CN110498314B - Health assessment method and system for elevator door system, electronic device and storage medium - Google Patents

Abstract

The invention discloses a health assessment method and system of an elevator door system, electronic equipment and a storage medium, wherein the health assessment method comprises the following steps: acquiring corresponding first historical operation data when the elevator door system is in a normal operation state; acquiring current operation data of an elevator door system in a current operation state; acquiring a first target model and a second target model; acquiring a first contact ratio between a current operation state and a normal operation state; and determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio. According to the invention, the multidimensional operation data of the elevator door system is monitored at the same time, and the real-time evaluation of the health state of the elevator door system is realized based on the Gaussian mixture model, so that the evaluation accuracy of the health state of the elevator door system is improved; in addition, a health state warning mechanism of the elevator door system is provided, so that a user or a maintenance worker can be timely reminded of carrying out proper maintenance measures on the elevator door in advance, and further the elevator door is prevented from being broken down.

Description

Health assessment method and system for elevator door system, electronic device and storage medium

Technical Field

The invention relates to the technical field of elevator management, in particular to a health assessment method and system for an elevator door system, electronic equipment and a storage medium.

Background

With the increase of high-rise buildings, elevators have become indispensable vertical transportation vehicles for people day by day. Through years of development, the national elevator holding capacity is greatly increased and is increasingly saturated. As the running time of the elevator accumulates, the probability of failure of the elevator increases significantly. Because the opening and closing actions of the elevator door motor are very frequent, the elevator door system is easy to break down, and the frequent failure of the elevator door system becomes the most main component of elevator accidents, wherein more than 80 percent of elevator failures and more than 70 percent of elevator failures are caused by the failure of the elevator door system; failure of the elevator door system can have serious consequences which are difficult to imagine, so that the operating state of the elevator door system needs to be monitored effectively in time.

At present, the treatment process that the elevator at home and abroad is maintained to an elevator door system through maintenance personnel comprises: 1) when the gantry crane fails, determining a failure source, and maintaining or replacing a failed part; 2) whether a door system fault occurs or not, the maintenance is carried out according to a specified flow at a fixed period. Such maintenance methods have the following problems: when the elevator door system has certain performance degradation, the performance degradation cannot be found in time, and the elevator door system is not maintained; or the door machine is maintained blindly according to the unified flow no matter how the actual situation is, so that the door machine accident is still possibly caused, unnecessary ladder stopping and maintenance cost is caused, and the labor cost is high.

In addition, the current sensor is used for acquiring a real-time current signal of the elevator door system at present, and whether the elevator door system has a fault is evaluated by judging whether the real-time current signal is in a normal threshold range, so that the maintenance mode has the problems of low evaluation accuracy and the like.

Disclosure of Invention

The invention aims to solve the technical problem that the evaluation method of the health state of the elevator door system in the prior art has the defect of low evaluation accuracy, and aims to provide a health evaluation method, a health evaluation system, electronic equipment and a storage medium of the elevator door system.

The invention solves the technical problems through the following technical scheme:

the invention provides a health assessment method of an elevator door system, which comprises the following steps:

s1, acquiring corresponding first historical operation data of an elevator door system in a normal operation state within a historical sampling time period;

s2, acquiring current operation data of the elevator door system in a current sampling time period in a current operation state;

s3, respectively taking the first historical operation data and the current operation data as training parameters, inputting the training parameters into a target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

s4, acquiring a first contact ratio between the current running state and the normal running state according to the first target model and the second target model;

s5, determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health.

Preferably, when the target model comprises a gaussian mixture model, the first target model is a first gaussian mixture model, and the second target model is a second gaussian mixture model;

step S4 includes:

s41, calculating according to the first Gaussian mixture model and the second Gaussian mixture model to obtain a first contact ratio between the current operation state and the normal operation state.

Preferably, when the first historical operating data includes a duration of the door opening process or a duration of the door closing process, and the mechanical energy average value, the step S1 specifically includes:

s11, acquiring the total times of door opening or closing of the elevator door system in a historical preset time period, and the first times of door opening or door closing when the holding time length meets a preset time length range;

s12, calculating the ratio of the first times to the total times;

s13, judging whether the ratio exceeds a first set threshold value or not, and if so, determining the preset time period as a historical intermediate time period;

s14, acquiring a historical target time period according to the mechanical energy average value corresponding to each historical intermediate time period;

and S15, taking the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in a normal operation state.

Preferably, step S14 includes:

sorting the mechanical energy average values corresponding to the historical intermediate time periods according to the sizes, and selecting the historical intermediate time period corresponding to the smallest mechanical energy average value as the historical target time period;

wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system; and/or the presence of a gas in the gas,

the historical preset time period is in days.

Preferably, after the step S15 and before the step S3, the method further comprises:

acquiring a corresponding first data matrix according to the first historical operating data;

carrying out standardization processing on the first data matrix to obtain a first characteristic matrix;

acquiring a corresponding second data matrix according to the current operation data;

carrying out standardization processing on the second data matrix to obtain a second feature matrix;

step S3 further includes:

and respectively taking the first characteristic matrix and the second characteristic matrix as training parameters, inputting the training parameters into the target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

Preferably, the first gaussian mixture model or the second gaussian mixture model comprises:

wherein g (x) represents the first Gaussian mixture model or the second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing d dimensions of the first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector μ_iSum covariance matrix sigma_i。

Preferably, in step S41, the first contact ratio is calculated by the following formula:

wherein CV represents the first degree of coincidence, g₁(x1) represents the first Gaussian mixture model, g₂(x2) represents the second Gaussian mixture model, x1 represents the first feature matrix, and x2 represents the second feature matrix.

Preferably, the health assessment method further comprises:

presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period respectively;

step S5 is followed by:

acquiring second historical operating data corresponding to the sub-health state in the historical sampling time period according to the first label;

acquiring third history operation data corresponding to the fault state in the history sampling time period according to the second label;

a third data matrix corresponding to the second historical operating data;

carrying out standardization processing on the third data matrix to obtain a third feature matrix;

acquiring a corresponding fourth data matrix according to the third history operation data;

carrying out standardization processing on the fourth data matrix to obtain a fourth feature matrix;

inputting the third feature matrix as a training parameter into a Gaussian mixture model for training to obtain a third Gaussian mixture model for representing the running condition of the elevator door system in a sub-health state;

inputting the fourth feature matrix as a training parameter into a Gaussian mixture model for training to obtain a fourth Gaussian mixture model for representing the running condition of the elevator door system in a fault state;

calculating a second overlap ratio between the sub-health state and the normal operation state according to the first Gaussian mixture model and the third Gaussian mixture model;

calculating according to the first Gaussian mixture model and the fourth Gaussian mixture model to obtain a third coincidence ratio between the fault state and the normal operation state;

setting a first early warning value according to the maximum value of the second superposition degrees;

wherein the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

setting a second early warning value according to the maximum value of the third overlap ratios;

and the second early warning value is used for early warning the condition that the door opening or closing fault occurs in the elevator door system.

Preferably, the health assessment method further comprises:

performing mean value filtering processing on the first coincidence degree by adopting sliding windows with a first width and a second width to obtain a fourth coincidence degree and a fifth coincidence degree corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period and the second width corresponds to a short period;

when the fourth overlap ratio is smaller than the first early warning value and the fifth overlap ratio is greater than or equal to the second early warning value, generating first warning information for representing that the door opening or closing function of the elevator door system is reduced;

and when the fifth overlap ratio is smaller than the second early warning value, generating second warning information for representing that the door opening or closing fault of the elevator door system occurs.

Preferably, the first historical operating data, the second historical operating data, the third historical operating data, and the current operating data each include at least one of gating signal data, current data, electrical energy data, power data, and speed data.

The invention also provides a health evaluation system of the elevator door system, which comprises a first historical data acquisition module, a current data acquisition module, a model acquisition module, a first contact ratio acquisition module and a health degree determination module;

the first historical data acquisition module is used for acquiring corresponding first historical operation data when the elevator door system is in a normal operation state within a historical sampling time period;

the current data acquisition module is used for acquiring current operation data of the elevator door system in a current sampling time period in a current operation state;

the model acquisition module is used for inputting the first historical operation data and the current operation data as training parameters into a target model for training to acquire a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

the first contact ratio obtaining module is used for obtaining a first contact ratio between the current operation state and the normal operation state according to the first target model and the second target model;

the health degree determining module is used for determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health.

Preferably, when the target model comprises a gaussian mixture model, the first target model is a first gaussian mixture model, and the second target model is a second gaussian mixture model;

the first contact ratio obtaining module is used for calculating and obtaining a first contact ratio between the current operation state and the normal operation state according to the first Gaussian mixture model and the second Gaussian mixture model.

Preferably, when the first historical operating data includes a holding time of a door opening process or a holding time of a door closing process, and a mechanical energy average value, the first historical data acquiring module includes a frequency acquiring unit, a ratio calculating unit, a judging unit, a target time period acquiring unit, and a historical data acquiring unit;

the frequency acquisition unit is used for acquiring the total frequency of opening or closing the door of the elevator door system in a historical preset time period and the first frequency of opening or closing the door with the holding time length meeting a preset time length range;

the ratio calculation unit is used for calculating the ratio of the first times to the total times;

the judging unit is used for judging whether the ratio exceeds a first set threshold value or not, and if so, determining the preset time period as a historical intermediate time period;

the target time period obtaining unit is used for obtaining historical target time periods according to the mechanical energy average value corresponding to each historical intermediate time period;

the historical data acquisition unit is used for taking the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in the normal operation state.

Preferably, the target time period obtaining unit is configured to sort the mechanical energy average value corresponding to each historical intermediate time period according to size, and select the historical intermediate time period corresponding to the smallest mechanical energy average value as the historical target time period;

wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system; and/or the presence of a gas in the gas,

the historical preset time period is in days.

Preferably, the health assessment system further comprises a first feature matrix acquisition module and a second feature matrix acquisition module;

the first characteristic matrix acquisition module is used for acquiring a corresponding first data matrix according to the first historical operating data and carrying out standardization processing on the first data matrix to acquire a first characteristic matrix;

the second characteristic matrix acquisition module is used for acquiring a corresponding second data matrix according to the current operating data and carrying out standardization processing on the second data matrix to acquire a second characteristic matrix;

the model acquisition module is used for inputting the first characteristic matrix and the second characteristic matrix as training parameters into the target model for training to acquire a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

Preferably, the first gaussian mixture model or the second gaussian mixture model comprises:

wherein g (x) represents the first Gaussian mixture model or the second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing d dimensions of the first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector μ_iSum covariance matrix sigma_i。

Preferably, the first contact ratio obtaining module calculates the first contact ratio by using the following formula:

wherein CV represents the first degree of coincidence, g₁(x1) represents the first Gaussian mixture model, g₂(x2) represents the second Gaussian mixture model, x1 represents the first feature matrix, and x2 represents the second feature matrix.

Preferably, the health assessment system is used for a label presetting module, a second historical data acquisition module, a third feature matrix acquisition module, a fourth feature matrix acquisition module, a second coincidence degree acquisition module, a third coincidence degree acquisition module and an early warning value setting module;

the label presetting module is used for respectively presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period;

the second historical data acquisition module is used for acquiring second historical operating data of the corresponding sub-health state in the historical sampling time period according to the first label;

the third history data acquisition module is used for acquiring third history operation data of a corresponding fault state in the history sampling time period according to the second label;

the third feature matrix acquisition module is used for carrying out standardization processing on a third data matrix according to the third data matrix corresponding to the second historical operating data to acquire a third feature matrix;

the fourth feature matrix acquisition module is used for acquiring a corresponding fourth data matrix according to the third history operation data, and carrying out standardization processing on the fourth data matrix to acquire a fourth feature matrix;

the model acquisition module is further used for inputting the third feature matrix as a training parameter into a Gaussian mixture model for training to acquire a third Gaussian mixture model for representing the running condition of the elevator door system in a sub-health state;

the model acquisition module is further used for inputting the fourth feature matrix as a training parameter into a Gaussian mixture model for training to acquire a fourth Gaussian mixture model for representing the running condition of the elevator door system in a fault state;

the second overlap ratio acquisition module is used for calculating a second overlap ratio between the sub-health state and the normal operation state according to the first Gaussian mixture model and the third Gaussian mixture model;

the third coincidence degree obtaining module is used for calculating a third coincidence degree between the fault state and the normal operation state according to the first Gaussian mixture model and the fourth Gaussian mixture model;

the early warning value setting module is used for setting a first early warning value according to the maximum value of the second superposition degrees;

wherein the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

the early warning value setting module is further used for setting a second early warning value according to the maximum value of the third overlap ratios;

and the second early warning value is used for early warning the condition that the door opening or closing fault occurs in the elevator door system.

Preferably, the health assessment system comprises a filtering processing module, a first alarm information generation module and a second alarm information generation module;

the filtering processing module is used for performing mean value filtering processing on the first contact ratio by adopting sliding windows with a first width and a second width to obtain a fourth contact ratio and a fifth contact ratio corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period and the second width corresponds to a short period;

the first warning information generation module is used for generating first warning information for representing that the door opening or closing function of the elevator door system is reduced when the fourth overlap ratio is smaller than the first warning value and the fifth overlap ratio is greater than or equal to the second warning value;

and the second warning information generating module is used for generating second warning information for representing that the door opening or closing fault of the elevator door system occurs when the fifth overlap ratio is smaller than the second warning value.

Preferably, the first historical operating data, the second historical operating data, the third historical operating data, and the current operating data each include at least one of gating signal data, current data, electrical energy data, power data, and speed data.

The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and operable on the processor, wherein the processor implements the health assessment method of the elevator door system when executing the computer program.

The invention also provides a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method of health assessment of an elevator door system.

The positive progress effects of the invention are as follows:

according to the method, a first Gaussian mixture model used for representing a normal operation condition and a second Gaussian mixture model used for representing a current operation condition are established based on multi-dimensional data of an elevator door system, so that a first contact ratio between the current operation condition and the normal operation condition is obtained, the health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, real-time and comprehensive monitoring and analysis are achieved, and therefore the evaluation accuracy of the health state of the elevator door system is improved; label data is not needed (manual intervention is not needed), so that the labor cost is reduced; in addition, a health state warning mechanism of the elevator door system is provided, so that a user or a maintenance worker can be timely reminded of carrying out proper maintenance measures on the elevator door in advance, and further the elevator door is prevented from being broken down.

Drawings

Fig. 1 is a flowchart of a health assessment method of an elevator door system according to embodiment 1 of the present invention.

Fig. 2 is a first flowchart of a health assessment method of an elevator door system according to embodiment 2 of the present invention.

Fig. 3 is a second flowchart of a health assessment method of an elevator door system according to embodiment 2 of the present invention.

Fig. 4 is a flowchart of a health assessment method of an elevator door system according to embodiment 3 of the present invention.

Fig. 5 is a block schematic diagram of a health assessment system of an elevator door system according to embodiment 4 of the present invention.

Fig. 6 is a schematic view of a first module of a health assessment system of an elevator door system according to embodiment 5 of the present invention.

Fig. 7 is a second block schematic diagram of a health assessment system of an elevator door system according to embodiment 5 of the present invention.

Fig. 8 is a block schematic diagram of a health assessment system of an elevator door system according to embodiment 6 of the present invention.

Fig. 9 is a schematic configuration diagram of an electronic device for implementing a health assessment method of an elevator door system according to embodiment 7 of the present invention.

Detailed Description

The invention is further illustrated by the following examples, which are not intended to limit the scope of the invention.

Example 1

As shown in fig. 1, the health assessment method of the elevator door system of the present embodiment includes:

s101, obtaining corresponding first historical operation data of an elevator door system in a normal operation state within a historical sampling time period;

s102, obtaining current operation data of the elevator door system in a current sampling time period in a current operation state;

the first historical operation data and the current operation data comprise but are not limited to door control signal data, current data, electric energy data, power data and speed data, namely the elevator operation data of multiple dimensions of the elevator door system are comprehensively monitored, and the operation state of the elevator door system is more comprehensively evaluated; and the phenomenon of over-fitting is not easy to generate during model training.

S103, respectively taking the first historical operation data and the current operation data as training parameters, inputting the training parameters into a target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

s104, acquiring a first contact ratio between the current operation state and the normal operation state according to the first target model and the second target model;

s105, determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health.

The evaluation process in the embodiment does not need manual intervention, namely, the degree of automation is higher, and further, the labor cost is reduced.

In the embodiment, the first Gaussian mixture model used for representing the normal operation condition and the second Gaussian mixture model used for representing the current operation condition are established through the multidimensional data based on the elevator door system, so that the first contact ratio between the current operation state and the normal operation state is obtained, the health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, the health state of the elevator door system is evaluated in real time, meanwhile, more comprehensive monitoring and analysis are realized, and the evaluation accuracy of the health state of the elevator door system is improved.

Example 2

The health assessment method of the elevator door system of the present embodiment is a further improvement of embodiment 1, specifically:

as shown in fig. 2, when the first historical operating data includes a holding time of a door opening process or a holding time of a door closing process, and a mechanical energy average value, the step S101 specifically includes:

s10101, acquiring the total times of opening or closing the door of the elevator door system in a historical preset time period, and the first times of opening or closing the door with the holding time length meeting the preset time length range;

wherein the historical preset time period is in days.

S10102, calculating the ratio of the first times to the total times;

s10103, judging whether the ratio exceeds a first set threshold value, and if so, determining a preset time period as a historical intermediate time period;

s10104, acquiring historical target time periods according to the mechanical energy average value corresponding to each historical intermediate time period;

specifically, the mechanical energy average value corresponding to each historical intermediate time period is sorted according to size, and the historical intermediate time period corresponding to the smallest mechanical energy average value is selected as the historical target time period.

Wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system;

s10105, taking the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in the normal operation state, namely automatically searching and positioning the normal state data in the historical sampling time period based on the original label-free historical operation data.

As shown in fig. 3, after step S102 and before step S103, the method further includes:

s1021, acquiring a corresponding first data matrix according to the first historical operating data;

carrying out standardization processing on the first data matrix to obtain a first characteristic matrix;

s1022, acquiring a corresponding second data matrix according to the current operating data;

carrying out standardization processing on the second data matrix to obtain a second feature matrix;

the method comprises the following steps of combining expert experience and a feature extraction algorithm, performing feature extraction and feature conversion processing on all operation data corresponding to each sampling time point in each sampling time period, and reserving a set number of feature parameters, such as the following 15 feature parameters: the current of the Q axis is accumulated, the current of the D axis is accumulated, the sum of speed errors, (the ratio of the square of the speed error to the count value of the speed error, the positive maximum value of the speed error, the negative maximum value of the speed error, the position where the positive maximum value of the speed error appears, the position where the negative maximum value of the speed error appears, the starting position of the current state, the starting position of the current end, the sum of positive values of Iq, the sum of negative values of Iq, the sum of positive values of Id, the sum of negative values of Id.

Acquiring the day with the most normal running state of the elevator door system in a historical sampling time period, and acquiring running data (15 characteristic parameters) corresponding to each sampling time point in the day to form a first data matrix X1; acquiring operation data (15 characteristic parameters) corresponding to each sampling time point in the current sampling time period to form a second data matrix X2;

the first data matrix X1 and the second data matrix X2 are normalized to obtain a first feature matrix and a second feature matrix corresponding to each other, and the normalization processing is specifically performed by adopting the following formulas:

x＝(X-μ)/σ

wherein X represents the first feature matrix or the second feature matrix, X represents the first data matrix or the second data matrix, μ represents the mean vector, and σ represents the covariance matrix.

Step S103 includes:

and S1031, inputting the first characteristic matrix and the second characteristic matrix into a target model for training respectively as training parameters, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

When the target model includes a gaussian mixture model, the first target model is a first gaussian mixture model and the second target model is a second gaussian mixture model. The objective model may also include any other model that can be implemented for characterizing the operation of the elevator door system.

Step S104 includes:

s1041, calculating according to the first Gaussian mixture model and the second Gaussian mixture model to obtain a first contact ratio between the current operation state and the normal operation state.

Specifically, the method comprises the following steps: the first gaussian mixture model or the second gaussian mixture model includes:

wherein g (x) represents a first Gaussian mixture model or a second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing a d-dimensional first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector mu_iSum covariance matrix sigma_i。

Specifically, EM (expectation maximization) algorithm is adopted for the parameter theta_iEstimating, specifically solving as follows:

(1) randomly initializing a model parameter theta;

(2) using Bayes' theorem, data feature vector x is used_nCalculating the posterior probability of the model i according to the current model parameter thetaThe following were used:

(3) maximum likelihood reestimation of model coefficients

And (3) calculating and converging to a stable solution by repeating the step (2) and the step (3) in the iteration process, wherein the stable solution corresponds to the maximum likelihood solution, and then the converged mean value, the covariance matrix and the preposed vector are obtained.

In addition, the selection of the number I of the mixed models in the Gaussian mixed model is determined by using a BIC (Bayesian information) criterion algorithm, and the specific formula is as follows:

wherein H_jRepresenting the jth candidate model, D representing the training features,

representing the maximum log likelihood function of the jth candidate model, k representing the number of the estimated parameters, n representing the size of the features, and finally establishing the optimal Gaussian mixture model (namely, the first Gaussian mixture model) g with the minimum Bayesian information criterion score₁(x) The Gaussian mixture model is the model which most accurately represents the normal operation condition of the elevator door.

In step S1041, the first contact ratio is calculated by using the following formula:

wherein CV represents a first degree of coincidence, g₁(x1) denotes a first Gaussian mixture model, g₂(x2) represents a second Gaussian mixture model, x1 represents a first feature matrix, and x2 represents a second feature matrix.

The CV value range is 0-1, and the higher the CV value is, the closer the current running condition of the elevator door system is to the normal state is; conversely, a lower CV value indicates that the more the current operating condition of the elevator door system is away from normal, some degradation may occur, requiring appropriate maintenance measures in real time.

In the embodiment, a first Gauss mixed model used for representing a normal operation condition and a second Gauss mixed model used for representing a current operation condition are established, so that a first contact ratio between a current operation state and a normal operation state is obtained, a health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, and comprehensive monitoring and analysis are realized by combining multidimensional data, so that the evaluation accuracy of the health state of the elevator door system is improved; and no label data (no manual intervention) is needed, thereby reducing the labor cost.

Example 3

As shown in fig. 4, the health assessment method of the elevator door system of the present embodiment is a further improvement of embodiment 2, specifically:

step S105 is followed by:

s106, respectively presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period;

the process of presetting the label is a process of setting corresponding relations among the sub-health state, the fault state and the corresponding operation data.

Sub-health conditions include, but are not limited to, wear of the slide, and increased friction due to the presence of foreign objects in the guide rails.

The fault state includes, but is not limited to, the condition that the opening and closing door is not in place, the car door does not drive the hall door, and the condition that the friction of the guide rail is severe to cause the abnormal speed of the opening and closing door.

S107, second historical operating data corresponding to the sub-health state in the historical sampling time period are obtained according to the first label;

s108, acquiring third history operation data corresponding to the fault state in the history sampling time period according to the second label;

the second historical operating data and the third historical operating data include, but are not limited to, gating signal data, current data, power data and speed data.

S109, a third data matrix corresponding to the second historical operating data;

carrying out standardization processing on the third data matrix to obtain a third feature matrix;

s1010, acquiring a corresponding fourth data matrix according to the third history operation data;

carrying out standardization processing on the fourth data matrix to obtain a fourth feature matrix;

s1011, inputting the third feature matrix as a training parameter into a Gaussian mixture model for training, and acquiring a third Gaussian mixture model for representing the running condition of the elevator door system in a sub-health state;

s1012, inputting the fourth feature matrix as a training parameter into a Gaussian mixture model for training, and acquiring a fourth Gaussian mixture model for representing the running condition of the elevator door system in a fault state;

s1013, calculating according to the first Gaussian mixture model and the third Gaussian mixture model to obtain a second overlap ratio between the sub-health state and the normal operation state;

s1014, calculating according to the first Gaussian mixture model and the fourth Gaussian mixture model to obtain a third coincidence degree between the fault state and the normal operation state;

s1015, setting a first early warning value according to the maximum value of the plurality of second overlapping degrees;

the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

s1016, setting a second early warning value according to the maximum value of the multiple third overlap ratios;

and the second early warning value is used for early warning the condition of door opening or door closing faults of the elevator door system.

Specifically, a third gaussian mixture model corresponding to each sampling time point in the sub-health state:

g_u1(x_u1)、g_u2(x_u2)···

and respectively calculating a plurality of second overlap ratios between the sub-health state and the normal operation state according to each third Gaussian mixture model and the first Gaussian mixture model:

CV_u1,CV_u2····

and a fourth Gaussian mixture model corresponding to each sampling time point in the fault state:

g_d1(x_d1)、g_d2(x_d2)···

respectively calculating a plurality of third overlap ratios between the fault state and the normal operation state according to each fourth Gaussian mixture model and the first Gaussian mixture model:

CV_d1,CV_d2···

CV_T1＝max(CV_d1,CV_d2······)

CV_T2＝max(CV_u1,CV_u2······)

according to CV of_T1Determining a first warning value according to CV_T2And determining a second early warning value.

S1017, performing mean value filtering processing on the first overlap ratio by adopting sliding windows with the first width and the second width to obtain a fourth overlap ratio and a fifth overlap ratio corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period (e.g., n takes the value of 500) and the second width corresponds to a short period (e.g., n takes the value of 5).

S1018, when the fourth overlap ratio is smaller than the first early warning value and the fifth overlap ratio is larger than or equal to the second early warning value, generating first warning information for representing that the door opening or closing function of the elevator door system is reduced;

and when the fifth overlap ratio is smaller than the second early warning value, generating second warning information for representing the door opening or closing fault of the elevator door system.

Obtaining the condition of door opening or door closing faults of the elevator door system in a sub-health state through fourth overlap ratio analysis; the condition that the door opening or closing function of the elevator door system is reduced in a fault state is obtained through the analysis of the fifth overlap ratio, so that timely warning is carried out, and personnel can conveniently carry out troubleshooting and processing in time.

The following is illustrated with reference to specific examples:

1) acquiring corresponding operation data of an elevator door system every day in the last half year, performing feature extraction and feature conversion on the operation data, and reserving 15 feature parameters;

2) acquiring the total door opening times of an elevator door system every day and the door opening times of which the keeping time length in the door opening process is equal to the preset time length (such as 377);

3) calculating the ratio of the door opening times to the total times, selecting the date corresponding to each day with the ratio being greater than 95%, then sequencing the average values of the mechanical energy corresponding to the days, selecting the day with the smallest average value of the mechanical energy as the day with the most normal operation of the average value of the mechanical energy in the last half year, assuming that the day is 2019-03-12, and obtaining a time sequence (namely a first data matrix) X1 corresponding to each sampling time point on the day of 2019-03-12;

4) acquiring a second data matrix X2 corresponding to each sampling time point in the current sampling time period in the current running state;

5) the first data matrix X1 and the second data matrix X2 are subjected to standardization processing, and a first feature matrix X1 and a second feature matrix X2 which correspond to each other are obtained respectively;

6) respectively taking the first characteristic matrix x1 and the second characteristic matrix x2 as training parameters, inputting the training parameters into a Gaussian mixture model for training, and acquiring a first Gaussian mixture model g for representing the normal operation condition of the elevator door system₁(x1) and method for characterizing elevator door systemsSecond Gaussian mixture model g of current operating conditions₂(x2)。

7) According to the first Gaussian mixture model g₁(x1) and a second Gaussian mixture model g₂(x2) calculating a first contact ratio CV between the current operating state and the normal operating state, for example, obtaining the first contact ratio CV corresponding to a certain current sampling time point as 0.5932.

Obtaining a third Gaussian mixture model of the running condition of the elevator door system in the sub-health state and a plurality of corresponding second coincidence degrees:

g_u1(x_u1)、g_u2(x_u2)···；CV_u1,CV_u2····

obtaining a fourth Gaussian mixture model of the running condition of the elevator door system in the fault state:

g_d1(x_d1)、g_d2(x_d2)···；CV_d1,CV_d2···

CV_T1＝max(CV_d1,CV_d2······)＝0.2541；

CV_T2＝max(CV_u1,CV_u2······)＝0.7806；

according to CV of_T1Determining the first warning value to be 0.3 according to CV_T2And determining the second early warning value to be 0.8.

8) Filtering the first coincidence degree CV corresponding to each sampling time point in the current sampling time period by adopting different-width (n is 5 and n is 500) mean value filtering to obtain a corresponding fourth coincidence degree CV_SAnd CV_L；

9) When CV is_LLess than 0.8 and CV_SWhen the door opening or closing function of the elevator door system is greater than or equal to 0.3, first warning information used for representing that the door opening or closing function of the elevator door system is reduced is generated; when CV is_SAnd when the value is less than 0.3, second alarm information for representing the door opening or closing fault of the elevator door system is generated.

In addition, the evaluation process of the health status of the door closing process of the elevator door system is similar to the above-described evaluation process of the health status of the door opening process, and thus, a detailed description thereof will be omitted.

In the embodiment, a first Gauss mixed model used for representing a normal operation condition and a second Gauss mixed model used for representing a current operation condition are established, so that a first contact ratio between a current operation state and a normal operation state is obtained, a health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, and comprehensive monitoring and analysis are realized by combining multidimensional data, so that the evaluation accuracy of the health state of the elevator door system is improved; label data is not needed (manual intervention is not needed), so that the labor cost is reduced; in addition, a health state warning mechanism of the elevator door system is provided, so that a user or a maintenance worker can be timely reminded of carrying out proper maintenance measures on the elevator door in advance, and further the elevator door is prevented from being broken down.

Example 4

As shown in fig. 5, the health evaluation system of the elevator door system of the present embodiment includes a first history data acquisition module 1, a current data acquisition module 2, a model acquisition module 3, a first contact ratio acquisition module 4, and a health degree determination module 5.

The first historical data acquisition module 1 is used for acquiring corresponding first historical operation data when the elevator door system is in a normal operation state within a historical sampling time period;

the current data acquisition module 2 is used for acquiring current operation data of the elevator door system in a current sampling time period in a current operation state;

the first historical operation data and the current operation data comprise but are not limited to door control signal data, current data, electric energy data, power data and speed data, namely the elevator operation data of multiple dimensions of the elevator door system are comprehensively monitored, and the operation state of the elevator door system is more comprehensively evaluated; and the phenomenon of over-fitting is not easy to generate during model training.

The model acquisition module 3 is used for inputting the first historical operation data and the current operation data as training parameters into a target model for training to acquire a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

the first contact ratio obtaining module 4 is used for obtaining a first contact ratio between the current operation state and the normal operation state according to the first target model and the second target model;

the health degree determining module 5 is used for determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health.

The evaluation process in the embodiment does not need manual intervention, namely, the degree of automation is higher, and further, the labor cost is reduced.

In the embodiment, the first Gaussian mixture model used for representing the normal operation condition and the second Gaussian mixture model used for representing the current operation condition are established, so that the first contact ratio between the current operation state and the normal operation state is obtained, the health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, the health state of the elevator door system is evaluated in real time, meanwhile, the multi-dimensional data is combined, more comprehensive monitoring and analysis are achieved, and the evaluation accuracy of the health state of the elevator door system is improved.

Example 5

The health assessment system of the elevator door system of the present embodiment is a further improvement of embodiment 4, specifically:

as shown in fig. 6, when the first historical operating data includes the holding time period of the door opening process or the holding time period of the door closing process, and the mechanical energy average value, the first historical data acquiring module 1 includes a number of times acquiring unit 6, a ratio calculating unit 7, a judging unit 8, a target time period acquiring unit 9, and a historical data acquiring unit 10.

The frequency acquiring unit 6 is used for acquiring the total frequency of door opening or closing of the elevator door system in a historical preset time period, and the first frequency of door opening or door closing with the holding time length meeting the preset time length range;

wherein the historical preset time period is in days.

The ratio calculation unit 7 is used for calculating the ratio of the first times to the total times;

the judging unit 8 is configured to judge whether the ratio exceeds a first set threshold, and if so, determine that the preset time period is a historical intermediate time period;

the target time period obtaining unit 9 is configured to obtain a historical target time period according to the mechanical energy average value corresponding to each historical intermediate time period;

specifically, the target time period obtaining unit is configured to sort the mechanical energy average value corresponding to each historical intermediate time period according to the size, and select the historical intermediate time period corresponding to the smallest mechanical energy average value as the historical target time period;

wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system.

The historical data acquiring unit 10 is configured to use the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in the normal operation state, that is, to automatically find and locate the normal state data in the historical sampling time period based on the original unlabeled historical operation data.

As shown in fig. 7, the health assessment system of the present embodiment further includes a first feature matrix acquisition module 11 and a second feature matrix acquisition module 12.

The first feature matrix acquisition module 11 is configured to acquire a corresponding first data matrix according to the first historical operating data, and perform standardization processing on the first data matrix to acquire a first feature matrix;

the second feature matrix obtaining module 12 is configured to obtain a corresponding second data matrix according to the current operating data, and perform standardization processing on the second data matrix to obtain a second feature matrix;

the method comprises the following steps of combining expert experience and a feature extraction algorithm, performing feature extraction and feature conversion processing on all operation data corresponding to each sampling time point in each sampling time period, and reserving a set number of feature parameters, such as the following 15 feature parameters: the current of the Q axis is accumulated, the current of the D axis is accumulated, the sum of speed errors, (the ratio of the square of the speed error to the count value of the speed error, the positive maximum value of the speed error, the negative maximum value of the speed error, the position where the positive maximum value of the speed error appears, the position where the negative maximum value of the speed error appears, the starting position of the current state, the starting position of the current end, the sum of positive values of Iq, the sum of negative values of Iq, the sum of positive values of Id, the sum of negative values of Id.

Acquiring the day with the most normal running state of the elevator door system in a historical sampling time period, and acquiring running data (15 characteristic parameters) corresponding to each sampling time point in the day to form a first data matrix X1; acquiring operation data (15 characteristic parameters) corresponding to each sampling time point in the current sampling time period to form a second data matrix X2;

the first data matrix X1 and the second data matrix X2 are normalized to obtain a first feature matrix and a second feature matrix corresponding to each other, and the normalization processing is specifically performed by adopting the following formulas:

x＝(X-μ)/σ

wherein X represents the first feature matrix or the second feature matrix, X represents the first data matrix or the second data matrix, μ represents the mean vector, and σ represents the covariance matrix.

The model obtaining module 3 is used for inputting the first feature matrix and the second feature matrix as training parameters into a target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

When the target model comprises a Gaussian mixture model, the first target model is a first Gaussian mixture model, and the second target model is a second Gaussian mixture model; the objective model may also include any other model that can be implemented for characterizing the operation of the elevator door system.

The first contact ratio obtaining module 4 is configured to calculate a first contact ratio between the current operation state and the normal operation state according to the first gaussian mixture model and the second gaussian mixture model.

Specifically, the first gaussian mixture model or the second gaussian mixture model includes:

wherein g (x) represents a first Gaussian mixture model or a second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing a d-dimensional first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector mu_iSum covariance matrix sigma_i。

Specifically, EM (expectation maximization) algorithm is adopted for the parameter theta_iEstimating, specifically solving as follows:

(1) randomly initializing a model parameter theta;

(2) using Bayes' theorem, data feature vector x is used_nAnd calculating the posterior probability of the model i according to the current model parameter theta, wherein the specific formula is as follows:

(3) maximum likelihood reestimation of model coefficients

And (3) calculating and converging to a stable solution by repeating the step (2) and the step (3) in the iteration process, wherein the stable solution corresponds to the maximum likelihood solution, and then the converged mean value, the covariance matrix and the preposed vector are obtained.

In addition, the selection of the number I of the mixed models in the Gaussian mixed model is determined by using a BIC (Bayesian information) criterion algorithm, and the specific formula is as follows:

wherein H_jRepresenting the jth candidate model, and D representing a training feature;

representing the maximum log likelihood function of the jth candidate model, k representing the number of the estimated parameters, n representing the size of the features, and finally establishing the optimal Gaussian mixture model (namely, the first Gaussian mixture model) g with the minimum Bayesian information criterion score₁(x) The Gaussian mixture model is the model which most accurately represents the normal operation condition of the elevator door.

The first contact ratio obtaining module 4 calculates the first contact ratio by using the following formula:

wherein CV represents a first degree of coincidence, g₁(x1) denotes a first Gaussian mixture model, g₂(x2) represents a second Gaussian mixture model, x1 represents a first feature matrix, and x2 represents a second feature matrix.

The CV value range is 0-1, and the higher the CV value is, the closer the current running condition of the elevator door system is to the normal state is; conversely, a lower CV value indicates that the more the current operating condition of the elevator door system is away from normal, some degradation may occur, requiring appropriate maintenance measures in real time.

In the embodiment, a first Gauss mixed model used for representing a normal operation condition and a second Gauss mixed model used for representing a current operation condition are established, so that a first contact ratio between a current operation state and a normal operation state is obtained, a health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, and comprehensive monitoring and analysis are realized by combining multidimensional data, so that the evaluation accuracy of the health state of the elevator door system is improved; and no label data (no manual intervention) is needed, thereby reducing the labor cost.

Example 6

As shown in fig. 8, the health assessment system of the elevator door system of the present embodiment is a further improvement of embodiment 5, specifically:

the health assessment system of the embodiment is used for a tag presetting module 13, a second historical data obtaining module 14, a third historical data obtaining module 15, a third feature matrix obtaining module 16, a fourth feature matrix obtaining module 17, a second coincidence degree obtaining module 18, a third coincidence degree obtaining module 19, an early warning value setting module 20, a filtering processing module 21, a first warning information generating module 22 and a second warning information generating module 23.

The label presetting module 13 is used for respectively presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period; the process of presetting the label is a process of setting corresponding relations among the sub-health state, the fault state and the corresponding operation data.

Sub-health conditions include, but are not limited to, wear of the slide, and increased friction due to the presence of foreign objects in the guide rails.

The fault state includes, but is not limited to, the condition that the opening and closing door is not in place, the car door does not drive the hall door, and the condition that the friction of the guide rail is severe to cause the abnormal speed of the opening and closing door.

The second historical data acquisition module 14 is configured to acquire second historical operating data corresponding to the sub-health state within a historical sampling time period according to the first tag;

the third history data acquisition module 15 is configured to acquire third history operation data of a corresponding fault state in a history sampling time period according to the second tag;

the second historical operating data and the third historical operating data include, but are not limited to, gating signal data, current data, power data and speed data.

The third feature matrix obtaining module 16 is configured to perform normalization processing on a third data matrix according to the third data matrix corresponding to the second historical operating data, so as to obtain a third feature matrix;

the fourth feature matrix obtaining module 17 is configured to obtain a corresponding fourth data matrix according to the third history operation data, and perform normalization processing on the fourth data matrix to obtain a fourth feature matrix;

the model obtaining module 3 is further configured to input the third feature matrix as a training parameter to the gaussian mixture model for training, and obtain a third gaussian mixture model for representing the operation condition of the elevator door system in the sub-health state;

the model obtaining module 3 is further configured to input the fourth feature matrix as a training parameter to the gaussian mixture model for training, and obtain a fourth gaussian mixture model for representing the operation condition of the elevator door system in a fault state;

the second overlap ratio obtaining module 18 is configured to calculate a second overlap ratio between the sub-health state and the normal operation state according to the first gaussian mixture model and the third gaussian mixture model;

the third overlap ratio obtaining module 19 is configured to calculate a third overlap ratio between the fault state and the normal operation state according to the first gaussian mixture model and the fourth gaussian mixture model;

the early warning value setting module 20 is configured to set a first early warning value according to a maximum value of the plurality of second overlap ratios;

the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

the early warning value setting module is also used for setting a second early warning value according to the maximum value of the plurality of third overlap ratios;

and the second early warning value is used for early warning the condition of door opening or door closing faults of the elevator door system.

Specifically, a third gaussian mixture model corresponding to each sampling time point in the sub-health state:

g_u1(x_u1)、g_u2(x_u2)···

and respectively calculating a plurality of second overlap ratios between the sub-health state and the normal operation state according to each third Gaussian mixture model and the first Gaussian mixture model:

CV_u1,CV_u2····

and a fourth Gaussian mixture model corresponding to each sampling time point in the fault state:

g_d1(x_d1)、g_d2(x_d2)···

respectively calculating a plurality of third overlap ratios between the fault state and the normal operation state according to each fourth Gaussian mixture model and the first Gaussian mixture model:

CV_d1,CV_d2···

CV_T1＝max(CV_d1,CV_d2······)

CV_T2＝max(CV_u1,CV_u2······)

according to CV of_T1Determining a first warning value according to CV_T2And determining a second early warning value.

The filtering processing module 21 is configured to perform mean filtering processing on the first overlap ratio by using sliding windows with the first width and the second width to obtain a fourth overlap ratio and a fifth overlap ratio corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period and the second width corresponds to a short period;

the first warning information generating module 22 is configured to generate first warning information for representing that the door opening or closing function of the elevator door system is reduced when the fourth overlap ratio is smaller than the first warning value and the fifth overlap ratio is greater than or equal to the second warning value;

the second warning information generating module 23 is configured to generate second warning information for indicating that the door opening or closing fault occurs in the elevator door system when the fifth overlap ratio is smaller than the second warning value.

Obtaining the condition of door opening or door closing faults of the elevator door system in a sub-health state through fourth overlap ratio analysis; the condition that the door opening or closing function of the elevator door system is reduced in a fault state is obtained through the analysis of the fifth overlap ratio, so that timely warning is carried out, and personnel can conveniently carry out troubleshooting and processing in time.

The following is illustrated with reference to specific examples:

1) acquiring corresponding operation data of an elevator door system every day in the last half year, performing feature extraction and feature conversion on the operation data, and reserving 15 feature parameters;

2) acquiring the total door opening times of an elevator door system every day and the door opening times of which the keeping time length in the door opening process is equal to the preset time length (such as 377);

3) calculating the ratio of the door opening times to the total times, selecting the date corresponding to each day with the ratio being greater than 95%, then sequencing the average values of the mechanical energy corresponding to the days, selecting the day with the smallest average value of the mechanical energy as the day with the most normal operation of the average value of the mechanical energy in the last half year, assuming that the day is 2019-03-12, and obtaining a time sequence (namely a first data matrix) X1 corresponding to each sampling time point on the day of 2019-03-12;

4) acquiring a second data matrix X2 corresponding to each sampling time point in the current sampling time period in the current running state;

5) the first data matrix X1 and the second data matrix X2 are subjected to standardization processing, and a first feature matrix X1 and a second feature matrix X2 which correspond to each other are obtained respectively;

6) respectively taking the first characteristic matrix x1 and the second characteristic matrix x2 as training parameters, inputting the training parameters into a Gaussian mixture model for training, and acquiring a first Gaussian mixture model g for representing the normal operation condition of the elevator door system₁(x1) and a second Gaussian mixture model g for characterizing the current behavior of the elevator door system₂(x2)。

7) According to the first Gaussian mixture model g₁(x1) and a second Gaussian mixture model g₂(x2) calculating a first contact ratio CV between the current operating state and the normal operating state, for example, obtaining the first contact ratio CV corresponding to a certain current sampling time point as 0.5932.

Obtaining a third Gaussian mixture model of the running condition of the elevator door system in the sub-health state and a plurality of corresponding second coincidence degrees:

g_u1(x_u1)、g_u2(x_u2)···；CV_u1,CV_u2····

obtaining a fourth Gaussian mixture model of the running condition of the elevator door system in the fault state:

g_d1(x_d1)、g_d2(x_d2)···；CV_d1,CV_d2···

CV_T1＝max(CV_d1,CV_d2······)＝0.2541；

CV_T2＝max(CV_u1,CV_u2······)＝0.7806；

according to CV of_T1Determining the first warning value to be 0.3 according to CV_T2And determining the second early warning value to be 0.8.

8) Filtering the first coincidence degree CV corresponding to each sampling time point in the current sampling time period by adopting different-width (n is 5 and n is 500) mean value filtering to obtain a corresponding fourth coincidence degree CV_SAnd CV_L；

9) When CV is_LLess than 0.8 and CV_SWhen the door opening or closing function of the elevator door system is greater than or equal to 0.3, first warning information used for representing that the door opening or closing function of the elevator door system is reduced is generated; when CV is_SAnd when the value is less than 0.3, second alarm information for representing the door opening or closing fault of the elevator door system is generated.

In addition, the evaluation process of the health status of the door closing process of the elevator door system is similar to the above-described evaluation process of the health status of the door opening process, and thus, a detailed description thereof will be omitted.

In the embodiment, a first Gauss mixed model used for representing a normal operation condition and a second Gauss mixed model used for representing a current operation condition are established, so that a first contact ratio between a current operation state and a normal operation state is obtained, a health degree corresponding to the current operation condition of the elevator door system is evaluated according to the first contact ratio, and comprehensive monitoring and analysis are realized by combining multidimensional data, so that the evaluation accuracy of the health state of the elevator door system is improved; label data is not needed (manual intervention is not needed), so that the labor cost is reduced; in addition, a health state warning mechanism of the elevator door system is provided, so that a user or a maintenance worker can be timely reminded of carrying out proper maintenance measures on the elevator door in advance, and further the elevator door is prevented from being broken down.

Example 7

Fig. 9 is a schematic structural diagram of an electronic device according to embodiment 7 of the present invention. The electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor implements the method of health assessment of an elevator door system of any of embodiments 1 to 3. The electronic device 30 shown in fig. 9 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.

As shown in fig. 9, the electronic device 30 may be embodied in the form of a general purpose computing device, which may be, for example, a server device. The components of the electronic device 30 may include, but are not limited to: the at least one processor 31, the at least one memory 32, and a bus 33 connecting the various system components (including the memory 32 and the processor 31).

The bus 33 includes a data bus, an address bus, and a control bus.

The memory 32 may include volatile memory, such as Random Access Memory (RAM)321 and/or cache memory 322, and may further include Read Only Memory (ROM) 323.

Memory 32 may also include a program/utility 325 having a set (at least one) of program modules 324, such program modules 324 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

The processor 31 executes various functional applications and data processing, such as a health assessment method of an elevator door system in any one of embodiments 1 to 3 of the present invention, by running a computer program stored in the memory 32.

The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). Such communication may be through input/output (I/O) interfaces 35. Also, model-generating device 30 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet) via network adapter 36. As shown in FIG. 9, network adapter 36 communicates with the other modules of model-generating device 30 via bus 33. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the model-generating device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

It should be noted that although in the above detailed description several units/modules or sub-units/modules of the electronic device are mentioned, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the units/modules described above may be embodied in one unit/module according to embodiments of the invention. Conversely, the features and functions of one unit/module described above may be further divided into embodiments by a plurality of units/modules.

Example 8

The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in the health assessment method of an elevator door system according to any of embodiments 1 to 3.

More specific examples, among others, that the readable storage medium may employ may include, but are not limited to: a portable disk, a hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.

In a possible embodiment, the invention can also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps of a method of health assessment of an elevator door system implementing any of embodiments 1 to 3, when the program product is run on the terminal device.

Where program code for carrying out the invention is written in any combination of one or more programming languages, the program code may execute entirely on the viewer device, partly on the viewer device, as a stand-alone software package, partly on the viewer device, partly on a remote device or entirely on the remote device.

While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.

Claims (18)

1. A health assessment method of an elevator door system, characterized in that the health assessment method comprises:

s1, acquiring corresponding first historical operation data of an elevator door system in a normal operation state within a historical sampling time period;

s2, acquiring current operation data of the elevator door system in a current sampling time period in a current operation state;

s3, respectively taking the first historical operation data and the current operation data as training parameters, inputting the training parameters into a target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

s4, acquiring a first contact ratio between the current running state and the normal running state according to the first target model and the second target model;

s5, determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health;

when the target model comprises a Gaussian mixture model, the first target model is a first Gaussian mixture model, and the second target model is a second Gaussian mixture model;

step S4 includes:

s41, calculating according to the first Gaussian mixture model and the second Gaussian mixture model to obtain a first contact ratio between the current operation state and the normal operation state;

when the first historical operating data includes the duration of the door opening process or the duration of the door closing process, and the mechanical energy average value, step S1 specifically includes:

s11, acquiring the total times of door opening or closing of the elevator door system in a historical preset time period, and the first times of door opening or door closing when the holding time length meets a preset time length range;

s12, calculating the ratio of the first times to the total times;

s13, judging whether the ratio exceeds a set threshold value or not, and if so, determining the preset time period as a historical intermediate time period;

s14, acquiring a historical target time period according to the mechanical energy average value corresponding to each historical intermediate time period;

and S15, taking the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in a normal operation state.

2. The method for health assessment of an elevator door system according to claim 1, wherein step S14 comprises:

sorting the mechanical energy average values corresponding to the historical intermediate time periods according to the sizes, and selecting the historical intermediate time period corresponding to the smallest mechanical energy average value as the historical target time period;

wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system; and/or the presence of a gas in the gas,

the historical preset time period is in days.

3. The method of health assessment of an elevator door system of claim 1, further comprising after step S15 and before step S3:

acquiring a corresponding first data matrix according to the first historical operating data;

carrying out standardization processing on the first data matrix to obtain a first characteristic matrix;

acquiring a corresponding second data matrix according to the current operation data;

carrying out standardization processing on the second data matrix to obtain a second feature matrix;

step S3 further includes:

and respectively taking the first characteristic matrix and the second characteristic matrix as training parameters, inputting the training parameters into the target model for training, and obtaining a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

4. The method of health assessment of an elevator door system of claim 3, wherein the first Gaussian mixture model or the second Gaussian mixture model comprises:

wherein g (x) represents the first Gaussian mixture model or the second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing d dimensions of the first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector μ_iSum covariance matrix sigma_i。

5. The method of evaluating the health of an elevator door system of claim 4, wherein the first contact ratio is calculated in step S41 using the following formula:

wherein CV represents the first degree of coincidence, g₁(x1) represents the first Gaussian mixture model, g₂(x2) represents the second Gaussian mixture model, x1 represents the first feature matrix, and x2 represents the second feature matrix.

6. The health assessment method of an elevator door system of claim 1, further comprising:

presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period respectively;

step S5 is followed by:

acquiring second historical operating data corresponding to the sub-health state in the historical sampling time period according to the first label;

acquiring third history operation data corresponding to the fault state in the history sampling time period according to the second label;

a third data matrix corresponding to the second historical operating data;

carrying out standardization processing on the third data matrix to obtain a third feature matrix;

acquiring a corresponding fourth data matrix according to the third history operation data;

carrying out standardization processing on the fourth data matrix to obtain a fourth feature matrix;

inputting the third feature matrix as a training parameter into a Gaussian mixture model for training to obtain a third Gaussian mixture model for representing the running condition of the elevator door system in a sub-health state;

inputting the fourth feature matrix as a training parameter into a Gaussian mixture model for training to obtain a fourth Gaussian mixture model for representing the running condition of the elevator door system in a fault state;

calculating a second overlap ratio between the sub-health state and the normal operation state according to the first Gaussian mixture model and the third Gaussian mixture model;

calculating according to the first Gaussian mixture model and the fourth Gaussian mixture model to obtain a third coincidence ratio between the fault state and the normal operation state;

setting a first early warning value according to the maximum value of the second superposition degrees;

wherein the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

setting a second early warning value according to the maximum value of the third overlap ratios;

and the second early warning value is used for early warning the condition that the door opening or closing fault occurs in the elevator door system.

7. The health assessment method of an elevator door system of claim 6, further comprising:

performing mean value filtering processing on the first coincidence degree by adopting sliding windows with a first width and a second width to obtain a fourth coincidence degree and a fifth coincidence degree corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period and the second width corresponds to a short period;

when the fourth overlap ratio is smaller than the first early warning value and the fifth overlap ratio is greater than or equal to the second early warning value, generating first warning information for representing that the door opening or closing function of the elevator door system is reduced;

and when the fifth overlap ratio is smaller than the second early warning value, generating second warning information for representing that the door opening or closing fault of the elevator door system occurs.

8. The elevator door system health assessment method of claim 6, wherein the first historical operating data, the second historical operating data, the third historical operating data, the current operating data each comprise at least one of gating signal data, current data, power data, and speed data.

9. A health assessment system of an elevator door system is characterized by comprising a first historical data acquisition module, a current data acquisition module, a model acquisition module, a first contact ratio acquisition module and a health degree determination module;

the first historical data acquisition module is used for acquiring corresponding first historical operation data when the elevator door system is in a normal operation state within a historical sampling time period;

the current data acquisition module is used for acquiring current operation data of the elevator door system in a current sampling time period in a current operation state;

the model acquisition module is used for inputting the first historical operation data and the current operation data as training parameters into a target model for training to acquire a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system;

the first contact ratio obtaining module is used for obtaining a first contact ratio between the current operation state and the normal operation state according to the first target model and the second target model;

the health degree determining module is used for determining the health degree corresponding to the current running condition of the elevator door system according to the first contact ratio;

wherein the first degree of coincidence is positively correlated with the degree of health;

when the target model comprises a Gaussian mixture model, the first target model is a first Gaussian mixture model, and the second target model is a second Gaussian mixture model;

the first contact ratio obtaining module is used for calculating and obtaining a first contact ratio between the current operation state and the normal operation state according to the first Gaussian mixture model and the second Gaussian mixture model;

when the first historical operating data comprises the holding time of the door opening process or the holding time of the door closing process and the mechanical energy average value, the first historical data acquisition module comprises a frequency acquisition unit, a ratio calculation unit, a judgment unit, a target time period acquisition unit and a historical data acquisition unit;

the frequency acquisition unit is used for acquiring the total frequency of opening or closing the door of the elevator door system in a historical preset time period and the first frequency of opening or closing the door with the holding time length meeting a preset time length range;

the ratio calculation unit is used for calculating the ratio of the first times to the total times;

the judging unit is used for judging whether the ratio exceeds a set threshold value, and if so, determining the preset time period as a historical intermediate time period;

the target time period obtaining unit is used for obtaining historical target time periods according to the mechanical energy average value corresponding to each historical intermediate time period;

the historical data acquisition unit is used for taking the operation data corresponding to each sampling time point in the historical target time period as the corresponding first historical operation data in the normal operation state.

10. The health assessment system of an elevator door system according to claim 9, wherein the target time period acquisition unit is configured to sort the average value of the mechanical energy corresponding to each of the historical intermediate time periods according to magnitude, and select the historical intermediate time period corresponding to the smallest average value of the mechanical energy as the historical target time period;

wherein the magnitude of the mechanical energy average is inversely related to the health of the operating condition of the elevator door system; and/or the presence of a gas in the gas,

the historical preset time period is in days.

11. The health assessment system of an elevator door system of claim 9, further comprising a first feature matrix acquisition module and a second feature matrix acquisition module;

the first characteristic matrix acquisition module is used for acquiring a corresponding first data matrix according to the first historical operating data and carrying out standardization processing on the first data matrix to acquire a first characteristic matrix;

the second characteristic matrix acquisition module is used for acquiring a corresponding second data matrix according to the current operating data and carrying out standardization processing on the second data matrix to acquire a second characteristic matrix;

the model acquisition module is used for inputting the first characteristic matrix and the second characteristic matrix as training parameters into the target model for training to acquire a first target model for representing the normal operation condition of the elevator door system and a second target model for representing the current operation condition of the elevator door system.

12. The health assessment system of an elevator door system of claim 11, wherein the first gaussian mixture model or the second gaussian mixture model comprises:

wherein g (x) represents the first Gaussian mixture model or the second Gaussian mixture model, h (x; θ)_i) Representing a single Gaussian function, x representing d dimensions of the first or second feature matrix, I representing the number of mixture models, p_iA priority vector representing a preset ith single Gaussian function

θ_iModel parameters representing the ith single Gaussian function, the model parameters including an average vector μ_iSum covariance matrix sigma_i。

13. The health assessment system of an elevator door system of claim 12, wherein the first contact ratio acquisition module calculates the first contact ratio using the formula:

wherein CV represents the first degree of coincidence, g₁(x1) represents the first Gaussian mixture model, g₂(x2) represents the second Gaussian mixture model, x1 represents the first feature matrix, and x2 represents the second feature matrix.

14. The health assessment system of an elevator door system of claim 9, wherein the health assessment system is for a tag presetting module, a second historical data acquisition module, a third feature matrix acquisition module, a fourth feature matrix acquisition module, a second overlap ratio acquisition module, a third overlap ratio acquisition module, and an early warning value setting module;

the label presetting module is used for respectively presetting a first label and a second label corresponding to the running data of the sub-health state and the fault state in the historical sampling time period;

the second historical data acquisition module is used for acquiring second historical operating data of the corresponding sub-health state in the historical sampling time period according to the first label;

the third history data acquisition module is used for acquiring third history operation data of a corresponding fault state in the history sampling time period according to the second label;

the third feature matrix acquisition module is used for carrying out standardization processing on a third data matrix according to the third data matrix corresponding to the second historical operating data to acquire a third feature matrix;

the fourth feature matrix acquisition module is used for acquiring a corresponding fourth data matrix according to the third history operation data, and carrying out standardization processing on the fourth data matrix to acquire a fourth feature matrix;

the model acquisition module is further used for inputting the third feature matrix as a training parameter into a Gaussian mixture model for training to acquire a third Gaussian mixture model for representing the running condition of the elevator door system in a sub-health state;

the model acquisition module is further used for inputting the fourth feature matrix as a training parameter into a Gaussian mixture model for training to acquire a fourth Gaussian mixture model for representing the running condition of the elevator door system in a fault state;

the second overlap ratio acquisition module is used for calculating a second overlap ratio between the sub-health state and the normal operation state according to the first Gaussian mixture model and the third Gaussian mixture model;

the third coincidence degree obtaining module is used for calculating a third coincidence degree between the fault state and the normal operation state according to the first Gaussian mixture model and the fourth Gaussian mixture model;

the early warning value setting module is used for setting a first early warning value according to the maximum value of the second superposition degrees;

wherein the first early warning value is used for early warning the condition that the door opening or closing function of the elevator door system is reduced;

the early warning value setting module is further used for setting a second early warning value according to the maximum value of the third overlap ratios;

and the second early warning value is used for early warning the condition that the door opening or closing fault occurs in the elevator door system.

15. The health assessment system of an elevator door system of claim 14, wherein the health assessment system comprises a filter processing module, a first alarm information generation module, and a second alarm information generation module;

the filtering processing module is used for performing mean value filtering processing on the first contact ratio by adopting sliding windows with a first width and a second width to obtain a fourth contact ratio and a fifth contact ratio corresponding to each sampling time point in the current sampling time period;

wherein the first width corresponds to a long period and the second width corresponds to a short period;

the first warning information generation module is used for generating first warning information for representing that the door opening or closing function of the elevator door system is reduced when the fourth overlap ratio is smaller than the first warning value and the fifth overlap ratio is greater than or equal to the second warning value;

and the second warning information generating module is used for generating second warning information for representing that the door opening or closing fault of the elevator door system occurs when the fifth overlap ratio is smaller than the second warning value.

16. The elevator door system health assessment system of claim 14, wherein the first historical operating data, the second historical operating data, the third historical operating data, the current operating data each comprise at least one of gating signal data, current data, electrical energy data, power data, and speed data.

17. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the method of health assessment of an elevator door system of any of claims 1-8.

18. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of health assessment of an elevator door system according to any of claims 1-8.

Images