data, realizes the diagnosis of air conditioning abnormality at a lower cost and higher reliability, timely discovers air conditioning abnormality, and improves the quality of air conditioning operation. The present invention uses the Internet of Things and the air conditioning housekeeper monitoring terminal to collect and store data in real time, and can understand the operating status of each air conditioner in the building at a low cost and in a refined manner, and realizes efficient and accurate estimation of air conditioning load power through the ETP model rolling update technology of the building air conditioning system. Based on this, further correlation analysis is performed with the actual measured value of the air conditioning load power in the same period, and the air conditioning abnormal state that meets a certain probability and has a certain fault tolerance rate is screened out, which provides a more accurate target for the operation and maintenance personnel to detect the air conditioner on site in a timely manner, which can greatly improve the accuracy and efficiency of air conditioning abnormal state diagnosis, and also greatly reduce the diagnosis cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flowchart of an air conditioning abnormality diagnosis method based on air conditioning terminal monitoring data of the present invention;

[0049] Figure 2Abnormal diagnosis result of an air conditioner based on monitored data of an air conditioner terminal according to an embodiment of the present invention

[0050] Figure 3 Block diagram of the structure of an air conditioner abnormal diagnosis system based on monitored data of an air conditioner terminal according to the present invention

[0051] In the figure, 201 - data monitoring and fusion module, 202 - preprocessing module, 203 - model identification module, 204 - power estimation module, 205 - air conditioner diagnosis module Specific embodiments

[0052] The present invention will be described in detail below with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, method, or functional transformation made by those of ordinary skill in the art based on these embodiments is included within the protection scope of the present invention

[0053] The terms used in the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "the", and "said" used in the present invention and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more of the associated listed items

[0054] As Figure 1 shown, an air conditioner abnormal diagnosis method based on monitored data of an air conditioner terminal includes the following steps

[0055] Step S1, data monitoring and fusion: Collect real-time monitored data of the air conditioner in the room where the corresponding air conditioner is located through the air conditioner terminals provided in each air conditioner in the building air conditioning system, and at the same time collect meteorological data, and fuse the meteorological data and the real-time monitored data of the air conditioner to form an initial data set

[0056] The real-time monitored data of the air conditioner collected by the air conditioner terminal in the room where the corresponding air conditioner is located includes indoor temperature, indoor humidity, air conditioner mode, air conditioner set temperature, air conditioner wind speed, air conditioner power, etc., and the meteorological data includes weather type, outdoor temperature, wind force, etc. The initial data set X = [X1, X2,..., X i ,..., X q , q is the dimension of the data type, X i ∈R p×1 contains samples at p moments, where p, q, and i are all positive integers. After the new air conditioner is installed and debugged or the old air conditioner is repaired and tested, it is confirmed that the air conditioner starts to operate normally, and this time is recorded as the initial operation time TM0 of the air conditioner, and TM1 is set as the sample data set node time

[0057] Step S102, preprocess the data: Preprocess the initial data set to obtain a preprocessed sample set.

[0058] Specifically, step S102 includes the following steps:

[0059] Step S1021, calculate the time-domain mean matrix of the data set X where is the time-domain mean of the i-th type of data, and calculate the standard deviation σ of the i-th type of data i :

[0060]

[0061] constitute the standard deviation vector σ = [σ1, σ2, …, σ i , …, σ q ,

[0062] If the data set X j = [X 1j , X 2j , …, X ij , …, X qj at the j-th moment satisfies equation (2), then delete the data set X j at the j-th moment as abnormal data:

[0063]

[0064] where |*| ρ represents the L ρ -norm of *, ρ is 0, 1, 2 or ∞, j is a positive integer, and K is a set positive integer.

[0065] Step S1022, check the time-domain coherence of each type of data after removing abnormal data. If the continuous missing exceeds m time periods or the continuous startup duration is less than h hours, then delete all sample data in the continuous startup periods of the two situations where the continuous missing exceeds m time periods or the air conditioner continuously starts up for less than h hours from the sample set to obtain a refined data sample set, where both m and h are set positive integers.

[0066] Step S1023, for the refined data sample set, use the interpolation method to fill in the sporadic missing data to obtain a preprocessed sample set. Here, the interpolation method can be the linear interpolation method, the mean interpolation method, the polynomial interpolation method, the spline interpolation method, etc.

[0067] After obtaining the preprocessed sample set, if the time is in the period from TM0 to TM1, then enter step S103; if the time > TM1, then enter step S104.

[0068] Step S103, identify the building air-conditioning system model: construct the ETP model of the building air-conditioning system, and calculate the equivalent thermal resistance and equivalent heat capacity of the building air-conditioning system.

[0069] Specifically, step S103 includes the following steps:

[0070] Step S1031, simulate the thermodynamics characteristics of the building air-conditioning system, and preliminarily construct the ETP model of the building air-conditioning system:

[0071]

[0072] Among them, T in is the indoor temperature, T out is the outdoor temperature, R is the equivalent thermal resistance, C is the equivalent heat capacity, P is the air-conditioning load power, and η is the air-conditioning energy efficiency ratio.

[0073] The ETP (equivalent thermal parameter) model is the most common air-conditioning load model at present and is widely used in various fields of air-conditioning load control. The ETP model of the building air-conditioning system can be a first-order model, and the air-conditioning energy efficiency ratio η can be determined according to the air-conditioning model.

[0074] Step S1032, extract the indoor temperature T in , outdoor temperature T out , and air-conditioning power P in the preprocessed sample set during the time period from TM0 to TM1, and construct the sample data matrix of the Jth continuous air-conditioning operation period N J is the number of sampling time points of the Jth group of sample data. Here, the air-conditioning monitoring data and meteorological data during the time period from TM0 to TM1 are collected for parameter identification of the ETP model of the building air-conditioning system.

[0075] Step S1033, use each group of sample data matrices Y J , and identify the ETP parameters: equivalent thermal resistance R and equivalent heat capacity C of each air-conditioning startup period according to the ETP model of the building air-conditioning system, and store them in the ETP data set V of the building air-conditioning system.

[0076] Step S1034, classify the ETP data set V of the building air-conditioning system, select the category with the largest data volume, and calculate the central value of all ETP parameters in this category, and use it as the equivalent thermal resistance R and equivalent heat capacity C of the building air-conditioning system.

[0077] Step S104, estimate the air-conditioning operating power: obtain the estimated value and measured value of the air-conditioning load power according to the ETP model of the building air-conditioning system and the preprocessed sample set.

[0078] Specifically, step S104 includes the following steps:

[0079] Step S1041: After time TM1, based on the real-time air-conditioning monitoring data and meteorological data collected by the air-conditioning terminal, extract the indoor temperature T from the preprocessed sample set obtained in step S102 in , outdoor temperature T out , and together with the equivalent thermal resistance R and equivalent heat capacity C obtained in step S1034, substitute them into the ETP model of the building air-conditioning system to obtain the estimated value of the air-conditioning load power for the Jth consecutive startup period N J is the number of sampling time points of the Jth group of sample data, and J is a positive integer.

[0080] Step S1042: Extract the measured value of the air-conditioning load power after time TM1 and during the Jth consecutive startup period from the preprocessed sample set obtained in step S102

[0081] Step S105: Diagnose the operating state of the air conditioner: Judge whether the operating state of the air conditioner is abnormal according to the correlation coefficient between the estimated value of the air-conditioning load power and the measured value of the air-conditioning load power.

[0082] Specifically, step S105 includes the following steps:

[0083] Step S1051: According to the latest L groups of estimated values P of the air-conditioning load power obtained in step S104 estimate and the measured values P of the air-conditioning load power real , calculate the correlation coefficient γ estimate between the estimated value P of the air-conditioning load power of the Jth group real and the measured value P of the air-conditioning load power J ;

[0084]

[0085] Among them, both J and L are positive integers and 1 ≤ J ≤ L, P estimate,i is the ith item of P estimate , P real,i is the ith item of P real , is the mean value of P estimate , is the mean value of P real .

[0086] Step S1052: Among the L correlation coefficients γ J , if more than α% are lower than the threshold δ, it is judged that the operating state of the air conditioner is abnormal and a warning is issued; otherwise, it is judged that the operating state of the air conditioner is normal.

[0087] Taking the measured data of the air-conditioning terminals in a certain building air-conditioning system as an example, after the air-conditioning is installed, the air-conditioning monitoring data and meteorological data for 15 days are collected first for parameter identification of the ETP model of the building air-conditioning system. Then, based on the ETP model and parameters of the building air-conditioning system, according to the real-time collected operation data of the building air-conditioning system and the corresponding meteorological data, the correlation between the predicted value and the measured value of the air-conditioning load power is calculated and analyzed in a rolling manner, and an air-conditioning anomaly alarm signal is given in a timely manner.

[0088] Among them, the time period from TM0 to TM1 is 15 days in total. The air-conditioning terminal can be various air-conditioning Internet of Things terminals such as air-conditioning butlers on the market. Specific parameters are selected according to experimental experience and can be selected as: L ρ -norm selects the L2-norm, K = 3. When eliminating abnormal data, if the continuous missing exceeds m = 1 time period or the continuous startup duration is less than h = 6 hours, then all sample data of the continuous startup time period where these two situations of continuous missing exceeding m time periods or the air-conditioning continuous startup being less than h hours are eliminated; the air-conditioning energy efficiency ratio η is determined according to the air-conditioning model to get η = 3.8; set the correlation coefficient γ J Among them, if more than 30% is lower than the threshold value of 0.7, it is judged that the air-conditioning operation state is abnormal.

[0089] After calculation, the thermal parameters of the building air-conditioning system are obtained: the equivalent thermal resistance R = 3.1 m2K / W, the equivalent heat capacity C = 1.5 J / ℃; the correlation coefficient γ J Such as Figure 2 shown, 30% of the 4 correlation coefficients γ J have fallen below the threshold value of 0.7, and the warning threshold for the abnormal air-conditioning state has been reached.

[0090] Corresponding to the foregoing embodiment of the air-conditioning anomaly diagnosis method based on air-conditioning terminal monitoring data, the present invention also provides an air-conditioning anomaly diagnosis device based on air-conditioning terminal monitoring data, including a data monitoring and fusion module 201, a preprocessing module 202, a model identification module 203, a power estimation module 204, and an air-conditioning diagnosis module 205, as Figure 3 shown.

[0091] The data monitoring and fusion module 201 is used to collect the air-conditioning monitoring data in the room where the corresponding air conditioner is located in real time through the air-conditioning terminals provided in each air conditioner, collect meteorological data, and fuse the meteorological data and the air-conditioning monitoring data to form an initial data set; the preprocessing module 202 is used to preprocess the initial data set to obtain a preprocessed sample set; the model identification module 203 is used to construct an ETP model of the building air-conditioning system and calculate the equivalent thermal resistance and equivalent heat capacity of the building air-conditioning system; the power estimation module 204 is used to obtain the estimated value and measured value of the air-conditioning load power according to the ETP model of the building air-conditioning system and the preprocessed sample set; the air-conditioning diagnosis module 205 is used to judge whether the operation state of the air conditioner is abnormal according to the correlation coefficient between the estimated value and measured value of the air-conditioning load power.

[0092] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment related to the method, and will not be elaborated here.

[0093] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only illustrative, and some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present invention. Those of ordinary skill in the art can understand and implement it without creative work.

[0094] In another embodiment of the present invention, a storage medium is provided, storing a program, which, when executed by a processor, implements the above-mentioned air-conditioning anomaly diagnosis method based on air-conditioning terminal monitoring data.

[0095] Optionally, when the program is executed by a processor, it can also be used to execute the technical solutions of any of the above-mentioned air-conditioning anomaly diagnosis methods provided in the embodiments of the present invention based on air-conditioning terminal monitoring data, and achieve the corresponding beneficial effects.

[0096] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments of the present invention can be implemented by means of software and the necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on this understanding, the technical solution of the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disc of a computer, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0097] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. This application is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed in the present invention. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present invention are pointed out by the claims of this application.

[0098] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. An air conditioner abnormal diagnosis method based on air conditioner terminal monitoring data, characterized in that, Including the following steps: Step S101, data monitoring and fusion: Collect the real-time monitoring data of the air conditioner in the room where the corresponding air conditioner is located through the air conditioner terminals of each air conditioner in the building air conditioning system, and at the same time collect meteorological data, and fuse the meteorological data and the real-time monitoring data of the air conditioner to form an initial data set; Step S102, preprocess the data: Preprocess the initial data set to obtain a preprocessed sample set; Step S103, identify the building air conditioning system model: Construct an ETP model of the building air conditioning system, and calculate the equivalent thermal resistance and equivalent heat capacity of the building air conditioning system; Step S104, estimate the air conditioner operating power: Obtain the estimated value and measured value of the air conditioner load power according to the ETP model of the building air conditioning system and the preprocessed sample set; Step S105, diagnose the air conditioner operating status: Judge whether the air conditioner operating status is abnormal according to the correlation coefficient between the estimated value and measured value of the air conditioner load power; In the step S101, the air conditioner terminal real-time monitoring and collection of the air conditioner monitoring data in the room where the corresponding air conditioner is located includes indoor temperature, indoor humidity, air conditioner mode, air conditioner set temperature, air conditioner wind speed and air conditioner power, and the meteorological data includes weather type, outdoor temperature, wind force; The initial data set X obtained in the step S101 is X = [X1, X2, …, X i , …, X q T , where q is the data type dimension, and X i ∈ R 1×p contains samples at p moments, where p, q, and i are all positive integers, and T represents matrix transpose;​ In the step S102, the "preprocessing the initial data set" includes the following steps: Step S1021, calculate the time-domain mean matrix of the dataset X where is the time-domain mean of the i-th type of data, and calculate the standard deviation σ of the i-th type of data i : Construct the standard deviation vector σ = [σ1, σ2, …, σ i , …, σ q T ,​ If the data set X at the j-th moment j =[X 1j ,X 2j ,…,X ij ,…,X qj T satisfies Equation (2), then the data set X at the j-th moment j is deleted as abnormal data:​ where |*| ρ denotes the L ρ -norm of *, ρ is 0, 1, 2, or ∞, j is a positive integer, and K is a set positive integer; Step S1022, check the time-domain coherence of various data after checking and eliminating abnormal data. If the continuous missing exceeds m time periods or the continuous startup duration is less than h hours, then all sample data in the time periods where these two situations of continuous missing exceeding m time periods or the air conditioner continuously starting up less than h hours are excluded from the sample set to obtain a refined data sample set, where m and h are both set positive integers; Step S1023, for the refined data sample set, use the interpolation method to fill in the sporadic missing data to obtain a preprocessed sample set; In the step S103, the "construct an ETP model of the building air conditioning system and calculate the equivalent thermal resistance and equivalent heat capacity of the building air conditioning system" includes the following steps: Step S1031, simulate the thermodynamics characteristics of the building air conditioning system, and initially construct an ETP model of the building air conditioning system: Among them, T in is the indoor temperature, T out is the outdoor temperature, R is the equivalent thermal resistance, C is the equivalent heat capacity, P is the air-conditioning load power, and η is the air-conditioning energy efficiency ratio; Step S1032, extract the indoor temperature T in the preprocessed sample set during the time period from TM0 to TM1 in , the outdoor temperature T out , and the air conditioner power P, and construct the sample data matrix for the Jth continuous operation period of the air conditioner N J is the number of sampling time points of the Jth group of sample data, TM0 is the initial operation time of the air conditioner, and TM1 is the node time of the sample data set; Step S1033, using each group of sample data matrices Y J , identify the ETP parameters of the equivalent thermal resistance R and the equivalent heat capacity C for each air conditioner startup period according to the ETP model of the building air conditioning system, and store them in the ETP dataset V of the building air conditioning system; Step S1034, classify the ETP data set V of the building air conditioning system, select the category with the largest data volume and calculate the central value of all ETP parameters in this category, and use it as the equivalent thermal resistance R and equivalent heat capacity C of the building air conditioning system; In the step S104, the "obtain the estimated value and measured value of the air conditioner load power according to the ETP model of the building air conditioning system and the preprocessed sample set" includes the following steps: Step S1041, after time TM1, according to the real-time air-conditioning monitoring data and meteorological data collected by the air-conditioning terminal, extract the indoor temperature T from the preprocessed sample set obtained in step S102 in , the outdoor temperature T out , together with the equivalent thermal resistance R and equivalent heat capacity C obtained in step S1034, substitute them into the building air-conditioning system ETP model to obtain the estimated value of the air-conditioning load power for the Jth consecutive startup period N J is the number of sampling time points of the Jth group of sample data, and J is a positive integer; Step S1042, extract the measured values of the air-conditioning load power during the Jth consecutive startup period after time TM1 from the preprocessed sample set obtained in step S102 In the step S105, the "judge whether the air conditioner operating status is abnormal according to the correlation coefficient between the estimated value and measured value of the air conditioner load power" includes the following steps: Step S1051: According to the latest L - group estimated values of air - conditioning load power P estimate and the measured values of air - conditioning load power P real , calculate the correlation coefficient γ estimate between the estimated value P real of the J - th group of air - conditioning load power and the measured value P J ; where J and L are both positive integers and 1 ≤ J ≤ L, P estimate,i is the i-th item of P estimate , P real,i is the i-th item of P real , is the mean of P estimate , is the mean of P real ; Step S1052, L correlation coefficients γ J If more than α% of them are lower than the threshold δ, it is determined that the operating state of the air conditioner is abnormal and a warning is issued. Otherwise, it is determined that the operating state of the air conditioner is normal.

2. The air conditioner abnormal diagnosis method based on air conditioner terminal monitoring data according to claim 1, wherein, In the step S1023, the interpolation method is linear interpolation method, mean interpolation method, polynomial interpolation method or spline interpolation method.

3. An air conditioner abnormal diagnosis device based on air conditioner terminal monitoring data, characterized in that, Applied to the air conditioner abnormal diagnosis method based on air conditioner terminal monitoring data according to any one of claims 1-2, including a data monitoring and fusion module, a preprocessing module, a model identification module, a power estimation module and an air conditioner diagnosis module; The data monitoring and fusion module is used to collect air-conditioning monitoring data in the room where the corresponding air conditioner is located in real time through the air-conditioning terminals provided in each air conditioner, collect meteorological data, and fuse the meteorological data and the air-conditioning monitoring data to form an initial data set; The preprocessing module is used to preprocess the initial data set to obtain a preprocessed sample set; The model identification module is used to construct an ETP model of the building air-conditioning system and calculate the equivalent thermal resistance and equivalent heat capacity of the building air-conditioning system; The power estimation module is used to obtain the estimated value and measured value of the air-conditioning load power according to the ETP model of the building air-conditioning system and the preprocessed sample set; The air-conditioning diagnosis module is used to judge whether the operation state of the air conditioner is abnormal according to the correlation coefficient between the estimated value and measured value of the air-conditioning load power.

4. A storage medium stores a program, characterized in that: When the program is executed by the processor, it implements the air-conditioning anomaly diagnosis method based on air-conditioning terminal monitoring data according to any one of claims 1-2.

Images