A smoothness detection system and method for a tundish crane

Abstract

The present application relates to the technical field of taphole crane, and particularly relates to a taphole crane stability detection system and method, which comprises: a crane body comprising a crane motor, a steel wire rope, a hook and a movable pressing plate, the steel wire rope extending from the crane motor, passing through a central hole of the movable pressing plate arranged at a taphole, and then connecting the hook and a slag ladle; a height travel switch arranged at the taphole, a normally closed contact of the height travel switch being mechanically linked with the movable pressing plate, and the normally closed contact being broken when the steel wire rope drives the movable pressing plate due to the shaking of the slag ladle; a contact signal acquisition module for acquiring the on-off state signal of the normally closed contact; a current signal acquisition module for acquiring real-time electric signals in the process of motor operation; and a control module electrically connected with the contact signal acquisition module and the current signal acquisition module, respectively, for receiving and processing the on-off state signal and the current signal, and calculating a stability comprehensive score. The present application provides a system and method for realizing the stability detection of the taphole crane at low cost.

Description

Technical Field

[0001] This invention relates to the field of slag discharge crane technology, and in particular to a slag discharge crane stability detection system and method. Background Technology

[0002] In mountainous bridge pile foundation engineering, bored pile muck removal cranes are widely used for vertical transportation of excavated soil during manual excavation. The raising and lowering of these cranes is controlled by the operator via a handheld button on an AC contactor. During the ascent of the muck bucket within the borehole, the narrow space inevitably causes collisions with the borehole wall, resulting in swaying of the wire rope and the bucket's swinging. Currently, there is a lack of effective means to monitor the operation of these cranes on construction sites. Operators can only visually judge whether the muck bucket is rising smoothly. When the muck bucket shakes violently or becomes stuck against the borehole wall, it cannot be detected in time, easily leading to safety accidents such as wire rope breakage, bucket overturning, or falling from the top. While existing technologies include load detection and sway monitoring solutions for ordinary cranes, the limited operating space, simple equipment, and small modification space of bored pile muck removal cranes make it difficult to install complex sensors. Therefore, there is an urgent need for a method that can detect stability using only the existing electrical signals of the equipment without adding dedicated sway or position sensors. Summary of the Invention

[0003] This invention provides a system and method for detecting the stability of a slag discharge crane, which can effectively solve the problems in the background art.

[0004] The present invention provides a slag discharge crane stability detection system, comprising: The crane body includes a crane motor, wire rope, hook and movable pressure plate. The wire rope extends from the crane motor, passes through the central hole of the movable pressure plate located at the slag discharge opening, and then connects to the hook and slag bucket. The height limit switch is located at the orifice. Its normally closed contact is mechanically linked to the movable pressure plate. When the wire rope moves the movable pressure plate due to the shaking of the slag bucket, the normally closed contact is broken. The contact signal acquisition module is used to acquire the on / off status signals of normally closed contacts; The current signal acquisition module is installed on the power supply line of the crane motor. It is used to acquire the real-time current signal during the operation of the motor and convert it into a voltage signal that is proportional to the primary current. The control module is electrically connected to the contact signal acquisition module and the current signal acquisition module, respectively, and is used to receive and process the on / off state signals and current signals, and calculate the overall stability score.

[0005] Furthermore, the movable pressure plate is a rectangular flat plate, which is hinged to the fixed bracket via a pivot, with one end being a free end. When the wire rope is centered, the movable pressure plate is in a balanced position, and the movable pressure plate does not contact the normally closed contact. When the wire rope sways and scrapes against the center hole of the movable pressure plate, the movable pressure plate swings around the pivot, and the free end moves towards the normally closed contact and pushes the contact, causing the normally closed contact to open.

[0006] Furthermore, the height limit switch is fixed to the slag discharge port by a mounting bracket; a reset spring is also provided between the movable pressure plate and the normally closed contact of the height limit switch.

[0007] This invention also provides a method for detecting the stability of a slag discharge crane, using the slag discharge crane stability detection system described above, comprising the following steps: S1: Continuously acquire the normally closed contact signal of the height limit switch at a set sampling frequency to form a contact state time series; S2: Continuously collect the current signal of the slag hoist motor at a set sampling frequency to form a current time series; S3: Triggered by the start signal of the slag removal crane's upward movement, the buffer for the contact state time series and current time series is started synchronously, and S4 to S8 are run once every set time during the upward movement. S4: Perform event detection on the cached contact state time series, identify the start time and duration of each contact disconnection event, and form a contact disconnection event sequence; S5: Perform transient spike detection on the cached current time series, identify spike pulses in the current waveform whose amplitude exceeds the dynamic threshold, record the peak time and amplitude of each spike pulse, and form a current spike event sequence. S6: Using the start time of each contact disconnection event as the anchor point, search for current spike events within the set time window after the anchor point, pair the successfully matched contact disconnection events with the current spike events as collision event pairs, record the delay time of each collision event pair, and calculate the average current spike density throughout the process. S7: Calculate the pairing rate, mean delay time, and coefficient of variation based on the delay time sequence of collision event pairs, and determine the validity of the data by combining the average current peak density throughout the process. S8: When the data validity is determined, calculate the rate of change of disconnection duration based on the contact disconnection duration sequence of the collision event pair, determine the comprehensive stability score, and execute the graded control response.

[0008] Further, in step S1, the acquisition of the on / off state signal of the normally closed contact of the height limit switch and the formation of the contact state time series are specifically as follows: The voltage signal across the normally closed contact of the height limit switch is collected. When the contact is closed, a low-level signal is collected and recorded as state value 0; when the contact is open, a high-level signal is collected and recorded as state value 1. Set a sampling frequency, continuously acquire the level signal at that sampling frequency, read the current level state each time, and arrange the corresponding state values ​​in chronological order to form a contact state time sequence; Software denoising is performed on the collected contact state time series: when a change in state value is detected, the state value is read again after a set delay to confirm. If the two reads are consistent, the state change is confirmed to be valid. If the two reads are inconsistent, it is determined to be a false change caused by contact jitter, and the previous state value is maintained unchanged.

[0009] Further, in step S2, acquiring the real-time current signal during the operation of the slag removal crane motor and forming a current time series specifically involves: Extract the current signal from one phase of the motor power supply line and convert the extracted current signal into an analog voltage signal that is proportional to the primary current. Set a sampling frequency, continuously acquire the voltage signal at that sampling frequency, and convert the voltage at each sampling point into a digital value through an analog-to-digital converter; Based on the pre-calibrated conversion factor, the digital quantity is converted into the corresponding effective current value; The continuously collected effective current values ​​are arranged in chronological order to form a current time series.

[0010] Furthermore, in step S5, detecting current spike events and calculating relative amplitudes specifically involves: Using a set time period as the sliding window length, calculate the arithmetic mean I of the current sample values ​​within the window for each sampling point. avg,i and sample standard deviation δ i Set the dynamic threshold to I avg,i +C·δ i Where C is a preset multiplier coefficient; For each sampling point in the current time series, if the current value exceeds the dynamic threshold of the window to which the sampling point belongs, and the current value of the sampling point is greater than the current values ​​of its adjacent forward sampling point and adjacent backward sampling point, then the sampling point is determined to be the current peak point. When multiple consecutive sampling points meet the peak determination criteria, only the sampling point with the largest amplitude is recorded as a current peak event, and the time t' of the peak value of each current peak is recorded. j and amplitude A j ; Extract the arithmetic mean of all current samples within the window preceding the appearance of the spike as the base value I. base,j Calculate the relative amplitude H of the current spike.j =A j / I base,j .

[0011] Further, in step S6, pairing the contact disconnection event with the current spike event and calculating the current spike density specifically involves: For the initial time t of each contact disconnection k Set the search window to [t] k +Δt min ,t k +Δt max The window boundary includes endpoint values; the peak current spike time t' is found within this window. j ; If there is one and only one current spike in the search window, then the contact disconnection event and the current spike event are paired as a collision event pair, and the delay duration L is recorded. k =t' j -t k ; If multiple current spikes exist within the search window, select the relative amplitude H. j The largest one is used as the pairing object; If no current spikes are found in the search window, the event is marked as unpaired. Each current spike can be paired with at most one contact disconnection event. When the search windows of multiple contact disconnection events overlap and point to the same current spike, the contact disconnection event with the shortest delay time will be paired first. If the delay time is the same, the contact disconnection events will be paired first according to the order of their occurrence time. After searching for all contact disconnection events, count the total number K of successfully paired event pairs. mt ; Calculate the average current peak density D throughout the entire process. peak =J / T eff T eff J represents the effective duration of the rising action, and J represents the total number of current spike events.

[0012] Further, in step S7, the pairing rate, mean delay duration, standard deviation of delay duration, and coefficient of variation of delay duration are calculated based on the delay duration sequence, and the data validity is determined as follows: If the number of disconnection events and the number of successful pairings meet the set minimum requirements, then proceed with the subsequent calculations. Extract the successfully paired delay sequence {L} k |k=1,2,...,K mt}, calculate the pairing rate P according to the following formula. mt Average delay time mu L , Delay duration standard deviation sigL and the coefficient of variation of delay duration (CV) L : P mt =K mt / K; mu L =(1 / K mt )×ΣL k ; sig L =sqrt((1 / (K mt -1))×Σ(L k -mu L ) 2 ); CV L =sig L / (mu L +eps); Where eps is a preset small positive number; When P mt Greater than or equal to the preset lower limit of the matching rate, CV L Less than the preset upper limit threshold for the coefficient of variation, and D peak Less than or equal to the preset peak density multiple threshold and (K / T) eff When the product of ) is used, the data is deemed valid and proceeds to step S8 to calculate the comprehensive stability score; otherwise, the data is deemed questionable or invalid.

[0013] Further, in step S8, the calculation of the rate of change of disconnection duration and the determination of the comprehensive stability score are specifically as follows: Successfully paired event pairs are divided into two subsets, the first and second parts, in chronological order. When K mt When the number is even, both the first and second segments contain K. mt / 2 event pairs, when K mt When the number is odd, discard the middle event pair, ensuring that each segment contains (K). mt -1) / 2 event pairs; Calculate the mean disconnection duration d of the preceding pairing event based on the contact disconnection duration sequence corresponding to the preceding pairing event. avf The mean disconnection duration d of the subsequent pairing event is calculated based on the contact disconnection duration sequence. avr ; Calculate the rate of change R of disconnection duration d =(d avr -d avf ) / (d avf +eps); Determine the current stability composite score (PSI): When R d When PSI is ≤0, PSI = 0; When 0 < Rd ≤R th At that time, PSI=(R d / R th )×S th ; When R d >R th At that time, PSI=100; Among them, R th S is the preset rate of change warning threshold. th The preset deceleration response triggers the initial score value.

[0014] The technical solution of this invention can achieve the following technical effects: This method utilizes the contact signals of the existing height limit switch of the slag removal crane and the motor power supply current signal for joint analysis. By pairing contact disconnection events with current spike events on the time axis, a data validity discrimination mechanism based on the collision physical conduction chain is established, eliminating false alarms caused by electrical interference. Data is only included in the evaluation when there is a stable time correspondence between the two signal channels; otherwise, invalid detection results are automatically masked. Based on this, the risk of overshoot is quantified by analyzing the changing trend of the contact disconnection duration (i.e., collision recovery time) during the ascent process. This method does not rely on subjective weighting coefficients in multi-index weighted synthesis; the evaluation result is directly determined by the measured physical quantities of the equipment's operating status. The entire detection process does not require the addition of new mechanical vibration, sway, or position sensors. It only needs to utilize the existing overload protection current transformer or standard current transformer to extract current signals, and realize the detection through signal acquisition and algorithm processing. It features low modification cost and wide applicability, and is suitable for the safety modification of various types of slag removal cranes for bored piles. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the slag discharge crane stability detection system of the present invention; Figure 2 This is a flowchart illustrating the method for detecting the stability of a slag discharge crane according to the present invention. Reference numerals in the attached diagram: 1. Wire rope; 2. Movable pressure plate; 3. Height limit switch; 4. Fixed bracket; 5. Return spring. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0019] This invention relates to a stability detection system for a slag discharge crane, such as... Figure 1 As shown, it includes the crane body, height limit switch 3, contact signal acquisition module, current signal acquisition module and control module.

[0020] The crane body includes a crane motor, a wire rope 1, a hook, and a movable pressure plate 2. The wire rope 1 extends from the crane motor, passes through the central hole of the movable pressure plate 2 located at the slag discharge opening, and connects to the hook and the slag bucket. The crane motor is a three-phase asynchronous motor, which drives the wire rope 1 to wind up and down via a winch. The hook is used to suspend the slag bucket, which is a cylindrical steel container used to load the slag excavated from the bottom of the hole.

[0021] The height limit switch 3 is located at the orifice, and its normally closed contact is mechanically linked to the movable pressure plate 2. When the wire rope 1 moves the movable pressure plate 2 due to the slag bucket's swaying, the normally closed contact is broken. The height limit switch 3 can cut off the rising circuit to achieve safety protection when the slag bucket is overturned. Furthermore, in this invention, the contact signal of the height limit switch 3 is also used as one of the signal sources for detecting the stability of the slag hoist.

[0022] The contact signal acquisition module is used to acquire the on / off status signal of normally closed contacts and convert the signal into a low-voltage level signal that the control module can recognize.

[0023] The current signal acquisition module is installed on the power supply line of the crane motor to collect the real-time current signal during motor operation and convert it into a voltage signal proportional to the primary current.

[0024] The control module is electrically connected to the contact signal acquisition module and the current signal acquisition module respectively. It is used to receive and process the on / off state signal and the current signal, calculate the comprehensive stability score and execute hierarchical control.

[0025] Preferably, the movable pressure plate 2 is a rectangular flat plate, which is hinged to the fixed bracket 4 via a rotating shaft, with one end being a free end. The fixed bracket 4 is installed at the slag discharge orifice, and the rotating shaft is horizontally positioned, allowing the movable pressure plate 2 to swing around the shaft in a vertical plane. A hole is provided in the center of the movable pressure plate 2, through which the steel wire rope 1 passes. The height limit switch 3 is fixed to the slag discharge orifice via a mounting bracket, and its normally closed contact is located on the swing path of the free end of the movable pressure plate 2. The contact of the height limit switch 3 faces the rotating end of the movable pressure plate 2. When the movable pressure plate 2 is in a balanced position with the wire rope 1 centered, the movable pressure plate 2 does not contact the normally closed contact. When the slag bucket shakes in the channel, the wire rope 1 shakes with the slag bucket. The deviated wire rope 1 contacts the inner wall of the central hole of the movable pressure plate 2 and pushes against the inner wall. Since one end of the movable pressure plate 2 is hinged by a rotating shaft, the movable pressure plate 2 swings around the rotating shaft as a whole. Its free end moves towards the normally closed contact of the height limit switch 3 and pushes against the contact, causing the normally closed contact to open.

[0026] Preferably, the height limit switch 3 is fixed to the slag discharge hole by a mounting bracket; a reset spring 5 is also provided between the movable pressure plate 2 and the normally closed contact of the height limit switch 3. The reset spring 5 is used to return the movable pressure plate 2 to the balance position after the wire rope 1 returns to center.

[0027] Preferably, the contact signal acquisition module includes an optocoupler isolation circuit and a signal conditioning circuit. The optocoupler isolation circuit is connected in parallel with the two normally closed contacts of the height limit switch 3. The function of the optocoupler isolation circuit is to convert the on / off state of the contacts into a low-voltage level signal that the control module can recognize, while realizing electrical isolation between the control circuit and the detection circuit, preventing high voltage from the control circuit from being introduced into the low-voltage detection circuit, and ensuring system safety.

[0028] When the normally closed contact is closed, it is recorded as state value 0; when the normally closed contact is open, it is recorded as state value 1. This low-voltage signal is filtered by the signal conditioning circuit and then output to the control module.

[0029] Preferably, the current signal acquisition module includes a current transformer and a signal conditioning circuit. The current transformer is installed on one phase of the three-phase power supply line of the crane motor using a non-contact installation method, without altering the original electrical circuit structure. The current transformer can be an existing overload protection current transformer or an additional standard current transformer. The current transformer proportionally transforms the large primary current into a small secondary current signal, which is then output to the signal conditioning circuit. The signal conditioning circuit includes an IV conversion circuit, an amplification circuit, and a filtering circuit, used to convert the current signal output by the current transformer into a 0-5V analog voltage signal. This analog voltage signal is linearly proportional to the primary current value and is output to the control module.

[0030] This invention also relates to a method for detecting the stability of a slag discharge crane, such as... Figure 2 As shown, the method comprises multiple steps from S1 to S8. Based on the aforementioned slag discharge crane stability detection system, it achieves quantitative detection and graded early warning control of the stability during the slag bucket's ascent process without adding dedicated swing sensors and position sensors, utilizing only the existing electrical signals of the equipment. The specific details of each step are as follows: S1: Acquire the on / off status signal of the normally closed contact of the height limit switch, and continuously collect this on / off status signal at a set sampling frequency to form a contact status time series. The contact signal reflects the change in wire rope tension. When the slag bucket shakes, causing the wire rope to drive the movable pressure plate, the normally closed contact opens. This signal is the first response to the collision event.

[0031] S2: Acquire the real-time current signal of the slag removal crane motor during operation, and continuously collect the current signal at a set sampling frequency to form a current time series. The spike pulses in the current signal indicate the sudden load change caused by the collision impact transmitted to the motor through the wire rope-drum-gearbox, and together with the contact signal, constitute two detectable features of the same collision event.

[0032] S3: Obtain the start signal for the lifting action of the slag removal crane. Using the start signal as a trigger, synchronously start the buffering of the contact state time series and current time series. During the lifting action, run S4 to S8 once every set time to determine the stability at this moment and determine which level of control strategy to adopt to avoid continuing to rise in an unstable state.

[0033] S4: Perform event detection on the buffered contact state time series, identify the start time of the contact transition from closed to open and the duration of each open event, forming a contact open event sequence. The contact open duration refers to the time length from the start of contact open to the moment of re-closing; this duration reflects the recovery speed of the wire rope tension after the slag bucket collides with the hole wall.

[0034] S5: Perform transient spike detection on the buffered current time series, identify spike pulses in the current waveform whose amplitude exceeds the dynamic threshold, record the peak time and amplitude of each spike pulse, and form a current spike event sequence. Current spikes are electrical response signals of collision impacts transmitted to the motor through the mechanical transmission chain, and their amplitude reflects the intensity of the collision impact.

[0035] S6: Using the start time of each contact disconnection event as the anchor point, search for current spike events within a set time window after the anchor point. Pair successfully matched contact disconnection events with current spike events as collision event pairs, record the delay time of each collision event pair, and calculate the average current spike density throughout the process. The delay time refers to the time interval between the contact disconnection time and the current spike time, determined by the physical conduction path of the mechanical impact. The pairing logic utilizes the sequential relationship of the collision signals on the time axis: contact disconnection occurs first, followed by the current spike.

[0036] S7: Calculate the pairing rate, mean delay time, and coefficient of variation based on the delay time sequence of collision event pairs. Combine this with the average current spike density throughout the process to determine the validity of the data. The pairing rate reflects the proportion of contact disconnection events that can be verified by current spikes; the coefficient of variation reflects the consistency of the delay time of each collision, used to determine whether there is a stable physical correlation between the two signal channels and to exclude invalid data caused by electrical interference.

[0037] S8: When the data validity is determined, the rate of change of disconnection duration is calculated based on the contact disconnection duration sequence of the collision event pair, the comprehensive stability score is determined, and a graded control response is executed. The score is mapped to four control levels: normal operation, warning alarm, deceleration control, and emergency braking, to achieve graded safety protection.

[0038] Preferably, in step S1, the acquisition of the on / off state signal of the normally closed contact of the height limit switch and the formation of a contact state time series are specifically as follows: The voltage signal across the normally closed contact of the height limit switch is collected. When the contact is closed, a low-level signal is collected and recorded as state value 0; when the contact is open, a high-level signal is collected and recorded as state value 1. The sampling frequency is set to 200Hz. The level signal is continuously acquired at this sampling frequency. Each time the current level state is read, the corresponding state value is arranged in chronological order to form a contact state time sequence.

[0039] Software noise reduction is performed on the collected contact state time series: when a change in state value is detected, the state value is read again after a set time of 5ms for confirmation. If the two reads are consistent, the state change is confirmed to be valid. If the two reads are inconsistent, it is determined to be a false change caused by contact jitter, and the previous state value is maintained unchanged.

[0040] Conventional mechanical contact switches, due to the springback effect of the metal contacts, generate multiple on / off bounces on the order of milliseconds during closing or opening. These bounces typically last 3–10 ms. Without noise reduction, each collision event might be mistakenly counted as multiple disconnections, leading to serious inaccuracies in subsequent pairing statistics. The "post-jump delay confirmation" method used in this section waits a set duration (e.g., 5 ms) after detecting a jump before reading the confirmation. This is equivalent to adding a 5 ms rectangular window to the timeline, filtering out all bounce pulses shorter than this window's duration, ensuring that only genuine steady-state jumps are counted in subsequent detections.

[0041] Preferably, in step S2, acquiring the real-time current signal during the operation of the slag removal crane motor and forming a current time series specifically involves: On one phase of the three-phase power supply line for the motor, the existing overload protection current transformer or a non-contact current transformer is used to extract the current signal, and the extracted current signal is converted into a 0-5V analog voltage signal that is proportional to the primary current.

[0042] The sampling frequency is set to 1000Hz, and the voltage signal is continuously acquired. The voltage value at each sampling point is converted into a digital value using a 12-bit analog-to-digital converter. Based on pre-calibrated conversion factors, the digital value is converted into the corresponding instantaneous current value, in amperes (A). The continuously acquired instantaneous current values ​​are arranged in chronological order to form a current time series.

[0043] The 1000Hz sampling frequency was selected based on the following: the typical rising edge duration of the current spike of the pile foundation slag removal crane motor is 5–30ms. A 1000Hz sampling rate ensures that each spike rising edge is covered by 5–30 sampling points, which is sufficient to accurately capture the peak time and amplitude of the spike. The rated current on the primary side of the current transformer corresponds to a voltage value between 0 and 5V, and the specific conversion factor is determined by on-site calibration. The 12-bit ADC has a resolution of approximately 1.22mV in the 0–5V range, which meets the accuracy requirements for current spike detection.

[0044] The current and voltage signals are timestamped using the same clock source to ensure that the two signals are strictly synchronized on the time axis when subsequent events are paired.

[0045] Preferably, in step S3, obtaining the rising action start signal and starting the data buffer specifically involves: The system acquires and continuously monitors the engagement status signal of the rising contactor. When the engagement status signal changes from disengagement to engagement, the rising contactor is determined to be engaged, and the current time is recorded as the start time of the rising action. Using this start time as the trigger point, a buffer is activated, and the contact status time series and current time series acquired after this time are synchronously written into the buffer. Simultaneously, a timer is started to record the engagement duration from the start time. When the engagement status signal of the rising contactor changes from engaged to disengagement, the rising contactor is determined to be disengaged, the current time is recorded as the end time of the rising action, the buffer is stopped, and all data from the start time to the end time is treated as a complete dataset for a single rising action.

[0046] If the rising contactor status signal completely fails (no changes in engagement or disengagement can be detected), the system will not initiate the comprehensive stability scoring path (because the start and end boundaries of the data buffer cannot be determined). In this case, the system only retains the full-scale current self-diagnostic criterion. This criterion does not require the rising or falling contactor status signal as a trigger condition in a complete failure scenario. If the full-scale current is detected for more than 5 seconds, it is determined to be a current channel fault, and the rising speed is limited to 50% of the rated speed. "Full-scale current" refers to the primary current of the current transformer reaching 1.2 times its rated current, corresponding to an analog voltage output of 5V.

[0047] When the termination signal of the rising contactor is lost but the start signal is known, the following alternative termination criterion is adopted: if the lowering contactor is detected to be engaged (indicating that the operator has switched to the lowering operation), or if no new contact disconnection event is detected for more than a preset time (e.g., 10s) and the current base value returns to the no-load level (indicating that the slag bucket has stopped rising), then it is determined that the current rising action has actually ended, the buffering is stopped, and the actual end time of this determination is used as the termination time of the rising action for subsequent calculations.

[0048] If both the lifting and lowering contactors are detected to be engaged simultaneously, the lifting and lowering power supply circuits will be immediately disconnected, a fault will be output, and crane operation will be prohibited until manual reset. The system will immediately stop buffering and discard any incomplete lifting data, resuming operation only when the lifting contactor is detected engaging again.

[0049] After the system is powered on or restarted, the cache is cleared. If the restart occurs during ascent, the system defaults to a conservative speed-limiting mode, limiting the ascent speed to 50% of the rated speed until the data cache for the next complete ascent is completed and the data validity is verified, at which point the original speed control is restored.

[0050] Preferably, in step S4, identifying the contact disconnection event specifically involves: Traverse the contact state time series in the buffer, starting from the start of the rising action, and detect the change of state value at each sampling point. When a state value changes from 0 to 1, record this moment as the start time t of a contact disconnection. k The continuous disconnection state continues until the state value changes from 1 back to 0. This moment is recorded as the contact recovery and closure moment. The difference between the recovery and closure moment and the disconnection start moment is calculated as the duration d of this contact disconnection event. k The unit is seconds (s). The time (t) for each contact disconnection event... k and d k Arranged in chronological order as the contact disconnection event sequence {t k ,d k |k=1,2,...,K}, where K is the total number of contact disconnection events.

[0051] If the contact is still open when the upward movement ends, the contact opening event is considered an incomplete event and is not included in the contact opening event count K. However, the contact status data during this period can still be used to calculate the safety redundancy condition of the contact opening ratio.

[0052] To avoid accidental contact disconnection caused by mechanical vibration and electrical interference during the moment the rising contactor engages, contact disconnection events detected within the first 0.5 seconds after the rising contactor engages are marked as start-up transient events and are not included in the contact disconnection event count K (these events are also not used in subsequent pairing calculations).

[0053] When the interval between the start times of two consecutive contact disconnections is less than 50ms, the two events are merged into one event. The earlier start time of the merged disconnection is taken, and the duration of the merged event is the difference between the second recovery closure time and the first disconnection start time, i.e., the total duration from the first disconnection to the final recovery closure. The principle of this merging is that after a collision between the slag bucket and the orifice wall, the wire rope may experience multiple rebound vibrations within 20-50ms, causing the contacts to repeatedly open and close. A merging threshold of 50ms effectively filters out multiple rebounds within the same collision event, avoiding miscounting a single collision as multiple events, and also avoiding the mismerging of independent collisions with intervals greater than 50ms. The duration after merging is uniformly based on the total duration of the entire process and remains unchanged in subsequent comparisons of the preceding and following segments. Therefore, the merging method does not affect the rate of change R of the subsequent disconnection duration. d The trend judgment is unaffected.

[0054] Preferably, in step S5, detecting current spike events and calculating relative amplitudes specifically involves: Using a sliding window length of 0.2s, with the end of the window aligned with the previous sampling point, the arithmetic mean I of the current sampled values ​​within the window is calculated for each sampling point. avg,i and sample standard deviation δ iThe duration of current spikes is approximately 0.1–0.3 seconds. A window length of 0.2 seconds ensures that both normal fluctuation data and spike data are included within the window for calculating the effective mean. The equivalent step size for updating each sampling point is 1 ms, which balances detection sensitivity with computational efficiency. No spike detection is performed within the first 0.2 seconds after the current sequence begins; formal judgment begins after the window data has accumulated completely.

[0055] Set the dynamic threshold to I avg,i +C·δ i C is a preset multiplier, usually taken as 3.5, which is an engineering value that achieves a good balance between false alarm rate and false negative rate in the field measured data.

[0056] For each sampling point in the current time series, if its current value exceeds the dynamic threshold of the window to which the sampling point belongs, and the current value of the sampling point is simultaneously greater than the current values ​​of its adjacent forward and backward sampling points, then the sampling point is determined to be a current peak. When multiple consecutive sampling points meet the peak determination criteria, only the sampling point with the largest amplitude is recorded as a current peak event, and the time t' of each current peak is recorded. j and amplitude A j .

[0057] For each identified current spike, the arithmetic mean of all current samples within the window (0.2s) preceding the spike is extracted as the base value I. base,j Calculate the relative amplitude H of the current spike. j =A j / I base,j .

[0058] The base current value varies under different load conditions, so A can be used directly. j It is impossible to compare the intensity of collision impacts across different operating conditions. j By normalizing by dividing by a base value, the degree of impact from collisions under different load levels can be evaluated uniformly. j The larger the value, the more severe the impact on the motor load from the collision. During subsequent pairing, if multiple current spikes exist within a search window, select H. j The largest one is used as the pairing object.

[0059] To avoid interference from the instantaneous inrush current (typically several times the rated current) generated during motor startup, current spike event detection is paused for the first 0.5 seconds after the rising contactor engages, but the sliding window data continues to accumulate. After 0.5 seconds, once the startup surge has attenuated to a stable region, the spike detection is then performed normally according to the same procedure.

[0060] Preferably, in step S6, pairing the contact disconnection event with the current spike event and calculating the current spike density specifically involves: For the initial time t of each contact disconnection k Set the search window to [t] k +Δt min ,t k +Δt max The window boundaries include endpoint values ​​(i.e., the delay duration equals Δt). min or Δt max (Even at this time, it is still considered a valid pairing). The selection of the search window boundary is based on the following: When a real collision occurs, the movable pressure plate is first driven by the wire rope, and the contact breaks; subsequently, the mechanical impact is transmitted to the motor shaft through the wire rope-drum-gearbox, causing a current spike. Therefore, the current spike should lag behind the contact breaking. Δt min =0.02s is the shortest physical delay from contact disconnection to the appearance of a detectable spike in motor current, Δt max =0.35s, which is the maximum time required for the complete mechanical transmission and electrical response of the impact through the wire rope-drum-gearbox-motor shaft. For Δt min and Δt max The optimal value can be determined experimentally. In this method, Δt can be determined through actual measurements using five different crane models. min =0.02s, Δt max =0.35s.

[0061] Search for the peak value of the current spike t' in the search window. j ; If there is one and only one current spike in the search window, then the contact disconnection event and the current spike event are paired as a collision event pair, and the delay duration L is recorded. k =t' j -t k ; If multiple current spikes exist within the search window, select the relative amplitude H. j The largest one is used as the pairing object; If no current spikes are found in the search window, the event is marked as unpaired.

[0062] To prevent the same current spike from being repeatedly paired, a deduplication rule is introduced: Each current spike can be paired with at most one contact disconnection event. When the search windows of multiple contact disconnection events overlap and point to the same current spike, the contact disconnection event with the shortest delay time will be paired first. If the delay time is the same, the contact disconnection events will be paired first according to their occurrence time (earlier events take precedence). After searching for all contact disconnection events, count the total number K of successfully paired event pairs. mt .

[0063] Calculate the average current peak density D throughout the entire process.peak =J / T eff T eff J represents the effective duration of the rising action, and J represents the total number of current spike events.

[0064] Current spike density D peak This reflects the frequency of current spikes occurring per unit effective time. This density value is used to help determine data quality; if D... peak Much greater than the contact disconnection event density K / T eff This indicates that there are numerous interference spikes in the current signal that are unrelated to the collision, which will affect the reliability of subsequent pairing. peak The larger the value of D, the more interference components may be present in the current signal. peak >3×(K / T) eff When the current spike density deviates significantly from the collision event density, it indicates that there are a large number of disturbances in the current signal that are unrelated to the collision, and the reliability of the data is questionable.

[0065] To unify J and T total The statistical caliber defines the effective duration T. eff =max(T total -0.5s,0), ​​where 0.5s is the start-up shielding duration. When T eff When ≤0, D is not calculated. peak T total The acquisition method is as follows: When the engagement and disengagement status signals of the rising contactor recorded in step S3 are both complete, T total Take the difference between the current cycle time and the start time of the ascent; when the start time signal is known and only the end time signal is lost (at which point the ascent has ended according to the backup termination criterion in step S3), use the estimation formula T. total =t K -t1+(t K -t1)×0.1, where t1 is the initial time of the first contact disconnection, t K The starting time of the last contact disconnection is extended by 10% to cover boundary events. This 10% is a conservative upper limit determined based on the maximum proportion of deviations between the first and last contact disconnection events and their starting and ending times from multiple actual measurements. When the starting time signal is also lost, or when K < 2 (there are not enough collision events for density comparison), T cannot be effectively estimated. total D is not calculated at this time. peak The system will not proceed to the subsequent comprehensive stability score calculation; Safety redundancy condition 1 cannot operate because the start time of the rising action cannot be determined; Safety redundancy condition 2 can still operate independently; Only the full-scale current self-diagnosis criterion is retained (see step S3).

[0066] Preferably, in step S7, the pairing rate, mean delay duration, standard deviation of delay duration, and coefficient of variation of delay duration are calculated based on the delay duration sequence, and the data validity is determined as follows: Determining data validity cannot proceed directly to statistical calculations from any pairing state; various boundary cases must be handled first. This is because the mathematical calculation of statistics (pairing rate, mean, standard deviation, coefficient of variation) depends on a sufficient sample size. When the total number of disconnected events K=0, the denominator of the pairing rate is zero, rendering it meaningless; when the number of successfully paired events K... mt When K = 0, the delay time sequence is empty, and no statistics can be calculated; when K mt When = 1, the denominator in the formula for calculating the standard deviation of the delay time is (K). mt -1) = 0, which is mathematically undefined. Therefore, before calculating the statistics, the above boundary cases must be classified and processed to ensure that the method can give a clear output state under various extreme data conditions, rather than crashing or producing meaningless results. The overall logic of boundary case processing is: no collision occurs when K = 0; K mt When =0, pairing fails completely; K mt When the value is 1, the sample size is insufficient. None of the three methods can support reliable calculation of the stationarity comprehensive score. However, because the safety redundancy condition operates independently without relying on data validity judgment, the safety protection in these boundary cases is still guaranteed by the safety redundancy condition. In these cases, the stationarity comprehensive score path skips directly instead of exiting with an error.

[0067] The specific boundary handling is as follows: When K=0 (no contact disconnection event), the scoring calculation step is skipped, and the score PSI is directly set to 0; however, the safety redundancy condition still operates independently, and the final PSI value is taken according to the S8 rule, which is the higher value. If the safety redundancy condition is not triggered, "No collision throughout" is output. When K=0 and J>0 (i.e., there is a current spike but no corresponding contact disconnection), "No collision throughout but there is a current spike, it is recommended to check if the contact signal is normal" is output, without directly ruling out contact jamming faults.

[0068] Only if K > 0 and K mt When the value is 0 (there are disconnected contacts but no successful pairing), both the stability comprehensive scoring path and the data validity judgment are invalid, and a maintenance prompt is directly output without proceeding to the subsequent stability comprehensive scoring calculation; however, the safety redundancy condition still operates independently.

[0069] When K mt When σ = 1 (only 1 paired event), the standard deviation sig cannot be calculated. L and coefficient of variation CV L If insufficient data is detected, an "Insufficient data" message will be output, and the system will not proceed to the subsequent comprehensive stability score calculation; however, the safety redundancy condition will still operate independently.

[0070] Only if K > 0 and K mt If the result is ≥2, continue with the following calculation.

[0071] Extract the successfully paired delay sequence {L} k |k=1,2,...,K mt}, calculate the pairing rate P according to the following formula. mt Average delay time mu L , Delay duration standard deviation sig L and the coefficient of variation of delay duration (CV) L : Pairing rate P mt =K mt / K; where K mt K represents the total number of successfully paired event pairs; K represents the total number of contact disconnection events. The calculation logic of the pairing rate stems from the physical nature of the collision process: a genuine collision event should produce a pairing of a contact disconnection event and a current spike event in time. If a contact disconnection event cannot find a corresponding current spike within a preset window, it cannot be verified as a genuine collision. mt =K mt / K directly measures the proportion of contact disconnection events that are verified as genuine collisions. P mt The larger the value, the more likely the two signal channels reflect the same physical process, and the higher the data reliability. mt When the value is less than 0.6, a large number of contact disconnection events have no corresponding spikes, and the data is unreliable.

[0072] Average delay duration mu L =(1 / K mt )×ΣL k ; mu L For k=1 to k=K mt The corresponding arithmetic mean of delay durations, in seconds; L k mu represents the k-th delay duration, measured in seconds. L This represents an estimate of the average conduction time of the impact from the point of occurrence to the motor, manifesting as a current spike. The mechanical conduction path of the same device is fixed, and the delay time of each impact should stably fluctuate around a central value, mu. L It is an estimate of the central value.

[0073] Delay duration standard deviation sig L =sqrt((1 / (K mt -1))×Σ(L k -mu L ) 2 ); If the delay times of each collision are similar, the mechanical transmission path is stable; otherwise, it is unstable. L It is a measure of the degree to which the delay time deviates from the mean, and the denominator uses (K) mt -1) To provide an unbiased estimate of the population standard deviation. sig L The smaller the value, the better the repeatability and consistency of the collision physics process. L The larger the value, the more likely there is a change in the impact transmission path or mispairing of non-collision spikes.

[0074] Delay duration coefficient of variation (CV) L =sig L / (mu L +eps); Where eps is a preset small positive number, eps=10 -6 's' is used to prevent division by zero errors. sig L The absolute value is subject to mu L The magnitude of the impact, simply using sig L It is impossible to compare the degree of dispersion across different operating conditions. CV L By dividing by mu L Normalization is performed to enable it to determine whether the physical association is reliable. CV L The smaller the value, the more consistent the transmission path, and the more reliable the data. CV L When the value is ≥0.3, the physical conduction path is unstable, and the data is questionable.

[0075] When P mt Greater than or equal to the preset lower limit of the matching rate (P) mt ≥0.6), CV L Less than the preset upper limit threshold for coefficient of variation (CV) L <0.3), and D peak Less than or equal to the preset peak density multiple threshold and (K / T) eff The product of (D) peak ≤3×(K / T eff If the data is deemed valid, proceed to step S8 to calculate the overall stability score; otherwise, the data is deemed questionable or invalid.

[0076] The selection criteria for each threshold are as follows: P mt A value ≥0.6 indicates that over 60% of contact disconnection events can achieve current spike verification (the lower limit of the matching rate for effective operating conditions in field measurement data is approximately 0.55–0.65, with 0.6 taken as the universal lower limit); CV LA value <0.3 indicates that the relative dispersion of the delay duration is less than 30% (a coefficient of variation below 0.3 is generally considered to have good consistency in engineering statistics); when both conditions are met, the physical correlation between the contact signal and the current signal is verified to be stable and reliable. The above threshold values ​​are example values; specific values ​​can be calibrated according to the actual working conditions and do not constitute a limitation on the protection range.

[0077] In step S8, when the data validity determination is passed, the rate of change of disconnection duration is calculated and the comprehensive stability score is determined as follows: Successfully paired event pairs are divided into two subsets, the first and second parts, in chronological order. When K mt When the number is even, both the first and second segments contain K. mt / 2 event pairs; when K mt When the number is odd, discard the middle event pair, ensuring that each segment contains (K). mt -1) / 2 event pairs, the midpoint of the time has the least impact on the mean of the first segment and the mean of the second segment, and is the most unbiased discarding method.

[0078] Calculate the mean disconnection duration d of the preceding pairing event based on the contact disconnection duration sequence corresponding to the preceding pairing event. avf The mean disconnection duration d of the subsequent pairing event is calculated based on the contact disconnection duration sequence. avr ; Calculate the rate of change R of disconnection duration d =(d avr -d avf ) / (d avf +eps); Contact disconnection duration d k This refers to the time between the initial opening and the reopening of the normally closed contact of the height limit switch. It represents the duration of the contact opening, the time during which the normally closed contact remains open while the movable pressure plate is laterally pushed by the wire rope. It reflects the speed at which the wire rope recovers from swaying or tension. During the ascent of the slag bucket within the channel, the length of the wire rope between the movable pressure plate and the slag bucket continuously shortens, thus reducing the lateral movement allowance provided by the orifice of the movable pressure plate to the wire rope. Under the same impact intensity, the wire rope requires a longer time to detach from the pushing state on the movable pressure plate; therefore, d... k It should show an increasing trend in the later stages of an upward trend. R d =(d avr -d avf ) / d avf d was quantified k The higher the relative rate of change as the height increases, the faster the working margin of the moving pressure plate is consumed, and the higher the risk of overshooting.

[0079] Determine the overall stability score (PSI): When Rd When PSI is ≤0, the movable pressure plate has sufficient space and PSI=0, indicating normal operation; When 0 < R d ≤R th At that time, PSI=(R d / R th )×S th This indicates that the working margin of the active pressure plate is gradually being consumed; When R d >R th At this time, PSI=100 indicates that the slag bucket is close to the top and emergency braking needs to be triggered; Among them, R th The preset rate of change warning threshold is set to 0.25; S th The preset deceleration response trigger initial score value is set to 60.

[0080] The overall stability score PSI is entirely determined by R d Decision. R d The calculation assumes that the data has been processed by P. mt CV L and D peak Verification refers to the pure physical quantity after eliminating interference. d When ≤0, the movable pressure plate has sufficient space; when 0 < R d When R is ≤0.25, the risk increases linearly to the warning boundary; d When the value is greater than 0.25, the space for the movable pressure plate is severely insufficient. The final PSI value is the higher of the overall stability score and the safety redundancy condition (when the overall stability score is valid) to ensure that the safety protection does not fail due to delays in the overall stability score path.

[0081] PSI=0 indicates normal operation; 0 < PSI < 60 indicates that the working margin of the movable pressure plate is gradually being consumed; PSI=60 indicates that the working margin of the movable pressure plate has been consumed to the warning boundary, triggering deceleration; PSI=100 indicates that the slag bucket is approaching the top, triggering emergency braking. When R d Exceeding the warning threshold R th At this point, the working margin of the movable pressure plate has been depleted to a critical state. The system then switches from a warning state to an emergency braking state, therefore the PSI is directly assigned a value of 100, without further linear interpolation. This threshold-triggered design ensures immediate braking measures are taken when a dangerous condition is reached, avoiding delays in the scoring transition zone.

[0082] After calculating the overall stability score (PSI), a graded control response can be implemented based on the PSI value. When PSI=0, no warning is triggered, and the crane operates normally; When 0 < PSI < 60, an intermittent audible and visual warning is issued, and the crane continues to operate; When PSI=60, an urgent audible and visual alarm is issued and a deceleration control signal is output, reducing the ascent speed to 50% of the rated speed. When PSI=100, the power supply circuit of the ascent contactor is immediately disconnected to implement emergency braking, while a continuous audible and visual alarm is issued, and the descent contactor control function is retained.

[0083] The lifting contactor must not be engaged again within 3 seconds after emergency braking. Based on on-site braking tests, the winch's inertial rotation to a complete stop takes approximately 0.8–1.1 seconds; the wire rope tension release takes approximately 0.5–0.8 seconds; adding an operator reaction margin, this is taken as 3 seconds. After the 3-second timeout, if the lifting button is still pressed, the lifting contactor must not engage; it must be pressed again by the operator after releasing the button before it can be started.

[0084] When data validity fails, calculate the frequency F of unpaired events. unmt =(KK mt ) / T eff If F unmt >2.0s -1 If the equipment is found to have abnormal vibration or electrical interference, a maintenance prompt will be output, and the ascent speed will be limited to 50% of the rated speed; if F unmt ≤2.0s -1 The disturbance was determined to be an occasional event, and the results of this test were ignored.

[0085] To ensure the safe operation of the crane, in addition to the calculated score of the Stability Performance Index (PSI), two safety redundancy conditions are set. When these conditions are met, the PSI is directly assigned a value of 100. 1. Starting 0.5 seconds after the start of the ascent action, each 1-second window is defined. When the contact disconnection rate of any three consecutive non-overlapping windows is greater than 0.7, the PSI is directly assigned a value of 100. The contact disconnection rate is defined as the total number of sampling points in the window where the contact is in the disconnected state divided by the total number of sampling points in the window. This condition is based on dynamic trends rather than a single instantaneous state, which can effectively eliminate false alarms caused by instantaneous jitter in a single window.

[0086] 2. When the moving average value of the current load factor λ avg >1.15 and the sliding standard deviation of current fluctuation δ avg <0.03×I rated At that time, PSI is directly assigned a value of 100. λ avg δ is the ratio of the arithmetic mean of the current samples to the rated current of the motor. avg I is the standard deviation of the current sample. rated This is the rated current of the motor. Where λ avg and δ avgAll calculations use a sliding window, with a window length of 1 second and a step size of 0.1 seconds. The end of the window is aligned with the previous sampling point of the current sampling point. The ratio of the arithmetic mean of the current samples within the window to the rated current of the motor is λ. avg The sample standard deviation of the current samples within the window is δ. avg .

[0087] The aforementioned safety redundancy conditions operate independently during their applicable period and are unaffected by data validity assessment results. Even if the collision detection path fails due to interference, these safety redundancy conditions can still operate independently, providing basic safety protection for the equipment.

[0088] Taking the manual excavation of the pylon foundation for a highway bridge as an example, the diameter of the bored pile is 2.5m, and the designed pile length is 68m. A slag removal crane is used for vertical transportation of the excavated soil. The crane motor has a rated power of 7.5kW and a rated current I... rated =18A. The crane's lifting speed is approximately 0.3 m / s, and the theoretical time for lifting a full bucket of excavated soil from the bottom of the hole (depth 68 m) to the top (depth 0 m) is approximately 227 seconds. Due to emergency braking triggered later in the lifting process, the actual total lifting time T total =212.5s.

[0089] Throughout the entire process of the slag bucket rising from the bottom to the top of the hole, intelligent stability detection is performed according to steps S1 to S8 as described above. The various thresholds used in this embodiment (0.2s sliding window, 3.5 times dynamic threshold, 0.6 lower limit of pairing rate, 0.3 upper limit of coefficient of variation, 0.25 warning threshold of rate of change, 0.7 threshold of contact disconnection ratio, 1.15 times overload threshold, and 0.03 times fluctuation threshold) were optimized and determined after statistical analysis of the operating data of the slag removal crane for bored piles. The measured data covered 5 different crane models and 3 different hole depth conditions (30m, 50m, and 68m), totaling more than 200 sets of valid operating data.

[0090] Data Acquisition and Caching: The contact status signal and current signal are collected according to steps S1 and S2 respectively. Step S3 activates the data buffer when the lifting contactor is detected to be engaged. In this embodiment, the actual total lifting action time T... total =212.5s, the buffer contains a total of 42,500 contact state sampling points (212.5s × 200Hz) and 212,500 current sampling points (212.5s × 1000Hz). The buffer adopts a ring buffer structure, retaining only the most recent data required for the current sliding calculation; when a complete statistical analysis is needed, the necessary data is then transferred to non-volatile memory.

[0091] Contact disconnection event detection: Traversing the contact state time series, a total of K=48 contact disconnection events were detected. The first contact disconnection started at t1=18.2s (the slag bucket first collides with the hole wall 18.2s after leaving the hole bottom), and the 48th contact disconnection started at t... 48 =208.0s. The average interval between adjacent events of the 48 events is approximately 4.04s, the maximum interval is approximately 7.8s, and the minimum interval is approximately 1.8s (greater than the 50ms merging threshold, no merging is required).

[0092] Current spike event detection: With a window length of 0.2s, the window ends aligned with the previous sampling point of the current sampling point, and the dynamic threshold is calculated point by point. The dynamic threshold is set to I. avg,i +3.5δ i A total of J=52 current spike events were detected. The relative amplitude H of 10 typical current spikes was calculated. j The values ​​are: 0.23, 0.19, 0.34, 0.28, 0.41, 0.18, 0.35, 0.22, 0.29, 0.32, with an average of 0.28, indicating that the slag bucket continuously collides with the orifice wall during its ascent within the channel. (According to D...) peak =J / T eff Calculate, T eff =212.5-0.5=212.0s, D peak =52 / 212.0≈0.245s -1 Contact disconnection event density K / T eff =48 / 212.0≈0.226s -1 D peak =0.245<3×0.226=0.679, indicating that the current peak density is within a reasonable range and there is no significant interference.

[0093] Perform pairing: For 48 contact disconnection events, set the search window [t] k +0.02s,t k Search for current spikes within +0.35s. Successfully paired K. mt =43 pairs, pairing rate P mt =43 / 48≈0.896, which is greater than the validity lower limit threshold of 0.6. All 5 unpaired contact disconnection events were without current spikes within the search window; when multiple current spikes existed within the window, the one with the largest relative amplitude was selected for pairing according to the rules.

[0094] The delay time L of the 43 successfully paired collision event pairs k The distribution is as follows: minimum value 0.034s, maximum value 0.148s, mean value mu. L =0.087s, standard deviation sig L=0.021s. Coefficient of variation (CV) of delay duration. L =sig L / (mu L +eps)=0.021 / 0.087≈0.241. CV L =0.241 < 0.3, the physical conduction path is stable, and the data is valid.

[0095] Data validity assessment: K=48>0, K mt =43≥2, proceed to calculation. P mt =0.896≥0.6, CV L =0.241<0.3, D peak =0.245≤3×(K / T eff The result is 0.679, all three conditions are met, and the data is considered valid.

[0096] Step S8 calculates the rate of change of disconnection duration and the comprehensive score of stability: 43 pairs of collision events are arranged in chronological order, K mt =43 is an odd number. Each segment contains (43-1) / 2=21 event pairs. The first segment takes the 1st to the 21st pair, and the second segment takes the 23rd to the 43rd pair. The 22nd pair is discarded.

[0097] The average duration d of the contact disconnection for the first 21 events avf =0.031s. The average duration d of the contact disconnection for the subsequent 21 events. avr =0.074s. Rate of change R of disconnection duration. d =(0.074-0.031) / (0.031+10 -6 R = 0.043 / 0.031001 ≈ 1.387. d =1.387>0.25, PSI=100.

[0098] Control Response: PSI=100, indicating a dangerous situation. Immediately disconnect the power supply circuit to the lifting contactor to implement emergency braking. After braking, the slag bucket stops rising, but the lowering contactor control function remains, continuously issuing audible and visual alarms. The lifting contactor must not be engaged again within 3 seconds. A fault record is generated: the trigger time is approximately 212.5 seconds after the lifting action begins, and the trigger reason is R. d (1.387) exceeds the warning threshold of 0.25, P mt =0.896, CV L =0.241.

[0099] Verification conclusion: In this embodiment, the average duration of the initial contact disconnection is only 0.031s, indicating that the wire rope can quickly reset after slack in the early stage of ascent; the duration of the later stage is 0.074s, indicating that the wire rope needs a longer time to disengage from the pushing action on the movable pressure plate after slack in the later stage of ascent. d =1.387, far exceeding the warning threshold of 0.25, indicating a severe lack of working margin for the movable pressure plate. The slag bucket was close to the top, therefore PSI=100 triggered emergency braking. The slag bucket rose and triggered braking at 212.5s, at which time the slag bucket was approximately 4.25m from the top of the hole; the stopping position after braking was approximately 3.9-4.1m.

Claims

1. A stability detection system for a slag discharge crane, characterized in that, include: The crane body includes a crane motor, a wire rope, a hook, and a movable pressure plate. The wire rope extends from the crane motor, passes through the central hole of the movable pressure plate located at the slag discharge opening, and then connects to the hook and the slag bucket. A height limit switch is installed at the orifice. Its normally closed contact is mechanically linked to the movable pressure plate. When the wire rope moves the movable pressure plate due to the shaking of the slag bucket, the normally closed contact is broken. The contact signal acquisition module is used to acquire the on / off status signals of normally closed contacts; The current signal acquisition module is installed on the power supply line of the crane motor. It is used to acquire the real-time current signal during the operation of the motor and convert it into a voltage signal that is proportional to the primary current. The control module is electrically connected to the contact signal acquisition module and the current signal acquisition module, respectively, and is used to receive and process the on / off state signal and the current signal, and calculate the overall stability score.

2. The slag discharge crane stability detection system according to claim 1, characterized in that, The movable pressure plate is a rectangular flat plate that is hinged to a fixed bracket via a pivot, with one end being a free end. When the wire rope is centered, the movable pressure plate is in a balanced position and does not contact the normally closed contact. When the wire rope sways and scrapes against the center hole of the movable pressure plate, the movable pressure plate swings around the pivot, and the free end moves toward the normally closed contact and pushes the contact, causing the normally closed contact to open.

3. The slag discharge crane stability detection system according to claim 2, characterized in that, The height limit switch is fixed to the slag discharge port by a mounting bracket; a reset spring is also provided between the movable pressure plate and the normally closed contact of the height limit switch.

4. A method for detecting the stability of a slag discharge crane, characterized in that, Using the slag discharge crane stability detection system as described in any one of claims 1 to 3, the steps include: S1: Continuously acquire the normally closed contact signal of the height limit switch at a set sampling frequency to form a contact state time series; S2: Continuously collect the current signal of the slag hoist motor at a set sampling frequency to form a current time series; S3: Triggered by the start signal of the slag removal crane's upward movement, the buffer for the contact state time series and current time series is started synchronously, and S4 to S8 are run once every set time during the upward movement. S4: Perform event detection on the cached contact state time series, identify the start time and duration of each contact disconnection event, and form a contact disconnection event sequence; S5: Perform transient spike detection on the cached current time series, identify spike pulses in the current waveform whose amplitude exceeds the dynamic threshold, record the peak time and amplitude of each spike pulse, and form a current spike event sequence. S6: Using the start time of each contact disconnection event as the anchor point, search for current spike events within the set time window after the anchor point, pair the successfully matched contact disconnection events with the current spike events as collision event pairs, record the delay time of each collision event pair, and calculate the average current spike density throughout the process. S7: Calculate the pairing rate, mean delay time, and coefficient of variation based on the delay time sequence of collision event pairs, and determine the validity of the data by combining the average current peak density throughout the process. S8: When the data validity is determined, calculate the rate of change of disconnection duration based on the contact disconnection duration sequence of the collision event pair, determine the comprehensive stability score, and execute the graded control response.

5. The method for detecting the stability of a slag discharge crane according to claim 4, characterized in that, In step S1, the acquisition of the on / off state signal of the normally closed contact of the height limit switch and the formation of the contact state time series are specifically as follows: The voltage signal across the normally closed contact of the height limit switch is collected. When the contact is closed, a low-level signal is collected and recorded as state value 0; when the contact is open, a high-level signal is collected and recorded as state value 1. Set a sampling frequency, continuously acquire the level signal at that sampling frequency, read the current level state each time, and arrange the corresponding state values ​​in chronological order to form a contact state time sequence; Software denoising is performed on the collected contact state time series: when a change in state value is detected, the state value is read again after a set delay to confirm. If the two reads are consistent, the state change is confirmed to be valid. If the two reads are inconsistent, it is determined to be a false change caused by contact jitter, and the previous state value is maintained unchanged.

6. The method for detecting the stability of a slag discharge crane according to claim 4, characterized in that, In step S2, acquiring the real-time current signal during the operation of the slag removal crane motor and forming a current time series specifically involves: Extract the current signal from one phase of the motor power supply line and convert the extracted current signal into an analog voltage signal that is proportional to the primary current. Set a sampling frequency, continuously acquire the voltage signal at that sampling frequency, and convert the voltage at each sampling point into a digital value through an analog-to-digital converter; Based on the pre-calibrated conversion factor, the digital quantity is converted into the corresponding effective current value; The continuously collected effective current values ​​are arranged in chronological order to form a current time series.

7. The method for detecting the stability of a slag discharge crane according to claim 4, characterized in that, In step S5, detecting current spike events and calculating relative amplitudes specifically involves: Using a set time period as the sliding window length, calculate the arithmetic mean I of the current sample values ​​within the window for each sampling point. avg,i and sample standard deviation δ i Set the dynamic threshold to I avg,i +C·δ i Where C is a preset multiplier coefficient; For each sampling point in the current time series, if the current value exceeds the dynamic threshold of the window to which the sampling point belongs, and the current value of the sampling point is greater than the current values ​​of its adjacent forward sampling point and adjacent backward sampling point, then the sampling point is determined to be the current peak point. When multiple consecutive sampling points meet the peak determination criteria, only the sampling point with the largest amplitude is recorded as a current peak event, and the time t' of the peak value of each current peak is recorded. j and amplitude A j ; Extract the arithmetic mean of all current samples within the window preceding the appearance of the spike as the base value I. base,j Calculate the relative amplitude H of the current spike. j =A j / I base,j .

8. The method for detecting the stability of a slag discharge crane according to claim 7, characterized in that, In step S6, pairing the contact disconnection event with the current spike event and calculating the current spike density specifically involves: For the initial time t of each contact disconnection k Set the search window to [t] k +Δt min ,t k +Δt max The window boundary includes endpoint values; the peak current spike time t' is found within this window. j ; If there is one and only one current spike in the search window, then the contact disconnection event and the current spike event are paired as a collision event pair, and the delay duration L is recorded. k =t' j -t k ; If multiple current spikes exist within the search window, select the relative amplitude H. j The largest one is used as the pairing object; If no current spikes are found in the search window, the event is marked as unpaired. Each current spike can be paired with at most one contact disconnection event. When the search windows of multiple contact disconnection events overlap and point to the same current spike, the contact disconnection event with the shortest delay time will be paired first. If the delay time is the same, the contact disconnection events will be paired first according to the order of their occurrence time. After searching for all contact disconnection events, count the total number K of successfully paired event pairs. mt ; Calculate the average current peak density D throughout the entire process. peak =J / T eff T eff J represents the effective duration of the rising action, and J represents the total number of current spike events.

9. The method for detecting the stability of a slag discharge crane according to claim 4, characterized in that, In step S7, the pairing rate, mean delay duration, standard deviation of delay duration, and coefficient of variation of delay duration are calculated based on the delay duration sequence, and the data validity is determined as follows: If the number of disconnection events and the number of successful pairings meet the set minimum requirements, then proceed with the subsequent calculations. Extract the successfully paired delay sequence {L} k |k=1,2,...,K mt }, calculate the pairing rate P according to the following formula. mt Average delay time mu L , Delay duration standard deviation sig L and the coefficient of variation of delay duration (CV) L : P mt =K mt / K; mu L =(1 / K mt )×ΣL k ; sig L =sqrt((1 / (K mt -1))×Σ(L k -mu L ) 2 ); CV L =sig L / (mu L +eps); Where eps is a preset small positive number; When P mt Greater than or equal to the preset lower limit of the matching rate, CV L Less than the preset upper limit threshold for the coefficient of variation, and D peak Less than or equal to the preset peak density multiple threshold and (K / T) eff When the product of ) is used, the data is deemed valid and proceeds to step S8 to calculate the comprehensive stability score; otherwise, the data is deemed questionable or invalid.

10. The method for detecting the stability of a slag discharge crane according to claim 9, characterized in that, In step S8, the calculation of the rate of change of disconnection duration and the determination of the comprehensive stability score are specifically as follows: Successfully paired event pairs are divided into two subsets, the first and second parts, in chronological order. When K mt When the number is even, both the first and second segments contain K. mt / 2 event pairs, when K mt When the number is odd, discard the middle event pair, ensuring that each segment contains (K). mt -1) / 2 event pairs; Calculate the mean disconnection duration d of the preceding pairing event based on the contact disconnection duration sequence corresponding to the preceding pairing event. avf The mean disconnection duration d of the subsequent pairing event is calculated based on the contact disconnection duration sequence. avr ; Calculate the rate of change R of disconnection duration d =(d avr -d avf ) / (d avf +eps); Determine the current stability composite score (PSI): When R d When PSI is ≤0, PSI = 0; When 0 < R d ≤R th At that time, PSI=(R d / R th )×S th ; When R d >R th At that time, PSI=100; Among them, R th S is the preset rate of change warning threshold. th The preset deceleration response triggers the initial score value.

Images