A tower crane remote control failure hierarchical emergency linkage control method and system

Abstract

The application discloses a kind of tower crane remote control failure hierarchical emergency linkage control method and system, belong to rail transit engineering construction field, including the following steps: S1, the heartbeat signal of main communication link is monitored on edge computing gateway, when the heartbeat signal is abnormal, the real-time swing angle data of lifting hook and the historical instruction sequence of remote control end are collected, the current lifting hook swing angle and the running speed of each mechanism are combined, and the dynamic hesitation period length is calculated;The application integrates four links of failure determination, slow-release braking, permission transfer and parameter self-calibration into a closed-loop system with bidirectional information flow. The output of each layer not only drives the next layer to execute, but also serves as a feedback signal to reverse-calibrate the judgment parameters of the previous layer. The real-time state data of the braking process provides an initial reference for the permission transfer, and the smoothness index of the transfer optimizes the damping parameter settings of the braking phase in reverse, forming a complete enhancement loop.

Description

Technical Field

[0001] This invention belongs to the field of rail transit engineering construction, specifically, it relates to a hierarchical emergency linkage control method and system for tower crane remote control failure. Background Technology

[0002] Currently, with the popularization of 5G, fiber optic communication, and automated control technologies, remote tower crane operation (including ground-based cab remote control and cloud-based centralized control) has become an important development direction for smart construction sites. However, the tower crane operating environment is complex, and failure scenarios such as strong electromagnetic interference, signal fading due to obstacles, and remote control link equipment malfunctions are difficult to eliminate. Once the remote control link is interrupted, if the heavy-load hook suspended high in the air is not handled properly, it can easily cause serious safety accidents such as severe tower shaking, structural fatigue damage, collision with surrounding obstacles, or even the entire machine overturning. Among these, failure scenarios such as communication interruption, abnormal commands, and equipment failure remain the core bottlenecks restricting its safety, and existing solutions mainly have the following shortcomings. 1. Existing technologies typically rely solely on the "on / off" status of communication heartbeat signals to determine the urgency level, failing to consider the tower crane's current motion status (such as load, swing amplitude, and speed). This leads to frequent emergency brake triggering even under low-risk conditions like no-load or low-speed operation, impacting production efficiency and exacerbating equipment fatigue due to unnecessary mechanical shocks.

[0003] 2. When the remote control fails and triggers an emergency stop, the anti-sway algorithm is forcibly interrupted, and the hook is in an uncontrolled inertial swinging state. The rigid braking of the mechanical brake not only fails to suppress the swaying, but may also induce more violent secondary swinging, increasing the risk of collision.

[0004] 3. There is a response gap of hundreds of milliseconds to several seconds between the failure of the remote link and the takeover by on-site personnel using the backup remote control. During this period, the tower crane is in an uncontrolled state without intervention; moreover, the takeover is prone to sudden changes in control parameters, causing hook jerking and motor shock.

[0005] 4. The heartbeat timeout threshold, braking parameters, etc. are all set to factory preset values ​​and cannot be dynamically adjusted according to the actual electromagnetic environment and working conditions on the construction site, resulting in the system being either "overly sensitive" or "slow to respond" in different scenarios. Summary of the Invention

[0006] To address the aforementioned problems and technical deficiencies, this invention employs the following technical solution: a tiered emergency response control method for tower crane remote control failure, comprising the following steps: S1. Step onto the edge computing gateway to monitor the heartbeat signal of the main communication link. When the heartbeat signal is abnormal, collect the real-time swing angle data of the hook and the historical command sequence of the remote control terminal. Combine the current hook swing angle and the operating speed of each mechanism to calculate the dynamic hesitation period. If the main communication link is restored within the dynamic hesitation period, the current operating state is maintained. If it is not restored after the timeout, a graded failure trigger signal is output according to the comprehensive situation of the hook swing angle and speed. S2. After receiving the graded failure trigger signal, identify the current load equivalent rotational inertia online, calculate the virtual damping coefficient and virtual stiffness coefficient, generate virtual damping compensation torque accordingly and inject it into the frequency converter of each mechanism; at the same time, control the frequency converter to execute the speed curve deceleration, and engage the mechanical brake after the speed drops below the threshold, and continuously broadcast the real-time operating status data of the hook to the outside during the braking process. S3. Simultaneously with the initiation of S2, the emergency wireless link is activated and an authorized readiness beacon is broadcast. Upon receiving the authorized readiness beacon, the on-site convenient terminal displays the current hook movement status and suggested operating direction. When the on-site convenient terminal sends a takeover request, it compares its joystick commands with the last valid command cached from the remote control terminal. Figure 1 Consistency comparison: when the deviation meets the preset condition, the control switch is completed by a ramp function transition method; S4. Record the full time-series data of each failure event from occurrence to completion of handling, forming an event slice; extract the key performance indicators in the event slice, compare and attribute them with the preset benchmark values, generate parameter correction suggestions, and feed them back to the corresponding links in S1 to S3.

[0007] Preferably, the duration T of the dynamic hesitation period in S1 is... wait The calculation formula is:

[0008] In the formula: T base The baseline waiting time; θ safe The safety swing angle threshold; θ current This is the current hook swing angle; To prevent the elimination of the zero constant; k is the velocity attenuation coefficient; v current This represents the maximum current speed among all institutions.

[0009] Furthermore, the historical instruction sequence is read as follows: read the handle instruction sequence in the last preset time period in the circular instruction buffer. If the sequence shows a monotonically decreasing trend of continuously pushing back to zero and the end instruction value is less than a preset proportion of the rated value, it is determined that the operator has the intention to actively stop work and is allowed to enter the dynamic hesitation period; otherwise, a graded failure trigger signal is directly output.

[0010] Preferably, the equivalent rotational inertia J of the load in S2 load The calculation formula is:

[0011] In the formula: G is the current lifting weight; R is the current amplitude; i is the reduction ratio of the reducer; n is the drum radius; η is the transmission efficiency.

[0012] Preferably, the virtual damping coefficient D in S2 virtual and virtual stiffness coefficient K virtual The calculation formula is:

[0013]

[0014] The virtual damping compensation torque T comp The calculation formula is:

[0015] In the formula: D base For the reference virtual damping system; K base J is the reference virtual stiffness coefficient; rated ω is the equivalent moment of inertia under rated load; m (t) represents the real-time speed of the motor; θ swing (t) represents the hook swing angle component in the corresponding direction.

[0016] Preferably, the graded failure triggering signal in S2 includes an L2 level signal and an L3 level signal; When an L2 level signal is received, the virtual damping compensation and speed curve deceleration are performed, and the mechanical brake is engaged after a preset time delay after the deceleration drops below the crawling speed threshold. When an L3 signal is received, the mechanical brake closing command is triggered synchronously, and the amplitude of the virtual damping compensation torque is increased to the preset multiple of L2 level. At the same time, the residual sway amplitude of the tower body is monitored, and when it exceeds the safety threshold, a reverse jog pulse is output to the slewing mechanism.

[0017] Preferably, the meaning described in S3 Figure 1 The formula for calculating the deviation Δ of consistency comparison is:

[0018] When the following conditions are met simultaneously, it is determined that the intent has been confirmed, and control is transferred: Condition 1: Local cmd The direction of change of (t) and Ref cmd The corresponding directions of motion are the same; Condition 2: Δ is less than the preset deviation threshold.

[0019] Furthermore, the control command Cmd for the ramp function transition described in S3 exec The calculation formula is:

[0020] In the formula: t switch T is the switching time; ramp t represents the ramp transition time; t represents the current time.

[0021] Preferably, the key performance indicators mentioned in S4 include: the effective avoidance rate indicator, which is measured by the percentage of times communication recovers automatically during the hesitation period without entering the graded braking phase; The braking balance index is measured by the ratio of the maximum swing amplitude of the hook during braking to the initial swing amplitude before braking. The smoothness index of the switchover is measured by the ratio of the peak value of the motor current surge at the moment of switching to the rated current. The time-series causal thrust engine locates the cause to the corresponding parameter link and generates correction suggestions based on the deviation direction of the above indicators from the preset benchmark value.

[0022] A tiered emergency response control system for tower crane remote control failure, employing the aforementioned tiered emergency response control method for tower crane remote control failure, includes: The remote control terminal is used to send control commands and receive operating data; The tower-mounted edge computing gateway includes a second communication module for establishing a main communication link with the remote control terminal, an inertial measurement unit for collecting hook swing angle data, a data buffer and processing module with a built-in time-series causal inference engine, and a frequency converter interface module for data interaction with the frequency converters of each mechanism; an emergency wireless link, set in parallel with the second communication module, which is automatically activated when the main communication link fails; a field portable terminal that communicates with the tower-mounted edge computing gateway through the emergency wireless link; and drive units for each mechanism, including hoisting frequency converter, luffing frequency converter, slewing frequency converter, and mechanical brake.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This tiered emergency linkage control method and system for tower crane remote control failure replaces the rigid logic of "heartbeat timeout for emergency stop" in traditional technology by introducing hook swing angle sensing and operator intent recognition as joint criteria. Under low-risk conditions such as hook stability and low-speed return to position, the system actively tolerates faults through a dynamic hesitation period mechanism, effectively avoiding unnecessary emergency braking caused by instantaneous jitter of 5G signals. The false trigger rate can be effectively reduced in typical construction site electromagnetic environments, while not affecting rapid response to truly dangerous conditions, achieving an optimized balance between safety protection and operational efficiency.

[0024] 2. The tiered emergency linkage control method and system for tower crane remote control failure identifies the equivalent rotational inertia of the load online and injects virtual damping compensation torque into the frequency converter, simulating the physical effect of a hydraulic damper connected in parallel on the motor shaft. The braking process itself becomes an active absorption stage of the hook's swing kinetic energy, rather than the laissez-faire approach after the anti-sway algorithm is interrupted in traditional solutions. During L2-level braking, the maximum hook swing amplitude is effectively reduced compared to the traditional direct brake solution, and the peak impact load on the tower structure is also effectively reduced. At the same time, the hook's motion status is continuously broadcast through the emergency link throughout the braking process, providing complete situational awareness data for subsequent manual takeover.

[0025] 3. The tiered emergency response control method and system for tower crane remote control failure, through a pre-broadcast wake-up and status push mechanism via emergency wireless links, reduces the time for on-site personnel to acquire situational awareness to near zero. It innovatively employs an "intent mirroring and comparison" mechanism, transforming the handover of control from a rigid on / off switch to a conditional, smooth transition. The average delay from the on-site personnel pressing the takeover confirmation button to actually gaining control is significantly reduced to approximately 48ms, an order of magnitude shorter than traditional solutions. The ramp function transition eliminates the step-like impact of commands at the moment of takeover, providing on-site personnel with an intuitive experience of "equipment following expectations," significantly reducing cognitive load and the probability of misoperation in emergency situations.

[0026] 4. The tiered emergency response control method and system for tower crane remote control failure utilizes an edge-side temporal causal inference engine to slice and record full data for each failure event, automatically locating the source of deviation and generating parameter correction suggestions. After the same hardware system is deployed and operated at different construction sites for several months, it will naturally develop differentiated parameter configurations, maintaining high sensitivity at sites with stable signals and automatically becoming more stable at sites with complex electromagnetic interference. Furthermore, the system is no longer a factory-fixed, closed system, but an open, evolving system that continuously approaches its theoretical optimal value as operating time and handling experience accumulate. Each failure event becomes a training sample for improving system performance.

[0027] 5. This tiered emergency response control method and system for tower crane remote control failure organically integrates four stages—failure judgment, slow-release braking, authority handover, and parameter self-calibration—into a closed-loop system with bidirectional information flow. The output of each layer not only drives the execution of the next layer but also serves as a feedback signal to calibrate the judgment parameters of the previous layer. Real-time status data during the braking process provides an initial benchmark for authority handover, and the smoothness index of the handover in turn optimizes the damping parameter settings during the braking phase, forming a complete reinforcement loop. This architecture solves the structural defects of the existing technology where the "detection-braking-takeover" stages are isolated from each other, providing a complete, reliable, and evolvable solution for emergency response to tower crane remote control failures. Attached Figure Description

[0028] In the attached diagram: Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Generally, the components of the embodiments of the present invention described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0030] Example 1: System Overall Architecture: This embodiment provides a tiered emergency response control system for tower crane remote control failure, comprising: Remote control terminal: Located in the ground control room or cloud-based centralized control center, it includes a control handle, a monitoring display screen, and a first communication module (5G CPE / fiber optic transceiver), used to send control commands to the tower crane and receive operating data; Tower-mounted edge computing gateway: Installed in the tower crane's cab or electrical cabinet, it serves as the core decision-making unit of this system. It includes: Second communication module: Establishes a main communication link with the remote control terminal; Inertial measurement unit: used to acquire real-time triaxial acceleration and swing angle data of the hook; Data buffering and processing module: Built-in time-series causal inference engine; Inverter interface module: Enables high-speed data exchange with inverters in various institutions.

[0031] Emergency wireless link: An independent data transmission radio using the 433MHz or 2.4GHz frequency band is connected in parallel with the second communication module and is automatically activated when the main communication link fails.

[0032] Portable terminal on site: a smart bracelet or handheld remote control worn by on-site commanders, with a built-in third communication module, communicating with the tower edge computing gateway through the emergency wireless link.

[0033] Each mechanism drive unit includes a hoisting inverter, a luffing inverter, a slewing inverter, and a mechanical brake.

[0034] The modules mentioned above interact with each other via industrial Ethernet or CAN bus, forming a four-layer linkage closed-loop control architecture.

[0035] Example 2: First Layer – Failure Prediction of Hook Attitude and Dynamic Delay of “Hesitation Period” This embodiment is mainly used to solve the problem of "single failure determination method and high false judgment rate" mentioned in the background technology; The second communication module of the tower edge computing gateway sends a heartbeat request frame to the remote control terminal at a first period of T1=50ms and waits for a response frame. Simultaneously, the inertial measurement unit collects the real-time swing angle θ of the hook in the lifting plane and luffing plane at a second period of T2=10ms. x θ y And calculate the resultant angle:

[0036] The data buffer and processing module has a pre-set circular command buffer that continuously stores the sequence of commands sent by the remote control terminal to each mechanism handle within the last 3 seconds. If the second communication module fails to receive a response frame within N consecutive heartbeat cycles (N=3), it determines that the main communication link has malfunctioned. At this point, the system does not immediately trigger braking, but instead enters the "dynamic failure level determination" process: Step S101: Read the handle command sequence within the last T=3 seconds in the circular command buffer. If the sequence shows a monotonically decreasing trend of "each mechanism handle continuously pushing back to the zero position", and the final command value is less than 5% of the rated value, then it is determined that the operator has "actively intends to stop work", and a flag is set. intent =1, otherwise Flag intent =0.

[0037] Step S102: Read the current hook swing angle θ current and the current speed v of each institution current (lifting speed v) h Amplitude speed v l Rotational speed v s (the maximum value in the range).

[0038] Step S103: Calculate the dynamic hesitation period T wait The formula is as follows:

[0039] Wherein: T base The baseline waiting time is 300ms (default); θ safe The safety swing angle threshold is preset to 1.5°; ε is a small constant to prevent division by zero, set to 0.01°; k is the velocity attenuation coefficient, preset to 0.05 s / m; v current The units are m / s or ° / s, and have been normalized.

[0040] Step S104: If Flag intent =1 and T wait If the value is greater than 0, then the cooling-off period timer is started. In T... waitDuring this period: The system maintains the current operation of all mechanisms; if the main communication link is restored during this period, the failure marker is cleared, the tower crane continues to respond to remote commands, and a "effective avoidance" event is recorded; if T wait If the timer expires and the system still hasn't recovered, an L2 level failure trigger signal will be output to the second layer. If Flag... intent =0 or θ current >θ threshold (If the default value is 5°), the hesitation period is skipped, and the L2 or L3 level failure trigger signal is directly output to the second layer.

[0041] Compared to the rigid "heartbeat timeout triggers emergency stop" judgment in existing technologies, this layer introduces hook posture perception and operation intention recognition as joint criteria. Under low-risk conditions such as low-speed, unloaded return of the tower crane and stable hook operation, the system exhibits "active fault tolerance," effectively avoiding unnecessary emergency braking caused by momentary jitter in the 5G signal. In typical construction site electromagnetic environments, the false trigger rate can be effectively reduced without affecting rapid response to truly dangerous conditions. Furthermore, the continuous heartbeat monitoring mechanism during the cooling-off period allows for a seamless return to normal operation after brief signal recovery, ensuring production continuity.

[0042] Example 3: Second Layer – Graded Slow-Release Braking Based on “Virtual Inertia Cancellation” This embodiment is mainly used to solve the problem of "lack of buffering in the braking process and forced interruption of the anti-shake algorithm" in the prior art.

[0043] A high-speed fieldbus connection (such as EtherCAT or Profinet) is established between the inverter interface module of the tower edge computing gateway and the inverters of each mechanism, supporting real-time writing of torque compensation commands. The data buffer and processing module is pre-loaded with online load inertia identification algorithms and virtual damping control algorithms.

[0044] Upon receiving an L2 or L3 failure trigger signal from the first layer, this layer executes the following graded braking procedure: Level 2 braking procedure (confirm interruption, execute slow-release stop): Step S201: Online identification of load inertia. Read the current lifting load G (provided by the torque limiter) and amplitude R, and calculate the equivalent rotational inertia J of the load referred to the hoisting motor shaft. load :

[0045] Where i is the reduction ratio of the reducer, n is the drum radius, and η is the transmission efficiency.

[0046] Step S202: Based on the identified J load Calculate the virtual damping coefficient D virtual and virtual stiffness coefficient K virtual，The following empirical formula is used in this embodiment:

[0047]

[0048] Among them, D base K base These are the baseline parameters under rated load, which are pre-set in the system after offline simulation calibration.

[0049] Step S203: Generate virtual damping compensation torque. The inverter interface module writes the torque compensation value T to the inverters of each mechanism at a third cycle T3 = 5ms. comp :

[0050] Where, ω m (t) represents the real-time motor speed (feedback from the frequency converter), θ swing (t) represents the hook swing angle component in the corresponding direction (provided by the inertial measurement unit).

[0051] The physical meaning of this compensating torque is: it simulates a virtual hydraulic damper connected in parallel on the motor shaft. When the hook swings to one side, the motor outputs a reverse damping torque to gradually dissipate the swing kinetic energy, rather than allowing it to be released in a concentrated manner at the moment of mechanical braking.

[0052] Step S204: Simultaneously, the frequency converter executes an S-shaped speed curve deceleration, with the target speed at zero, and the jerk is limited to a preset value to avoid secondary excitation during the deceleration process.

[0053] Step S205: After the speed of each mechanism drops to below the crawling speed threshold (e.g., 2% of the rated speed), the mechanical brake is engaged after a 50ms delay to complete the final locking.

[0054] Level L3 braking procedure (emergency condition, forced anchoring): When an Level L3 trigger signal is received (e.g., first-level determination S≥S2, or encoder overspeed / runaway detected), step S206: Based on steps S201-S203, the following actions are triggered simultaneously: immediately issue a closing command to the mechanical brake; issue a maximum braking torque command to the eddy current retarder (if equipped); virtual damping compensation torque T comp The amplitude is increased to 1.5 times that of L2 level.

[0055] Step S207: Using data from the tower top RTK positioning module and inertial measurement unit, monitor the elastic deformation displacement of the tower top in real time. When the residual sway amplitude of the tower exceeds the preset safety threshold, the edge computing gateway sends a set of short, small-amplitude reverse jogging pulses to the slewing mechanism through the inverter interface module to actively counteract the elastic potential energy of the tower and suppress residual swaying after shutdown.

[0056] Throughout the braking process, the data buffering and processing module packages the following real-time data at a period of T4=20ms and broadcasts it externally via the emergency wireless link: the current hook swing angle θ. current Current speed v of each institution h v l v s The current amplitude of the virtual damping compensation torque; braking progress flag (L2 in progress / L2 completed / L3 in execution). This broadcast data serves as the initial state reference for the handover of third-level permissions.

[0057] Therefore, compared to the existing technology's approach of "failure means interruption, anti-sway, and forced braking," this solution achieves a deep integration of the braking process and active anti-sway. The injection of virtual damping compensation torque makes the braking process itself an "active absorption" phase of hook sway, rather than a "laissez-faire" phase. Furthermore, during L2-level braking, the maximum hook swing is effectively reduced compared to traditional direct braking solutions, and the peak impact load on the tower structure is also effectively reduced. More importantly, the braking process is no longer a "black box" period; the real-time motion status of the hook is continuously broadcast externally via the emergency link, providing complete situational awareness for on-site personnel and solving the pain point of "facing unknown motion status during takeover" in existing technologies.

[0058] Example 4: Layer 3 – Zero-Delay Radio Wave Handover Protocol Based on “Motion Intent Mirroring” This embodiment is mainly used to solve the problem in the prior art of "control vacuum during authority handover and step jump in control quantity at the moment of takeover".

[0059] The portable terminal on-site includes a built-in third communication module, display screen, and control joystick. The data buffer and processing module of the tower-mounted edge computing gateway has a "reference trajectory buffer" to cache the intentional commands of the remote control terminal just before a failure.

[0060] The emergency wireless link adopts a half-duplex mode, with a power-on standby power consumption of less than 0.5W. It is in a sleep listening state when the tower crane is running normally and is only awakened and activated when the second-level braking process is started.

[0061] Work style: Phase 1: Intent mirroring lock at the moment of failure. When the first layer determines that the main communication link is abnormal and triggers the second layer braking, the on-tower edge computing gateway synchronously performs the following operations: Step S301: Immediately read the last valid frame of handle command data (i.e., the last valid control intent issued by the remote control terminal) from the circular command buffer, store it in the reference trajectory buffer, and mark it as "Intent Reference Value Ref". cmd ".

[0062] Step S302: Continuously transmit “Authorization Ready” beacon frames in broadcast mode via the emergency wireless link. The frame contains: the tower crane’s unique identifier; the current braking level (L2 / L3); and an encrypted authorization token.

[0063] Phase Two: Activation and Status Synchronization of Field Terminals Step S303: Upon receiving the "Authorization Ready" beacon frame, the portable terminal worn by on-site personnel is prompted by a vibration motor, and the display screen automatically lights up and displays: a visual graphic of the current hook swing angle (from the θ broadcast by the second layer). current Data-driven); current velocity values ​​and direction indications for each mechanism; Intended reference value Ref cmd The corresponding suggested operation direction (indicated by arrow icons).

[0064] Step S304: After the on-site personnel confirm the takeover intention, they press the "Takeover Confirmation" button on the portable terminal. The portable terminal then sends a "Takeover Request" frame to the tower edge computing gateway and begins uploading the real-time position value of the local joystick.

[0065] Phase 3: Comparison of Motion Intent and Seamless Switching The core of this stage lies in not directly switching control, but rather in "intentional" action. Figure 1 "Consistency verification" only checks whether the on-site personnel's operating intention matches the remote operator's final intention. Figure 1 The transfer of permissions should only be completed when a smooth transition is possible.

[0066] Step S305: After receiving the "takeover request," the tower-based edge computing gateway initiates the intent comparison process. It continuously receives local joystick position values. cmd (t) and the Ref in the reference trajectory buffer cmd Perform real-time comparison and calculate the deviation Δ:

[0067] Step S306: When both of the following conditions are met simultaneously, the action is deemed "intent confirmation," and a switch of control is executed: Condition 1: Local cmd The direction of change of (t) (the direction of joystick movement) and Ref cmdThe corresponding directions of motion are the same; Condition 2: Δ < Δ threshold , where Δ threshold The preset deviation threshold is set to 15% of the rated command value.

[0068] Step S307: At the moment of control switch, the PLC actually executes the instruction Cmd. exec Transition using the following ramp function:

[0069] Where t switch For the switching time, T ramp This is the ramp transition time, preset to 50ms.

[0070] This ramp transition ensures that the control volume is transferred from Ref. cmd Smooth transition to Local cmd This completely eliminates the instruction step impact caused by "hard switching" in the background technology.

[0071] Step S308: After the switch is completed, the virtual damping compensation torque accumulated during the second-level braking process is gradually released to zero, and the control is completely transferred to the on-site portable terminal.

[0072] Phase Four: Direct Emergency Takeover (Bypass Mode) If, in step S303, on-site personnel observe a dangerous movement of the hook, without waiting for intent comparison, they can directly press and hold the "Emergency Takeover" button on the portable terminal for more than 3 seconds. The system will skip steps S305-S306 and directly execute the switching process of steps S307-S308, but T ramp Reduced to 20ms for the fastest response time.

[0073] Compared to the sequential and delayed process of "detecting anomalies → repositioning → powering on and matching codes → seizing control" in existing technologies, this solution reduces the time for on-site personnel to acquire situational awareness to almost zero through pre-broadcast wake-up and status push via the emergency link. Furthermore, the introduction of intent mirroring and comparison mechanisms transforms the transfer of control from a "brutal grab" to a "conditional confirmation," ensuring the continuity of control at the moment of takeover.

[0074] The average delay between pressing the "takeover confirmation" button and actually gaining control has been effectively reduced. More importantly, on-site personnel will have an intuitive experience of "equipment following expectations" during operation, because the suggested operation direction displayed by the system is highly consistent with their judgment, greatly reducing the cognitive load and probability of misoperation in emergency situations.

[0075] Example 5: Fourth Layer – Edge-Side Temporal Causality Analysis Engine and Parameter Self-Calibration This embodiment is mainly used to solve the problem of "fixed system parameters and lack of adaptive capability" in the prior art; The data buffering and processing module of the tower-based edge computing gateway embeds a lightweight temporal causal inference engine. This engine is implemented based on decision trees or small-scale neural networks, with model parameters controlled to within 1MB, and can run directly on the embedded processor in real time. The data buffering and processing module is equipped with a non-volatile memory area (such as ferroelectric RAM) for persistently storing full slice data of each failure event.

[0076] Work away from: Step S401: Record full data slices of the failure event Whenever the system goes through a complete failure emergency process (from the first layer determining the failure to the third layer completing the handover of permissions or braking), the time-series causal inference engine automatically extracts the full data for the following time period to form an "event slice": Start time T start The moment when the first abnormal heartbeat is detected in the first layer is moved forward by 5 seconds; the end time T end : 10 seconds after the completion of the third-level authority handover or the completion of the second-level braking.

[0077] Each event slice contains time-series data in the following dimensions (sampling frequency 500Hz): Communication layer: main link bit error rate, signal strength indication, heartbeat response delay; Control layer: handle command values ​​of each mechanism, actual output frequency / torque of the frequency converter, virtual damping compensation value; Mechanical layer: hook three-axis acceleration, swing angle, motor speed, tower top displacement; Interaction layer: field terminal activation delay, joystick deviation Δ, ramp switching time.

[0078] Step S402: Causal Attribution and Performance Deviation Analysis Once an event slice is recorded, the time-series causal inference engine executes the following analysis process offline: Sub-step S4021: Extract the "key performance indicators" for this event, including: Accidental trigger flag: if the cooling-off period T wait If communication is restored automatically during the event and L2 braking is not initiated, the event is considered "effective avoidance" and is marked as 1; otherwise, it is marked as 0.

[0079] Braking smoothness: Measured by the ratio of the maximum swing amplitude of the hook during braking to the initial swing amplitude before braking.

[0080] Smoothness of switching: Measured by the ratio of the peak motor current surge at the moment of switching to the rated current.

[0081] Sub-step S4022: Compare the above indicators with the pre-stored "ideal benchmark value" to determine the source of deviation.

[0082] Step S403: Generation and confirmation of parameter correction suggestions Step S4031: The temporal causal inference engine generates a set of parameter correction suggestions based on the attribution results. For example: Suggested correction: T base The adjustment from 300ms to 450ms is based on the following: In the last 10 events, T wait The average value was 412ms, and 80% of the effective recovery occurred in the 350-500ms range, with a confidence level of 0.82.

[0083] Step S4032: The suggestion is uploaded to the cloud monitoring platform through the tower crane's maintenance communication link (such as a 4G IoT module) and reviewed and confirmed by authorized management personnel.

[0084] Step S4033: After confirmation, the corrected parameters are sent to the edge computing gateway on the tower via an encrypted link to update the operating parameters of the corresponding module; at the same time, the corrected record is written to the non-volatile storage area as prior knowledge for subsequent attribution analysis.

[0085] Step S404: Model parameter transfer and generalization For multiple tower cranes of the same model and batch, the cloud platform can aggregate time-series slice data of each individual machine and perform cross-device statistical analysis. After feature engineering, it can extract optimized parameter combinations with generalization capabilities and deploy them in batches to other equipment in the same batch, achieving collaborative evolution of "single machine experience, group benefit".

[0086] Through the above technical solution, this embodiment endows the system with the ability to "learn from experience," solving the inherent problem in existing technologies where fixed parameters cannot adapt to dynamic changes at construction sites. After the same hardware system is deployed and operated at different construction sites for several months, it will naturally form differentiated parameter configurations. At newly built construction sites with stable 5G signals, the system maintains high sensitivity; at renovation sites with complex electromagnetic interference, the system automatically becomes more "stable and patient." This adaptive capability makes the emergency control system of this method no longer a closed system that is fixed at the factory, but an open and evolving system that continuously optimizes its response strategy as operating time increases and handling experience accumulates. Every failure event becomes a training sample for the system's "growth." After long-term operation, key indicators such as the system's false trigger rate, braking stability, and takeover smoothness will continuously approach the theoretical optimal value.

[0087] Example 6: Overall Workflow of Four-Layer Linkage To make the linkage relationship of this method clearer, the following describes a complete emergency handling process for remote control failure: At time T0: The tower crane is operating normally. The remote control terminal sends control commands to the edge computing gateway on the tower via the 5G link, and the hook is running smoothly.

[0088] At time T1: The 5G signal was momentarily interrupted due to obstruction, and the second communication module did not receive a response frame for three consecutive heartbeat cycles.

[0089] At time T2 (T1+150ms): The first level of failure assessment is initiated. It is detected that the operator's last 3-second command sequence is a continuous deceleration to zero, and the hook swing angle is only 0.8°. The calculated T... wait = 520ms, startup cooling-off period.

[0090] At time T3 (T2+200ms): The 5G signal recovers automatically during the cooling-off period, the second communication module receives the heartbeat response again, the first layer clears the failure flag, the system seamlessly returns to normal operation, and an "effective avoidance" event is recorded. The fourth layer records the timing data of this event, confirming T. base The settings are reasonable.

[0091] Let's consider another scenario: At time T3' (T2+520ms): The cooling-off period ends, but the 5G signal has not yet been restored. The first layer outputs an L2 level failure trigger signal to the second layer.

[0092] At time T4 (T3'+10ms): Second layer startup. Identify the current load inertia, calculate the virtual damping coefficient, and inject reverse compensation torque into the frequency converter; simultaneously, the emergency wireless link is activated, and broadcasts the "authorized ready" beacon and real-time hook swing data begin.

[0093] At time T5 (T4+2s): On-site personnel receive a vibration alert from the portable terminal, observe that the display shows the hook is slowly decelerating and the swing amplitude is controllable, and press "Confirm Takeover".

[0094] At time T6 (T5+48ms): The third layer completes the intent comparison, and the control is smoothly transferred to the field terminal. The field personnel continue to operate the tower crane to complete the current hoisting task.

[0095] Time T7 (T6+30s): Task completed, tower crane safely shut down, event slice generated on the fourth layer, attribution analysis confirms that all indicators of this handling are better than the baseline value, no parameter adjustment is required.

[0096] Throughout the process, the data flow and control flow between each layer are intertwined and progressively advance, forming a complete closed-loop linkage emergency control chain.

[0097] The embodiments described above are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications, improvements, and substitutions without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. A tiered emergency response control method for tower crane remote control failure, characterized in that, Includes the following steps: S1. Step onto the edge computing gateway to monitor the heartbeat signal of the main communication link. When the heartbeat signal is abnormal, collect the real-time swing angle data of the hook and the historical command sequence of the remote control terminal. Combine the current swing angle of the hook with the operating speed of each mechanism to calculate the dynamic hesitation period. If the main communication link is restored within the specified dynamic hesitation period, the current operating state will be maintained; if it is not restored after the timeout, a graded failure trigger signal will be output based on the comprehensive situation of the hook swing angle and speed. S2. After receiving the graded failure trigger signal, identify the current load equivalent rotational inertia online, calculate the virtual damping coefficient and virtual stiffness coefficient, generate virtual damping compensation torque accordingly and inject it into the frequency converter of each mechanism; at the same time, control the frequency converter to execute the speed curve deceleration, and engage the mechanical brake after the speed drops below the threshold, and continuously broadcast the real-time operating status data of the hook to the outside during the braking process. S3. Simultaneously with the initiation of S2, the emergency wireless link is activated and an authorized ready beacon is broadcast. After receiving the authorized ready beacon, the on-site convenient terminal displays the current hook movement status and suggested operation direction. When the on-site convenient terminal sends a takeover request, its joystick command is compared with the last valid command from the cached remote control terminal to ensure consistency of intent. If the deviation meets the preset condition, the control switch is completed using a ramp function transition method. S4. Record the full time-series data of each failure event from occurrence to completion of handling, forming an event slice; extract the key performance indicators in the event slice, compare and attribute them with the preset benchmark values, generate parameter correction suggestions, and feed them back to the corresponding links in S1 to S3.

2. The hierarchical emergency linkage control method for tower crane remote control failure according to claim 1, characterized in that, The dynamic hesitation period T mentioned in S1 wait The calculation formula is: In the formula: T base The baseline waiting time; θ safe The safety swing angle threshold; θ current This is the current hook swing angle; To prevent the elimination of the zero constant; k is the velocity attenuation coefficient; v current This represents the maximum current speed among all institutions.

3. The graded emergency linkage control method for tower crane remote control failure according to claim 2, characterized in that, The historical instruction sequence is read as follows: read the handle instruction sequence in the last preset time period in the circular instruction buffer. If the sequence shows a monotonically decreasing trend of continuously pushing back to zero and the end instruction value is less than the preset proportion of the rated value, it is determined that the operator has the intention to actively stop work and is allowed to enter the dynamic hesitation period. Otherwise, output the graded failure trigger signal directly.

4. The hierarchical emergency linkage control method for tower crane remote control failure according to claim 1, characterized in that, The equivalent rotational inertia of the load described in S2 is J. load The calculation formula is: In the formula: G is the current lifting weight; R is the current amplitude; i is the reduction ratio of the reducer; n is the drum radius; η is the transmission efficiency.

5. A graded emergency linkage control method for tower crane remote control failure according to claim 4, characterized in that, The virtual damping coefficient D mentioned in S2 virtual and virtual stiffness coefficient K virtual The calculation formula is: The virtual damping compensation torque T comp The calculation formula is: In the formula: D base For the reference virtual damping system; K base J is the reference virtual stiffness coefficient; rated ω is the equivalent moment of inertia under rated load; m (t) represents the real-time speed of the motor; θ swing (t) represents the hook swing angle component in the corresponding direction.

6. The hierarchical emergency linkage control method for tower crane remote control failure according to claim 1, characterized in that, The graded failure triggering signals described in S2 include L2 level signals and L3 level signals; When an L2 level signal is received, the virtual damping compensation and speed curve deceleration are performed, and the mechanical brake is engaged after a preset time delay after the deceleration drops below the crawling speed threshold. When an L3 signal is received, the mechanical brake closing command is triggered synchronously, and the amplitude of the virtual damping compensation torque is increased to the preset multiple of L2 level. At the same time, the residual sway amplitude of the tower body is monitored, and when it exceeds the safety threshold, a reverse jog pulse is output to the slewing mechanism.

7. A graded emergency linkage control method for tower crane remote control failure according to claim 1, characterized in that, The formula for calculating the deviation Δ of the intent consistency comparison described in S3 is as follows: When the following conditions are met simultaneously, it is determined that the intent has been confirmed, and control is transferred: Condition 1: Local cmd The direction of change of (t) and Ref cmd The corresponding directions of motion are the same; Condition 2: Δ is less than the preset deviation threshold.

8. A graded emergency linkage control method for tower crane remote control failure according to claim 7, characterized in that, The control command Cmd for the ramp function transition in S3 exec The calculation formula is: In the formula: t switch T is the switching time; ramp t represents the ramp transition time; t represents the current time.

9. A graded emergency linkage control method for tower crane remote control failure according to claim 1, characterized in that, The key performance indicators mentioned in S4 include: the effective avoidance rate, which is measured by the percentage of times communication recovers automatically during the hesitation period without entering the graded braking phase. The braking balance index is measured by the ratio of the maximum swing amplitude of the hook during braking to the initial swing amplitude before braking. The smoothness index of the switchover is measured by the ratio of the peak value of the motor current surge at the moment of switching to the rated current. The time-series causal thrust engine locates the cause to the corresponding parameter link and generates correction suggestions based on the deviation direction of the above indicators from the preset benchmark value.

10. A tiered emergency response control system for tower crane remote control failure, characterized in that, The method for hierarchical emergency linkage control of tower crane remote control failure as described in any one of claims 1-9 includes: The remote control terminal is used to send control commands and receive operating data; The tower-mounted edge computing gateway includes a second communication module for establishing a main communication link with the remote control terminal, an inertial measurement unit for collecting hook swing angle data, a data buffer and processing module with a built-in time-series causal inference engine, and a frequency converter interface module for data interaction with the frequency converters of each mechanism; an emergency wireless link, set in parallel with the second communication module, which is automatically activated when the main communication link fails; a field portable terminal that communicates with the tower-mounted edge computing gateway through the emergency wireless link; and drive units for each mechanism, including hoisting frequency converter, luffing frequency converter, slewing frequency converter, and mechanical brake.

Images