CN116050717B - A big data-based data management method for elevator working condition cloud platform - Google Patents

Abstract

The present invention relates to the technical field of electrical digital data processing, and in particular to a big data-based cloud platform data management method for elevator operating conditions; obtain the serialized data corresponding to the door system on each floor when the elevator is running on the same day; The opening time from opening to closing of the elevator door; obtain the standard opening time, and calculate the first probability of an abnormality in the opening time, the abnormality includes delayed closing abnormality and early closing abnormality; calculate the probability that the early closing abnormality and delayed closing abnormality are caused by human factors , correct the first probability to obtain the second probability; divide the serialized data into normal data and suspected abnormal data; calculate the abnormal degree of suspected abnormal data according to the second probability; obtain the storage time according to the abnormal degree, and compress the suspected abnormal data And perform storage corresponding to the storage time. The invention can adaptively adjust the storage time of different data.

Description

A big data-based data management method for elevator working condition cloud platform

technical field

The invention relates to the technical field of electrical digital data processing, in particular to a big data-based cloud platform data management method for elevator operating conditions.

Background technique

With the continuous development of urban construction and the increasing number of high-rise buildings, elevators, as a means of transportation running vertically in high-rise buildings, are inseparable from people's daily life. Therefore, the operation safety inspection of elevators is very important; traditional elevator maintenance is mainly through professional Personnel carry out regular maintenance on the elevator, which cannot meet the needs of on-demand maintenance, and has a certain delay. With the development of the Internet of Things, it has become the mainstream method of elevator maintenance to realize intelligent analysis of elevator working conditions based on big data by collecting various data during elevator operation, which can meet the needs of on-demand maintenance.

However, when collecting various data during elevator operation, the elevator often collects a lot of data due to its mobility, and new data is generated every moment. The establishment of the elevator operation data is of great significance; the current management of the storage time of the elevator operation data is only to set a fixed storage time and regularly clean up the out-of-timeliness data, and the utilization rate of its storage space is low; at the same time, some elevator maintenance has high Data with high value and reference significance has the same storage time as ordinary data, which cannot highlight the importance of data with high value and reference significance, resulting in the inability to quickly obtain valuable data information during elevator maintenance, resulting in low maintenance efficiency.

Contents of the invention

In order to solve the above-mentioned technical problem of low utilization rate of storage space caused by the fixed storage time, the purpose of the present invention is to provide a data management method based on big data on the elevator working condition cloud platform, and the adopted technical solution is as follows:

Obtain the serialized data corresponding to the door system on each floor when the elevator is running on the same day; collect the opening time corresponding to each elevator door opening to closing on each floor; obtain the standard opening time, and calculate the opening time according to the difference between the opening time and the standard opening time The first probability of an abnormality in the duration; the abnormality includes a delayed closing exception and an early closing exception;

Set the waiting area, and calculate the delay closing time based on the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, the number of people who move out of the elevator after the elevator door is opened and enter the elevator, and the number of people who are fully loaded in the elevator. The probability that the abnormality is caused by human factors; according to the probability that the delayed closing abnormality is caused by human factors, calculate the probability that the early closing abnormalities are caused by human factors;

The first probability is corrected according to the probability that the delayed closing abnormality and the early closing abnormality are caused by human factors, and the second probability is obtained;

According to the difference between the opening time of two adjacent elevator doors from opening to closing on each floor and the difference threshold, determine whether there is suspected abnormality in each opening time, and then divide the serialized data corresponding to each floor into normal data and suspected abnormal data ;

According to the second probability corresponding to the suspected abnormality in the opening time, calculate the abnormality degree corresponding to the suspected abnormal data; calculate the storage time corresponding to the suspected abnormal data storage according to the abnormality degree; compress the suspected abnormal data and store the corresponding storage time.

Preferably, the method for obtaining the standard opening duration is: setting the number of days of historical collection, and for one of the days of historical collection, under the normal operation state of the elevator, divide the time length from the start of operation to the stop of the elevator in one day into at least Two time periods, collect the opening time corresponding to each elevator door opening to closing on each floor, and count the number of elevator door openings in each time period and the corresponding opening time of each elevator door opening to closing in units of floors Duration, calculate the average value of the opening duration corresponding to each time period, and record it as the first average opening duration;

If the current day is a working day, then select the first average opening duration corresponding to each working day in each time period in the historical collection days, calculate the average value of the first average opening duration corresponding to all working days in each time period, and divide it into It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and it is recorded as the standard opening time corresponding to each new time period of the working day;

If the day is a rest day, then select the first average open duration corresponding to each rest day in each time period in the historical collection days, calculate the average value of the first average open duration corresponding to all rest days in each time period, and divide it It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and recorded as the standard opening time corresponding to each new time period on the rest day.

Preferably, the method of calculating the storage time corresponding to the suspected abnormal data storage according to the abnormality degree is: setting a fixed storage time and a first fixed storage time, and obtaining the maximum abnormality degree among the abnormalities corresponding to the suspected abnormal data of all floors and the minimum abnormal degree; for the abnormal degree corresponding to the suspected abnormal data of a certain floor, the difference between the abnormal degree and the minimum abnormal degree is calculated as the first difference, and the difference between the maximum abnormal degree and the minimum abnormal degree is calculated as It is recorded as the second difference, the ratio of the first difference and the second difference is recorded as the characteristic factor, the product of the characteristic factor and the first fixed storage time is calculated, and the sum of the product and the fixed storage time is used as the floor The corresponding storage time when the suspected abnormal data is stored.

Preferably, according to the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, the number of people who move out of the elevator and enter the elevator after the elevator door is opened, and the full number of people in the elevator, the delay is calculated. The probability that the abnormality is caused by human factors, including: calculating the ratio of the sum of the number of people who have moved out of the elevator and the number of people entering the elevator after the elevator door is opened, and the ratio of the number of people who are twice the elevator's full load, to obtain the first feature; Calculate the elevator door The ratio of the number of people entering the waiting area after opening to the sum of the number of people in the waiting area and the adjustment parameter when the elevator door is not opened can be used to obtain the second feature, and the product of the first feature and the second feature is normalized and mapped to obtain the delay Probability that when the shutdown abnormality is caused by human factors; among them, the adjustment parameter is greater than 0.

Preferably, according to the probability that the delayed closing abnormality is caused by human factors, the method for calculating the probability that the early closing abnormalities are caused by human factors is as follows: take the natural constant as the base number, and take the probability of negative delayed closing abnormalities as the index The value of the exponential function as the probability of an early shutdown anomaly caused by an artifact.

Preferably, according to the difference between the opening time and the standard opening time, the first probability of abnormal opening time is calculated, including: for any one floor, the opening time corresponding to the opening of the elevator door once to the closing time, obtaining the time when the elevator door is opened this time belongs to A new period of time, get the standard opening time corresponding to the opening time, calculate the absolute value of the difference between the opening time and the standard opening time, get the difference between the opening time and the standard opening time, calculate the difference between the opening time and the standard opening time The ratio of the sum between the openings, and the ratio is recorded as the first probability that an abnormality occurs during the opening time.

Preferably, the first probability is corrected according to the probability that the delayed closing abnormality and the early closing abnormality are caused by human factors, and the method for obtaining the second probability is: for any floor, the opening time corresponding to one elevator door opening to closing , to determine the size of the opening time and the corresponding standard opening time, when the opening time is greater than or equal to the standard opening time, calculate the difference between 1 and the probability that the delayed closing abnormality is caused by human factors, and compare the first probability with the difference The product of is recorded as the second probability; when the opening time is less than the standard opening time, calculate the difference between 1 and the probability that the early closing abnormality is caused by human factors, and record the product of the first probability and the difference as the second probability.

Preferably, the method for judging whether there is a suspected abnormality in each opening duration according to the difference between the opening durations of two adjacent elevator doors from opening to closing in each floor and the difference threshold is: for any floor, compare the The difference between the opening time of two adjacent elevator doors from opening to closing and the difference threshold, if the difference is less than the difference threshold, it is determined that there is no suspected abnormality in the opening time of two adjacent elevator doors from opening to closing; if the difference If it is greater than or equal to the difference threshold, it is determined that there is a suspected abnormality in the opening time between two adjacent elevator doors opening and closing.

Preferably, the method of calculating the abnormality degree corresponding to the suspected abnormal data according to the second probability corresponding to the suspected abnormality in the opening time length is: calculating the average value of the second probability corresponding to the suspected abnormality in all the opening time lengths, and recording the average value is the degree of abnormality corresponding to suspected abnormal data.

Embodiments of the present invention have at least the following beneficial effects:

The present invention collects the opening time of each elevator door on each floor from opening to closing; through the difference between the opening time and the standard opening time, the first probability of abnormal opening time is calculated; different degrees of difference correspond to different first probabilities, The calculation of the first probability is only obtained by the difference between the opening time and the standard opening time, which is one-sided, because the abnormal opening time is not only caused by the fault of the elevator itself, but also may be caused by human factors, for example, when the elevator is running normally When there are too many people who need to take the elevator in a certain period of time, there is a great possibility that the opening time will be abnormal; since the abnormality includes early closing abnormality and delayed closing abnormality, the present invention calculates the early closing abnormality and delayed closing abnormality. The probability that the abnormality is caused by human factors is corrected to the first probability to obtain the second probability; the calculation of the second probability excludes the interference factors caused by human factors such as early closing abnormalities and delayed closing abnormalities, which in turn make the opening time abnormal The second probability can more truly reflect that the abnormality is caused by a failure of the elevator itself. At the same time, the present invention calculates the degree of abnormality of suspected abnormal data through the second probability; the storage time of suspected abnormal data is obtained according to the degree of abnormality; wherein the degree of abnormality can reflect the degree of importance of suspected abnormal data, and different degrees of abnormality represent different degrees of importance of suspected abnormal data The suspected abnormal data of different abnormal degrees are stored for different storage times, which realizes the self-adaptive adjustment of the storage time and improves the utilization rate of the storage space. Since the second probability can more truly reflect that the abnormality is caused by the fault of the elevator itself, the degree of abnormality calculated according to the second probability can more truly reflect that the degree of abnormality is caused by the fault of the elevator itself, thus making the calculation The storage time is more precise.

Description of drawings

In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description The drawings are only some embodiments of the present invention, and those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of the steps of an embodiment of a data management method for an elevator operating condition cloud platform based on big data in the present invention.

Detailed ways

In order to further explain the technical means and effects of the present invention to achieve the intended purpose of the invention, below in conjunction with the accompanying drawings and preferred embodiments, the solution proposed according to the present invention, its specific implementation, structure, features and effects are described in detail described as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, the particular features, structures or characteristics of one or more embodiments may be combined in any suitable manner.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field of the invention.

Referring to Fig. 1, it shows a kind of flow chart of the steps of the cloud platform data management method based on big data of elevator working condition that one embodiment of the present invention provides, and this method comprises the following steps:

Step 1, obtain the serialized data corresponding to the door system on each floor when the elevator is running on the same day; collect the opening time corresponding to each elevator door opening to closing on each floor; obtain the standard opening time, according to the standard opening time and opening time and standard The difference in the opening time is used to calculate the first probability of an abnormality in the opening time; the abnormality includes delayed closing abnormality and early closing abnormality.

Specifically, the working condition data corresponding to the door system on each floor when the elevator is running on the same day is collected. The specific type of working condition data is determined by the implementer according to the actual situation, and will not be described again. Then use the serialization method to convert the working condition data into serialized data to obtain the corresponding serialized data of the door system on each floor when the elevator is running on the same day; the serialization method is a known technology and will not be described in detail.

When data is stored or transmitted, it needs to be converted into a binary string that can be stored and transmitted. The binary string is serialized data. Therefore, the serialization method is used to convert the working condition data into serialized data for subsequent serialization. compressed data for storage.

The method to obtain the standard opening time is: set the number of days of historical collection, and for one of the days of historical collection, in the normal operation state of the elevator, divide the length of time from the start of the elevator to the stop of the elevator into at least two time periods, Collect the opening time corresponding to each elevator door opening to closing of the elevator on each floor, and use the floor as a unit to count the number of elevator door openings in each time period and the opening time corresponding to each elevator door opening to closing, and calculate each time The average value of the opening durations corresponding to the segments is recorded as the first average opening duration.

If the current day is a working day, then select the first average opening duration corresponding to each working day in each time period in the historical collection days, calculate the average value of the first average opening duration corresponding to all working days in each time period, and divide it into It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and recorded as the standard opening time corresponding to each new time period on a working day.

If the day is a rest day, then select the first average open duration corresponding to each rest day in each time period in the historical collection days, calculate the average value of the first average open duration corresponding to all rest days in each time period, and divide it It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and recorded as the standard opening time corresponding to each new time period on the rest day.

It should be noted that, as other implementation manners, the value of the standard opening duration can also be manually set directly according to the state of the elevator in normal operation, or the value of the standard opening duration can be obtained based on experience.

In this embodiment, the number of days of historical collection is set as one week, and the length of time corresponding to the time period is 30 minutes. The implementer can adjust the number of days of historical collection and the length of time corresponding to the time period.

Taking an elevator used by an enterprise as an example, the method of obtaining the standard opening time is explained in detail:

In the normal running state of the elevator, the length of time from the start of the elevator to the stop of the elevator in a day is divided into each time period,

，

，式中，

表示一天中第1 个30分钟对应的第一平均开启时长，

表示一天中第2个30分钟对应的第一平均开启时长，

表示一天中第n个30分钟对应的第一平均开启时长，n表示一天中电梯开始运行至停止运 行的时间长度被30分钟划分得到的时间段的数量。 According to the number of floors, take a certain floor as an example, count the number of times the elevator door is opened every 30 minutes in a day and the corresponding opening time of each elevator door from opening to closing. If the last elevator door is opened and then closed within 30 minutes If the time node is not within the time period corresponding to this 30 minutes, then the opening time corresponding to the opening and closing of the elevator doors this time is placed in the next 30 minutes, and the number of times the elevator doors are opened every 30 minutes and the time when the elevator doors are opened to each time are counted. Close the corresponding opening time, calculate the average of the corresponding opening time every 30 minutes, record it as the first average opening time, and obtain a set consisting of the first average opening time corresponding to every 30 minutes in a day

,

, where,

Indicates the first average opening time corresponding to the first 30 minutes of a day,

Indicates the first average opening time corresponding to the second 30 minutes of a day,

Indicates the first average opening time corresponding to the nth 30 minutes of a day, and n indicates the number of time periods obtained by dividing the time length from the start of the elevator to the stop of the elevator in a day by 30 minutes.

Due to the large difference in the working conditions of the enterprise using elevators on weekdays and rest days, the standard opening time corresponding to workdays and rest days is different. The standard opening time corresponding to the rest day is distinguished.

For the set A corresponding to each day in the historical collection days:

，

，式中，

表示工作日对应的第1个30分钟 的第二平均开启时长，

表示工作日对应的第2个30分钟的第二平均开启时长；

表示工作 日对应的第n个30分钟的第二平均开启时长；n表示一天中电梯开始运行至停止运行的时间 长度被30分钟划分得到的时间段的数量。 If the current day is a working day, select the set A corresponding to each working day, calculate the average value of the first average opening duration of the corresponding time period in the set, and record it as the second average opening duration corresponding to each time period , to get a set consisting of the second average turn-on duration

,

, where,

Indicates the second average opening time of the first 30 minutes corresponding to the working day,

Indicates the second average opening time of the second 30 minutes corresponding to the working day;

Indicates the second average opening time of the nth 30 minutes corresponding to the working day; n indicates the number of time periods obtained by dividing the time length from the start of the elevator to the stop of the elevator in a day by 30 minutes.

Then all the second average open durations are clustered to obtain a set number of categories, the present embodiment utilizes the K-means clustering algorithm to cluster, so that the clusters are 2 to obtain two different categories; K-means The clustering algorithm is a well-known technology and will not be described in detail; in the actual operation process, the implementer can also select other clustering algorithms for clustering; one of the categories corresponds to the peak period of elevator use in working days, such as commuting time, Therefore, the opening time corresponding to each elevator door opening to closing during the peak period is longer; the other category corresponds to the off-peak period, such as the going out and entering of some staff during the working period, so each time the elevator door is opened to closing during the off-peak period The corresponding opening time is shorter; based on this, calculate the average value of all second average opening time corresponding to the two categories, then the category corresponding to the larger average value is the peak period, and the category corresponding to the smaller average value is the off-peak period.

,

，式中，

表示工作日的第1个新时间段的标准开启时长；

表示工作日 的第2个新时间段的标准开启时长；

表示工作日的第m个新时间段的标准开启时长；m表 示工作日的新时间段的数量。 Finally, the time periods are merged to obtain each new time period, and then the standard opening duration is calculated; specifically, if the second average opening duration corresponding to two adjacent time periods is of the same category, the two adjacent time periods are merged into a new time period, calculate the average value of the second average opening time corresponding to each time period in the new time period, and record it as the standard opening time corresponding to each new time period of the working day; it should be noted that if there are two adjacent If the time periods are not merged, record the two time periods as new time periods respectively, and the second average opening duration corresponding to the two time periods is the standard opening time corresponding to the new time period. So far, each new time of the working day is obtained A collection of standard open durations for segments

,

, where,

Indicates the standard opening time of the first new time period on a working day;

Indicates the standard opening time of the second new time period of the working day;

Indicates the standard opening time of the mth new time period on a working day; m represents the number of new time periods on a working day.

In the same way, the standard opening time of each new time period of the rest day is obtained. The method for obtaining the standard opening duration of each new time period on a rest day is the same as the method for obtaining the standard opening duration of each new time period on a weekday, and will not be repeated here.

After the standard opening time is obtained, the first probability that the opening time is abnormal is calculated according to the difference between the opening time and the standard opening time.

In this embodiment, it is determined that all opening durations may be abnormal, so the first probability of each opening duration being abnormal is calculated.

If the current day is a working day, for any one floor, one of the opening durations corresponding to the elevator door opening to closing, obtain the new time period to which the elevator door opening time belongs, obtain the standard opening duration corresponding to the opening duration, and calculate the opening duration The absolute value of the difference between the length of time and the standard opening time, the difference between the opening time and the standard opening time is obtained, the ratio of the difference to the sum between the opening time and the standard opening time is calculated, and the ratio is recorded as the abnormality of the opening time. first probability.

The first probability is expressed by the formula:

表示电梯在任意一个楼层中第x次电梯门开启到关闭对应的开启时长的 第一概率，

表示电梯在该楼层第x次电梯门开启到关闭对应的开启时长，且第x次电梯门 开启的时间所属的新时间段为第j个新时间段；

表示第j个新时间段的标准开启时长；

，

，m表示工作日的新时间段的数量；

表示电梯在该楼层电梯 门开启的次数；

表示求取绝对值的函数。 in,

Indicates the first probability of the opening time corresponding to the opening and closing of the elevator door for the xth time on any floor,

Indicates the opening time corresponding to the opening and closing of the elevator door for the xth time on the floor, and the new time period to which the xth elevator door is opened is the jth new time period;

Indicates the standard opening time of the jth new time period;

,

, m represents the number of new time slots in weekdays;

Indicates the number of times the elevator door is opened on this floor;

Represents a function that finds the absolute value.

时，说明此开启时长小于标准开启时长，即表征此开启时长有可能出 现提前关闭异常，

表征该开启时长与标准开启时长之间的差异，差异越大，说明 此开启时长出现提前关闭异常的可能性越高，则对应的第一概率的取值越大；当

时，说明此开启时长大于标准开启时长，即表征此开启时长有可能出现延时关闭异常，

表征该开启时长与标准开启时长之间的差异，差异越大，说明此开启时长出现延 时关闭异常的可能性越高，则对应的第一概率的取值越大。 when

, it means that the opening time is shorter than the standard opening time, which means that the opening time may be abnormally closed in advance.

Represents the difference between the opening duration and the standard opening duration. The greater the difference, the higher the possibility of early closing abnormalities in this opening duration, and the greater the value of the corresponding first probability; when

, it means that the opening time is longer than the standard opening time, which means that the opening time may have a delayed closing abnormality.

Indicates the difference between the opening time and the standard opening time. The greater the difference, the higher the possibility of delayed closing abnormality in this opening time, and the greater the value of the corresponding first probability.

Similarly, if the day is a rest day, calculate the first probability that the opening time is abnormal according to the standard opening time and the difference between the opening time and the standard opening time, and obtain the first probability corresponding to the rest day and get the rest day of the working day The method for the corresponding first probability is the same, and will not be repeated here.

Step 2, set the waiting area, according to the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, the number of people who move out of the elevator and enter the elevator after the elevator door is opened, and the full number of people in the elevator, calculate The probability that the delay closing abnormality is caused by human factors; according to the probability that the delay closing abnormality is caused by human factors, calculate the probability that the early closing abnormalities are caused by human factors.

Since the first probability of an abnormality in the opening time of the elevator door in step 1 is obtained by comparing the opening time with the standard opening time obtained by analyzing the opening time of the elevator in normal operation in the past, the opening time of the elevator and the standard opening time The reasons for the big difference can be divided into human reasons, that is, there are many people who need to take the elevator, the elevator door is opened for a long time due to multiple activations of the elevator door opening button, or the opening time of the elevator is short due to the small number of people. Another reason is that the elevator itself is abnormal, that is, when there are many people who need to take the elevator, the elevator door closes early, resulting in a short opening time, or when there are few people, the elevator door closes delayed, resulting in a long opening time. Therefore, it is necessary to obtain the probability that the abnormality is caused by human factors when the opening time is abnormal.

In this embodiment, a waiting area is set, and the area of the waiting area is set by the implementer according to the actual situation, and will not be repeated here. For example, when the elevator can carry a large number of people, the area of the waiting area is larger; when the elevator can carry a small number of people, the area of the waiting area is small. According to the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, the number of people who move out of the elevator after the elevator door is opened and the number of people entering the elevator, and the full number of people in the elevator, it is calculated that the delayed closing abnormality is caused by human factors. The probability. Among them, the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, and the number of people who move out of the elevator and enter the elevator after the elevator door is opened are determined by using instance segmentation technology from the monitoring video data inside and outside the elevator on the same day Acquisition and instance segmentation techniques are well-known techniques and will not be repeated here.

The calculation method of the probability that the delayed closing abnormality is caused by human factors is as follows: calculate the ratio of the sum of the number of people moving out of the elevator and the number of people entering the elevator corresponding to the number of people entering the elevator after the elevator door is opened, and twice the full load of the elevator, and get the first Features: Calculate the ratio of the number of people entering the waiting area after the elevator door is opened to the sum of the number of people in the waiting area and the adjustment parameters when the elevator door is not opened, to obtain the second feature, and normalize the product of the first feature and the second feature The mapping is used to obtain the probability that the delay shutdown abnormality is caused by human factors.

The probability that the delayed shutdown abnormality is caused by human factors is expressed by the formula:

表示延时关闭异常为人为因素造成的概率，G表示电梯门未开启时等待区 中的人数，L表示电梯门开启后进入等待区中的人数，F表示电梯门开启后移出电梯的人数，

表示电梯门开启后进入电梯的人数，H表示电梯的满载人数；

表示调节参数，调节参数大 于0，本实施例中

，调节参数的加入是为了避免出现分母为0的情况；

表示第一特 征，

表示第二特征；

表示归一化处理函数。 In the formula,

Indicates the probability that the delayed closing abnormality is caused by human factors, G indicates the number of people in the waiting area when the elevator door is not opened, L indicates the number of people entering the waiting area after the elevator door is opened, F indicates the number of people who move out of the elevator after the elevator door is opened,

Indicates the number of people entering the elevator after the elevator door is opened, and H indicates the full number of people in the elevator;

Indicates the adjustment parameter, the adjustment parameter is greater than 0, in this embodiment

, the adjustment parameter is added to avoid the situation where the denominator is 0;

represents the first feature,

Indicates the second characteristic;

Indicates the normalization processing function.

表示电梯此次对应的开启时长内电梯内部的人员流动转换人数，

表示电 梯内部的最大人员流动转换人数，故第一特征

越大，表征人员进入电梯的耗时越长，延 时关闭异常为人为因素造成的概率越大；第二特征

表示后续进入等待区中的人数在电 梯门未开启时等待区中的人数（开始等待的人数）中的占比，当开始等待的人数较大时，后 续进入等待区中的人数对开始等待的人群进入电梯造成的影响较小，即对电梯的开启时长 的影响较小，而当开始等待的人数较小时，后续进入人数对开始等待的人群进入电梯造成 的影响较大，即对电梯的开启时长的影响较大，L值越大，需要的开启时长越长，因此

越 大，进电梯的耗时越长，延时关闭异常为人为因素造成的概率越大。基于此，计算延时关闭 异常为人为因素造成的概率R。

Indicates the number of personnel flow and conversion inside the elevator within the corresponding opening time of the elevator this time,

Indicates the maximum number of people flowing and converting inside the elevator, so the first feature

The larger the , the longer it takes for a person to enter the elevator, and the greater the probability that the delayed closing abnormality is caused by human factors; the second characteristic

Indicates the proportion of the number of people who subsequently enter the waiting area to the number of people in the waiting area when the elevator door is not opened (the number of people who start waiting). The impact caused by the crowd entering the elevator is small, that is, the impact on the opening time of the elevator is small, and when the number of people waiting at the beginning is small, the subsequent number of people entering the elevator has a greater impact on the crowd entering the elevator, that is, the opening time of the elevator is relatively small. The time length has a greater impact, the larger the L value, the longer the required opening time, so

The larger the value, the longer it takes to enter the elevator, and the greater the probability that the delayed closing abnormality is caused by human factors. Based on this, calculate the probability R that the delayed shutdown abnormality is caused by human factors.

，

表示提前关闭异常为人为因素造成的概率，

表示延时关闭 异常为人为因素造成的概率，

为自然常数。当延时关闭异常为人为因素造成的概率越大 时，则表明提前关闭异常为人为因素造成的概率越小，当延时关闭异常为人为因素造成的 概率越小时，则表明提前关闭异常为人为因素造成的概率越大，所以提前关闭异常为人为 因素造成的概率与延时关闭异常为人为因素造成的概率两者之间呈现负相关关系，基于 此，获取提前关闭异常为人为因素造成的概率。 According to the probability that the delayed closing abnormality is caused by human factors, calculate the probability that the early closing abnormalities are caused by human factors; specifically, an exponential function with the natural constant as the base and the probability of negative delayed closing abnormalities as the index The value of is taken as the probability that the early closing abnormality is caused by human factors; the probability that the early closing abnormality is caused by human factors is expressed as:

,

Indicates the probability that the early shutdown exception is caused by human factors,

Indicates the probability that the delayed shutdown exception is caused by human factors,

is a natural constant. When the probability that the delayed closing abnormality is caused by human factors is higher, it indicates that the probability that the early closing abnormality is caused by human factors is smaller; when the probability that the delayed closing abnormality is caused by human factors is smaller, it indicates that the early closing abnormality is caused by human factors The greater the probability caused by factors, so there is a negative correlation between the probability of early closing abnormalities caused by human factors and the probability of delayed closing abnormalities caused by human factors. Based on this, the probability of early closing abnormalities caused by human factors is obtained. .

In step 3, the first probability is corrected according to the probabilities caused by human factors corresponding to the delayed closing anomaly and the early closing anomaly to obtain the second probability.

The method for calculating the second probability is as follows: For any floor, one of the corresponding opening durations from opening to closing of the elevator door, judge the size of the opening duration and the corresponding standard opening duration, and when the opening duration is greater than or equal to the standard opening duration, calculate 1 The difference between the probability of the delayed closing exception and the probability caused by human factors, and the product of the first probability and the difference is recorded as the second probability; when the opening time is less than the standard opening time, calculate 1 and the early closing abnormality is a human factor The resulting probability difference, the product of the first probability and the difference is recorded as the second probability.

The method for obtaining the second probability when the current day is a rest day is consistent with the method for obtaining the second probability when the current day is a working day, and will not be repeated. This embodiment specifically expresses the calculation formula of the second probability when the current day is a working day. The formula for calculating the probability is:

表示电梯在任意一个楼层中第x次电梯门开启到关闭对应的开启时长的 第二概率；

表示电梯在任意一个楼层中第x次电梯门开启到关闭对应的开启时长的第一 概率，

表示电梯在该楼层第x次电梯门开启到关闭对应的开启时长，且第x次电梯门开启 的时间所属的新时间段为第j个新时间段；

表示第x次电梯门开启到关闭对应的开启时长 出现延时关闭异常时，延时关闭异常为人为因素造成的概率；

表示第x次电梯门开启到关 闭对应的开启时长出现提前关闭异常时，提前关闭异常为人为因素造成的概率；

表示第j 个新时间段的标准开启时长，

，

，m表示工作日的新时间段的数 量；

表示电梯在该楼层电梯门开启的次数。 In the formula,

Indicates the second probability of the opening time corresponding to the opening time of the elevator door opening to closing for the xth time on any floor;

Indicates the first probability of the opening time corresponding to the opening and closing of the elevator door for the xth time on any floor,

Indicates the opening time corresponding to the opening and closing of the elevator door for the xth time on the floor, and the new time period to which the xth elevator door is opened is the jth new time period;

Indicates the probability that the delayed closing abnormality is caused by human factors when the delayed closing abnormality occurs in the opening time corresponding to the xth elevator door opening to closing;

Indicates the probability that the early closing abnormality is caused by human factors when the opening time corresponding to the opening time of the xth elevator door to closing occurs early;

Indicates the standard opening time of the jth new time period,

,

, m represents the number of new time slots in weekdays;

Indicates the number of times the elevator door is opened on this floor.

时，说明此开启时长大于标准开启时长，即表征此开启时长有可能出 现延时关闭异常，

表示第x次电梯门开启到关闭对应的开启时长出现延时关闭异常时，延 时关闭异常为人为因素造成的概率，

表示第x次电梯门开启到关闭对应的开启时长出现 异常的第一概率，

的值越大，表征

的可信度越低，即说明该开启时长出现延时关闭异常 越有可能为人为因素造成的，延时关闭异常越不可能是由电梯本身出现故障导致的，因此， 以

表示第x次电梯门开启到关闭对应的开启时长的第二概率，排除了延时关闭异 常为人为因素造成的干扰因素，进而使得开启时长出现异常的第二概率能够更加真实的反 映出异常是由电梯本身出现故障导致的。 when

, it means that the opening time is longer than the standard opening time, which means that the opening time may have a delayed closing abnormality.

Indicates the probability that the delayed closing abnormality is caused by human factors when the delayed closing abnormality occurs in the opening time corresponding to the xth elevator door opening to closing,

Indicates the first probability that the opening time corresponding to the xth elevator door opening to closing is abnormal,

The larger the value, the representation

The lower the reliability of , that is to say, the more likely the delayed closing abnormality of the opening time is caused by human factors, and the less likely the delayed closing abnormality is caused by the failure of the elevator itself. Therefore, the

Indicates the second probability of the opening time corresponding to the opening time of the elevator door from opening to closing for the xth time, and eliminates the interference factor caused by human factors such as the delayed closing abnormality, so that the second probability of abnormal opening time can more truly reflect that the abnormality is Caused by failure of the elevator itself.

时，说明此开启时长小于标准开启时长，即表征此开启时长有可 能出现提前关闭异常；

表示第x次电梯门开启到关闭对应的开启时长出现提前关闭异常 时，提前关闭异常为人为因素造成的概率，

表示第x次电梯门开启到关闭对应的开启时长 出现异常的第一概率，

的值越大，表征

的可信度越低，即说明该开启时长出现提前关闭 异常越有可能为人为因素造成的，提前关闭异常越不可能是由电梯本身出现故障导致的， 因此，以

表示第x次电梯门开启到关闭对应的开启时长的第二概率，排除了提前 关闭异常为人为因素造成的干扰因素，进而使得开启时长出现异常的第二概率能够更加真 实的反映出该异常是由电梯本身出现故障导致的。 Similarly, when

, it means that the opening time is shorter than the standard opening time, which means that the opening time may be abnormally closed in advance;

Indicates the probability that the early closing abnormality is caused by human factors when the opening time corresponding to the opening time of the xth elevator door is opened to closing.

Indicates the first probability that the opening time corresponding to the xth elevator door opening to closing is abnormal,

The larger the value, the representation

The lower the reliability of , it means that the early closing abnormality of the opening time is more likely to be caused by human factors, and the earlier closing abnormality is less likely to be caused by the failure of the elevator itself. Therefore, the

Indicates the second probability of the opening time corresponding to the opening time of the elevator door from opening to closing for the xth time, excluding the interference factors caused by human factors caused by the early closing abnormality, so that the second probability of abnormal opening time can more truly reflect that the abnormality is Caused by failure of the elevator itself.

Step 4: According to the difference between the opening time of two adjacent elevator doors from opening to closing on each floor and the difference threshold, determine whether there is any suspected abnormality in each opening time, and then divide the serialized data corresponding to each floor into normal data and Suspected abnormal data.

This embodiment only describes the method for obtaining the corresponding difference threshold when the day is a working day. The method for obtaining the difference threshold corresponding to a rest day is the same as the method for obtaining the difference threshold corresponding to a working day, and will not be repeated here.

If the current day is a working day, the method to obtain the difference threshold is:

，将步骤1中非 高峰期对应的所有第二平均开启时长的均值记为

，

表示高峰期每次电梯门的开 启时长；

表示非高峰期每次电梯门的开启时长，则

为差异阈值。 Record the mean value of all second average opening durations corresponding to the peak period in step 1 as

, record the average of all the second average opening durations corresponding to the off-peak period in step 1 as

,

Indicates the opening time of each elevator door during the peak period;

Indicates the opening time of each elevator door during non-peak hours, then

is the difference threshold.

Then, according to the difference between the opening time of two adjacent elevator doors from opening to closing in each floor and the difference threshold, the method of judging whether there is a suspected abnormality in each opening time is as follows: for any floor, compare the two adjacent elevator doors on the floor The difference between the opening time of the elevator door from opening to closing and the difference threshold. If the difference is less than the difference threshold, it is determined that the opening time of two adjacent elevator doors from opening to closing is suspected to be abnormal; if the difference is greater than or equal to the difference threshold, Then it is determined that there is a suspected abnormality in the opening time between the opening and closing of the elevator door twice. Wherein, the difference between the opening durations of two adjacent elevator doors on the floor from opening to closing is the absolute value of the difference between the two opening durations.

In this embodiment, it is judged in chronological order whether there is a suspected abnormality in each opening duration, that is, first comparing the difference between the opening duration of the elevator door for the first time and the opening duration of the elevator door for the second time and the difference threshold, if the difference is less than difference threshold, it is determined that the opening duration of the two elevator doors is similar data, and it is determined that there is no suspected abnormality in the opening duration of the two elevator doors; if the difference is greater than or equal to the difference threshold, it is determined that the opening duration of the two elevator doors is not similar According to the data, it is determined that the opening time of the elevator doors for these two times is suspected to be abnormal.

It should be noted that, in order to further accurately obtain normal data and suspected abnormal data, when the difference between the opening duration of the first elevator door and the opening duration of the second elevator door is less than the difference threshold, the average of the two opening durations is calculated Represent new data, calculate the difference between the new data and the opening time of the third elevator door, compare the difference with the difference threshold, if the difference is less than the difference threshold, it is determined that the opening time of the third elevator door and the first elevator door The opening time of the door and the opening time of the second elevator door are similar data, and it is determined that there is no suspected abnormality in the opening time of the third elevator door; and so on, until the opening time of each elevator door in a floor is judged .

After the determination of the opening time of each elevator door in a floor is completed, the serialized data corresponding to the floor is divided into normal data and suspected abnormal data; the corresponding data in the serialized data that is suspected to be abnormal in the opening time is recorded as For suspected abnormal data, the corresponding data in the serialized data that does not appear suspected abnormal during the opening time is recorded as normal data. For normal data, normal data is less important. Therefore, in order to improve the compression rate, set a fixed storage time. Normal data is compressed by run-length encoding and stored for a fixed storage time. Run-length coding compression is a well-known technology, which is not within the protection scope of the present invention, and will not be repeated here. The value of the fixed storage time is 30 days, and the implementer can adjust it according to the actual situation.

Step 5, according to the second probability corresponding to the suspected abnormality in the opening time, calculate the degree of abnormality corresponding to the suspected abnormal data; calculate the storage time corresponding to the storage of the suspected abnormal data according to the degree of abnormality; compress the suspected abnormal data and store the corresponding storage time .

The method of calculating the degree of abnormality is as follows: For the suspected abnormal data corresponding to any floor; calculate the average value of the second probability corresponding to the suspected abnormality of all opening durations, and record the average value as the abnormal degree corresponding to the suspected abnormal data.

，

表示第1个楼层的疑似异常数据对应的异常程度，

表示第2个楼层的 疑似异常数据对应的异常程度，

表示第q个楼层的疑似异常数据对应的异常程度；获取所 有楼层的疑似异常数据对应的异常程度中的最大异常程度与最小异常程度；对于某一个楼 层的疑似异常数据对应的异常程度，计算该异常程度与最小异常程度的差值将其记为第一 差值，计算最大异常程度与最小异常程度的差值将其记为第二差值，将第一差值与第二差 值的比值记为特征因子，计算特征因子与第一固定储存时间的乘积，将所述乘积与固定储 存时间的和作为该楼层的疑似异常数据储存时对应的储存时间。 Then calculate the storage time corresponding to the suspected abnormal data storage according to the abnormality degree. The calculation method of the storage time is: set the first fixed storage time, and the abnormality degrees corresponding to the suspected abnormal data of all floors form a set

,

Indicates the degree of abnormality corresponding to the suspected abnormal data on the first floor,

Indicates the degree of abnormality corresponding to the suspected abnormal data on the second floor,

Indicates the degree of abnormality corresponding to the suspected abnormal data of the qth floor; obtains the maximum and minimum degree of abnormality in the degree of abnormality corresponding to the suspected abnormal data of all floors; for the degree of abnormality corresponding to the suspected abnormal data of a certain floor, calculate the The difference between the degree of abnormality and the minimum degree of abnormality is recorded as the first difference, the difference between the maximum degree of abnormality and the minimum degree of abnormality is calculated as the second difference, and the ratio of the first difference to the second difference Denote it as a characteristic factor, calculate the product of the characteristic factor and the first fixed storage time, and use the sum of the product and the fixed storage time as the corresponding storage time when storing the suspected abnormal data of this floor.

The formula for storage time is expressed as:

表示第t个楼层的疑似异常数据对应的储存时间，

表示固定储存时间， 本实施例中，

；

表示第t个楼层的疑似异常数据对应的异常程度，

表示最小异 常程度，

表示最大异常程度，

表示第一固定储存时间，本实施例中，

，实施者 可对第一固定储存时间的取值进行调整，保证第一固定储存时间的取值大于固定储存时间 的取值；

表示第一差值；

表示第二差值；

表示特征因子，特征 因子表征的是对该异常程度与最小异常程度的差值进行归一化处理。式中，

， q表示楼层的数量。 in,

Indicates the storage time corresponding to the suspected abnormal data on the tth floor,

Indicates a fixed storage time, in this embodiment,

;

Indicates the degree of abnormality corresponding to the suspected abnormal data on the tth floor,

Indicates the minimum degree of abnormality,

Indicates the maximum degree of abnormality,

Indicates the first fixed storage time, in this embodiment,

, the implementer can adjust the value of the first fixed storage time to ensure that the value of the first fixed storage time is greater than the value of the fixed storage time;

represents the first difference;

Indicates the second difference;

Represents the eigenfactor, and the eigenfactor characterizes the normalization process of the difference between the abnormal degree and the minimum abnormal degree. In the formula,

, q represents the number of floors.

表征的是该第t个楼层的疑似异常数据对应的重要程度，异常程度越 高，表征重要程度越高，疑似异常数据对后续的电梯安全检查越重要，则需要疑似异常数据 对应的储存时间应当越长；反之，异常程度越低，表征重要程度越低，疑似异常数据对后续 的电梯安全检查越不重要，则需要疑似异常数据对应的储存时间应当越短。因此，根据异常 程度计算疑似异常数据对应的储存时间；不同的异常程度对应的疑似异常数据的储存时间 不同，通过异常程度计算疑似异常数据的储存时间，实现了对疑似异常数据的储存时间的 自适应调整，能够提高电梯运行数据存储空间的利用率，完成电梯工况云平台数据的高效 存储。 Abnormal degree

It represents the degree of importance corresponding to the suspected abnormal data on the tth floor. The higher the degree of abnormality, the higher the importance of the representation. The more important the suspected abnormal data is to the subsequent elevator safety inspection, the storage time corresponding to the suspected abnormal data should be Conversely, the lower the degree of abnormality, the lower the importance of the representation, and the less important the suspected abnormal data is to the subsequent elevator safety inspection, the shorter the storage time corresponding to the suspected abnormal data should be. Therefore, the storage time corresponding to the suspected abnormal data is calculated according to the degree of abnormality; the storage time of the suspected abnormal data corresponding to different abnormal degrees is different, and the storage time of the suspected abnormal data is calculated by the degree of abnormality, which realizes the automatic storage time of the suspected abnormal data. Adaptation and adjustment can improve the utilization rate of elevator operation data storage space and complete the efficient storage of elevator operating condition cloud platform data.

，

，

表征第1个楼层的疑似异常数据对应的储存时 间，

表示第2个楼层的疑似异常数据对应的储存时间，

表示第q个楼层的疑似异常数据 对应的储存时间。 Then obtain the set of storage time of suspected abnormal data in the serialized data corresponding to all floors of the elevator door system on that day

,

,

Represents the storage time corresponding to the suspected abnormal data on the first floor,

Indicates the storage time corresponding to the suspected abnormal data on the second floor,

Indicates the storage time corresponding to the suspected abnormal data on the qth floor.

After the storage time of the suspected abnormal data corresponding to each floor is obtained, the suspected abnormal data is compressed and stored corresponding to the storage time, that is, the suspected abnormal data is compressed by run-length encoding and stored corresponding to the storage time. Run-length coding compression is a well-known technology, which is not within the protection scope of the present invention, and will not be repeated here.

Since elevator faults can be divided into door system faults, operating system faults and command system faults, the above method only divides the serialized data corresponding to the door system into normal data and suspected abnormal data, and stores different types of data for different storage times; Therefore, the present invention also includes obtaining serialized data corresponding to the operating system and the command system, and the faults of the operating system and the command system are only affected by the elevator itself.

For the running system, the running speed of the elevator is a fixed value, and there are only two states of normal speed and abnormal speed. Therefore, the serialized data corresponding to the running system can be divided into normal data and abnormal data according to the actual running speed of the elevator. When the actual running speed is a fixed value, the corresponding data in the serialized data is recorded as normal data, when the actual running speed of the elevator is not a fixed value, the corresponding data in the serialized data is recorded as abnormal data, and the normal data is run-length encoded compress and store for a fixed storage time, perform run-length coding compression on the abnormal data and store for a first fixed storage time.

For the instruction system, the instruction only has two states of response and non-response, so the serialized data corresponding to the instruction system is divided into normal data and abnormal data, that is, the corresponding data in the serialized data when the instruction responds is recorded as normal data, and the When the instruction does not correspond, the corresponding data in the serialized data is recorded as abnormal data, the normal data is compressed by run-length coding and stored for a fixed storage time, and the abnormal data is compressed by run-length coding and stored for a first fixed storage time.

The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still implement the foregoing embodiments Modifications to the technical solutions described in the examples, or equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of each embodiment of the application, and should be included in the scope of the technical solutions of the embodiments of the application. within the scope of protection.

Claims (9)

1. a kind of elevator operation condition cloud platform data management method based on big data, it is characterized in that, the method comprises the following steps: Obtain the serialized data corresponding to the door system on each floor when the elevator is running on the same day; collect the opening time corresponding to each elevator door opening to closing on each floor; obtain the standard opening time, and calculate the opening time according to the difference between the opening time and the standard opening time The first probability of an abnormality in the duration; the abnormality includes a delayed closing exception and an early closing exception; Set the waiting area, and calculate the delay closing time based on the number of people in the waiting area when the elevator door is not opened, the number of people entering the waiting area after the elevator door is opened, the number of people who move out of the elevator after the elevator door is opened and enter the elevator, and the number of people who are fully loaded in the elevator. The probability that the abnormality is caused by human factors; according to the probability that the delayed closing abnormality is caused by human factors, calculate the probability that the early closing abnormalities are caused by human factors; The first probability is corrected according to the probability that the delayed closing abnormality and the early closing abnormality are caused by human factors, and the second probability is obtained; According to the difference between the opening time of two adjacent elevator doors from opening to closing on each floor and the difference threshold, determine whether there is suspected abnormality in each opening time, and then divide the serialized data corresponding to each floor into normal data and suspected abnormal data ; According to the second probability corresponding to the suspected abnormality in the opening time, calculate the abnormality degree corresponding to the suspected abnormal data; calculate the storage time corresponding to the suspected abnormal data storage according to the abnormality degree; compress the suspected abnormal data and store the corresponding storage time. 2. a kind of elevator operating condition cloud platform data management method based on big data according to claim 1, is characterized in that, the method for described acquisition standard opening duration is: setting history collection days number, for the history collection days number On one day, when the elevator is in normal operation, the length of time from the start of the elevator to the stop of the elevator in one day is divided into at least two time periods, and the opening time corresponding to the opening and closing of the elevator door on each floor is collected. As a unit, count the number of times the elevator door is opened in each time period and the opening time corresponding to each elevator door from opening to closing, calculate the average value of the opening time corresponding to each time period, and record it as the first average opening time; If the current day is a working day, then select the first average opening duration corresponding to each working day in each time period in the historical collection days, calculate the average value of the first average opening duration corresponding to all working days in each time period, and divide it into It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and it is recorded as the standard opening time corresponding to each new time period of the working day; If the day is a rest day, then select the first average open duration corresponding to each rest day in each time period in the historical collection days, calculate the average value of the first average open duration corresponding to all rest days in each time period, and divide it It is recorded as the second average opening duration corresponding to each time period; all the second average opening durations are clustered to obtain a set number of categories, if the second average opening duration corresponding to two adjacent time periods is the same category, then Two adjacent time periods are merged into a new time period, and the average value of the second average opening time corresponding to each time period in the new time period is calculated, and recorded as the standard opening time corresponding to each new time period on the rest day. 3. A kind of cloud platform data management method based on big data of elevator working condition according to claim 1, it is characterized in that, the method for calculating the corresponding storage time when the suspected abnormal data is stored according to the degree of abnormality is: set a fixed The storage time and the first fixed storage time are used to obtain the maximum abnormality and the minimum abnormality among the abnormalities corresponding to the suspected abnormal data of all floors; for the abnormality corresponding to the suspected abnormal data of a certain floor, calculate the abnormality and the minimum abnormality The difference of the degree is recorded as the first difference, the difference between the maximum abnormal degree and the minimum abnormal degree is calculated as the second difference, and the ratio of the first difference to the second difference is recorded as the characteristic factor, The product of the feature factor and the first fixed storage time is calculated, and the sum of the product and the fixed storage time is used as the corresponding storage time when the suspected abnormal data of the floor is stored. 4. A kind of cloud platform data management method based on big data of elevator working condition according to claim 1, it is characterized in that, according to the number of people in the waiting area when the elevator door is not opened, enter the waiting area after the elevator door is opened The number of people, the number of people who move out of the elevator and the number of people entering the elevator after the elevator door is opened, and the number of people who are fully loaded in the elevator, and calculate the probability that the delayed closing abnormality is caused by human factors, including: calculating the corresponding number of people who have moved out of the elevator after the elevator door is opened and the number of people entering the elevator The ratio between the sum of the number of people and twice the full number of people in the elevator is obtained to obtain the first characteristic; calculate the sum of the number of people entering the waiting area after the elevator door is opened and the number of people in the waiting area when the elevator door is not opened and the adjustment parameters Ratio to obtain the second feature, and perform normalized mapping on the product of the first feature and the second feature to obtain the probability that the delayed shutdown abnormality is caused by human factors; wherein, the adjustment parameter is greater than 0. 5. according to claim 1 or 4, a kind of cloud platform data management method based on big data of elevator operating conditions, is characterized in that, according to the probability that the delayed closing abnormality is caused by human factors, it is calculated that the early closing abnormality is caused by human factors The method of the probability is as follows: take the natural constant as the base number, and take the value of the exponential function whose index is the probability of the negative delayed closing abnormality as the human factor as the probability of the early closing abnormality as the human factor. 6. A kind of cloud platform data management method based on big data of elevator operating conditions according to claim 2, is characterized in that, according to the difference of opening duration and standard opening duration, calculates the abnormal first probability of opening duration, comprising: For the opening time corresponding to one elevator door opening to closing on any floor, obtain the new time period of the opening time of the elevator door, obtain the standard opening time corresponding to the opening time, and calculate the difference between the opening time and the standard opening time The absolute value of the value is obtained to obtain the difference between the opening time and the standard opening time, and the ratio between the difference and the sum between the opening time and the standard opening is calculated, and the ratio is recorded as the first probability that the opening time is abnormal. 7. according to claim 1 or 2, a kind of cloud platform data management method based on big data of elevator working condition, it is characterized in that, according to the probability that delay closing abnormality and closing abnormality in advance are caused by human factors to the first The probability is corrected, and the method to obtain the second probability is as follows: For any one floor, one of the elevator doors is opened to close corresponding to the opening time length, judge the size of the opening time length and the corresponding standard opening time length, when the opening time length is greater than or equal to the standard opening time length , calculate the difference between 1 and the probability that the delayed closing abnormality is caused by human factors, and record the product of the first probability and the difference as the second probability; when the opening time is less than the standard opening time, calculate 1 and close in advance The abnormality is the difference in probability caused by human factors, and the product of the first probability and the difference is recorded as the second probability. 8. A kind of cloud platform data management method based on big data of elevator working condition according to claim 1, it is characterized in that, according to the difference between the opening duration of two adjacent elevator doors in each floor from opening to closing and the difference threshold, the method for judging whether there is a suspected abnormality in each opening time is: for any floor, compare the difference between the opening time of two adjacent elevator doors on the floor from opening to closing with the difference threshold, if the difference If it is less than the difference threshold, it is determined that there is no suspected abnormality in the opening time of two adjacent elevator doors from opening to closing; if the difference is greater than or equal to the difference threshold, it is determined that there is a suspected abnormality in the opening time of two adjacent elevator doors from opening to closing. 9. A big data-based cloud platform data management method for elevator operating conditions according to claim 1, characterized in that, according to the second probability corresponding to the suspected abnormality in the opening duration, the abnormality degree corresponding to the suspected abnormal data is calculated The method is: calculate the average value of the second probability corresponding to the suspected abnormality of all the opening durations, and record the average value as the degree of abnormality corresponding to the suspected abnormal data.

Images