Elevator fault diagnosis method and device based on machine learning, medium and equipment

Abstract

The invention provides an elevator fault diagnosis method and device based on machine learning, a medium and equipment, and relates to the technical field of elevator fault diagnos.The elevator fault diagnosis method comprises the steps that firstly, time sequence data, environment data and operation and maintenance record data in historical operation of an elevator are obtained, and time sequence features and environment features are extracted; the operation and maintenance record data are evaluated based on an expert scoring method, and the elevator fault probability reflecting the elevator operation quality is generated; then, constructing and training a deep learning network model by using the features and the fault probabilities as labels; and finally, current time sequence data and environment data of the elevator are obtained in real time, corresponding features are extracted and input into the trained model, the current elevator fault probability is obtained, the current running state of the elevator is judged by comparing the current elevator fault probability with a preset fault threshold value, and therefore efficient and accurate fault diagnosis and early warning are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator fault diagnosis, and in particular to an elevator fault diagnosis method based on machine learning. Background Art

[0002] As an indispensable vertical transportation tool in modern buildings, the safety and reliability of elevators directly affect people's daily life and work efficiency. With the acceleration of urbanization, the frequency of elevator use continues to increase, and the resulting elevator failure problem has become increasingly prominent. In recent years, with the development of sensor technology, data acquisition technology, and deep learning algorithms, data-driven elevator fault monitoring and diagnosis methods have gradually become a research hotspot.

[0003] In the current elevator industry, fault diagnosis and predictive maintenance technologies generally rely on traditional experience-based methods, usually focusing on a single type of data, such as operating parameters or historical fault records. This approach is inadequate when faced with complex elevator systems and cannot accurately capture the multi-dimensional information of fault occurrence.

[0004] The prior art has the following deficiencies:

[0005] Existing elevator fault diagnosis technologies generally rely on traditional periodic inspections and manual inspections, which have several obvious shortcomings. First, traditional methods often make it difficult to monitor the operating status of elevators in real time, resulting in potential faults that cannot be discovered in time, which may lead to more serious safety accidents. In addition, due to the subjectivity and uncertainty of manual inspections, there may be deviations in the judgment of faults, increasing the probability of misjudgment and missed fault detection. In addition, the efficiency of the use of operation and maintenance record data is also low, and there is a lack of effective feature extraction and analysis methods, which cannot fully utilize historical data for fault prediction. In addition, due to the lack of effective data fusion and analysis methods, the early warning system for elevator faults often has the problem of slow response, resulting in extended maintenance time after the fault occurs, affecting the safety and experience of users. Existing technologies are also obviously insufficient in real-time monitoring and dynamic adaptability, and cannot meet the efficient and intelligent requirements for fault diagnosis during the operation of modern elevators.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide an elevator fault diagnosis method, device, medium and equipment based on machine learning to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides an elevator fault diagnosis method based on machine learning, and the specific steps include:

[0009] Step 1: Acquire multiple sets of time series data, environmental data, and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current, and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator;

[0010] Step 2: Extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator failure probability that reflects the quality of elevator operation;

[0011] Step 3: Build an elevator fault model based on the deep learning network, take the time series features and environmental features as input, and use the corresponding elevator fault probabilities as labels to train the elevator fault model;

[0012] Step 4: Obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features. Input the time series features and environmental features into the trained elevator fault model to obtain the current elevator failure probability. Compare the elevator failure probability with the fault threshold to determine the current elevator status.

[0013] Furthermore, the specific logic for obtaining time series data, environment data, and operation and maintenance record data is as follows:

[0014] An acceleration sensor is installed at the bottom of the elevator car in the elevator shaft to obtain the acceleration data of the elevator, a speed sensor is installed on the motor shaft to obtain the running speed data of the elevator, a current sensor is installed at the power input line of the motor to obtain the current data of the elevator motor, a pressure sensor is installed at the bottom of the elevator car to obtain the load data of the elevator, the collected elevator acceleration, running speed, current data and load data are standardized, a temperature sensor and a humidity sensor are installed on the top of the elevator car to obtain the temperature and humidity data in the elevator, and the maintenance record of the elevator is derived from the operation and maintenance record data.

[0015] Furthermore, the specific logic for extracting time series features, environmental features, and operation and maintenance features is as follows:

[0016] The mean and standard deviation of the elevator's acceleration, running speed, current and load data are used as time series features, and the formula for calculating the mean acceleration is:

[0017]

[0018] Among them, a iis the acceleration of the elevator at the i-th collection time, i is the index of the collection time, and i = 1, 2, ..., N, N is the number of collection times, is the mean acceleration;

[0019] The running speed, current and load of the elevator at the i-th acquisition moment are calibrated as v i ,I i , P i , and the average running speed is obtained in the same way Current mean Load Average

[0020] The calculation of the standard deviation of acceleration is based on the formula:

[0021]

[0022] Among them, σ a is the acceleration standard deviation, and similarly, the running speed standard deviation σ v , current standard deviation σ I , load standard deviation σ P ; Taking the average temperature and humidity of the elevator as the environmental characteristics, the formula for calculating the average temperature is:

[0023]

[0024] Among them, T i is the temperature of the elevator at the i-th collection time, is the mean temperature;

[0025] The humidity of the elevator at the i-th collection time is calibrated as H i , and the mean humidity is obtained in the same way

[0026] Furthermore, the specific logic for constructing the elevator failure model is:

[0027] Based on the deep learning network model, the mean and standard deviation of the time series data are extracted as time series features, and the mean of the environmental data is used as the environmental features. The time series features and the environmental features are concatenated to form a feature vector, which is input into the elevator fault model. The elevator failure probability generated by the expert scoring method is used as the label, and the output layer uses the softmax function to obtain the elevator failure probability. The accuracy of the current model prediction is judged by the precision rate. When the precision rate reaches 0.9, it means that the elevator fault model training is completed.

[0028] Furthermore, the specific logic for determining the current elevator status is:

[0029] Collect the time series data and environmental data of the current elevator, calculate the mean and standard deviation of the time series data, and the mean of the environmental data as the time series features and environmental features respectively, input the time series features and environmental features into the trained elevator fault model, obtain the elevator failure probability in the current state, compare the elevator failure probability predicted by the model with the preset fault threshold, if the elevator failure probability predicted by the model is greater than the fault threshold, judge that the current elevator has a fault, if the elevator failure probability predicted by the model is lower than the fault prediction, judge that the current elevator is in a safe state.

[0030] The present invention further provides an elevator fault diagnosis device based on machine learning, wherein the elevator fault diagnosis device is used to implement the above-mentioned elevator fault diagnosis method, comprising:

[0031] A data acquisition module is used to obtain multiple sets of time series data, environmental data and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator;

[0032] The fault probability acquisition module is used to extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator fault probability that reflects the quality of elevator operation;

[0033] The fault model building module is used to build an elevator fault model based on a deep learning network, taking time series features and environmental features as inputs and the corresponding elevator fault probabilities as labels to train the elevator fault model;

[0034] The real-time fault diagnosis module is used to obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features, input the time series features and environmental features into the trained elevator fault model, obtain the current elevator failure probability, compare the elevator failure probability with the fault threshold, and judge the current elevator status.

[0035] The present invention further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the elevator fault diagnosis method based on machine learning as described above.

[0036] The present invention further provides a device, comprising a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, the above-mentioned elevator fault diagnosis method based on machine learning can be implemented.

[0037] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0038] This method solves the problems of insufficient accuracy and timeliness in traditional elevator fault diagnosis methods by introducing multimodal data fusion and deep learning technology. By integrating multiple data sources from sensors, environmental monitoring, and operation and maintenance records, a comprehensive feature representation is constructed, thereby greatly improving the sensitivity and accuracy of fault identification. At the same time, through an adaptive learning mechanism, the solution can monitor the operating status of the elevator in real time, warn of potential faults in advance, reduce elevator downtime, and optimize maintenance costs. Ultimately, the use of this solution can significantly improve the safety and reliability of elevators and provide users with a better service experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0040] Figure 2 It is a schematic diagram of the structure of the device of the present invention. DETAILED DESCRIPTION

[0041] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0042] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0043] Example:

[0044] See also Figure 1 The present invention provides an elevator fault diagnosis method based on machine learning, and the specific steps include:

[0045] Step 1: Acquire multiple sets of time series data, environmental data, and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current, and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator;

[0046] In this embodiment, the specific logic for obtaining time series data, environmental data, and operation and maintenance record data is as follows:

[0047] An acceleration sensor is installed at the bottom of the elevator car in the elevator shaft to obtain the acceleration data of the elevator, a speed sensor is installed on the motor shaft to obtain the running speed data of the elevator, a current sensor is installed at the power input line of the motor to obtain the current data of the elevator motor, a pressure sensor is installed at the bottom of the elevator car to obtain the load data of the elevator, the collected elevator acceleration, running speed, current data and load data are standardized, a temperature sensor and a humidity sensor are installed on the top of the elevator car to obtain the temperature and humidity data in the elevator, and the maintenance record of the elevator is derived from the operation and maintenance record data.

[0048] The acceleration and running speed of the elevator help monitor the operating status of the elevator and identify whether there are abnormal movement patterns, such as abnormal acceleration or deceleration. The current and load data can reflect the energy consumption of the elevator under different working conditions. Abnormal current values ​​may indicate motor failure or increased resistance, while load monitoring helps ensure that the elevator operates within the rated load range. Environmental data such as temperature and humidity can reveal the impact of external conditions on elevator operation and component wear. For example, excessive temperature or humidity may cause performance degradation or failure of elevator equipment. Elevator maintenance records provide past failures and repairs of the elevator, providing an important basis for identifying potential elevator failures.

[0049] Step 2: Extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator failure probability that reflects the quality of elevator operation;

[0050] In this embodiment, the specific logic for extracting the time series features, environmental features, and operation and maintenance features is as follows:

[0051] The mean and standard deviation of the elevator's acceleration, running speed, current and load data are used as time series features, and the formula for calculating the mean acceleration is:

[0052]

[0053] Among them, a i is the acceleration of the elevator at the i-th collection time, i is the index of the collection time, and i = 1, 2, ..., N, N is the number of collection times, is the mean acceleration;

[0054] The running speed, current and load of the elevator at the i-th acquisition moment are calibrated as v i ,I i , P i , and the average running speed is obtained in the same way Current mean Load Average

[0055] The calculation of the standard deviation of acceleration is based on the formula:

[0056]

[0057] Among them, σ a is the acceleration standard deviation, and similarly, the running speed standard deviation σ v , current standard deviation σ I , load standard deviation σ P ; Taking the average temperature and humidity of the elevator as the environmental characteristics, the formula for calculating the average temperature is:

[0058]

[0059] Among them, T i is the temperature of the elevator at the i-th collection time, is the mean temperature;

[0060] The humidity of the elevator at the i-th collection time is calibrated as H i , and the mean humidity is obtained in the same way

[0061] The expert scoring method is used to evaluate the elevator operation and maintenance record data, and generate the elevator failure probability that reflects the quality of elevator operation. It has multiple advantages, such as strong professionalism, high data quality, strong flexibility, high comprehensive evaluation ability, improved model accuracy, strong real-time performance, strong early warning ability, and high transparency. By introducing the professional knowledge and experience of experts, the accuracy and reliability of the failure probability assessment can be effectively improved, providing solid technical support for elevator fault diagnosis and preventive maintenance.

[0062] By calculating the mean, a large amount of time series data can be simplified into several representative features, making the analysis more efficient. These mean features can reflect the overall performance status of the elevator within a certain period of time. The mean provides information about the central trend of the data, which helps to understand the normal operation status of the various parameters of the elevator. The standard deviation describes the degree of dispersion of the data and helps to identify the fluctuation of the parameters. The mean and standard deviation can help to identify the difference between normal operation and abnormal state. For example, if the mean acceleration or current of the elevator deviates significantly from the normal range, it may indicate that the equipment has a fault or potential problem. The mean and standard deviation as input features help the training and stability of the deep learning model. Standardized input features can accelerate the convergence of the model and improve the prediction performance of the model. By analyzing the mean and standard deviation features of the elevator in different time periods, the operation mode of the elevator can be identified, which helps to understand its performance under different working conditions and provide a basis for further fault diagnosis.

[0063] Step 3: Build an elevator fault model based on the deep learning network, take the time series features and environmental features as input, and use the corresponding elevator fault probabilities as labels to train the elevator fault model;

[0064] In this embodiment, the specific logic for constructing the elevator fault model is:

[0065] Based on the deep learning network model, the mean and standard deviation of the time series data are extracted as time series features, and the mean of the environmental data is used as the environmental features. The time series features and the environmental features are concatenated to form a feature vector, which is input into the elevator fault model. The elevator failure probability generated by the expert scoring method is used as the label, and the output layer uses the softmax function to obtain the elevator failure probability. The accuracy of the current model prediction is judged by the precision rate. When the precision rate reaches 0.9, it means that the elevator fault model training is completed.

[0066] The deep learning model can automatically extract complex feature patterns, which is more flexible and efficient than traditional methods, reducing the workload of manual feature engineering. The labels generated by the expert scoring method can provide high-quality annotations, and combined with the powerful learning ability of the deep learning model, the accuracy and reliability of fault diagnosis can be improved. Using precision as an evaluation indicator can objectively reflect the actual effect of the model in fault prediction. When the precision reaches 0.9, it indicates that the model has a high degree of credibility in identifying faults.

[0067] Step 4: Obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features. Input the time series features and environmental features into the trained elevator fault model to obtain the current elevator failure probability. Compare the elevator failure probability with the fault threshold to determine the current elevator status.

[0068] In this embodiment, the specific logic for determining the current elevator status is:

[0069] Collect the time series data and environmental data of the current elevator, calculate the mean and standard deviation of the time series data, and the mean of the environmental data as the time series features and environmental features respectively, input the time series features and environmental features into the trained elevator fault model, obtain the elevator failure probability in the current state, compare the elevator failure probability predicted by the model with the preset fault threshold, if the elevator failure probability predicted by the model is greater than the fault threshold, judge that the current elevator has a fault, if the elevator failure probability predicted by the model is lower than the fault prediction, judge that the current elevator is in a safe state.

[0070] The preset fault threshold here can be the lowest elevator failure probability generated based on the expert scoring method in the collected historical data when a failure occurs. When the failure probability exceeds the threshold, the system can provide early warning information in a timely manner to avoid safety accidents caused by elevator failures.

[0071] See also Figure 2 The present invention further provides an elevator fault diagnosis device based on machine learning, wherein the elevator fault diagnosis device based on machine learning is used to implement the above-mentioned elevator fault diagnosis method based on machine learning, including:

[0072] A data acquisition module is used to obtain multiple sets of time series data, environmental data and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator;

[0073] The fault probability acquisition module is used to extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator fault probability that reflects the quality of elevator operation;

[0074] The fault model building module is used to build an elevator fault model based on a deep learning network, taking time series features and environmental features as inputs and the corresponding elevator fault probabilities as labels to train the elevator fault model;

[0075] The real-time fault diagnosis module is used to obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features, input the time series features and environmental features into the trained elevator fault model, obtain the current elevator failure probability, compare the elevator failure probability with the fault threshold, and judge the current elevator status.

[0076] The present invention further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned elevator fault diagnosis method based on machine learning.

[0077] The present invention further provides a device, comprising a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, the above-mentioned elevator fault diagnosis method based on machine learning can be implemented.

[0078] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0079] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0080] 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, and may be located in one place or distributed on multiple 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.

[0081] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for elevator fault diagnosis based on machine learning, characterized in that: The specific steps include: Step 1: Acquire multiple sets of time series data, environmental data, and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current, and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator; Step 2: Extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator failure probability that reflects the quality of elevator operation; Step 3: Build an elevator fault model based on the deep learning network, take the time series features and environmental features as input, and use the corresponding elevator fault probabilities as labels to train the elevator fault model; Step 4: Obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features. Input the time series features and environmental features into the trained elevator fault model to obtain the current elevator failure probability. Compare the elevator failure probability with the fault threshold to determine the current elevator status.

2. The elevator fault diagnosis method based on machine learning according to claim 1, characterized in that: The specific logic for obtaining time series data, environment data, and operation and maintenance record data is as follows: An acceleration sensor is installed at the bottom of the elevator car in the elevator shaft to obtain the acceleration data of the elevator, a speed sensor is installed on the motor shaft to obtain the running speed data of the elevator, a current sensor is installed at the power input line of the motor to obtain the current data of the elevator motor, a pressure sensor is installed at the bottom of the elevator car to obtain the load data of the elevator, the collected elevator acceleration, running speed, current data and load data are standardized, a temperature sensor and a humidity sensor are installed on the top of the elevator car to obtain the temperature and humidity data in the elevator, and the maintenance record of the elevator is derived from the operation and maintenance record data.

3. The elevator fault diagnosis method based on machine learning according to claim 2 is characterized in that: The specific logic for extracting time series features, environmental features, and operation and maintenance features is as follows: The mean and standard deviation of the elevator's acceleration, running speed, current and load data are used as time series features, and the formula for calculating the mean acceleration is: Among them, a i is the acceleration of the elevator at the i-th collection time, i is the index of the collection time, and i = 1, 2, ..., N, N is the number of collection times, is the mean acceleration; The running speed, current and load of the elevator at the i-th acquisition moment are calibrated as v i ,I i , P i , and the average running speed is obtained in the same way Current mean Load Average The calculation of the standard deviation of acceleration is based on the formula: Among them, σ a is the acceleration standard deviation, and similarly, the running speed standard deviation σ v , current standard deviation σ I , load standard deviation σ P ; Taking the mean temperature and humidity of the elevator as the environmental characteristics, the formula for calculating the mean temperature is: Among them, T i is the temperature of the elevator at the i-th collection time, is the mean temperature; The humidity of the elevator at the i-th collection time is calibrated as H i , and the mean humidity is obtained in the same way 4. The elevator fault diagnosis method based on machine learning according to claim 3 is characterized in that: The specific logic for building the elevator failure model is: Based on the deep learning network model, the mean and standard deviation of the time series data are extracted as time series features, and the mean of the environmental data is used as the environmental features. The time series features and the environmental features are concatenated to form a feature vector, which is input into the elevator fault model. The elevator failure probability generated by the expert scoring method is used as the label, and the output layer uses the softmax function to obtain the elevator failure probability. The accuracy of the current model prediction is judged by the precision rate. When the precision rate reaches 0.9, it means that the elevator fault model training is completed.

5. The elevator fault diagnosis method based on machine learning according to claim 4 is characterized in that: The specific logic for judging the current elevator status is: Collect the time series data and environmental data of the current elevator, calculate the mean and standard deviation of the time series data, and the mean of the environmental data as the time series features and environmental features respectively, input the time series features and environmental features into the trained elevator fault model, obtain the elevator failure probability in the current state, compare the elevator failure probability predicted by the model with the preset fault threshold, if the elevator failure probability predicted by the model is greater than the fault threshold, judge that the current elevator has a fault, if the elevator failure probability predicted by the model is lower than the fault prediction, judge that the current elevator is in a safe state.

6. A method for elevator fault diagnosis based on machine learning, characterized in that: The elevator fault diagnosis device is used to implement the elevator fault diagnosis method according to any one of claims 1 to 5, comprising: A data acquisition module is used to obtain multiple sets of time series data, environmental data and operation and maintenance record data in the historical operation of the elevator, wherein the time series data is the acceleration, running speed, current and load data of the elevator, the environmental data is the temperature and humidity data in the elevator, and the operation and maintenance record data is the maintenance record of the elevator; The fault probability acquisition module is used to extract the time series features and environmental features of the time series data and environmental data, and evaluate the elevator operation and maintenance record data based on the expert scoring method to generate the elevator fault probability that reflects the quality of elevator operation; The fault model building module is used to build an elevator fault model based on a deep learning network, taking time series features and environmental features as inputs and the corresponding elevator fault probabilities as labels to train the elevator fault model; The real-time fault diagnosis module is used to obtain the current time series data and environmental data of the elevator and extract the time series features and environmental features, input the time series features and environmental features into the trained elevator fault model, obtain the current elevator failure probability, compare the elevator failure probability with the fault threshold, and judge the current elevator status.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the machine learning-based elevator fault diagnosis method according to any one of claims 1 to 5 is implemented.

8. A device, characterized in that: The device includes a processor and a storage medium, wherein the storage medium stores a computer program. When the computer program is executed by the processor, the machine learning-based elevator fault diagnosis method described in any one of claims 1 to 5 can be implemented.

Images