Clothing processing equipment fault prediction method and device, equipment and storage medium
By generating training data sets and using the hidden Markov model for fault prediction, the problem of difficulty in observing clothing processing equipment in the prior art is solved, and the efficiency of prediction of equipment failures and reliability experiments is improved.
Patent Information
- Application Number
- CN202311637831.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-30
- Publication Date
- 2025-05-30
AI Technical Summary
In existing reliability experiments, it is difficult to observe the laundry processing equipment continuously for a long time, which makes it difficult to know the specific situation in the event of a failure, which may lead to missed observations and the collection of key data.
By obtaining the running data of the same type of clothing processing equipment during each running cycle, a training data set is generated, and a fault prediction model is trained based on the implicit Markov model to predict whether a failure occurs in the next running cycle.
It realizes the prediction of failures of clothing processing equipment, reduces the need for long-term continuous observation, saves manpower and material resources, and improves the efficiency and accuracy of reliability experiments.
Smart Images
Figure CN120067846A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the technical field of electrical appliances, and particularly relates to a method, device, equipment and storage medium for predicting faults of clothing treatment equipment. Background Art
[0002] With the continuous improvement of production technology level and the continuous growth of user needs, users have higher and higher requirements for clothing treatment equipment. How to improve the reliability of the operation of clothing treatment equipment is an issue that electrical appliance manufacturers need to pay attention to. Usually during the product R & D process, R & D personnel will conduct reliability experiments on clothing treatment equipment, using various environmental test equipment to simulate the operating environment to verify whether it meets the expected quality goals in R & D, design, and manufacturing, so as to evaluate the overall product, determine the reliability life of the product, and improve the product.
[0003] In the existing reliability experiments, it is necessary to observe the washing machine before a possible failure occurs and collect relevant fault data in advance.
[0004] However, the reliability experiment cycle is usually long, and experimenters cannot continuously observe the machine beside the machine for a long time, making it difficult to know the specific situation when the clothing treatment equipment fails, which may lead to missing observations and collecting important experimental phenomena and key data. Summary of the Invention
[0005] In view of the above problems, this application provides a method, device, equipment and storage medium for predicting faults of clothing treatment equipment.
[0006] In a first aspect, this application provides a method for predicting faults of clothing treatment equipment, the method comprising:
[0007] Obtain the operation data of the same type of clothing treatment equipment in each operation cycle, and obtain the operation data set corresponding to the type according to the operation data, wherein the operation data includes the setting data and measurement data in the current operation cycle;
[0008] Obtain the state classification corresponding to the operation data of different operation cycles in the operation data set, and generate a training data set according to the operation data of different operation cycles and their corresponding state classifications;
[0009] Train a fault prediction model according to the training data set to obtain a trained fault prediction model, wherein the fault prediction model is a Hidden Markov Model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
[0010] Optionally, the training the fault prediction model according to the training data set includes:
[0011] Classify different states as the hidden states of the hidden Markov model, and obtain the initial probability matrix of the hidden states and the transition probability matrix of the hidden states;
[0012] Take the operation data in each operation cycle as the observation states of the hidden Markov model, and obtain the generation probability matrix of the observation states;
[0013] Construct an initial fault prediction model according to the initial probability matrix of the hidden states, the transition probability matrix of the hidden states, and the generation probability matrix of the observation states;
[0014] Train the initial fault prediction model according to the training data set to obtain a trained fault prediction model.
[0015] Optionally, the constructing an initial fault prediction model according to the initial probability matrix of the hidden states, the transition probability matrix of the hidden states, and the generation probability matrix of the observation states includes:
[0016] Model the operation data x(N) and the state classification y(N) of the Nth operation cycle according to the initial probability matrix of the hidden states, the transition probability matrix of the hidden states, and the generation probability matrix of the observation states, to obtain an initial fault prediction model; where N is a positive integer, and the state classification y(N) of the Nth operation cycle is the hidden variable of the operation data x(N), and y(N) depends on the state classification y(N - 1) of the (N - 1)th operation cycle.
[0017] Optionally, the obtaining the operation data set corresponding to the type according to the operation data includes:
[0018] Preprocess the operation data, where the preprocessing includes at least one of missing value processing, outlier cleaning, or duplicate value cleaning, to obtain preprocessed data;
[0019] Perform standardization processing on the preprocessed data to obtain standardized data, and use the standardized data as the operation data set corresponding to the type.
[0020] Optionally, the setting data includes at least one of the operation program information, the set water temperature, or the set water level of the laundry treatment device in the current operation cycle; the measurement data includes at least one of the load weight, the amount of laundry treatment agent put in, the motor frequency, the motor temperature, the operation duration, or the operation vibration signal of the laundry treatment device in the current operation cycle.
[0021] Optionally, the obtaining the state classification corresponding to the operation data of different operation cycles in the operation data set includes:
[0022] Input the operation data of different operation cycles in the operation dataset into the state classification model respectively, and obtain the state classification output by the state classification model as the state classification corresponding to the input operation data.
[0023] In a second aspect, the present application provides a method for predicting faults of a clothing treatment device, and the method includes:
[0024] Obtain the operation data of the clothing treatment device in the Nth operation cycle and the previous M - 1 operation cycles, and the state classification of the (N - M)th operation cycle; where N is a positive integer, and M is less than or equal to N - 1;
[0025] Obtain the type corresponding to the clothing treatment device, and input the operation data in the Nth operation cycle and the state classification of the (N - 1)th operation cycle into the fault prediction model corresponding to the type; where the fault prediction model is a model trained by the method described in the first aspect;
[0026] Obtain the state classification of the Nth operation cycle output by the fault prediction model;
[0027] If the state classification is a fault type, send a fault alarm to the user, and store the operation data in the Nth operation cycle in a preset storage location.
[0028] In a third aspect, the present application provides a device for predicting faults of a clothing treatment device, and the device includes:
[0029] An acquisition module, configured to acquire the operation data of the clothing treatment devices of the same type in each operation cycle, and obtain the operation dataset corresponding to the type according to the operation data, where the operation data includes the set data and the measured data in the current operation cycle;
[0030] A judgment module, configured to obtain the state classification corresponding to the operation data of different operation cycles in the operation dataset, and generate a training dataset according to the operation data of different operation cycles and their corresponding state classifications;
[0031] A training module, configured to train a fault prediction model according to the training dataset, and obtain a trained fault prediction model, where the fault prediction model is a hidden Markov model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
[0032] In a fourth aspect, the present application provides a device for predicting faults of a clothing treatment device, including: a memory and at least one processor;
[0033] The memory is used to store a computer program;
[0034] The at least one processor is configured to run a computer program stored in the memory to implement the methods described in the first aspect and the second aspect.
[0035] In a fifth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored;
[0036] When the computer program is executed by a processor, the methods described in the first aspect and the second aspect are implemented.
[0037] The method for predicting faults of a clothing treatment device provided by the embodiments of the present application obtains the operation data of clothing treatment devices of the same type in each operation cycle, and obtains an operation data set corresponding to the type according to the operation data, where the operation data includes setting data and measurement data in the current operation cycle; obtains the state classifications corresponding to the operation data of different operation cycles in the operation data set, and generates a training data set according to the operation data of different operation cycles and their corresponding state classifications; trains a fault prediction model according to the training data set to obtain a trained fault prediction model, where the fault prediction model is a hidden Markov model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle, so that experimenters can collect and observe the operation data and fault phenomena of the device in time according to the prediction results, eliminating the need for a large amount of manpower and material resources spent on long-term continuous observation and recording of the experiment, and providing strong support for the smooth and efficient progress of the reliability experiment of the clothing treatment device. Description of the Drawings
[0038] The drawings here are incorporated into the description and form a part of this description, showing embodiments consistent with the present application, and are used together with the description to explain the principles of the present application.
[0039] Figure 1 Schematic flow chart of the method for predicting faults of a clothing treatment device provided by the present application Figure 1 ;
[0040] Figure 2 Schematic flow chart of the method for predicting faults of a clothing treatment device provided by the present application Figure 2 ;
[0041] Figure 3 Schematic structural diagram of a device for predicting faults of a clothing treatment device provided by the present application;
[0042] Figure 4 Schematic structural diagram of a device for predicting faults of a clothing treatment device provided by the present application.
[0043] Through the above-mentioned accompanying drawings, specific embodiments of the present application have been shown, and will be described in more detail hereinafter. These drawings and the written description are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Description of the Embodiments
[0044] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the accompanying drawings in the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without making creative efforts shall fall within the scope of protection of the present application.
[0045] In the description of the present invention, the terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned accompanying drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein.
[0046] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific manner.
[0047] With the continuous improvement of production technology levels and the continuous growth of user demands, users' requirements for clothing treatment devices are also getting higher and higher. How to improve the reliability of the operation of clothing treatment devices is an issue that electrical appliance manufacturers need to pay attention to. Usually during the product R & D process, R & D personnel will conduct reliability experiments on clothing treatment devices, using various environmental test equipment to simulate the operating environment to verify whether it meets the expected quality goals in R & D, design, and manufacturing, so as to evaluate the overall product, determine the reliability life of the product, and improve the product.
[0048] In the existing reliability experiments, it is necessary to observe the washing machine before a possible failure occurs and collect relevant fault data in advance.
[0049] However, the reliability experiment cycle is usually long, and the experimenters cannot continuously observe the machine beside the machine for a long time, and it is difficult to know the specific situation when the clothing treatment device fails, which may lead to missing observations and collecting important experimental phenomena and key data.
[0050] In view of the above problems, the present application proposes the following technical concept: collect the operation data of a large number of clothing processing devices at different operation durations, and divide the collected operation data according to operation cycles of equal duration; perform state classification marking on the operation data within different operation cycles, and the state classification may include a normal state and a fault state; establish a hidden Markov model based on the marked operation data, and use the operation data of different operation cycles after marking to train a fault prediction model to obtain a trained fault prediction model. This fault prediction model can predict the state classification in a future period of time. Experimental personnel can collect and observe the operation data and fault phenomena of the device in a timely manner according to the prediction results, so that the reliability experiment can obtain more effective experimental records, and further make the experimental results more accurate and reference-worthy, and save a large amount of manpower and material resources.
[0051] Figure 1 Flow schematic of the clothing processing device fault prediction method provided by the embodiment of the present application Figure 1 ; As Figure 1 shown, the clothing processing device fault prediction method provided by the embodiment of the present application includes:
[0052] S101: Obtain the operation data of the same type of clothing processing device in each operation cycle, and obtain the operation data set corresponding to the type according to the operation data.
[0053] Among them, the operation data includes the setting data and measurement data in the current operation cycle.
[0054] Specifically, the data collection device reads the setting information of the clothing processing program according to the clothing processing program executed in the current operation cycle of the clothing processing device, and determines the setting data of the clothing processing device in the current operation cycle according to the setting information; collects relevant physical signals during the operation of the clothing processing device through a measurement device pre-set on the clothing processing device, and determines the corresponding measurement data according to the physical signals; after the data collection device obtains the operation data of the same type of clothing processing device in each operation cycle, performs data processing on the operation data according to a preset algorithm, and stores the operation data with a unified format after processing to a preset storage location as the operation data set corresponding to the clothing processing device of the type.
[0055] S102: Obtain the state classification corresponding to the operation data of different operation cycles in the operation data set, and generate a training data set according to the operation data of different operation cycles and their corresponding state classifications.
[0056] Among them, the state classification corresponding to the operation data of different operation cycles is used to indicate whether the clothing processing device is likely to have a fault from the current cycle to a future period of time.
[0057] Specifically, the data recognition device determines the status classifications corresponding to the operation data in different operation cycles in the operation data set according to a preset status classification method, and stores the operation data in different operation cycles and their corresponding status classification labels in corresponding positions as a training data set. The method for obtaining the status classification can be, for example, determining the status classification corresponding to the operation data according to the fault records in different operation cycles; it can also be inputting the operation data into a pre-trained status classification model and obtaining the status classification output by the model as the status classification corresponding to the operation data. The embodiments of the present application do not limit the specific method for determining the status classification corresponding to the operation data.
[0058] S103: Train a fault prediction model according to the training data set to obtain a trained fault prediction model.
[0059] Among them, the fault prediction model is a hidden Markov model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
[0060] Specifically, the model training device first models the hidden Markov model to determine the initial parameters of the model, then trains the hidden Markov model according to the training data set, and adjusts the model-related parameters according to the training metrics until the training metrics and the validation metrics meet the design requirements of the model, ends the training, and stores the trained fault prediction model. The embodiments of the present application do not limit the specific implementation manner, training method, and corresponding metrics of the hidden Markov model.
[0061] The clothing processing equipment fault prediction method provided by the embodiments of the present application collects the operation data of the same type of clothing processing equipment in each operation cycle, adds labels to the operation data in each operation cycle after preprocessing according to their corresponding status classifications, generates a training data set from different operation data and their corresponding labels, and trains a fault prediction model based on a hidden Markov model according to the training data set to obtain a trained fault prediction model. This fault prediction can predict the status classification of the clothing processing equipment in the next cycle. Experimenters can collect and observe the operation data and fault phenomena of the equipment in a timely manner according to the prediction results, saving a large amount of manpower and material resources spent on long-term continuous observation and recording of the experiment, and providing strong support for the smooth and efficient progress of the reliability experiment of the clothing processing equipment.
[0062] Figure 2 It is a flow schematic diagram of the clothing processing equipment fault prediction method provided by the embodiments of the present application Figure 2 ; as Figure 2 shown, the clothing processing equipment fault prediction method provided by the embodiments of the present application includes:
[0063] S201: Obtain the operation data of clothes treatment devices of the same type in each operation cycle.
[0064] Among them, the operation data includes the setting data and measurement data in the current operation cycle.
[0065] Optionally, the setting data includes at least one of the operation program information, set water temperature, or set water level of the clothes treatment device in the current operation cycle; the measurement data includes at least one of the load weight, clothes treatment agent dosage, motor frequency, motor temperature, operation duration, or operation vibration signal of the clothes treatment device in the current operation cycle.
[0066] The implementation manner of step S201 is similar to that of step S202, and will not be elaborated herein in this embodiment.
[0067] S202: Preprocess the operation data to obtain preprocessed data.
[0068] Among them, the preprocessing includes at least one of missing value processing, outlier cleaning, or duplicate value cleaning.
[0069] Specifically, the data processing device identifies the missing values, outliers, and duplicate values in the dataset according to a preset data filtering algorithm; the data filtering algorithm can, for example, be to judge the relationship between the value of a certain data item and multiple preset value ranges, or to identify the data change rule and judge whether the value of a certain data item conforms to the data change rule, etc.; after identifying that the data has missing values, the samples containing missing values are excluded or imputed according to the type of missing values; after identifying that the data has outliers or duplicate values, the samples containing outliers or duplicate values are excluded or replaced, etc.; the data after the preprocessing is completed is used as the preprocessed data.
[0070] S203: Standardize the preprocessed data to obtain standardized data, and use the standardized data as the operation dataset corresponding to the type.
[0071] Specifically, the data processing device constructs features for each data item in the preprocessed data according to a preset data standardization algorithm, selects and extracts the features of the constructed data, and performs grading and quantization processing on the extracted values. Finally, the preprocessed data corresponding to each operation cycle in the preprocessed data is converted and merged into data vectors with the same format, and the set of data vectors corresponding to different operation cycles is used as the operation dataset corresponding to the type.
[0072] S204: Input the operation data of different operation cycles in the operation dataset into the state classification model respectively, and obtain the state classification output by the state classification model as the state classification corresponding to the input operation data.
[0073] Among them, the state classification can be, for example, a healthy state and a failure state.
[0074] Specifically, the data processing device inputs the operation data of different operation cycles in the operation dataset into the state classification model respectively. The state classification model can be, for example, a device state classification model based on historical data, a fault diagnosis model of an expert system based on an inference engine, etc., and obtains the state classification output by the state classification model as the state classification corresponding to the input operation data.
[0075] S205: Generate a training dataset according to the operation data of different operation cycles and their corresponding state classifications.
[0076] The implementation manner of step S205 is similar to that of step S102, and will not be elaborated here in this embodiment.
[0077] Optionally, the training dataset can also be divided into a training set, a validation set, and a test set according to a ratio for the corresponding processes in the subsequent model training steps.
[0078] S206: Use different state classifications as the hidden states of the hidden Markov model, and obtain the initial probability matrix of the hidden states and the transition probability matrix of the hidden states.
[0079] Specifically, the data processing device defines each different state classification of the type of clothing processing device as the hidden state of the hidden Markov model, and determines the initial probability distribution of each hidden state of the hidden Markov model and the transition probability between each hidden state according to the occurrence frequency of different state classifications in the training dataset and the total number of samples, and determines the initial probability matrix of the hidden states and the transition probability matrix of the hidden states according to the initial probability and the transition probability.
[0080] S207: Use the operation data in each operation cycle as the observed states of the hidden Markov model, and obtain the generation probability matrix of the observed states.
[0081] Specifically, the data processing device defines the operation data in each operation cycle of the type of clothing processing device as different observed states according to a preset observed classification index, and determines the generation probability of different observed states according to the occurrence frequency of different observed states in the dataset and the total number of samples, and determines the generation probability matrix of the observed states according to the generation probability of different observed states.
[0082] S208: Construct an initial fault prediction model based on the initial probability matrix of the hidden state, the transition probability matrix of the hidden state, and the generation probability matrix of the observation state.
[0083] Optionally, a specific method for constructing the initial fault prediction model may be, for example:
[0084] Based on the initial probability matrix of the hidden state, the transition probability matrix of the hidden state, and the generation probability matrix of the observation state, model the operation data x(N) and the state classification y(N) of the Nth operation cycle to obtain an initial fault prediction model.
[0085] Wherein, the state classification y(N) of the Nth operation cycle is a hidden variable of the operation data x(N), and y(N) depends on the state classification y(N - 1) of the (N - 1)th operation cycle; the fault prediction model may be, for example, a self-recurrent neural network model.
[0086] The input of the model is the operation data within the Nth operation cycle and the state classification of the (N - 1)th operation cycle, and the output value is the state classification of the Nth operation cycle. The self-recurrent neural network model can determine, according to the operation data within the Nth operation cycle and the state classification of the (N - 1)th operation cycle, the state classification of the Nth operation cycle that is most likely to make the operation data within the Nth operation cycle the currently acquired value when determining the state classification of the (N - 1)th operation cycle.
[0087] S209: Train the initial fault prediction model according to the training data set to obtain a trained fault prediction model.
[0088] The implementation manner of step S209 is similar to that of step S103, and will not be elaborated herein in this embodiment.
[0089] Optionally, to ensure the accuracy of the prediction result of the fault prediction model, the training data can be updated at a preset time interval, and the updated training data can be used to iteratively train the fault prediction model to update the fault prediction model.
[0090] S210: Obtain the operation data of the clothing processing device in the Nth operation cycle and the previous M - 1 operation cycles, and the state classification of the (N - M)th operation cycle.
[0091] Wherein, N and M are positive integers, and M is less than or equal to N - 1.
[0092] Specifically, the data processing device reads the operation data of the laundry processing device recorded in advance for a total of M operation cycles in the Nth operation cycle and the previous cycles, and obtains the status classification of the (N - M)th operation cycle, where the status classification of the (N - M)th operation cycle is a fault classification determined to conform to the actual situation of the (N - M)th operation cycle.
[0093] S211: Obtain the type corresponding to the laundry processing device, and input the operation data in the Nth operation cycle and the status classification of the (N - M)th operation cycle into the fault prediction model corresponding to the type.
[0094] Among them, the fault prediction model is a model trained by the method described in steps S201 to S209.
[0095] S212: Obtain the status classification of the Nth operation cycle output by the fault prediction model.
[0096] Specifically, the data processing device inputs the operation data in the Nth operation cycle and the status classification of the (N - M)th operation cycle into the fault prediction model corresponding to the type. The model first outputs the status classification of the (N - M + 1)th operation cycle according to the status classification of the (N - M)th operation cycle and the operation data of the (N - M + 1)th operation cycle, enters a self-loop, and outputs the status classification of the (N - M + 2)th operation cycle according to the status classification of the (N - M + 1)th operation cycle and the operation data of the (N - M + 2)th operation cycle... and so on, until the status classification of the Nth operation cycle is output as the fault prediction result of the laundry processing device in the Nth operation cycle to a future period of time.
[0097] For example, the data processing device obtains the operation data of a certain model of washing machine in the 2nd to 5th operation cycles, and the status classification of the 1st operation cycle, for example, is a healthy state; and inputs the above data into the fault prediction model. The fault prediction model determines the status classification of the 2nd operation cycle according to the status classification of the 1st operation cycle and the operation data of the 2nd operation cycle; then determines the status classification of the 2nd operation cycle according to the status classification of the 2nd operation cycle and the operation data of the 3rd operation cycle... and so on, until the status classification of the 5th operation cycle is determined.
[0098] S213: If the status classification is a fault type, send a fault alarm to the user, and store the operation data in the Nth operation cycle in a preset storage location.
[0099] Specifically, when the data processing device determines that the status classification is a fault type, it controls the communication device to send a fault alarm to the user to prompt the user to observe the fault phenomenon, and stores the operation data in the Nth operation cycle at a preset storage location as the data for the reliability experiment record.
[0100] The clothing processing equipment fault prediction method provided by the embodiment of the present application constructs a neural network hidden Markov model according to historical data and trains the model. The hidden state of the model forms a memory in the form of self-loop. The status classification of each operation cycle depends on its past state. Through self-loop, a memory of the status classification of each operation cycle is formed and combined with the current operation data to predict the status classification from the current operation cycle to a period of time in the future, avoiding the deviation caused by relying solely on the current operation data for fault prediction; it can be understood that the clothing processing equipment fault prediction method provided by the embodiment of the present application can be applied not only to the reliability experiment scenario in the research and development and design stage, but also to the application scenario of the clothing processing equipment in use to predict the possible faults of the clothing processing equipment in the use scenario and prompt the user or maintenance personnel to take corresponding measures, so as to reduce the failure rate and improve the maintenance efficiency, and improve the user satisfaction of the clothing processing equipment.
[0101] Figure 3 It is a schematic structural diagram of a clothing processing equipment fault prediction device provided by an embodiment of the present application; as Figure 3 shown, the present application provides a clothing processing equipment fault prediction device. The clothing processing equipment fault prediction device 300 includes:
[0102] An acquisition module 301, configured to acquire the operation data of the same type of clothing processing equipment in each operation cycle, and obtain an operation data set corresponding to the type according to the operation data, where the operation data includes the setting data and measurement data in the current operation cycle;
[0103] A judgment module 302, configured to acquire the status classification corresponding to the operation data of different operation cycles in the operation data set, and generate a training data set according to the operation data of different operation cycles and their corresponding status classifications;
[0104] A training module 303, configured to train a fault prediction model according to the training data set to obtain a trained fault prediction model, where the fault prediction model is a hidden Markov model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
[0105] Optionally, the training module 303 is specifically configured to use different status classifications as the hidden states of the hidden Markov model, and obtain the initial probability matrix of the hidden states and the transition probability matrix of the hidden states; use the operation data in each operation cycle as the observation states of the hidden Markov model, and obtain the generation probability matrix of the observation states; construct an initial fault prediction model according to the initial probability matrix of the hidden states, the transition probability matrix of the hidden states, and the generation probability matrix of the observation states; and train the initial fault prediction model according to the training data set to obtain a trained fault prediction model.
[0106] Optionally, the training module 303 is specifically configured to model the operation data x(N) and the status classification y(N) of the Nth operation cycle according to the initial probability matrix of the hidden states, the transition probability matrix of the hidden states, and the generation probability matrix of the observation states, to obtain an initial fault prediction model; where N is a positive integer, and the status classification y(N) of the Nth operation cycle is a hidden variable of the operation data x(N), and y(N) depends on the status classification y(N-1) of the (N-1)th operation cycle.
[0107] Optionally, the acquisition module 301 is specifically configured to preprocess the operation data, where the preprocessing includes at least one of missing value processing, outlier cleaning, or duplicate value cleaning, to obtain preprocessed data; perform standardization processing on the preprocessed data to obtain standardized data, and use the standardized data as the operation data set corresponding to the type.
[0108] Optionally, the setting data includes at least one of the operation program information, the set water temperature, or the set water level of the laundry treatment device in the current operation cycle; the measurement data includes at least one of the load weight, the laundry treatment agent dosage, the motor frequency, the motor temperature, the operation duration, or the operation vibration signal of the laundry treatment device in the current operation cycle.
[0109] Optionally, the judgment module 302 is specifically configured to input the operation data of different operation cycles in the operation data set into the status classification model respectively, and obtain the status classification output by the status classification model as the status classification corresponding to the input operation data.
[0110] Optionally, the acquisition module 301 is further configured to acquire the operation data of the laundry treatment device in the Nth operation cycle and the previous M-1 operation cycles, and the status classification of the (N-M)th operation cycle; where N is a positive integer, and M is less than or equal to N-1;
[0111] The determination module 302 is further configured to obtain the type corresponding to the clothing treatment device, and input the operation data in the Nth operation cycle and the status classification in the (N-1)th operation cycle into the fault prediction model corresponding to the type; wherein, the fault prediction model is a model trained by the method embodiment corresponding to Figure 1 - Figure 2 ; obtain the status classification of the Nth operation cycle output by the fault prediction model; when the status classification is a fault type, send a fault alarm to the user, and store the operation data in the Nth operation cycle at a preset storage location.
[0112] Figure 4 FIG. is a schematic structural diagram of a clothing treatment device fault prediction device provided by an embodiment of the present application, as Figure 4 shown. The clothing treatment device fault prediction device 400 includes: at least one processor 401 and a memory 402; in addition, the clothing treatment device fault prediction device 400 may further have a communication interface 404 for receiving and sending instructions.
[0113] Wherein, the processor 401, the memory 402 and the communication interface 404 are connected through a bus 403;
[0114] Wherein, the computer program is stored in the memory 402 and is configured to be executed by the processor 401 to implement the clothing treatment device fault prediction method provided by any embodiment corresponding to the present application Figure 1 - Figure 2 ;
[0115] Figure 4 The clothing treatment device fault prediction device shown in the embodiment can be used to execute the technical solutions of the above method embodiments, and its implementation principles and technical effects are similar, and will not be described in detail here.
[0116] In addition, this embodiment also provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the clothing treatment device fault prediction method of the above embodiment.
[0117] In several embodiments provided by the present invention, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection of the device or unit may be in an electrical, mechanical or other form.
[0118] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0119] In addition, the functional units in each embodiment of the present invention may be integrated into one processing unit, may exist separately as individual physical units, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of a combination of hardware and software functional units.
[0120] The above-mentioned integrated units implemented in the form of software functional units can be stored in a computer-readable storage medium. The above-mentioned software functional units stored in a storage medium include several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute some steps of the methods in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.
[0121] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the device described above can refer to the corresponding process in the foregoing method embodiments and will not be elaborated here.
[0122] So far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.
Claims
1. A method for predicting faults in a laundry treatment device, characterized in that, the method comprises: obtaining the operation data of laundry treatment devices of the same type in each operation cycle, and obtaining an operation data set corresponding to the type according to the operation data, wherein the operation data includes setting data and measurement data in the current operation cycle; obtaining the state classifications corresponding to the operation data of different operation cycles in the operation data set, and generating a training data set according to the operation data of different operation cycles and their corresponding state classifications; training a fault prediction model according to the training data set to obtain a trained fault prediction model, wherein the fault prediction model is a Hidden Markov Model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
2. The method according to claim 1, characterized in that, the training of the fault prediction model according to the training data set includes: taking different state classifications as the hidden states of the Hidden Markov Model, and obtaining a hidden state initial probability matrix and a hidden state transition probability matrix; taking the operation data in each operation cycle as the observation states of the Hidden Markov Model, and obtaining an observation state generation probability matrix; constructing an initial fault prediction model according to the hidden state initial probability matrix, the hidden state transition probability matrix and the observation state generation probability matrix; training the initial fault prediction model according to the training data set to obtain a trained fault prediction model.
3. The method according to claim 2, characterized in that, the constructing of the initial fault prediction model according to the hidden state initial probability matrix, the hidden state transition probability matrix and the observation state generation probability matrix includes: modeling the operation data x(N) and the state classification y(N) of the Nth operation cycle according to the hidden state initial probability matrix, the hidden state transition probability matrix and the observation state generation probability matrix to obtain an initial fault prediction model; wherein N is a positive integer, and the state classification y(N) of the Nth operation cycle is a hidden variable of the operation data x(N), and y(N) depends on the state classification y(N - 1) of the (N - 1)th operation cycle.
4. The method according to claim 1, characterized in that, the obtaining of the operation data set corresponding to the type according to the operation data includes: performing preprocessing on the operation data, the preprocessing including at least one of missing value processing, outlier cleaning or duplicate value cleaning, to obtain preprocessed data; performing standardization processing on the preprocessed data to obtain standardized data, and taking the standardized data as the operation data set corresponding to the type.
5. The method according to claim 1, characterized in that, the setting data includes at least one of the operation program information, the set water temperature or the set water level of the laundry treatment device in the current operation cycle; the measurement data includes at least one of the load weight, the amount of laundry treatment agent dispensed, the motor frequency, the motor temperature, the operation duration or the operation vibration signal of the laundry treatment device in the current operation cycle.
6. The method according to claim 1, wherein, obtaining the status classifications corresponding to the operation data of different operation cycles in the operation dataset includes: inputting the operation data of different operation cycles in the operation dataset into a status classification model respectively, and obtaining the status classification output by the status classification model as the status classification corresponding to the input operation data.
7. A method for predicting faults of a laundry treatment device, wherein, the method includes: obtaining the operation data of the laundry treatment device in the Nth operation cycle and the previous M - 1 operation cycles and the status classification of the (N - M)th operation cycle; wherein, N is a positive integer, and M is less than or equal to N - 1; obtaining the type corresponding to the laundry treatment device, and inputting the operation data in the Nth operation cycle and the status classification in the (N - 1)th operation cycle into the fault prediction model corresponding to the type; wherein, the fault prediction model is a model trained by the method according to any one of claims 1 to 6; obtaining the status classification of the Nth operation cycle output by the fault prediction model; if the status classification is a fault type, sending a fault alarm to the user, and storing the operation data in the Nth operation cycle at a preset storage location.
8. A device for predicting faults of a laundry treatment device, wherein, the device includes: an obtaining module, configured to obtain the operation data of the laundry treatment devices of the same type in each operation cycle, and obtain the operation dataset corresponding to the type according to the operation data, wherein the operation data includes the set data and the measured data in the current operation cycle; a judging module, configured to obtain the status classifications corresponding to the operation data of different operation cycles in the operation dataset, and generate a training dataset according to the operation data of different operation cycles and their corresponding status classifications; a training module, configured to train a fault prediction model according to the training dataset, and obtain a trained fault prediction model, wherein the fault prediction model is a hidden Markov model, and the fault prediction model is used to predict whether a fault will occur in the next operation cycle.
9. A device for predicting faults of a laundry treatment device, wherein, it includes: a memory and at least one processor; the memory is used for storing a computer program; the at least one processor is used for running the computer program stored in the memory to implement the method according to any one of claims 1 - 7.
10. A computer - readable storage medium, wherein, a computer program is stored thereon; when the computer program is executed by a processor, it implements the method according to any one of claims 1 - 7.