Elevator accident prediction method, equipment, medium and product

Abstract

The invention discloses an elevator accident prediction method and device, a medium and a product, and relates to the field of elevator accident prediction.The method comprises the steps that working parameters of a to-be-detected elevator are obtained; the working parameters comprise an operation parameter, a management parameter and a riding parameter; inputting the working parameters into an elevator accident prediction model to obtain a fault occurrence probability corresponding to the to-be-detected elevator; the elevator accident prediction model is a deep learning network; when the fault occurrence probability exceeds a preset threshold value, the elevator accident type is determined according to the fault occurrence probability; the elevator accident type serves as a target variable, and each working parameter serves as a characteristic variable; respectively calculating standardized mutual information values of the target variable and each characteristic variable, and determining a maximum standardized mutual information value; and determining a to-be-optimized fault combination according to the maximum standardized mutual information value. According to the invention, the elevator accident can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the field of elevator accident prediction, and in particular to an elevator accident prediction method, equipment, medium and product. Background Art

[0002] As an important vertical transportation tool for high-rise buildings, elevators play an increasingly important role in the national economy and people's daily lives. Elevators are complex mechatronics systems. As a special device, they not only need to be comfortable to ride and dock accurately, but also meet the requirements of speed and convenience, and try to shorten the waiting time for maintenance and the downtime of accidents.

[0003] Elevators play the same role as cars in high-rise buildings, but accidents involving elevators are more dangerous than accidents involving cars, seriously endangering the safety of human life and property. As a special equipment, the state has set strict requirements for the quality and safety of elevators. Although the reliability of modern elevators has been greatly improved, during operation, accidents such as falls, shearing / extrusion, and top impact still occur from time to time. Various places have successively issued elevator maintenance and use specifications, but there is still a lack of accident prediction and accident hazard prevention.

[0004] Therefore, how to design a method, device, medium and product that can accurately predict elevator accidents has become a technical problem that urgently needs to be solved in this field. Summary of the invention

[0005] The purpose of the present invention is to provide an elevator accident prediction method, device, medium and product, which can accurately predict elevator accidents.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] In a first aspect, the present invention provides an elevator accident prediction method, the elevator accident prediction method comprising:

[0008] Obtain the working parameters of the elevator to be tested; the working parameters include: operating parameters, management parameters and riding parameters; the operating parameters include: the elevator's operating speed, acceleration, the elevator's open and closed state, the sensor's state and reading, the elevator's use frequency and load capacity; the management parameters include: safety program files and operating instructions; the riding parameters include: operating behavior parameters and riding behavior parameters.

[0009] The working parameters are input into an elevator accident prediction model to obtain the corresponding failure probability of the elevator to be tested; the elevator accident prediction model is a deep learning network.

[0010] When the fault occurrence probability exceeds a preset threshold, the elevator accident type is determined according to the fault occurrence probability.

[0011] The elevator accident type is used as the target variable, and each of the working parameters is used as a characteristic variable.

[0012] The standardized mutual information value between the target variable and each of the feature variables is calculated respectively, and the maximum standardized mutual information value is determined.

[0013] A fault combination to be optimized is determined according to the maximum normalized mutual information value.

[0014] Optionally, the method for determining the elevator accident prediction model specifically includes:

[0015] Acquire a data set; the data set includes: each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case; the historical working parameters include: historical operating parameters, historical management parameters and historical riding parameters; the historical operating parameters include: the historical operating speed of the elevator, the historical acceleration, the historical opening and closing status of the elevator, the historical status and readings of the sensor, the historical frequency of use of the elevator and the historical load capacity; the management parameters include: historical safety program files and historical work instructions; the riding parameters include: historical operating behavior parameters and historical riding behavior parameters.

[0016] According to the data set, the failure probability corresponding to each type of elevator accident case is determined.

[0017] Taking the historical working parameters as input and the failure probability corresponding to the historical working parameters as output, the deep learning network is trained to obtain an elevator accident prediction model.

[0018] Optionally, obtain a data set, including:

[0019] Obtain each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case.

[0020] Characterization analysis and statistical analysis are performed on each type of elevator accident case and the historical operating parameters of the elevator corresponding to each type of elevator accident case to obtain a data set; the characterization analysis is used to associate the historical operating parameters of the elevator with the elevator failure and accident records; the statistical analysis is used to determine the probability pattern of accident occurrence and evaluate the frequency and conditions of accident occurrence under different conditions.

[0021] Optionally, according to the data set, determining the probability of occurrence corresponding to each type of elevator accident case specifically includes:

[0022] Identify all contributing factors for each type of elevator accident case.

[0023] All accident causes are coded and represented to obtain accident cause codes.

[0024] According to the accident cause codes, accident trees for different types of elevator accidents are established.

[0025] According to each of the accident trees, the occurrence probability of each top-level event is calculated; the top-level event is the starting point of the accident tree analysis; the occurrence probability of each top-level event is the failure occurrence probability of different types of elevator accident cases.

[0026] Optionally, according to each of the accident trees, the probability of occurrence of each top-level event is calculated, specifically including:

[0027] Identify top events.

[0028] Identify several direct causes that led to the top-level event.

[0029] According to each direct cause, the corresponding basic event is determined; the basic event is the bottom-level cause that leads to the accident.

[0030] A logical relationship is constructed based on the direct cause and the corresponding basic event.

[0031] A probability value is determined for each of the elementary events.

[0032] The occurrence probability of each of the top-level events is determined according to the probability value of each of the basic events and the logical relationship.

[0033] Optionally, the calculation formula of the normalized mutual information value is:

[0034]

[0035] Among them, NMI(X;Y) is the standardized mutual information value between the target variable and the feature variable, I(X;Y) is the mutual information between the target variable and the feature variable, H(X) is the entropy of the feature variable, H(Y) is the entropy of the target variable, X is the feature variable, and Y is the target variable.

[0036] Optionally, the elevator accident prediction method further includes:

[0037] According to the maximum normalized mutual information value, a risk assessment is performed on the elevator to be tested to obtain a risk assessment result.

[0038] Based on the risk assessment results, preventive measures and maintenance strategies are generated.

[0039] In a second aspect, the present invention provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described elevator accident prediction methods.

[0040] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described elevator accident prediction methods.

[0041] In a fourth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned elevator accident prediction methods.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] The present invention provides an elevator accident prediction method, device, medium and product. First, by inputting the working parameters of the elevator to be tested into the elevator accident prediction model, the corresponding fault probability of the elevator to be tested can be obtained; secondly, when the fault probability exceeds the preset threshold, the type of elevator accident can be determined; then, by mutual information analysis, the elevator accident type is used as the target variable, each working parameter is used as the characteristic variable, the maximum standardized mutual information value is determined, and the fault combination to be optimized is determined according to the maximum standardized mutual information value. The present invention uses a macro-micro risk fusion trigger mechanism to predict the probability of elevator accidents, which can integrate macro and micro risk factors, so as to more comprehensively capture the conditions and situations that may lead to accidents. Macro risks may include a wide range of operating environments and external conditions, such as elevator operating parameters and management parameters; while micro risks focus on specific components and operation levels, such as riding parameters. By integrating information at these two levels, more accurate risk assessment can be achieved, ensuring that the prediction system can trigger timely warnings, thereby more effectively preventing the occurrence of elevator accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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 work.

[0045] Figure 1 This is an application environment diagram of an elevator accident prediction method in one embodiment of the present invention.

[0046] Figure 2 A schematic flow chart of an elevator accident prediction method provided by an embodiment of the present invention.

[0047] Figure 3 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] 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.

[0049] 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.

[0050] The accuracy of current prediction models relies heavily on high-quality and large amounts of instance data. In many cases, the available data may be incomplete, outdated, or lack standardization, which in turn affects the accuracy of the prediction. In practice, elevator data may be contaminated by noise or contain outliers. The processing of such data requires complex preprocessing methods, otherwise the performance of the prediction model may be degraded.

[0051] The present invention uses a macro-micro risk fusion trigger mechanism to predict the probability of elevator accidents in order to integrate risk factors at the macro and micro levels, thereby more comprehensively capturing the conditions and situations that may lead to accidents. Macro risks may include a wide range of operating environments and external conditions, such as elevator usage frequency, maintenance history, building characteristics, etc., while micro risks focus on specific components and operational levels, such as minor failures in the elevator control system or operator behavior. By integrating information from these two levels, a more accurate risk assessment can be achieved, ensuring that the prediction system can trigger timely warnings, thereby more effectively preventing the occurrence of elevator accidents.

[0052] In summary, accident prediction of special equipment (taking elevators as an example) based on macro-micro analysis, combined with emergency decision-making for subsequent accidents, ultimately forms a recommended method for accident prediction and emergency decision-making for special equipment. Through typical accident cases and their emergency decision-making completion method verification, it provides the core technology of accident intelligent prediction and emergency decision-making for portable equipment for quality and safety inspection and supervision. Promote the scientific, systematic, digital and intelligent research on the recommended method of accident intelligent prediction and emergency decision-making for typical electromechanical special equipment.

[0053] The elevator accident prediction method provided by the embodiment of the present invention can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the working parameters of the elevator to be tested to the server 104. After the server 104 receives the working parameters of the elevator to be tested, for the working parameters of the elevator to be tested, the server 104 inputs the working parameters into the elevator accident prediction model to obtain the corresponding fault probability of the elevator to be tested; the elevator accident prediction model is a deep learning network; when the fault probability exceeds a preset threshold, the elevator accident type is determined according to the fault probability; the elevator accident type is used as the target variable, and each of the working parameters is used as a feature variable; the standardized mutual information value of the target variable and each of the feature variables is calculated respectively, and the maximum standardized mutual information value is determined; according to the maximum standardized mutual information value, the fault combination to be optimized is determined. The server 104 can feedback the obtained fault combination to be optimized to the terminal 102. In addition, in some embodiments, the elevator accident prediction method can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform elevator accident prediction based on the working parameters of the elevator to be tested, or the server 104 can obtain the working parameters of the elevator to be tested from the data storage system and perform elevator accident prediction based on the working parameters of the elevator to be tested.

[0054] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones and tablet computers, IoT devices and portable wearable devices. IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0055] In an exemplary embodiment, Figure 2 As shown, a method for predicting elevator accidents is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In an embodiment of the present invention, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the method, which includes the following steps S1 to S6. Among them:

[0056] S1: Obtain the working parameters of the elevator to be tested; the working parameters include: operating parameters, management parameters and riding parameters; the operating parameters include: the elevator's operating speed, acceleration, the elevator's open and closed state, the sensor's state and reading, the elevator's use frequency and load capacity; the management parameters include: safety program files and operating instructions; the riding parameters include: operating behavior parameters and riding behavior parameters.

[0057] S2: Input the working parameters into an elevator accident prediction model to obtain the corresponding failure probability of the elevator to be tested; the elevator accident prediction model is a deep learning network.

[0058] S3: When the probability of the fault occurring exceeds a preset threshold, the type of elevator accident is determined according to the probability of the fault occurring.

[0059] S4: taking the elevator accident type as the target variable and each of the working parameters as the characteristic variable;

[0060] S5: Calculate the standardized mutual information value between the target variable and each of the feature variables respectively, and determine the maximum standardized mutual information value.

[0061] S6: Determine the fault combination to be optimized according to the maximum normalized mutual information value.

[0062] The implementation of the above steps S1 to S6 can predict the probability of elevator accidents and the causes of key accidents, which is of great significance for digital intelligent analysis and prediction of elevator accidents. Compared with the prior art, the present invention has the following main advantages:

[0063] 1. More comprehensive. Existing technologies usually focus on one of macro-statistics or micro-analysis. For example, existing accident prediction methods may only rely on historical accident data to predict future risks, while ignoring potential system failures or micro-level details.

[0064] 2. Higher accuracy: Macro data helps identify overall trends, while micro data helps identify specific potential failure points. The present invention has higher accuracy.

[0065] 3. This method can develop more targeted maintenance strategies and reduce the probability of accidents.

[0066] In another exemplary embodiment of the present invention, in step S1, the safety program file includes: maintenance records, annual inspection and safety inspection reports, operation logs, fault and repair records, parts replacement records, operator training records, technical parameters and elevator configuration files, management specifications of the user unit or property, etc.

[0067] In another exemplary embodiment of the present invention, in step S2, the method for determining the elevator accident prediction model specifically includes:

[0068] S21: Obtain a data set; the data set includes: each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case; the historical working parameters include: historical operating parameters, historical management parameters and historical riding parameters; the historical operating parameters include: the historical operating speed of the elevator, the historical acceleration, the historical opening and closing status of the elevator, the historical status and readings of the sensor, the historical usage frequency of the elevator and the historical load capacity; the management parameters include: historical safety program files and historical work instructions; the riding parameters include: historical operating behavior parameters and historical riding behavior parameters.

[0069] S22: Determine the failure occurrence probability corresponding to each type of elevator accident case according to the data set.

[0070] S23: Taking the historical working parameters as input and the failure probability corresponding to the historical working parameters as output, the deep learning network is trained to obtain an elevator accident prediction model.

[0071] In another exemplary embodiment of the present invention, in step S21, the following steps are specifically included:

[0072] S211: Obtain each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case.

[0073] S212: Performing characterization analysis and statistical analysis on each type of elevator accident case and the historical operating parameters of the elevator corresponding to each type of elevator accident case to obtain a data set; the characterization analysis is used to associate the historical operating parameters of the elevator with the elevator failure and accident records; the statistical analysis is used to determine the probability pattern of accident occurrence and evaluate the frequency and conditions of accident occurrence under different conditions.

[0074] Specifically, based on the database of special equipment accident causes, we collected recent elevator accident case reports and organized them into Excel data sets. Characteristic analysis includes: associating the general properties of elevator operation, such as frequency of use, passenger capacity, and operating speed, with elevator failure and accident records; statistical analysis includes: using these data to determine the probability pattern of accident occurrence, and evaluating the frequency and conditions of accident occurrence under different conditions, such as the accident rate within a specific time period, or the failure rate associated with a specific maintenance activity.

[0075] In another exemplary embodiment of the present invention, in step S22, specifically including:

[0076] S221: Identify all accident causes for each type of elevator accident case.

[0077] S222: Encode and characterize all accident causes to obtain accident cause codes.

[0078] S223: Establishing accident trees for different types of elevator accidents according to the accident cause codes.

[0079] S224: According to each of the accident trees, the occurrence probability of each top-level event is calculated; the top-level event is the starting point of the accident tree analysis; the occurrence probability of each top-level event is the failure occurrence probability of different types of elevator accident cases.

[0080] Specifically, based on the elevator accident case database, all possible causes of accidents are statistically enumerated, and all accident causes (bottom events) are encoded and represented to facilitate the subsequent construction of the accident tree, and the existence probability of each factor in each accident is analyzed. Based on all possible accident causes, an accident tree of different types of elevator accidents is established, and the minimum cut set of the accident tree is calculated using Python. After the minimum cut set is disjointly made, the probability of occurrence of the top event of each tree is calculated, that is, the probability of occurrence of different types of elevator accidents.

[0081] In this embodiment, the causes of the accident include: direct causes, indirect causes, fundamental causes and root causes; the direct causes include: the cause person's illegal operation behavior, the cause person's illegal riding behavior, the cause person's riding behavior that does not violate the rules but causes the accident, the cause person's illegal command behavior, the cause object is in an illegal state, the cause object is in a state that does not violate the rules but causes the accident, environmental illegal conditions, the perpetrator's illegal operation behavior, the perpetrator's unlicensed work behavior, the perpetrator's illegal command behavior, the perpetrator's illegal state; the indirect causes include: insufficient safety knowledge, low safety awareness, and poor product and service awareness of suppliers; the root causes include: lack of safety guiding ideology, lack of safety management and use system, failure to establish or imperfect or lack of safety procedure documents, failure to establish or imperfect or lack of work instructions; the root causes include: poor safety management system, inadequate safety training, inadequate physical hidden danger inspection and poor safety performance.

[0082] In another exemplary embodiment of the present invention, in step S224, specifically including:

[0083] S2241: Identify top-level events.

[0084] S2242: Determine several direct causes that lead to the occurrence of the top-level event.

[0085] S2243: Determine a corresponding basic event based on each direct cause; the basic event is the most basic cause that leads to the accident.

[0086] S2244: Construct a logical relationship based on the direct cause and the corresponding basic event.

[0087] S2245: Determine the probability value of each of the basic events.

[0088] S2246: Determine the occurrence probability of each of the top-level events according to the probability value of each of the basic events and the logical relationship.

[0089] In this embodiment, a logical structure of fault tree analysis identification and visual display of elevator failures is established, which is decomposed from the top-level event (such as elevator accident) to the basic causes that may cause the event (such as electrical failure, mechanical failure, maintenance error, etc.). The detailed steps of the fault tree analysis method (FAT) are as follows:

[0090] A1: Identify the top events.

[0091] The top-level event is the starting point of the accident tree analysis, which represents the main failure or accident of concern, such as "elevator fall", "elevator trapped", etc. Identifying the top-level event is the first step in constructing the accident tree.

[0092] A2: Identify the immediate cause.

[0093] Identify the direct causes that lead to the top-level events, which are considered as the first-level events. For example, the direct causes of an elevator fall may include "brake system failure", "elevator overload", "control system failure", etc.

[0094] A3: Break it down into basic events.

[0095] For each direct cause, further trace the causes behind it until you reach the basic event that cannot be further subdivided. The basic event is the lowest level cause of the accident, which may include "brake wear", "overload sensor failure", "sudden power failure", etc.

[0096] A4: Build logical relationships.

[0097] The events in the accident tree are connected by logic gates (such as AND gates, OR gates) to represent the logical relationship between different events. For example, if the top event is caused by two direct causes, the two direct causes are connected to the top event through an AND gate. If the top event can be caused by any one of multiple causes alone, these causes are connected to the top event through an OR gate.

[0098] A5: Quantitative analysis.

[0099] By assigning probability values ​​to the elementary events in the accident tree, the probability of the top-level event occurring can be calculated. This requires sufficient data to estimate the probability of each elementary event occurring.

[0100] In addition, based on the accident tree analysis method, it also includes:

[0101] A6: Identify the critical path.

[0102] Accident tree analysis can identify the critical paths leading to the top events, that is, the sequence of events that is most likely to lead to an accident. This helps determine which basic events have the greatest impact on system safety.

[0103] A7 Develop mitigation measures.

[0104] Based on the analysis results, preventive measures can be formulated for events on the critical path to reduce the probability of accidents. This may include improving design, adding redundant systems, strengthening maintenance and inspection, etc.

[0105] In addition, the present invention also determines the causes of elevator key accidents and takes precautions according to the size of the standardized mutual information NMI. The present invention uses mutual information analysis to evaluate the degree of interdependence between faults, thereby identifying which fault combinations are most likely to cause system failure.

[0106] The detailed steps of mutual information (MI) are as follows: I(X; Y) represents mutual information, that is, the amount of information one variable contains about another variable:

[0107] I(X;Y)=H(X)-H(X|Y);

[0108] The calculation formulas for H(X) and H(X|Y) are as follows:

[0109] H(X)=-∑ x∈X p(x)log2p(x);

[0110] H(X|Y)=-∑ x∈X ∑ y∈Y p(x, y)log2p(x|y);

[0111] Arranging the above formula, we can get:

[0112]

[0113] Among them, H(X) represents the entropy of discrete variables, and H(X|Y) represents the entropy of the conditional probability distribution of variable X given variable Y. Entropy is a measure of the uncertainty of the variable. The larger the entropy, the greater the uncertainty of the variable. Mutual information can be interpreted as the amount by which the uncertainty of variable X is reduced due to the introduction of variable Y. Therefore, mutual information can be used to describe the degree of correlation between variables X and Y, and it can also reflect the nonlinear correlation between feature variables. The closer the relationship between variables X and Y, the greater the reduction, the larger I(X; Y), and the closer the relationship between variables X and Y. When I(X; Y) is equal to zero, variables X and Y are independent of each other. p(x), p(y) are marginal distributions, p(x, y) is a joint distribution; p(x|y) is the probability distribution of variable x conditional given variable y. In this embodiment, X is a feature variable and Y is a target variable.

[0114] By limiting the mutual information to [0,1], the standardized mutual information value NMI can be obtained. The calculation formula is:

[0115]

[0116] Among them, NMI(X;Y) is the standardized mutual information value between the target variable and the feature variable, I(X;Y) is the mutual information between the target variable and the feature variable, H(X) is the entropy of the feature variable, H(Y) is the entropy of the target variable, X is the feature variable, and Y is the target variable.

[0117] Mutual information can reveal the interdependence between elevator operating parameters and accident occurrence. The following are the steps and methods for using mutual information for analysis:

[0118] B1: Data collection and preprocessing.

[0119] First, it is necessary to collect various data related to elevator operation, including but not limited to:

[0120] B11: Elevator running speed and acceleration.

[0121] B12: The open and closed status of the elevator.

[0122] B13: Status and readings of important sensors (e.g., overload sensor, door lock sensor, position sensor, etc.).

[0123] B14: Elevator usage frequency and load capacity.

[0124] B15: The time and type of failure or accident.

[0125] The collected data is pre-processed to clean outliers, fill in missing values, and may need to be standardized for easier analysis.

[0126] B2: Select the target variable.

[0127] Determine the target of the analysis, that is, the type of elevator accident you want to predict or analyze. These can be specific accident types such as elevator entrapment, elevator fall, door failure, etc. These accident types are used as target variables (Y) and other elevator operation parameters are used as feature variables (X).

[0128] B3: Calculate the normalized mutual information value.

[0129] The standardized mutual information value is calculated using the formula for the standardized mutual information value between the target variable (Y) and each feature variable (X). The standardized mutual information value can quantify the amount of uncertainty reduction of another variable under the premise of knowing the information of one variable. A high standardized mutual information value means that there is a strong mutual dependence between the feature variable (such as elevator speed) and the accident type.

[0130] B4: Analysis and interpretation.

[0131] B41 Feature selection: By comparing the mutual information values ​​between different feature variables and the target variable, those variables that are most relevant to the type of accident can be selected. This helps to identify the key factors that affect the occurrence of accidents.

[0132] B42 Pattern Recognition: By observing which parameters have higher mutual information values ​​with accident types, some non-intuitive causal relationships or correlations can be revealed, such as a specific operating mode may be related to the occurrence of an accident.

[0133] B43 Risk assessment: Based on the mutual information value, the risk of specific types of accidents occurring in elevators under different operating conditions can be assessed.

[0134] In another exemplary embodiment of the present invention, the elevator accident prediction method further includes:

[0135] S7: performing risk assessment on the elevator to be tested according to the maximum normalized mutual information value to obtain a risk assessment result.

[0136] S8: Generate preventive measures and maintenance strategies based on the risk assessment results.

[0137] In other words, the analysis using mutual information also includes:

[0138] B5: Implement preventive measures.

[0139] Based on the analysis results, targeted preventive measures and maintenance strategies can be developed, such as:

[0140] Set safety thresholds for operating parameters based on key factors.

[0141] Optimize maintenance and inspection schedules, paying special attention to those components and parameters associated with a high risk of accidents.

[0142] Implement real-time monitoring and early warning systems to provide early warning of operating conditions that may lead to accidents.

[0143] B6: Continuous monitoring.

[0144] Finally, through continuous data collection and analysis, accident-related factors and risks can be identified more accurately, thereby improving the safety and reliability of elevators.

[0145] The present invention provides an elevator accident prediction method based on macro-micro analysis, which can predict the probability of elevator accidents and key accident causes, and is of great significance for digital intelligent analysis and prediction of elevator accidents. From a macro perspective, all possible accident causes are statistically listed based on the elevator accident case library, and all accident causes are encoded to facilitate the subsequent extraction of bottom events for constructing an accident tree and analyzing the existence probability of each factor in each accident; from a micro perspective, accident trees for different types of elevator accidents are established based on all possible accident causes, and the probability of occurrence of different types of accidents is calculated. In addition, based on the accident causes obtained from the macro analysis, the key causes of elevator accidents are determined in the form of correlation through the mutual information method, providing suggestions for subsequent accident prevention and emergency decision-making.

[0146] The present invention also provides an application scenario, which applies the above-mentioned elevator accident prediction method. Specifically: The elevator accident prediction method provided in this embodiment can be applied in the elevator accident prediction scenario. The elevator accident prediction scenario includes: parameter acquisition link, fault probability determination link and fault combination determination link to be optimized; the acquired working parameters of the elevator to be tested are input into the elevator accident prediction model to obtain the fault probability corresponding to the elevator to be tested, and then the elevator accident type is determined according to the fault probability, and finally the fault combination to be optimized can be determined by mutual information analysis method.

[0147] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 3As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the working parameters of the elevator to be tested. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting elevator accidents is implemented.

[0148] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the above-mentioned method embodiments are implemented when the processor executes the computer program.

[0150] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the above-mentioned method embodiments are implemented.

[0151] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the above-mentioned method embodiments are implemented.

[0152] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0153] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0154] The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0155] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0156] 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. A method for predicting elevator accidents, characterized in that: The elevator accident prediction method comprises: Obtaining the working parameters of the elevator to be tested; the working parameters include: operating parameters, management parameters and riding parameters; the operating parameters include: the running speed, acceleration, the opening and closing state of the elevator, the state and reading of the sensor, the frequency of use and the load of the elevator; the management parameters include: safety program files and operating instructions; the riding parameters include: operating behavior parameters and riding behavior parameters; Inputting the working parameters into an elevator accident prediction model to obtain a corresponding fault probability of the elevator to be tested; the elevator accident prediction model is a deep learning network; When the probability of the fault occurring exceeds a preset threshold, determining the type of elevator accident according to the probability of the fault occurring; The elevator accident type is used as a target variable, and each of the working parameters is used as a characteristic variable; Calculating the standardized mutual information value of the target variable and each of the feature variables respectively, and determining the maximum standardized mutual information value; A fault combination to be optimized is determined according to the maximum normalized mutual information value.

2. The elevator accident prediction method according to claim 1, characterized in that: The method for determining the elevator accident prediction model specifically includes: Acquire a data set; the data set includes: each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case; the historical working parameters include: historical operating parameters, historical management parameters and historical riding parameters; the historical operating parameters include: the historical operating speed of the elevator, the historical acceleration, the historical opening and closing state of the elevator, the historical state and reading of the sensor, the historical frequency of use of the elevator and the historical load capacity; the management parameters include: historical safety program files and historical work instructions; the riding parameters include: historical operating behavior parameters and historical riding behavior parameters; Determine the probability of failure occurrence corresponding to each type of elevator accident case according to the data set; Taking the historical working parameters as input and the failure probability corresponding to the historical working parameters as output, the deep learning network is trained to obtain an elevator accident prediction model.

3. The elevator accident prediction method according to claim 2, characterized in that: Get the dataset, including: Obtain each type of elevator accident case and the historical working parameters of the elevator corresponding to each type of elevator accident case; Characterization analysis and statistical analysis are performed on each type of elevator accident case and the historical operating parameters of the elevator corresponding to each type of elevator accident case to obtain a data set; the characterization analysis is used to associate the historical operating parameters of the elevator with the elevator failure and accident records; the statistical analysis is used to determine the probability pattern of accident occurrence and evaluate the frequency and conditions of accident occurrence under different conditions.

4. The elevator accident prediction method according to claim 2, characterized in that: According to the data set, the probability of failure corresponding to each type of elevator accident case is determined, including: Identify all accident contributing factors for each type of elevator accident case; All accident causes are coded and represented to obtain accident cause codes; According to the accident cause codes, establish accident trees for different types of elevator accidents; According to each of the accident trees, the occurrence probability of each top-level event is calculated; the top-level event is the starting point of the accident tree analysis; the occurrence probability of each top-level event is the failure occurrence probability of different types of elevator accident cases.

5. The elevator accident prediction method according to claim 4, characterized in that: According to each of the accident trees, the probability of occurrence of each top-level event is calculated, including: Identify top events; Identify several direct causes that led to the occurrence of the top-level event; According to each direct cause, determine the corresponding basic event; the basic event is the lowest level cause that leads to the accident; Constructing a logical relationship based on the direct cause and the corresponding basic event; determining a probability value for each of the elementary events; The occurrence probability of each of the top-level events is determined according to the probability value of each of the basic events and the logical relationship.

6. The elevator accident prediction method according to claim 1, characterized in that: The calculation formula of the normalized mutual information value is: Among them, NMI(X;Y) is the standardized mutual information value between the target variable and the feature variable, I(X;Y) is the mutual information between the target variable and the feature variable, H(X) is the entropy of the feature variable, H(Y) is the entropy of the target variable, X is the feature variable, and Y is the target variable.

7. The elevator accident prediction method according to claim 1, characterized in that: The elevator accident prediction method further includes: Performing risk assessment on the elevator to be tested according to the maximum normalized mutual information value to obtain a risk assessment result; Based on the risk assessment results, preventive measures and maintenance strategies are generated.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the elevator accident prediction method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the elevator accident prediction method described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the elevator accident prediction method described in any one of claims 1 to 7 is implemented.

Images