CN117948744B - Electronic expansion valve health management system based on cloud interconnection - Google Patents
Electronic expansion valve health management system based on cloud interconnection Download PDFInfo
- Publication number
- CN117948744B CN117948744B CN202410226818.9A CN202410226818A CN117948744B CN 117948744 B CN117948744 B CN 117948744B CN 202410226818 A CN202410226818 A CN 202410226818A CN 117948744 B CN117948744 B CN 117948744B
- Authority
- CN
- China
- Prior art keywords
- data
- module
- crayfish
- expansion valve
- electronic expansion
- 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
- 230000036541 health Effects 0.000 title claims abstract description 58
- 241000238017 Astacoidea Species 0.000 claims abstract description 67
- 238000003745 diagnosis Methods 0.000 claims abstract description 49
- 238000005457 optimization Methods 0.000 claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000012937 correction Methods 0.000 claims abstract description 10
- 238000013508 migration Methods 0.000 claims abstract description 10
- 230000005012 migration Effects 0.000 claims abstract description 10
- 238000004891 communication Methods 0.000 claims abstract description 8
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 7
- 230000002159 abnormal effect Effects 0.000 claims abstract description 5
- 230000004927 fusion Effects 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 32
- 238000007726 management method Methods 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 20
- 238000013523 data management Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 15
- 238000010257 thawing Methods 0.000 claims description 11
- 229920006395 saturated elastomer Polymers 0.000 claims description 10
- 238000011161 development Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 7
- 230000002431 foraging effect Effects 0.000 claims description 6
- 238000013526 transfer learning Methods 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000013256 Gubra-Amylin NASH model Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 230000006872 improvement Effects 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 230000007257 malfunction Effects 0.000 claims 1
- 230000006855 networking Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005057 refrigeration Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 239000003507 refrigerant Substances 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B49/00—Arrangement or mounting of control or safety devices
- F25B49/02—Arrangement or mounting of control or safety devices for compression type machines, plants or systems
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60H—ARRANGEMENTS OF HEATING, COOLING, VENTILATING OR OTHER AIR-TREATING DEVICES SPECIALLY ADAPTED FOR PASSENGER OR GOODS SPACES OF VEHICLES
- B60H1/00—Heating, cooling or ventilating [HVAC] devices
- B60H1/00485—Valves for air-conditioning devices, e.g. thermostatic valves
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60H—ARRANGEMENTS OF HEATING, COOLING, VENTILATING OR OTHER AIR-TREATING DEVICES SPECIALLY ADAPTED FOR PASSENGER OR GOODS SPACES OF VEHICLES
- B60H1/00—Heating, cooling or ventilating [HVAC] devices
- B60H1/00642—Control systems or circuits; Control members or indication devices for heating, cooling or ventilating devices
- B60H1/00978—Control systems or circuits characterised by failure of detection or safety means; Diagnostic methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/0475—Generative networks
-
- 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
- G06N3/08—Learning methods
- G06N3/096—Transfer learning
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B2500/00—Problems to be solved
- F25B2500/06—Damage
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B2500/00—Problems to be solved
- F25B2500/19—Calculation of parameters
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B2500/00—Problems to be solved
- F25B2500/22—Preventing, detecting or repairing leaks of refrigeration fluids
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F25—REFRIGERATION OR COOLING; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS; MANUFACTURE OR STORAGE OF ICE; LIQUEFACTION SOLIDIFICATION OF GASES
- F25B—REFRIGERATION MACHINES, PLANTS OR SYSTEMS; COMBINED HEATING AND REFRIGERATION SYSTEMS; HEAT PUMP SYSTEMS
- F25B2600/00—Control issues
- F25B2600/25—Control of valves
- F25B2600/2513—Expansion valves
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Thermal Sciences (AREA)
- Mechanical Engineering (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses an electronic expansion valve health management system based on cloud network interconnection, which comprises a data acquisition module, a data correction module, a CPU module, a WIFI communication module, a central integrated processing cloud platform and a vehicle-mounted display module; the method comprises the steps of processing missing and abnormal data by using a 3D convolutional neural network according to acquired running state data of vehicle-mounted air conditioner electronic expansion valves of different users; then transmitting the corrected data to a central integrated processing cloud platform; the health diagnosis module in the cloud platform optimizes the parameters of the information generation countermeasure network model InfoGAN by utilizing a crayfish optimization algorithm, and outputs a preliminary fault diagnosis result of the vehicle-mounted air conditioner electronic expansion valve; and fusion federal migration learning is carried out, a global model is established, and the primary result is polymerized to form a final diagnosis result. The invention can realize health management of the electronic expansion valves of different users on the basis of protecting data safety and data privacy, and can also adjust the parameters of the diagnosis model in real time, thereby improving the diagnosis precision and reliability.
Description
Technical Field
The invention relates to an electronic expansion valve health management technology, in particular to an electronic expansion valve health management system based on cloud network interconnection.
Background
The electronic expansion valve is an important flow regulating device in the vehicle-mounted air conditioner. The method and the device realize accurate and rapid adjustment by sensing parameters of the refrigeration system and adjusting the opening according to a control algorithm so as to control the flow of the refrigerant. The function and performance of the electronic expansion valve are critical to the operation efficiency and refrigeration effect of the vehicle-mounted air conditioner. However, electronic expansion valves may suffer from various failures and performance degradation in the event of long-term environmental changes in the open air. For example: when the electronic expansion valve bears an excessive pressure difference and is not controlled in place, a leakage phenomenon can occur; when the valve of the electronic expansion valve receives interference pulse current, the limit spring is deformed, and spring sound is generated; when the electronic expansion valve is started or stopped, the rotor is out of step due to inertia; the mechanical part of the electronic expansion valve is blocked or the valve is in misoperation caused by wiring errors and other electrical faults.
Traditional vehicle-mounted electronic expansion valve health management methods rely mainly on periodic inspection and maintenance, as well as on empirical judgment to manage electronic expansion valves. This method has the following problems: firstly, relying on manual inspection, the workload is large and fine abnormal conditions are easy to ignore; secondly, the health state of the electronic expansion valve cannot be monitored and diagnosed in real time; accordingly, there is a need in the art for a more intelligent, efficient method of monitoring, diagnosing, and optimizing the health status of an electronic expansion valve.
Disclosure of Invention
The invention aims to: the invention provides an electronic expansion valve health management system based on cloud network interconnection, which is used for realizing intelligent processing of vehicle-mounted air conditioner electronic expansion valve data and accurate identification of fault types by combining a federal migration learning algorithm, a crayfish optimization algorithm and information to generate an countermeasure network model.
The technical scheme is as follows: the invention relates to an electronic expansion valve health management system based on cloud network interconnection, which comprises:
And a data acquisition module: detecting data representing the running state of an electronic expansion valve of the vehicle-mounted air conditioner;
And a data correction module: the missing value and the abnormal value in the data acquisition module are processed by using the 3D convolutional neural network CNN, and the processed data are transmitted to the CPU module;
CPU module: the data management and feature extraction module is used for receiving the data processed by the data correction module, uploading the data to the central integrated processing cloud platform by utilizing the WIFI communication module, and realizing data transmission with the vehicle-mounted display module;
WIFI communication module: the system is used for realizing data transmission between the CPU module and the central integrated processing cloud platform module;
Central integrated processing cloud platform: the system comprises a data management and feature extraction module, a health diagnosis module and a decision management module; the data management and feature extraction module stores the stream data corrected by the data correction module, extracts input features and stores features closely related to the health state of the vehicle-mounted expansion valve for the health diagnosis module to use; the health diagnosis module utilizes a crayfish optimization algorithm to optimize parameters of the information generation countermeasure network model InfoGAN and outputs a preliminary fault diagnosis result of the vehicle-mounted air conditioner electronic expansion valve; and fusion federal migration learning is carried out, a global model is established, and the primary result is polymerized to form a final diagnosis result; the decision management module stores the final diagnosis result of the health diagnosis module, classifies and decides the fault data of the electronic expansion valves of the vehicle-mounted air conditioners of different types, gives a health report, and feeds the health state back to the user CPU module;
And the vehicle-mounted display module: and receiving and displaying the diagnosis result of the health state of the expansion valve of the vehicle-mounted air conditioner received from the CPU module.
Further, the data of the electronic expansion valve operation state comprises an evaporator saturation temperature, a condenser saturation temperature, a defrosting temperature, a fan working current intensity and a compressor working frequency.
Further, the data management and feature extraction module comprises a plurality of data integration servers, each server represents different automobile brand manufacturers, and different servers store different user data to realize monitoring data sharing.
Furthermore, the health diagnosis module takes the extracted characteristics of the saturated temperature of the evaporator, the saturated temperature of the condenser, the defrosting temperature, the working current of the fan and the frequency data of the compressor as input to identify and diagnose the leakage, noise, motor step-out and valve misoperation of the electronic expansion valve.
Further, the health diagnosis module utilizes a crayfish optimization algorithm to generate the parameters of the countermeasure network model InfoGAN for optimizing the information, and the implementation process is as follows:
Optimizing important parameters of infoGAN, including the network structure, learning rate and regularization coefficient of the generator and the discriminator, wherein the important parameters of the infoGAN model correspond to the positions of the crayfish in the crayfish optimization algorithm; the positions are updated continuously through different behaviors of the crayfish, the algorithm iterates and loops continuously, diagnosis accuracy of expansion valve leakage, noise, motor step out and valve misoperation is used as an optimization target, the optimal position of each crayfish is obtained, and an optimal solution is output, namely an optimal network structure, a learning rate and a regular coefficient of an optimized infoGAN model;
firstly, initializing the position of each crayfish by a crayfish optimization algorithm, namely initializing the network structure, learning rate and regularization parameters of a generator and a discriminator of a infoGAN model, and then updating the position of the crayfish according to different behaviors, wherein the formula is as follows:
Xi,j＝lbj+(ubj-lbj)×rand
Wherein, individual i=1, 2, … M; dimension j=1, 2, … dim; x i,j is the position of individual i in the j dimension; lb j represents the lower bound of the j-th dimension variable; ub j represents the upper bound of the j-th dimension variable; rand is a random number between [0,1 ];
In the exploration stage, when T is more than 30 ℃, the position updating formula is as follows:
Wherein X β is a global optimal position obtained along with iteration; x η is the optimal position of the current population; x α is the position of entering the hole to avoid summer heat; wherein the temperature discrimination formula is: t=rand×15+20;
the phenomenon that crayfish fights for a cave occurs in the exploration stage, and the position updating formula is as follows:
Wherein, when the rand is less than 0.5, the phenomenon of competing for the cave does not occur; t is the current iteration number; t+1 is the next iteration number; The position of the individual i in the j dimension at the t+1st time of the iteration; The position of the individual i in the j dimension at the t-th time of iteration; c 2 is a falling curve C 2 =2- (T/T);
in the development stage, when T is more than 30 ℃ and rand is more than or equal to 0.5, the competition stage is entered, and the position updating formula is as follows:
Where z represents random individuals of crayfish, each crayfish competing with each other by the formula z=round (rand× (M-1)) +1, and X i of the crayfish makes a change based on the position of X z of the other crayfish;
In the development stage, when T is less than 30 ℃, the food is fed into the foraging stage, the size of the food is judged, and the definition formula is as follows:
Wherein Q is C 3 =3 is a food factor; f i is the fitness value of the ith crayfish; f food is food position fitness value; and the crayfish food size judgment is derived from the size of the largest food;
When (when) When the food is too big, the crayfish bites the food by the first paw, and the formula is:
When the food is smaller, the crayfish adopts the second paw foot and the third paw foot to eat alternately, and the process is simulated by using a sine function and a cosine function, and the formula is as follows:
wherein p is foraging intake and is defined as: Mu is the temperature most suitable for crayfish; sigma and C 1 are used to control crayfish intake at different temperatures;
When (when) When eating directly, the formula is as follows:
Through the temperature regulation exploration and development process, the global optimization capacity is effectively improved, the randomness is higher, the search precision can be better improved, and the error rate is reduced.
Further, the implementation process of the preliminary fault diagnosis result of the electronic expansion valve of the vehicle-mounted air conditioner is as follows:
Inputting the data acquired by the data management and feature extraction module into an optimized infoGAN model, acquiring the data of the saturated temperature of the evaporator, the saturated temperature of the condenser, the defrosting temperature, the working current of the fan and the frequency of the compressor after feature extraction, transmitting the data to a generator G, respectively transmitting the generated data of G to a discriminator D and a classifier C, re-extracting the data by the classifier C to acquire an output C and generate a result x, and classifying faults;
Firstly, a fixed characteristic c and a random variable e are obtained and transmitted to a generator, a sample G (e, c) is obtained, and an objective function is obtained as follows:
minGmaxDV1(D,G)＝V(D,G)-λI(c；G(e,c))
Wherein V 1 (D; G) is an objective function of the infoGAN model; v (D; G) is an objective function of the GAN model; lambda is a super parameter; in the function I (C; G (e, C)) there is an edge probability which is difficult to calculate, and an approximate distribution F 1 (G, Q) is added to replace the edge probability, so that mutual information between the generator G and the output C of the classifier C is represented, and the calculation formula is as follows:
minG,QmaxDV1(D,G,Q)＝V(D,G)-λF1(G；Q)
wherein E is the desired; x is model inputs e and c; p (c) is a probability distribution of c; h (c) is the information entropy of c; q (c|x) is the auxiliary network Q; p G(e,c) is the probability distribution of x.
Further, the fusion federal migration learning in the health diagnosis module is to cooperatively establish a global model by using a plurality of automobile users and an integrated server, which specifically comprises the following steps:
Through the participation of K automobile users in federation, all modeling tasks can be regarded as the optimization process of a loss function, the loss function is accumulated by diagnostic errors of diagnostic models corresponding to infoGAN of the automobile users, the loss function is represented by min (f (alpha)), f (alpha) is a global diagnostic error of a infoGAN model of a current task, and alpha is a current infoGAN model parameter;
the federation transfer learning does not need centralized data, and the federation transfer learning objective function is defined as:
Wherein F i (alpha) is the fault loss of the ith user in a brand automobile, and the calculation model is a InfoGAN model optimized by a crayfish improvement algorithm; n i is the total sample size of the i-th vehicle-mounted air conditioner electronic expansion valve.
The beneficial effects are that: compared with the prior art, the invention has the beneficial effects that: the invention can effectively utilize the health data of the vehicle-mounted air conditioner electronic expansion valve of different users, and improves the accuracy and the robustness of health management; according to the invention, parameter optimization is performed through the information generation countermeasure network and the crayfish optimization algorithm, so that the robustness of the model can be effectively ensured; the health management of the electronic expansion valves of different users can be effectively realized under the condition of ensuring the data safety through federal migration learning, and the parameters of the diagnosis model can be adjusted in real time according to the running state change of the expansion valve, so that the diagnosis precision and reliability are improved.
Drawings
FIG. 1 is a frame diagram of the present invention;
FIG. 2 is a flow chart of the optimization algorithm for the model parameters of InfoGAN for the crayfish.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a cloud interconnection-based electronic expansion valve health management system, which can realize real-time monitoring and diagnosis of the health state of an electronic expansion valve and optimize control parameters of the expansion valve under complex environment and working load by utilizing infoGAN models and a crayfish optimization algorithm, so that the reliability of an air conditioning system is improved.
The data acquisition module is used for detecting characteristics representing the running state of the electronic expansion valve of the vehicle-mounted air conditioner, including but not limited to evaporator saturation temperature, condenser saturation temperature, defrosting temperature, fan working current and compressor frequency data.
The data acquisition module comprises an evaporator saturation temperature sensor, a condenser saturation temperature sensor, a defrosting temperature sensor, a fan working current sensor and a compressor frequency sensor. The evaporator saturation temperature sensor is used for collecting the thermal resistance sensor and measuring data of the saturation temperature of the evaporator in the evaporator; condenser saturation temperature sensor: the device is used for collecting data of the thermal resistance sensor and measuring the saturation temperature of the condenser in the condenser; defrosting temperature sensor: the air conditioner temperature data acquisition unit is used for acquiring data of air conditioner temperature in a defrosting state; fan operating current sensor: be used for gathering drive vehicle-mounted air conditioner fan data of current intensity of operation; compressor frequency sensor: the system is used for collecting data of the working frequency of the vehicle-mounted air conditioner compressor.
The data correction module processes missing values and abnormal values caused by extreme environmental factors, sensor faults, communication network abnormality and human factors in the data acquisition module by using the 3D convolutional neural network CNN, and the processed data are transmitted to the CPU module.
The CPU module is used for receiving the data corrected by the data correction module, uploading the data to the data management and feature extraction module in the central integrated processing cloud platform by using the WIFI communication module, and realizing data transmission between the data management and feature extraction module and the data display module.
The WIFI communication module is used for realizing data transmission between the vehicle-mounted CPU module and the central integrated processing cloud platform module.
The vehicle-mounted display module receives and displays the health state of the vehicle-mounted air conditioner expansion valve received from the CPU module.
The central integrated processing cloud platform comprises a data management and feature extraction module, a health diagnosis module and a decision management module. The data management and feature extraction module stores the stream data corrected by the data correction module, extracts input features and stores features closely related to the health state of the vehicle-mounted expansion valve for the health diagnosis module to use. The health diagnosis module fuses federal migration learning and a crayfish improvement infoGAN model to carry out fault diagnosis on the vehicle-mounted air conditioner electronic expansion valve, so that diagnosis accuracy is improved. The decision management module stores the diagnosis result of the health diagnosis module, classifies and decides the fault data of the electronic expansion valves of the vehicle-mounted air conditioners of different types, gives a health report, and feeds the health state back to the user CPU module.
The data management and feature extraction module comprises a plurality of data integration servers, each server represents different automobile brand manufacturers, different servers store different user data, the function of monitoring data sharing is achieved, the function of health management model parameter sharing can be achieved, the problem of data information leakage can be avoided, a model is built through sharing parameter cooperation, and data transmission safety is guaranteed.
The health diagnosis module takes the evaporator saturated temperature, the condenser saturated temperature, the defrosting temperature, the fan working current and the compressor frequency data after the feature extraction as input, identifies and diagnoses leakage, noise, motor step-out and valve misoperation of the electronic expansion valve, and improves the diagnosis precision of the infoGAN model by utilizing a crayfish optimization algorithm.
Optimizing important parameters of the infoGAN model, including the network structure, the learning rate and the regularization coefficient of the generator and the discriminator, wherein the important parameters of the infoGAN model correspond to the positions of the crayfish in the crayfish optimization algorithm; the positions are updated continuously through different behaviors of the crayfish, the algorithm iterates and loops continuously, diagnosis accuracy of expansion valve leakage, noise, motor step out and valve misoperation is used as an optimization target, the optimal position of each crayfish is obtained, and an optimal solution is output, namely an optimal network structure, a learning rate and a regular coefficient of an optimized infoGAN model.
As shown in fig. 2, the health diagnosis module optimizes InfoGAN model parameters by using a crayfish optimization algorithm, and the specific implementation process is as follows:
s1: firstly, initializing corresponding parameters of a crayfish optimization algorithm, wherein the formula is as follows:
Xi,j＝lbj+(ubj-lbj)×rand
Wherein, individual i=1, 2, … M; dimension j=1, 2, … dim; x i,j is the position of individual i in the j dimension; lb j represents the lower bound of the j-th dimension variable; ub j represents the upper bound of the j-th dimension variable; rand is a random number between 0, 1.
S2: in the exploration stage, when T is more than 30 ℃, the position updating formula is as follows:
Wherein X β is a global optimal position obtained along with iteration; x η is the optimal position of the current population; x α is the position of entering the hole to avoid summer heat; wherein the temperature discrimination formula is: t=rand×15+20.
In this stage, crayfish fights for a cave, and the position updating formula is as follows:
Wherein, when the rand is less than 0.5, the phenomenon of competing for the cave does not occur; t is the current iteration number; t+1 is the next iteration number; The position of the individual i in the j dimension at the t+1st time of the iteration; The position of the individual i in the j dimension at the t-th time of iteration; c 2 is a falling curve C 2 =2- (T/T).
S3: in the development stage, when T is more than 30 ℃ and rand is more than or equal to 0.5, the competition stage is entered, and the position updating formula is as follows:
Where z represents random individuals of crayfish, each crayfish will compete with the other by the formula z=round (rand× (M-1)) +1, and X i of the crayfish will change based on the position of X z of the other crayfish.
In the development stage, when T is less than 30 ℃, the food is fed into the foraging stage, the size of the food is judged, and the definition formula is as follows:
Wherein Q is C 3 =3 is a food factor; f i is the fitness value of the ith crayfish, namely the accuracy of health management; f food is food position fitness value; and the crayfish-to-food size judgment is derived from the size of the largest food.
S4: when (when)When the food is too big, the crayfish bites the food by the first paw, and the formula is:
When the food is smaller, the crayfish adopts the second paw foot and the third paw foot to eat alternately, and the process is simulated by using a sine function and a cosine function, and the formula is as follows:
wherein p is foraging intake and is defined as: mu is the temperature most suitable for crayfish; sigma and C 1 were used to control crayfish intake at different temperatures.
When (when)When eating directly, the formula is as follows:
Through the temperature regulation exploration and development process, the global optimization capacity is effectively improved, the randomness is higher, the search precision can be better improved, and the error rate is reduced.
Inputting the data acquired by the data management and feature extraction module into an optimized infoGAN model, acquiring the data of the saturated temperature of the evaporator, the saturated temperature of the condenser, the defrosting temperature, the working current of the fan and the frequency of the compressor after feature extraction, transmitting the data to a generator G, respectively transmitting the generated data of G to a discriminator D and a classifier C, re-extracting the data by the classifier C to obtain an output C, generating a result x, and classifying faults.
Firstly, a fixed characteristic c and a random variable e are obtained and transmitted to a generator, a sample G (e, c) is obtained, and an objective function can be obtained as follows:
minGmaxDV1(D,G)＝V(D,G)-λI(c；G(e,c))
Wherein V 1 (D; G) is an objective function of the infoGAN model; v (D; G) is an objective function of the GAN model; lambda is a super parameter; the I (c; G (e, c)) in the function has edge probability which is difficult to calculate, an approximate distribution F 1 (G, Q) is added to replace the edge probability, mutual information of the generator G and the discriminator D is represented, and a final fitness value, namely a fitness value in a crayfish optimization algorithm, can be obtained by adding mutual information control, and the fitness value is calculated as follows:
minG,QmaxDV1(D,G,Q)＝V(D,G)-λF1(G；Q)
Wherein E is the desired; x is model inputs e and c; p (c) is a probability distribution of c; h (c) is the information entropy of c; q (·) is the auxiliary network Q; p (c) is a probability distribution of c; p G(e,c) is the probability distribution of x. And the health diagnosis module fuses federal migration learning to cooperatively establish a global model by utilizing a plurality of automobile users and an integrated server. Through the participation of K automobile users in federation, all modeling tasks can be regarded as an optimization process of a loss function, the loss function is expressed by min (f (alpha)), f (alpha) is a infoGAN model global diagnosis error of a current task, and alpha is a current model parameter.
Since federation transfer learning does not require centralized data, the federation transfer learning objective function can be defined as follows:
Wherein F i (alpha) is the fault loss of the ith user in a brand automobile, and the calculation model is InfoGAN optimized by a crayfish optimization algorithm; n i is the total sample size of the i-th vehicle-mounted air conditioner electronic expansion valve. Outputting a preliminary fault diagnosis result of the electronic expansion valve of the vehicle-mounted air conditioner by the optimized InfoGAN; and establishing a global model based on fusion federal migration learning, and performing polymerization treatment on the primary result to form a final diagnosis result.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that: the technical scheme described in the foregoing embodiments can be modified or some of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. An electronic expansion valve health management system based on cloud interconnection, which is characterized by comprising:
And a data acquisition module: detecting data representing the running state of an electronic expansion valve of the vehicle-mounted air conditioner;
And a data correction module: the missing value and the abnormal value in the data acquisition module are processed by using the 3D convolutional neural network CNN, and the processed data are transmitted to the CPU module;
CPU module: the data management and feature extraction module is used for receiving the data processed by the data correction module, uploading the data to the central integrated processing cloud platform by utilizing the WIFI communication module, and realizing data transmission with the vehicle-mounted display module;
WIFI communication module: the system is used for realizing data transmission between the CPU module and the central integrated processing cloud platform module;
Central integrated processing cloud platform: the system comprises a data management and feature extraction module, a health diagnosis module and a decision management module; the data management and feature extraction module stores the stream data corrected by the data correction module, extracts input features and stores features closely related to the health state of the vehicle-mounted expansion valve for the health diagnosis module to use; the health diagnosis module utilizes a crayfish optimization algorithm to optimize parameters of the information generation countermeasure network model InfoGAN and outputs a preliminary fault diagnosis result of the vehicle-mounted air conditioner electronic expansion valve; and fusion federal migration learning is carried out, a global model is established, and the primary result is polymerized to form a final diagnosis result; the decision management module stores the final diagnosis result of the health diagnosis module, classifies and decides the fault data of the electronic expansion valves of the vehicle-mounted air conditioners of different types, gives a health report, and feeds the health state back to the user CPU module;
And the vehicle-mounted display module: and receiving and displaying the diagnosis result of the health state of the expansion valve of the vehicle-mounted air conditioner received from the CPU module.
2. The cloud networking based electronic expansion valve health management system of claim 1, wherein said electronic expansion valve operational status data comprises evaporator saturation temperature, condenser saturation temperature, defrost temperature, fan operating current intensity and compressor operating frequency.
3. The cloud interconnection-based electronic expansion valve health management system of claim 1, wherein the data management and feature extraction module comprises a plurality of data integration servers, each server represents a different automobile brand manufacturer, and different servers store different user data to realize monitoring data sharing.
4. The cloud interconnection-based electronic expansion valve health management system of claim 1, wherein the health diagnosis module takes as input the extracted characteristics of the evaporator saturation temperature, the condenser saturation temperature, the defrosting temperature, the fan working current and the compressor frequency data, and identifies and diagnoses leakage, noise, motor step-out and valve malfunction of the electronic expansion valve.
5. The cloud interconnection-based electronic expansion valve health management system of claim 1, wherein the health diagnosis module utilizes a crayfish optimization algorithm to generate the challenge network model InfoGAN parameter optimization implementation process for the information as follows:
Optimizing important parameters of infoGAN, including the network structure, learning rate and regularization coefficient of the generator and the discriminator, wherein the important parameters of the infoGAN model correspond to the positions of the crayfish in the crayfish optimization algorithm; the positions are updated continuously through different behaviors of the crayfish, the algorithm iterates and loops continuously, diagnosis accuracy of expansion valve leakage, noise, motor step out and valve misoperation is used as an optimization target, the optimal position of each crayfish is obtained, and an optimal solution is output, namely an optimal network structure, a learning rate and a regular coefficient of an optimized infoGAN model;
firstly, initializing the position of each crayfish by a crayfish optimization algorithm, namely initializing the network structure, learning rate and regularization parameters of a generator and a discriminator of a infoGAN model, and then updating the position of the crayfish according to different behaviors, wherein the formula is as follows:
Xi,j＝lbj+(ubj-lbj)×rand
Wherein, individual i=1, 2, … M; dimension j=1, 2, … dim; x i,j is the position of individual i in the j dimension; lb j represents the lower bound of the j-th dimension variable; ub j represents the upper bound of the j-th dimension variable; rand is a random number between [0,1 ];
In the exploration stage, when T is more than 30 ℃, the position updating formula is as follows:
Wherein X β is a global optimal position obtained along with iteration; x η is the optimal position of the current population; x α is the position of entering the hole to avoid summer heat; wherein the temperature discrimination formula is: t=rand×15+20;
the phenomenon that crayfish fights for a cave occurs in the exploration stage, and the position updating formula is as follows:
Wherein, when the rand is less than 0.5, the phenomenon of competing for the cave does not occur; t is the current iteration number; t+1 is the next iteration number; The position of the individual i in the j dimension at the t+1st time of the iteration; The position of the individual i in the j dimension at the t-th time of iteration; c 2 is a falling curve C 2 =2- (T/T);
in the development stage, when T is more than 30 ℃ and rand is more than or equal to 0.5, the competition stage is entered, and the position updating formula is as follows:
Where z represents random individuals of crayfish, each crayfish competing with each other by the formula z=round (rand× (M-1)) +1, and X i of the crayfish makes a change based on the position of X z of the other crayfish;
In the development stage, when T is less than 30 ℃, the food is fed into the foraging stage, the size of the food is judged, and the definition formula is as follows:
Wherein Q is C 3 =3 is a food factor; f i is the fitness value of the ith crayfish; f food is food position fitness value; and the crayfish food size judgment is derived from the size of the largest food;
When (when) When the food is too big, the crayfish bites the food by the first paw, and the formula is:
When the food is smaller, the crayfish adopts the second paw foot and the third paw foot to eat alternately, and the process is simulated by using a sine function and a cosine function, and the formula is as follows:
wherein p is foraging intake and is defined as: Mu is the temperature most suitable for crayfish; sigma and C 1 are used to control crayfish intake at different temperatures;
When (when) When eating directly, the formula is as follows:
Through the temperature regulation exploration and development process, the global optimization capacity is effectively improved, the randomness is higher, the search precision can be better improved, and the error rate is reduced.
6. The cloud interconnection-based electronic expansion valve health management system of claim 1, wherein the preliminary fault diagnosis result implementation process of the vehicle-mounted air conditioner electronic expansion valve is as follows:
Inputting the data acquired by the data management and feature extraction module into an optimized infoGAN model, acquiring the data of the saturated temperature of the evaporator, the saturated temperature of the condenser, the defrosting temperature, the working current of the fan and the frequency of the compressor after feature extraction, transmitting the data to a generator G, respectively transmitting the generated data of G to a discriminator D and a classifier C, re-extracting the data by the classifier C to acquire an output C and generate a result x, and classifying faults;
Firstly, a fixed characteristic c and a random variable e are obtained and transmitted to a generator, a sample G (e, c) is obtained, and an objective function is obtained as follows:
minGmaxDV1(D,G)＝V(D,G)-λI(c；G(e,c))
Wherein V 1 (D; G) is an objective function of the infoGAN model; v (D; G) is an objective function of the GAN model; lambda is a super parameter; in the function I (C; G (e, C)) there is an edge probability which is difficult to calculate, and an approximate distribution F 1 (G, Q) is added to replace the edge probability, so that mutual information between the generator G and the output C of the classifier C is represented, and the calculation formula is as follows:
minG,QmaxDV1(D,G,Q)＝V(D,G)-λF1(G；Q)
wherein E is the desired; x is model inputs e and c; p (c) is a probability distribution of c; h (c) is the information entropy of c; q (c|x) is the auxiliary network Q; p G(e,c) is the probability distribution of x.
7. The cloud interconnection-based electronic expansion valve health management system of claim 1, wherein the health diagnosis module fuses federal migration learning to cooperatively establish a global model by using a plurality of automobile users and an integrated server, and specifically comprises:
Through the participation of K automobile users in federation, all modeling tasks can be regarded as the optimization process of a loss function, the loss function is accumulated by diagnostic errors of diagnostic models corresponding to infoGAN of the automobile users, the loss function is represented by min (f (alpha)), f (alpha) is a global diagnostic error of a infoGAN model of a current task, and alpha is a current infoGAN model parameter;
the federation transfer learning does not need centralized data, and the federation transfer learning objective function is defined as:
Wherein F i (alpha) is the fault loss of the ith user in a brand automobile, and the calculation model is a InfoGAN model optimized by a crayfish improvement algorithm; n i is the total sample size of the i-th vehicle-mounted air conditioner electronic expansion valve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410226818.9A CN117948744B (en) | 2024-02-29 | 2024-02-29 | Electronic expansion valve health management system based on cloud interconnection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410226818.9A CN117948744B (en) | 2024-02-29 | 2024-02-29 | Electronic expansion valve health management system based on cloud interconnection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117948744A CN117948744A (en) | 2024-04-30 |
| CN117948744B true CN117948744B (en) | 2024-07-05 |
Family
ID=90794484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410226818.9A Active CN117948744B (en) | 2024-02-29 | 2024-02-29 | Electronic expansion valve health management system based on cloud interconnection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117948744B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115950045A (en) * | 2022-12-02 | 2023-04-11 | 青岛海信日立空调系统有限公司 | Multi-split air conditioning system and control method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9207007B1 (en) * | 2009-10-05 | 2015-12-08 | Robert J. Mowris | Method for calculating target temperature split, target superheat, target enthalpy, and energy efficiency ratio improvements for air conditioners and heat pumps in cooling mode |
| US9644877B2 (en) * | 2012-11-22 | 2017-05-09 | Mitsubishi Electric Corporation | Air-conditioning apparatus and operation control method therefor |
| WO2020008247A1 (en) * | 2018-07-04 | 2020-01-09 | L&T Valves Limited | System and method for health diagnosis and maintenance of industrial units |
| WO2023197711A1 (en) * | 2022-04-11 | 2023-10-19 | 青岛海信日立空调系统有限公司 | Multi-split air-conditioning system, fault positioning method, and fault diagnosis model training method |
| KR20230167589A (en) * | 2022-06-02 | 2023-12-11 | (주)코젠 | Fault Diagnosis System based on Deep Learning Model for Smart IoT Valve |
-
2024
- 2024-02-29 CN CN202410226818.9A patent/CN117948744B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115950045A (en) * | 2022-12-02 | 2023-04-11 | 青岛海信日立空调系统有限公司 | Multi-split air conditioning system and control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117948744A (en) | 2024-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7188255B2 (en) | LEARNING METHOD, LEARNING PROGRAM AND LEARNING DEVICE | |
| CN106682814B (en) | Wind turbine generator fault intelligent diagnosis method based on fault knowledge base | |
| CN111275288A (en) | XGboost-based multi-dimensional data anomaly detection method and device | |
| JP2022524244A (en) | Predictive classification of future behavior | |
| CN111881617B (en) | Data processing method, performance evaluation method and system of wind generating set | |
| JP4641537B2 (en) | Data classification method and apparatus | |
| CN115013298B (en) | Real-time performance online monitoring system and monitoring method of sewage pump | |
| CN111723925B (en) | Fault diagnosis method, device, equipment and medium for on-road intelligent train air conditioning unit | |
| CN109947786A (en) | Failure checks method, apparatus and computer readable storage medium automatically in wind power plant | |
| CN111103137A (en) | Wind turbine gearbox fault diagnosis method based on deep neural network | |
| CN116992322B (en) | Smart city data center management system | |
| CN116502072A (en) | Robust fault diagnosis method for key components of wind generating set under complex variable working conditions | |
| Elnour et al. | Novel actuator fault diagnosis framework for multizone hvac systems using 2-D convolutional neural networks | |
| CN117948744B (en) | Electronic expansion valve health management system based on cloud interconnection | |
| CN117708627A (en) | Intelligent dome adjusting method and system for all-sky-domain telescope | |
| WO2020230422A1 (en) | Abnormality diagnosis device and method | |
| CN109840563B (en) | Fan fault diagnosis method based on three-dimensional membership function fuzzy logic system | |
| Dey et al. | Unsupervised learning techniques for HVAC terminal unit behaviour analysis | |
| CN116451121A (en) | Switch fault diagnosis method, system, equipment and medium | |
| CN113757922A (en) | Deep learning-based air conditioning system energy-saving control method, device and equipment and computer medium | |
| CN115480556A (en) | Intelligent online fault diagnosis system for heating and ventilation electromechanical equipment | |
| CN116070669A (en) | Workshop energy consumption prediction method and management system based on improved deep belief network | |
| US20180087489A1 (en) | Method for windmill farm monitoring | |
| CN110287594B (en) | Aero-engine state diagnosis method based on neural network algorithm | |
| CN118035873B (en) | Fault diagnosis method for parallel convolution improved triple network |
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 |