CN1477353A - Fuzzy diagnosis method for air conditioner refigerator failure - Google Patents
Fuzzy diagnosis method for air conditioner refigerator failure Download PDFInfo
- Publication number
- CN1477353A CN1477353A CNA031164536A CN03116453A CN1477353A CN 1477353 A CN1477353 A CN 1477353A CN A031164536 A CNA031164536 A CN A031164536A CN 03116453 A CN03116453 A CN 03116453A CN 1477353 A CN1477353 A CN 1477353A
- Authority
- CN
- China
- Prior art keywords
- fault
- temperature
- kinds
- fuzzy
- compressor
- 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.)
- Pending
Links
Abstract
The fuzzy diagnosis method of air-conditioner refrigerator fault is characterized by that it uses PC machine as core and adopts the fuzzy diagnosis technology to build up fuzzy mapping from element set to remark set of the object to be jadged so as to resolve the fuzzy relationship. For brevity's sake. The present invention provides an example for utilizing PC machine and the fuzzy relationship to obtain the output of the fault signal so as to implement fuzzy diagnosis.
Description
Technical field
What the present invention relates to is a kind of air conditioner refrigerating device method for diagnosing faults.Particularly a kind of is the air conditioner refrigerating device fuzzy Diagnosis of Fault method that fault is discerned on the basis with the fuzzy mathematics.Belong to mechanical fault diagnosis and Refrigeration ﹠ Air-Conditioning technical field.
Background technology
The application of air conditioner refrigerating has been penetrated into fields of society, and air-conditioning refrigeration system is also complicated day by day, and equipment is tending towards variation and maximizes, and the automatic control degree improves day by day.The performance quality of refrigeration machine has a significant impact air-conditioning system, need carry out real-time online to its state and detect and fault diagnosis.At present, in the domestic air conditioner refrigerating station, many refrigeration machines only adopt single-sensor that absolute parameter (as temperature, pressure etc.) is monitored when operation, each monitored signal all has been set up thresholding, if the signal that samples surpasses thresholding, system just sends alarm signal, the place one's entire reliance upon sensitivity of sensor of its sensitivity.And when actual breaking down, refrigeration machine can show various features, and does not have absolute thresholding, and this has just caused the contradiction between warning sensitivity and the rate of false alarm.Therefore, carry out the failure mechanism of air conditioner refrigerating device and research and analyse, set up effectively, fault judgment model accurately, the optimization operation of refrigeration system real time and on line monitoring, the prediction of fault tendency and system is seemed very important.Because the complexity that concerns between refrigeration machine diversity structure and failure symptom and the failure cause, promptly with a kind of failure symptom often corresponding the various faults reason, with a kind of failure cause corresponding again the various faults symptom, make traditional diagnostic method expose their limitation in actual applications.And the fuzzy diagnosis method is the fault recognition method based on fuzzy mathematics, and problem to be solved is to judge the fault that is taken place according to observed symptom in the refrigeration machine running.Because this diagnostic method relatively meets the human mode of thinking, can overcome the difficulty that the machinery equipment complexity is brought, therefore has stronger practical value.In the machinery equipment Application in Fault Diagnosis some examples are arranged at present, but in air-conditioning refrigeration system, be not reported.
Summary of the invention
For deficiency and the defective that overcomes prior art, the present invention is used for the fault diagnosis of air conditioner refrigerating device with the fault Vague Diagnosis Technique, and this method is a core with the PC, sets up a FUZZY MAPPING from the element set of passing judgment on object to the comment collection.Behind each signal normalization of the required detection of air conditioner refrigerating device that will read in from data acquisition board, be quantized into several grades, obtain the value of its degree of membership according to fuzzy theory by lookup table mode, the weight according to various parameters is fused into a fuzzy quantity again, separates the output that obtains fault-signal at last.Concrete diagnostic method is as follows:
1, the at first supposition system parameter that need detect is condensation temperature, evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, six parameters of cooling water temperature; Phylogenetic fault is 7 kinds, and these 7 kinds of failure causes are: cold-producing medium is very few; The refrigeration system line clogging; The stall of vaporization chamber blower fan; Air inlet and exhaust valve of compressor leaks; The condenser cooling water flow is too little; Expansion valve is not opened or is got clogged; System oil return is bad.
2, concrete fault being carried out single factor passes judgment on.When for example condensation temperature is changed to Δ T, for fault 1, ask the expert to do judge, have 50% people to think to break down 1 possibility big, have 30% people to think in the possibility, having 10% people to think may be little, and remaining 10% people thinks no possibility.Then this Δ T to the evaluation result of fault 1 is: (0.5,0.3,0.1,0.1).
All can make possibility equally to different Δ T passes judgment on.For for simplicity, it is 0~8 grade point that Δ T is normalized into, and each grade point is made possibility respectively pass judgment on, and obtains condensation temperature and changes single factor of fault 1 passed judgment on and be subordinate to kilsyth basalt F
11Obtaining these six parameters of evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, cooling water temperature equally changes and is subordinate to kilsyth basalt F for fault 1
12, F
13... F
16Can obtain the be subordinate to kilsyth basalt of these six parameters variations for all the other six faults equally, the corresponding 7 kinds of conclusions of such 6 kinds of parameters are formed 42 single factor unijunctions opinions altogether and are subordinate to kilsyth basalt, and we exist in the PC these tables in order to searching.
3, determine weight allocation.Because each parameter is different for the influence of different faults, we will analyze its weight separately for this reason, obtain the weight allocation Table A.For example: for fault 1, the possibility maximum that takes place when condensation temperature, evaporating temperature change greatly, its weight should be maximum; Think by analysis that again indoor air temperature in air conditioned building changes weight and should take second place, Compressor Inlet Temperature, compressor exit temperature weight are less, and cooling water temperature then has nothing to do with it fully, and the weight allocation to 6 parameters of this fault can obtain so
In like manner, also there is different weight relationships to be for other 6 kinds of faults
These 7 weight allocation tables are existed in the PC, in order to inquiry and modification.
After fault occurred, six sensor input change in information amounts obtained grade point separately through normalization, quantification, corresponding to inquiring about being subordinate in the kilsyth basalt of this fault, draw the degree of membership value of this moment to separately.So obtain the multifactorial evaluation fuzzy relation matrix
Further carry out normalized, calculate the evaluation result of fault this moment
4, the utilization gravity model appoach carries out the reverse gelatinization to fuzzy vector, if the fault probability of happening is 90% when setting " possibility is big ", the fault probability of happening is 50% when " in the possibility ", the fault probability of happening is 10% when " possibility is little ", the fault probability of happening is 0 when " no possibility ", then can calculate fault probability of happening E this moment.
5, the last diagnostic of fault.The result of following formula is 7 fuzzy vectors, when not considering concerning between each fault, sticks with paste relational matrix (being passed judgment on by expert and statistics) inquiry according to the reverse of each fault oneself, obtains the alert levels of each fault oneself.Find that in practice certain contact is arranged between each fault, therefore, we need summarize this contact, sum up a fault relationship table.Each the fault probability of happening matrix that obtains at last can be stuck with paste relational matrix (being passed judgment on by expert and statistics) inquiry according to the reverse of each fault oneself, can obtain each fault alarm rank.
In sum, Vague Diagnosis Technique of the present invention is core with the PC, at first sets up a FUZZY MAPPING from the element set of passing judgment on object to the comment collection, promptly obtains fuzzy relation.For for simplicity, it is 0~8 grade point that Δ T is normalized into, each grade point is made possibility respectively pass judgment on, obtain condensation temperature, evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, these six parameters of cooling water temperature and change the kilsyth basalt that is subordinate to for fault 1.Equally, the phylogenetic fault of supposition is 7 kinds earlier, and the corresponding 7 kinds of conclusions of such 6 kinds of parameters are formed 42 single factor unijunction opinions altogether and are subordinate to kilsyth basalt, and we exist in the PC these tables in order to searching.Obtain the value of its degree of membership according to fuzzy theory by lookup table mode, the weight according to various parameters is fused into a fuzzy quantity again, and last ambiguity solution obtains the output of fault-signal.
Because said method has adopted the integrated and many information fusion treatment technology of multisensor, this Vague Diagnosis Technique has the following advantages:
1. reduced rate of false alarm.
2. owing to emphasized the comprehensive judgement of various information, accomplished the unification of sensitivity and rate of false alarm.
3. help under adverse circumstances, using.
The specific embodiment:
Come technical scheme of the present invention is further described below in conjunction with a specific embodiment
For refrigerating capacity 650kW idle call refrigerator system, the fuzzy fault diagnosis is roughly as follows:
1. on refrigeration machine, arrange condensation temperature, evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, six sensors of cooling water temperature, 6 signals that collect are changed send into PC by data collecting system.Suppose that the fault that refrigeration system takes place is 7 kinds.These 7 kinds of failure causes are: cold-producing medium is very few; The refrigeration system line clogging; The stall of vaporization chamber blower fan; Air inlet and exhaust valve of compressor leaks; The condenser cooling water flow is too little; Expansion valve is not opened or is got clogged; System oil return is bad.
2. system carries out single factor judge to concrete fault earlier.When for example condensation temperature is changed to Δ T, for fault 1, ask the expert to do judge, have 50% people to think to break down 1 possibility big, have 30% people to think in the possibility, having 10% people to think may be little, and remaining 10% people thinks no possibility.Then this Δ T to the evaluation result of fault 1 is: (0.5,0.3,0.1,0.1).
All can make possibility equally to different Δ T passes judgment on.For for simplicity, it is 0~8 grade point that Δ T is normalized into, and each grade point is made possibility respectively pass judgment on, and obtains condensation temperature and changes single factor of fault 1 passed judgment on and be subordinate to kilsyth basalt F
11Obtaining evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, cold temperature, these six parameters of cooling water temperature equally changes and is subordinate to kilsyth basalt F for fault 1
12, F
13... F
16In like manner obtain the be subordinate to kilsyth basalt of these six parameters variations to all the other six faults.The corresponding 7 kinds of conclusions of such 6 kinds of parameters are formed 42 single factor unijunction opinions altogether and are subordinate to kilsyth basalt, and we exist in the PC these tables in order to searching.System carries out single factor to concrete fault earlier and passes judgment on.
3. determine weight allocation
Each parameter is different for the influence of different faults, and we will analyze its weight separately for this reason, obtain the weight allocation Table A.For example: for fault 1, the possibility maximum that takes place when condensation temperature, evaporating temperature change greatly, its weight should be maximum; Think by analysis that again cold variations in temperature weight should take second place, Compressor Inlet Temperature, compressor exit temperature weight are less, and cooling water temperature then has nothing to do with it fully, and the weight allocation to 6 parameters of this fault can be made as so:
In like manner, also there is different weight relationships to be for other 6 kinds of faults
These 7 weight allocation tables are existed in the PC, in order to inquiry and modification.
4. single failure multifactorial evaluation
With a simulated failure is the flow process that example is made a concrete analysis of fuzzy diagnosis.After simulated failure three occurs, six sensors input change in information amounts through normalization, quantize to obtain grade point separately, be respectively 7,5,3,3,2,1, to being subordinate to kilsyth basalt F31, F32 corresponding to this fault separately,, to inquire about among the F36, the degree of membership value that draws this moment is respectively (0.8,0.1,0.1,0.0), (0.3,0.5,0.1,0.1), (0.3,0.3,0.2,0.2), (0.1,0.4,0.4,0.1), (0.1,0.2,0.5,0.2), (0.0,0.1,0.3,0.5).So obtain the multifactorial evaluation fuzzy relation matrix
The weight allocation Table A 3 of this moment through inquiry is:
Then fault 3 evaluation results are:
Further carry out getting after the normalized:
From top result as can be seen, the proportion maximum of " possibility is big " relatively tallies with the actual situation.
In like manner, calculate the evaluation result of all the other 6 kinds of faults this moment
5. the last diagnostic of fault
At first, the utilization gravity model appoach carries out the reverse gelatinization to fuzzy vector, if the fault probability of happening is 90% when setting " possibility is big ", the fault probability of happening is 50% when " in the possibility ", the fault probability of happening is 10% when " possibility is little ", the fault probability of happening is 0 when " no possibility ", and then can calculating at this moment, fault 3 probability of happening E3 are:
E
3=0.45×90%+0.22×50%+0.22×10%+0.01×0=53.7%
In like manner calculate other Trouble tickets and trigger the probability E that gives birth to
1, E
2..., E
7
The result of following formula is 7 fuzzy vectors, when not considering concerning between each fault, sticks with paste relational matrix (being passed judgment on by expert and statistics) inquiry according to the reverse of each fault oneself, obtains the alert levels of each fault oneself.
6. find in practice that certain contact is arranged between each fault, therefore, we need summarize this contact, sum up a fault relationship table, are shown below:
ω in the following formula
IjCause the probability of j item fault when being the generation of i item fault.Obvious 0≤ω
Ij≤ 1, and ω
Ij=1.So each final fault probability of happening matrix D is:
From following formula, the final probability that can get each fault i generation is:
At this moment, can be according to predefined each thresholding, the final probable value of each fault is converted into different alert levels, go to trigger corresponding warning behavior.
Claims (1)
1. an air conditioner refrigerating device method for diagnosing faults is characterized in that adopting the fuzzy mathematics diagnostic method, and this method comprises:
(1) on refrigeration machine, arranges condensation temperature, evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, six sensors of cooling water temperature, send into PC by data collecting system, suppose that phylogenetic fault is 7 kinds, these 7 kinds of failure causes are: cold-producing medium is very few; The refrigeration system line clogging; The stall of vaporization chamber blower fan; Air inlet and exhaust valve of compressor leaks; The condenser cooling water flow is too little; Expansion valve is not opened or is got clogged; System oil return is bad, and the corresponding 7 kinds of conclusions of such 6 kinds of parameters are formed 42 single factor unijunction opinions altogether and are subordinate to kilsyth basalt, and we exist in the PC these tables in order to searching;
(2) all can make possibility equally to the change amount signal of different sensors passes judgment on, for for simplicity, it is 0~8 grade point that the change amount signal of 6 sensors is normalized into, each grade point is made possibility respectively pass judgment on, obtain condensation temperature, evaporating temperature, Compressor Inlet Temperature, compressor exit temperature, indoor air temperature in air conditioned building, these six parameters of cooling water temperature and change and be subordinate to kilsyth basalt for fault;
(3) each parameter is different for the influence of different faults, and we will analyze its weight separately for this reason, obtains the weight allocation Table A and deposits PC in;
(4) reverse is stuck with paste and is handled, when not considering concerning between each fault, stick with paste the relational matrix inquiry, obtain the alert levels of each fault oneself according to the reverse of each fault oneself, in practice, find, certain contact is arranged between each fault, and therefore, we need summarize this contact, sum up a fault relationship table, each the fault probability of happening matrix that obtains at last can be stuck with paste the relational matrix inquiry according to the reverse of each fault oneself, can obtain the alert levels of each fault.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA031164536A CN1477353A (en) | 2003-04-17 | 2003-04-17 | Fuzzy diagnosis method for air conditioner refigerator failure |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA031164536A CN1477353A (en) | 2003-04-17 | 2003-04-17 | Fuzzy diagnosis method for air conditioner refigerator failure |
Publications (1)
Publication Number | Publication Date |
---|---|
CN1477353A true CN1477353A (en) | 2004-02-25 |
Family
ID=34152608
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA031164536A Pending CN1477353A (en) | 2003-04-17 | 2003-04-17 | Fuzzy diagnosis method for air conditioner refigerator failure |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN1477353A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101410677B (en) * | 2006-03-27 | 2010-12-08 | 大金工业株式会社 | Refrigeration system |
WO2011015135A1 (en) * | 2009-08-04 | 2011-02-10 | 华为技术有限公司 | Method and device for detecting system fault |
CN102175282A (en) * | 2011-01-24 | 2011-09-07 | 长春工业大学 | Method for diagnosing fault of centrifugal air compressor based on information fusion |
CN102270278A (en) * | 2011-07-21 | 2011-12-07 | 广东电网公司佛山供电局 | Method and device for forecasting equipment failure based on infrared temperature measurement |
CN105466143A (en) * | 2014-09-12 | 2016-04-06 | 苏州三星电子有限公司 | Detection method and system for refrigerator |
CN105975995A (en) * | 2016-05-26 | 2016-09-28 | 山东省计算中心(国家超级计算济南中心) | Fuzzy-preference-relation-based multi-vibration-signal fusion method |
CN106014946A (en) * | 2016-06-16 | 2016-10-12 | 珠海格力电器股份有限公司 | Compressor service life control method and device and refrigerating system |
CN106121980A (en) * | 2016-06-29 | 2016-11-16 | 珠海格力电器股份有限公司 | Determination method, device and the refrigeration system of a kind of compressor extent of deterioration |
CN106369879A (en) * | 2015-07-23 | 2017-02-01 | 青岛海尔空调电子有限公司 | Oil return control method and device for multi-split air-conditioning system |
CN112682928A (en) * | 2020-12-29 | 2021-04-20 | 珠海格力电器股份有限公司 | Control method, air conditioner, processor and storage medium |
CN116975502A (en) * | 2023-09-21 | 2023-10-31 | 天津津轨汇海科技发展有限公司 | Energy efficiency monitoring management system for electromechanical equipment of subway train air conditioning system |
-
2003
- 2003-04-17 CN CNA031164536A patent/CN1477353A/en active Pending
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101410677B (en) * | 2006-03-27 | 2010-12-08 | 大金工业株式会社 | Refrigeration system |
WO2011015135A1 (en) * | 2009-08-04 | 2011-02-10 | 华为技术有限公司 | Method and device for detecting system fault |
CN101626275B (en) * | 2009-08-04 | 2013-03-27 | 华为技术有限公司 | Method and device for detecting system fault |
CN102175282A (en) * | 2011-01-24 | 2011-09-07 | 长春工业大学 | Method for diagnosing fault of centrifugal air compressor based on information fusion |
CN102175282B (en) * | 2011-01-24 | 2012-07-25 | 长春工业大学 | Method for diagnosing fault of centrifugal air compressor based on information fusion |
CN102270278A (en) * | 2011-07-21 | 2011-12-07 | 广东电网公司佛山供电局 | Method and device for forecasting equipment failure based on infrared temperature measurement |
CN105466143A (en) * | 2014-09-12 | 2016-04-06 | 苏州三星电子有限公司 | Detection method and system for refrigerator |
CN105466143B (en) * | 2014-09-12 | 2018-03-02 | 苏州三星电子有限公司 | The detection method and system of a kind of refrigerator |
CN106369879A (en) * | 2015-07-23 | 2017-02-01 | 青岛海尔空调电子有限公司 | Oil return control method and device for multi-split air-conditioning system |
CN106369879B (en) * | 2015-07-23 | 2019-10-22 | 青岛海尔空调电子有限公司 | A kind of multi-line system method for controlling oil return and device |
CN105975995B (en) * | 2016-05-26 | 2019-03-15 | 山东省计算中心(国家超级计算济南中心) | More vibration signal fusion methods based on fuzzy preference relation |
CN105975995A (en) * | 2016-05-26 | 2016-09-28 | 山东省计算中心(国家超级计算济南中心) | Fuzzy-preference-relation-based multi-vibration-signal fusion method |
CN106014946A (en) * | 2016-06-16 | 2016-10-12 | 珠海格力电器股份有限公司 | Compressor service life control method and device and refrigerating system |
CN106121980A (en) * | 2016-06-29 | 2016-11-16 | 珠海格力电器股份有限公司 | Determination method, device and the refrigeration system of a kind of compressor extent of deterioration |
CN112682928A (en) * | 2020-12-29 | 2021-04-20 | 珠海格力电器股份有限公司 | Control method, air conditioner, processor and storage medium |
CN116975502A (en) * | 2023-09-21 | 2023-10-31 | 天津津轨汇海科技发展有限公司 | Energy efficiency monitoring management system for electromechanical equipment of subway train air conditioning system |
CN116975502B (en) * | 2023-09-21 | 2023-12-19 | 天津津轨汇海科技发展有限公司 | Energy efficiency monitoring management system for electromechanical equipment of subway train air conditioning system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Grimmelius et al. | On-line failure diagnosis for compression refrigeration plants | |
US7188482B2 (en) | Fault diagnostics and prognostics based on distance fault classifiers | |
CN100549574C (en) | Fault diagnosis and prediction based on the distance fault grader | |
Kocyigit | Fault and sensor error diagnostic strategies for a vapor compression refrigeration system by using fuzzy inference systems and artificial neural network | |
CN106006344B (en) | Staircase On-line Fault early warning system and method for diagnosing faults | |
JP4265982B2 (en) | Equipment diagnostic equipment, refrigeration cycle equipment, refrigeration cycle monitoring system | |
CN1477353A (en) | Fuzzy diagnosis method for air conditioner refigerator failure | |
Wang et al. | A robust fault detection and diagnosis strategy for centrifugal chillers | |
Braun | Automated fault detection and diagnostics for vapor compression cooling equipment | |
CN109858140B (en) | Fault diagnosis method for water chilling unit based on information entropy discrete Bayesian network | |
CN104089364A (en) | On-line defrosting detection and control system and method for heat pump type electric automobile air conditioner | |
CN102175282A (en) | Method for diagnosing fault of centrifugal air compressor based on information fusion | |
CN107806690A (en) | A kind of domestic air conditioning method for diagnosing faults based on Bayesian network | |
CN114484731A (en) | Method and device for diagnosing faults of central air conditioner based on stacking fusion algorithm | |
CN102789676A (en) | Method for designing industrial alarm on basis of alarm evidence fusion | |
CN110500709A (en) | The online method of discrimination of conditioners running conditions | |
CN110779264B (en) | Defrosting control system and method based on fan vibration and information fusion | |
CN114487361A (en) | Oil on-line monitoring system | |
CN116720150B (en) | Mechanical refrigeration system fault diagnosis method and system | |
CN203940577U (en) | A kind of pump type heat electric automobile air conditioner online measurement and control system that defrosts | |
CN109872511A (en) | A kind of adaptive two-stage alarm method for axial displacement mutation monitoring | |
Laughman | Fault detection methods for vapor-compression air conditioners using electrical measurements | |
WO2021256119A1 (en) | Abnormality detection device, program, and electrical device system | |
CN115356133A (en) | Bogie operation condition on-line monitoring, real-time classification and identification algorithm | |
Halm-Owoo et al. | Applications of fault detection and diagnostic techniques for refrigeration and air conditioning: A review of basic principles |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |