Overload detection method for concrete core drilling machine

Abstract

The invention relates to an overload detection method for a concrete drilling and coring machine, and the method comprises the steps: firstly obtaining multi-source data when the drilling and coring machine works, and enabling the data to be obtained in real time; and fusing the obtained multi-source data to obtain a comprehensive load characterization value, and calling a dynamic overload threshold model based on the operation parameters to calculate a real-time overload threshold. And then comparing the comprehensive load characterization value with a real-time overload threshold value to obtain a comparison result. And judging whether the working state of the core drilling machine is steady-state overload or instantaneous impact overload based on a comparison result. And different control is executed based on different overload types, so that the core drilling machine can be effectively protected when being subjected to instantaneous high harm, and the core drilling machine can be relieved and kept in a high continuous working state when being subjected to slow overload.

Description

Technical Field

[0001] This application relates to the field of engineering machinery safety monitoring technology, and in particular to an overload detection method for a concrete core drilling machine. Background Technology

[0002] Concrete core drilling machines are crucial equipment used in building construction, road testing, and quality assessment to obtain cylindrical concrete samples. In actual operation, due to the potential for uneven aggregate composition, reinforcement, or strength variations within the concrete, drilling rigs frequently encounter sudden loads or sustained overloads. Traditional overload protection typically relies on mechanical safety clutches or simple electrical overcurrent protection. These methods are slow to respond, have low protection accuracy, and cannot distinguish between overload types, often leading to drill bit jamming or damage, and even equipment failure and personal injury accidents.

[0003] Existing concrete core drilling machines often employ electronic protection methods based on a single sensor (such as monitoring only current or torque). However, these solutions make the single signal susceptible to interference from equipment vibration, power grid fluctuations, etc., easily leading to false alarms or missed alarms. Furthermore, the use of fixed thresholds cannot adapt to changes in operating conditions caused by different concrete grades, different drill bit diameters, and drill bit wear. Summary of the Invention

[0004] Therefore, it is necessary to provide an overload detection method for concrete core drilling machines to address the problems of inaccurate overload detection and limited applicability caused by relying on a single detection data during operation.

[0005] This application provides an overload detection method for a concrete core drilling machine, including: Real-time acquisition of multi-source data during the operation of the core drilling machine, including the current signal of the drive motor, the spindle torque signal, and the pressure signal of the hydraulic feed system; The collected multi-source data are fused to generate a comprehensive load characterization value; Based on the job parameters, a dynamic overload threshold model is invoked to calculate a real-time overload threshold. The comprehensive load characterization value is compared with the real-time overload threshold, and the steady-state overload or instantaneous impact overload is determined based on the comparison result. Based on the determined overload type, the corresponding hierarchical response control is executed.

[0006] Furthermore, the process of fusing the collected multi-source data to generate a comprehensive load characterization value includes: The current signal, torque signal, and pressure signal are normalized to convert them into dimensionless scalar values. The comprehensive load characterization value V is calculated based on V = α·I + β·T + γ·P; Where I is the normalized current signal; T is the normalized torque signal; P is the normalized pressure signal; α, β, γ are preset weighting coefficients, and satisfy α+β+γ=1.

[0007] Furthermore, the method for determining α, β, and γ is as follows: Calibration tests were performed on the core drilling machine under no-load and multiple known load conditions, and the values ​​of each signal were recorded. Calculate the Pearson correlation coefficient between each signal value and the actual load; The proportion of each signal's correlation coefficient in the total correlation coefficient is used as the weighting coefficient for that signal.

[0008] Furthermore, the step of calling the dynamic overload threshold model based on the job parameters to calculate a real-time overload threshold includes: Using historical successful drilling data as the training set, a dynamic overload threshold model is constructed using a feedforward neural network. The input features of the model include at least the concrete design grade, drill bit diameter, and drilling depth. The output of the model is the recommended benchmark threshold. At the start of the current operation, input the concrete design grade, drill bit diameter, and real-time drilling depth for the current operation, and the model will output a baseline threshold. The real-time overload threshold is obtained by multiplying the baseline threshold by a preset safety factor.

[0009] Furthermore, the condition for the steady-state overload is: The comprehensive load characterization value continuously exceeds the real-time overload threshold for a first predetermined time.

[0010] Furthermore, in the overload response step, the first-level response executed for steady-state overload includes: Control the hydraulic feed system of the core drilling machine to reduce the drill bit feed speed to K1 of the original speed; At the same time, activate the yellow warning light on the control panel; If the comprehensive load characterization value falls below the real-time overload threshold within a second predetermined time period, the drill bit will automatically return to the normal feed rate.

[0011] Furthermore, after executing the first-level response, if the comprehensive load characterization value does not decrease within a second predetermined time, or continues to rise and exceeds a higher second overload threshold, then a second-level response is executed, including: Immediately control the hydraulic feed system to completely stop the drill bit feed; The spindle motor is kept rotating for at least a third predetermined time second to perform chip removal.

[0012] Furthermore, the step of comparing the comprehensive load characterization value with the real-time overload threshold and determining steady-state overload or instantaneous impact overload based on the comparison result includes: Calculate the rate of change of the comprehensive load characterization value within a unit time Δt. If the rate of change exceeds a preset rate of change threshold, it is determined to be an instantaneous impact overload. Its corresponding overload response is: immediately control the hydraulic feed system to make the drill bit quickly retract a predetermined distance S.

[0013] Furthermore, the overload detection method for the concrete core drilling machine also includes: A prediction algorithm based on exponential smoothing is used to predict the trend of change of the comprehensive load characterization value in the fourth predetermined time period in the future. When the predicted value indicates that the comprehensive load characterization value will exceed the real-time overload threshold, the drill bit feed rate is actively reduced to K2 of the original rate before the overload occurs.

[0014] Furthermore, the overload detection method for the concrete core drilling machine also includes: After each drilling operation is successfully completed, the entire process data of that operation, including input features, comprehensive load characterization value sequence and final result, is stored as a new data sample in the historical database. The feedforward neural network model is periodically retrained using an updated historical database to optimize its weight parameters.

[0015] This application relates to an overload detection method for a concrete core drilling machine. The method involves first acquiring multi-source data during the core drilling machine's operation, which is acquired in real-time. The multi-source data is then fused to obtain a comprehensive load characterization value. Based on the operating parameters, a dynamic overload threshold model is used to calculate a real-time overload threshold. The comprehensive load characterization value is then compared with the real-time overload threshold to obtain the comparison result. Based on the comparison result, it is determined whether the core drilling machine's operating state is a steady-state overload or a transient impact overload. Different controls are implemented based on different overload types to effectively protect the core drilling machine from instantaneous high-risk situations and to mitigate and maintain a high level of continuous operation when subjected to slow overloads. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating an overload detection method for a concrete core drilling machine provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] like Figure 1 As shown, in one embodiment of this application, the overload detection method of the concrete core drilling machine includes the following steps S100 to S500.

[0019] S100 collects multi-source data in real time during the operation of the core drilling machine. The multi-source data includes the current signal of the drive motor, the spindle torque signal, and the pressure signal of the hydraulic feed system.

[0020] Specifically, multi-source data refers to multiple signals collected from sensors in different physical locations and based on different principles, which can indirectly or directly reflect the borehole load.

[0021] The current signal from the drive motor reflects the motor output, the torque signal from the spindle reflects the cutting resistance, and the pressure signal from the hydraulic feed system reflects the propulsion force. Combining these three data sources allows for a comprehensive description of the load state of the core drilling machine from three dimensions: electrical drive, mechanical drive, and hydraulic actuation.

[0022] S200 integrates the collected multi-source data to generate a comprehensive load characterization value.

[0023] Specifically, the comprehensive load characterization value is a dimensionless scalar value generated from the aforementioned multi-source data through a specific data fusion algorithm. Its purpose is to integrate load information from multiple dimensions and different units into a more stable and accurate indicator representing the overall load level of the core drilling machine, while also serving as a unified basis for all subsequent judgments and controls.

[0024] S300 uses a dynamic overload threshold model based on job parameters to calculate a real-time overload threshold.

[0025] S400 compares the comprehensive load characterization value with the real-time overload threshold and determines whether it is a steady-state overload or an instantaneous impact overload based on the comparison result.

[0026] Specifically, steady-state overloads are typically caused by encountering high-strength concrete zones or minor drill bit wear, and are characterized by a sustained, slow increase in load value beyond a safe threshold. Such overloads can be protected against using a relatively mild, graded response.

[0027] Instantaneous impact overload is usually caused by a drill bit suddenly striking a reinforcing bar, large particles, or a cavity. It is characterized by a sudden change in load value within a very short time, with an extremely high rate of change. This type of overload is highly damaging to equipment and requires a rapid and decisive response to achieve protection.

[0028] The S500 executes corresponding graded response control based on the determined overload type.

[0029] Specifically, graded response control refers to a set of control commands executed in progressively heavier, escalating manner based on the type and severity of the overload. This avoids direct shutdown, instead maintaining operational continuity and improving efficiency as much as possible while ensuring equipment safety.

[0030] In this embodiment, multi-source data is first acquired in real-time during the operation of the core drilling machine. This multi-source data is then fused to obtain a comprehensive load characterization value. Based on the operating parameters, a dynamic overload threshold model is used to calculate a real-time overload threshold. The comprehensive load characterization value is then compared with the real-time overload threshold to obtain the comparison result. Based on the comparison result, it is determined whether the core drilling machine's operating state is a steady-state overload or a transient impact overload. Different controls are implemented based on different overload types to effectively protect the core drilling machine from transient high-risk situations and to mitigate and maintain a high level of continuous operation when subjected to slow overloads.

[0031] In one embodiment of this application, the step of fusing the collected multi-source data to generate a comprehensive load characterization value includes the following steps S201 to S202.

[0032] S201, the current signal, torque signal and pressure signal are normalized respectively to convert them into dimensionless scalar values.

[0033] Specifically, normalization is a data preprocessing method whose main purpose is to map raw signal data of different dimensions and orders of magnitude to the interval [0,1] through linear transformation, thereby eliminating the influence of dimensions and making subsequent weighted fusion calculations easier.

[0034] For example, if the current signal range is 0-100A, then I = I_actual / 100; The torque signal range is 0-500 Nm, therefore T = Tactual / 500; The pressure signal range is 0-40MPa, therefore P = Pactual / 40.

[0035] S202, calculate the comprehensive load characterization value V based on V=α·I+β·T+γ·P; Where I is the normalized current signal; T is the normalized torque signal; P is the normalized pressure signal; α, β, γ are preset weighting coefficients, and satisfy α+β+γ=1.

[0036] Specifically, the weighting coefficients α, β, and γ assign weights to each normalized signal, and their magnitudes characterize the importance of that signal in constructing the overall load representation value. Higher weights indicate a greater influence of that signal on the final result. Using the Pearson correlation coefficient to determine the weights ensures that the signal with the best linear relationship to the true load has the greatest influence.

[0037] In this embodiment, the generated comprehensive load characterization value has higher reliability and anti-interference ability through normalization processing of multi-source data and correlation-based weight allocation.

[0038] In one embodiment of this application, the method for determining α, β, γ is as follows: S202a to S202c.

[0039] S202a, calibration tests were performed under no-load and multiple known load conditions of the core drilling machine, and the values ​​of each signal were recorded.

[0040] The core drilling machine was fixed on the experimental platform and connected to a standard load measuring device. With the drive motor running stably at its rated speed, at least four different known load points Li (i=1,2,3,4) were simulated by applying different braking torques. After stabilization at each load point, data from at least 100 sampling points were continuously collected, and the average value was calculated as the final signal value for that load point.

[0041] S202b, calculate the Pearson correlation coefficient between each signal value and the actual load.

[0042] Specifically, the Pearson correlation coefficient is a statistical indicator used to quantify the degree of linear correlation between two variables, with a value between -1 and 1. It is used to accurately measure the strength of the linear relationship between each normalized mechanical signal, such as I, T, and P, and the actual load. The closer the absolute value of the coefficient is to 1, the stronger the linear relationship.

[0043] The normalized value and load value of each data source are substituted into the Pearson correlation coefficient calculation formula to obtain the correlation coefficient of each data source.

[0044] S202c uses the proportion of the correlation coefficient of each signal in the total correlation coefficient as the weighting coefficient of that signal.

[0045] Specifically, each weighting factor is determined by the proportion of the correlation coefficient of each data source in the total correlation coefficient.

[0046] In one embodiment of this application, the step of calling the dynamic overload threshold model based on the job parameters to calculate a real-time overload threshold includes the following steps S301 to S305.

[0047] S301 uses historical successful drilling data as a training set and employs a feedforward neural network to construct a dynamic overload threshold model. The input features of the model include at least the concrete design grade, drill bit diameter, and drilling depth. The output of the model is the recommended benchmark threshold.

[0048] S302, at the start of the current operation, input the concrete design grade, drill bit diameter and real-time drilling depth of the current operation, and the model outputs a benchmark threshold.

[0049] S303, multiply the benchmark threshold by a preset safety factor to obtain the real-time overload threshold.

[0050] S304. After each drilling operation is successfully completed, the entire process data of the operation, including input features, comprehensive load characterization value sequence and final result, is stored as a new data sample in the historical database.

[0051] S305, The feedforward neural network model is periodically retrained using the updated historical database to optimize its weight parameters.

[0052] Specifically, the dynamic overload threshold model is a computational model that takes operational parameters as input and outputs recommended thresholds. Unlike fixed thresholds, it is dynamic, meaning that its output value automatically adjusts with changes in input conditions, such as concrete strength and drill bit size, thereby matching the protection standard with the actual situation of the current operation.

[0053] Feedforward neural networks are the preferred algorithm for implementing the dynamic model. They are computational models that mimic the structure of neural networks in the human brain.

[0054] Construction of the dynamic overload threshold model: Use a feedforward neural network with a 3-8-1 structure.

[0055] The input layer has 3 neurons, corresponding to: concrete grade, such as C30 is 30, drill bit diameter (mm), and drilling depth (mm).

[0056] The output layer has one neuron and a corresponding baseline threshold V.

[0057] The model was trained using 500 sets of historical data.

[0058] Furthermore, for every 50 successful drilling operations, new data is automatically added to the training set, and the neural network model is incrementally trained to update its weights.

[0059] In this embodiment, by constructing a model with self-learning capabilities, the overload protection standard is upgraded from static setting to dynamic adaptation, which greatly improves the intelligence level and protection accuracy under different operating conditions.

[0060] In one embodiment of this application, the condition for the steady-state overload is: The comprehensive load characterization value continuously exceeds the real-time overload threshold for a first predetermined time.

[0061] Specifically, the first predetermined time is a delay condition for determining steady-state overload. Once the load exceeds the threshold, this time must be consistently met to be considered a valid steady-state overload, rather than a momentary fluctuation. This effectively filters out brief interference signals and prevents malfunctions.

[0062] The overload response step includes the following first-level response for steady-state overload: Control the hydraulic feed system of the core drilling machine to reduce the drill bit feed speed to K1 of the original speed; Specifically, 50% <K1<80%。

[0063] At the same time, activate the yellow warning light on the control panel; If the comprehensive load characterization value falls below the real-time overload threshold within a second predetermined time period, the drill bit will automatically return to the normal feed rate.

[0064] Specifically, the first-level response is an initial mitigation measure for steady-state overload, with the aim of reducing speed as a warning and attempting to restore operation without interrupting the overload.

[0065] After executing the first-level response, if the comprehensive load characterization value does not decrease within a second predetermined time, or continues to rise and exceeds a higher second overload threshold, then the second-level response is executed, including: Immediately control the hydraulic feed system to completely stop the drill bit feed; The spindle motor is kept rotating for at least a third predetermined time second to perform chip removal.

[0066] Specifically, the second-level response is an upgraded protection when the first-level response is ineffective. Its core function is to stop the feed and ensure safety.

[0067] For example, if the steady-state overload lasts for 3 seconds, the first-level response is executed, reducing the feed rate to 70%. If the feed rate recovers within 5 seconds, it will automatically reset. If the feed rate does not recover, the second-level response is executed, stopping the feed and rotating the spindle for 8 seconds to remove chips.

[0068] The step of comparing the comprehensive load characterization value with the real-time overload threshold and determining steady-state overload or instantaneous impact overload based on the comparison result includes the following S401: S401, calculate the rate of change of the comprehensive load characterization value within a unit time Δt. If the rate of change exceeds a preset rate of change threshold, it is determined to be an instantaneous impact overload.

[0069] Specifically, the corresponding overload response is to immediately control the hydraulic feed system, causing the drill bit to quickly retract a predetermined distance S. This is a rapid protection mechanism specifically designed for instantaneous impact overloads. By controlling the drill bit to quickly retract a safe distance, it removes it from strong interaction with hard objects, thereby preventing serious accidents such as cutting or jamming.

[0070] The rate of change threshold is the core criterion for judging instantaneous impact overload. It sets a limit value for the rate of load growth. When the instantaneous rate of change of the load exceeds this threshold, regardless of its absolute value, it is immediately judged as a dangerous impact.

[0071] In one embodiment of this application, the overload detection method of the concrete core drilling machine further includes the following steps S601 to S602.

[0072] S601, using a prediction algorithm based on exponential smoothing, to predict the trend of change of the comprehensive load characterization value in the fourth predetermined time period in the future.

[0073] Specifically, exponential smoothing is a classic method for time series forecasting. Its core idea is to assign greater weight to recent data and less weight to older data, with the weights decreasing exponentially, thereby sensitively capturing the latest trends in data changes.

[0074] Holt linear trend exponential smoothing method is chosen because it can simultaneously capture the level of data, such as the magnitude of the current value and the rate of increase or decrease of the trend, making it suitable for loads that may exhibit a continuous growth or decline trend.

[0075] The trend of change in the fourth predetermined time period in the future does not refer to a single future point, but rather to the overall direction and magnitude of the change in the comprehensive load characterization value V from the current moment to the fourth predetermined time period in the future. This trend is quantified by predicting the specific value V at the fourth predetermined time period in the future and comparing it with the current value.

[0076] S602, when the predicted value indicates that the comprehensive load characterization value will exceed the real-time overload threshold, the drill bit feed rate is actively reduced to K2 of the original rate before the overload occurs.

[0077] Specifically, 70% <K2<90%。

[0078] In this embodiment, by predicting overload trends at the algorithm level and proactively and slightly reducing the load before the overload actually occurs, potential overload risks are mitigated, achieving an improvement from protection to avoidance.

[0079] The technical features of the above embodiments can be combined arbitrarily, and the execution order of the method steps is not restricted. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0080] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An overload detection method for a concrete core drill, characterized by, The overload detection method of the concrete core drill includes: Real-time acquisition of multi-source data during the operation of the core drill, the multi-source data including a current signal of a driving motor, a torque signal of a main shaft, and a pressure signal of a hydraulic feeding system; Fusion of the acquired multi-source data to generate a comprehensive load characteristic value; Calling of a dynamic overload threshold model based on operation parameters to calculate a real-time overload threshold; Comparison of the comprehensive load characteristic value with the real-time overload threshold to determine a steady-state overload or a transient impact overload according to a comparison result; Based on the determined overload type, corresponding hierarchical response control is performed.

2. The method of overload detection for a concrete core drill according to claim 1, wherein, The fusion of the acquired multi-source data to generate a comprehensive load characteristic value includes: Normalization processing of the current signal, the torque signal, and the pressure signal respectively to convert them into dimensionless scalar values; Calculation of the comprehensive load characteristic value V based on V = α·I + β·T + γ·P; Wherein, I is the normalized current signal; T is the normalized torque signal; P is the normalized pressure signal; α, β, γ are preset weighting coefficients, and α + β + γ = 1.

3. The method of overload detection for a concrete core drill according to claim 2, wherein, The determination method of α, β, γ includes: Calibration testing under no-load and multiple known load states of the core drill to record the values of each signal; Calculation of the Pearson correlation coefficient between each signal value and the actual load; The proportion of the correlation coefficient of each signal in the total correlation coefficient is taken as the weighting coefficient of the signal.

4. The method of overload detection for a concrete core drill according to claim 1, wherein, The calling of a dynamic overload threshold model based on operation parameters to calculate a real-time overload threshold includes: Using historical successful drilling operation data as a training set, a dynamic overload threshold model is constructed by using a feedforward neural network, the input features of the model including at least a concrete design grade, a drill bit diameter, and a drilling depth, and the output of the model being a recommended reference threshold; At the beginning of the current operation, the concrete design grade, the drill bit diameter, and the real-time drilling depth of the current operation are input, and a reference threshold is output by the model; The reference threshold is multiplied by a preset safety factor to obtain the real-time overload threshold.

5. The method of overload detection for a concrete core drill according to claim 1, wherein, The condition for determining a steady-state overload is that the time during which the comprehensive load characteristic value continuously exceeds the real-time overload threshold reaches a first predetermined time. In the overload response step, the first-level response performed for a steady-state overload includes:

6. The method of overload detection for a concrete core drill according to claim 5, wherein, Controlling the hydraulic feeding system of the core drill to reduce the drill bit feeding speed to K1 of the original speed; At the same time, activating the yellow warning light on the control panel; Within a second predetermined time, if the comprehensive load characteristic value falls below the real-time overload threshold, the drill bit is automatically restored to the normal feeding speed. After the first-level response is performed, if the comprehensive load characteristic value does not fall within the second predetermined time, or continues to rise and exceeds a higher second overload threshold, a second-level response is performed, including:

7. The method of overload detection for a concrete core drill according to claim 6, wherein, Immediately controlling the hydraulic feeding system to completely stop the drill bit feeding; Maintaining the main shaft motor to continue rotating for no less than a third predetermined time to perform chip removal. The comparison of the comprehensive load characteristic value with the real-time overload threshold to determine a steady-state overload or a transient impact overload includes:

8. The method of overload detection for a concrete core drill according to claim 1, wherein, ​ The rate of change of the comprehensive load characteristic value in a unit time Δt is calculated, and if the rate of change exceeds a preset rate of change threshold, it is determined that there is an instantaneous impact overload; The corresponding overload response is to immediately control the hydraulic feeding system to make the drill bit quickly retreat a predetermined distance S.

9. The method of overload detection for a concrete core drill according to claim 1, wherein, The overload detection method of the concrete core drill also includes: A prediction algorithm based on the exponential smoothing method is used to predict the trend of the comprehensive load characteristic value in a future time period of a fourth predetermined time; When the predicted value indicates that the comprehensive load characteristic value will exceed the real-time overload threshold, the drill bit feeding speed is actively reduced to K2 of the original speed before the overload occurs.

10. The method of overload detection for a concrete core drill according to claim 4, wherein, The overload detection method of the concrete core drill also includes: After each drilling operation is successfully completed, the whole process data of the operation, including the input characteristics, the comprehensive load characteristic value sequence and the final result, are stored as a new data sample in the historical database; The updated historical database is regularly used to retrain the feedforward neural network model to optimize its weight parameters.

Images