Elevator fault prediction and diagnosis method based on Internet of Things

Abstract

The invention discloses an elevator fault prediction and diagnosis method based on the Internet of Things, and relates to the technical field of elevator fault prediction and diagnosis, and the method comprises the following steps: firstly, through a high-precision sensor and data acquisition equipment, obtaining various operation data generated in the elevator operation process in real time, and through the Internet of Things technology, obtaining the operation data in real time; data are transmitted to a central monitoring system in real time, continuity and real-time performance of the data are ensured, and a solid foundation is provided for follow-up analysis. Elevator data are collected in real time through the high-precision sensor and the Internet of Things and transmitted to the central monitoring system, and continuous monitoring and intelligent evaluation are achieved. The system recognizes abnormity through the speed change rate and the acceleration oscillation index, dynamically adjusts the detection frequency, optimizes resources, ensures timely early warning and early intervention, and guarantees the safety of the elevator.

Description

Technical Field

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

[0002] Elevator fault prediction and diagnosis based on the Internet of Things uses the Internet of Things technology to monitor the operating status of the elevator in real time and collect data. Through the synergy of sensors, edge computing devices and cloud platforms, a comprehensive elevator status perception network is formed. Sensors can collect key data during the operation of the elevator, such as vibration, temperature, current, voltage and other parameters, and transmit them to the cloud or edge devices. With the help of machine learning and big data analysis technology, the system can establish a fault model and analyze these data, identify potential abnormal signals or fault modes, and achieve early prediction of elevator faults. In this way, not only can early warning be given before the fault occurs to avoid unexpected downtime, but also the cause of the fault can be quickly diagnosed, the maintenance time can be shortened, and the safety and reliability of the elevator can be improved. Ultimately, this predictive maintenance model based on the Internet of Things has greatly improved the operation and maintenance efficiency of the elevator and reduced the economic losses and personal safety risks caused by faults.

[0003] The prior art has the following deficiencies: Existing technologies usually use fixed frequencies to analyze key data in elevator operation and identify potential abnormal signals or failure modes, which can meet most application scenarios. However, when the elevator experiences drastic changes in speed or acceleration in a short period of time (such as sudden stops or rapid rise and fall), fixed frequency analysis may not be able to respond in time. Such drastic acceleration or speed changes usually indicate serious problems with the traction system or braking system. If the monitoring system fails to capture these anomalies quickly, it may cause the elevator to stop suddenly or even fall rapidly, increasing the risk of elevator falls and posing a huge threat to the safety of passengers. Especially in high-rise buildings, the consequences of this delayed response can be extremely fatal.

[0004] 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

[0005] The purpose of the present invention is to provide an elevator fault prediction and diagnosis method based on the Internet of Things, which collects elevator data in real time through high-precision sensors and Internet of Things technology and transmits it to a central monitoring system to achieve continuous monitoring and intelligent evaluation of the elevator's operating status. When an emergency stop or rapid change occurs, the system can quickly identify the abnormality through the speed change rate and acceleration oscillation index, and generate a safety factor for early warning to prevent accidents. At the same time, the detection frequency is dynamically adjusted to optimize resource utilization, the monitoring frequency is increased under abnormal conditions, data changes are carefully captured, early intervention of faults is ensured, monitoring accuracy and efficiency are improved, and safe operation of elevators is guaranteed to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: an elevator fault prediction and diagnosis method based on the Internet of Things, comprising the following steps: First, through high-precision sensors and data acquisition equipment, various operating data generated during the operation of the elevator are obtained in real time. Through the Internet of Things technology, the data is transmitted to the central monitoring system in real time to ensure the continuity and real-time nature of the data, providing a solid foundation for subsequent analysis; Extract features from the acquired elevator operation data, and use the pre-learned machine learning model to intelligently evaluate the elevator's operating status based on the extracted features; According to the evaluation results of the machine learning model, the elevator operation status is divided into normal operation status and abnormal operation status; For elevators in normal operation, an optimized detection frequency is set according to their operating characteristics and historical data, and the elevator operation data is continuously monitored and analyzed; For elevators in abnormal operating states, the detection frequency is dynamically adjusted based on the optimized detection frequency set under normal operating conditions to capture abnormal changes in detail.

[0007] Preferably, feature extraction is performed on the acquired elevator operation data, and the extracted features include the rate of change of the elevator speed in a short period of time and the frequency and amplitude of the acceleration change of the elevator during operation. Under the detection window, after analyzing the rate of change of the elevator speed in a short period of time and the frequency and amplitude of the acceleration change of the elevator during operation, a speed change rate index and an acceleration oscillation index are generated respectively, and the speed change rate index and the acceleration oscillation index are input into a pre-learned machine learning model, and a safe operation coefficient is generated by the machine learning model, and the operating status of the elevator is intelligently evaluated by the safe operation coefficient.

[0008] Preferably, within the detection window, the change rate of the elevator speed in a short period of time is analyzed, and the specific steps of generating the speed change rate index are as follows: In the detection window, acceleration information is used to derive each time point The instantaneous velocity change rate at the moment is calculated as follows: , where is the instantaneous rate of change of velocity, The elevator is at the time The speed at a moment is a quantity that changes with time and reflects the motion state of the elevator at a certain point in time. is the adjustment coefficient; The instantaneous speed change rate at each time point Accumulate to obtain the speed change trend within the detection window. The calculation expression is as follows: , where is the cumulative velocity change influence coefficient, is the total number of sampling points, It is Second sampling point, is the time interval between adjacent sampling points, It's time point The instantaneous rate of change of velocity on is an exponential decay factor, is the natural base, is the weight decay coefficient; In order to capture anomalies more sensitively, the accumulated speed change influence coefficient is transformed nonlinearly, and the nonlinear speed change rate index is defined. The calculation expression is as follows: , where is the weighted nonlinear velocity change rate index, is the amplification factor used to adjust the overall value range of the weighted nonlinear velocity change rate index. is a nonlinear index; Based on the nonlinear speed change rate index, a dynamic adjustment factor is introduced to make the detection more sensitive to fluctuations. The speed change rate index after dynamic adjustment is defined, and the calculation expression is as follows: , where is the velocity change rate exponent, is the dynamic adjustment coefficient.

[0009] Preferably, in the detection window, the frequency and amplitude of acceleration changes of the elevator during operation are analyzed, and the specific steps of generating the acceleration oscillation index are as follows: In the detection window, the elevator is at the time point The acceleration at the moment is calibrated as , define the elevator acceleration at different time intervals The calculation expression for the instantaneous change in is as follows: , where is the increment of acceleration change, is the acceleration during the previous time interval; The increment of acceleration change On this basis, based on the cumulative impact of acceleration fluctuations over time, different weight coefficients are introduced to reflect the importance of acceleration changes over time. The calculation expression is as follows: , where is the weighted acceleration change, indicating the time point The weighted acceleration change at time, It's time point The weight coefficient of the moment, It's time point The change in acceleration at is the total number of time points; The nonlinear activation function is applied to further amplify the instantaneous acceleration fluctuation. The calculation expression is as follows: , where is the amplified instantaneous acceleration change value, is the magnification factor, is the nonlinear amplification factor, is the natural base; Based on the amplified instantaneous acceleration change value , define the final acceleration oscillation index, the calculation expression is as follows: , where is the acceleration oscillation index, It's at the time The amplified instantaneous acceleration change.

[0010] Preferably, the safety operation coefficient generated after analyzing various data generated during the operation of the elevator is compared and analyzed with a preset safety operation coefficient reference threshold, and the elevator operation state is divided, and the division steps are as follows: If the safety operation coefficient is greater than or equal to a preset safety operation coefficient reference threshold, the elevator operation state is classified as an abnormal operation state; If the safety operation coefficient is less than a preset safety operation coefficient reference threshold, the elevator operation state is classified as a normal operation state.

[0011] Preferably, the optimized detection frequency set under normal operating conditions is calibrated as , ensuring a balance between monitoring efficiency and data processing; Based on the rate of change index and acceleration oscillation index The frequency adjustment coefficient is dynamically determined based on the abnormality level. The calculation expression is as follows: , where is the frequency adjustment factor, is the adjustment scale factor, is the rate of change index The preset reference threshold value, is the acceleration oscillation index The preset reference threshold value of Adjustment factor according to frequency Dynamically adjust the detection frequency, the calculation expression is as follows: , where is the adjusted detection frequency; Under abnormal conditions, the system should maintain real-time monitoring and continuously evaluate the speed change rate index and acceleration oscillation index Dynamically update the frequency adjustment coefficient , and adjust the detection frequency in real time accordingly When the elevator is detected to be back to normal state, the detection frequency is restored to , reducing the usage of system resources.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention collects data in real time through high-precision sensors, and uses the Internet of Things technology to transmit the data to the central monitoring system. The solution realizes continuous monitoring and intelligent evaluation of the elevator's operating status. When the elevator has drastic changes in speed or acceleration, such as sudden stops, rapid rises or falls, the system can quickly identify potential abnormalities through the speed change rate index and the acceleration oscillation index. This early warning mechanism provides an efficient response means for elevator fault detection. Even when the elevator is in a sudden abnormal state, it can quickly capture abnormal features and output the corresponding safe operation coefficient. This enables the system to issue an early warning before the risk develops further, avoiding possible serious safety accidents, especially preventing elevators from stopping or falling rapidly in high-rise buildings, providing strong protection for passenger safety.

[0013] The present invention further optimizes the monitoring efficiency of the elevator operation status by dynamically adjusting the detection frequency. Under normal conditions, the system can set an appropriate detection frequency according to the elevator's operating characteristics and historical data to avoid data transmission and processing loads caused by excessively high frequencies, thereby improving the system's resource utilization and operating stability. When the elevator is in an abnormal operating state, the system will dynamically adjust in real time based on the set optimized detection frequency, increase the detection frequency, and more carefully capture subtle changes in the operating data. This flexible frequency adjustment mechanism not only improves the accuracy and timeliness of monitoring, but also ensures that the system can capture anomalies early in the precursor stage of a fault, facilitate targeted analysis and precise intervention, and ultimately greatly improve the overall effect of elevator operation status monitoring and fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 The present invention is a method flow chart of an elevator fault prediction and diagnosis method based on the Internet of Things. DETAILED DESCRIPTION

[0016] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0017] The present invention provides Figure 1 The elevator fault prediction and diagnosis method based on the Internet of Things shown includes the following steps: First, through high-precision sensors and data acquisition equipment, various operating data generated during the operation of the elevator are obtained in real time. Through the Internet of Things technology, the data is transmitted to the central monitoring system in real time to ensure the continuity and real-time nature of the data, providing a solid foundation for subsequent analysis; High-precision sensors and data acquisition equipment can capture various key parameters during the operation of the elevator, including speed, acceleration, vibration, temperature, current, voltage, etc. Each sensor has a specific sensing range and sensitivity, and can detect subtle changes in the elevator's operating status. These sensors are usually installed in key parts such as the traction machine, guide rails, and elevator cars to ensure that the collected data accurately reflects the actual operation of the elevator. Through this arrangement, all-round and no-dead-angle monitoring of the elevator operation can be achieved, providing comprehensive data support for analyzing the health status of the elevator.

[0018] Through the Internet of Things technology, these collected operating data can be transmitted to the central monitoring system in real time to achieve data continuity and real-time. The Internet of Things system connects sensors, edge computing devices and cloud platforms into a unified network, which can automatically and quickly collect, transmit, process and store data. The real-time transmission function ensures that the central monitoring system can instantly obtain the latest status of the elevator and avoid the safety risks caused by information delays. This data link not only ensures the integrity of the data, but also provides a solid foundation for subsequent analysis and fault diagnosis, enabling the system to issue early warnings before a fault occurs, thereby improving the overall safety and operating efficiency of the elevator.

[0019] Extract features from the acquired elevator operation data, and use the pre-learned machine learning model to intelligently evaluate the elevator's operating status based on the extracted features; Extract features from the acquired elevator operation data. The extracted features include the rate of change of the elevator speed in a short period of time and the frequency and magnitude of the acceleration change during the operation of the elevator. Under the detection window, after analyzing the rate of change of the elevator speed in a short period of time and the frequency and magnitude of the acceleration change during the operation of the elevator, generate a speed change rate index and an acceleration oscillation index respectively. Input the speed change rate index and the acceleration oscillation index into a pre-learned machine learning model, generate a safe operation coefficient through the machine learning model, and intelligently evaluate the operation status of the elevator through the safe operation coefficient. A dramatic change in elevator speed in a short period of time usually indicates that the elevator is in an abnormal operating state, because the normal acceleration and deceleration of the elevator is strictly designed to follow a smooth speed change curve to ensure the safety and comfort of passengers. When the speed change rate fluctuates dramatically, such as the elevator suddenly rises, stops or falls rapidly, it often indicates that there may be an abnormality in the traction system, control system or braking system. This abnormal speed change may be caused by problems such as unstable power output of the traction motor, signal interference of the control system, failure of the braking system or wear of the stressed parts. Such a dramatic speed fluctuation will make the elevator run unsteady, causing the risk of passengers losing balance or even getting injured, and will also increase the wear of key components and aggravate the aging of the system. If the speed change rate cannot be kept stable, it is difficult for the monitoring system to achieve effective early warning or predictive maintenance. Therefore, a sharp speed fluctuation in a short period of time is usually an early sign of elevator failure and should be taken seriously so that it can be checked and maintained as soon as possible to avoid more serious failures and safety risks.

[0020] Under the detection window, the change rate of the elevator speed in a short period of time is analyzed, and the specific steps for generating the speed change rate index are as follows: In the detection window, acceleration information is used to derive each time point The instantaneous velocity change rate at the moment is calculated as follows: , where is the instantaneous rate of change of velocity, The elevator is at the time The speed at a moment is a quantity that changes with time and reflects the motion state of the elevator at a certain point in time. is the adjustment coefficient; The reason for using acceleration information for derivation is that acceleration directly reflects the trend of speed change over time, that is, the rate of increase or decrease of the elevator speed. By analyzing acceleration, sudden acceleration or deceleration in elevator operation can be captured more sensitively, which are often early signals of potential elevator failures. Acceleration can reveal the rate of change of speed in a short period of time, and drastic fluctuations in the rate of change of speed usually mean that there is an abnormality in the traction system or control system. In addition, acceleration information provides real-time feedback on speed changes, which is more suitable for monitoring the dynamic state of the elevator than speed itself, ensuring that the system can identify abnormal conditions in a timely manner and issue early warnings.

[0021] Adjustment coefficient The role here is to balance the first derivative of the elevator speed (Acceleration) and the second derivative The impact of jerk on the rate of change index of velocity is used to achieve comprehensive sensitivity to different types of anomalies. Specifically, The larger the value, the higher the weight of the second-order derivative, and the more sensitive the formula is to sudden changes in acceleration, which helps to capture drastic changes in acceleration in a short period of time; this usually indicates that the system is at risk of sudden failure. The value increases the weight of the first-order derivative (i.e., the rate of change of speed), focusing more on the overall change trend of the elevator speed, which is used to identify abnormalities in the overall operating status. , the sensitivity of the formula to different abnormal modes can be flexibly adjusted according to actual monitoring needs.

[0022] The instantaneous speed change rate at each time point Accumulate to obtain the speed change trend within the detection window. The calculation expression is as follows: , where is the cumulative velocity change influence coefficient, is the total number of sampling points, It is Second sampling point, is the time interval between adjacent sampling points, It's time point The instantaneous rate of change of velocity on is an exponential decay factor used to give the rate of change of speed at different time points Assign different weights, is the natural base, is the weight decay coefficient; Cumulative speed change influence coefficient It is the result of gradually accumulating the instantaneous speed change rate of the elevator within the detection window, which is used to reflect the overall speed change trend of the elevator during the window. Its main function is to integrate the speed changes at multiple time points into a comprehensive indicator, providing a quantitative measure of the magnitude of the elevator speed change. When the elevator is in an abnormal operating state, the cumulative speed change impact coefficient will increase significantly, indicating that the elevator has experienced a large speed fluctuation within the time window, which helps to identify potential faults or abnormal conditions.

[0023] Weight decay coefficient It is a parameter that controls the weight of data at different sampling points, and is used to gradually reduce the influence of data at earlier time points during the accumulation process. , data close to the current moment gets a higher weight, while the weight of more distant historical data gradually decreases. The role of the weight decay coefficient is to make the cumulative speed change influence coefficient more biased towards current and recent data changes, so as to more sensitively capture the latest abnormal features in elevator operation, reduce the interference of outdated data, and improve the accuracy of anomaly detection.

[0024] In order to capture anomalies more sensitively, the accumulated speed change influence coefficient is transformed nonlinearly, and the nonlinear speed change rate index is defined. The calculation expression is as follows: , where is the weighted nonlinear velocity change rate index, is the amplification factor used to adjust the overall value range of the weighted nonlinear velocity change rate index. is a nonlinear index; The nonlinear speed rate index (NSVRI) is a further processing of the elevator speed rate index (SVRI). It enhances the sensitivity to abnormal speed changes through nonlinear amplification technology. NSVRI introduces a power exponent in the calculation, which makes the accumulated speed rate of change exponentially amplified when there is an abnormal fluctuation. This processing can effectively highlight drastic speed changes, so that small changes remain stable, while sudden abnormal phenomena such as acceleration and deceleration will significantly increase the value of NSVRI, thereby improving the monitoring system's ability to distinguish abnormal conditions and helping to quickly warn of potential faults.

[0025] The nonlinear index is the key parameter used in the NSVRI calculation formula to achieve this nonlinear amplification, which is usually greater than 1 (for example, 2 or 3). Its function is to amplify the accumulated speed change influence coefficient by power, so that when the elevator operation state is close to abnormal, the value of NSVRI increases rapidly. This exponential parameter gives the system the ability to respond sensitively to large speed fluctuations, can identify abnormalities earlier, and avoid the system from falsely reporting small changes in smooth operation.

[0026] Based on the nonlinear speed change rate index, a dynamic adjustment factor is introduced to make the detection more sensitive to fluctuations. The speed change rate index after dynamic adjustment is defined, and the calculation expression is as follows: , where is the velocity change rate exponent, is the dynamic adjustment coefficient, which is used to control the impact of dynamic weights on the overall index; The dynamic adjustment coefficient is a key factor used to adaptively adjust the monitoring parameters according to the real-time situation, ensuring the sensitivity of the monitoring system to abnormal changes. In the calculation of the speed change rate index, the dynamic adjustment coefficient is usually associated with the derivative of the change trend, and can enlarge or reduce the value of the monitoring index according to the rate of change of the current speed or other key parameters. This means that when the speed or acceleration of the elevator changes rapidly, the dynamic adjustment coefficient will automatically increase, making the monitoring system more responsive and sensitive to the change; while in a stable state, it will decrease to prevent false alarms. This dynamic adaptive adjustment method improves the accuracy and real-time performance of the system, ensures more accurate detection of sudden anomalies, and effectively reduces false alarms caused by normal small fluctuations, thereby realizing intelligent monitoring of the elevator's operating status.

[0027] Under the detection window, the larger the speed change rate index performance value generated after analyzing the rate of change of the elevator speed in a short period of time, the more dramatic the change in the elevator speed has been in a short period of time, which usually means that the elevator is in an abnormal operating state. This higher rate of change may reflect an abnormality in the traction system, control system or braking system, causing the elevator to experience abnormal phenomena such as rapid acceleration and deceleration, sudden stops, etc., which poses a threat to passenger safety or causes excessive wear of the equipment. On the contrary, if the speed change rate index performance value is small, it means that the elevator speed changes more smoothly, in line with the expected acceleration and deceleration pattern, indicating that the elevator is in a normal operating state.

[0028] The frequency and amplitude of acceleration changes during the operation of the elevator increase sharply in a short period of time, which usually indicates that the elevator is in an abnormal operating state. This sudden acceleration fluctuation means that there may be a fault in the elevator's traction system, braking system or control system. Under normal circumstances, the acceleration of the elevator changes relatively smoothly, ensuring passenger comfort and avoiding excessive wear of mechanical parts. However, when the acceleration increases sharply or fluctuates frequently in a short period of time, it may be caused by brake system failure, unbalanced traction machine load or abnormal control system signal. Such changes may also indicate wear or looseness of the suspension cable or guide rail system, causing the elevator to deviate, shake or even emergency stop during operation. These violent acceleration fluctuations not only affect the operating stability, but also increase the risk of mechanical shock, accelerate component wear, and even cause sudden failures, posing a direct threat to passenger safety.

[0029] In the detection window, the frequency and amplitude of acceleration changes during the operation of the elevator are analyzed, and the specific steps for generating the acceleration oscillation index are as follows: In the detection window, the elevator is at the time point The acceleration at the moment is calibrated as , define the elevator acceleration at different time intervals The calculation expression for the instantaneous change in is as follows: , where is the increment of acceleration change, expressed in time The instantaneous change increment of the elevator acceleration at time, that is, in the time interval The change in the acceleration of the internal elevator, is the acceleration of the previous time interval, at time Time interval before time The acceleration value of The increment of acceleration change On this basis, based on the cumulative impact of acceleration fluctuations over time, different weight coefficients are introduced to reflect the importance of acceleration changes over time. The calculation expression is as follows: , where is the weighted acceleration change, indicating the time point The weighted acceleration change at time, It's time point The weight coefficient at each moment is used to adjust the contribution of the acceleration change at different moments to the total result. It's time point The change in acceleration at is the total number of time points; The cumulative effect of acceleration fluctuations over time refers to the combined effect of multiple changes in acceleration over a period of time on the overall operating status of the elevator. When the elevator experiences continuous or frequent acceleration fluctuations during operation, even if the amplitude of each fluctuation is small, the accumulation of these changes may have a significant impact on the stability and operating safety of the elevator. By cumulatively calculating acceleration fluctuations, it is possible to identify whether the elevator has an abnormal fluctuation trend within a specific time period, thereby effectively judging whether the elevator's operating status is normal.

[0030] The nonlinear activation function is applied to further amplify the instantaneous acceleration fluctuation. The calculation expression is as follows: , where is the amplified instantaneous acceleration change value, is the magnification factor, is the nonlinear amplification factor, is the natural base; The amplification factor is a parameter used to adjust the overall amplitude of the acceleration change. Its purpose is to amplify or reduce the value of the acceleration change as a whole. In abnormality detection, the introduction of the amplification factor can help highlight the amplitude of the acceleration change, making small changes more significant and easier to identify. If the fluctuations in the elevator operation are small, the amplification factor can amplify these fluctuations to a detectable range, thereby improving the sensitivity of monitoring and enabling the system to detect potential abnormalities earlier.

[0031] The nonlinear amplification factor is used to nonlinearly amplify the acceleration change, usually through nonlinear transformations such as exponential or logarithmic functions. Nonlinear amplification allows larger changes to be amplified more significantly, while smaller changes are relatively compressed or remain unchanged, which is very critical for anomaly detection. For example, in elevator operation data, normal acceleration fluctuations do not need to be over-amplified, but sudden and violent fluctuations need to be significantly amplified to attract attention. The nonlinear amplification factor can respond more sensitively to sudden abnormal fluctuations without increasing noise interference.

[0032] The main difference between the amplification factor and the nonlinear amplification factor is the difference in the amplification method. The amplification factor is a linear factor that amplifies all acceleration changes by the same proportion, which is suitable for uniformly adjusting the entire data set to make the overall changes more obvious. The nonlinear amplification factor uses a nonlinear function (such as an exponential or logarithmic) to differentially amplify the changes, with a higher degree of amplification for larger changes and a relatively small impact on smaller changes, thereby highlighting abnormal changes. This nonlinear amplification is suitable for highlighting abnormal situations, making violent fluctuations more significant, while not causing too much interference to normal fluctuations.

[0033] Based on the amplified instantaneous acceleration change value , define the final acceleration oscillation index, the calculation expression is as follows: , where is the acceleration oscillation index, It's at the time The amplified instantaneous acceleration change; In the detection window, the larger the acceleration oscillation index value generated by analyzing the frequency and amplitude of the acceleration changes during the operation of the elevator, the higher the value, which usually indicates that the elevator is in an abnormal operation state. This is because a larger mutation index reflects frequent and drastic fluctuations in acceleration, which may be caused by failures in the traction system, braking system or control system, causing the elevator to experience problems such as emergency stops, jitters or abnormal acceleration and deceleration. In normal operation, the acceleration changes are relatively stable, and the mutation index value is small, indicating that the start, stop and operation process of the elevator meets the design specifications, and the mechanical and control systems are operating normally.

[0034] The machine learning model is not limited here and can achieve the speed change rate index and acceleration oscillation index Generate a safety operation factor after comprehensive analysis The machine learning model can be used. In order to implement the technical solution of the present invention, the present invention provides a specific implementation method: Safety operation factor The generation logic is as follows: , where , Velocity rate index and acceleration oscillation index The preset scaling factor of , Both are greater than 0.

[0035] It can be seen from the calculation expression of the safe operation coefficient that, under the detection window, the larger the speed change rate index performance value generated after analyzing the rate of change of the elevator speed in a short period of time, the larger the acceleration oscillation index performance value generated after analyzing the frequency and amplitude of the acceleration change of the elevator during operation, indicating that the larger the safety operation coefficient performance value generated after analyzing the various data generated during the operation of the elevator under the detection window, the greater the elevator is in an abnormal operation state, otherwise it indicates that the elevator is in a normal operation state.

[0036] According to the evaluation results of the machine learning model, the elevator operation status is divided into normal operation status and abnormal operation status; The safety operation coefficient generated by analyzing the various data generated during the operation of the elevator is compared with the preset safety operation coefficient reference threshold, and the elevator operation status is divided. The division steps are as follows: If the safety operation coefficient is greater than or equal to a preset safety operation coefficient reference threshold, the elevator operation state is classified as an abnormal operation state; If the safety operation coefficient is less than the preset safety operation coefficient reference threshold, the elevator operation state is classified as a normal operation state; Normal operation means that all parameters of the elevator are within the safe range and no special treatment is required; abnormal operation means that some parameters are beyond the normal range, there may be potential risks, and further attention and treatment are required. For elevators in normal operation, an optimized detection frequency is set according to their operating characteristics and historical data, and the elevator operation data is continuously monitored and analyzed; This frequency should reduce the burden of data processing and transmission as much as possible and save system resources while ensuring monitoring accuracy. At the same time, it should ensure that potential abnormal signals can be captured in time. Under the newly set optimized detection frequency, the elevator operation data is continuously monitored and analyzed. Even in normal conditions, it is necessary to keep an eye on the data and use predictive maintenance technology to identify potential abnormal trends or early signs of failure so that preventive measures can be taken in advance to avoid failures.

[0037] For elevators in abnormal operation, the detection frequency is dynamically adjusted based on the optimized detection frequency set under normal operation to capture abnormal changes in detail; For elevators in abnormal operation, the detection frequency is dynamically adjusted based on the optimized detection frequency set under normal operation. The specific steps for capturing abnormal changes in detail are as follows: The optimized detection frequency set under normal operating conditions is calibrated as , ensuring a balance between monitoring efficiency and data processing; Based on the rate of change index and acceleration oscillation index The frequency adjustment coefficient is dynamically determined based on the abnormality level. The calculation expression is as follows: , where is the frequency adjustment factor, is the adjustment scale factor, is the rate of change index The preset reference threshold value, is the acceleration oscillation index The preset reference threshold value of Adjust the scale factor It is a parameter used to control the adjustment range of the detection frequency. When dynamically adjusting the detection frequency, Determines the sensitivity of frequency adjustment, that is, the degree to which the detection frequency increases under abnormal conditions. Its function is to balance system resources and monitoring accuracy: by adjusting The size of the proportional coefficient can increase the detection frequency in severe abnormal situations where frequent data collection is required, and moderately increase the frequency in minor abnormal situations to avoid wasting system resources caused by excessive frequency. In simple terms, adjusting the proportional coefficient This ensures that the adjustment of detection frequency can not only respond quickly to major anomalies, but also flexibly adapt to different degrees of anomalies, achieving efficient real-time monitoring.

[0038] Frequency adjustment factor The value of the speed change rate index and acceleration oscillation index The abnormal degree of changes dynamically. When the two are larger, the frequency adjustment coefficient The higher the value, the higher the detection frequency of the system will be, thus detecting the elevator operation status more frequently.

[0039] Adjustment factor according to frequency Dynamically adjust the detection frequency, the calculation expression is as follows: , where It is the adjusted detection frequency, used for higher density real-time monitoring under abnormal conditions; When the elevator enters an abnormal state, the adjusted detection frequency Will be based on For example, the optimized detection frequency set under normal operation for ,like , then the adjusted detection frequency The frequency will change to 3Hz, that is, data is collected three times per second. This dynamic adjustment can capture abnormal changes in more detail.

[0040] Under abnormal conditions, the system should maintain real-time monitoring and continuously evaluate the speed change rate index and acceleration oscillation index Dynamically update the frequency adjustment coefficient , and adjust the detection frequency in real time accordingly , when it is detected that the elevator returns to normal state (i.e. speed change rate index and acceleration oscillation index If the detection frequency is restored to below the threshold, the detection frequency is restored to , reducing the usage of system resources.

[0041] The present invention collects data in real time through high-precision sensors, and uses the Internet of Things technology to transmit the data to the central monitoring system. The solution realizes continuous monitoring and intelligent evaluation of the elevator's operating status. When the elevator has drastic changes in speed or acceleration, such as sudden stops, rapid rises or falls, the system can quickly identify potential abnormalities through the speed change rate index and the acceleration oscillation index. This early warning mechanism provides an efficient response means for elevator fault detection. Even when the elevator is in a sudden abnormal state, it can quickly capture abnormal features and output the corresponding safe operation coefficient. This enables the system to issue an early warning before the risk develops further, avoiding possible serious safety accidents, especially preventing elevators from stopping or falling rapidly in high-rise buildings, providing strong protection for passenger safety.

[0042] The present invention further optimizes the monitoring efficiency of the elevator operation status by dynamically adjusting the detection frequency. Under normal conditions, the system can set an appropriate detection frequency according to the elevator's operating characteristics and historical data to avoid data transmission and processing loads caused by excessively high frequencies, thereby improving the system's resource utilization and operating stability. When the elevator is in an abnormal operating state, the system will dynamically adjust in real time based on the set optimized detection frequency, increase the detection frequency, and more carefully capture subtle changes in the operating data. This flexible frequency adjustment mechanism not only improves the accuracy and timeliness of monitoring, but also ensures that the system can capture anomalies early in the precursor stage of a fault, facilitate targeted analysis and precise intervention, and ultimately greatly improve the overall effect of elevator operation status monitoring and fault diagnosis.

[0043] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An elevator fault prediction and diagnosis method based on the Internet of Things, characterized in that: The following steps are involved: First, through high-precision sensors and data acquisition equipment, various operating data generated during the operation of the elevator are obtained in real time. Through the Internet of Things technology, the data is transmitted to the central monitoring system in real time to ensure the continuity and real-time nature of the data, providing a solid foundation for subsequent analysis; Extract features from the acquired elevator operation data, and use the pre-learned machine learning model to intelligently evaluate the elevator's operating status based on the extracted features; According to the evaluation results of the machine learning model, the elevator operation status is divided into normal operation status and abnormal operation status; For elevators in normal operation, an optimized detection frequency is set according to their operating characteristics and historical data, and the elevator operation data is continuously monitored and analyzed; For elevators in abnormal operating states, the detection frequency is dynamically adjusted based on the optimized detection frequency set under normal operating conditions to capture abnormal changes in detail.

2. The elevator fault prediction and diagnosis method based on the Internet of Things according to claim 1 is characterized in that: Feature extraction is performed on the acquired elevator operation data, and the extracted features include the rate of change of the elevator speed in a short period of time and the frequency and amplitude of the acceleration change of the elevator during operation. Under the detection window, the rate of change of the elevator speed in a short period of time and the frequency and amplitude of the acceleration change of the elevator during operation are analyzed, and a speed change rate index and an acceleration oscillation index are generated respectively. The speed change rate index and the acceleration oscillation index are input into a pre-learned machine learning model, and a safe operation coefficient is generated through the machine learning model. The operating status of the elevator is intelligently evaluated through the safe operation coefficient.

3. The elevator fault prediction and diagnosis method based on the Internet of Things according to claim 2 is characterized in that: Under the detection window, the change rate of the elevator speed in a short period of time is analyzed, and the specific steps for generating the speed change rate index are as follows: In the detection window, acceleration information is used to derive each time point The instantaneous velocity change rate at the moment is calculated as follows: , where is the instantaneous rate of change of velocity, The elevator is at the time The speed at a moment is a quantity that changes with time and reflects the motion state of the elevator at a certain point in time. is the adjustment coefficient; The instantaneous speed change rate at each time point Accumulate to obtain the speed change trend within the detection window. The calculation expression is as follows: , where is the cumulative velocity change influence coefficient, is the total number of sampling points, It is Second sampling point, is the time interval between adjacent sampling points, It's time point The instantaneous rate of change of velocity on is an exponential decay factor, is the natural base, is the weight decay coefficient; In order to capture anomalies more sensitively, the accumulated speed change influence coefficient is transformed nonlinearly, and the nonlinear speed change rate index is defined. The calculation expression is as follows: , where is the weighted nonlinear velocity change rate index, is the amplification factor used to adjust the overall value range of the weighted nonlinear velocity change rate index. is a nonlinear index; Based on the nonlinear speed change rate index, a dynamic adjustment factor is introduced to make the detection more sensitive to fluctuations. The speed change rate index after dynamic adjustment is defined, and the calculation expression is as follows: , where is the velocity change rate exponent, is the dynamic adjustment coefficient.

4. The elevator fault prediction and diagnosis method based on the Internet of Things according to claim 2 is characterized in that: In the detection window, the frequency and amplitude of acceleration changes during the operation of the elevator are analyzed, and the specific steps for generating the acceleration oscillation index are as follows: In the detection window, the elevator is at the time point The acceleration at the moment is calibrated as , define the elevator acceleration at different time intervals The calculation expression of the instantaneous change in is as follows: , where is the increment of acceleration change, is the acceleration during the previous time interval; The increment of acceleration change On this basis, based on the cumulative impact of acceleration fluctuations over time, different weight coefficients are introduced to reflect the importance of acceleration changes over time. The calculation expression is as follows: , where is the weighted acceleration change, indicating the time point The weighted acceleration change at time, It's time point The weight coefficient of the moment, It's time point The acceleration change at is the total number of time points; The nonlinear activation function is applied to further amplify the instantaneous acceleration fluctuation. The calculation expression is as follows: , where is the amplified instantaneous acceleration change value, is the magnification factor, is the nonlinear amplification factor, is the natural base; Based on the amplified instantaneous acceleration change value , define the final acceleration oscillation index, the calculation expression is as follows: , where is the acceleration oscillation index, It's at the time The amplified instantaneous acceleration change.

5. The elevator fault prediction and diagnosis method based on the Internet of Things according to claim 2 is characterized in that: The safety operation coefficient generated by analyzing the various data generated during the operation of the elevator is compared with the preset safety operation coefficient reference threshold, and the elevator operation status is divided. The division steps are as follows: If the safety operation coefficient is greater than or equal to a preset safety operation coefficient reference threshold, the elevator operation state is classified as an abnormal operation state; If the safety operation coefficient is less than a preset safety operation coefficient reference threshold, the elevator operation state is classified as a normal operation state.

6. The elevator fault prediction and diagnosis method based on the Internet of Things according to claim 5 is characterized in that: The optimized detection frequency set under normal operating conditions is calibrated as , ensuring a balance between monitoring efficiency and data processing; Based on the rate of change index and acceleration oscillation index The frequency adjustment coefficient is dynamically determined based on the abnormality level. The calculation expression is as follows: , where is the frequency adjustment factor, is the adjustment scale factor, is the rate of change index The preset reference threshold value, is the acceleration oscillation index The preset reference threshold value of Adjustment factor according to frequency Dynamically adjust the detection frequency, the calculation expression is as follows: , where is the adjusted detection frequency; Under abnormal conditions, the system should maintain real-time monitoring and continuously evaluate the speed change rate index and acceleration oscillation index Dynamically update the frequency adjustment coefficient , and adjust the detection frequency in real time accordingly When the elevator is detected to be back to normal state, the detection frequency is restored to , reducing the usage of system resources.

Images