CN117563144A - Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument - Google Patents
Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument Download PDFInfo
- Publication number
- CN117563144A CN117563144A CN202311642160.1A CN202311642160A CN117563144A CN 117563144 A CN117563144 A CN 117563144A CN 202311642160 A CN202311642160 A CN 202311642160A CN 117563144 A CN117563144 A CN 117563144A
- Authority
- CN
- China
- Prior art keywords
- condition
- acquiring
- infrared therapeutic
- data
- parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000001225 therapeutic effect Effects 0.000 title claims abstract description 139
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000011156 evaluation Methods 0.000 claims abstract description 99
- 230000002159 abnormal effect Effects 0.000 claims abstract description 70
- 230000015556 catabolic process Effects 0.000 claims abstract description 50
- 238000006731 degradation reaction Methods 0.000 claims abstract description 50
- 230000036541 health Effects 0.000 claims abstract description 42
- 238000012216 screening Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 32
- 238000012549 training Methods 0.000 claims description 25
- 230000004927 fusion Effects 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 210000002569 neuron Anatomy 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000000513 principal component analysis Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 241000288105 Grus Species 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012821 model calculation Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 206010033675 panniculitis Diseases 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 210000004304 subcutaneous tissue Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000010792 warming Methods 0.000 description 2
- 102000016938 Catalase Human genes 0.000 description 1
- 108010053835 Catalase Proteins 0.000 description 1
- 206010061218 Inflammation Diseases 0.000 description 1
- 208000029549 Muscle injury Diseases 0.000 description 1
- 206010052428 Wound Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000019522 cellular metabolic process Effects 0.000 description 1
- 208000037976 chronic inflammation Diseases 0.000 description 1
- 230000006020 chronic inflammation Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 230000008338 local blood flow Effects 0.000 description 1
- 210000002540 macrophage Anatomy 0.000 description 1
- 210000003470 mitochondria Anatomy 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000000242 pagocytic effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000000554 physical therapy Methods 0.000 description 1
- 238000001243 protein synthesis Methods 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 230000014616 translation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
-
- 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
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
-
- 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
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
- A61N2005/0658—Radiation therapy using light characterised by the wavelength of light used
- A61N2005/0659—Radiation therapy using light characterised by the wavelength of light used infrared
Abstract
The invention discloses a method and a system for evaluating the condition of an infrared therapeutic apparatus and predicting the residual life, wherein the method comprises the following steps: acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, and screening evaluation indexes from the preprocessed historical operation parameters by utilizing maximum condition mutual information; acquiring index parameters under normal working conditions by using a clustering method, constructing a normal operation model, acquiring real-time residual data to identify abnormal working conditions, and acquiring health condition assessment scores corresponding to the abnormal working conditions according to a parameter deviation scoring system; and acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics, constructing a residual life prediction model, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics. The invention utilizes the residual error of the real-time operation data and the reference operation data to identify the abnormal working condition, can accurately evaluate the health condition of the infrared therapeutic apparatus, avoids the faults of midway downtime and the like of the infrared therapeutic apparatus, and ensures the stable operation.
Description
Technical Field
The invention relates to the technical field of operation condition assessment, in particular to a method and a system for assessing the condition and predicting the residual life of an infrared therapeutic apparatus.
Background
The infrared therapeutic apparatus is a medical apparatus which can penetrate human skin by infrared rays to generate heat effect on muscle and subcutaneous tissue, thereby achieving therapeutic effect. After being absorbed by human body, the infrared ray resonates with water molecules in the body, so that organism cells are at high vibration level, resonance and warming effect are generated, the temperature of deep subcutaneous tissue parts is increased, warming effect is generated, capillary vessels are expanded, the catalase activity of cell mitochondria is increased, and cell metabolism and protein synthesis are enhanced. Is often used as an auxiliary physical therapy for relieving muscle injury pain and the like. When used for treating chronic inflammation, it can promote local blood circulation, increase phagocytic function of cells and macrophages, control infection, and promote inflammation dissipation. In addition, the utility model can promote the regeneration and repair of wounds.
The key parameters reflecting the health condition of the equipment are continuously collected under the normal operation condition of the infrared therapeutic equipment by utilizing data perception, and real-time data, historical data and offline data are deeply analyzed so as to accurately identify the change of the equipment state and find possible problems. The application of the infrared therapeutic apparatus health condition assessment key technology not only can comprehensively and accurately monitor the running state of the equipment, but also can predict the future state of the equipment. By the method, early detection and early warning of equipment condition problems can be achieved, and therefore running risks of equipment are reduced. At present, the infrared therapeutic apparatus lacks health state evaluation, greatly influences the therapeutic effect, and cannot achieve the purpose of real-time monitoring in the therapeutic process, so how to evaluate the health state and predict the residual life of the infrared therapeutic apparatus through key parameters is a problem to be solved.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method and a system for evaluating the condition of an infrared therapeutic apparatus and predicting the residual life.
The first aspect of the invention provides a method for evaluating the condition and predicting the residual life of an infrared therapeutic apparatus, which comprises the following steps:
acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
carrying out abnormal working condition identification of the infrared therapeutic apparatus according to the residual data, and obtaining a health condition assessment score corresponding to the abnormal working condition according to a parameter deviation scoring system;
and acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
In the scheme, the characteristic selection is carried out in the preprocessed historical operation parameters by utilizing the maximum condition mutual information, and the evaluation indexes are screened, specifically:
acquiring state parameters, environment parameters and energy conversion parameters of the infrared therapeutic apparatus within preset time, preprocessing the acquired operation parameters, removing missing data and abnormal data in the operation parameters, acquiring start-stop conditions of the infrared therapeutic apparatus, and removing stop data in the operation parameters;
constructing a historical operation parameter data set through reserved operation parameters, acquiring a state code corresponding to each operation parameter in the historical operation parameter data set, extracting corresponding working condition information according to the state code, and generating a data tag of each historical operation parameter by utilizing the working condition information;
clustering historical operation parameters according to different data tags, constructing operation parameter subsets corresponding to the different data tags, extracting parameter features corresponding to each operation parameter in the operation parameter subsets, and performing transverse comparison between the different operation parameter subsets on the parameter features;
acquiring operation parameters with deviation larger than a preset threshold value through comparison results, taking the unselected operation parameters as evaluation indexes to be selected according to the acquired operation parameters as target evaluation indexes, and calculating mutual information between the target evaluation indexes and the evaluation indexes to be selected;
Adding the to-be-selected evaluation index corresponding to the maximum mutual information into a preset index set, calculating the maximum condition mutual information of the rest to-be-selected evaluation indexes according to the target evaluation index and the preset index set, and repeating iterative calculation until the evaluation indexes in the preset index set reach the preset number;
and acquiring the evaluation index of the health condition of the infrared therapeutic instrument according to the evaluation index in the preset index set and the target evaluation index.
In the scheme, index parameters under normal working conditions are obtained according to the evaluation indexes by using a clustering method, and a normal operation model is constructed according to the index parameters, specifically:
acquiring the clustering number through normal working conditions and abnormal working conditions, determining an initial clustering center in working condition data labels of historical operation parameters to perform clustering, acquiring Euclidean distances from different working condition data labels to the initial clustering center, and performing clustering distribution according to the Euclidean distances;
acquiring a final clustering result according to iterative clustering, extracting working condition data labels in the class clusters corresponding to the normal working conditions, marking, and screening index parameters under the normal working conditions by using the evaluation index under the marked working condition data labels;
constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, setting a corresponding number of mixed kernel functions according to the number of marked index parameters, initializing weighted multi-kernel, acquiring multi-kernel weights through multi-kernel learning, fusing the index parameters under normal working conditions, and generating a fusion parameter set;
And constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and hidden layer neurons of the normal operation model, training the normal operation model through the fusion parameter set, and outputting the normal operation model which meets the standard.
In this scheme, carry out the unusual operating mode discernment of infrared therapeutic instrument according to residual error data, specifically do:
acquiring training data for normalization processing, acquiring evaluation data corresponding to the training data through a normal operation model, acquiring reference residual error data according to the absolute value of the difference value between the training data and the evaluation data, and acquiring reference probability distribution representing normal operation of the infrared therapeutic apparatus according to the reference residual error data;
acquiring real-time operation parameters of the infrared therapeutic apparatus, importing the real-time operation parameters into the normal operation model to acquire real-time residual error data, and acquiring target probability distribution representing the current operation condition of the infrared therapeutic apparatus according to the real-time residual error data;
and calculating relative entropy according to the reference probability distribution and the target probability distribution, and when the relative entropy is larger than a preset threshold value, proving that the current running condition is different from the normal running condition, and the current infrared therapeutic instrument has potential faults and is an abnormal working condition.
In the scheme, a health condition assessment score corresponding to an abnormal working condition is obtained according to a parameter deviation scoring system, and specifically comprises the following steps:
a score interval corresponding to the deviation score of the preset parameter is constructed, a weighted neural network is constructed to acquire nonlinear relations between different score intervals and target probability distribution of the current running condition of the infrared therapeutic instrument, and the deviation score is acquired by utilizing the nonlinear relations;
generating initial weights of abnormal working conditions according to the deviation between the relative entropy of the reference probability distribution and the target probability distribution and the reference threshold value, and introducing a multi-head attention mechanism to acquire self-attention weights of real-time residual data corresponding to the target probability distribution;
and combining the initial weight and the self-attention weight to obtain the parameter weight corresponding to each index parameter, and constructing a parameter deviation scoring system according to the deviation score and combining the parameter weight and the reference value to generate a health condition assessment score corresponding to the current abnormal working condition.
In this scheme, corresponding degradation characteristics are extracted according to the initial remaining life interval, and a deep learning method is utilized to construct a remaining life prediction model, specifically:
acquiring a historical abnormal working condition instance by a big data means, acquiring corresponding scoring areas of the historical abnormal working condition instance according to the parameter deviation scoring system, and clustering the historical abnormal working condition instance by using the scoring areas;
Obtaining average residual life of corresponding class clusters among different scoring areas, carrying out principal component analysis according to the average residual life, taking the operation parameter with the highest principal component score as a principal component parameter, and carrying out principal component direction projection by utilizing the principal component parameter to obtain a corresponding operation parameter scatter diagram;
selecting an operation parameter from the operation parameter scatter diagram according to a preset range to obtain a degradation characteristic, and matching the degradation characteristic with a corresponding scoring area and the average residual life;
acquiring an initial residual life interval of the infrared therapeutic instrument according to a health condition evaluation score of the current operation working condition of the infrared therapeutic instrument and a score interval falling into the health condition evaluation score, and reading corresponding degradation characteristics according to the residual life interval;
carrying out feature coding reconstruction on the degradation features according to a stacked self-encoder, introducing an attention mechanism in feature decoding to weight and characterize the importance degree of the degradation features, acquiring the reconstructed degradation features, and importing the reconstructed degradation features into a BiGRU network to construct a residual life prediction model;
and capturing the time dependence of the reconstructed degradation characteristic sequence through the BiGRU network, and outputting a residual life prediction result of the infrared therapeutic instrument through the full-connection layer.
The second aspect of the present invention also provides a system for evaluating the condition of an infrared therapeutic apparatus and predicting the remaining life, the system comprising: the infrared therapeutic instrument condition evaluation and residual life prediction method comprises a memory and a processor, wherein the memory comprises an infrared therapeutic instrument condition evaluation and residual life prediction method program, and the infrared therapeutic instrument condition evaluation and residual life prediction method program realizes the following steps when being executed by the processor:
acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
carrying out abnormal working condition identification of the infrared therapeutic apparatus according to the residual data, and obtaining a health condition assessment score corresponding to the abnormal working condition according to a parameter deviation scoring system;
and acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
The invention discloses a method and a system for evaluating the condition of an infrared therapeutic apparatus and predicting the residual life, wherein the method comprises the following steps: acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, and screening evaluation indexes from the preprocessed historical operation parameters by utilizing maximum condition mutual information; acquiring index parameters under normal working conditions by using a clustering method, constructing a normal operation model, acquiring real-time residual data to identify abnormal working conditions, and acquiring health condition assessment scores corresponding to the abnormal working conditions according to a parameter deviation scoring system; and acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics, constructing a residual life prediction model, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics. The invention utilizes the residual error of the real-time operation data and the reference operation data to identify the abnormal working condition, can accurately evaluate the health condition of the infrared therapeutic apparatus, avoids the faults of midway downtime and the like of the infrared therapeutic apparatus, and ensures the stable operation.
Drawings
FIG. 1 is a flow chart of a method for evaluating the condition and predicting the remaining life of an infrared therapeutic apparatus according to the present invention;
FIG. 2 is a flow chart of a method for constructing a normal operation model according to index parameters according to the present invention;
FIG. 3 shows a flow chart of an abnormal condition identification method of the infrared therapeutic apparatus of the present invention;
FIG. 4 is a block diagram of an infrared therapeutic apparatus condition assessment and remaining life prediction system of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
FIG. 1 is a flow chart of a method for evaluating the condition and predicting the remaining life of an infrared therapeutic apparatus according to the present invention.
As shown in fig. 1, a first aspect of the present invention provides a method for evaluating a condition of an infrared therapeutic apparatus and predicting remaining life, including:
s102, acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
S104, acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
s106, identifying abnormal working conditions of the infrared therapeutic apparatus according to the residual data, and acquiring health condition assessment scores corresponding to the abnormal working conditions according to a parameter deviation scoring system;
s108, acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
The method comprises the steps of acquiring state parameters such as equipment temperature, current and the like, environment parameters where equipment is located, energy conversion parameters such as output power and the like of an infrared therapeutic instrument in preset time, preprocessing the acquired operation parameters, removing missing data and abnormal data in the operation parameters, and judging whether the abnormal information is the abnormal data according to whether the output power is smaller than or equal to zero or not; acquiring start-stop conditions of the infrared therapeutic apparatus, and eliminating stop data in the operation parameters; constructing a historical operation parameter data set through reserved operation parameters, acquiring state codes, such as 'operation', 'stoppage', 'failure', and the like, corresponding to each operation parameter in the historical operation parameter data set, extracting corresponding working condition information according to the state codes, and generating a data tag of each historical operation parameter by utilizing the working condition information; clustering historical operation parameters according to different data tags, constructing operation parameter subsets corresponding to the different data tags, extracting parameter features corresponding to each operation parameter in the operation parameter subsets, and performing transverse comparison between the different operation parameter subsets on the parameter features; acquiring operation parameters with deviation larger than a preset threshold value through comparison results, taking the unselected operation parameters as evaluation indexes x to be selected according to the acquired operation parameters as target evaluation indexes y, and calculating mutual information between the target evaluation indexes and the evaluation indexes to be selected; adding a to-be-selected evaluation index corresponding to the maximum mutual information into a preset index set s, and calculating maximum condition mutual information I (X, y|s) of the rest to-be-selected evaluation indexes according to the target evaluation index and the preset index set, wherein the condition mutual information defines the mutual information among random variables X, Y as condition mutual information I (X, y|Z) under the condition that three random variables X, Y, Z and a random variable Z are known; repeating iterative calculation until the evaluation indexes in the preset index set reach the preset number; according to the evaluation indexes in the preset index set and the target evaluation indexes, the evaluation indexes of the health condition of the infrared therapeutic instrument are obtained, and the index redundancy in the preset index set can be ensured to be minimized.
FIG. 2 shows a flow chart of a method of constructing a normal operation model according to the index parameters of the present invention.
According to the embodiment of the invention, index parameters under normal working conditions are obtained according to the evaluation indexes by using a clustering method, and a normal operation model is constructed according to the index parameters, specifically:
s202, acquiring the number of clusters through normal working conditions and abnormal working conditions, determining initial cluster centers in working condition data labels of historical operation parameters to perform clustering, acquiring Euclidean distances from different working condition data labels to the initial cluster centers, and performing cluster distribution according to the Euclidean distances;
s204, obtaining a final clustering result according to iterative clustering, extracting working condition data labels in the class clusters corresponding to the normal working conditions, marking, and screening index parameters under the normal working conditions by using the evaluation indexes under the marked working condition data labels;
s206, constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, setting a corresponding number of mixed kernel functions according to the number of marked index parameters, initializing weighted multi-kernel, acquiring multi-kernel weights through multi-kernel learning, fusing the index parameters under normal working conditions, and generating a fusion parameter set;
S208, constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and hidden layer neurons of the normal operation model, training the normal operation model through the fusion parameter set, and outputting the normal operation model which meets the standard.
It should be noted that, index parameters under normal working conditions are obtained through a clustering algorithm, and a clustering result is evaluated through a contour coefficient, and when the contour coefficient reaches a preset standard, the clustering result is output. Introducing multi-core learning, constructing a multi-core learning space, placing index parameters into the multi-core learning space for self-adaptation and fusion, constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, optimizing weight information of different mixed kernel functions, and utilizing the acquired optimal kernel function information to realize adaptation and fusion of the index parameters; and constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and the hidden layer neuron number corresponding to the model by utilizing a genetic algorithm or a particle swarm algorithm, acquiring optimized model parameters, generating a training set and a testing set according to the fusion parameter set, and performing model training to acquire the normal operation model based on the NAR dynamic neural network.
FIG. 3 shows a flow chart of an abnormal condition identification method of the infrared therapeutic apparatus of the present invention.
According to the embodiment of the invention, the abnormal working condition of the infrared therapeutic apparatus is identified according to the residual data, specifically:
s302, acquiring training data for normalization processing, acquiring evaluation data corresponding to the training data through a normal operation model, acquiring reference residual data according to the absolute value of the difference value between the training data and the evaluation data, and acquiring reference probability distribution representing normal operation of an infrared therapeutic apparatus according to the reference residual data;
s304, acquiring real-time operation parameters of the infrared therapeutic apparatus, importing the real-time operation parameters into the normal operation model to acquire real-time residual error data, and acquiring target probability distribution representing the current operation condition of the infrared therapeutic apparatus according to the real-time residual error data;
s306, calculating relative entropy according to the reference probability distribution and the target probability distribution, and when the relative entropy is larger than a preset threshold value, proving that the current running condition is different from the normal running condition, and the current infrared therapeutic apparatus has potential faults and is an abnormal working condition.
It should be noted that, the average value and variance of the residual data are used to obtain the corresponding probability distribution, the relative entropy is more sensitive to the slight and early faults, the obtained relative entropy is used as the judgment index of the abnormal working condition to obtain the running condition change information of the infrared therapeutic apparatus from the probability distribution difference layer, and the calculation amount and calculation process of the abnormal condition identification are simplified.
The method includes the steps that a score interval corresponding to a preset parameter deviation score is constructed, a weighted neural network is constructed to obtain nonlinear relations between different score intervals and target probability distribution of the current running condition of the infrared therapeutic instrument, and the deviation score is obtained by the nonlinear relations; generating initial weights of abnormal working conditions according to the deviation between the relative entropy of the reference probability distribution and the target probability distribution and the reference threshold value, and introducing a multi-head attention mechanism to acquire self-attention weights of real-time residual data corresponding to the target probability distribution; and combining the initial weight and the self-attention weight to obtain the parameter weight corresponding to each index parameter, and constructing a parameter deviation scoring system according to the deviation score and combining the parameter weight and the reference value to generate a health condition assessment score corresponding to the current abnormal working condition.
The historical abnormal working condition examples are obtained through big data means, corresponding scoring areas of the historical abnormal working condition examples are obtained according to the parameter deviation scoring system, and the historical abnormal working condition examples are clustered through the scoring areas; obtaining average residual life of corresponding class clusters among different scoring areas, carrying out principal component analysis according to the average residual life, taking the operation parameter with the highest principal component score as a principal component parameter, and carrying out principal component direction projection by utilizing the principal component parameter to obtain a corresponding operation parameter scatter diagram; selecting an operation parameter from the operation parameter scatter diagram according to a preset range to obtain a degradation characteristic, and matching the degradation characteristic with a corresponding scoring area and the average residual life; and acquiring an initial residual life span of the infrared therapeutic instrument according to a health condition evaluation score of the current operation working condition of the infrared therapeutic instrument and a score zone which falls into the health condition evaluation score, and reading corresponding degradation characteristics according to the residual life span.
Stacking and constructing a multi-layer noise reduction self-coding network on the basis of a self-coder network to generate a deep nonlinear mapping capability depth architecture, taking a hidden layer learned by a previous self-coding network as an input layer of a next self-coding network, carrying out feature coding reconstruction on the degradation features according to the stacked self-coder, reducing dimensionality through feature reconstruction and feature fusion, reducing model calculation amount, introducing a attention mechanism to weight and characterize the importance degree of the degradation features in feature decoding, acquiring the reconstructed degradation features, and introducing the reconstructed degradation features into a BiGRU network to construct a residual life prediction model, wherein the BiGRU network is composed of GRUs with unidirectional and opposite directions, and can further improve the information memory capability and the accuracy of long-time sequence prediction; and capturing the time dependence of the reconstructed degradation characteristic sequence through the BiGRU network, and outputting a residual life prediction result of the infrared therapeutic instrument through the full-connection layer.
According to the embodiment of the invention, the infrared therapeutic apparatus under the abnormal working condition is classified and divided according to the health condition evaluation score and the residual life prediction result of the abnormal working condition, the use data of the infrared therapeutic apparatus in the current preset time period is obtained, the average use duration of the infrared therapeutic apparatus is analyzed according to the use data, the comparison is carried out according to the average use duration and the residual life prediction result corresponding to different categories, and the category corresponding to the infrared therapeutic apparatus smaller than the average use duration is eliminated; training LSTM network learning time dependence according to historical demand information of the infrared therapeutic apparatus, constructing a demand prediction model, acquiring the infrared therapeutic apparatus prediction demand of a preset time step, setting use priority for the infrared therapeutic apparatus of different categories according to deviation information of category characteristics corresponding to the infrared therapeutic apparatus of different categories through the prediction demand, and preferentially using the infrared therapeutic apparatus, so that the infrared therapeutic apparatus meeting the demand and having the minimum residual life can avoid faults such as midway downtime and the like of the infrared therapeutic apparatus.
FIG. 4 is a block diagram of an infrared therapeutic apparatus condition assessment and remaining life prediction system of the present invention.
The second aspect of the present invention also provides an infrared therapeutic apparatus condition assessment and remaining life prediction system 4, comprising: a memory 41, a processor 42, wherein the memory includes an infrared therapeutic apparatus condition evaluation and remaining life prediction method program, and the infrared therapeutic apparatus condition evaluation and remaining life prediction method program when executed by the processor realizes the following steps:
acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
carrying out abnormal working condition identification of the infrared therapeutic apparatus according to the residual data, and obtaining a health condition assessment score corresponding to the abnormal working condition according to a parameter deviation scoring system;
And acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
According to the embodiment of the invention, index parameters under normal working conditions are obtained according to the evaluation indexes by using a clustering method, and a normal operation model is constructed according to the index parameters, specifically:
acquiring the clustering number through normal working conditions and abnormal working conditions, determining an initial clustering center in working condition data labels of historical operation parameters to perform clustering, acquiring Euclidean distances from different working condition data labels to the initial clustering center, and performing clustering distribution according to the Euclidean distances;
acquiring a final clustering result according to iterative clustering, extracting working condition data labels in the class clusters corresponding to the normal working conditions, marking, and screening index parameters under the normal working conditions by using the evaluation index under the marked working condition data labels;
constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, setting a corresponding number of mixed kernel functions according to the number of marked index parameters, initializing weighted multi-kernel, acquiring multi-kernel weights through multi-kernel learning, fusing the index parameters under normal working conditions, and generating a fusion parameter set;
And constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and hidden layer neurons of the normal operation model, training the normal operation model through the fusion parameter set, and outputting the normal operation model which meets the standard.
It should be noted that, index parameters under normal working conditions are obtained through a clustering algorithm, and a clustering result is evaluated through a contour coefficient, and when the contour coefficient reaches a preset standard, the clustering result is output. Introducing multi-core learning, constructing a multi-core learning space, placing index parameters into the multi-core learning space for self-adaptation and fusion, constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, optimizing weight information of different mixed kernel functions, and utilizing the acquired optimal kernel function information to realize adaptation and fusion of the index parameters; and constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and the hidden layer neuron number corresponding to the model by utilizing a genetic algorithm or a particle swarm algorithm, acquiring optimized model parameters, generating a training set and a testing set according to the fusion parameter set, and performing model training to acquire the normal operation model based on the NAR dynamic neural network.
According to the embodiment of the invention, the abnormal working condition of the infrared therapeutic apparatus is identified according to the residual data, specifically:
acquiring training data for normalization processing, acquiring evaluation data corresponding to the training data through a normal operation model, acquiring reference residual error data according to the absolute value of the difference value between the training data and the evaluation data, and acquiring reference probability distribution representing normal operation of the infrared therapeutic apparatus according to the reference residual error data;
acquiring real-time operation parameters of the infrared therapeutic apparatus, importing the real-time operation parameters into the normal operation model to acquire real-time residual error data, and acquiring target probability distribution representing the current operation condition of the infrared therapeutic apparatus according to the real-time residual error data;
and calculating relative entropy according to the reference probability distribution and the target probability distribution, and when the relative entropy is larger than a preset threshold value, proving that the current running condition is different from the normal running condition, and the current infrared therapeutic instrument has potential faults and is an abnormal working condition.
It should be noted that, the average value and variance of the residual data are used to obtain the corresponding probability distribution, the relative entropy is more sensitive to the slight and early faults, the obtained relative entropy is used as the judgment index of the abnormal working condition to obtain the running condition change information of the infrared therapeutic apparatus from the probability distribution difference layer, and the calculation amount and calculation process of the abnormal condition identification are simplified.
The method includes the steps that a score interval corresponding to a preset parameter deviation score is constructed, a weighted neural network is constructed to obtain nonlinear relations between different score intervals and target probability distribution of the current running condition of the infrared therapeutic instrument, and the deviation score is obtained by the nonlinear relations; generating initial weights of abnormal working conditions according to the deviation between the relative entropy of the reference probability distribution and the target probability distribution and the reference threshold value, and introducing a multi-head attention mechanism to acquire self-attention weights of real-time residual data corresponding to the target probability distribution; and combining the initial weight and the self-attention weight to obtain the parameter weight corresponding to each index parameter, and constructing a parameter deviation scoring system according to the deviation score and combining the parameter weight and the reference value to generate a health condition assessment score corresponding to the current abnormal working condition.
The historical abnormal working condition examples are obtained through big data means, corresponding scoring areas of the historical abnormal working condition examples are obtained according to the parameter deviation scoring system, and the historical abnormal working condition examples are clustered through the scoring areas; obtaining average residual life of corresponding class clusters among different scoring areas, carrying out principal component analysis according to the average residual life, taking the operation parameter with the highest principal component score as a principal component parameter, and carrying out principal component direction projection by utilizing the principal component parameter to obtain a corresponding operation parameter scatter diagram; selecting an operation parameter from the operation parameter scatter diagram according to a preset range to obtain a degradation characteristic, and matching the degradation characteristic with a corresponding scoring area and the average residual life; and acquiring an initial residual life span of the infrared therapeutic instrument according to a health condition evaluation score of the current operation working condition of the infrared therapeutic instrument and a score zone which falls into the health condition evaluation score, and reading corresponding degradation characteristics according to the residual life span.
Stacking and constructing a multi-layer noise reduction self-coding network on the basis of a self-coder network to generate a deep nonlinear mapping capability depth architecture, taking a hidden layer learned by a previous self-coding network as an input layer of a next self-coding network, carrying out feature coding reconstruction on the degradation features according to the stacked self-coder, reducing dimensionality through feature reconstruction and feature fusion, reducing model calculation amount, introducing a attention mechanism to weight and characterize the importance degree of the degradation features in feature decoding, acquiring the reconstructed degradation features, and introducing the reconstructed degradation features into a BiGRU network to construct a residual life prediction model, wherein the BiGRU network is composed of GRUs with unidirectional and opposite directions, and can further improve the information memory capability and the accuracy of long-time sequence prediction; and capturing the time dependence of the reconstructed degradation characteristic sequence through the BiGRU network, and outputting a residual life prediction result of the infrared therapeutic instrument through the full-connection layer.
The third aspect of the present invention also provides a computer-readable storage medium having embodied therein an infrared therapeutic apparatus condition evaluation and remaining life prediction method program which, when executed by a processor, implements the steps of the infrared therapeutic apparatus condition evaluation and remaining life prediction method as described above.
In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. The method for evaluating the condition and predicting the residual life of the infrared therapeutic apparatus is characterized by comprising the following steps of:
acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
carrying out abnormal working condition identification of the infrared therapeutic apparatus according to the residual data, and obtaining a health condition assessment score corresponding to the abnormal working condition according to a parameter deviation scoring system;
And acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
2. The method for evaluating the condition and predicting the residual life of an infrared therapeutic apparatus according to claim 1, wherein the characteristic selection is performed in the preprocessed historical operation parameters by using the maximum condition mutual information, and the evaluation indexes are selected specifically:
acquiring state parameters, environment parameters and energy conversion parameters of the infrared therapeutic apparatus within preset time, preprocessing the acquired operation parameters, removing missing data and abnormal data in the operation parameters, acquiring start-stop conditions of the infrared therapeutic apparatus, and removing stop data in the operation parameters;
constructing a historical operation parameter data set through reserved operation parameters, acquiring a state code corresponding to each operation parameter in the historical operation parameter data set, extracting corresponding working condition information according to the state code, and generating a data tag of each historical operation parameter by utilizing the working condition information;
Clustering historical operation parameters according to different data tags, constructing operation parameter subsets corresponding to the different data tags, extracting parameter features corresponding to each operation parameter in the operation parameter subsets, and performing transverse comparison between the different operation parameter subsets on the parameter features;
acquiring operation parameters with deviation larger than a preset threshold value through comparison results, taking the unselected operation parameters as evaluation indexes to be selected according to the acquired operation parameters as target evaluation indexes, and calculating mutual information between the target evaluation indexes and the evaluation indexes to be selected;
adding the to-be-selected evaluation index corresponding to the maximum mutual information into a preset index set, calculating the maximum condition mutual information of the rest to-be-selected evaluation indexes according to the target evaluation index and the preset index set, and repeating iterative calculation until the evaluation indexes in the preset index set reach the preset number;
and acquiring the evaluation index of the health condition of the infrared therapeutic instrument according to the evaluation index in the preset index set and the target evaluation index.
3. The method for evaluating the condition and predicting the residual life of an infrared therapeutic apparatus according to claim 1, wherein the method for obtaining the index parameters under the normal working condition according to the evaluation index by using a clustering method, and constructing a normal operation model according to the index parameters is as follows:
Acquiring the clustering number through normal working conditions and abnormal working conditions, determining an initial clustering center in working condition data labels of historical operation parameters to perform clustering, acquiring Euclidean distances from different working condition data labels to the initial clustering center, and performing clustering distribution according to the Euclidean distances;
acquiring a final clustering result according to iterative clustering, extracting working condition data labels in the class clusters corresponding to the normal working conditions, marking, and screening index parameters under the normal working conditions by using the evaluation index under the marked working condition data labels;
constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, setting a corresponding number of mixed kernel functions according to the number of marked index parameters, initializing weighted multi-kernel, acquiring multi-kernel weights through multi-kernel learning, fusing the index parameters under normal working conditions, and generating a fusion parameter set;
and constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and hidden layer neurons of the normal operation model, training the normal operation model through the fusion parameter set, and outputting the normal operation model which meets the standard.
4. The method for evaluating the condition and predicting the residual life of an infrared therapeutic apparatus according to claim 1, wherein the identifying of the abnormal condition of the infrared therapeutic apparatus is performed according to the residual data, specifically:
Acquiring training data for normalization processing, acquiring evaluation data corresponding to the training data through a normal operation model, acquiring reference residual error data according to the absolute value of the difference value between the training data and the evaluation data, and acquiring reference probability distribution representing normal operation of the infrared therapeutic apparatus according to the reference residual error data;
acquiring real-time operation parameters of the infrared therapeutic apparatus, importing the real-time operation parameters into the normal operation model to acquire real-time residual error data, and acquiring target probability distribution representing the current operation condition of the infrared therapeutic apparatus according to the real-time residual error data;
and calculating relative entropy according to the reference probability distribution and the target probability distribution, and when the relative entropy is larger than a preset threshold value, proving that the current running condition is different from the normal running condition, and the current infrared therapeutic instrument has potential faults and is an abnormal working condition.
5. The method for evaluating the condition and predicting the remaining life of an infrared therapeutic apparatus according to claim 1, wherein the method for acquiring the health condition evaluation score corresponding to the abnormal condition according to the parameter deviation scoring system is specifically as follows:
a score interval corresponding to the deviation score of the preset parameter is constructed, a weighted neural network is constructed to acquire nonlinear relations between different score intervals and target probability distribution of the current running condition of the infrared therapeutic instrument, and the deviation score is acquired by utilizing the nonlinear relations;
Generating initial weights of abnormal working conditions according to the deviation between the relative entropy of the reference probability distribution and the target probability distribution and the reference threshold value, and introducing a multi-head attention mechanism to acquire self-attention weights of real-time residual data corresponding to the target probability distribution;
and combining the initial weight and the self-attention weight to obtain the parameter weight corresponding to each index parameter, and constructing a parameter deviation scoring system according to the deviation score and combining the parameter weight and the reference value to generate a health condition assessment score corresponding to the current abnormal working condition.
6. The method for evaluating the condition of an infrared therapeutic apparatus and predicting the residual life according to claim 1, wherein the method for constructing the residual life prediction model by using the deep learning method is characterized by extracting corresponding degradation characteristics according to the initial residual life interval, and specifically comprising the following steps:
acquiring a historical abnormal working condition instance by a big data means, acquiring corresponding scoring areas of the historical abnormal working condition instance according to the parameter deviation scoring system, and clustering the historical abnormal working condition instance by using the scoring areas;
obtaining average residual life of corresponding class clusters among different scoring areas, carrying out principal component analysis according to the average residual life, taking the operation parameter with the highest principal component score as a principal component parameter, and carrying out principal component direction projection by utilizing the principal component parameter to obtain a corresponding operation parameter scatter diagram;
Selecting an operation parameter from the operation parameter scatter diagram according to a preset range to obtain a degradation characteristic, and matching the degradation characteristic with a corresponding scoring area and the average residual life;
acquiring an initial residual life interval of the infrared therapeutic instrument according to a health condition evaluation score of the current operation working condition of the infrared therapeutic instrument and a score interval falling into the health condition evaluation score, and reading corresponding degradation characteristics according to the residual life interval;
carrying out feature coding reconstruction on the degradation features according to a stacked self-encoder, introducing an attention mechanism in feature decoding to weight and characterize the importance degree of the degradation features, acquiring the reconstructed degradation features, and importing the reconstructed degradation features into a BiGRU network to construct a residual life prediction model;
and capturing the time dependence of the reconstructed degradation characteristic sequence through the BiGRU network, and outputting a residual life prediction result of the infrared therapeutic instrument through the full-connection layer.
7. An infrared therapeutic apparatus condition assessment and remaining life prediction system, comprising: the infrared therapeutic instrument condition evaluation and residual life prediction method comprises a memory and a processor, wherein the memory comprises an infrared therapeutic instrument condition evaluation and residual life prediction method program, and the infrared therapeutic instrument condition evaluation and residual life prediction method program realizes the following steps when being executed by the processor:
Acquiring historical operation parameters of the infrared therapeutic apparatus under different working conditions, preprocessing the historical operation parameters, performing feature selection in the preprocessed historical operation parameters by utilizing maximum condition mutual information, and screening evaluation indexes;
acquiring index parameters under normal working conditions according to the evaluation indexes by using a clustering method, constructing a normal operation model according to the index parameters, and inputting real-time operation parameters of an infrared therapeutic instrument as a model to acquire real-time residual error data;
carrying out abnormal working condition identification of the infrared therapeutic apparatus according to the residual data, and obtaining a health condition assessment score corresponding to the abnormal working condition according to a parameter deviation scoring system;
and acquiring an initial residual life interval according to the health condition evaluation score, extracting corresponding degradation characteristics according to the initial residual life interval, constructing a residual life prediction model by using a deep learning method, and carrying out residual life prediction of the infrared therapeutic instrument by combining characteristic parameters corresponding to the degradation characteristics.
8. The system for evaluating the condition of an infrared therapeutic apparatus and predicting the residual life according to claim 7, wherein the clustering method is used for obtaining index parameters under normal working conditions according to the evaluation index, and constructing a normal operation model according to the index parameters, specifically:
Acquiring the clustering number through normal working conditions and abnormal working conditions, determining an initial clustering center in working condition data labels of historical operation parameters to perform clustering, acquiring Euclidean distances from different working condition data labels to the initial clustering center, and performing clustering distribution according to the Euclidean distances;
acquiring a final clustering result according to iterative clustering, extracting working condition data labels in the class clusters corresponding to the normal working conditions, marking, and screening index parameters under the normal working conditions by using the evaluation index under the marked working condition data labels;
constructing a mixed kernel function by combining a Gaussian kernel function with an inverse cosine kernel function, setting a corresponding number of mixed kernel functions according to the number of marked index parameters, initializing weighted multi-kernel, acquiring multi-kernel weights through multi-kernel learning, fusing the index parameters under normal working conditions, and generating a fusion parameter set;
and constructing a normal operation model through the NAR dynamic neural network, optimizing the time delay order and hidden layer neurons of the normal operation model, training the normal operation model through the fusion parameter set, and outputting the normal operation model which meets the standard.
9. The system for evaluating the condition of an infrared therapeutic apparatus and predicting the remaining life according to claim 7, wherein the identifying of the abnormal condition of the infrared therapeutic apparatus is performed according to the residual data, specifically:
Acquiring training data for normalization processing, acquiring evaluation data corresponding to the training data through a normal operation model, acquiring reference residual error data according to the absolute value of the difference value between the training data and the evaluation data, and acquiring reference probability distribution representing normal operation of the infrared therapeutic apparatus according to the reference residual error data;
acquiring real-time operation parameters of the infrared therapeutic apparatus, importing the real-time operation parameters into the normal operation model to acquire real-time residual error data, and acquiring target probability distribution representing the current operation condition of the infrared therapeutic apparatus according to the real-time residual error data;
and calculating relative entropy according to the reference probability distribution and the target probability distribution, and when the relative entropy is larger than a preset threshold value, proving that the current running condition is different from the normal running condition, and the current infrared therapeutic instrument has potential faults and is an abnormal working condition.
10. The system for evaluating the condition of an infrared therapeutic apparatus and predicting the remaining life according to claim 7, wherein the health evaluation score corresponding to the abnormal condition is obtained according to the parameter deviation scoring system, specifically:
a score interval corresponding to the deviation score of the preset parameter is constructed, a weighted neural network is constructed to acquire nonlinear relations between different score intervals and target probability distribution of the current running condition of the infrared therapeutic instrument, and the deviation score is acquired by utilizing the nonlinear relations;
Generating initial weights of abnormal working conditions according to the deviation between the relative entropy of the reference probability distribution and the target probability distribution and the reference threshold value, and introducing a multi-head attention mechanism to acquire self-attention weights of real-time residual data corresponding to the target probability distribution;
and combining the initial weight and the self-attention weight to obtain the parameter weight corresponding to each index parameter, and constructing a parameter deviation scoring system according to the deviation score and combining the parameter weight and the reference value to generate a health condition assessment score corresponding to the current abnormal working condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311642160.1A CN117563144A (en) | 2023-12-04 | 2023-12-04 | Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311642160.1A CN117563144A (en) | 2023-12-04 | 2023-12-04 | Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117563144A true CN117563144A (en) | 2024-02-20 |
Family
ID=89886177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311642160.1A Pending CN117563144A (en) | 2023-12-04 | 2023-12-04 | Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117563144A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100191681A1 (en) * | 2009-01-23 | 2010-07-29 | Oxfordian, L.L.C. | Prognostics and health management method for aging systems |
| US20140182350A1 (en) * | 2013-01-03 | 2014-07-03 | Dexcom, Inc. | End of life detection for analyte sensors |
| JP2014149840A (en) * | 2014-03-12 | 2014-08-21 | Hitachi Ltd | Abnormality detection method and system therefor |
| CN112836920A (en) * | 2020-12-01 | 2021-05-25 | 北方联合电力有限责任公司乌拉特发电厂 | Coal electric unit energy efficiency state evaluation method and device and coal electric unit system |
| CN113221441A (en) * | 2020-12-24 | 2021-08-06 | 山东鲁能软件技术有限公司 | Method and device for health assessment of power plant equipment |
| CN113221442A (en) * | 2020-12-24 | 2021-08-06 | 山东鲁能软件技术有限公司 | Construction method and device of health assessment model of power plant equipment |
| CN113361186A (en) * | 2021-04-28 | 2021-09-07 | 山东大学 | Complete data-based wind turbine generator fault diagnosis method |
| CN113656910A (en) * | 2021-08-06 | 2021-11-16 | 电子科技大学 | Rolling bearing health index curve construction method based on AFF-AAKR fusion |
| CN114548195A (en) * | 2020-11-26 | 2022-05-27 | 中兴通讯股份有限公司 | Data annotation processing method and device, storage medium and electronic device |
| CN114742345A (en) * | 2022-02-28 | 2022-07-12 | 山东鲁软数字科技有限公司 | Power distribution equipment health evaluation method, system and device |
| US20220283576A1 (en) * | 2021-03-03 | 2022-09-08 | Aktiebolaget Skf | Automatic diagnosis method, system and storage medium for equipment |
| CN116990692A (en) * | 2023-09-28 | 2023-11-03 | 深圳康普盾科技股份有限公司 | Lithium battery health condition assessment and residual life prediction method and system |
-
2023
- 2023-12-04 CN CN202311642160.1A patent/CN117563144A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100191681A1 (en) * | 2009-01-23 | 2010-07-29 | Oxfordian, L.L.C. | Prognostics and health management method for aging systems |
| US20140182350A1 (en) * | 2013-01-03 | 2014-07-03 | Dexcom, Inc. | End of life detection for analyte sensors |
| JP2014149840A (en) * | 2014-03-12 | 2014-08-21 | Hitachi Ltd | Abnormality detection method and system therefor |
| CN114548195A (en) * | 2020-11-26 | 2022-05-27 | 中兴通讯股份有限公司 | Data annotation processing method and device, storage medium and electronic device |
| WO2022111284A1 (en) * | 2020-11-26 | 2022-06-02 | 中兴通讯股份有限公司 | Data labeling processing method and apparatus, and storage medium and electronic apparatus |
| CN112836920A (en) * | 2020-12-01 | 2021-05-25 | 北方联合电力有限责任公司乌拉特发电厂 | Coal electric unit energy efficiency state evaluation method and device and coal electric unit system |
| CN113221442A (en) * | 2020-12-24 | 2021-08-06 | 山东鲁能软件技术有限公司 | Construction method and device of health assessment model of power plant equipment |
| CN113221441A (en) * | 2020-12-24 | 2021-08-06 | 山东鲁能软件技术有限公司 | Method and device for health assessment of power plant equipment |
| US20220283576A1 (en) * | 2021-03-03 | 2022-09-08 | Aktiebolaget Skf | Automatic diagnosis method, system and storage medium for equipment |
| CN113361186A (en) * | 2021-04-28 | 2021-09-07 | 山东大学 | Complete data-based wind turbine generator fault diagnosis method |
| CN113656910A (en) * | 2021-08-06 | 2021-11-16 | 电子科技大学 | Rolling bearing health index curve construction method based on AFF-AAKR fusion |
| CN114742345A (en) * | 2022-02-28 | 2022-07-12 | 山东鲁软数字科技有限公司 | Power distribution equipment health evaluation method, system and device |
| CN116990692A (en) * | 2023-09-28 | 2023-11-03 | 深圳康普盾科技股份有限公司 | Lithium battery health condition assessment and residual life prediction method and system |
Non-Patent Citations (3)
| Title |
|---|
| 刘大伟;戴苏榕;: "PHM技术在液压伺服系统中的应用", 航空电子技术, no. 02, 15 June 2016 (2016-06-15) * |
| 曹现刚 等: "综采设备健康状态评估研究综述", 工矿自动化, 15 September 2023 (2023-09-15) * |
| 梁博渊;刘伟;杨欣桐;: "变压器健康状况评估与剩余寿命预测", 电网与清洁能源, no. 11, 25 November 2010 (2010-11-25) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bashir et al. | MV5: a clinical decision support framework for heart disease prediction using majority vote based classifier ensemble | |
| CN116386869B (en) | Disease critical degree assessment method based on multiple variables | |
| CN112017789B (en) | Triage data processing method, triage data processing device, triage data processing equipment and triage data processing medium | |
| CN106934199A (en) | A kind of psychological pressure appraisal procedure and device | |
| Pagán et al. | Grammatical evolutionary techniques for prompt migraine prediction | |
| CN114023449A (en) | Diabetes risk early warning method and system based on depth self-encoder | |
| CN113270197A (en) | Health prediction method, system and storage medium based on artificial intelligence | |
| Karimifard et al. | A robust method for diagnosis of morphological arrhythmias based on Hermitian model of higher-order statistics | |
| CN114403899B (en) | Depression detection device combining brain neuron spike potential and local field potential | |
| CN115668395A (en) | System and method for processing retinal signal data and identifying conditions | |
| Mortensen et al. | Multi-class stress detection through heart rate variability: A deep neural network based study | |
| Harikumar et al. | Genetic algorithm optimization of fuzzy outputs for classification of epilepsy risk levels from EEG signals | |
| Kim et al. | Electrocardiogram authentication method robust to dynamic morphological conditions | |
| CN117563144A (en) | Method and system for evaluating condition and predicting residual life of infrared therapeutic instrument | |
| Saif et al. | Feature engineering based performance analysis of ML and DL algorithms for Botnet attack detection in IoMT | |
| CN112057068A (en) | Epilepsia pathological data classification method and device and storage medium | |
| KR20210152254A (en) | Methods and apparatuses for diagnosing parkinson’s disease using speech data based on class probability output network | |
| CN116864121A (en) | Health risk screening system | |
| CN114093521B (en) | Random forest based method and system for estimating blood sugar by reconstructing homogenized samples | |
| He et al. | Unsupervised feature learning based on autoencoder for epileptic seizures prediction | |
| Sohankar et al. | Systematic analysis of liveness detection methods in biometrie security systems | |
| CN114707420A (en) | Credit fraud behavior identification method, device, equipment and storage medium | |
| Cai et al. | Local metric learning for EEG-based personal identification | |
| CN117476183B (en) | Construction system of autism children rehabilitation effect AI evaluation model | |
| Luo | Analysis and Research of Network Information Security Evaluation Model Based on Machine Learning Algorithm |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Guo Yongqiang Inventor after: Wang Qingmei Inventor after: Luo Xudong Inventor before: Luo Xudong |
|
| CB03 | Change of inventor or designer information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240429 Address after: 274712 Central Health Center, Liji Town, Yuncheng County, Heze City, Shandong Province Applicant after: Guo Yongqiang Country or region after: China Address before: 516200 Building A, Zhongxing Industrial Park, Tangxian Avenue East District, Xinxu Town, Huiyang District, Huizhou City, Guangdong Province Applicant before: Huizhou Lingsheng Medical Technology Co.,Ltd. Country or region before: China |