CN107806690A - A kind of domestic air conditioning method for diagnosing faults based on Bayesian network - Google Patents

A kind of domestic air conditioning method for diagnosing faults based on Bayesian network Download PDF

Info

Publication number
CN107806690A
CN107806690A CN201710824183.2A CN201710824183A CN107806690A CN 107806690 A CN107806690 A CN 107806690A CN 201710824183 A CN201710824183 A CN 201710824183A CN 107806690 A CN107806690 A CN 107806690A
Authority
CN
China
Prior art keywords
failure
domestic air
bayesian network
sign
air conditioning
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.)
Granted
Application number
CN201710824183.2A
Other languages
Chinese (zh)
Other versions
CN107806690B (en
Inventor
赵阳
李婷婷
张学军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201710824183.2A priority Critical patent/CN107806690B/en
Publication of CN107806690A publication Critical patent/CN107806690A/en
Application granted granted Critical
Publication of CN107806690B publication Critical patent/CN107806690B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Air Conditioning Control Device (AREA)

Abstract

The invention provides a kind of domestic air conditioner fault detection and diagnosis method based on Bayesian network.The Bayesian network describe most of domestic air conditioner typical fault structure-characterizedly and it relies on complicated causality between occurrence condition and failure symptom;Conditional probability table representated by the directed edge of the Bayesian network quantitatively describes the probable value in causality.This method can merge etiologic diagnosis information and quantitative data, make full use of the knowledge experience of industry specialists and diagnose the additional information of object, improve fault diagnosis efficiency and the degree of accuracy, realize diagnostic message it is imperfect, it is uncertain in the case of carry out accurate fault diagnosis.Bayesian network provided by the present invention being capable of the most of domestic air conditioner failure of effective detection diagnosis.

Description

A kind of domestic air conditioning method for diagnosing faults based on Bayesian network
Technical field
The invention belongs to air-conditioning system fault detection and diagnosis and artificial intelligence field, it is related to that information is uncertain, not comprehensive feelings Knowledge reasoning under condition, more particularly to a kind of domestic air conditioner method for diagnosing faults and technology based on Bayesian network.
Background technology
With the generally raising of people's living standard, the usage amount of domestic air conditioner increases year by year, it has also become uses at present Most commonly used air conditioning form.Because domestic air conditioner is in actual moving process, various failures may be produced, influence interior Hot comfort simultaneously brings energy waste.Traditional air conditioner fault diagnosis scheme is exactly the expertise by engineer, to special Industry competency profiling is higher, and wastes time and energy.Developing intellectual resource diagnostic tool is to improving field diagnostic efficiency, and based on high in the clouds Domestic air conditioner fault detection and diagnosis, have great importance.
The operation of air-conditioning is dynamic, time-varying, a nonlinear process, and the form of expression of failure, which has, not to be known Property.Domestic air conditioner does not often record service data, can not use physical modeling to its operation characteristic carry out accurate description, it is necessary to New fault detection and diagnosis algorithm is proposed in theoretic.
Bayesian network is one of current uncertain knowledge expression and the maximally effective theoretical model in reasoning field.It is description Complicated influence relation provides approach between domestic air conditioner failure and failure symptom, can describe between various stochastic variables Relation, by the knowledge, practical experience and diagnostic reasoning thinking of using for reference domain expert.
The content of the invention
The present invention is intended to provide a kind of information deficiency and the domestic air conditioner method for diagnosing faults under uncertain condition.The party Method is based on Bayes net algorithm, refines efficient Bayesian diagnostic network by building, and propose to be based on the Bayesian network The fault detection and diagnosis flow of network, realize and fault detection and diagnosis efficiently and rapidly are carried out to domestic air conditioner.
In view of this, the domestic air conditioning method for diagnosing faults of the invention based on Bayesian network walks including following basic operation Suddenly:
S1:The Bayesian diagnostic network for domestic air conditioning fault diagnosis is built, the diagnostic network is by additional information, failure Formed with the class node of sign three, established and contacted by directed edge according to air-conditioning failure mechanism and characteristic between three class nodes, formed Topological structure;Wherein each additional information node represents a kind of O&M situation of domestic air conditioning;Each malfunctioning node represents family expenses One potentially possible failure of air-conditioning;Each sign node represents the sign that domestic air conditioning breaks down;
S2:Conditional probability between prior probability and node is set to each node in Bayesian diagnostic network;
S3:The prognostic information of on-site collection domestic air conditioner to be diagnosed, Bayes is inputted when abnormal sign be present Network;
S4:The posterior probability of each malfunctioning node is updated, and finds two maximum failure P of posterior probability1stAnd P2nd, wherein P1st>P2nd
S5:Judge P1stAnd P2ndDifference whether exceed threshold value;If exceeding, by P1stCorresponding failure output is used as should Failure corresponding to abnormal sign;If not exceeded, then obtaining other based on Cost Benefit Principle measurement is used for the attached of auxiliary judgment Add information or prognostic information, and after being re-entered Bayesian network, return to step S4.
Further, other additional informations or prognostic information for being used for auxiliary judgment are obtained based on Cost Benefit Principle measurement Specific method is as follows:
With failure P1stRelated sign and the collection of additional information are combined into E1st, with failure P2ndRelated sign and additional letter The collection of breath is combined into E2nd, U=E1st∪E2nd, T=E1st∩E2nd, subset T supplementary set is C in UUT, to supplementary set CUAll indications in T Complexity is obtained with additional information to be ranked up, in-site measurement is easiest to the sign or additional information obtained, for aiding in sentencing Disconnected domestic air conditioning failure.
Further, described threshold value is set to 30%.
Further, described posterior probability is calculated by prior probability and conditional probability by Bayesian formula.
The domestic air conditioner method for diagnosing faults based on Bayesian network of the present invention, at least has the advantages that:
(1) set out based on graph theory thought, build domestic air conditioner fault diagnosis Bayesian network, describe qualitative and quantitative Failure and its rely on causality complicated between condition and sign performance and conditional relationship;
(2) Bayesian network has incorporated priori and additional information carries out Multi-source Information Fusion, possesses diagnostic message It is imperfect it is uncertain in the case of be capable of the ability of efficiently fault diagnosis;
(3) algorithm provided by the present invention can carry out Efficient fault diagnosis in the case of information deficiency, can reduce air-conditioning The cost of device fault diagnosis, it can be used in developing live auxiliary diagnostic tool and the remote diagnosis instrument based on high in the clouds.
Brief description of the drawings
Fig. 1 is the domestic air conditioner fault diagnosis network topological diagram established using Bayesian network.
Fig. 2 is the process for using figure of Bayes's fault detection and diagnosis algorithmic tool.
Embodiment
Make further explaination and explanation to the present invention with reference to the accompanying drawings and examples.
Domestic air conditioning method for diagnosing faults based on Bayesian network comprises the following steps:
(1) by carrying out going deep into anatomy to mechanism such as domestic air-conditioning system failure, signs, tieed up by expertise, scene Personnel's experience, historic survey result etc. are repaiied, summarizes typical fault list, the failure symptom list of domestic air conditioner.Fully obtain Take the O&M information of domestic air conditioner to be diagnosed, including running status, air-conditioner repair record and the operation note that air conditioner is current Record etc., forms the additional information list of auxiliary diagnosis.
(2) tissue is carried out to additional information, fault message, prognostic information, determined according to air-conditioning failure mechanism and characteristic attached Add the causality between information and failure, failure and sign, above three list is abstracted as malfunctioning node, sign node respectively With additional information node, according to the causality of foregoing determination between three class nodes, pass through directed edge and establish contact, form topology Structure.Wherein additional information node is represented with C={ C1, C2 ... ..., Ct }, illustrates the O&M situation of domestic air conditioning;Each Additional information node represents a kind of O&M situation of domestic air conditioning, the filling maintenance event of such as refrigerant, and domestic air conditioning Maintenance record, the cleaning treatment of such as maintenance of valve and replacing, screen pack and heat exchanger.Malfunctioning node with F=F1, F2 ... ..., Fn } to represent, express and diagnose the potential possible breakdown of object, each malfunctioning node represents one of domestic air conditioning Potentially possible failure, such as lack of refrigerant, incoagulable gas, fouling of evaporator.Sign node with E=E1, E2 ... ..., Em } represent, describe the common sign for identifying domestic air conditioning failure, evident information during for diagnosing, each sign Node represents the sign that domestic air conditioning breaks down, such as evaporating temperature, condensation temperature, the frozen air inlet and outlet temperature difference.
Thus domestic air conditioner fault diagnosis bayesian network structure is established.Bayesian network is by directed acyclic graph and condition What probability distribution table two parts were formed, this two parts represents the structure and parameter of Bayesian network respectively.Wherein, oriented no circle Figure is its qualitative part, and conditional probability is its dosing section, is premise and basis that data reasoning calculates.Directed acyclic graph is by saving Point and directed edge composition.Node variable can be the abstract of any problem, can be discrete or continuous.Discrete Bayesian network In network there is the state of different values, referred to as node in each node, conventional be two state of value (it is normal and abnormal, or Be with it is no), also have the state of more than three.In our current research, mainly additional information, failure and the failure symptom that node refers to, Each node has different status and appearances, and these status and appearances all can be considered an event;Directed edge between node represents Cross correlation between node, and the relationship strength that conditional probability can be used between description node.
(3) conditional probability between the prior probability and node of each node in Bayesian diagnostic network is determined.The present embodiment In, the method to set up of prior probability and conditional probability is as follows:
First, bayesian network structure can be determined by study mechanism and specificity analysis, it is determined that network node is opened up Flutter after order, it is thus necessary to determine that the conditional probability between the prior probability and node of each node.For setting for these probability numbers It is fixed, the degree that can be distinguished according to people, it is tentatively arranged to 7 grades:90%th, 75%, 60%, 50%, 25%, 10% and 1%, expert, domestic air conditioner operation maintenance personnel and the measuring technology personnel in consulting HVAC field etc., allow it to be selected from above-mentioned 7 values Most suitable probability numbers are selected, integrate multiple expertises, preliminary assignment is carried out to all prior probabilities and conditional probability.
Then, by practical operation situation of various air-conditioning system failure Survey and Inquiry results and research object etc., to elder generation Probability is tested further to be corrected.For example, filter screen dirty stifled, fouling of evaporator, the probability of happening of condenser fouling are 10% Left and right, still, because air conditioner filter gauze directly contacts with room air, and because its mesh is fine and closely woven, the possibility blocked Maximum, therefore its prior probability is arranged to 12%;And because the air that evaporator touches is to filter it by air conditioner filter gauze Air afterwards, dust impurity is less, and the air that condenser touches is the outdoor air without any processing, and dust is more, So the probability of condenser fouling is arranged to 11%, the probability of fouling of evaporator is arranged to 6%.
Amendment and adjustment finally, for conditional probability numerical value in terms of two, it is necessary to carry out, first, utilizing various air-conditioning systems The history data and investigation statisticses data of system, determine that the relation between node is strong and weak;Second, by specificity analysis, according to certain On direct/indirect influence of different signs and the difference of influence order when failure occurs, relative size relation is determined.For example, Indoor fan blocks or during the failure such as belt slippage, can cause increase and the evaporating temperature of the frozen air inlet and outlet temperature difference Reduction, according to expertise, the two probability are all between 75% and 60%.But due to indoor fan once occur it is above-mentioned Failure, most direct consequence are exactly that blast reduces, and air-supply air quantity reduces, now evaporating temperature or normal value, evaporation and heat-exchange amount Also it is normal, therefore frozen air inlet and outlet temperature difference increase, to ensure wind side heat exchange amount (cpM Δs t) is equal to evaporation and heat-exchange amount.When cold When the temperature difference increase of jelly air ports is also not enough to make up heat exchange amount reduction in wind side caused by air-supply air quantity is reduced, evaporation and heat-exchange amount Reduce, evaporating temperature reduces.So when fan trouble causes the rotating speed to reduce, the conditional probability for causing the frozen air temperature difference to increase will The conditional probability that slightly above evaporating temperature reduces.Certainly, the determination of prior probability and conditional probability can also use others Research method, the implementation in the present embodiment is above are only, is not intended to limit the present invention.
After the conditional probability that each sign occurs when the prior probability and failure that failure occurs occur determines, Bayes Network algorithm updates the posterior probability of the node including malfunctioning node, equation below based on Bayes' theorem:
For example, Bi can represent a kind of failure, such as lack of refrigerant, incoagulable gas, fouling of evaporator, A can be with table Show a kind of failure symptom, such as evaporating temperature, condensation temperature, the frozen air inlet and outlet temperature difference, the prior probability P that failure Bi occurs (Bi) and A occurs in the case where Bi occurs conditional probability P (A | Bi) is investigated, specially by servicing manual, historical data What family's experience and air-conditioning mechanism characteristic obtained.The size can that posterior probability P (Bi | A) is calculated is represented when observing The possibility that a certain failure occurs during sign A.
After completing above-mentioned diagnostic network model construction, you can carry out real-time diagnosis to domestic air conditioner failure.Utilize pattra leaves This formula, it is current in known all or part of sign node state table, the faulty probability of happening of institute can be gone out with reasoning and calculation, with This foundation made a definite diagnosis as further fault diagnosis or failure.
(4) when carrying out domestic air conditioning fault detection and diagnosis using Bayesian network, the fault diagnosis for using for reference industry specialists is thought Dimension, by purposefully obtaining part of nodes state, realizes accurate fault diagnosis under less amount of diagnostic information.Referring to It is the process for using figure of Bayes's fault detection and diagnosis algorithmic tool shown in Fig. 2.Specifically, fault detection and diagnosis are being carried out When, be primarily based on can live real-time collecting domestic air conditioner to be diagnosed prognostic information, when in the absence of abnormal sign, it is believed that Domestic air conditioner fault-free, normal operation.When abnormal sign be present, failure symptom node and the additional information section that can be obtained The observation of point, Bayesian network is inputted, updates the posterior probability of remaining node not observed, and find posterior probability maximum Two malfunctioning node P1stAnd P2nd, wherein P1st>P2nd, the most possible failure of the abnormal sign be the two failures it One.
(5) P is judged1st-P2ndWhether more than 30%, if exceeding, show P1stFor most probable failure, therefore by P1stIt is right The failure answered exports the failure corresponding to as the abnormal sign.If not less than 30%, P cannot be distinguished by1stAnd P2ndBased on whichever Want failure, it is therefore desirable to other additional informations or prognostic information are obtained, for aiding in determining whether.And the acquisition of this type of information Need to be based on Cost Benefit Principle, guarantee while cost minimization is considered as far as possible efficiently, rapidly confirm failure.This In embodiment, the acquisition methods of other additional informations or prognostic information are as follows:
With failure P1stRelated sign and the collection of additional information are combined into E1st, with failure P2ndRelated sign and additional letter The collection of breath is combined into E2nd, U=E1st∪E2nd, T=E1st∩E2nd, subset T supplementary set is C in UUT, the element in the supplementary set are It is all to distinguish failure P1stAnd P2ndSign or additional information.To supplementary set CUAll indications and additional information obtain difficult in T Easy degree is ranked up, the information that typically information that air-conditioning is torn machine open and can got can will be needed to be considered as more difficult acquisition, and incite somebody to action The i.e. obtainable information of machine need not be torn open and be considered as the information for being easier to obtain.Selection is easiest to the sign or additional information obtained, scene The observation of the information is measured, and after being re-entered Bayesian network, updates the posterior probability and failure of each node again Node P1stAnd P2nd, the deterministic process of this step is repeated again, untill the failure corresponding to the abnormal sign is exported.
Embodiment described above is a kind of preferable scheme of the present invention, and so it is not intended to limiting the invention.Have The those of ordinary skill of technical field is closed, without departing from the spirit and scope of the present invention, various changes can also be made Change and modification.Therefore the technical scheme that all modes for taking equivalent substitution or equivalent transformation are obtained, the guarantor of the present invention is all fallen within In the range of shield.

Claims (4)

  1. A kind of 1. domestic air conditioning method for diagnosing faults based on Bayesian network, it is characterised in that:
    S1:The Bayesian diagnostic network for domestic air conditioning fault diagnosis is built, the diagnostic network is by additional information, failure and sign 1000003 class nodes form, and are established and contacted by directed edge according to air-conditioning failure mechanism and characteristic between three class nodes, form topology Structure;Wherein each additional information node represents a kind of O&M situation of domestic air conditioning;Each malfunctioning node represents domestic air conditioning A potentially possible failure;Each sign node represents the sign that domestic air conditioning breaks down;
    S2:Conditional probability between prior probability and node is set to each node in Bayesian diagnostic network;
    S3:The prognostic information of on-site collection domestic air conditioner to be diagnosed, Bayesian network is inputted when abnormal sign be present;
    S4:The posterior probability of each malfunctioning node is updated, and finds two maximum failure P of posterior probability1stAnd P2nd, wherein P1st> P2nd
    S5:Judge P1stAnd P2ndDifference whether exceed threshold value;If exceeding, by P1stCorresponding failure output is used as the exception Failure corresponding to sign;If not exceeded, other additional letters for being used for auxiliary judgment are then obtained based on Cost Benefit Principle measurement Breath or prognostic information, and after being re-entered Bayesian network, return to step S4.
  2. 2. the domestic air conditioning method for diagnosing faults based on Bayesian network as claimed in claim 1, it is characterised in that:Based on into It is as follows for the additional information or prognostic information specific method of auxiliary judgment that the measurement of this principle of effectiveness obtains other:
    With failure P1stRelated sign and the collection of additional information are combined into E1st, with failure P2ndRelated sign and additional information Collection is combined into E2nd, U=E1st∪E2nd, T=E1st∩E2nd, subset T supplementary set is C in UUT, to supplementary set CUAll indications and attached in T Acquisition of information complexity is added to be ranked up, in-site measurement is easiest to the sign or additional information obtained, for auxiliary judgment man With air-conditioning failure.
  3. 3. the domestic air conditioning method for diagnosing faults based on Bayesian network as claimed in claim 1, it is characterised in that:Described Threshold value is set to 30%.
  4. 4. the domestic air conditioning method for diagnosing faults based on Bayesian network as claimed in claim 1, it is characterised in that:Described Posterior probability is calculated by prior probability and conditional probability by Bayesian formula.
CN201710824183.2A 2017-09-13 2017-09-13 A kind of domestic air conditioning method for diagnosing faults based on Bayesian network Active CN107806690B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710824183.2A CN107806690B (en) 2017-09-13 2017-09-13 A kind of domestic air conditioning method for diagnosing faults based on Bayesian network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710824183.2A CN107806690B (en) 2017-09-13 2017-09-13 A kind of domestic air conditioning method for diagnosing faults based on Bayesian network

Publications (2)

Publication Number Publication Date
CN107806690A true CN107806690A (en) 2018-03-16
CN107806690B CN107806690B (en) 2019-11-22

Family

ID=61591337

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710824183.2A Active CN107806690B (en) 2017-09-13 2017-09-13 A kind of domestic air conditioning method for diagnosing faults based on Bayesian network

Country Status (1)

Country Link
CN (1) CN107806690B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108614536A (en) * 2018-06-11 2018-10-02 云南中烟工业有限责任公司 A kind of complex network construction method of cigarette primary processing technology key factor
CN109740905A (en) * 2018-12-26 2019-05-10 国网陕西省电力公司电力科学研究院 A kind of secondary probability of malfunction estimation method of polynary power grid based on Bayesian model
CN111207484A (en) * 2019-12-13 2020-05-29 浙江大学 Central air-conditioning system fault diagnosis method based on object-oriented Bayesian network
CN111319425A (en) * 2020-03-18 2020-06-23 郑州新基业汽车电子有限公司 Vehicle-mounted air conditioning system of electric automobile
CN111461497A (en) * 2020-03-12 2020-07-28 许昌许继风电科技有限公司 Wind turbine generator early warning method and system with intelligent diagnosis function
CN112766047A (en) * 2020-12-29 2021-05-07 广东麦德克斯科技有限公司 Fault diagnosis method for refrigeration system and refrigeration device
CN113310171A (en) * 2021-05-24 2021-08-27 浙江大学 Central air-conditioning system fault detection and diagnosis method based on Bayesian network unit
US20210328861A1 (en) * 2018-09-10 2021-10-21 Nippon Telegraph And Telephone Corporation Diagnostic apparatus
US11531669B2 (en) 2020-08-21 2022-12-20 Siemens Industry, Inc. Systems and methods to assess and repair data using data quality indicators
US11625016B2 (en) 2020-08-21 2023-04-11 Siemens Industry, Inc. Systems and methods for HVAC equipment predictive maintenance using machine learning

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101915234A (en) * 2010-07-16 2010-12-15 西安交通大学 Method for diagnosing compressor-associated failure based on Bayesian network
CN103197177A (en) * 2013-03-20 2013-07-10 山东电力集团公司济宁供电公司 Transformer fault diagnosis analysis method based on bayesian network
US20160041070A1 (en) * 2014-08-05 2016-02-11 01dB-METRAVIB, Société par Actions Simplifiée Automatic Rotating-Machine Fault Diagnosis With Confidence Level Indication

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101915234A (en) * 2010-07-16 2010-12-15 西安交通大学 Method for diagnosing compressor-associated failure based on Bayesian network
CN103197177A (en) * 2013-03-20 2013-07-10 山东电力集团公司济宁供电公司 Transformer fault diagnosis analysis method based on bayesian network
US20160041070A1 (en) * 2014-08-05 2016-02-11 01dB-METRAVIB, Société par Actions Simplifiée Automatic Rotating-Machine Fault Diagnosis With Confidence Level Indication

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
YANG ZHAO, JIN WEN, SHENGWEI WANG: "Diagnostic Bayesian networks for diagnosing air handling units faultse Part II: Faults in coils and sensors", 《APPLIED THERMAL ENGINEERING》 *
YANG ZHAO,JINWEN, FU XIAO, XUEBIN YANG, SHENGWEIWANG: "Diagnostic Bayesian networks for diagnosing air handling units faults part I: Faults in dampers,fans,filters and sensors", 《APPLIED THERMAL ENGINEERING》 *
吴文忠: "基于贝叶斯方法分析蒸汽管网保温层损坏问题", 《区域供热》 *

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108614536A (en) * 2018-06-11 2018-10-02 云南中烟工业有限责任公司 A kind of complex network construction method of cigarette primary processing technology key factor
CN108614536B (en) * 2018-06-11 2020-10-27 云南中烟工业有限责任公司 Complex network construction method for key factors of cigarette shred making process
US20210328861A1 (en) * 2018-09-10 2021-10-21 Nippon Telegraph And Telephone Corporation Diagnostic apparatus
US11876670B2 (en) * 2018-09-10 2024-01-16 Nippon Telegraph And Telephone Corporation Diagnostic apparatus
CN109740905A (en) * 2018-12-26 2019-05-10 国网陕西省电力公司电力科学研究院 A kind of secondary probability of malfunction estimation method of polynary power grid based on Bayesian model
CN109740905B (en) * 2018-12-26 2021-06-15 国网陕西省电力公司电力科学研究院 Multivariate power grid secondary fault probability estimation method based on Bayesian model
CN111207484A (en) * 2019-12-13 2020-05-29 浙江大学 Central air-conditioning system fault diagnosis method based on object-oriented Bayesian network
CN111207484B (en) * 2019-12-13 2021-01-19 浙江大学 Central air-conditioning system fault diagnosis method based on object-oriented Bayesian network
CN111461497A (en) * 2020-03-12 2020-07-28 许昌许继风电科技有限公司 Wind turbine generator early warning method and system with intelligent diagnosis function
CN111319425A (en) * 2020-03-18 2020-06-23 郑州新基业汽车电子有限公司 Vehicle-mounted air conditioning system of electric automobile
US11531669B2 (en) 2020-08-21 2022-12-20 Siemens Industry, Inc. Systems and methods to assess and repair data using data quality indicators
US11625016B2 (en) 2020-08-21 2023-04-11 Siemens Industry, Inc. Systems and methods for HVAC equipment predictive maintenance using machine learning
CN112766047A (en) * 2020-12-29 2021-05-07 广东麦德克斯科技有限公司 Fault diagnosis method for refrigeration system and refrigeration device
CN113310171A (en) * 2021-05-24 2021-08-27 浙江大学 Central air-conditioning system fault detection and diagnosis method based on Bayesian network unit

Also Published As

Publication number Publication date
CN107806690B (en) 2019-11-22

Similar Documents

Publication Publication Date Title
CN107806690B (en) A kind of domestic air conditioning method for diagnosing faults based on Bayesian network
Kim et al. A review of fault detection and diagnostics methods for building systems
Gunay et al. Cluster analysis-based anomaly detection in building automation systems
Guo et al. Optimized neural network-based fault diagnosis strategy for VRF system in heating mode using data mining
Kocyigit Fault and sensor error diagnostic strategies for a vapor compression refrigeration system by using fuzzy inference systems and artificial neural network
Zhao et al. Diagnostic Bayesian networks for diagnosing air handling units faults–Part II: Faults in coils and sensors
Gunay et al. Characterization of a building's operation using automation data: A review and case study
Wang et al. A system-level fault detection and diagnosis strategy for HVAC systems involving sensor faults
Katipamula et al. Methods for fault detection, diagnostics, and prognostics for building systems—A review, part II
Wang et al. Fault detection based on Bayesian network and missing data imputation for building energy systems
Hu et al. A machine learning bayesian network for refrigerant charge faults of variable refrigerant flow air conditioning system
US20060042277A1 (en) Fault diagnostics and prognostics based on distance fault classifiers
Zhou et al. A model‐based fault detection and diagnosis strategy for HVAC systems
Li et al. Identification and isolation of outdoor fouling faults using only built-in sensors in variable refrigerant flow system: A data mining approach
Yu et al. A VRF charge fault diagnosis method based on expert modification C5. 0 decision tree
Zhou et al. Anomaly detection method of daily energy consumption patterns for central air conditioning systems
CN110500709B (en) Air conditioner running state online judging method
Lauro et al. Building fan coil electric consumption analysis with fuzzy approaches for fault detection and diagnosis
Yan et al. A Decentralized Boltzmann-machine-based fault diagnosis method for sensors of Air Handling Units in HVACs
CN112766047B (en) Fault diagnosis method for refrigeration system and refrigeration device
Chen et al. Using discrete Bayesian networks for diagnosing and isolating cross-level faults in HVAC systems
Guo et al. Modularized PCA method combined with expert-based multivariate decoupling for FDD in VRF systems including indoor unit faults
Li et al. An efficient online wkNN diagnostic strategy for variable refrigerant flow system based on coupled feature selection method
Yang et al. HVAC equipment, unitary: Fault detection and diagnosis
Es-sakali et al. Advanced predictive maintenance and fault diagnosis strategy for enhanced HVAC efficiency in buildings

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