Elevator fault diagnosis system based on AI intelligent data analysis

Abstract

The present invention discloses an elevator fault diagnosis system based on AI intelligent data analysis. The present invention relates to the field of elevator fault analysis technology, and solves the technical problem of lack of effective data mining and analysis means and inability to provide strong support for fault diagnosis. The present invention integrates various types of elevator fault phenomena, causes, involved components and solutions through a fault knowledge graph based on multi-channel data sources, using advanced data mining and knowledge extraction technology. This provides comprehensive and structured knowledge support for fault diagnosis. Whether it is a common fault or a rare fault, it can be quickly matched based on the graph, improving the diagnostic efficiency, combining historical fault records with current fault phenomena for comprehensive analysis, generating appearance analysis results and concrete analysis results, which can not only quickly grasp the reasons behind the intuitive and visible external appearances, but also dig deep into the deep-seated fault hidden sources hidden in the internal structure of the elevator and the electrical circuits.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator fault analysis, and in particular to an elevator fault diagnosis system based on AI intelligent data analysis. Background Art

[0002] With the acceleration of urbanization, elevators, as an indispensable means of vertical transportation in modern high-rise buildings, have continued to increase in number. The safe and stable operation of elevators is directly related to people's daily travel convenience and life safety. However, during the long-term operation of elevators, due to many factors such as mechanical wear, electrical aging, environmental changes, and frequent use, failures often occur.

[0003] The patent application with publication number CN119038346A discloses a method and device for diagnosing faults in an elevator system, including monitoring the vibration signals of the x-axis, y-axis and z-axis respectively through a mobile terminal fixed in the elevator in the start-up, acceleration, uniform speed, deceleration and stop states of the elevator; extracting the time domain features and frequency domain features in the vibration signal, wherein the time domain features include the variance, root mean square and kurtosis factor of the vibration signal, and the frequency domain features include the spectrum energy features obtained by wavelet packet decomposition and the elevator frequency features; using a support vector machine to classify the time domain features and the frequency domain features; using Shapley values ​​to quantify the contribution of each feature in the time domain features and the frequency domain features; screening and obtaining key features for fault diagnosis according to the order of the contribution from large to small; and using the support vector machine to diagnose the faults of the elevator according to the key features.

[0004] However, some existing elevator fault diagnosis mainly relies on the experience of maintenance personnel. When faced with a fault, maintenance personnel rely on the case knowledge accumulated in the past, through on-site observation and the use of simple detection tools, to check the mechanical parts and electrical systems of the elevator. This method is inefficient. For some complex faults, it often takes a lot of time and repeated attempts to find the root cause. The lack of effective data mining and analysis methods cannot provide strong support for fault diagnosis. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an elevator fault diagnosis system based on AI intelligent data analysis, which solves the problem of lack of effective data mining and analysis methods and inability to provide strong support for fault diagnosis.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an elevator fault diagnosis system based on AI intelligent data analysis, comprising:

[0007] The fault graph matching and analysis module is used to match the fault information transmitted by the fault parameter acquisition module with the fault knowledge graph to determine the fault cause, and generate a secondary analysis signal and fault cause information, and at the same time transmit the fault cause to the fault diagnosis information output module and transmit the secondary analysis signal to the comprehensive analysis and processing module;

[0008] A comprehensive analysis processing module is used to process the acquired secondary analysis signal, obtain historical fault records, and perform comprehensive processing in combination with the fault phenomenon corresponding to the elevator fault information to generate a representation analysis result and a concrete analysis result, and transmit the concrete analysis result to the concrete performance analysis module, and generate corresponding fault cause information by processing the representation analysis result, and then transmit the fault cause information to the fault diagnosis information output module;

[0009] The concrete performance analysis module is used to analyze the obtained concrete analysis results. It obtains the associated parameters corresponding to the fault information, and comprehensively analyzes the fault parameters and the associated parameters to screen the non-external appearance records to obtain pre-selected records. At the same time, the fault cause is determined based on the different fault frequencies in the pre-selected records, and the fault cause information is generated, and then the fault cause information is transmitted to the fault diagnosis information output module.

[0010] As a further solution of the present invention, it also includes a fault parameter acquisition module, which is used to acquire fault information and transmit the acquired fault information to the fault spectrum matching analysis module;

[0011] The fault diagnosis information output module is used to display the acquired fault cause information to the corresponding operator.

[0012] As a further solution of the present invention, the specific manner in which the fault spectrum matching analysis module generates the secondary analysis signal and the fault cause information is:

[0013] Obtain the fault knowledge graph, and then match the fault information of the elevator with the fault knowledge graph. If the fault knowledge graph contains the same situation as the fault information, obtain the same fault information in the fault knowledge graph and generate fault cause information. Otherwise, if the fault knowledge graph does not contain the same situation as the fault information, generate a secondary analysis signal.

[0014] As a further solution of the present invention, the specific manner in which the comprehensive analysis and processing module processes the secondary analysis signal is:

[0015] Obtain the fault information corresponding to the secondary analysis signal, and simultaneously obtain the historical fault records, then obtain the fault parameters corresponding to the fault information and record them as target parameters, and use the target parameters as the standard to filter the historical fault records to obtain the records to be analyzed, and simultaneously obtain the fault phenomena corresponding to the records to be analyzed, and classify them according to the fault phenomena to obtain external phenomenon records and non-external phenomenon records, then obtain the fault phenomena corresponding to the fault information, and classify and judge the fault phenomena corresponding to the fault information to obtain phenomenon analysis results and concrete analysis results, and analyze the phenomenon analysis results at the same time.

[0016] As a further solution of the present invention, the specific method in which the comprehensive analysis processing module analyzes the appearance analysis result is:

[0017] Obtain the external phenomenon record, and at the same time obtain the fault phenomenon corresponding to the fault information, and use the fault phenomenon as a standard to filter the external phenomenon record to obtain a secondary screening record, then obtain the fault feature corresponding to the fault information, and at the same time use the fault feature as a standard to filter the secondary screening record to obtain specific cause information, and transmit the specific cause information to the fault diagnosis information output module.

[0018] As a further solution of the present invention, the specific manner in which the concrete performance analysis module analyzes the concrete analysis result is:

[0019] Acquire the fault parameters corresponding to the fault information, acquire the associated parameters corresponding to the fault parameters, and acquire the parameter characteristics of the associated parameters, then combine the fault characteristics corresponding to the fault parameters and the parameter characteristics corresponding to the associated parameters to obtain the combined characteristics, and screen the combined characteristics and non-external appearance records to obtain pre-selected records, and analyze the pre-selected records.

[0020] As a further solution of the present invention, the specific manner in which the concrete performance analysis module analyzes the pre-selected records is:

[0021] All pre-selected records are obtained, and the historical faults corresponding to the pre-selected records are obtained at the same time, and the historical faults are obtained and classified to obtain fault classification information, which is denoted by i, and i=1, 2, ..., j, where j represents the number of fault types. Then any group of fault classification information is obtained as the target object, and the fault frequency corresponding to the target object is calculated. At the same time, the fault frequency is compared with the preset value, and the fault classification information corresponding to the fault frequency greater than the preset value is screened and recorded as the pre-selected fault classification information, and the pre-selected fault classification information is analyzed at the same time.

[0022] As a further solution of the present invention, the specific manner in which the concrete performance analysis module analyzes the pre-selected fault classification information is as follows:

[0023] All pre-selected fault classification information is obtained, and at the same time, the parameter changes corresponding to the pre-selected fault classification information are sorted out to obtain the abnormal parameter interval, and then the fault parameters corresponding to the fault information are obtained, and the change values ​​of the fault parameters within the time period t are obtained, and the obtained change values ​​are matched with the abnormal parameter interval, the fault cause is determined to generate fault cause information, and the generated fault cause information is transmitted to the fault diagnosis information output module.

[0024] The present invention provides an elevator fault diagnosis system based on AI intelligent data analysis. Compared with the prior art, it has the following beneficial effects:

[0025] The present invention integrates various elevator fault phenomena, causes, involved components and solutions through fault knowledge graphs based on multi-channel data sources and advanced data mining and knowledge extraction technologies. This provides comprehensive and structured knowledge support for fault diagnosis. Whether it is a common fault or a rare fault, it can be quickly matched based on the graph, improving diagnostic efficiency.

[0026] Combined with the historical fault records and the current fault phenomenon for comprehensive analysis, the appearance analysis results and the concrete analysis results are generated, which can not only quickly grasp the reasons behind the intuitive external appearance, but also dig deep into the deep-seated fault hidden causes hidden in the internal structure and electrical circuits of the elevator, avoiding misdiagnosis and missed diagnosis. The concrete performance analysis module innovatively introduces the concept of associated parameters, obtains the fault parameters and their associated parameters corresponding to the fault information, combines the features and filters with non-external appearance records to obtain pre-selected records, and then combines the fault frequency analysis to determine the cause of the fault. This process makes full use of the inherent connection between the elevator operation data to accurately locate the cause of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a block diagram of the system principle of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] For example, see Figure 1 The present application provides an elevator fault diagnosis system based on AI intelligent data analysis, including: a fault parameter acquisition module, a fault spectrum matching analysis module, a comprehensive analysis and processing module, a concrete performance analysis module and a fault diagnosis information output module, and combines Figure 1 It can be known that the above functional modules are electrically connected in a unidirectional manner.

[0030] Fault parameter acquisition module, which is used to acquire fault information and transmit the acquired fault information to the fault map matching analysis module. The fault information is acquired through the real-time performance of the elevator, including fault code, car operation status (such as abnormal vibration, jamming, overspeed, etc.), electrical parameter fluctuations, abnormal opening and closing of car doors and other comprehensive details.

[0031] Fault graph matching analysis module, which is used to match the acquired fault information with the fault knowledge graph to determine the cause of the fault, and generate a secondary analysis signal and fault cause information, while transmitting the fault cause to the fault diagnosis information output module and the secondary analysis signal to the comprehensive analysis processing module.

[0032] First, we obtain the fault knowledge graph. Relying on multi-channel data sources such as professional elevator fault databases, industry maintenance case collections, and technical information provided by elevator manufacturers, we use advanced data mining and knowledge extraction technologies to accurately construct a detailed fault knowledge graph covering various types of elevator fault phenomena, causes, involved components, and solutions. Then, we match the elevator fault information with the fault knowledge graph. If the fault knowledge graph contains the same situation as the fault information, we obtain the same fault information in the fault knowledge graph and generate fault cause information. Otherwise, if the fault knowledge graph does not contain the same situation as the fault information, we generate a secondary analysis signal.

[0033] The fault diagnosis information output module is used to display the acquired fault cause information to the corresponding operator.

[0034] Embodiment 2, as Embodiment 2 of the present invention, is implemented on the basis of Embodiment 1, and differs from Embodiment 1 in that:

[0035] The comprehensive analysis processing module is used to process the acquired secondary analysis signals. It obtains historical fault records and performs comprehensive processing on the fault phenomena corresponding to the elevator fault information to generate representational analysis results and concrete analysis results. At the same time, the concrete analysis results are transmitted to the concrete performance analysis module. The corresponding fault cause information is generated by processing the representational analysis results, and then the fault cause information is transmitted to the fault diagnosis information output module.

[0036] The fault information corresponding to the secondary analysis signal is obtained, and the historical fault records are obtained at the same time. Then, the fault parameters corresponding to the fault information are obtained and recorded as target parameters. The historical fault records are screened based on the target parameters to obtain the records to be analyzed. Here, the records with the same target parameters in the historical fault records are screened, and the same means that they are all the same parameter abnormalities. For example, if the fault information shows that the traction motor current instantly soars and the car is accompanied by a slight lateral displacement during the starting and acceleration stage of the elevator, the target parameters identified by the system at this time are "abnormal traction motor current during the starting and acceleration stage" and "abnormal lateral displacement parameters of the car".

[0037] Subsequently, these target parameters are used as a yardstick for precise screening to conduct a detailed screening of historical fault records, and the records in the historical records that present the same abnormal characteristics as the target parameters are picked out one by one and gathered into a set of records to be analyzed. That is to say, only those fault records that have had similar situations as "abnormal traction motor current during the start-up acceleration phase" or "abnormal lateral displacement parameters of the car" in the past will be included in the category of records to be analyzed. At the same time, the fault phenomena corresponding to the records to be analyzed are obtained, and classified according to the fault phenomena to obtain external phenomenon records and non-external phenomenon records. Then, the fault phenomena corresponding to the fault information are obtained, and at the same time, the fault phenomena corresponding to the fault information are classified and judged to obtain the phenomenon analysis results and the concrete analysis results. The phenomenon analysis results here correspond to the external phenomenon records obtained by the above classification, and the concrete analysis results correspond to the non-external phenomenon records, and the obtained phenomenon analysis results are analyzed; these fault phenomena include both external manifestations that can be directly observed, such as obvious jams when the car door is closed, excessive gaps between the floor door and the car door, etc., and non-external phenomena that can only be detected with the help of professional detection tools, such as slight bulging of a specific capacitor on a circuit board in the control cabinet, hidden wear of the steel wire inside the traction rope, etc.

[0038] For example, if the current fault symptom is the car door jamming when closing, then the surface analysis results may point to external factors such as the accumulation of debris on the door sill and deformation of the car door guide rail; if the fault symptom is abnormal noise and unstable speed during elevator operation, the specific analysis results may dig deep into deep-seated causes such as wear of the traction machine bearings and parameter drift of the control system.

[0039] Obtain the external phenomenon record, and at the same time obtain the fault phenomenon corresponding to the fault information, and use the fault phenomenon as a standard to filter the external phenomenon record to obtain a secondary screening record, then obtain the fault feature corresponding to the fault information, and the fault feature here is specifically manifested as the change of the corresponding fault parameter, and at the same time use the fault feature as a standard to filter the secondary screening record to obtain specific cause information, and transmit the specific cause information to the fault diagnosis information output module.

[0040] The concrete performance analysis module is used to analyze the obtained concrete analysis results. It obtains the associated parameters corresponding to the fault information, and comprehensively analyzes the fault parameters and the associated parameters to screen the non-external appearance records to obtain pre-selected records. At the same time, the fault cause is determined based on the different fault frequencies in the pre-selected records, and the fault cause information is generated. Then the fault cause information is transmitted to the fault diagnosis information output module.

[0041] Obtain the fault parameters corresponding to the fault information, and at the same time obtain the associated parameters corresponding to the fault parameters, where the associated parameters are represented as parameters that change with the fault parameters, and obtain the parameter characteristics of the associated parameters, then combine the fault characteristics corresponding to the fault parameters and the parameter characteristics corresponding to the associated parameters to obtain the combined characteristics, for example, the current surge is set to exceed the rated current by 1.2 times or more, the current fluctuation frequency is abnormal with more than 10 fluctuations per minute, the high temperature at the rear end of the traction machine is defined as 10°C or more higher than the average temperature at the front end, and the heating rate is fast, which is set to a heating amplitude of 8°C or more per minute, then further mark the obtained parameters as combined characteristics, and at the same time, screen the combined characteristics and non-external appearance records to obtain pre-selected records;

[0042] Then, all pre-selected records are obtained, and at the same time, historical faults corresponding to the pre-selected records are obtained, and the historical faults are obtained and classified to obtain fault classification information, and here the same historical faults are combined and classified, and the label is recorded as i, and i=1, 2, ..., j, where j represents the number of fault types, and then any group of fault classification information is obtained as the target object, and the fault frequency corresponding to the target object is calculated, and the fault frequency here is the mean fault frequency of the target object as a whole, and at the same time, the fault frequency is compared with the preset value, and the specific value of the preset value is set by the operator, and the fault classification information corresponding to the fault frequency greater than the preset value is screened and recorded as the pre-selected fault classification information;

[0043] All pre-selected fault classification information is obtained, and at the same time, the parameter changes corresponding to the pre-selected fault classification information are sorted out to obtain the abnormal parameter interval, and the abnormal parameter interval here is expressed as a parameter change value, then the fault parameter corresponding to the fault information is obtained, and the change value of the fault parameter within the time period t is obtained, the obtained change value is matched with the abnormal parameter interval, the fault cause is determined to generate fault cause information, and the generated fault cause information is transmitted to the fault diagnosis information output module.

[0044] For example, for the pre-selected fault classification information involving elevator traction motor faults, through the analysis of historical data, it was found that when similar faults occurred in the past, the traction motor current often soared from 80% to 120% of the rated current within 5 seconds after starting, and the fluctuation frequency reached more than 15 times per minute. At the same time, the traction motor casing temperature increased by 20℃-30℃ within 10 minutes of operation. Based on this, the abnormal parameter range for this type of fault can be accurately defined: the current change range is from 80%-120% of the rated current within 5 seconds of starting, the fluctuation frequency is ≥15 times per minute, and the temperature change range is an increase of 20℃-30℃ within 10 minutes of operation;

[0045] Next, the system quickly captures the fault parameters corresponding to the current elevator fault information, which requires a thorough understanding of the monitoring data of each key part of the elevator in real-time operation, such as the real-time current value of the traction machine, the running speed of the car, the pressure data of the guide rail, etc. Then, focus on a specific time period t to accurately track the dynamic changes of these fault parameters. The selection of the time period t depends on the characteristics of the fault type. For some instantaneous explosive faults, t may be only a few seconds; for progressive faults, it may be extended to several minutes or even longer.

[0046] For example, if t is set to 6 seconds and the traction motor current is continuously monitored during this period, it is found that it gradually increases from 90% to 110% of the rated current, with a fluctuation frequency of 18 times per minute. At the same time, the temperature sensor monitors that the traction motor casing temperature has increased by 22°C within these 6 seconds.

[0047] Finally, the acquired fault parameter change values ​​within the time period t are compared and matched one by one with the abnormal parameter intervals carefully sorted out before. Once it is found that the current fault parameter change value falls precisely within a certain abnormal parameter interval, the system can quickly determine the precise cause of the elevator failure based on the historical fault causes and solutions associated with the pre-selected fault classification information, and instantly generate detailed and clearly organized fault cause information.

[0048] The fault diagnosis information output module is used to display the acquired fault cause information to the corresponding operator.

[0049] Embodiment 3, as the embodiment 3 of the present invention, focuses on combining the implementation processes of embodiment 1 and embodiment 2 for implementation.

[0050] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0051] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. Elevator fault diagnosis system based on AI intelligent data analysis, characterized by: include: The fault graph matching and analysis module is used to match the fault information transmitted by the fault parameter acquisition module with the fault knowledge graph to determine the fault cause, and generate a secondary analysis signal and fault cause information, and at the same time transmit the fault cause to the fault diagnosis information output module and transmit the secondary analysis signal to the comprehensive analysis and processing module; A comprehensive analysis processing module is used to process the acquired secondary analysis signal, obtain historical fault records, and perform comprehensive processing in combination with the fault phenomenon corresponding to the elevator fault information to generate a representation analysis result and a concrete analysis result, and transmit the concrete analysis result to the concrete performance analysis module, and generate corresponding fault cause information by processing the performance analysis result, and then transmit the fault cause information to the fault diagnosis information output module; The concrete performance analysis module is used to analyze the obtained concrete analysis results. It obtains the associated parameters corresponding to the fault information, and comprehensively analyzes the fault parameters and the associated parameters to screen the non-external appearance records to obtain pre-selected records. At the same time, the fault cause is determined based on the different fault frequencies in the pre-selected records, and the fault cause information is generated, and then the fault cause information is transmitted to the fault diagnosis information output module.

2. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 1 is characterized in that: It also includes a fault parameter acquisition module, which is used to acquire fault information and transmit the acquired fault information to the fault spectrum matching analysis module; The fault diagnosis information output module is used to display the acquired fault cause information to the corresponding operator.

3. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 1 is characterized in that: The specific method in which the fault spectrum matching analysis module generates secondary analysis signals and fault cause information is as follows: Obtain the fault knowledge graph, and then match the fault information of the elevator with the fault knowledge graph. If the fault knowledge graph contains the same situation as the fault information, obtain the same fault information in the fault knowledge graph and generate fault cause information. Otherwise, if the fault knowledge graph does not contain the same situation as the fault information, generate a secondary analysis signal.

4. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 1 is characterized in that: The specific way in which the comprehensive analysis and processing module processes the secondary analysis signal is as follows: Obtain the fault information corresponding to the secondary analysis signal, and simultaneously obtain the historical fault records, then obtain the fault parameters corresponding to the fault information and record them as target parameters, and use the target parameters as the standard to filter the historical fault records to obtain the records to be analyzed, and simultaneously obtain the fault phenomena corresponding to the records to be analyzed, and classify them according to the fault phenomena to obtain external phenomenon records and non-external phenomenon records, then obtain the fault phenomena corresponding to the fault information, and classify and judge the fault phenomena corresponding to the fault information to obtain phenomenon analysis results and concrete analysis results, and analyze the phenomenon analysis results at the same time.

5. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 4 is characterized in that: The specific way in which the comprehensive analysis processing module analyzes the results of the appearance analysis is as follows: Obtain the external phenomenon record, and at the same time obtain the fault phenomenon corresponding to the fault information, and use the fault phenomenon as a standard to filter the external phenomenon record to obtain a secondary screening record, then obtain the fault feature corresponding to the fault information, and at the same time use the fault feature as a standard to filter the secondary screening record to obtain specific cause information, and transmit the specific cause information to the fault diagnosis information output module.

6. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 1 is characterized in that: The specific way in which the concrete performance analysis module analyzes the concrete analysis results is as follows: Acquire the fault parameters corresponding to the fault information, acquire the associated parameters corresponding to the fault parameters, and acquire the parameter characteristics of the associated parameters, then combine the fault characteristics corresponding to the fault parameters and the parameter characteristics corresponding to the associated parameters to obtain the combined characteristics, and screen the combined characteristics and non-external appearance records to obtain pre-selected records, and analyze the pre-selected records.

7. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 6 is characterized in that: The specific way in which the concrete performance analysis module analyzes the pre-selected records is as follows: All pre-selected records are obtained, and the historical faults corresponding to the pre-selected records are obtained at the same time, and the historical faults are obtained and classified to obtain fault classification information, which is denoted by i, and i=1, 2, ..., j, where j represents the number of fault types. Then any group of fault classification information is obtained as the target object, and the fault frequency corresponding to the target object is calculated. At the same time, the fault frequency is compared with the preset value, and the fault classification information corresponding to the fault frequency greater than the preset value is screened and recorded as the pre-selected fault classification information, and the pre-selected fault classification information is analyzed at the same time.

8. The elevator fault diagnosis system based on AI intelligent data analysis according to claim 7 is characterized in that: The specific way in which the concrete performance analysis module analyzes the pre-selected fault classification information is as follows: All pre-selected fault classification information is obtained, and at the same time, the parameter changes corresponding to the pre-selected fault classification information are sorted out to obtain the abnormal parameter interval, and then the fault parameters corresponding to the fault information are obtained, and the change values ​​of the fault parameters within the time period t are obtained, and the obtained change values ​​are matched with the abnormal parameter interval, the fault cause is determined to generate fault cause information, and the generated fault cause information is transmitted to the fault diagnosis information output module.

Images