Method of controlling exit congestion of passenger conveyor

Abstract

This invention discloses a method for controlling congestion at the exit of a passenger transport device. Pressure sensors are embedded in the side linings on the left and right sides of the exit of the passenger transport device. The pressure signals of the pressure sensors are detected, and the pressure values ​​are calculated based on the pressure signals from the left and right sides. When the pressure value exceeds a preset threshold, an exit congestion warning signal for the passenger transport device is output, and the passenger transport device implements control and management based on the pressure value.

Description

Technical Field

[0001] This invention relates to the field of passenger transport device technology, and more specifically to a method for controlling congestion at the exit of a passenger transport device. Background Technology

[0002] Passenger transport devices such as escalators or moving walkways have inherent risks as continuous transport equipment: the steps continuously transport passengers to the exit, and once the exit is congested, it can cause crushing and stampede accidents within seconds, which is extremely risky in large passenger flow scenarios; continuous operation at the rated speed will exacerbate congestion, and safety interventions such as deceleration and stopping the escalator are required.

[0003] To monitor congestion at passenger transport exits, Reference 1 (CN208200104U) discloses a recognition scheme based on machine vision and deep learning. This scheme can only identify abnormal states such as falls, lacking crowd detection and pressure perception capabilities. The core problems are twofold: First, strong environmental interference; cameras are prone to severe obstruction when crowds are dense, and factors such as changes in station lighting and large luggage can easily lead to missed detections or false alarms. Second, it lacks mechanical sensing characteristics; as a non-contact monitoring method, it can only judge congestion from a spatial geometric perspective and cannot directly detect the lateral physical pressure intensity experienced by passengers at the exit. In the initial stages of actual stampedes, the sudden increase in physical pressure often precedes visually identifiable deformation features, and current technology cannot quantitatively analyze these key mechanical indicators. Reference 1's scheme recognizes falls and congestion only after the fact, resulting in a severely delayed response. The abnormal state judgment of this scheme requires the fulfillment of the established condition of "a falling posture appearing in the image / dense stacking of people," meaning that image comparison and escalator stopping can only be completed and the escalator can only be triggered after a fall has occurred and a crowding situation has formed. However, the entire process from the gathering of people at the escalator exit to the crowding and pushing, and then to the fall, takes only a few seconds. The delayed detection means that the escalator stopping operation can only play a role in mitigating losses after the fact, and cannot achieve early prevention of accidents. The scheme in Reference 1 has obvious privacy protection defects due to its technical implementation method. When applied in public areas such as shopping malls and subways, it poses serious legal compliance risks of privacy leakage. This type of scheme continuously collects high-definition images of the escalator area through cameras, which can capture sensitive privacy information such as passengers' facial features and personal belongings. The use of such technology in public areas must strictly comply with the requirements of relevant laws and regulations such as the Personal Information Protection Law, which not only significantly increases the compliance costs of technology application, but also brings high legal risks of privacy leakage.

[0004] References 2 (CN108805093A) and 3 (US20180118522A1) disclose a gravity-sensing-based crowd estimation scheme. This type of scheme estimates the number of pedestrians using gravity sensors on escalator covers to determine congestion, but it has significant technical flaws. First, it lacks the ability to detect squeezing pressure, relying solely on load to indirectly calculate passenger numbers and infer congestion, which is completely disconnected from actual squeezing pressure. The cover load only reflects the total weight of static people at the exit, failing to reflect the lateral squeezing and pushing forces generated when crowds are congested—forces that are the core cause of falls and stampedes at escalator exits. Furthermore, the passenger count calculation relies on a fixed value of "average individual weight," and the significant weight difference between adults and children in reality leads to large calculation errors, failing to accurately reflect the physical intensity of congestion. Secondly, the detection is subject to numerous interferences and has blind spots. The weighing sensors are easily interfered with by non-personnel heavy objects such as suitcases and shopping carts on the cover, causing the load to be converted into an inflated number of passengers, which in turn leads to misjudgment of congestion. Through-beam sensors are easily blocked by passengers' clothing and belongings, causing signal misjudgment. The superposition of multiple errors will further amplify the overall judgment deviation. At the same time, the detection range is limited to the cover area and cannot cover the entire crowded area on the side of the escalator exit handrail, resulting in obvious detection blind spots.

[0005] The solution in Reference 3 is a static detection mode based on the condition of people remaining and congestion already formed, resulting in a response that lags behind the development of congestion. This solution determines "people remaining" by waiting for people to stay on the steps for more than 1-2 seconds and "congestion" by waiting for people to occupy more than 50% of the width, meaning intervention is only triggered after a static accumulation of people has formed at the exit. However, congestion at escalator exits develops gradually from a dynamic process of people gathering and pushing against each other on the handrails. This solution cannot capture this early warning sign and can only react passively after congestion has formed, missing the best opportunity for early evacuation. Furthermore, the solution in Reference 3 has a single hierarchical dimension and lacks precision in intermediate speed regulation, making it a coarse-grained control method. This scheme only sets two fixed actions: "deceleration at 50% width occupancy and stopping at 90% width occupancy." The deceleration range is a preset value and cannot adaptively adjust the speed adjustment range according to the actual physical intensity (compression force) of the crowding. Furthermore, the "personnel width occupancy percentage" is not directly related to the actual compression force, easily leading to misjudgments such as "high percentage but loose personnel with no compression" or "low percentage but tightly packed personnel with severe compression," resulting in a mismatch between speed adjustment and stopping operations and actual safety risks. The sensor deployment location in Reference 3 is unreasonable, easily damaged, and has high maintenance costs. The sensors are deployed on the surface / inside of the escalator steps, a high-frequency stepping area, and are constantly affected by personnel stepping on them, luggage running over them, and liquid immersion from spilled drinks, making them extremely prone to damage and malfunction, requiring frequent replacement. Moreover, the large number of grid-like sensors (calculated at 1 meter width and 10cm spacing, 10 sensors are needed per row) results in a large workload for subsequent inspection, replacement, and maintenance, leading to high overall maintenance costs. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method for controlling congestion at the exit of a passenger transport device. Pressure sensors are embedded in the side panels on both sides of the exit of the passenger transport device. The pressure signals from the pressure sensors are detected, and pressure values ​​are calculated based on the pressure signals from both sides. When the pressure value exceeds a preset threshold, an exit congestion warning signal for the passenger transport device is output, and the passenger transport device is managed and controlled according to the pressure value.

[0007] Preferably, the pressure sensor is a matrix pressure sensor.

[0008] Preferably, the method for calculating the pressure value based on the pressure signals from the left and right sides includes: Step S1: Calculate the bilateral synchronous average pressure. ; Step S2, based on bilateral synchronous average pressure Calculate three normalized risk factors; Step S3: Based on the three normalized risk factors, a weighted fusion and amplitude limit are performed to calculate the final stress value. .

[0009] Preferably, the bilateral synchronous average pressure is calculated. The formula is: ; This is the effective pressure value on the left. This represents the effective pressure value on the right side. ; .

[0010] Preferably, the three normalized risk factors are: pressure amplitude factor. Pressure rise rate factor Duration factor ; Preferably, pressure amplitude factor The calculation formula is: ; The preset basic trigger threshold, This is the preset rated threshold; Pressure rise rate factor The calculation formula is: Real-time ascent rate , The critical rate of ascent; This is a preset sliding calculation window; Duration factor The calculation formula is: ; The valid duration for continuously satisfying the verification conditions, This is the duration saturation value; Preferably, the final pressure value is calculated by weighted fusion and amplitude limiting. The method is as follows: ; Preset Value limit, Three consecutive windows Correction factor ,otherwise , This is the preset window threshold; ; These are the weighting coefficients. .

[0011] Preferably, the passenger conveying device is based on the final pressure value. Three-level gradient speed control is implemented. when Maintain normal operation at the rated frequency; when The operating frequency will be smoothly reduced at a preset speed; when This causes the passenger transport device to stop within a preset distance.

[0012] Compared with the prior art, the present invention has the following technical effects: 1. Real-time and direct sensing of compressive force: The pressure sensor is directly integrated into the side lining plates on both sides of the escalator exit, which can directly capture the lateral compressive force generated by passengers on the side lining plates when crowds are crowded. It eliminates the need for indirect methods such as visual recognition and load conversion, and completely solves the core defects of existing technologies such as "non-mechanical sensing" and "inability to quantify compressive force". The sensor has a fast response speed and can accurately capture the key signal of the sudden increase in compressive force in the early stage of a stampede. It breaks through the limitation of existing technologies that "can only identify existing deformation / static accumulation", and provides accurate and real-time mechanical data support for subsequent early warning and control.

[0013] 2. Improved accuracy of pressure detection: The dual-sided collaborative triggering logic effectively eliminates interference from single-sided accidental touches (single person leaning, object collision, accidental scratches). Only when the pressure sensors on both side panels detect pressure signals simultaneously and the signals reach the preset basic threshold is it determined to be "precursor to congestion or congestion state". This avoids the detection deviations of existing technologies (such as gravity sensing being easily interfered with by heavy objects and visual recognition being prone to missed detections and false alarms), and greatly improves the accuracy and reliability of pressure detection.

[0014] 3. Complete coverage of the detection area, eliminating blind spots: The pressure sensor is integrated into the side lining plate, which can cover the entire congested area on the escalator exit side. This solves the problem of blind spots in the existing gravity sensing solution, which "only covers the cover plate area". It can comprehensively capture the pressure changes at each position of the exit, ensuring no pressure monitoring is missed, and further improving the comprehensiveness of the detection.

[0015] 4. Reduce compliance costs and enhance the feasibility of technology application: There is no need to invest in the encryption, storage, and protection of privacy information (such as encrypted cameras and privacy desensitization algorithms), nor is there any need to deal with legal risks and disputes related to privacy leaks, which greatly reduces the compliance costs of technology application; at the same time, it avoids application restrictions caused by privacy protection issues, and can be freely applied in various public areas such as shopping malls, subways, and stations, which enhances the feasibility and promotion value of the technology solution. Attached Figure Description

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Figure 1 This is a schematic diagram showing the installation position of the pressure sensor in the exit congestion control method of the passenger transport device of the present invention. Detailed Implementation

[0017] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can fully understand other advantages and technical effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through different specific embodiments, and the details in this specification can also be applied based on different viewpoints, with various modifications or changes made without departing from the overall design concept of the invention. It should be noted that, unless otherwise specified, the following embodiments and features can be combined with each other. The following exemplary embodiments of the present invention can be implemented in many different forms and should not be construed as being limited to the specific embodiments set forth herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical solutions of these exemplary embodiments to those skilled in the art.

[0018] This specific embodiment provides a method for controlling congestion at the exit of a passenger transport device. By directly sensing the magnitude of crowding and squeezing pressure at the exit, the congestion at the exit of the passenger transport device can be monitored. The following description uses an escalator as an example.

[0019] like Figure 1As shown, multiple pressure sensors are embedded in the side panels on both sides of the escalator exit. The sensors are flush with the surface of the side panels, without protruding or affecting passenger passage or the escalator's appearance. The sensors are strategically arranged along the height of the escalator exit to fully cover the areas most prone to lateral compression when crowded. When the escalator moves upward, the pressure signal from the upper pressure sensor is detected; when the escalator moves downward, the pressure signal from the lower pressure sensor is detected.

[0020] The system adopts a dual-sided collaborative triggering logic: only when the pressure sensor of the left liner and the pressure sensor of the right liner simultaneously detect a valid pressure signal and the signal amplitude exceeds the preset basic trigger threshold is it determined to be a real congestion and squeezing signal, and an exit congestion warning signal of the passenger conveyor is output, thereby eliminating interference such as single person leaning, object scraping, and one-sided collision.

[0021] Meanwhile, the system employs a filtering algorithm on the pressure signal to eliminate false signals such as escalator vibration and slight environmental disturbances. By directly collecting lateral compressive force through the side lining plate, this embodiment achieves real-time, direct, and quantitative detection of the escalator exit, with no detection blind spots and unaffected by obstructions, lighting, or load conversion, fundamentally overcoming the shortcomings of existing visual and gravity detection solutions.

[0022] The arrangement of the pressure sensor in this invention has the following technical advantages: 1. Reduce hardware deployment costs and simplify deployment process: The sensors are integrated inside the escalator, eliminating the need for high-definition cameras, high-performance processors and a large number of grid sensors. This significantly reduces the number of hardware devices and equipment purchase costs. The embedded installation method eliminates the need for complicated debugging processes, making deployment easier and faster. This reduces labor costs for deployment and addresses the shortcomings of existing technologies that are "difficult and costly to deploy".

[0023] 2. Extend sensor lifespan and reduce maintenance costs: The side panels on both sides are not high-frequency stepping areas, and the embedded installation can avoid damage to the sensor from stepping, crushing, collision, etc., while reducing environmental interference such as beverage spills and dust accumulation; during later maintenance, there is no need to disassemble the bed cover, pedals, and other components. Only the surface of the side panels needs to be inspected. The maintenance process is simple, which greatly reduces the maintenance workload and maintenance costs, and solves the core defects of existing technology such as "easily damaged sensors and high maintenance costs".

[0024] 3. Improve scene adaptability and expand application scope: The simplified hardware structure can be adapted to different types of escalators (such as shopping mall escalators, subway escalators, and station escalators), without the need to adjust the hardware deployment plan according to the scene.

[0025] A specific exemplary technical solution is as follows: Matrix pressure sensors are respectively installed in the side lining plates on the left and right sides of the escalator exit. Left matrix pressure sensor (M rows N columns array); right-side matrix pressure sensor It is the same size as the one on the left.

[0026] Real-time pressure value of a single sensing unit (left matrix, row i, column j) (Right matrix, row i, column j). The basic trigger threshold is... This indicates the minimum effective pressure, eliminating accidental activation. The preset rated threshold. This represents the maximum effective pressure. The preset congestion trend cycle. Slide calculation window .

[0027] The raw array data from the single-sided matrix pressure sensor is converted into single-sided effective pressure values, and invalid interfering cells are removed. The specific method is as follows: Effective cells are screened, and only pressure values ​​are retained. The sensing unit eliminates invalid points with no pressure or slight pressure.

[0028] Step S1: Calculate the bilateral synchronous average pressure. Effective pressure value on the left side The calculation formula is as follows: ; Effective pressure value on the right side The calculation formula is as follows: ; Bilateral synchronous average pressure The calculation formula is as follows: ; Step S2 involves quantifying core risk characteristics based on bilateral synchronous average pressure. Calculate three normalized risk factors (range 0~1). 1. Pressure amplitude factor (Current congestion level): ; 2. Pressure rise rate factor (Core of Congestion Trend Prediction): Real-time rate of ascent , The critical rate of ascent; ; 3. Duration factor (Anti-interference): The valid duration for continuously satisfying the verification conditions, This is the duration saturation value; .

[0029] Step S3: The final P-value is quantized and output. After weighted fusion and limiting, the output is the final pressure value that can be directly used for control. value; Basic integration (core trend has the highest weight): ; These are the weighting coefficients. .

[0030] Trend enhancement correction: 3 consecutive windows Correction factor ,otherwise ;

[0031] Limiting output final value( for (Value upper limit, to prevent overflow due to sensor malfunction):

[0032] Passenger transport device based on final pressure value The value is controlled.

[0033] Specifically, based on the final pressure value The system employs a three-level gradient stepless speed control with tiered values. The passenger conveyor sets three control thresholds based on real-time compressive pressure collected from the side lining plates. Normal threshold; Warning threshold; Limit threshold.

[0034] And execute the following gradient control logic: Normal operating state ( The escalator should be kept running at its rated frequency without interference to ensure efficient passage.

[0035] Deceleration and evacuation state ( The control cabinet controls the frequency converter and uses variable frequency stepless speed regulation to smoothly reduce the escalator's operating frequency according to the preset speed, so as to achieve gradual deceleration and avoid sudden speed reduction that may cause discomfort to passengers.

[0036] Emergency safety shutdown state ( If the system determines that there is a serious risk of crushing, it controls the frequency converter to perform electrical smooth braking, so that the escalator stops smoothly within a short distance, preventing people from falling due to inertia due to sudden stop and avoiding secondary accidents.

[0037] This embodiment achieves precise matching between the squeezing force and the control action, balancing safety and passage efficiency, and solving the problems of crude control and high risk of direct elevator stoppage in existing technologies.

[0038] The above-mentioned hierarchical control strategy has the following technical effects: 1. Achieve gradient-based crowd control to address the problem of coarse-grained control: The three-level gradient control logic (corresponding to normal, warning, and danger levels of crowd control) can accurately match different control actions according to the actual size of the crowd control, completely solving the defects of existing technologies such as "single level, no gradient-based crowd control" and "direct escalator shutdown". It achieves precise control with "the more severe the congestion, the stronger the control", taking into account both safety and traffic efficiency.

[0039] 2. Stepless speed regulation adapts to actual risks, improving safety and traffic efficiency: Variable frequency stepless speed regulation technology can adaptively adjust the escalator's running speed according to the dynamic changes in the squeezing pressure, rather than a fixed deceleration range, ensuring that the speed adjustment action is precisely matched with the actual squeezing pressure; in the early stages of congestion, only the speed is reduced without stopping the escalator, which can ensure the escalator's traffic efficiency and avoid the inefficiency caused by "stopping the escalator directly in the early stages of congestion" in existing technologies; when the danger threshold is reached, the escalator can smoothly decelerate and stop, avoiding the inertial falls caused by sudden stops, eliminating secondary safety accidents, and improving operational safety.

[0040] 3. Avoid mismatch between control actions and actual risks: Use the squeezing force amplitude as the classification basis to replace indirect indicators such as "width ratio and load" in existing technologies. This ensures that the classification standard is directly related to the actual congestion intensity, avoids misjudgments such as "high ratio but no squeezing" or "low ratio but severe squeezing", and makes speed adjustment and elevator stopping actions more in line with actual safety risks, thereby improving the accuracy of control.

[0041] The present invention has been described in detail above through specific embodiments and examples, but these are not intended to limit the invention. Many modifications and improvements can be made by those skilled in the art without departing from the principles of the invention, and these should also be considered within the scope of protection of the present invention.

Claims

1. A method for controlling congestion at the exit of a passenger transport device, characterized in that, Pressure sensors are embedded in the side panels on both sides of the passenger conveyor exit. The pressure signals from the pressure sensors are detected, and the pressure values ​​are calculated based on the pressure signals from both sides. When the pressure value exceeds a preset threshold, an exit congestion warning signal for the passenger conveyor is output, and the passenger conveyor is controlled and operated according to the pressure value.

2. The exit congestion control method for a passenger transport device according to claim 1, characterized in that, The pressure sensor is a matrix pressure sensor.

3. The exit congestion control method for a passenger transport device according to claim 2, characterized in that, Methods for calculating pressure values ​​based on pressure signals from both sides include: Step S1: Calculate the bilateral synchronous average pressure. ; Step S2, based on bilateral synchronous average pressure Calculate three normalized risk factors; Step S3: Based on the three normalized risk factors, a weighted fusion and amplitude limit are performed to calculate the final stress value. .

4. The exit congestion control method for a passenger transport device according to claim 3, characterized in that, Calculate the bilateral synchronous average pressure The formula is: ; This is the effective pressure value on the left. This represents the effective pressure value on the right side. ; .

5. The exit congestion control method for a passenger transport device according to claim 3, characterized in that, The three normalized risk factors are: stress amplitude factor Pressure rise rate factor Duration factor .

6. The exit congestion control method for a passenger transport device according to claim 5, characterized in that, Pressure amplitude factor The calculation formula is: ; The preset basic trigger threshold, This is the preset rated threshold; Pressure rise rate factor The calculation formula is: Real-time ascent rate , The critical rate of ascent; This is a preset sliding calculation window; Duration factor The calculation formula is: ; The valid duration for continuously satisfying the verification conditions, This is the duration saturation value.

7. The exit congestion control method for a passenger transport device according to claim 6, characterized in that, The final pressure value is calculated by weighted fusion and amplitude limiting. The method is as follows: ; The preset upper limit of value, 3 consecutive windows Correction factor ,otherwise , This is the preset window threshold; ; These are the weighting coefficients. .

8. The exit congestion control method for a passenger transport device according to claim 3, characterized in that, Passenger transport device based on final pressure value Three-level gradient speed control is implemented. when Maintain normal operation at the rated frequency; when The operating frequency will be smoothly reduced at a preset speed; when This causes the passenger transport device to stop within a preset distance.

Images