Passenger retention detection method

Abstract

The invention discloses a passenger retention detection method. The method comprises the following steps: S1, carrying out weighing self-learning; s2, the elevator continuously runs for multiple times in a no-load state, the elevator load is calculated through the deformation of an elevator damping device, a data set is obtained, the control panel controls the load data to be ranked from small to large, and the median of the load data serves as a reference number A; s3, when the elevator carries passengers, the real-time load B is measured through deformation of the damping device and compared with the reference number A, when the three conditions that the real-time load B is larger than or equal to 105% A, the elevator is in a non-operation state and the duration time is larger than or equal to 120 S are met at the same time, a passenger retention alarm is triggered, otherwise, the elevator continues to operate continuously, load data in a no-load state are judged, and the passenger retention alarm is given out. Updating the reference number A in real time; s4, when a passenger retention alarm is triggered; and S5, after detecting that the person leaves, resetting the passenger retention alarm. The detection effect can be improved, and missing report of passenger retention information is avoided.

Description

Technical Field

[0001] The present invention relates to a detection method, in particular to a method for detecting passenger detention. Background Art

[0002] With the continuous development of social economy and technology, household elevators have gradually emerged. Townhouses, villas, self-built houses, etc. are all occasions where household elevators are used. Household elevators solve the inconvenience of going up and down stairs and carrying items at home, especially bringing great convenience to the elderly with mobility difficulties. At the same time, the health of family members is the focus of every family. When family members (especially the elderly or those with sudden illnesses who are alone) take the elevator at home and suddenly get ill (such as myocardial infarction) or fall, other family members cannot obtain the dangerous situation in time, resulting in a delay in the treatment time.

[0003] Currently, the main solutions for detecting passenger detention are to identify human body information through infrared recognition cameras, infrared human body sensors, etc. These methods not only have high costs but are also easily interfered. Or, as disclosed in the patent with the application number 202110194912.7, a device for automatically warning of over-time detention of personnel in an elevator. When a person is detained in the elevator for over time, there is gravity of the person in the elevator. At this time, the load-bearing plate bears the gravity, and the load-bearing plate transmits the gravity to the lower rope head plate, and the lower rope head plate transmits the gravity to the weighing sensor. The weighing sensor senses the gravity and generates a reaction. At this time, the weighing sensor transmits a warning signal to the elevator alarm through the data line. However, there are differences in the forces received on different floors during the operation of the elevator, and for the structure of the sleeve plus spring, after a long time of elastic pressure, the spring has fatigue (resulting in permanent deformation), and it cannot be compared with the initial value after a long time; there are limitations, resulting in the inability to accurately detect passenger detention information. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for detecting passenger detention. It can improve the detection effect and avoid missing the detection of passenger detention information.

[0005] The technical solution of the present invention: A method for detecting passenger detention includes the following steps:

[0006] S1. Conduct weighing self-learning. When at the position origin, detect the deformation amount of the elevator shock absorber when the elevator is unloaded and when there are different loads in the elevator, and obtain the linear relationship between the deformation amount of the elevator shock absorber and the load.

[0007] S2. The elevator runs continuously multiple times in an unloaded state. Calculate the elevator load through the deformation amount of the elevator shock absorber to obtain a data set. Remove the unreasonable data in the data set, and control the load data by the control panel to be sorted from small to large, and take the median of the load data as the reference number A.

[0008] S3. When the elevator is carrying passengers, measure the real-time load B through the deformation of the shock absorption device, and compare it with the reference number A. When the three conditions of real-time load B ≥ 105% A, the elevator is in a non-operating state, and the duration ≥ 120S are simultaneously met, trigger the passenger retention alarm. Otherwise, the elevator continues to operate continuously, judge the load data in the no-load state, and update the reference number A in real time;

[0009] S4. When the passenger retention alarm is triggered, transmit this event to the Internet of Things cloud platform through the Internet of Things transmission device, and the Internet of Things cloud platform pushes this event to the mobile phone APP or mini-program of family members;

[0010] S5. After detecting that the personnel have left, reset the passenger retention alarm.

[0011] In the foregoing method for detecting passenger retention, the specific steps of S1 are as follows:

[0012] At the position origin, record the no-load weight as Q, the first load weight as Q1, the second load weight as Q2, and combine the deformation of the corresponding elevator shock absorption device to calculate the linear relationship between the load weight and the elevator shock absorption device.

[0013] In the foregoing method for detecting passenger retention, the number of no-load state operations in S2 can be 5 times. Calculate the corresponding load weights, which are respectively recorded as A1, A2, A3, A4, and A5. Remove the unreasonable data, and control the load data to be sorted from small to large by the control panel, and take the median.

[0014] In the foregoing method for detecting passenger retention, the specific method for removing unreasonable data is as follows:

[0015] C1. Calculate the mean value and the deviation s;

[0016] C2. Determine the suspicious value;

[0017] C3. Calculate the statistic G and compare it with the critical value;

[0018] C4. Judge and remove the unreasonable data.

[0019] In the foregoing method for detecting passenger retention, in C1 The calculation formulas for the mean value and the deviation s are as follows:

[0020]

[0021] In the foregoing method for detecting passenger retention, determining the suspicious value means finding the data point Ad with the largest absolute value of the residual. The specific calculation formula is as follows:

[0022]

[0023] In the foregoing method for detecting passenger retention, the specific calculation formula of C3 is as follows:

[0024]

[0025] And according to the confidence level (such as α = 0.05 or 0.01) and the sample size n, look up the Grubbs critical value table G(α, n).

[0026] In the foregoing method for detecting passenger retention, the specific method of judgment and rejection is that if G ≥ G(α, n), it is determined that Ad is an outlier and rejected; otherwise, it is retained.

[0027] In the foregoing method for detecting passenger retention, the specific steps for real-time updating of the reference number A in S3 are as follows: During the operation of the elevator, record the load data when the elevator is in the no-load state into the data set, and delete the earliest load data in the data set, remove the unreasonable data in the data set, and sort the load data from small to large by the control panel. Take the median of the load data as the reference number A. If the collected data is greater than / less than the reference number for 3 consecutive times, these 3 data are valid; otherwise, automatically filter the data in the data set, re-perform the weighing self-learning, and reset the reference data A.

[0028] Compared with the prior art, the present invention has the following advantages:

[0029] 1. By performing weighing self-learning (i.e., through the existing elevator shock absorption device and detecting its deformation amount under different loads), the relationship between the deformation amount of the elevator shock absorption device and the elevator load can be studied, and whether there are people staying in the elevator car can be judged by the deformation amount of the elevator detection device, without adding infrared sensors and cameras in the car, thereby reducing the detection cost;

[0030] 2. Through multiple runs of the elevator in the no-load mode, the no-load load of the elevator is detected and calculated, and after removing unreasonable data, the median of multiple groups of load data is used as the reference number A, so that the comparison between the real-time load B and the reference number A during the passenger-carrying operation of the elevator is more accurate;

[0031] 3. By real-time updating the reference number A with the data during the overload operation of the elevator, the detection reliability is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flowchart of the present invention;

[0033] Figure 2 is a linear relationship diagram of the load weight and the elevator shock absorption device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, but it shall not be used as a basis for limiting the present invention.

[0035] Embodiment. A method for detecting passenger retention is configured as Figure 1-2 shown, and includes the following steps:

[0036] S1. Perform weighing self-learning. When at the position origin, detect the deformation amount of the elevator shock absorption device when the elevator is unloaded and when there are different loads in the elevator, and obtain the linear relationship between the deformation amount of the elevator shock absorption device and the load. The specific steps of S1 are as follows: when at the position origin, record the unloaded weight as Q, the first loaded weight as Q1, and the second loaded weight as Q2, and combine the corresponding deformation amounts of the elevator shock absorption device to calculate the linear relationship between the loaded weight and the elevator shock absorption device.

[0037] S2. The elevator runs continuously in the unloaded state for multiple times. Calculate the elevator load through the deformation amount of the elevator shock absorption device, remove unreasonable data, and control the load data by the control panel to be sorted from small to large. Take the median of the load data as the reference number A. In step S2, first, through multiple unloaded runs, detect and calculate the load during the unloaded operation of the elevator. For example, the number of unloaded runs in this application can be 5 times, and the measured loads are recorded as A1, A2, A3, A4, and A5 respectively. Then the data set consists of A1, A2, A3, A4, and A5. Remove unreasonable data, and control the load data by the control panel to be sorted from small to large, and take the median. Since there are significant differences in the unloaded load at different floors during the elevator operation, especially when the elevator runs from bottom to top, even if the elevator is in the unloaded state, the force on the elevator shock absorption device gradually increases. By taking the median, this interference can be reduced, making the detection result more accurate. And before taking the median value, unreasonable data will be removed to reduce the influence of incorrect data on the reference number A. The specific steps for real-time updating the reference number A in S3 are as follows: during the elevator operation process and when the elevator is in the unloaded state, record the load data in the data set, delete the earliest load data in the data set, remove unreasonable data in the data set, and control the load data by the control panel to be sorted from small to large. Take the median of the new load data as the reference number A. If the collected data is greater than / less than the reference number for 3 consecutive times, then these 3 data are valid; otherwise, automatically filter the data in the data set, re-perform weighing self-learning, and reset the reference data A.

[0038] The specific method for removing unreasonable data in this application is as follows:

[0039] C1. Calculate the mean and the deviation s; in C1 The calculation formulas for the mean and the deviation s are:

[0040]

[0041] C2. Determine the suspect value; determining the suspect value means finding the data point Ad with the largest absolute residual value. The specific calculation formula is as follows:

[0042]

[0043] C3. Calculate the statistic G and compare it with the critical value; the specific calculation formula for C3 is as follows:

[0044]

[0045] And according to the confidence level (such as α = 0.05 or 0.01) and the sample size n, look up the Grubbs critical value table G(α,n).

[0046] C4. Judge and eliminate the unreasonable data. The specific method of judgment and elimination is that if G ≥ G(α,n), determine Ad as an outlier and eliminate it; otherwise, keep it.

[0047] S3. When the elevator is carrying passengers, measure the real-time load B through the deformation of the shock absorption device and compare it with the reference number A. When the three conditions that the real-time load B ≥ 105% A, the elevator is in a non-operating state, and the duration ≥ 120S are simultaneously satisfied, trigger the passenger detention alarm. Otherwise, the elevator continues to run continuously, judge the load data in the no-load state, and update the reference number A in real time;

[0048] S4. When the passenger detention alarm is triggered, transmit this event to the Internet of Things cloud platform through the Internet of Things transmission device, and the Internet of Things cloud platform pushes this event to the mobile phone APP or small program of family members;

[0049] S5. After detecting that the person has left, reset the passenger detention alarm.

[0050] The working principle of this application includes a pre - operation test phase and a detection and real - time update phase during use. The test phase includes calculating and obtaining the linear relationship between the deformation amount of the elevator shock - absorbing device and the load, and obtaining the load data in the no - load state of the elevator. Through multiple runs in the no - load state and removing unreasonable data, after sorting the remaining data from smallest to largest, the median is taken as the reference number A, which can reduce the influence of floors on the elevator operation. And when the elevator is running, the reference number A can be updated in real - time. That is, during the use of the elevator, there will be no - load operation situations. At this time, the subsequent elevator loads in the no - load state are recorded as A6, A7, A8... Ai in sequence, and unreasonable data are removed. For example, at the beginning, the reference number is taken as the median of A1, A2, A3, A4, A5, and after operation, the reference number is taken as the median of A2, A3, A4, A5, A6, and so on. This can reduce the interference caused by the permanent deformation of the shock - absorbing device during use. Further considering the influence of the environment on the rubber shock - absorbing pad in the shock - absorbing device, if the collected data is greater than / less than the reference number for 3 consecutive times, then these 3 data are valid; otherwise, the earliest data is automatically filtered out, and the weight self - learning is restarted, and the reference number A is reset, with better reliability.

Claims

1. A method for detecting stranded passengers, characterized in that: The following steps are involved: S1. Perform weighing self-learning. At the position origin, the deformation of the elevator shock absorbing device is detected when the elevator is unloaded and when different loads are installed in the elevator, and the linear relationship between the deformation of the elevator shock absorbing device and the load is obtained; S2. The elevator runs continuously for multiple times in an empty state, and the elevator load is calculated by the deformation of the elevator shock absorbing device to obtain a data set. The unreasonable data in the data set is removed, and the control panel controls the order of the load data from small to large, and the median of the load data is used as the benchmark number A; S3. When the elevator is carrying passengers, the real-time load B is measured by the deformation of the shock absorbing device and compared with the reference number A. When the three conditions of real-time load B ≥ 105% A and the elevator is in a non-operating state and the duration is ≥ 120S are met at the same time, the passenger detention alarm is triggered. Otherwise, the elevator continues to operate, the load data in the no-load state is judged, and the reference number A is updated in real time; S4. When a passenger stranded alarm is triggered, the event is transmitted to the IoT cloud platform through the IoT transmission device, and the IoT cloud platform pushes the event to the mobile phone APP or mini program of family members; S5. After detecting that a person has left, the passenger retention alarm is reset.

2. A passenger detention detection method according to claim 1, characterized in that: The specific steps of S1 are: At the origin of the position, the no-load weight is recorded as Q, the first loading weight is recorded as Q1, and the second loading weight is recorded as Q2. Combined with the deformation of the corresponding elevator shock absorbing device, the linear relationship between the loading weight and the elevator shock absorbing device is calculated.

3. A passenger detention detection method according to claim 1, characterized in that: The number of no-load state operations in S2 can be 5 times, and the corresponding loads are calculated and recorded as A1, A2, A3, A4, and A5 respectively. The unreasonable data is removed, and the control panel controls the load data to be sorted from small to large, and the median is taken.

4. A passenger detention detection method according to claim 3, characterized in that: The specific method for removing unreasonable data is: C1, calculate the mean and deviation s; C2, determine suspicious values; C3, calculate the statistic G and compare it with the critical value; C4, judge and eliminate unreasonable data.

5. A passenger detention detection method according to claim 4, characterized in that: The C1 The calculation formula of the deviation s is:

6. A passenger detention detection method according to claim 4, characterized in that: The determination of the suspicious value is to find the data point Ad with the largest absolute value of the residual. The specific calculation formula is:

7. A passenger detention detection method according to claim 4, characterized in that: The specific calculation formula of C3 is: And according to the confidence level (such as α = 0.05 or 0.01) and the sample size n, check the Grubbs critical value table G (α, n).

8. A passenger detention detection method according to claim 4, characterized in that: The specific method of judgment and elimination is: if G≥G(α,n), Ad is judged as an abnormal value and eliminated; otherwise, it is retained.

9. A passenger detention detection method according to claim 1, characterized in that: The specific steps of updating the reference number A in real time in S3 are as follows: recording the load data of the elevator in the process of operation and when the elevator is in an unloaded state into the data set, deleting the earliest load data in the data set, removing unreasonable data in the data set, and controlling the load data to be sorted from small to large by the control panel, and taking the median of the load data as the reference number A. If the collected data is greater than / less than the reference number for three consecutive times, the three data are valid, otherwise the data set data is automatically filtered out, the weighing self-learning is performed again, and the reference data A is reset.

Images