CN109668588A - Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor - Google Patents

Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor Download PDF

Info

Publication number
CN109668588A
CN109668588A CN201910144270.2A CN201910144270A CN109668588A CN 109668588 A CN109668588 A CN 109668588A CN 201910144270 A CN201910144270 A CN 201910144270A CN 109668588 A CN109668588 A CN 109668588A
Authority
CN
China
Prior art keywords
sensor
virtual
refrigeration unit
air cooling
cooling refrigeration
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
Application number
CN201910144270.2A
Other languages
Chinese (zh)
Inventor
李冬辉
高龙
李丁
尹海燕
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.)
Tianjin University
Original Assignee
Tianjin University
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 Tianjin University filed Critical Tianjin University
Priority to CN201910144270.2A priority Critical patent/CN109668588A/en
Publication of CN109668588A publication Critical patent/CN109668588A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01DMEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D18/00Testing or calibrating of apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computer systems based on biological models
    • G06N3/02Computer systems based on biological models using neural network models
    • G06N3/04Architectures, e.g. interconnection topology
    • G06N3/049Temporal neural nets, e.g. delay elements, oscillating neurons, pulsed inputs

Abstract

The present invention relates to a kind of air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor, comprising the following steps: correlation analysis is carried out to any two sensor in air cooling refrigeration unit;According to correlation analysis as a result, being grouped to refrigeration unit sensor and obtaining different groups;Virtual-sensor is constructed in each group;By virtual-sensor reading compared with respective sensor actual read number, if virtual-sensor reading mutates, there are failures for respective sensor.The present invention has rational design, it uses correlation analysis to analyze existing correlation between air cooling refrigeration unit sensor, the strong sensor of correlation is divided into one group, and virtual-sensor is constructed using long short-term memory Recognition with Recurrent Neural Network in group, by judging whether virtual-sensor reading mutates, to accurately and rapidly be positioned to refrigeration unit multisensor failure, energy waste is avoided, the service life of equipment is improved.

Description

Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor
Technical field
The invention belongs to air cooling refrigeration unit multi-sensor technology field, especially a kind of wind based on virtual-sensor Cold type refrigeration unit sensor fault diagnosis method.
Background technique
With increasingly sharpening for world energy sources shortage problem, energy conservation and environmental protection has become current mostly important project, and builds Industry is built as high-energy source and consumes industry, the influence to ecological environment can not be ignored.Energy consumption in construction industry is very big by one It is derived partly from the energy consumption of building air conditioning after coming into operation, when air conditioning sensor breaks down, can lead to the energy of air-conditioning system Loss-rate increases by 50% under normal circumstances.Therefore, energy saving reliability service of the refrigeration unit multiple sensor faults diagnosis to air-conditioning system It plays a crucial role, it is in the urgent need to address at present for how detecting refrigeration unit multisensor failure accurately and timely Problem.
Summary of the invention
It is an object of the invention to overcome the deficiencies in the prior art, propose that a kind of design is reasonable and can sufficiently excavate refrigeration machine The air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor of group multisensor behavioral characteristics.
The present invention solves its technical problem and adopts the following technical solutions to achieve:
A kind of air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor, comprising the following steps:
Step 1 carries out correlation analysis to any two sensor in air cooling refrigeration unit;
Step 2, according to correlation analysis as a result, being grouped to refrigeration unit sensor and obtaining different groups;
Step 3 constructs virtual-sensor in each group;
Step 4 reads virtual-sensor compared with respective sensor actual read number, if virtual-sensor reading occurs to dash forward Become, then there are failures for respective sensor.
The sensor includes temperature sensor, flow sensor and the pressure sensor of air cooling refrigeration unit.
The step 1 carries out correlation analysis using pearson related coefficient method.
When the step 2 is grouped refrigeration unit sensor, the sensor by correlation greater than 0.5 is grouped into the same group Not;If a sensor is both less than 0.5 with other all the sensors correlations, the maximum sensor of property associated therewith is chosen Place group is group where the sensor.
The concrete methods of realizing of the step 3 are as follows: in different groups, recycle nerve net using long short-term memory respectively Network constructs virtual-sensor.
The method of the long short-term memory Recognition with Recurrent Neural Network construction virtual-sensor are as follows: if containing n sensing in group Device then constructs n long short-term memory Recognition with Recurrent Neural Network respectively, wherein neural network input is all sensings in same group The measured value at T moment before device, neural network output are the corresponding virtual-sensor of different sensors, and T is any greater than 0 Number.
The judgment method that the step 4 virtual-sensor reading mutates are as follows: if current time virtual-sensor and reality The difference of last moment virtual-sensor of the difference of border sensor reading greater than twice and real sensor reading, then it is assumed that virtual to pass Sensor reading mutates.
The advantages and positive effects of the present invention are:
The present invention has rational design, uses correlation analysis to existing phase between air cooling refrigeration unit sensor Closing property is analyzed, and the strong sensor of correlation is divided into one group, and long short-term memory Recognition with Recurrent Neural Network structure is utilized in group Virtual-sensor is built out, by judging whether virtual-sensor reading mutates, thus accurately and rapidly more to refrigeration unit Sensor fault is positioned, and energy waste is avoided, and improves the service life of equipment.
Detailed description of the invention
Fig. 1 is process flow diagram of the invention.
Specific embodiment
The embodiment of the present invention is further described below in conjunction with attached drawing.
A kind of air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor, as shown in Figure 1, include with Lower step:
Step 1 carries out correlation point to any two sensor in air cooling refrigeration unit by pearson related coefficient Analysis.
Sensor of the invention includes all the sensors such as temperature, flow, the pressure of air cooling refrigeration unit.For difference The fault diagnosis of seed type sensor must train different long short-term memory Recognition with Recurrent Neural Network models, only deposit for synchronization The sensor failure the case where.
Step 2, according to correlation analysis as a result, be grouped to refrigeration unit sensor, wherein correlation is greater than 0.5 Sensor is same group.
In this step, according to correlation analysis as a result, being grouped to refrigeration unit sensor, wherein correlation is greater than 0.5 sensor is same group, if a certain sensor C and other all the sensors correlations are both less than 0.5, selection and its Group where the sensor of correlation maximum is group where C.
Step 3 constructs virtual-sensor in each group.
In this step, virtual-sensor is constructed using long short-term memory Recognition with Recurrent Neural Network in each group, in difference Group in, if containing n sensor, be respectively necessary for a long short-term memory Recognition with Recurrent Neural Network of construction n, wherein neural network Input is the measured value at T moment before all the sensors in same group, and T is the arbitrary number greater than 0, long short-term memory circulation mind It is the corresponding virtual-sensor of different sensors through network output.
Step 4 reads virtual-sensor compared with respective sensor actual read number, if virtual-sensor reading occurs to dash forward Become, then there are failures for respective sensor.
In this step, if the difference of current time virtual-sensor and real sensor reading is greater than twice of last moment The difference of virtual-sensor and real sensor reading, then it is assumed that virtual-sensor reading mutates.
The present invention does not address place and is suitable for the prior art.
It is emphasized that embodiment of the present invention be it is illustrative, without being restrictive, therefore packet of the present invention Include and be not limited to embodiment described in specific embodiment, it is all by those skilled in the art according to the technique and scheme of the present invention The other embodiments obtained, also belong to the scope of protection of the invention.

Claims (7)

1. a kind of air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor, it is characterised in that including following Step:
Step 1 carries out correlation analysis to any two sensor in air cooling refrigeration unit;
Step 2, according to correlation analysis as a result, being grouped to refrigeration unit sensor and obtaining different groups;
Step 3 constructs virtual-sensor in each group;
Step 4 reads virtual-sensor compared with respective sensor actual read number, if virtual-sensor reading mutates, Then there are failures for respective sensor.
2. the air cooling refrigeration unit sensor fault diagnosis method according to claim 1 based on virtual-sensor, Be characterized in that: the sensor includes temperature sensor, flow sensor and the pressure sensor of air cooling refrigeration unit.
3. the air cooling refrigeration unit sensor fault diagnosis method according to claim 1 based on virtual-sensor, Be characterized in that: the step 1 carries out correlation analysis using pearson related coefficient method.
4. the air cooling refrigeration unit sensor fault diagnosis method according to claim 1 based on virtual-sensor, It is characterized in that: when the step 2 is grouped refrigeration unit sensor, the sensor that correlation is greater than 0.5 being divided into same Group;If a sensor is both less than 0.5 with other all the sensors correlations, the maximum sensing of property associated therewith is chosen Group where device is group where the sensor.
5. the air cooling refrigeration unit sensor fault diagnosis method according to claim 1 based on virtual-sensor, It is characterized in that: the concrete methods of realizing of the step 3 are as follows: in different groups, recycle nerve using long short-term memory respectively Net structure virtual-sensor.
6. the air cooling refrigeration unit sensor fault diagnosis method according to claim 5 based on virtual-sensor, It is characterized in that: the method for the long short-term memory Recognition with Recurrent Neural Network construction virtual-sensor are as follows: if containing n sensing in group Device then constructs n long short-term memory Recognition with Recurrent Neural Network respectively, wherein neural network input is all sensings in same group The measured value at T moment before device, neural network output are the corresponding virtual-sensor of different sensors, and T is any greater than 0 Number.
7. the air cooling refrigeration unit sensor fault diagnosis method according to claim 1 based on virtual-sensor, It is characterized in that: the judgment method that the step 4 virtual-sensor reading mutates are as follows: if current time virtual-sensor and reality The difference of last moment virtual-sensor of the difference of border sensor reading greater than twice and real sensor reading, then it is assumed that virtual to pass Sensor reading mutates.
CN201910144270.2A 2019-02-27 2019-02-27 Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor Pending CN109668588A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910144270.2A CN109668588A (en) 2019-02-27 2019-02-27 Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910144270.2A CN109668588A (en) 2019-02-27 2019-02-27 Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor

Publications (1)

Publication Number Publication Date
CN109668588A true CN109668588A (en) 2019-04-23

Family

ID=66151856

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910144270.2A Pending CN109668588A (en) 2019-02-27 2019-02-27 Air cooling refrigeration unit sensor fault diagnosis method based on virtual-sensor

Country Status (1)

Country Link
CN (1) CN109668588A (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101331504A (en) * 2005-11-18 2008-12-24 卡特彼勒公司 Process model based virtual sensor system and method
CN101681155A (en) * 2007-06-15 2010-03-24 卡特彼勒公司 Virtual sensor system and method
EP2413205A1 (en) * 2009-03-27 2012-02-01 Honda Motor Co., Ltd. Controller for plant
CN103646179A (en) * 2013-12-19 2014-03-19 宜春市脉恩多能科技有限公司 Method for measuring refrigerating capacity of air conditioner by virtual sensor
CN105446821A (en) * 2015-11-11 2016-03-30 哈尔滨工程大学 Improved neural network based fault diagnosis method for intelligent underwater robot propeller
CN106640548A (en) * 2016-12-19 2017-05-10 北京金风科创风电设备有限公司 State monitoring method and device for wind generating set

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101331504A (en) * 2005-11-18 2008-12-24 卡特彼勒公司 Process model based virtual sensor system and method
CN101681155A (en) * 2007-06-15 2010-03-24 卡特彼勒公司 Virtual sensor system and method
EP2413205A1 (en) * 2009-03-27 2012-02-01 Honda Motor Co., Ltd. Controller for plant
CN103646179A (en) * 2013-12-19 2014-03-19 宜春市脉恩多能科技有限公司 Method for measuring refrigerating capacity of air conditioner by virtual sensor
CN105446821A (en) * 2015-11-11 2016-03-30 哈尔滨工程大学 Improved neural network based fault diagnosis method for intelligent underwater robot propeller
CN106640548A (en) * 2016-12-19 2017-05-10 北京金风科创风电设备有限公司 State monitoring method and device for wind generating set

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王路瑶 等: "基于长短期记忆神经网络的数据中心空调系统传感器故障诊断", 《化工学报》 *

Similar Documents

Publication Publication Date Title
Liu et al. Wind power plant prediction by using neural networks
CN103414251B (en) A kind of substation network communication analogue system based on IEC61850
CN108931387B (en) Fault diagnosis method based on multi-sensor signal analysis and capable of providing accurate diagnosis decision
CN104698837B (en) A kind of time-varying linear structure operational modal parameter recognition methods, device and application
CN104483575B (en) Self-adaptive load event detection method for noninvasive power monitoring
CN106199305B (en) Underground coal mine power supply system dry-type transformer insulation health state evaluation method
CN105846780B (en) A kind of photovoltaic module method for diagnosing faults based on decision-tree model
CN103176077B (en) The reliability fast assessing method of a kind of numerical control finished circuit board under environment comprehensive effect
CN103631201B (en) A kind of numerically-controlled machine subsystem reliability Impact analysis method
CN106777984B (en) A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm
CN102707256B (en) Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter
CN102945311B (en) A kind of method that functional fault directed graph carries out fault diagnosis
CN105894133B (en) Wind turbine component maintenance and spare part demand forecasting method
CN102981452B (en) Method for modeling and evaluating reliability of three types of functional components of numerical control machine tool
CN101916311B (en) Model development and simulation test system and method for rail transit automatic pilot system
CN103983453A (en) Differentiating method of fault diagnosis of executing mechanism of aeroengine and sensor
CN104454785B (en) A kind of engineering machinery hydraulic cylinder fault diagnosis system and the fault sample signal acquisition method being suitable for therewith
CN105759201B (en) High-voltage circuitbreaker self-diagnosing method based on exceptional sample identification
CN101382439B (en) Multi-parameter self-confirming sensor and state self-confirming method thereof
CN107222362B (en) Automatic test platform for finished vehicle CAN network and optimization method thereof
CN1321328C (en) Wavelet diagnostic system for initial failure of electromotor and method for diagnosing malfunction of electromotor
CN104443427A (en) Aircraft flutter prediction system and method
CN106100579B (en) A kind of photovoltaic plant method for diagnosing faults based on data analysis
CN105183952B (en) A kind of power transmission network method for diagnosing faults based on separation time Fuzzy Petri Net
CN103278328B (en) Method for diagnosing failure of hydraulic variable-pitch system of wind turbine generator based on fuzzy Petri net

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