An elevator on-demand maintenance method and system based on Bayesian and risk order number
Through Bayesian method and elevator monitoring data, the elevator risk priority is calculated and maintenance measures are automatically judged, which solves the problem of unreasonable maintenance cycle in the existing elevator maintenance methods, and achieves on-demand maintenance and cost reduction of elevators.
Patent Information
- Application Number
- CN202210448953.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-26
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-04-26
AI Technical Summary
The existing elevator maintenance methods rely on expert experience and subjective judgment, resulting in unreasonable maintenance cycles, over-maintenance or under-maintenance, waste of resources and high maintenance costs.
Using Bayesian and risk order methods, the elevator's own monitoring data acquisition function and dynamic data are used to calculate the severity, occurrence and difficulty detection of safety risks, determine the elevator risk priority, and automatically determine whether maintenance is required Measures to reduce manual intervention.
It realizes on-demand maintenance, reduces maintenance costs, avoids under-maintenance or over-maintenance caused by excessive or short maintenance cycles, and improves maintenance efficiency and resource utilization.
Smart Images

Figure CN114897192B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of elevator maintenance, and in particular to an elevator on-demand maintenance method and system based on Bayesian and risk sequence numbers. Background Art
[0002] Generally speaking, when an elevator just enters the wear and tear period, the elevator equipment is not seriously worn, and performing maintenance according to the cycle set by the maintenance company is obviously over-maintenance. However, when the elevator has been used for a long time and enters the high-risk area or the frequency of use is extremely high, performing maintenance according to the cycle set by the maintenance company is obviously under-maintenance. Therefore, it is very necessary to know how to maintain the elevator on demand.
[0003] In the paper "Equipment Maintenance Strategy Method Based on FEMA", the FEMA method was applied to the sintering machine of the steel enterprise. The maintenance strategy was formulated by calculating the risk value RPN, but the possibility of risk occurrence was mostly based on past experience and insufficient data support. In the existing on-demand maintenance methods, on the one hand, they rely more on the evaluation of expert groups, and determine the on-demand maintenance plan through subjective thinking and experience judgment. Different experts often make different decisions, resulting in a variety of on-demand maintenance plans, but not necessarily effective. On the other hand, with the increasingly advanced elevator safety monitoring technology, when designing on-demand maintenance plans, the monitoring technology is not well utilized, resulting in a waste of resources. When the maintenance cycle is too short, the elevator is in a long-term over-maintenance state, and the elevator maintenance cost is high. Summary of the invention
[0004] The purpose of the present invention is to provide an elevator on-demand maintenance method and system based on Bayesian and risk sequence numbers, which can utilize the elevator's own monitoring data, avoid manual intervention to a lesser extent, realize elevator on-demand maintenance, and reduce the cost of elevator maintenance.
[0005] To achieve the above object, the present invention provides the following solutions:
[0006] An elevator on-demand maintenance method based on Bayesian and risk order number, characterized in that the elevator on-demand maintenance method comprises:
[0007] Get the functions of the elevators that need maintenance;
[0008] Determining potential failure modes of the functions based on the functions of the elevator requiring maintenance;
[0009] Obtain dynamic data of elevators that require maintenance;
[0010] Determining a safety risk severity score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data;
[0011] Determining the safety risk occurrence score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data;
[0012] Determining a safety risk difficulty-to-detect score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data;
[0013] Determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score, and the safety risk difficulty detection score;
[0014] Determining a threshold value of the elevator risk priority number;
[0015] Based on the elevator risk priority number and the threshold of the elevator risk priority number, determine whether maintenance measures need to be taken; when the elevator risk priority number is greater than the threshold of the elevator risk priority number, take maintenance measures; when the elevator risk priority number is less than or equal to the threshold of the elevator risk priority number, do not take maintenance measures.
[0016] Optionally, based on the potential failure mode of the function and the dynamic data, the specific steps of determining the occurrence score of the elevator safety risk requiring maintenance are as follows:
[0017] Obtain the past wear and tear data of elevators that require maintenance;
[0018] Based on the previous loss data of the elevator that needs maintenance, the predicted value s in the previous loss data is k and the observed value h k Approximately establish the observation equation;
[0019] Based on the observation equation, the predicted value s k After finding the mean and variance, construct the likelihood function;
[0020] Based on the current observation value h k and the previous observation h k-1 Establish the equation of state;
[0021] Based on the state equation, h k-1 After finding the mean and variance, construct the prior distribution function;
[0022] The probability of occurrence of the safety risk is determined based on the likelihood function, the prior distribution function and the Bayesian formula, and the safety risk occurrence score of the elevator requiring maintenance is determined based on the probability of occurrence of the safety risk.
[0023] Optionally, the observation equation is: k =ah k+b+θ, where a and b are the coefficients of the observation equation, θ is the residual, n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, and k=0,1,2…n.
[0024] Optionally, the likelihood function is: k (s k |h k )~N(ah k +b,σ 2 ), where f k is the likelihood function, ah k +b is s k The mean value, σ 2 For k The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0025] Optionally, the state equation is: k =a k-1 h k-1 +b k-1 +w k , where a k-1 、b k-1 is the coefficient of the state equation, w k is the residual of the observation equation, a k-1 、b k-1 and w k With h k-1 , n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0026] Optionally, the prior distribution function is: g k (h k |h0)~N(ch0+d,τ 2 ), where g k is the prior distribution function, c and d are fixed values, τ 2 h k-1 The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0027] Optionally, the calculation formula for the elevator risk priority number is: elevator risk priority number = safety risk severity score × safety risk occurrence score × safety risk difficulty detection score.
[0028] Optionally, the threshold for determining the elevator risk priority number is specifically: maximum elevator risk priority number×confidence level.
[0029] Optionally, the maximum elevator risk priority number calculation formula is: maximum elevator risk priority number = maximum safety risk severity score × maximum safety risk occurrence score × maximum safety risk difficulty detection score.
[0030] Based on the above method, the present invention also provides an elevator on-demand maintenance system based on Bayesian and risk sequence number, characterized in that the system comprises:
[0031] The elevator function acquisition module is used to obtain the functions of the elevators that need maintenance;
[0032] An elevator function potential failure mode determination module, connected to the elevator function acquisition module, for determining a potential failure mode of the function based on the function of the elevator requiring maintenance;
[0033] Elevator dynamic data acquisition module, used to obtain dynamic data of elevators that need maintenance;
[0034] an elevator safety risk severity score determination module, connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, and used to determine the elevator safety risk severity score requiring maintenance based on the potential failure mode of the function and the dynamic data;
[0035] an elevator safety risk occurrence score determination module, connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, and used to determine the elevator safety risk occurrence score that requires maintenance based on the potential failure mode of the function and the dynamic data;
[0036] An elevator safety risk difficulty detection score determination module, connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, for determining the elevator safety risk difficulty detection score that requires maintenance based on the potential failure mode of the function and the dynamic data;
[0037] an elevator risk priority number determination module, connected to the elevator safety risk severity score determination module, the elevator safety risk occurrence score determination module, and the elevator safety risk difficulty detection score determination module, and used to determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score, and the safety risk difficulty detection score;
[0038] A threshold determination module for the elevator risk priority number, used to determine the threshold of the elevator risk priority number;
[0039] The elevator maintenance judgment module is connected to the elevator risk priority number determination module and the elevator risk priority number threshold determination module, and is used to judge whether maintenance measures need to be taken based on the elevator risk priority number and the elevator risk priority number threshold; when the elevator risk priority number is greater than the elevator risk priority number threshold, maintenance measures are taken; when the elevator risk priority number is less than or equal to the elevator risk priority number threshold, no maintenance measures are taken.
[0040] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0041] The present invention provides an elevator on-demand maintenance method and system based on Bayesian and risk sequence numbers, which comprises the following steps: obtaining the function of an elevator requiring maintenance; determining the potential failure mode of the function based on the function of the elevator requiring maintenance; obtaining dynamic data of the elevator requiring maintenance; determining the safety risk severity score, the safety risk occurrence score and the safety risk difficulty detection score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data; determining the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score and the safety risk difficulty detection score; determining the threshold of the elevator risk priority number; judging whether maintenance measures need to be taken based on the elevator risk priority number and the threshold of the elevator risk priority number; taking maintenance measures when the elevator risk priority number is greater than the threshold of the elevator risk priority number; and not taking maintenance measures when the elevator risk priority number is less than or equal to the threshold of the elevator risk priority number. The present invention can avoid under-maintenance or over-maintenance of the elevator due to an overly long or short maintenance cycle, and utilizes the dynamic monitoring data of the elevator to achieve on-demand maintenance of the elevator, thereby reducing the maintenance cost of the elevator. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present invention 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 of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0043] Figure 1 This is a flow chart of an elevator on-demand maintenance method based on Bayesian and risk sequence number according to an embodiment of the present invention;
[0044] Figure 2 A flow chart for determining the safety risk occurrence score of the elevator requiring maintenance according to an embodiment of the present invention;
[0045] Figure 3 The present invention is a schematic diagram of the structure of an on-demand elevator maintenance system based on Bayesian and risk sequence number according to an embodiment of the present invention. DETAILED DESCRIPTION
[0046] 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.
[0047] The purpose of the present invention is to provide an elevator on-demand maintenance method and system based on Bayesian and risk sequence numbers, which can utilize the dynamic monitoring data of the elevator to achieve on-demand protection of the elevator.
[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0049] FEMA (Failure Mode and Effects Analysis) is often used in the product design stage. It analyzes the components of products, equipment, parts, and each process of the production process one by one to find out all failure modes and analyze their possible results, so as to take necessary measures. It can also be widely used in the evaluation of system design. The design of elevator maintenance plan based on FEMA is based on the equipment usage process data, predicting the possible failure mode of the equipment, and designing the system through the reliability, maintainability, and monitorability of the equipment under each failure mode. The Bayesian method is widely used in the field of reasoning and is considered to be one of the most effective theoretical models that can accurately evaluate the state of the object.
[0050] See attached picture Figure 1 The present invention provides an elevator on-demand maintenance method based on Bayesian and risk sequence number, and the elevator on-demand maintenance method comprises:
[0051] S1. Obtain the functions of the elevator that needs maintenance.
[0052] S2. Determine potential failure modes of the functions of the elevator that require maintenance based on the functions of the functions.
[0053] Specifically, when the elevator requiring maintenance is a traction sheave elevator, the function of the elevator requiring maintenance is the traction function, and a potential failure mode thereof is insufficient traction force.
[0054] S3. Obtain dynamic data of elevators that require maintenance.
[0055] Specifically, when the elevator requiring maintenance is a traction sheave elevator, the dynamic data of the traction force is obtained through the sensor data for detecting the acceleration of the elevator operation in the elevator.
[0056] S4. Determine the safety risk severity score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data.
[0057] Specifically, you can refer to the cases of relevant risk scenarios in GB 24803.1-2009 "Basic Safety Requirements for Elevators" or the risk severity evaluation method in GB / T 20900-2007 "Methods for Risk Assessment and Reduction of Elevators, Escalators and Moving Walkways", analyze the severity of elevator safety risks, and refer to Table 1 to score the severity of elevator safety risks to obtain the elevator safety risk severity score.
[0058] Table 1 Elevator safety risk severity score table
[0059]
[0060]
[0061] S5. See attached figure Figure 2 , based on the potential failure mode of the function and the dynamic data, determining the occurrence score of the elevator safety risk that requires maintenance, the specific steps are:
[0062] S51. Obtaining the previous wear and tear data of the elevator that needs maintenance.
[0063] S52, based on the previous loss data of the elevator that needs maintenance, the predicted value s in the previous loss data k and the observed value h k The observation equation is approximately established, and the observation equation is: k =ah k +b+θ, where a and b are the coefficients of the observation equation, θ is the residual, n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, and k=0,1,2…n.
[0064] S53, based on the observation equation, the predicted value s k After finding the mean and variance, the likelihood function is constructed, and the likelihood function is: f k (s k |h k )~N(ah k +b,σ 2 ), where f k is the likelihood function, ah k +b is s k The mean value, σ 2 For k The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0065] S54, based on the current observation value h k and the previous observation h k-1 Establish a state equation, the state equation is: k =a k-1 h k-1 +b k-1 +w k , where a k-1 , b k-1 is the coefficient of the state equation, w k is the residual of the observation equation, a k-1 , b k-1 and w k With h k-1 changes with the change of , n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0066] Specifically, the state equation reflects the random disturbance of the state variable.
[0067] S55. Based on the state equation, h k-1 After calculating the mean and variance, a priori distribution function is constructed, and the priori distribution function is: g k (h k |h0)~N(ch0+d,τ 2 ), where g k is the prior distribution function, c and d are fixed values, τ 2 h k-1 The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
[0068] S56. Determine the probability of occurrence of the safety risk based on the likelihood function, the prior distribution function and the Bayesian formula, and determine the safety risk occurrence score of the elevator requiring maintenance based on the safety risk occurrence probability.
[0069] Specifically, the steps for determining the probability of occurrence of the security risk are:
[0070] Determine the probability density of the security risk occurrence, the calculation formula of the probability density of the security risk occurrence is:
[0071]
[0072] Among them, h k is the current observation value, h k-1 is the observed value at the previous moment, h0 is the observed value at the initial moment, f k is the likelihood function, g kis the prior distribution function, φ is the observed value h k The probability density of safety risks under the above conditions.
[0073] Based on the safety risk probability density function, a probability distribution function is determined.
[0074] Specifically, the probability density function is indefinitely integrated to determine the probability distribution function.
[0075] Based on the probability distribution function and the observed value h k , determine the probability of safety risk occurrence, where k = 1…n, n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy.
[0076] Specifically, n is determined based on relevant technical standards and the actual situation of the maintenance company.
[0077] Specifically, based on the safety risk occurrence probability, the safety risk occurrence score of the elevator requiring maintenance is determined according to Table 2.
[0078] Specifically, when using it, the prior distribution of the state equation is known, and the posterior information of the observed value at a certain moment can be obtained, then the predicted value at a certain moment can be obtained by continuous recursion, that is, the predicted value s k The probability distribution function of the risk factors in Table 2 is used to quantitatively predict the probability of occurrence, rather than subjective assumptions, to obtain the elevator safety risk occurrence score.
[0079] The present invention takes a traction sheave elevator as an example to explain the above step S5 in detail. When the elevator requiring maintenance is a traction sheave elevator, the safety risk occurrence score of the elevator requiring maintenance is determined based on the potential failure mode of the function and the dynamic data. The specific steps are as follows:
[0080] Step 1: Fit the regression equation based on the previous data, that is, the observation equation can be approximately fitted based on the predicted value of traction s and the observed value of acceleration h of the data set obtained from the elevator sensor.
[0081] Specifically, an equation is fitted by actually detecting the traction force and acceleration, thereby obtaining an observation equation for the relationship between the traction force and acceleration.
[0082] Step 2: Obtain the state equation. The acceleration value at the current moment is fitted with the acceleration value at the previous moment to form the state equation.
[0083] Step three, construct the likelihood function of the observation equation and the state equation, that is, calculate the mean and variance of the two equations, and write out their normal distribution function (according to the central limit theorem, when there are enough data, the overall distribution tends to normal distribution).
[0084] Step 4: Substitute the two distribution functions into the Bayesian formula, obtain the initial prior distribution function based on prior knowledge or other theories, and then gradually recursively obtain the probability of the corresponding traction force appearing at a certain acceleration.
[0085] Step 5: Obtain the elevator safety risk occurrence score according to Table 2.
[0086] Table 2 Elevator safety risk occurrence score table
[0087] The likelihood of the risk occurring Probability Value Very high P≥30% 5 high 10<P≤30% 4 medium 1%<P≤10% 3 Low 0.1%<P≤1% 2 Can be ignored P<0.1% 1
[0088] S6. Based on the potential failure mode of the function and the dynamic data, determine the difficulty-to-detect score of the elevator safety risk that requires maintenance, as shown in Table 3 for details.
[0089] Specifically, based on the potential failure modes of the functions of the elevator requiring maintenance and the dynamic data, according to the actual operation of the maintenance company and referring to Table 3, the difficulty-to-detect score of the safety risk of the elevator requiring maintenance is determined.
[0090] Table 3 Elevator safety risk difficulty detection score table
[0091]
[0092]
[0093] Specifically, both Table 1 and Table 3 make judgments based on the harmfulness of corresponding risks in the elevator standards and the detection conditions in actual situations, and are less affected by human subjectivity.
[0094] S7. Determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score and the safety risk difficulty detection score; the elevator risk priority number calculation formula is: Elevator Risk Priority Number (RPN) = Safety Risk Severity Score (S) × Safety Risk Occurrence Score (O) × Elevator Safety Risk Difficulty Detection (D), the larger the RPN value, the greater the risk.
[0095] S8. Determine the threshold of the elevator risk priority number, and the threshold of the elevator risk priority number is determined specifically as follows: maximum elevator risk priority number × confidence level. The maximum elevator risk priority number is calculated as follows: maximum elevator risk priority number = maximum safety risk severity score × maximum safety risk occurrence score × maximum safety risk difficulty detection score.
[0096] S9. Based on the elevator risk priority number and the threshold of the elevator risk priority number, determine whether maintenance measures need to be taken; when the elevator risk priority number is greater than the threshold of the elevator risk priority number, take maintenance measures; when the elevator risk priority number is less than or equal to the threshold of the elevator risk priority number, do not take maintenance measures.
[0097] Specifically, the standard TSGT5002-2017 "Elevator Maintenance Rules" arranges different levels of maintenance plans for different maintenance intervals. Based on the elevator IoT dynamic monitoring system, when one of the data transmitted back by the monitoring system exceeds the threshold of the elevator risk priority number after calculation, the corresponding maintenance should be carried out immediately. If it does not exceed the threshold, the maintenance interval can be extended.
[0098] Specifically, when the elevator that needs maintenance is a traction wheel elevator, the traction failure score is S=5, D=1 according to the actual situation. The current probability of occurrence is calculated to be 0.5% by the Bayesian method. Therefore, according to Table 2, O=2, then RPN=10 is calculated. It is known that RPNmax, the maximum elevator risk priority number = maximum safety risk severity score × maximum safety risk occurrence score × maximum elevator risk difficulty score = 125, and the threshold of the elevator risk priority number = RPNmax×0.05=6.25. At this time, RPN>the threshold of the elevator risk priority number, so maintenance is required. In the present invention, O is a variable, and S and D remain basically unchanged after research and confirmation.
[0099] The threshold of the elevator risk priority number is determined based on the 95% confidence principle, and RPNmax×(1-0.95) is the threshold of the elevator risk priority number. That is to say, when the RPN value of a certain fault is greater than the threshold, corresponding maintenance measures should be taken.
[0100] Specifically, the confidence level may be determined according to the actual operation conditions of the maintenance company.
[0101] See attached picture Figure 3 Based on the above method, the present invention also provides an elevator on-demand maintenance system based on Bayes and risk sequence number, the system comprising: an elevator function acquisition module 1, an elevator function potential failure mode determination module 2, an elevator dynamic data acquisition module 3, an elevator safety risk severity score determination module 4, an elevator safety risk occurrence score determination module 5, an elevator safety risk difficulty detection score determination module 6, an elevator risk priority number determination module 7, an elevator risk priority number threshold determination module 8, and an elevator maintenance judgment module 9.
[0102] The elevator function acquisition module 1 is used to acquire the function of the elevator that needs maintenance.
[0103] The elevator function potential failure mode determination module 2 is connected to the elevator function acquisition module 1 and is used to determine the potential failure mode of the function based on the function of the elevator that needs maintenance.
[0104] The elevator dynamic data acquisition module 3 is used to acquire the dynamic data of the elevator that needs maintenance.
[0105] The elevator safety risk severity score determination module 4 is connected to the elevator function potential failure mode determination module 2 and the elevator dynamic data acquisition module 3, and is used to determine the elevator safety risk severity score that requires maintenance based on the potential failure mode of the function and the dynamic data.
[0106] The elevator safety risk occurrence score determination module 5 is connected to the elevator function potential failure mode determination module 2 and the elevator dynamic data acquisition module 3, and is used to determine the elevator safety risk occurrence score that requires maintenance based on the potential failure mode of the function and the dynamic data.
[0107] The elevator safety risk difficulty-detection score determination module 6 is connected to the elevator function potential failure mode determination module 2 and the elevator dynamic data acquisition module 3, and is used to determine the elevator safety risk difficulty-detection score that requires maintenance based on the potential failure mode of the function and the dynamic data.
[0108] The elevator risk priority number determination module 7 is connected to the elevator safety risk severity score determination module 4, the elevator safety risk occurrence score determination module 5, and the elevator safety risk difficulty detection score determination module 6, and is used to determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score, and the elevator safety risk difficulty detection score. The elevator risk priority number calculation formula is: elevator risk priority number = safety risk severity score × safety risk occurrence score × elevator safety risk difficulty detection.
[0109] The elevator risk priority number threshold determination module 8 is used to determine the elevator risk priority number threshold.
[0110] The elevator maintenance judgment module 9 is connected to the elevator risk priority number determination module 7 and the elevator risk priority number threshold determination module 8, and is used to determine whether maintenance measures need to be taken based on the elevator risk priority number and the threshold of the elevator risk priority number. When the elevator risk priority number is greater than the threshold of the elevator risk priority number, maintenance measures are taken; when the elevator risk priority number is less than or equal to the threshold of the elevator risk priority number, no maintenance measures are taken.
[0111] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0112] The present invention provides an elevator on-demand maintenance method and system based on Bayesian and risk sequence numbers. The Bayesian probability prediction method is used to quantitatively analyze the elevator operation status data transmitted back by the elevator Internet of Things and convert it into a probability state, so that it can objectively evaluate the probability of risk occurrence and avoid human subjective assumptions. In addition, the FEMA method can be used to comprehensively evaluate the operating status of the elevator system, increase the diversity of the entire evaluation system, fully consider its actual situation, and correct the shortcomings of traditional evaluation methods. The present invention also makes full use of the elevator safety status monitoring system based on the Internet of Things and makes full use of data resources.
[0113] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. An elevator on-demand maintenance method based on Bayesian and risk order number, characterized in that: The elevator on-demand maintenance method comprises: Get the functions of the elevators that need maintenance; Determining potential failure modes of the functions based on the functions of the elevator requiring maintenance; Obtain dynamic data of elevators that require maintenance; Determining a safety risk severity score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data; Based on the potential failure mode of the function and the dynamic data, the occurrence score of the elevator safety risk requiring maintenance is determined, and the specific steps are: Obtain the past wear and tear data of elevators that require maintenance; Based on the previous loss data of the elevator that needs maintenance, the predicted value s in the previous loss data is k and the observed value h k Approximately establish the observation equation; Based on the observation equation, the predicted value s k After finding the mean and variance, construct the likelihood function; Based on the current observation value h k and the previous observation h k-1 Establish the equation of state; Based on the state equation, h k-1 After finding the mean and variance, construct the prior distribution function; Determine the safety risk occurrence probability based on the likelihood function, the prior distribution function and the Bayesian formula, and determine the safety risk occurrence score of the elevator requiring maintenance based on the safety risk occurrence probability; Determining a safety risk difficulty-to-detect score of the elevator requiring maintenance based on the potential failure mode of the function and the dynamic data; Determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score, and the safety risk difficulty detection score; Determining a threshold value of the elevator risk priority number; Based on the elevator risk priority number and the threshold of the elevator risk priority number, determine whether maintenance measures need to be taken; when the elevator risk priority number is greater than the threshold of the elevator risk priority number, take maintenance measures; when the elevator risk priority number is less than or equal to the threshold of the elevator risk priority number, 10006 No maintenance measures are taken.
2. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 1 is characterized in that: The observation equation is: k =ah k +b+θ, where a and b are the coefficients of the observation equation, θ is the residual, n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, and k=0,1,2…n.
3. The elevator on-demand maintenance method based on Bayesian and risk order number according to claim 1 is characterized in that: The likelihood function is: k (s k |h k )~N(ah k +b,σ 2 ), where f k is the likelihood function, ah k +b is s k The mean value, σ 2 For k The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
4. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 1, characterized in that: The state equation is: k =a k-1 h k-1 +b k-1 +w k , where a k-1 、b k-1 is the coefficient of the state equation, w k is the residual of the observation equation, a k-1 、b k-1 and w k With h k-1 , n is the highest predicted value and number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
5. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 1, characterized in that: The prior distribution function is: k (h k |h0)~N(ch0+d,τ 2 ), where g k is the prior distribution function, c and d are fixed values, τ 2 h k-1 The variance of , n is the highest predicted value and the number of observed values within the maximum predictable range while maintaining accuracy, k = 0, 1, 2…n.
6. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 1, characterized in that: The calculation formula for the elevator risk priority number is: elevator risk priority number = safety risk severity score × safety risk occurrence score × safety risk difficulty detection score.
7. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 1, characterized in that: The threshold for determining the elevator risk priority number is specifically: maximum elevator risk priority number×confidence.
8. The on-demand elevator maintenance method based on Bayesian and risk order number according to claim 7, characterized in that: The calculation formula for the maximum elevator risk priority number is: maximum elevator risk priority number = maximum safety risk severity score × maximum safety risk occurrence score × maximum safety risk difficulty detection score.
9. An elevator on-demand maintenance system based on Bayesian and risk order number, characterized in that: The system comprises: The elevator function acquisition module is used to obtain the functions of the elevators that need maintenance; An elevator function potential failure mode determination module, connected to the elevator function acquisition module, for determining a potential failure mode of the function based on the function of the elevator requiring maintenance; Elevator dynamic data acquisition module, used to obtain dynamic data of elevators that need maintenance; an elevator safety risk severity score determination module, connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, and used to determine the elevator safety risk severity score requiring maintenance based on the potential failure mode of the function and the dynamic data; The elevator safety risk occurrence score determination module is connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, and is used to determine the elevator safety risk occurrence score that needs maintenance based on the potential failure mode of the function and the dynamic data, specifically including: Obtain the past wear and tear data of elevators that require maintenance; Based on the previous loss data of the elevator that needs maintenance, the predicted value s in the previous loss data is k and the observed value h k Approximately establish the observation equation; Based on the observation equation, the likelihood function is constructed after calculating the mean and variance of the predicted value sk; Based on the current observation value h k and the previous observation h k-1 Establish the equation of state; Based on the state equation, h k-1 After finding the mean and variance, construct the prior distribution function; Determine the safety risk occurrence probability based on the likelihood function, the prior distribution function and the Bayesian formula, and determine the safety risk occurrence score of the elevator requiring maintenance based on the safety risk occurrence probability; An elevator safety risk difficulty detection score determination module, connected to the elevator function potential failure mode determination module and the elevator dynamic data acquisition module, for determining the elevator safety risk difficulty detection score that requires maintenance based on the potential failure mode of the function and the dynamic data; an elevator risk priority number determination module, connected to the elevator safety risk severity score determination module, the elevator safety risk occurrence score determination module, and the elevator safety risk difficulty detection score determination module, and used to determine the elevator risk priority number based on the safety risk severity score, the safety risk occurrence score, and the elevator safety risk difficulty detection score; A threshold determination module for the elevator risk priority number, used to determine the threshold of the elevator risk priority number; The elevator maintenance judgment module is connected to the elevator risk priority number determination module and the elevator risk priority number threshold determination module, and is used to judge whether maintenance measures need to be taken based on the elevator risk priority number and the elevator risk priority number threshold; when the elevator risk priority number is greater than the elevator risk priority number threshold, maintenance measures are taken; when the elevator risk priority number is less than or equal to the elevator risk priority number threshold, no maintenance measures are taken.
