CN116142913A - A method and system for analyzing equipment health status based on big data - Google Patents

Abstract

The present invention relates to the technical field of equipment health status analysis, specifically a method and system for equipment health status analysis based on big data, including a historical elevator database analysis module, an abnormal data set analysis module, a matching index calculation module, and an equipment operation effective time extraction module , Maintenance cycle calculation module and early warning response module; historical elevator database analysis module is used to obtain historical elevator fault data; abnormal data set analysis module analyzes abnormal data set corresponding to faulty components; matching index calculation module analyzes health state set and historical power failure The matching index of the abnormal data set in the data; the effective operation time extraction module of the equipment extracts the effective operation time of the equipment corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold; the maintenance period calculation module analyzes and monitors the maintenance period of the elevator equipment in real time; early warning The response module is used for early warning response according to the maintenance period.

Description

A method and system for analyzing equipment health status based on big data

technical field

The invention relates to the technical field of equipment health status analysis, in particular to a big data-based equipment health status analysis method and system.

Background technique

At present, high-load, large-capacity, and long-term use of elevators are common in China. The number of old elevators has increased sharply, and the main responsibility for elevator safety cannot be effectively implemented. In particular, vicious competition within the maintenance industry and low-cost employment directly affect the quality of elevator maintenance. , the safety of the elevator cannot be guaranteed; maintenance is an important link in the life chain of the elevator. For an elevator, the development, production, and installation are only one-off, while the maintenance is the three links of development, production, and installation. Extension, and it is the work that needs to be repeated after the elevator is put into use; but at present, most of the maintenance work on elevator equipment is carried out based on basic standards. For example, the management of elevators in the park or apartment is to maintain equipment in a fixed period , and in the process of maintenance, there is no targeted maintenance of a certain component that may be abnormal. The maintenance work is relatively regular, the maintenance content is not refined, and in many cases it is solved after the elevator equipment has a problem. It is difficult to find problems in time and minimize the difficulty of problem handling, which brings inconvenience to users and brings a lot of uncertainty to the maintenance work of maintenance personnel.

Contents of the invention

The purpose of the present invention is to provide a method and system for analyzing equipment health status based on big data, so as to solve the problems raised in the above-mentioned background technology.

In order to solve the above technical problems, the present invention provides the following technical solutions: a method for analyzing the health status of equipment based on big data, comprising the following steps:

Step S1: Obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the historical elevator fault data in the historical elevator database that is the same as the elevator model and elevator production source of the real-time monitoring elevator equipment. The elevator running state data and parameter monitoring data in the first monitoring cycle;

Step S2: Analyze the abnormal data set corresponding to the faulty component based on the historical elevator fault data;

Step S3: Based on the abnormal data set analysis, monitor the health state set of the elevator equipment in real time, and analyze the matching index between the healthy state set and the abnormal data set in the historical power failure data, and extract the equipment corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold The effective running time; the effective running time of the equipment refers to the continuous normal running time of the elevator equipment before failure in the historical data;

Step S4: Analyze and monitor the maintenance period of the elevator equipment in real time based on the effective operating time of the equipment, and perform an early warning response according to the maintenance period.

Further, step S2 includes the following analysis steps:

Step S21: Obtain the initial elevator data in the historical elevator database that is the same as the real-time monitoring elevator equipment elevator model and elevator production source. The initial elevator data is the elevator data that meets the elevator operation standards. The elevator equipment is the elevator equipment to be analyzed, and the initial elevator data is determined to ensure that there is no equipment failure safety problem before the elevator is installed and operated;

Step S22: extract the elevator running state data of the elevator equipment to be analyzed in the first monitoring period, the elevator running state data includes the temperature of the traction machine and the control cabinet, the wear degree of the elevator cable, the running speed of the elevator body and the vibration frequency of the car; Elevator running status data refers to the data that can determine that the elevator is in an abnormal state by comparing and judging the running status data with the preset abnormality early warning data; and the above data correspond to the power control data, transmission data and elevator body data respectively. Targetedly determine whether the status data of the elevator is abnormal from three aspects of the elevator equipment;

There are many elevator components, so there are many problems with elevator failures. Simply determining the elevator status from three aspects is to accurately analyze the content of the parameters to make the analysis data more targeted, instead of blindly obtaining all monitoring data for analysis; analyzing abnormal status data is in the elevator. The "sub-health" diagnosis of the equipment before the "unhealthy" state;

Step S23: When the operating status data in the first monitoring cycle does not meet the preset abnormal warning data, the abnormal warning data does not meet the preset abnormal warning data means that it does not belong to the preset abnormal warning data range or is greater than the preset abnormal warning data, and obtain the abnormal state The time of data generation is D0, the starting time of the first monitoring cycle is extracted as the starting time of the parameter analysis cycle, and D0 is the end time of the parameter analysis cycle to constitute the parameter analysis cycle, and the parameter analysis cycle is smaller than the first monitoring cycle; when the first When there is no running state data in the monitoring cycle that does not meet the preset standard state data, let the first monitoring cycle be the parameter analysis cycle;

Step S24: Based on the parameter analysis period, extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis period; determine the abnormal data set corresponding to the historical fault component based on the parameter monitoring data; user behavior data.

Determining the parameter analysis cycle is to determine the time relationship between the abnormal elevator running status data and human influence. If there is abnormal elevator running status data in the monitoring cycle, then the human influence data before the abnormal analysis can be targeted to determine the abnormal elevator running status data. If there is no abnormal state data that can be directly obtained during the monitoring cycle, then the entire monitoring cycle before the fault occurs will be used as the analysis object; and the analysis parameter monitoring data is because the elevator equipment itself of the same model and the same production source The loss of the components used is basically the same, the difference is the human influence on the components caused by the user during use, so the analysis of the human influence can effectively analyze the change relationship of the life of the elevator equipment components.

Further, step S2 also includes the following analysis steps:

Obtain the video image of the elevator running when the user exists within the monitoring range in the parameter monitoring data, and extract the user's minimum leg bending degree vi and the user's position coordinates of the i-th frame image in the video image, and the leg bending degree is It refers to the bending angle captured in the monitoring data between the user's thigh and calf when it is greater than zero and less than one hundred and eighty degrees. The user's position coordinates represent the user's position with the user's head, and the coordinate system is The bottom edge of the monitoring image is taken as the abscissa, the left edge is the ordinate, and the intersection point of the bottom edge and the left edge is the origin; the analysis of the minimum leg bending means that there may be one or more people in the elevator , when there are multiple people and all can capture the screen, only the minimum value can be used to indicate whether there is abnormal user behavior in the elevator;

When the number of users whose position coordinates are captured in the video image is the same as the number of users whose leg bending degree is captured, the number m1 of the image frames with the minimum leg bending degree in m frames of images and the minimum leg bending degree m1 are obtained. The folding degree corresponds to the total number of image frames m2 in the user's existing image, i is less than or equal to m, and m represents the total number of frames of the video image; calculate the first abnormal data index g1, g1=m1/m2; set the first abnormal data threshold g0 , extract the minimum value g1min of g1 when the jth elevator equipment to be analyzed corresponds to the kth faulty component in the parameter monitoring data when g1 is greater than g0 in the parameter analysis period, and generate the first data pair {faulty component k, g1min, time Tk }; The analysis of the minimum leg bending is to analyze whether there is a behavior of the user jumping in the elevator, because jumping in the elevator is likely to cause accidents in the elevator operation; at the same time, it will also cause additional pressure on the running cables; Reduce the service life of the elevator and cause elevator failure;

When the number of users whose position coordinates are captured in the video image is different from the number of users whose legs are bent, calculate the vertical difference between the user's position coordinates in two adjacent frames of images in m frames, and extract m-1 The number of image frames of vertical differences that are not 0 among the vertical differences is p1, calculate the second abnormal data index f1, f1=m-p1, set the second abnormal data index threshold f0, and extract the jth parameter to be analyzed in the analysis period Elevator equipment corresponding to the kth fault component has the minimum value f1min of f1 when f1 is less than f0 in the parameter monitoring data, and generates the second data pair {faulty component k, f1min, time Tk}; the analysis of the longitudinal difference is because it cannot capture When the user's legs are in condition, the monitoring can be used to capture the user's position coordinates to determine whether the user is running and jumping in the elevator;

Obtain the video image of the elevator staying on the leveling floor within the monitoring range of the user in the parameter monitoring data, and capture the number of times when the elevator door is closed when the elevator door is closed in the video image. The user's limbs or objects; obtain the total number z of normal opening and closing of the elevator in the parameter analysis cycle, and the number z1 of the number of times the elevator is closed by foreign objects, calculate the third abnormal data index s1, s1=z1/z, and set the third abnormal data Index threshold s0, extract the minimum value s1min of s1 when the jth elevator equipment to be analyzed corresponds to the kth faulty component in the parameter monitoring data when s1 is greater than s0 in the parameter analysis period, and generate the third data pair {faulty component k, s1min , time Tk}; analysis of foreign object blocking is to judge whether there is an external influence when the elevator is opening and closing the door, which leads to the behavior that the elevator cannot be closed normally. The elevator is closed multiple times in a short time and the closing is blocked. It is also one of the reasons that may cause the elevator failure;

Based on the first data pair, the second data pair and the third data pair, an abnormal data set G of the kth type of faulty component is generated, G={faulty component k, {g1min, f1min, s1min}}.

Further, step S3 includes the following analysis steps:

Step S31: Obtain the video images of the real-time monitoring elevator equipment user presence monitoring range when the elevator is running and when the elevator stays on the leveling floor, and calculate the real-time first abnormal data index g1', the real-time second abnormal data index f1' and the real-time third abnormal data index respectively. Abnormal data index s1', generate health state set G', G'={g1', f1', s1'};

Step S32: If g1'-g1min<0, then output "g1'-g1min"=0, if g1'-g1min≥0, then output "g1'-g1min"=g1'-g1min

If f1’-f1min>0, then output “f1’-f1min”=0, if f1’-f1min≤0, then output “f1’-f1min”=|f1’-f1min|;

If s1’-s1min<0, then output “s1’-s1min”=0, if s1’-s1min≥0, then output “s1’-s1min”=s1’-s1min;

Step S33: using the formula:

Calculate the matching index c of the health state set and the abnormal data set, a={1,2,3}, where a represents the number of elements in the health state set G';

Step S34: Set the matching index threshold c0. When c is greater than or equal to c0 and is the maximum value, extract the faulty component k in the abnormal data set that matches the health state set to predict the faulty component k in real time, and extract the equipment operation recorded by the faulty component k Valid duration h0.

Further, step S4 includes the following analysis steps:

Step S41: Obtain the abnormal data index of the abnormal data set corresponding to the faulty component k, and calculate the average abnormal data index b0; obtain the real-time abnormal data index of the health state set corresponding to the faulty component k in real time, and calculate the real-time average abnormal data index b0' ;

Step S42: Using the formula: h1=(h0*b0)/b0', calculate the maintenance period h1 of the real-time predicted equipment operation; when the matching real-time predicted faulty component is not unique, select the minimum value of the maintenance period of the real-time predicted equipment operation as maintenance Warranty period h1; the above formula indicates that the abnormal data index and the maintenance cycle are inversely related. When the abnormal data index increases, the corresponding maintenance cycle will decrease under ideal conditions, because the increase of the abnormal data index indicates that there is a user in the elevator. If the behavior of causing damage is intensified, the degree of damage to the elevator will increase, the corresponding elevator life will be reduced, and the maintenance time will need to be shortened from the original time;

Step S43: Obtain the real-time monitored period d1, if d1≥h1, immediately transmit the maintenance warning response guarantee corresponding to the faulty component k; if d1<h1, then analyze the real-time prediction fault number I of the current elevator equipment history record of the faulty component, If I=0, the duration d2=0 of the output rate of change, if I≥1, then make d2 equal to the average value of the time difference of two adjacent identical fault parts in the historical elevator fault data; Whether there is a difference in the running time of the equipment caused by the first failure and the second failure of the same faulty component, because equipment components may have an impact on the subsequent service life after multiple failures, such as the first failure at an interval of 30 days , the second time is a failure at an interval of 25 days, and the third time is a failure at an interval of 20 days. The maintenance period of the equipment is getting shorter and shorter due to its own damage;

Step S44: When h1-d1-d2≤0, immediately transmit the maintenance warning response corresponding to the faulty component k; when h1-d1-d2>0, transmit the corresponding faulty component k after h1-d1-d2 days Maintenance warning response.

An equipment health status analysis system based on big data, including a historical elevator database analysis module, an abnormal data collection analysis module, a matching index calculation module, an equipment operation effective time extraction module, a maintenance cycle calculation module, and an early warning response module;

The historical elevator database analysis module is used to obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the same historical elevator fault data as the real-time monitoring elevator equipment elevator model and elevator production source in the historical elevator database;

The abnormal data set analysis module is used to analyze the abnormal data set of corresponding faulty components based on historical elevator fault data;

The matching index calculation module is used to analyze the matching index of the health state set and the abnormal data set in the historical power failure data;

The effective device operating time extraction module is used to extract the effective operating time of the device corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold;

The maintenance cycle calculation module is used to analyze and monitor the maintenance cycle of elevator equipment in real time based on the effective operation time of the equipment;

The early warning response module is used for early warning response according to the maintenance period.

Further, the abnormal data set analysis module includes an elevator running state data acquisition unit, a parameter analysis period determination unit and an abnormal data set generation unit;

The elevator running state data acquisition unit is used for running state data based on the elevator equipment to be analyzed;

The parameter analysis cycle determination unit is used to determine the parameter analysis cycle based on the abnormally divided cycle of the elevator running state data;

The abnormal data set generation unit is used to extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis cycle, analyze the first abnormal data index, the second abnormal data index and the third abnormal data index to generate an abnormal data set.

Further, the matching index calculation module includes a health state set acquisition unit, an index output analysis unit and a matching index output unit;

The health state set acquisition unit is used to obtain the real-time first abnormal data index, the real-time second abnormal data index and the real-time third abnormal data index to form a health state set;

The index output analysis unit is used to analyze the numerical relationship between the real-time abnormal value index and the historical abnormal data index;

The matching index output unit is used to calculate the matching index and output corresponding real-time predicted faulty components according to the matching index, and extract the effective operation time of the equipment recorded by the faulty components.

Further, the maintenance cycle calculation module includes an average abnormal data index calculation unit, a real-time average abnormal data index calculation unit and a maintenance cycle output unit;

The average abnormal data index calculation unit is used to obtain the abnormal data index existing in the abnormal data set corresponding to the faulty component, and calculate the average abnormal data index;

The real-time average abnormal data index calculation unit is used to obtain the real-time abnormal data index of the health state set corresponding to the real-time predicted fault component, and calculate the real-time average abnormal data index;

The maintenance cycle output unit is used to calculate the maintenance cycle, and transmit the maintenance time and the maintenance signal corresponding to the faulty component according to the time relationship between the monitored cycle and the maintenance cycle.

Compared with the prior art, the beneficial effects achieved by the present invention are: the present invention determines the influence relationship of human factors reflected in the historical fault data by analyzing the historical fault data of the elevator of the same model, because in the normal use of the elevator It is inevitable that the use of the equipment will cause the aging of the equipment, and the difference is the user's standardization degree when using the elevator. The present invention analyzes the abnormal data collection of historical fault data and monitors the health status collection in real time. Judging the effective running time of faulty components and equipment corresponding to the most matching abnormal data set under the real-time health state set, so that the maintenance time required for real-time elevator operation can be inferred, and the maintenance personnel can be given according to this time and the corresponding predicted faulty components Point out targeted maintenance parts during the work process, and at the same time, it can effectively predict the failure before the problem occurs, discover the "sub-health" of the equipment in time, and reduce the difficulty of problem handling to a minimum.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the attached picture:

Fig. 1 is a schematic structural diagram of a device health status analysis system based on big data in the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Please refer to Fig. 1, the present invention provides technical scheme: a kind of equipment health status analysis method based on big data, comprises the following steps:

Step S1: Obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the historical elevator fault data in the historical elevator database that is the same as the elevator model and elevator production source of the real-time monitoring elevator equipment. The elevator running state data and parameter monitoring data in the first monitoring cycle;

Step S2: Analyze the abnormal data set corresponding to the faulty component based on the historical elevator fault data;

Step S2 includes the following analysis steps:

Step S21: Obtain the initial elevator data in the historical elevator database that is the same as the real-time monitoring elevator equipment elevator model and elevator production source. The initial elevator data is the elevator data that meets the elevator operation standards. The elevator equipment is the elevator equipment to be analyzed, and the initial elevator data is determined to ensure that there is no equipment failure safety problem before the elevator is installed and operated;

Step S22: extract the elevator running state data of the elevator equipment to be analyzed in the first monitoring period, the elevator running state data includes the temperature of the traction machine and the control cabinet, the wear degree of the elevator cable, the running speed of the elevator body and the vibration frequency of the car; Elevator running status data refers to the data that can determine that the elevator is in an abnormal state by comparing and judging the running status data with the preset abnormality early warning data; and the above data correspond to the power control data, transmission data and elevator body data respectively. Targetedly determine whether the status data of the elevator is abnormal from three aspects of the elevator equipment;

For example, set the preset abnormal warning temperature range [w1,w2] of the traction machine and control cabinet; too high temperature and too low temperature will have a certain impact on the elevator control equipment; when the temperature of the traction machine and control cabinet belongs to the abnormal warning In the temperature range, it means that the traction machine and the control cabinet have abnormalities, but at this time it will not cause the failure of the elevator equipment;

The wear degree of the elevator cable can be averaged according to the cross-sectional diameter of the elevator cable measured many times. When the initial diameter is a1 and the current average diameter is a2, the cable wear degree (a1-a2)/a1 is calculated. When the cable wear When the wear degree is greater than 7%, you need to replace the new cable; you can set the preset abnormal early warning wear degree in this application to [4%, 5%],

The running speed of the elevator body refers to the rated speed of the fast elevator as [1M/S, 2M/S], and the rated speed of the high-speed elevator as [2M/S, 4M/S], etc. Different types of elevators set different standard operating speeds ;The preset abnormal warning speed of the fast elevator can be set to 1±0.5 and 2±0.5; and when the running speed of the elevator body refers to the running speed of the elevator from the leveling floor in the building to another leveling floor;

The vibration frequency of the car can be set to a preset abnormal warning vibration frequency b;

There are many elevator components, so there are many problems with elevator failures. Simply determining the elevator status from three aspects is to accurately analyze the content of the parameters to make the analysis data more targeted, instead of blindly obtaining all monitoring data for analysis; analyzing abnormal status data is in the elevator. The "sub-health" diagnosis of the equipment before the "unhealthy" state;

Step S23: When the operating status data in the first monitoring cycle does not meet the preset abnormal warning data, the abnormal warning data does not meet the preset abnormal warning data means that it does not belong to the preset abnormal warning data range or is greater than the preset abnormal warning data, and obtain the abnormal state The time of data generation is D0, the starting time of the first monitoring cycle is extracted as the starting time of the parameter analysis cycle, and D0 is the end time of the parameter analysis cycle to constitute the parameter analysis cycle, and the parameter analysis cycle is smaller than the first monitoring cycle; when the first When there is no running state data in the monitoring cycle that does not meet the preset standard state data, let the first monitoring cycle be the parameter analysis cycle;

Step S24: Based on the parameter analysis period, extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis period; determine the abnormal data set corresponding to the historical fault component based on the parameter monitoring data; user behavior data.

Determining the parameter analysis cycle is to determine the time relationship between the abnormal elevator running status data and human influence. If there is abnormal elevator running status data in the monitoring cycle, then the human influence data before the abnormal analysis can be targeted to determine the abnormal elevator running status data. If there is no abnormal state data that can be directly obtained during the monitoring cycle, then the entire monitoring cycle before the fault occurs will be used as the analysis object; and the analysis parameter monitoring data is because the elevator equipment itself of the same model and the same production source The loss of the components used is basically the same, the difference is the human influence on the components caused by the user during use, so the analysis of the human influence can effectively analyze the change relationship of the life of the elevator equipment components.

Step S2 also includes the following analysis steps:

Obtain the video image of the elevator running when the user exists within the monitoring range in the parameter monitoring data, and extract the user's minimum leg bending degree vi and the user's position coordinates of the i-th frame image in the video image, and the leg bending degree is It refers to the bending angle captured in the monitoring data between the user's thigh and calf when it is greater than zero and less than one hundred and eighty degrees. The user's position coordinates represent the user's position with the user's head, and the coordinate system is The bottom edge of the monitoring image is taken as the abscissa, the left edge is the ordinate, and the intersection point of the bottom edge and the left edge is the origin; the analysis of the minimum leg bending means that there may be one or more people in the elevator , when there are multiple people and all can capture the screen, only the minimum value can be used to indicate whether there is abnormal user behavior in the elevator;

When the number of users whose position coordinates are captured in the video image is the same as the number of users whose leg bending degree is captured, the number m1 of the image frames with the minimum leg bending degree in m frames of images and the minimum leg bending degree m1 are obtained. The folding degree corresponds to the total number of image frames m2 in the user's existing image, i is less than or equal to m, and m represents the total number of frames of the video image; calculate the first abnormal data index g1, g1=m1/m2; set the first abnormal data threshold g0 , extract the minimum value g1min of g1 when the jth elevator equipment to be analyzed corresponds to the kth faulty component in the parameter monitoring data when g1 is greater than g0 in the parameter analysis period, and generate the first data pair {faulty component k, g1min, time Tk }; The analysis of the minimum leg bending is to analyze whether there is a behavior of the user jumping in the elevator, because jumping in the elevator is likely to cause accidents in the elevator operation; at the same time, it will also cause additional pressure on the running cables; Reduce the service life of the elevator and cause elevator failure;

When the number of users whose position coordinates are captured in the video image is different from the number of users whose legs are bent, calculate the vertical difference between the user's position coordinates in two adjacent frames of images in m frames, and extract m-1 The number of image frames of vertical differences that are not 0 among the vertical differences is p1, calculate the second abnormal data index f1, f1=m-p1, set the second abnormal data index threshold f0, and extract the jth parameter to be analyzed in the analysis period Elevator equipment corresponding to the kth fault component has the minimum value f1min of f1 when f1 is less than f0 in the parameter monitoring data, and generates the second data pair {faulty component k, f1min, time Tk}; the analysis of the longitudinal difference is because it cannot capture When the user's legs are in condition, the monitoring can be used to capture the user's position coordinates to determine whether the user is running and jumping in the elevator;

Obtain the video image of the elevator staying on the leveling floor within the monitoring range of the user in the parameter monitoring data, and capture the number of times when the elevator door is closed when the elevator door is closed in the video image. The user's limbs or objects; obtain the total number z of normal opening and closing of the elevator in the parameter analysis cycle, and the number z1 of the number of times the elevator is closed by foreign objects, calculate the third abnormal data index s1, s1=z1/z, and set the third abnormal data Index threshold s0, extract the minimum value s1min of s1 when the jth elevator equipment to be analyzed corresponds to the kth faulty component in the parameter monitoring data when s1 is greater than s0 in the parameter analysis period, and generate the third data pair {faulty component k, s1min , time Tk}; analysis of foreign object blocking is to judge whether there is an external influence when the elevator is opening and closing the door, which leads to the behavior that the elevator cannot be closed normally. The elevator is closed multiple times in a short time and the closing is blocked. It is also one of the reasons that may cause the elevator failure;

Based on the first data pair, the second data pair and the third data pair, an abnormal data set G of the kth type of faulty component is generated, G={faulty component k, {g1min, f1min, s1min}}.

As shown in the embodiment: when there is only one kind of data pair in the abnormal data set, {g1min, f1min, s1min} in the abnormal data set G can only select one existing data pair to generate; When the two data are aligned, {g1min, f1min, s1min}} in the abnormal data set G are sorted in chronological order according to the corresponding time Tk of the respective data pairs.

Step S3: Based on the abnormal data set analysis, monitor the health state set of the elevator equipment in real time, and analyze the matching index between the healthy state set and the abnormal data set in the historical power failure data, and extract the equipment corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold The effective running time; the effective running time of the equipment refers to the continuous normal running time of the elevator equipment before failure in the historical data;

Step S3 comprises the following analysis steps:

Step S31: Obtain the video images of the real-time monitoring elevator equipment user presence monitoring range when the elevator is running and when the elevator stays on the leveling floor, and calculate the real-time first abnormal data index g1', the real-time second abnormal data index f1' and the real-time third abnormal data index respectively. The abnormal data index s1' generates a health state set G', G'={g1', f1', s1'}; the calculation method of the real-time abnormal data index is the same as that of the historical data corresponding index;

Step S32: If g1'-g1min<0, then output "g1'-g1min"=0, if g1'-g1min≥0, then output "g1'-g1min"=g1'-g1min

If f1’-f1min>0, then output “f1’-f1min”=0, if f1’-f1min≤0, then output “f1’-f1min”=|f1’-f1min|;

If s1’-s1min<0, then output “s1’-s1min”=0, if s1’-s1min≥0, then output “s1’-s1min”=s1’-s1min;

Step S33: using the formula:

Calculate the matching index c of the health state set and the abnormal data set, a={1,2,3}, a represents the number of elements in the health state set G'; here when there is only one element number in the health state set such as g1 ', the selected abnormal data set must contain at least g1min, analyze the matching index with the abnormal data set with g1min, then at this time a=1, only need to calculate g1'-g1min in the molecule; select the most matching abnormal data That is, the maximum value corresponding to the faulty part can represent the possible damaged part analyzed in real-time state;

Step S34: Set the matching index threshold c0. When c is greater than or equal to c0 and is the maximum value, extract the faulty component k in the abnormal data set that matches the health state set to predict the faulty component k in real time, and extract the equipment operation recorded by the faulty component k Valid duration h0.

Step S4: Analyze and monitor the maintenance period of the elevator equipment in real time based on the effective operating time of the equipment, and perform an early warning response according to the maintenance period.

Step S4 comprises the following analysis steps:

Step S41: Obtain the abnormal data index of the abnormal data set corresponding to the faulty component k, and calculate the average abnormal data index b0; obtain the real-time abnormal data index of the health state set corresponding to the faulty component k in real time, and calculate the real-time average abnormal data index b0' ;

Step S42: Using the formula: h1=(h0*b0)/b0', calculate the maintenance period h1 of the real-time predicted equipment operation; when the matching real-time predicted faulty component is not unique, select the minimum value of the maintenance period of the real-time predicted equipment operation as maintenance Warranty period h1; the above formula indicates that the abnormal data index and the maintenance cycle are inversely related. When the abnormal data index increases, the corresponding maintenance cycle will decrease under ideal conditions, because the increase of the abnormal data index indicates that there is a user in the elevator. If the behavior of causing damage is intensified, the degree of damage to the elevator will increase, the corresponding elevator life will be reduced, and the maintenance time will need to be shortened from the original time;

Step S43: Obtain the real-time monitored period d1, if d1≥h1, immediately transmit the maintenance warning response guarantee corresponding to the faulty component k; if d1<h1, then analyze the real-time prediction fault number I of the current elevator equipment history record of the faulty component, If I=0, the duration d2=0 of the output rate of change, if I≥1, then make d2 equal to the average value of the time difference of two adjacent identical fault parts in the historical elevator fault data; Whether there is a difference in the running time of the equipment caused by the first failure and the second failure of the same faulty component, because equipment components may have an impact on the subsequent service life after multiple failures, such as the first failure at an interval of 30 days , the second time is a failure at an interval of 25 days, and the third time is a failure at an interval of 20 days. The maintenance period of the equipment is getting shorter and shorter due to its own damage;

Step S44: When h1-d1-d2≤0, immediately transmit the maintenance warning response corresponding to the faulty component k; when h1-d1-d2>0, transmit the corresponding faulty component k after h1-d1-d2 days Maintenance warning response.

If the computing department predicts that the maintenance cycle of the equipment is 3 months, the current monitoring period is 40 days, and the change rate is 10 days, then 90-40-10=40>0, then the corresponding fault will be checked after 40 days of current monitoring Components carry out early warning response maintenance.

An equipment health status analysis system based on big data, including a historical elevator database analysis module, an abnormal data collection analysis module, a matching index calculation module, an equipment operation effective time extraction module, a maintenance cycle calculation module, and an early warning response module;

The historical elevator database analysis module is used to obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the same historical elevator fault data as the real-time monitoring elevator equipment elevator model and elevator production source in the historical elevator database;

The abnormal data set analysis module is used to analyze the abnormal data set of the corresponding faulty component based on the historical elevator fault data;

The matching index calculation module is used to analyze the matching index of the health state set and the abnormal data set in the historical power failure data;

The effective device operating time extraction module is used to extract the effective operating time of the device corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold;

The maintenance cycle calculation module is used to analyze and monitor the maintenance cycle of elevator equipment in real time based on the effective operation time of the equipment;

The early warning response module is used for early warning response according to the maintenance period.

Further, the abnormal data set analysis module includes an elevator running state data acquisition unit, a parameter analysis period determination unit and an abnormal data set generation unit;

The elevator running state data acquisition unit is used for running state data based on the elevator equipment to be analyzed;

The parameter analysis cycle determination unit is used to determine the parameter analysis cycle based on the abnormally divided cycle of the elevator running state data;

The abnormal data set generation unit is used to extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis cycle, analyze the first abnormal data index, the second abnormal data index and the third abnormal data index to generate an abnormal data set.

Further, the matching index calculation module includes a health state set acquisition unit, an index output analysis unit and a matching index output unit;

The health state set acquisition unit is used to obtain the real-time first abnormal data index, the real-time second abnormal data index and the real-time third abnormal data index to form a health state set;

The index output analysis unit is used to analyze the numerical relationship between the real-time abnormal value index and the historical abnormal data index;

The matching index output unit is used to calculate the matching index and output corresponding real-time predicted faulty components according to the matching index, and extract the effective operation time of the equipment recorded by the faulty components.

Further, the maintenance cycle calculation module includes an average abnormal data index calculation unit, a real-time average abnormal data index calculation unit and a maintenance cycle output unit;

The average abnormal data index calculation unit is used to obtain the abnormal data index existing in the abnormal data set corresponding to the faulty component, and calculate the average abnormal data index;

The real-time average abnormal data index calculation unit is used to obtain the real-time abnormal data index of the health state set corresponding to the real-time predicted fault component, and calculate the real-time average abnormal data index;

The maintenance cycle output unit is used to calculate the maintenance cycle, and transmit the maintenance time and the maintenance signal corresponding to the faulty component according to the time relationship between the monitored cycle and the maintenance cycle.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is a relationship between these entities or operations. any such actual relationship or order exists between them. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or apparatus.

Finally, it should be noted that: the above is only a preferred embodiment of the present invention, and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, it still The technical solutions recorded in the foregoing embodiments may be modified, or some technical features thereof may be equivalently replaced. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (9)

1. A method for analyzing equipment health status based on big data, characterized in that, comprising the following steps: Step S1: Obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the same historical elevator fault data as the elevator model and elevator production source of the real-time monitoring elevator equipment in the historical elevator database. Elevator running status data and parameter monitoring data in the first monitoring period before the failure; Step S2: Analyze the abnormal data set corresponding to the faulty component based on the historical elevator fault data; Step S3: Based on the abnormal data set analysis, monitor the health state set of the elevator equipment in real time, and analyze the matching index between the healthy state set and the abnormal data set in the historical power failure data, and extract the equipment corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold Effective running time; the effective running time of the equipment refers to the sustainable normal running time of the elevator equipment in the historical data before failure; Step S4: Analyze and monitor the maintenance period of the elevator equipment in real time based on the effective operating time of the equipment, and perform an early warning response according to the maintenance period. 2. A method for analyzing the health status of equipment based on big data according to claim 1, wherein said step S2 comprises the following analysis steps: Step S21: Obtain the initial elevator data in the historical elevator database that is the same as the real-time monitoring elevator equipment elevator model and elevator production source. The initial elevator data is elevator data that conforms to the elevator operation standard specification, so that there are initial elevator data and historical elevator faults The elevator equipment of the data is the elevator equipment to be analyzed; Step S22: extract the elevator running status data of the elevator equipment to be analyzed in the first monitoring period, the elevator running status data includes the temperature of the traction machine and the control cabinet, the wear degree of the elevator cable, the running speed of the elevator body and the vibration of the car Frequency; the elevator running state data refers to the data that can determine that the elevator is in an abnormal state by comparing and judging the running state data and the preset abnormal early warning data; Step S23: When the operating status data in the first monitoring period does not meet the preset abnormal warning data, the said abnormal warning data does not belong to the preset abnormal warning data range or is greater than the preset abnormal warning data, and obtain The time when the abnormal state data is generated is D0, the starting time in the first monitoring cycle is extracted as the starting time of the parameter analysis cycle, and D0 is the end time of the parameter analysis cycle to form a parameter analysis cycle, and the parameter analysis cycle is smaller than the first monitoring cycle ; When there is no operating state data in the first monitoring cycle that does not meet the preset standard state data, the first monitoring cycle is the parameter analysis cycle; Step S24: Based on the parameter analysis cycle, extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis cycle; determine the abnormal data set corresponding to the historical fault component based on the parameter monitoring data; the parameter monitoring data refers to the user entering the elevator monitoring device Captured user behavior data. 3. A kind of equipment health state analysis method based on big drama according to claim 2, it is characterized in that: described step S2 also comprises the following analysis steps: Obtain the video image when the elevator is running within the monitoring range of the user in the parameter monitoring data, extract the user's minimum leg bending degree vi and the user's position coordinates of the i-th frame image in the video image, and the leg bending The degree refers to the bending angle captured in the monitoring data between the user's thigh and calf when it is greater than zero and less than one hundred and eighty degrees. The user's position coordinates represent the user's position with the user's head, And the coordinate system is formed by taking the lower bottom of the monitoring image as the abscissa, the left side as the ordinate, and the intersection of the lower bottom and the left side as the origin; When the number of users whose position coordinates are captured in the video image is the same as the number of users whose leg bending degree is captured, the number m1 of the image frames with the minimum leg bending degree in m frames of images and the minimum leg bending degree m1 are obtained. The folding degree corresponds to the total number of image frames m2 in the user's existing image, i is less than or equal to m, and m represents the total number of frames of the video image; calculate the first abnormal data index g1, g1=m1/m2; set the first abnormal data threshold g0 , extract the minimum value g1min of g1 when the jth elevator equipment to be analyzed corresponds to the kth faulty component in the parameter monitoring data when g1 is greater than g0 in the parameter analysis period, and generate the first data pair {faulty component k, g1min, time Tk }; When the number of users whose position coordinates are captured in the video image is different from the number of users whose legs are bent, calculate the vertical difference between the user's position coordinates in two adjacent frames of images in m frames, and extract m-1 The number of image frames of vertical differences that are not 0 among the vertical differences is p1, calculate the second abnormal data index f1, f1=m-p1, set the second abnormal data index threshold f0, and extract the jth parameter to be analyzed in the analysis period The elevator equipment corresponds to the kth type of faulty component, and the minimum value f1min of f1 exists when f1 is less than f0 in the parameter monitoring data, and generates the second data pair {faulty component k, f1min, time Tk}; Obtain the video image of the elevator staying on the leveling floor within the monitoring range of the user in the parameter monitoring data, and capture the number of times the elevator door is blocked by foreign objects in the video image. The user's limbs or objects when opening; obtain the total number z of normal opening and closing of the elevator in the parameter analysis cycle, and the number z1 of the number of times the elevator is closed by foreign objects, calculate the third abnormal data index s1, s1=z1/z, set the third Abnormal data index threshold s0, extract the minimum value s1min of s1 when s1 is greater than s0 in the parameter monitoring data corresponding to the jth elevator equipment to be analyzed corresponding to the kth type of faulty component in the parameter analysis period, and generate the third data pair {faulty component k , s1min, time Tk}; Based on the first data pair, the second data pair and the third data pair, an abnormal data set G of the kth type of faulty component is generated, G={faulty component k, {g1min, f1min, s1min}}. 4. A method for analyzing equipment health status based on big data according to claim 1, characterized in that: said step S3 comprises the following analysis steps: Step S31: Obtain the video images of the real-time monitoring elevator equipment user presence monitoring range when the elevator is running and when the elevator stays on the leveling floor, and calculate the real-time first abnormal data index g1', the real-time second abnormal data index f1' and the real-time third abnormal data index respectively. Abnormal data index s1', generate health state set G', G'={g1', f1', s1'}; Step S32: If g1'-g1min<0, then output "g1'-g1min"=0, if g1'-g1min≥0, then output "g1'-g1min"=g1'-g1min If f1’-f1min>0, then output “f1’-f1min”=0, if f1’-f1min≤0, then output “f1’-f1min”=|f1’-f1min|; If s1’-s1min<0, then output “s1’-s1min”=0, if s1’-s1min≥0, then output “s1’-s1min”=s1’-s1min; Step S33: using the formula:

Calculate the matching index c of the health state set and the abnormal data set, a={1,2,3}, where a represents the number of elements in the health state set G'; Step S34: Set the matching index threshold c0. When c is greater than or equal to c0 and is the maximum value, extract the faulty component k in the abnormal data set that matches the health state set to predict the faulty component k in real time, and extract the equipment operation recorded by the faulty component k Valid duration h0. 5. A method for analyzing the health status of equipment based on big data according to claim 4, characterized in that: said step S4 comprises the following analysis steps: Step S41: Obtain the abnormal data index of the abnormal data set corresponding to the faulty component k, and calculate the average abnormal data index b0; obtain the real-time abnormal data index of the health state set corresponding to the faulty component k in real time, and calculate the real-time average abnormal data index b0' ; Step S42: Using the formula: h1=(h0*b0)/b0', calculate the maintenance period h1 of the real-time predicted equipment operation; when the matching real-time predicted faulty component is not unique, select the minimum value of the maintenance period of the real-time predicted equipment operation as maintenance Guarantee period h1; Step S43: Obtain the real-time monitored period d1, if d1≥h1, immediately transmit the maintenance warning response guarantee corresponding to the faulty component k; if d1<h1, then analyze the real-time prediction fault number I of the current elevator equipment history record of the faulty component, If I=0, output rate of change duration d2=0, if I≥1, then make d2 equal to the average value of the time difference of two adjacent identical fault parts in the historical elevator fault data; Step S44: When h1-d1-d2≤0, immediately transmit the maintenance warning response corresponding to the faulty component k; when h1-d1-d2>0, transmit the corresponding faulty component k after h1-d1-d2 days Maintenance warning response. 6. A kind of equipment health status analysis system based on big data of a kind of equipment health status analysis method based on big data described in any one of claims 1-5, it is characterized in that, comprises history elevator database analysis module, Abnormal data set analysis module, matching index calculation module, equipment operation effective time extraction module, maintenance cycle calculation module and early warning response module; The historical elevator database analysis module is used to obtain the elevator model and elevator production source of the real-time monitoring elevator equipment, and extract the same historical elevator fault data as the real-time monitoring elevator equipment elevator model and elevator production source in the historical elevator database; The abnormal data set analysis module is used to analyze the abnormal data set corresponding to the faulty component based on historical elevator fault data; The matching index calculation module is used to analyze the matching index of the health state set and the abnormal data set in the historical power failure data; The effective operating time extraction module of the device is used to extract the effective operating time of the device corresponding to the abnormal data set whose matching index is greater than or equal to the matching index threshold; The maintenance period calculation module is used to analyze and monitor the maintenance period of the elevator equipment in real time based on the effective duration of equipment operation; The early warning response module is used for performing early warning response according to the maintenance period. 7. A big data-based equipment health status analysis system according to claim 6, characterized in that: the abnormal data set analysis module includes an elevator running state data acquisition unit, a parameter analysis cycle determination unit and an abnormal data set generation unit; The elevator running state data acquisition unit is used for running state data based on the elevator equipment to be analyzed; The parameter analysis cycle determination unit is used to determine the parameter analysis cycle based on the abnormally divided cycle of the elevator running state data; The abnormal data set generation unit is used to extract the parameter monitoring data corresponding to the elevator equipment to be analyzed in the parameter analysis cycle, analyze the first abnormal data index, the second abnormal data index and the third abnormal data index to generate an abnormal data set. 8. A device health status analysis system based on big data according to claim 7, wherein the matching index calculation module includes a health status set acquisition unit, an index output analysis unit, and a matching index output unit; The health state set acquisition unit is used to obtain a real-time first abnormal data index, a real-time second abnormal data index and a real-time third abnormal data index to form a health state set; The index output analysis unit is used to analyze the numerical relationship between the real-time abnormal numerical index and the historical abnormal data index; The matching index output unit is used to calculate the matching index and output the corresponding real-time predicted faulty component according to the matching index, and extract the effective operation time of the equipment recorded by the faulty component. 9. A big data-based equipment health status analysis system according to claim 8, wherein the maintenance cycle calculation module includes an average abnormal data index calculation unit, a real-time average abnormal data index calculation unit, and a maintenance cycle calculation unit. Periodic output unit; The average abnormal data index calculation unit is used to obtain the abnormal data index existing in the abnormal data set corresponding to the faulty component, and calculate the average abnormal data index; The real-time average abnormal data index calculation unit is used to obtain the real-time abnormal data index of the health state set corresponding to the predicted fault component in real time, and calculate the real-time average abnormal data index; The maintenance cycle output unit is used to calculate the maintenance cycle, and transmit the maintenance time and the maintenance signal corresponding to the faulty component according to the time relationship between the monitored cycle and the maintenance cycle.

Images