Drill rod power tong motor torque self-adaptive control system based on load prediction
By using a load-predictive adaptive torque control system for drill pipe tongs motors, combined with an ARIMA model and PID control, the nonlinear and time-varying problems of drill pipe tongs motor torque control are solved, achieving precise torque matching and adaptive load adjustment, thus improving the safety and efficiency of drilling operations.
Patent Information
- Application Number
- CN202610077444.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-21
- Publication Date
- 2026-02-24
AI Technical Summary
PID control algorithms struggle to adapt to nonlinear, time-varying load torque changes in drill pipe power tongs motor torque control, leading to overshoot, oscillation, or slow response, which affects the safety and efficiency of drilling operations.
A load-prediction-based adaptive torque control system for drill pipe power tongs motors is adopted. Through data acquisition, fusion input acquisition, machine torque prediction, and torque control modules, the system uses an ARIMA model and PID control algorithm, combined with the drill pipe diameter and geological hardness, to dynamically adjust the motor torque to adapt to load changes.
It achieves accurate control of the drill pipe power tong motor torque, reduces torque error, improves operational safety and efficiency, and adapts to load changes under different working conditions.
Smart Images
Figure CN121567017A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of motor torque control technology, and more specifically to an adaptive control system for drill pipe power tong motor torque based on load prediction. Background Technology
[0002] In oil drilling and well workover operations, the drill pipe power tong is the core equipment for threading and unthreading drill pipe or casing. The motor torque control of the drill pipe power tong directly determines operational safety, efficiency, and equipment lifespan, making it a crucial technical aspect for ensuring the stable operation of drilling projects. Improper torque control can lead to loose threaded connections due to insufficient torque, potentially causing drilling fluid leaks in the high-pressure, high-vibration downhole environment. Conversely, excessive torque can easily cause thread "sticking" or overstretching, resulting in permanent damage or even rendering the drill pipe unusable. Generally, a PID control algorithm is used to adjust the motor output based on the deviation between the actual feedback torque and the set torque.
[0003] However, the control effect of PID control algorithm heavily depends on accurate system model and fixed PID parameters. The drill pipe fastening and unfastening process is a typical nonlinear and time-varying process. The load torque is strongly disturbed by various factors such as thread condition, lubrication conditions, speed, temperature and wear of the drill pipe itself. This makes it difficult for fixed PID parameters to maintain the best control performance under various working conditions, and problems such as overshoot, oscillation or slow response are likely to occur. Summary of the Invention
[0004] This invention provides a load prediction-based adaptive torque control system for drill pipe power tongs motors to address the problem that the nonlinear relationship between torque and rotation angle easily causes overshoot, oscillation, or slow response in PID control algorithms, leading to inaccurate torque control of the drill pipe power tongs motor. The specific technical solution adopted is as follows: One embodiment of the present invention provides a drill pipe power tong motor torque adaptive control system based on load prediction, the system comprising the following modules: The data acquisition module is used to collect working condition data, drill pipe diameter and geological hardness of the drill pipe power tong at different acquisition times, and obtain standard values for each type of working condition data. The working condition data includes the motor torque, tong head torque, rotation angle, rotation speed, vibration, temperature and position of the drill pipe power tong. The fusion input acquisition module is used to calculate the mutual information of each type of working condition data based on the correlation between all working condition data, delete some working condition data, record any acquisition time as the target acquisition time, calculate the fusion input of the target acquisition time based on the values and mutual information of the types of working condition data retained at the target acquisition time, and construct the fusion input sequence of the target acquisition time. The machine torque prediction module is used to calculate the performance of each type of working condition data retained at the target acquisition time based on the differences between the fused inputs of the acquisition time adjacent to the target acquisition time and the differences between each type of working condition data. Combined with the motor torque of the drill pipe power tong, the module calculates the corrected prediction value of the motor torque at each acquisition time. The torque control module is used to calculate the dynamic feedforward gain at the time of data acquisition based on the drill pipe diameter and geological hardness. Combined with the corrected predicted value of the motor torque at the time of data acquisition, it calculates the adaptive proportional coefficient at the time of data acquisition and controls the torque of the drill pipe power tong motor based on the adaptive proportional coefficient.
[0005] Furthermore, the specific steps involved in deleting some operating condition data are as follows: When the mutual information of a type of operating condition data is less than or equal to a preset correlation threshold, the operating condition data of that type is deleted.
[0006] Furthermore, the method for obtaining the fusion input at the target acquisition time is as follows: The working condition data of any one type that is retained is recorded as the target working condition data. The ratio of the mutual information of the target working condition data to the sum of the mutual information of all retained types of working condition data is recorded as the first ratio of the target working condition data. Based on the first ratio of the target working condition data at the target acquisition time, the dynamic weight of the target working condition data at the target acquisition time is calculated. The sum of the products of the dynamic weights of all types of operating condition data retained and the operating condition data at the target acquisition time is recorded as the correlation influence of the operating condition data at the target acquisition time. The sum of the product of the preset first coefficient and the motor torque at the target acquisition time and the correlation influence of the operating condition data at the target acquisition time is recorded as the fusion input at the target acquisition time.
[0007] Furthermore, the method for obtaining the dynamic weights is as follows: The product of the difference between the number 1 and the preset first coefficient and the first ratio of the target operating condition data at the target acquisition time is recorded as the dynamic weight of the target operating condition data at the target acquisition time.
[0008] Furthermore, the data included in the fusion input sequence at the target acquisition time is as follows: The target acquisition time and the fusion input of all acquisition times prior to the target acquisition time.
[0009] Furthermore, the method for calculating the performance of each type of working condition data retained at the target acquisition time is as follows: The average of the absolute values of the differences between the fused input at the target acquisition time and the fused input at two adjacent acquisition times is recorded as the first average value at the target acquisition time; the average of the absolute values of the differences between the target operating condition data at the target acquisition time and the target operating condition data at two adjacent acquisition times is recorded as the second average value of the target operating condition data at the target acquisition time. The third mean of the target operating condition data is calculated based on the difference between the first mean of the target acquisition time and all acquisition times prior to the target acquisition time and the second mean of the target operating condition data. The absolute value of the difference between the fused input at the target acquisition time and the mean of all fused inputs in the fused input sequence is denoted as the second absolute value at the target acquisition time. The positive correlation between the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time is denoted as the performance of the target operating condition data at the target acquisition time.
[0010] Furthermore, the method for determining the third mean is as follows: The absolute value of the difference between the first mean of the target acquisition time and the second mean of the target operating condition data is recorded as the first absolute value of the target operating condition data at the target acquisition time; the mean of the first absolute values of the target operating condition data at the target acquisition time and all acquisition times before the target acquisition time is recorded as the third mean of the target operating condition data.
[0011] Furthermore, the specific steps for calculating the corrected predicted value of the motor torque at each acquisition moment, based on the motor torque of the drill pipe power tongs, are as follows: The performance of the target operating condition data at the target acquisition time is used as the prediction weight of the ARIMA model, and the ARIMA model is used to calculate the predicted value of the motor torque at the next adjacent acquisition time of the target acquisition time. The absolute value of the difference between the motor torque at the target acquisition time and the predicted value of the motor torque is denoted as the predicted torque difference at the target acquisition time. The average value of the predicted torque differences at the target acquisition time and all previous target acquisition times is denoted as the error coefficient at the target acquisition time. The product of the sum of the number 1 and the error coefficient at the target acquisition time and the motor torque at the target acquisition time is denoted as the corrected predicted value of the motor torque at the target acquisition time.
[0012] Furthermore, the formula for calculating the dynamic feedforward gain at the acquisition time is: in, Represents the dynamic feedforward gain at the acquisition moment; and These are the preset second and third coefficients, and the sum of the second and third coefficients is 1. The diameter of the drill rod at the time of data collection; Indicates the geological hardness at the time of sampling; This indicates the maximum value of the drill pipe diameter for the drill pipe power tong motor; This indicates the maximum geological hardness that the drill pipe power tong motor can handle.
[0013] Furthermore, the method for calculating the adaptive proportional coefficient at the acquisition time by combining the corrected predicted value of the motor torque at the acquisition time, and controlling the torque of the drill pipe power tong motor based on the adaptive proportional coefficient, includes the following: The product of the corrected predicted value of the motor torque at the acquisition time and the dynamic feedforward gain is denoted as the adaptive proportional coefficient at the acquisition time. The adaptive proportional coefficient at the acquisition time is used as the proportional coefficient value of the PID control algorithm, and the torque control of the drill pipe power tong motor is achieved by using the PID control algorithm.
[0014] The beneficial effects of this invention are: First, data with strong correlations are selected from the operating condition data, while data with weak correlations are deleted. The strength of the correlations is evaluated, and the evaluation result is the mutual information of each type of operating condition data. Based on the degree of influence between different types of operating condition data, the strength of the correlation influence of different types of data at each acquisition time is calculated, and the fused input and fused input sequence at the target acquisition time are obtained. Then, based on the changes in the fused input in the fused input sequence and the changes in the types of operating condition data retained at each acquisition time, the influence of the types of operating condition data retained at each acquisition time on the motor torque is evaluated, and the performance of each type of operating condition data retained at the target acquisition time is obtained. Based on the performance, the drill pipe power tong is... The motor torque is corrected, and the corrected predicted value of the motor torque at each acquisition time is calculated. Since the motor torque of the drill pipe power tong motor is affected by the drill pipe diameter and geological hardness, the adaptive proportional coefficient at the acquisition time is jointly determined based on the corrected predicted value of the motor torque at the acquisition time, the drill pipe diameter, and the geological hardness. This balances the influence of some adverse geological conditions on the torque control of the drill pipe power tong motor. Based on the adaptive proportional coefficient, the torque of the drill pipe power tong motor is controlled, solving the problem that the nonlinear relationship between torque and rotation angle can easily cause overshoot, oscillation, or slow response of the PID control algorithm, which in turn leads to inaccurate torque control of the drill pipe power tong motor, thus improving the accuracy of the torque control of the drill pipe power tong motor. 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 of 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 flowchart illustrating an embodiment of the drill pipe power tong motor torque adaptive control system provided by the present invention. Figure 2 This is a schematic diagram of the adaptive torque control system for drill pipe power tongs motor based on load prediction, provided in one embodiment of the present invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] Please see Figure 1 The diagram illustrates a flowchart of a drill pipe power tong motor torque adaptive control system based on load prediction, according to an embodiment of the present invention. Figure 2 A schematic diagram of a drill pipe power tong motor torque adaptive control system based on load prediction is shown in an embodiment of the present invention. The system includes: a data acquisition module, a fusion input acquisition module, a machine torque prediction module, and a torque control module.
[0019] The data acquisition module collects operating data, drill pipe diameter, and geological hardness of the drill pipe power tong at different acquisition times, and obtains standard values for each type of operating data. The operating data includes the motor torque, tong head torque, rotation angle, rotation speed, vibration, temperature, and position of the drill pipe power tong.
[0020] Data on the working conditions of the drill pipe power tongs at different times were collected, including drill pipe diameter and geological hardness. The working conditions data included the motor torque, tong head torque, rotation angle, speed, vibration, temperature and position of the drill pipe power tongs.
[0021] Specifically, the motor torque of the drill pipe power tong is obtained using a rotary torque sensor on the motor output shaft; the tong head torque is obtained using a static or dynamic torque sensor installed on the torque transmission path of the tong head; the rotation angle is obtained using an incremental rotary encoder installed on the motor shaft or tong head drive shaft; the rotation speed is obtained using a tachogenerator or Hall sensor; the vibration is obtained using an ICP accelerometer installed on the tong body or motor housing; the temperature is obtained using a PT100 thermal resistor installed near the motor windings, bearings, or tong jaws; and the position, the clamping position of the drill pipe power tong's tong head, is obtained using an LVDT linear displacement sensor.
[0022] In this embodiment, mean filtering is used to denoise each type of working condition data. Mean filtering is a well-known technique and will not be described in detail here. As another implementation, while achieving the goal of data denoising, the implementer may use other methods in the prior art, such as median filtering, for data denoising. This application does not impose any special restrictions. In this embodiment, working condition data, drill pipe diameter, and geological hardness were collected within one hour, and the data acquisition frequency was 100Hz.
[0023] It is understandable that for each type of working condition data, there is a standard value preset according to the working condition.
[0024] Thus, the standard values for the working condition data of the drill pipe power tong at different acquisition times and for each type of working condition data are obtained.
[0025] The fusion input acquisition module calculates the mutual information of each type of working condition data based on the correlation between all working condition data, deletes some working condition data, records any acquisition time as the target acquisition time, calculates the fusion input of the target acquisition time based on the values and mutual information of the types of working condition data retained at the target acquisition time, and constructs the fusion input sequence of the target acquisition time.
[0026] In automated drilling processes, the stability and speed of each tightening operation directly impact the overall progress. The torque control process of the drill pipe tong motor, implemented using a PID control algorithm, cannot anticipate changes in the motor load, often resulting in slow response or excessive caution, potentially leading to prolonged tightening times or retrying failures. Furthermore, the drill pipe tong's transmission mechanism contains nonlinear factors such as backlash and friction, and the torque-angle relationship during tightening is also nonlinear. This often leads to insufficient adjustment and slow response from the PID control algorithm at small deviations, while at large deviations, it suffers from over-adjustment due to integral saturation and prolonged recovery time. Therefore, predicting and proactively adjusting the drill pipe tong's motor load is crucial to prevent abnormal operation caused by adjustment lag.
[0027] First, when controlling the torque of the drill pipe power tong motor, the torque variation is affected by various factors such as drill pipe specifications, formation hardness, and operating stage. It is necessary to first select data with strong correlation in the working condition data and evaluate the strength of the correlation.
[0028] A target matrix is established based on all operating condition data. The mutual information of each type of operating condition data is calculated based on the target matrix and the standard values for each type of operating condition data. If the mutual information of a certain type of operating condition data is greater than a preset correlation threshold, that type of operating condition data is retained; if the mutual information of a certain type of operating condition data is less than or equal to the preset correlation threshold, that type of operating condition data is deleted.
[0029] In this embodiment, the correlation threshold is set to 0.25; the process of establishing the target matrix and calculating the mutual information are well-known techniques and will not be described in detail; the correlation between the retained types of operating condition data and other types of operating condition data is relatively strong, while the correlation between the deleted types of operating condition data and other types of operating condition data is relatively weak.
[0030] The retained operating condition data of any type is designated as the target operating condition data. The ratio of the mutual information of the target operating condition data to the sum of the mutual information of all retained operating condition data types is designated as the first ratio of the target operating condition data. The first ratio of any retained operating condition data of any type can be obtained in the same way. Any acquisition time is designated as the target acquisition time. The product of the difference between the number 1 and the preset first coefficient and the first ratio of the target operating condition data at the target acquisition time is designated as the dynamic weight of the target operating condition data at the target acquisition time. The sum of the products of the dynamic weights of all retained operating condition data types and the operating condition data at the target acquisition time is designated as the correlation influence of the operating condition data at the target acquisition time. The sum of the product of the preset first coefficient and the motor torque at the target acquisition time and the correlation influence of the operating condition data at the target acquisition time is designated as the fusion input at the target acquisition time.
[0031] In this embodiment, the first coefficient is set to 0.3. To ensure the accuracy of the analysis, if the number of data acquisition times before the target acquisition time is less than 100, the target acquisition time will not be analyzed.
[0032] The dynamic weight of the target operating condition data is used to evaluate the degree of influence between the target operating condition data and other types of operating condition data. The greater the dynamic weight of the target operating condition data, the more important the target operating condition data is relative to other types of operating condition data.
[0033] The same method can be used to obtain the fusion input at any acquisition time, and the fusion inputs at the target acquisition time and all acquisition times before the target acquisition time can be arranged in the order of acquisition to obtain the fusion input sequence at the target acquisition time.
[0034] At this point, the fusion input sequence at the target acquisition time has been obtained.
[0035] The machine torque prediction module calculates the performance of each type of working condition data retained at the target acquisition time based on the differences between the fused inputs of acquisition times adjacent to the target acquisition time and the differences between each type of working condition data retained. Combined with the motor torque of the drill pipe power tong, it calculates the corrected prediction value of the motor torque at each acquisition time.
[0036] Based on the changes in the fused input in the fused input sequence and the changes in the types of operating condition data retained at each acquisition time, the impact of the types of operating condition data retained at each acquisition time on the motor torque is evaluated.
[0037] The average of the absolute values of the differences between the fused input at the target acquisition time and the fused input at two adjacent acquisition times is denoted as the first mean of the target acquisition time. The average of the absolute values of the differences between the target operating condition data at the target acquisition time and the target operating condition data at two adjacent acquisition times is denoted as the second mean of the target operating condition data at the target acquisition time. The absolute value of the difference between the first mean of the target acquisition time and the second mean of the target operating condition data is denoted as the first absolute value of the target operating condition data at the target acquisition time. The first absolute value of the target operating condition data at any acquisition time can be obtained in the same way. The average of the first absolute values of the target operating condition data at the target acquisition time and all acquisition times before the target acquisition time is denoted as the third mean of the target operating condition data. The absolute value of the difference between the fused input at the target acquisition time and the mean of all fused inputs in the fused input sequence is denoted as the second absolute value of the target acquisition time. The positive correlation result of the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time is denoted as the performance of the target operating condition data at the target acquisition time.
[0038] It is understood that a positive correlation is applied to the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time. This ensures that the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time are positively correlated with the performance of the target operating condition data at the target acquisition time. It is understood that the positive correlation in this application refers to the relationship between the independent and dependent variables. The independent variables are the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time, and the dependent variable is the performance of the target operating condition data at the target acquisition time. A positive correlation means that the dependent variable increases (decreases) as the independent variable increases (decreases), and can be an additive or multiplicative relationship.
[0039] Preferably, as an embodiment of this application, the normalized value of the product of the second absolute value, the first mean value and the third mean value of the target operating condition data at the target acquisition time is denoted as the performance of the target operating condition data at the target acquisition time.
[0040] In this embodiment, the sigmoid function is used to calculate the normalized value. The sigmoid function is a well-known technique and will not be described in detail here. As other implementation methods, implementers can use other methods of the prior art, such as the tanh function.
[0041] Among them, the two adjacent acquisition times of the target acquisition time are the previous adjacent acquisition time and the next adjacent acquisition time.
[0042] The larger the second absolute value of the target acquisition time, the greater the difference between the combined input and all fused input values within the fused input sequence at the target acquisition time. This results in a greater change in the motor torque of the drill pipe power tong at the target acquisition time. Therefore, when predicting the motor load of the drill pipe power tong, the influence weight of the data at the target acquisition time should be greater, thus increasing the expressiveness of the target operating condition data at the target acquisition time. Similarly, the larger the first mean value of the target acquisition time, the greater the change in motor torque at the target acquisition time, further increasing the expressiveness of the target operating condition data at the target acquisition time.
[0043] The performance of the target working condition data at the target acquisition time is used as the prediction weight of the ARIMA model. The ARIMA model is used to process the motor torque of the drill pipe power tong at the target acquisition time and all acquisition times before the target acquisition time, and the predicted value of the motor torque at the next adjacent acquisition time of the target acquisition time is calculated.
[0044] The use of the ARIMA model for data prediction is a well-known technique and will not be elaborated further.
[0045] The load torque is strongly affected by various factors such as thread condition, lubrication conditions, speed, temperature, and wear of the drill rod itself, resulting in a non-linear change in the value of the motor torque. Therefore, in order to reduce the deviation of the motor torque prediction, the value is further corrected based on the predicted value of the motor torque at different acquisition times.
[0046] The absolute value of the difference between the motor torque at the target acquisition time and the predicted value of the motor torque is denoted as the predicted torque difference at the target acquisition time. The average value of the predicted torque differences at the target acquisition time and all previous target acquisition times is denoted as the error coefficient at the target acquisition time. The product of the sum of the number 1 and the error coefficient at the target acquisition time and the motor torque at the target acquisition time is denoted as the corrected predicted value of the motor torque at the target acquisition time.
[0047] The same method can be used to obtain the corrected predicted value of the motor torque at any acquisition time.
[0048] At this point, the corrected predicted values of the motor torque at all acquisition times have been obtained.
[0049] The torque control module calculates the dynamic feedforward gain at the time of data acquisition based on the drill pipe diameter and geological hardness. It then calculates the adaptive proportional coefficient at the time of data acquisition by combining the corrected predicted value of the motor torque at the time of data acquisition. Based on the adaptive proportional coefficient, it controls the torque of the drill pipe power tong motor.
[0050] Based on the corrected predicted value of the motor torque, drill pipe diameter, and geological hardness at the acquisition time, the adaptive scaling factor at the acquisition time is calculated. The formula for calculating the adaptive scaling factor at the acquisition time is: in, The adaptive scaling factor indicates the acquisition time. Represents the dynamic feedforward gain at the acquisition moment; This represents the corrected predicted value of the motor torque at the time of data acquisition. and These are the preset second and third coefficients, and the sum of the second and third coefficients is 1. In this embodiment, the values of the second and third coefficients are 0.6 and 0.4, respectively. The diameter of the drill rod at the time of data collection; Indicates the geological hardness at the time of sampling; This indicates the maximum value of the drill pipe diameter for the drill pipe power tong motor; This indicates the maximum geological hardness that the drill pipe power tong motor can handle.
[0051] The motor torque of the drill pipe power tong motor is affected by the drill pipe diameter and geological hardness. Therefore, based on the corrected predicted value of the motor torque at the time of data acquisition, the drill pipe diameter, and the geological hardness, an adaptive proportional coefficient is jointly determined at the time of data acquisition to balance the impact of some adverse geological conditions on the torque control of the drill pipe power tong motor.
[0052] The adaptive proportional coefficient at the acquisition time is used as the proportional coefficient value of the PID control algorithm. The PID control algorithm is then used to process the motor torque of the drill pipe power tong to achieve torque control of the drill pipe power tong motor.
[0053] The PID control algorithm is a well-known technology and will not be described in detail here.
[0054] The torque control of the drill pipe power tong motor achieved through the above steps can achieve: (1) ensuring that the torque strictly matches the drill pipe specifications during threading and unthreading, so that the static torque error is less than or equal to ±3% FS, and accurately tracking the torque command; (2) suppressing torque fluctuations, ensuring that the torque fluctuation amplitude is less than or equal to ±5% of the rated torque in case of sudden load changes such as drill pipe jamming or changes in clamping force; (3) achieving overload protection, quickly reducing torque when the torque exceeds 1.2 times the rated torque to avoid drill pipe deformation and motor burnout; (4) adapting to changes in working conditions, that is, being compatible with the torque requirements of drill pipes of different diameters and different well depths.
[0055] The ±3% FS is the error range defined by the Full Scale, which ensures that the accuracy of torque control can be quantified.
[0056] It should be noted that the motor speed of the drill pipe power tong is usually less than 50 rpm when it is in operation, while the torque can reach tens of thousands; the impact load will cause a sudden change in torque at the moment of unhooking.
[0057] Thus, the torque of the drill pipe power tong motor is controlled based on the load.
[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A drill pipe power tong motor torque adaptive control system based on load prediction, characterized in that, The system includes the following modules: The data acquisition module is used to collect working condition data, drill pipe diameter and geological hardness of the drill pipe power tong at different acquisition times, and obtain standard values for each type of working condition data. The working condition data includes the motor torque, tong head torque, rotation angle, rotation speed, vibration, temperature and position of the drill pipe power tong. The fusion input acquisition module is used to calculate the mutual information of each type of working condition data based on the correlation between all working condition data, delete some working condition data, record any acquisition time as the target acquisition time, calculate the fusion input of the target acquisition time based on the values and mutual information of the types of working condition data retained at the target acquisition time, and construct the fusion input sequence of the target acquisition time. The machine torque prediction module is used to calculate the performance of each type of working condition data retained at the target acquisition time based on the differences between the fused inputs of the acquisition time adjacent to the target acquisition time and the differences between each type of working condition data. Combined with the motor torque of the drill pipe power tong, the module calculates the corrected prediction value of the motor torque at each acquisition time. The torque control module is used to calculate the dynamic feedforward gain at the time of data acquisition based on the drill pipe diameter and geological hardness. Combined with the corrected predicted value of the motor torque at the time of data acquisition, it calculates the adaptive proportional coefficient at the time of data acquisition and controls the torque of the drill pipe power tong motor based on the adaptive proportional coefficient.
2. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The specific steps involved in deleting certain operating condition data are as follows: When the mutual information of a type of operating condition data is less than or equal to a preset correlation threshold, the operating condition data of that type is deleted.
3. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The method for obtaining the fusion input at the target acquisition time is as follows: The working condition data of any one type that is retained is recorded as the target working condition data. The ratio of the mutual information of the target working condition data to the sum of the mutual information of all retained types of working condition data is recorded as the first ratio of the target working condition data. Calculate the dynamic weight of the target operating condition data at the target acquisition time based on the first ratio of the target operating condition data at the target acquisition time. The sum of the products of the dynamic weights of all types of operating condition data retained and the operating condition data at the target acquisition time is recorded as the correlation influence of the operating condition data at the target acquisition time. The sum of the product of the preset first coefficient and the motor torque at the target acquisition time and the correlation influence of the operating condition data at the target acquisition time is recorded as the fusion input at the target acquisition time.
4. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 3, characterized in that, The method for obtaining the dynamic weights is as follows: The product of the difference between the number 1 and the preset first coefficient and the first ratio of the target operating condition data at the target acquisition time is recorded as the dynamic weight of the target operating condition data at the target acquisition time.
5. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The data included in the fusion input sequence at the target acquisition time is as follows: The target acquisition time and the fusion input of all acquisition times prior to the target acquisition time.
6. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 3, characterized in that, The method for calculating the performance of each type of working condition data retained at the target acquisition time is as follows: The average of the absolute values of the differences between the fused input at the target acquisition time and the fused input at two adjacent acquisition times is recorded as the first average value at the target acquisition time; the average of the absolute values of the differences between the target operating condition data at the target acquisition time and the target operating condition data at two adjacent acquisition times is recorded as the second average value of the target operating condition data at the target acquisition time. The third mean of the target operating condition data is calculated based on the difference between the first mean of the target acquisition time and all acquisition times prior to the target acquisition time and the second mean of the target operating condition data. The absolute value of the difference between the fused input at the target acquisition time and the mean of all fused inputs in the fused input sequence is denoted as the second absolute value at the target acquisition time. The positive correlation between the second absolute value, the first mean, and the third mean of the target operating condition data at the target acquisition time is denoted as the performance of the target operating condition data at the target acquisition time.
7. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 6, characterized in that, The method for determining the third mean is as follows: The absolute value of the difference between the first mean of the target acquisition time and the second mean of the target operating condition data is recorded as the first absolute value of the target operating condition data at the target acquisition time; the mean of the first absolute values of the target operating condition data at the target acquisition time and all acquisition times before the target acquisition time is recorded as the third mean of the target operating condition data.
8. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The specific steps for calculating the corrected predicted value of the motor torque at each data acquisition moment, based on the motor torque of the drill pipe power tongs, are as follows: The performance of the target operating condition data at the target acquisition time is used as the prediction weight of the ARIMA model, and the ARIMA model is used to calculate the predicted value of the motor torque at the next adjacent acquisition time of the target acquisition time. The absolute value of the difference between the motor torque at the target acquisition time and the predicted value of the motor torque is denoted as the predicted torque difference at the target acquisition time. The average value of the predicted torque differences at the target acquisition time and all target acquisition times before the target acquisition time is denoted as the error coefficient at the target acquisition time. The product of the sum of the error coefficients of the number 1 and the target acquisition time and the motor torque at the target acquisition time is denoted as the corrected predicted value of the motor torque at the target acquisition time.
9. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The formula for calculating the dynamic feedforward gain at the acquisition time is: in, Represents the dynamic feedforward gain at the acquisition moment; and These are the preset second and third coefficients, and the sum of the second and third coefficients is 1. The diameter of the drill rod at the time of data collection; Indicates the geological hardness at the time of sampling; This indicates the maximum value of the drill pipe diameter for the drill pipe power tong motor; This indicates the maximum geological hardness that the drill pipe power tong motor can handle.
10. The drill pipe power tong motor torque adaptive control system based on load prediction according to claim 1, characterized in that, The method for calculating the adaptive proportional coefficient at the acquisition time by combining the corrected predicted value of the motor torque at the acquisition time, and controlling the torque of the drill pipe power tong motor based on the adaptive proportional coefficient, includes the following: The product of the corrected predicted value of the motor torque at the acquisition time and the dynamic feedforward gain is denoted as the adaptive proportional coefficient at the acquisition time. The adaptive proportional coefficient at the acquisition time is used as the value of the proportional coefficient of the PID control algorithm, and the torque control of the drill pipe power tong motor is achieved by using the PID control algorithm.