CN111412579B - Air conditioning unit fault type diagnosis method and system based on big data - Google Patents
Air conditioning unit fault type diagnosis method and system based on big data Download PDFInfo
- Publication number
- CN111412579B CN111412579B CN202010225020.4A CN202010225020A CN111412579B CN 111412579 B CN111412579 B CN 111412579B CN 202010225020 A CN202010225020 A CN 202010225020A CN 111412579 B CN111412579 B CN 111412579B
- Authority
- CN
- China
- Prior art keywords
- fault
- air conditioning
- conditioning unit
- data
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004378 air conditioning Methods 0.000 title claims abstract description 101
- 238000003745 diagnosis Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims description 55
- 230000008439 repair process Effects 0.000 claims abstract description 52
- 238000012423 maintenance Methods 0.000 claims abstract description 22
- 238000005265 energy consumption Methods 0.000 claims abstract description 21
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 20
- 238000011161 development Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 38
- 238000012544 monitoring process Methods 0.000 claims description 37
- 239000013598 vector Substances 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 11
- 230000005611 electricity Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 210000001503 joint Anatomy 0.000 claims description 7
- 238000013499 data model Methods 0.000 claims description 5
- 230000000977 initiatory effect Effects 0.000 claims description 5
- 238000007726 management method Methods 0.000 claims description 5
- 230000015654 memory Effects 0.000 claims description 5
- 230000006403 short-term memory Effects 0.000 claims description 5
- 230000007787 long-term memory Effects 0.000 claims description 3
- 206010063385 Intellectualisation Diseases 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 10
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009290 primary effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 108010076504 Protein Sorting Signals Proteins 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F11/32—Responding to malfunctions or emergencies
- F24F11/38—Failure diagnosis
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/52—Indication arrangements, e.g. displays
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/56—Remote control
- F24F11/58—Remote control using Internet communication
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/008—Subject matter not provided for in other groups of this subclass by doing functionality tests
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2130/00—Control inputs relating to environmental factors not covered by group F24F2110/00
- F24F2130/10—Weather information or forecasts
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2140/00—Control inputs relating to system states
- F24F2140/60—Energy consumption
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Mechanical Engineering (AREA)
- Combustion & Propulsion (AREA)
- Chemical & Material Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Biomedical Technology (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
Abstract
The invention solves the problem of fault type diagnosis of the air conditioning unit and improves the level of maintenance, refinement and intellectualization of important equipment in the building. Butting the big data of the running state of the air conditioning unit; collecting real-time power consumption data, calculating unit energy consumption, and collecting real-time weather information; determining a fault risk curve for a repair reporting system; automatically extracting fault situations; establishing an artificial intelligence fault diagnosis model; and carrying out fault diagnosis on the air conditioning unit. Aiming at the characteristics of the air conditioning unit, a fine air conditioning unit sensor Internet of things is connected, an intelligent model for diagnosing the specific reasons of the occurrence of the fault and the fault development stage is provided on the basis, the accuracy rate of the algorithm can reach 80%, and the automatic field engineering application can be supported.
Description
Technical Field
The invention belongs to the field of building operation and maintenance, and relates to a fault type diagnosis method and system for an air conditioning unit based on Internet of things monitoring big data.
Background
The air conditioning unit is common large-scale equipment in buildings, and especially plays a vital role in temperature and humidity regulation of complex public buildings such as hospitals, airport buildings, libraries, commercial office buildings and the like. If the air conditioning unit breaks down, the environment in the building is directly deteriorated, and negative experience is brought to users. Therefore, how to diagnose the fault of the air conditioning unit equipment in time to rapidly guide maintenance is a problem to be solved urgently.
The traditional unit fault diagnosis method can identify fault hidden dangers to a certain extent by depending on field manual inspection and master-slave experience judgment, but has the defects of time and labor consumption, non-objectivity and easiness in omission. With the popularization of Building Automation Systems (BAS), in recent years, important devices such as air conditioning units can automatically acquire real-time operation states and reflect the real-time operation states to a unified information-based central control System. This provides a material basis for more advanced solutions, and the existing technology and its limitations can be described as follows: (1) semi-automated diagnosis and prognosis based on statistical rules: after the operation parameters are automatically captured, a preset fault judgment rule is input, and the fault judgment rule is reported in a data center in a unified way, but the specific fault type still needs to be judged manually, and the subjective experience is relied on; (2) the data mining method comprises the following steps: a model of fault and monitoring data is established by adopting data mining algorithms such as a decision tree, a support vector machine and the like, the speed is high, the accuracy is general, and the deep relation behind the monitoring data cannot be fitted due to insufficient model scale; (3) the artificial intelligence method comprises the following steps: more intelligent diagnostic models are built based primarily on varieties of artificial neural networks, but their input is essentially limited to a single point-in-time data or one-dimensional signal sequence. According to the reliability theory, the occurrence of the complex unit fault is a process which gradually develops along with time, and the evolution process of various monitoring data in the fault time period must be considered.
In addition, when the prior art is used for processing the problems of the related equipment, the adopted data is often not rich enough, the actual running state of the equipment cannot be accurately restored, the result accuracy is questionable, and the confidence coefficient of the actual application is not high. This is mainly due to: (1) the sensors are not fine, and the number and the fine degree of monitoring point positions are limited by the technology, so that the acquired state information is not comprehensive; (2) only the BA data of the unit is extracted, and other valuable operation data are not included, for example, the abnormal fluctuation of the energy consumption curve of a large unit is important abnormal information, the real-time weather condition can also influence the operation of an air conditioner, and the like. But existing methods rarely take these external variables into account; (3) the existing database technology generally cannot bear the storage and operation of extremely large monitoring data, only a part of data can be abandoned, and information can be lost to a certain extent.
In order to solve the problems, the application provides a set of novel air conditioning unit fault type diagnosis method and system based on big data. Aiming at the characteristics of the air conditioning unit, a fine air conditioning unit sensor Internet of things is built, and a large amount of monitoring data is stored and operated by adopting a big data technology. The core algorithm introduces the LSTM network to solve the problem of complex relationship among various monitoring data, and simultaneously, the time sequence characteristics of the monitoring data can be considered, so that not only can the specific reason of the fault occurrence be given, but also the stage to which the potential fault develops can be predicted before the fault does not occur. And finally, an automatic field engineering application scheme is given.
Disclosure of Invention
In view of the above, the invention provides a set of novel air conditioning unit fault type diagnosis and method and system based on big data, and aims at the characteristics of an air conditioning unit, a fine air conditioning unit sensor internet of things is built, and a big data technology is adopted to store and operate massive monitoring data. The core algorithm of the invention solves the problem of complex relationship among various monitoring data, and simultaneously can consider the time sequence characteristic of the monitoring data and give out the specific reason of the fault. The invention solves the problems of low automation degree, incomplete data and unreliable results in the aspect of fault diagnosis of the air conditioning unit, greatly improves the operation and maintenance efficiency of the air conditioning unit, can comprehensively reflect the operation condition of the unit, and improves the maintenance, refinement and intelligentization levels of important equipment in buildings.
According to a first aspect of the invention, a big data-based fault type diagnosis method for an air conditioning unit is provided, which is characterized by comprising the following steps: step 101, butt joint of big data of the running state of an air conditioning unit; step 102, collecting real-time electricity consumption data, calculating energy consumption, and collecting real-time weather information; 103, determining a fault risk curve for a repair system, and automatically extracting a fault situation; step 104, establishing an artificial intelligence fault type diagnosis model; and 105, diagnosing the fault type of the fault, and preferably reporting the result to an operation and maintenance management department.
Further, the step 101 further includes: the running state big data of the air conditioning unit is collected by arranging sensors in all the air conditioning fresh air units.
Further, the step 102 further includes: setting an intelligent ammeter to acquire real-time power consumption data of the air conditioning unit; establishing an energy consumption data model of the equipment to extract an energy consumption standardized value; and collecting real-time weather information.
Further, the step 103 further includes: the method comprises the following steps that the repair reporting system is connected with an existing repair reporting server or a newly-built software system, collects fault repair reporting records related to an air conditioning unit and serves as a basis for determining the fault time T and the type of the air conditioning unit, and determines the fault type K and the development rule thereof corresponding to a repair work order through a keyword matching method, so that a fault risk curve P is determined as { pi }; the extracted fault situation comes from the repair system, is a real, influencing fault recognized by the logistical personnel, and does not use the data of the alarm sensors of the machine itself. The fault situation is a set of a series of parameters, and comprises various parameters related to the fault within a range of forward pushing 24 hours from the fault occurrence time; the initiation of each repair order triggers an automatic extraction of the fault condition.
Further, the step 104 further includes: step 401: preprocessing the acquired data to generate a training set; step 402: adopting a long-short term memory network to construct an artificial intelligence algorithm model; step 403: model training and parameter tuning are performed.
Further, the step 105 further includes: inputting data acquired by a sensor into a diagnotor, if the predicted value of the obtained fault probability is greater than 0.5, sending an alarm message as a fault, and calculating the corresponding fault type from the output vector output of the diagnotor; and the alarm message is sent to a manager through the building operation and maintenance client or the mobile phone.
Further, the loss function in the long-short term memory network is taken as Mean Absolute Error (MAE), which is an average value of the distance between the predicted value of the model and the true value of the sample, and is defined as follows:
According to a second aspect of the present invention, there is provided a big data based fault type diagnosis system for an air conditioning unit, comprising: the state monitoring module is used for butting the big running state data of the air conditioning unit; the data acquisition module is used for acquiring real-time electricity consumption data, calculating energy consumption and acquiring real-time weather information; the repair reporting module is used for determining a fault risk curve and automatically extracting a fault situation for the repair reporting system; the diagnostic module is used for establishing an artificial intelligent fault type diagnosis model; and the alarm module diagnoses the fault and reports the result to the operation and maintenance management department.
Further, the status monitoring module further comprises: the big data of the running state of the air conditioning unit is collected through a sensor arranged in the butt joint air conditioning fresh air unit.
Further, the data acquisition module further comprises: the method comprises the steps that a butt joint intelligent ammeter collects real-time electricity consumption data of an air conditioning unit; establishing an energy consumption data model of the equipment to extract an energy consumption standardized value; and collecting real-time weather information.
Further, the repair module further comprises: the repair reporting system is in butt joint with an existing repair reporting server or a newly-built software system, collects fault repair reporting records related to the air conditioning unit, and uses the fault repair reporting records as a basis for determining the fault time T and the type of the air conditioning unit, and determines the fault type K and the development rule thereof corresponding to a repair reporting work order through a keyword matching method, so that a fault risk curve P is determined to be { pi }. The extracted fault conditions come from the repair system, are real occurring, influencing, faults recognized by the logistical personnel, rather than using the machine's own alarm sensor data. The fault situation is a set of a series of parameters, and comprises various parameters related to the fault within a range of forward pushing 24 hours from the fault occurrence time; the initiation of each repair order triggers an automatic extraction of the fault condition.
Further, the diagnostic module further comprises: a preprocessing module: the training set is used for preprocessing the acquired data to generate a training set; an algorithm module: the method is used for constructing an artificial intelligence algorithm model by adopting a long-term and short-term memory network; an execution module: for performing model training and parameter adjustments.
Further, the alarm module further comprises: inputting data acquired by a sensor into a diagnostor, if the predicted value of the fault development stage is larger than 0.5, sending an alarm message as a fault, and calculating the corresponding fault type from the output vector output of the diagnostor; and the alarm message is sent to a manager through the building operation and maintenance client or the mobile phone.
Further, the air conditioning unit fault type diagnosis system based on big data further comprises that a loss function in the long and short term memory network is taken as an average Absolute Error (MAE), namely an average value of distances between a model predicted value and a sample true value, and is defined as follows:
According to the scheme, a fine air conditioning unit sensor Internet of things is built according to the characteristics of the air conditioning unit, an intelligent model for diagnosing the specific reasons of the fault is provided on the basis of the fine air conditioning unit sensor Internet of things, and the automatic field engineering application can be supported.
LSTM networks have the ability to analyze the time-series characteristics of monitored data, both to diagnose the specific cause of a fault and to assess the risk of a fault before it has not occurred. The accuracy of the LSTM model with the specific time window, which is preferred by the application, on the fault prediction can reach 80%, and other models have the problem that the fault type prediction result greatly fluctuates. Therefore, compared with the prior art, the method and the device have obvious advantages in the aspects of accuracy and fault diagnosis stability. Based on the characteristics of the LSTM network, the method and the device are creatively applied to the field of fault type diagnosis of the air conditioning unit with big data, so that the fault diagnosis effect of the air conditioning unit is obviously improved.
Practical engineering application shows that the characteristics extracted by the scheme of the application, particularly the modeling of the environmental characteristics, can reflect the running condition of the unit more comprehensively. This is a new building equipment analysis model. The invention solves the problem of fault type diagnosis of the air conditioning unit and improves the level of maintenance, refinement and intellectualization of important equipment in the building.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood and to implement them in accordance with the contents of the description, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention.
In the drawings:
figure 1 shows a schematic flow diagram of the method of the present invention.
Figure 2 shows another method flow diagram of the present invention.
Fig. 3 is a schematic diagram of the point location of the internet of things monitored by the air conditioning unit.
Fig. 4 is a schematic diagram of a data warehouse structure of monitoring data.
Fig. 5 is a schematic diagram of a part of repair order of the air conditioning unit.
Fig. 6 is a schematic diagram of a fault risk curve type.
Fig. 7 is a schematic diagram of the extraction range of a fault condition.
FIG. 8 is a diagram illustrating the training results of the diagnostic device.
Fig. 9 is a schematic diagram of mobile phone end fault alarming and processing.
Fig. 10 is a schematic diagram of a fault diagnosis system.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the specific embodiments of the present invention and the accompanying drawings. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention relates to a method and a device for diagnosing fault types of an air conditioning unit.
The process of the present invention will be described first with reference to FIG. 1. As shown in fig. 1, a big data-based fault type diagnosis method for an air conditioning unit is provided, which is characterized by comprising the following steps: step 101, collecting big data of the running state of an air conditioning unit; step 102, collecting real-time electricity consumption data, calculating energy consumption, and collecting real-time weather information; 103, determining a fault risk curve for a repair system, and automatically extracting a fault situation; step 104, establishing an artificial intelligence fault type diagnosis model; and 105, diagnosing the fault type of the air conditioner unit.
Next, another flow of the present invention is described with reference to FIG. 2, and the following is described:
step 1: the data of the internet of things are monitored in a docking site, sensors are arranged in all air conditioning fresh air handling units in a building, and the types of the sensors which are available at present comprise: the method comprises the steps of cold and hot water valve opening feedback, air supply temperature, temperature setting, cold and hot water valve forced value, fresh air valve mode, fresh air valve state, manual and automatic state, running state, fault alarm, primary effect filter screen alarm, intermediate effect filter screen alarm, fresh air valve opening state, fresh air valve closing state, winter and summer switching, cold and hot water valve mode, system starting and stopping and the like, and the big data of the running state of the air conditioning unit is collected. The return frequency of all data interfaces should be high, for example, in this embodiment, the network transmission interval is preferably 1 minute mostly, which ensures real-time monitoring of the unit operation state. The point locations for monitoring the internet of things are preferably as shown in fig. 3, which also gives an example of data for each sensor.
In the step, because the number of the sensors is large and the acquisition frequency is high, the accumulated data volume is extremely large, and the traditional database can not be used for storage. The star-shaped data warehouse which is more suitable for an analysis algorithm is a storage and extraction medium of big data of the Internet of things, and the structure of the star-shaped data warehouse is shown in FIG. 4.
Step 2: and (3) butting the intelligent electric meters, adding a special monitoring loop for the air conditioning unit in the building electricity loop, and acquiring real-time electricity consumption data of the corresponding air conditioning unit by using the digital electric meter. Because the monitoring value of the power consumption is the state quantity, the return frequency can be properly reduced compared with the sensor, and 5-minute intervals are preferably selected in the case, so that the accuracy and the fault tolerance of the power consumption data are ensured. The method can acquire real-time weather data while acquiring real-time power consumption data.
And step 3: establishing an energy consumption data model of the equipment based on the real-time power consumption data in the step 2, associating each unit to a corresponding energy consumption loop, and extracting the energy consumption standard of the air conditioning unit per hour according to the corresponding relationChange value ei. For example, the hourly power consumption of a loop of a unit is {55.14,51.05, … …,53.27} in 24 hours, and the average is 52.28, then e1＝55.14/52.28＝1.055，e2＝51.05/52.28＝0.976，……,e24＝5 3.27/52.28＝1.019。
And 4, step 4: establishing a warranty system, connecting the existing repair server or building the software system, and collecting the fault repair record related to the air conditioning unit in the system as a basis for determining the fault time T and the type of the air conditioning unit, as shown in fig. 5. Specifically, a fault category K corresponding to the repair work order and a development rule thereof are determined through a keyword matching method, and then a fault risk curve P is determined as { P ═ Pi}. For example, according to the system reliability theory, bearing damage belongs to a medium-occurrence fault, and the section of the risk curve rises quickly, as shown in fig. 6 (a); the dirty blockage of the condensate pipe belongs to the late-onset failure, and the risk curve of the dirty blockage is accelerated to rise, as shown in fig. 6 (b); while drive belt loosening is an early failure, the risk curve shows a gradual upward trend, as shown in fig. 6 (c). And automatically extracting the fault condition of the air conditioning unit. The extracted fault situation comes from the repair system and is a real, influencing fault recognized by the logistical personnel, rather than using the machine's own alarm sensors. The definition of "failure case" herein is: a fault situation is a set of parameters that contains various parameters associated with a fault in a 24 hour range that is pushed forward by the time the fault occurred. As shown in fig. 7, before and after the occurrence of the fault, the fault can be divided into 4 stages, namely, a normal use stage before the fault, a development stage, a maintenance stage, and a normal use stage after the fault. The fault situation extraction is only for the development stage, and the risk curve has a value only at this stage. Specifically, the fault condition includes an air conditioning unit name D, a fault type K, and a fault risk P ═ PiTime of failure T, weather data W ═ WiAnd monitoring data set S ═ S of Internet of thingsijThe normalized value of energy consumption E ═ Ei}. Wherein S is obtained in step 1, E is obtained by calculation in step 2, { K, P, T } is obtained in step 4, and W is obtained by inquiring weather data such as local temperature, humidity and the like in real time.
And 5, automatically extracting the fault condition of the air conditioning unit. For the initiation of the repair work order of each unit, an automatic extraction process of the fault condition is triggered once. The time window within each fault condition should be 30 minutes or 60 minutes, depending on the characteristics of the equipment, with a corresponding value of i of 1-48 or 1-24.
Besides the corresponding situations at all the fault moments, a certain proportion of fault-free situations should be set as training data when no fault exists. Considering that the time of a unit without faults accounts for about 90%, the selected number is more than the number in the case of faults. The ratio of the presence or absence of failure in this case was 1: 1.8. The acquisition time for a fault-free situation is randomly chosen.
Step 6: and establishing an artificial intelligence fault type diagnosis model, and then executing the training of the model. The method can be divided into the following three steps:
step 6.1: and generating a training set. And preprocessing the sensor big data. Firstly, inputting a variable input, performing different calculations on different variable types, and normalizing numerical variables such as a cold and hot water valve forced value, temperature setting, air supply temperature, energy consumption value, air temperature and humidity by using min-max according to the following formula (1); some numerical variables with clear upper and lower limits, such as the forced value of the cold and hot water valve and the air humidity, are normalized according to the formula (2). And (4) converting all the state value variables and the classification variables into one-hot vectors, wherein the length of the vectors is the total number of the states, and the specific method is calculated according to the formula (3). And splicing all the variables in the same dimension, and continuously splicing all the vectors along the length direction to form a one-dimensional input vector. For example, a numeric variable occupies the 1 st to 8 th elements of an input vector, a run state has 2 states, thus occupying the 9 th and 10 th elements of an input vector, and so on.
Then the output variable output is nt+1Dimension vector of where ntThe total number of fault types is determined according to the following method: (1) if the current fault is of the xth type, the xth element of output is equal to 1; (2) the remaining 1 dimension is the fault risk curve value pi at the current moment; (3) the remaining elements of output are 0. For the air conditioning unit fault, about 20 types of fault types are determined by the industry, and after being screened by field experts, 10 types of fault types can be represented by the sensor state, so that the types can be used for fault diagnosis, and the specific details are shown in the following table. There is also a "no fault" type, so n is determined heretThe total dimension of output is 12, 11.
TABLE 1 diagnosable failure types
Step 6.2: an artificial intelligence algorithm model is constructed, and an algorithm of an air conditioning unit fault type diagnostor is established, which is hereinafter referred to as a diagnostor. The main artificial intelligence algorithm is Long Short-Term Memory network (LSTM), which is a modified Recurrent Neural Network (RNN), all RNNs have repeated network modules, and can accept a sequence of tandem associations as input to predict the future behavior of the sequence. However, in the standard RNN, this duplicate network module is of a very simple structure, e.g. only contains one tanh layer, which results in its fitting capability being limited, and the output value can be affected by all previous inputs, i.e. long term dependence problems, which is not reasonable. And the LSTM divides the input into long-term influence and short-term influence, so that the influence of more recent data on the result is larger, more distant data can be selectively forgotten, the characteristics of time sequence data are more met, and the LSTM can be used for fitting the situation that the unit fails.
The network model herein includes one each of an input layer, an output layer, and a fully connected layer, where the hidden element of the fully connected layer is taken to be 32. The Optimizer is Adam Optimizer. The loss function is taken as the Mean Absolute Error (MAE), i.e. the average of the distance between the predicted value of the model and the true value of the sample, and is defined as follows:
The remaining parameters, which include activation functions, forgetting rates, random silence rates, etc., are taken to be common values used in the industry when modeling.
Step 6.3: model training and parameter tuning are performed. A total of 512 samples are trained per training period (epoch), with 183 being fault conditions. The number of training passes determines the termination condition by observing the change in the value of the loss function. In total 2000 epochs were trained in cases, and the next batch was trained using the minimatch training method, i.e., calculating the loss and update weights after every 20 samples were trained. The accuracy achievable by the model is high, and the results of the validation set are shown in fig. 8, which shows that the development of the fault can be basically identified. The final fitness of the training sample is 98.8%, and the accuracy of fault type diagnosis on test data is about 95%.
And 7: and (5) diagnosing and predicting, and reporting the result to an operation and maintenance management department for processing. Specifically, sensor data of each unit is input into a diagnostor in real time, if the predicted value of the fault development stage is larger than 0.5, the diagnostor is used as a fault to send out an alarm, and meanwhile, the corresponding fault type is calculated from the output vector output of the diagnostor. Then, the alarm message is pushed to the manager through the building operation and maintenance client and the mobile phone, and an abnormal alarm processing flow is entered, as shown in fig. 9.
Meanwhile, the data of the monitoring sensor of the air conditioning unit is continuously converted and integrated in a data warehouse, and if a new air conditioning maintenance work order exists, the fault situation is automatically extracted and added into the existing training set. The analysis engine is set to retrain a new diagnotor every 15 days to increase its accuracy.
Aiming at the characteristics of the air conditioning unit, a fine air conditioning unit sensor Internet of things is built, an intelligent model for diagnosing the specific reasons and the fault probability of the fault is provided on the basis, and the automatic field engineering application can be supported. The LSTM network has the capability of analyzing the time sequence characteristics of monitoring data, not only can diagnose the specific reasons of fault occurrence, but also can evaluate the fault risk before the fault does not occur, and the accuracy and the fault diagnosis stability are obviously improved. In addition, the characteristics extracted by the technical scheme, especially the modeling of the environmental characteristics, can comprehensively reflect the running condition of the unit, and improve the maintenance, refinement and intelligentization level of important equipment in the building
The modular structure of the system of the method is shown in fig. 10 and described as follows:
module 1: and the state monitoring module comprises a unit monitoring sensor network. The sensors are arranged in all air conditioning fresh air handling units in the building, and the types of the sensors to be arranged include but are not limited to: the method comprises the steps of cold and hot water valve opening feedback, air supply temperature, temperature setting, cold and hot water valve forced value, fresh air valve mode, fresh air valve state, manual and automatic state, running state, fault alarm, primary effect filter screen alarm, intermediate effect filter screen alarm, fresh air valve opening state, fresh air valve closing state, winter and summer switching, cold and hot water valve mode, system starting and stopping and the like, and the big data of the running state of the air conditioning unit is collected. The return frequency of all data interfaces should be high, for example, in this embodiment, the network transmission interval is mostly 1 minute, which ensures real-time monitoring of the unit running state. The point of monitoring the internet of things is shown in fig. 3, and a data example of each sensor is also given in the figure.
In the module, because the number of the sensors is large and the acquisition frequency is high, the accumulated data volume is extremely large, and the traditional database can not be used for storage. A star-shaped data warehouse which is more suitable for an analysis algorithm is used as a storage and extraction medium of big data of the Internet of things.
And (3) module 2: and the data acquisition module is used for monitoring the power consumption of the on-site air conditioning unit. In the special electricity consumption monitoring loop of the air conditioning unit for building installation and construction, a digital ammeter is used for acquiring real-time electricity consumption data of the corresponding air conditioning unit. Because the monitoring value of the power consumption is the state quantity, the return frequency can be properly increased compared with the sensor, and the accuracy and the fault-tolerant capability of the power consumption data are ensured. The data acquisition module can also be used for acquiring real-time weather data, and a module is not shown in the figure.
And a module 3: and the repair reporting module is used for butting the existing repair reporting server or building the software system, collecting fault repair records related to the air conditioning unit in the system and taking the fault repair records as a basis for determining the fault time T and the type of the air conditioning unit. Specifically, the fault category K corresponding to the repair work order and the development rule thereof are determined by a keyword matching method, and then the fault risk curve P is determined as { pi }. For example, according to the system reliability theory, the transmission belt loosens to be an early fault, and the risk curve shows that the rising trend is gradually slowed down; the dirty blockage of the condensate pipe belongs to the late-onset fault, the risk curve of the dirty blockage is accelerated to rise, and the like.
And (4) module: and the automatic fault extraction module is used for integrating and processing the big data. And continuously performing conversion integration on the monitoring sensor data of the air conditioning unit in a data warehouse, and if a new air conditioning maintenance work order exists, automatically extracting a fault situation and adding the fault situation into the existing training set. The analysis engine is set to retrain a new diagnotor every 15 days to increase its accuracy.
And a module 5: and the diagnostic module is an air conditioning unit fault type diagnostic device. The module is a core software module, establishes an artificial intelligence fault diagnosis model and then executes network training. Here, the following three modules can be divided:
module 5.1: and a preprocessing module. And generating a training set. And preprocessing the sensor big data. Firstly, inputting variable input, and for different variable types, some numerical variables are normalized by min-max according to the following formula (1), such as air supply temperature and energy consumption value, and partial numerical variables with definite upper and lower limits, such as cold and hot water valve forced value and air humidity, are normalized according to the formula (2). The state value variables and the classification variables are all converted into one-hot vectors, the length of the vectors is the total number of states, and the specific method is the formula (3). And continuously splicing all the vectors along the length direction to form a one-dimensional input vector. For example, a numeric variable occupies the 1 st to 8 th elements of an input vector, a run state has 2 states, thus occupying the 9 th and 10 th elements of an input vector, and so on.
Then, an output variable output is an nt + 1-dimensional vector, wherein nt is the total number of fault types and is determined according to the following method: (1) if the current fault is of the xth type, the xth element of output is equal to 1; (2) the remaining 1 dimension is the fault risk curve value p at the current timei(ii) a (3) The remaining elements of output are 0. For the air conditioning unit fault, about 20 types of fault types are determined by the industry, and after being screened by field experts, 10 types of fault types can be represented by the sensor state, so that the types can be used for fault diagnosis. There is also a "no fault" type, so n is determined heretThe total dimension of output is 12, 11.
Module 5.2: and (5) an algorithm module. An artificial intelligence algorithm model is constructed, and an air conditioning unit fault type diagnostor is established, which is hereinafter referred to as a diagnostor. The main artificial intelligence algorithm is a Long Short-Term Memory network (LSTM), and the model is selected because of the analysis requirement of time series and can also solve the Long-Term dependence problem of the general recurrent neural network. The network model herein includes one each of an input layer, an output layer, and a fully connected layer, where the hidden element of the fully connected layer is taken to be 32. The Optimizer is Adam Optimizer. The loss function is taken as the Mean Absolute Error (MAE), i.e. the average of the distance between the predicted value of the model and the true value of the sample. The remaining parameters, which include activation functions, forgetting rates, random silence rates, etc., are taken to be common values used in the industry when modeling.
Module 5.3: and executing the module. Model training and parameter tuning are performed. A total of 512 samples are trained per training period (epoch), with 183 being fault conditions. The number of training passes determines the termination condition by observing the change in the value of the loss function. In total 2000 epochs were trained in cases, and the next batch was trained using the minimatch training method, i.e., calculating the loss and update weights after every 20 samples were trained. The accuracy achievable by the model is high, and the results of the validation set are shown in fig. 8, which shows that the development of the fault can be basically identified. The fitting degree of the final training sample is 98.8%, the fault type diagnosis accuracy rate on the test data is about 95%, and the average accuracy rate of the fault probability is about 87%.
And a module 6: and the alarm module is connected with the fault diagnosis application module. And diagnosing the fault type, and reporting the result to an operation and maintenance management department for processing. Specifically, sensor data of each unit is input into a diagnostor in real time, if the predicted value of the obtained fault stage is larger than 0.5, the diagnostor is used as a fault to send out an alarm, and meanwhile, the corresponding fault type of the diagnostor is calculated from the output vector output of the diagnostor. Then, the alarm message is pushed to a manager through the building operation and maintenance client and the mobile phone, and an abnormal alarm processing flow is entered, as shown in fig. 9.
Aiming at the characteristics of the air conditioning unit, the invention builds a fine air conditioning unit sensor Internet of things, provides an intelligent model for diagnosing the specific reasons and the fault probability of the fault on the basis of the fine air conditioning unit sensor Internet of things, and can support the automatic field engineering application. LSTM networks have the ability to analyze the time-series characteristics of monitored data, both to diagnose the specific cause of a fault and to assess the risk of a fault before it has not occurred. Compared with the prior art, the method has the advantages of accuracy and fault diagnosis stability. In addition, the characteristics extracted by the technical scheme, particularly the modeling of the environmental characteristics, can reflect the running condition of the unit more comprehensively. The invention solves the problem of fault type diagnosis of the air conditioning unit and improves the level of maintenance, refinement and intellectualization of important equipment in the building.
According to the novel debugging unit fault type diagnosis method and system based on the big data, provided by the invention, aiming at the characteristics of the air conditioning unit, a fine air conditioning unit sensor Internet of things is built, and a big data technology is adopted to store and operate massive monitoring data. The core algorithm solves the problem of complex relation among various monitoring data, and can give specific reasons for faults by considering the time sequence characteristics of the monitoring data. The invention solves the problems of low automation degree, incomplete data and unreliable results in the aspect of fault diagnosis of the air conditioning unit.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and any simple modification, equivalent change and modification made to the above embodiment according to the technical spirit of the present invention are still within the scope of the technical solution of the present invention.
Claims (9)
1. A big data-based air conditioning unit fault type diagnosis method is characterized by comprising the following steps:
101, establishing a fine air conditioning unit sensor Internet of things, and butting large running state data of an air conditioning unit;
step 102, adding a monitoring loop special for an air conditioning unit in a building power utilization loop, acquiring real-time power consumption data, calculating energy consumption, and acquiring real-time weather information;
103, determining a fault risk curve for a repair system, and automatically extracting a fault situation; the step 103 further comprises: the method comprises the following steps that the repair reporting system is connected with an existing repair reporting server or a newly-built software system, collects fault repair reporting records related to an air conditioning unit and serves as a basis for determining the fault time T and the type of the air conditioning unit, and determines the fault type K and the development rule thereof corresponding to a repair reporting work order through a keyword matching method, so that a fault risk curve P is determined as { pi };
104, constructing an artificial intelligence algorithm model, establishing an artificial intelligence fault type diagnosis model, and an intelligent model which can diagnose the specific reasons of the fault and the fault probability and can support the automatic field engineering application; wherein, the main artificial intelligence algorithm is LSTM; the step 104 further comprises: step 401: preprocessing the acquired data to generate a training set; step 402: adopting a long-short term memory network to construct an artificial intelligence algorithm model; step 403: performing model training and parameter adjustment;
and 105, diagnosing the fault type of the air conditioner unit.
2. The method of claim 1, wherein the step 101 further comprises:
the running state big data of the air conditioning unit is collected by butting the sensors arranged in the air conditioning fresh air unit.
3. The method of claim 1, wherein the step 102 further comprises:
the method comprises the steps of butt joint of an intelligent ammeter and collection of real-time electricity consumption data of the air conditioning unit;
establishing an energy consumption data model of the equipment, and extracting an energy consumption standardized value;
and collecting real-time weather information.
4. The method of claim 1, wherein in step 103:
the automatic extraction of the fault situation comes from a repair reporting system, is a fault which is really generated, influenced and identified by logistics personnel, and does not adopt data of an alarm sensor of the machine;
the fault situation is a set of a series of parameters, and comprises various parameters related to the fault within a range of forward pushing 24 hours from the fault occurrence time;
the initiation of each repair order triggers an automatic extraction of the fault condition.
5. The method of claim 1, wherein said step 105 further comprises:
inputting data acquired by a sensor into a diagnostor, if the predicted value of the fault development stage is larger than 0.5, sending an alarm message as a fault, and calculating the corresponding fault type from the output vector output of the diagnostor;
and the alarm message is sent to a manager through the building operation and maintenance client or the mobile phone.
6. The method of claim 1,
the loss function in the long-short term memory network is taken as the average absolute error MAE, namely the average value of the distance between the model predicted value and the sample true value, and is defined as the following formula:
7. An air conditioning unit fault type diagnostic system based on big data is characterized by comprising:
the state monitoring module is used for building a fine air conditioning unit sensor Internet of things and butting the big data of the running state of the air conditioning unit;
the data acquisition module is used for adding a monitoring loop special for the air conditioning unit in the building power utilization loop, acquiring real-time power consumption data, calculating energy consumption and acquiring real-time weather information;
the repair reporting module is used for determining a fault risk curve and automatically extracting a fault situation for the repair reporting system; the repair module further comprises: the method comprises the following steps that the repair reporting system is connected with an existing repair reporting server or a newly-built software system, collects fault repair reporting records related to an air conditioning unit and serves as a basis for determining the fault time T and the type of the air conditioning unit, and determines the fault type K and the development rule thereof corresponding to a repair work order through a keyword matching method, so that a fault risk curve P is determined as { pi }; the extracted fault situation comes from a repair reporting system, is a fault which really occurs, generates influence and is identified by logistics personnel, and does not adopt data of an alarm sensor of the machine;
the diagnostic module is used for constructing an artificial intelligence algorithm model, establishing an artificial intelligence fault type diagnosis model, diagnosing the specific reasons of the fault and the intelligent model of the fault probability, and supporting the automatic field engineering application; wherein, the main artificial intelligence algorithm is LSTM; the diagnostic module further comprises: a preprocessing module: the training set is used for preprocessing the acquired data to generate a training set; an algorithm module: the method is used for constructing an artificial intelligence algorithm model by adopting a long-term and short-term memory network; an execution module: for performing model training and parameter adjustments;
and the alarm module diagnoses the fault and reports the result to the operation and maintenance management department.
8. The system of claim 7, wherein the condition monitoring module further comprises:
the method comprises the steps that sensors are arranged in a butt joint air conditioning fresh air handling unit, and big data of the running state of the air conditioning fresh air handling unit are collected;
and/or, the data acquisition module further comprises:
the method comprises the steps of butt joint of an intelligent ammeter and collection of real-time electricity consumption data of the air conditioning unit;
establishing an energy consumption data model of the equipment, and extracting an energy consumption standardized value;
collecting real-time weather information;
and/or, in the repair reporting module:
the method comprises the following steps that the repair reporting system is connected with an existing repair reporting server or a newly-built software system, collects fault repair reporting records related to an air conditioning unit and serves as a basis for determining the fault time T and the type of the air conditioning unit, and determines the fault type K and the development rule thereof corresponding to a repair work order through a keyword matching method, so that a fault risk curve P is determined as { pi }; the extracted fault situation comes from a repair reporting system, is a fault which really occurs, generates influence and is identified by logistics personnel, and does not adopt data of an alarm sensor of the machine;
the fault situation is a set of a series of parameters, and comprises various parameters related to the fault within a range of forward pushing 24 hours from the fault occurrence time;
the initiation of each repair work order triggers the automatic extraction of the fault situation once;
and/or, the alarm module further comprises:
inputting data acquired by a sensor into a diagnostor, if the predicted value of the fault development stage is larger than 0.5, sending an alarm message as a fault, and calculating the corresponding fault type from the output vector output of the diagnostor;
and the alarm message is sent to a manager through the building operation and maintenance client or the mobile phone.
9. The system of claim 7, wherein in the algorithm module: the loss function in the long-short term memory network is taken as the average absolute error MAE, namely the average value of the distance between the model predicted value and the sample true value, and is defined as the following formula:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010225020.4A CN111412579B (en) | 2020-03-26 | 2020-03-26 | Air conditioning unit fault type diagnosis method and system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010225020.4A CN111412579B (en) | 2020-03-26 | 2020-03-26 | Air conditioning unit fault type diagnosis method and system based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111412579A CN111412579A (en) | 2020-07-14 |
| CN111412579B true CN111412579B (en) | 2022-03-08 |
Family
ID=71491433
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010225020.4A Active CN111412579B (en) | 2020-03-26 | 2020-03-26 | Air conditioning unit fault type diagnosis method and system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111412579B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112283876A (en) * | 2020-10-30 | 2021-01-29 | 青岛海尔空调电子有限公司 | Air conditioner fault prediction method and air conditioner |
| CN113446716B (en) * | 2021-07-05 | 2022-09-16 | 成都地铁运营有限公司 | Self-diagnosis method and system for rail transit ventilation air conditioner mode |
| CN113834183B (en) * | 2021-08-18 | 2023-03-21 | 青岛海尔空调器有限总公司 | Control method and device for air conditioner and server |
| CN114135991B (en) * | 2021-12-03 | 2023-03-31 | 中国海洋大学 | Temperature preset control and equipment early warning method for subway station public area |
| CN114967796A (en) * | 2022-04-13 | 2022-08-30 | 安徽领电智能科技有限公司 | Intelligent building equipment management system |
| CN115899964A (en) * | 2022-12-22 | 2023-04-04 | 北京航天智造科技发展有限公司 | Multidimensional-based intelligent air conditioner monitoring method and system |
| CN116384980B (en) * | 2023-05-25 | 2023-08-25 | 杭州青橄榄网络技术有限公司 | Repair reporting method and system |
| CN116678072B (en) * | 2023-06-05 | 2024-05-10 | 耀昶嵘相变材料科技(广东)有限公司 | Fault processing method and terminal of central air conditioning system and central air conditioning system |
| CN117073129B (en) * | 2023-10-17 | 2024-01-26 | 山东意林建筑规划设计研究院有限公司 | Fault diagnosis method and system for heating ventilation air conditioning system |
| CN117993894B (en) * | 2024-04-03 | 2024-06-14 | 四川航天职业技术学院(四川航天高级技工学校) | Rail transit operation and maintenance state data processing method and system based on artificial intelligence |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0287992A (en) * | 1988-09-22 | 1990-03-28 | Mitsubishi Electric Corp | Inverter for air-conditioner |
| JPH07324799A (en) * | 1994-06-01 | 1995-12-12 | Matsushita Seiko Co Ltd | Abnormal content confirming device of air conditioner |
| CN1443980A (en) * | 2003-04-17 | 2003-09-24 | 上海交通大学 | Central air-conditioner monitoring and fault diagnosis system based on network technique |
| JP2007249684A (en) * | 2006-03-16 | 2007-09-27 | Ricoh Co Ltd | Facility management support system, facility management support method, and program |
| CN103616863A (en) * | 2013-11-22 | 2014-03-05 | 青岛海尔软件有限公司 | Risk forecast method for household air conditioner and household refrigerator |
| DE112013003702T5 (en) * | 2012-07-27 | 2015-04-23 | Myine Electronics, Inc. | A system and method for using a personal electronic device to wirelessly connect a remote diagnostic site to a home appliance for troubleshooting purposes |
| CN104748327A (en) * | 2015-03-23 | 2015-07-01 | 同济大学 | Public building air conditioner energy efficiency diagnosis system based on power consumption data |
| CN109229034A (en) * | 2017-07-11 | 2019-01-18 | 现代自动车株式会社 | Integral connection tube manages method and its online vehicles |
| CN109460873A (en) * | 2018-11-14 | 2019-03-12 | 北京未来科学城科技发展有限公司 | Air-conditioning system running optimizatin method and apparatus |
| CN109902153A (en) * | 2019-04-02 | 2019-06-18 | 杭州安脉盛智能技术有限公司 | Equipment fault diagnosis method and system based on natural language processing and reasoning by cases |
| CN110119851A (en) * | 2019-05-23 | 2019-08-13 | 上海建工四建集团有限公司 | A kind of building Mechatronic Systems intelligent fault prediction technique and system |
-
2020
- 2020-03-26 CN CN202010225020.4A patent/CN111412579B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0287992A (en) * | 1988-09-22 | 1990-03-28 | Mitsubishi Electric Corp | Inverter for air-conditioner |
| JPH07324799A (en) * | 1994-06-01 | 1995-12-12 | Matsushita Seiko Co Ltd | Abnormal content confirming device of air conditioner |
| CN1443980A (en) * | 2003-04-17 | 2003-09-24 | 上海交通大学 | Central air-conditioner monitoring and fault diagnosis system based on network technique |
| JP2007249684A (en) * | 2006-03-16 | 2007-09-27 | Ricoh Co Ltd | Facility management support system, facility management support method, and program |
| DE112013003702T5 (en) * | 2012-07-27 | 2015-04-23 | Myine Electronics, Inc. | A system and method for using a personal electronic device to wirelessly connect a remote diagnostic site to a home appliance for troubleshooting purposes |
| CN103616863A (en) * | 2013-11-22 | 2014-03-05 | 青岛海尔软件有限公司 | Risk forecast method for household air conditioner and household refrigerator |
| CN104748327A (en) * | 2015-03-23 | 2015-07-01 | 同济大学 | Public building air conditioner energy efficiency diagnosis system based on power consumption data |
| CN109229034A (en) * | 2017-07-11 | 2019-01-18 | 现代自动车株式会社 | Integral connection tube manages method and its online vehicles |
| CN109460873A (en) * | 2018-11-14 | 2019-03-12 | 北京未来科学城科技发展有限公司 | Air-conditioning system running optimizatin method and apparatus |
| CN109902153A (en) * | 2019-04-02 | 2019-06-18 | 杭州安脉盛智能技术有限公司 | Equipment fault diagnosis method and system based on natural language processing and reasoning by cases |
| CN110119851A (en) * | 2019-05-23 | 2019-08-13 | 上海建工四建集团有限公司 | A kind of building Mechatronic Systems intelligent fault prediction technique and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111412579A (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111412579B (en) | Air conditioning unit fault type diagnosis method and system based on big data | |
| CN109146093B (en) | Power equipment field investigation method based on learning | |
| CN109240244B (en) | Data-driven equipment running state health degree analysis method and system | |
| CN108199795B (en) | A kind of monitoring method and device of equipment state | |
| CN110320892A (en) | The sewage disposal device fault diagnosis system and method returned based on Lasso | |
| CN109766583A (en) | Based on no label, unbalanced, initial value uncertain data aero-engine service life prediction technique | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| RU2626780C1 (en) | Method and system of remote monitoring energy installations | |
| CN109858140B (en) | Fault diagnosis method for water chilling unit based on information entropy discrete Bayesian network | |
| CN114047719A (en) | Remote monitoring and evaluating system and operation method for rural domestic sewage treatment facility | |
| CN112529320A (en) | Intelligent diagnosis system for air compressor cluster | |
| Qiu et al. | Data mining based framework to identify rule based operation strategies for buildings with power metering system | |
| CN113036913B (en) | Method and device for monitoring state of comprehensive energy equipment | |
| CN111669123B (en) | Method and device for fault diagnosis of photovoltaic string | |
| CN110119851A (en) | A kind of building Mechatronic Systems intelligent fault prediction technique and system | |
| CN111581597A (en) | Wind turbine generator gearbox bearing temperature state monitoring method based on self-organizing kernel regression model | |
| CN110764474B (en) | Method and system for monitoring running state of equipment | |
| CN116224137A (en) | UPS equipment fault early warning method and system | |
| CN111103137A (en) | Wind turbine gearbox fault diagnosis method based on deep neural network | |
| CN114266197A (en) | Method for diagnosing equipment fault of hydropower station | |
| CN110033102A (en) | A kind of huge hydroelectric power plant has the intelligent diagnosing method and expert system of learning functionality | |
| CN115827411A (en) | Online monitoring and operation and maintenance evaluation system and method for automation equipment | |
| CN118051827A (en) | Power grid fault prediction method based on deep learning | |
| CN115859791A (en) | Method and device for estimating residual life of aero-engine | |
| CN117113157B (en) | Platform district power consumption fault detection system based on artificial intelligence |
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 |