Petroleum drilling machine driving-based permanent magnet motor state monitoring control system

Abstract

The invention discloses a permanent magnet motor state monitoring control system based on petroleum drilling machine driving, and relates to the technical field of permanent magnet motor state monitoring. Comprising a data acquisition and construction module, a data preprocessing and feature extraction module, a dynamic monitoring and trend analysis module, a deep learning evaluation module and a fuzzy control and real-time compensation module, the collected real-time data are stored in a time sequence mode, a data set is established, and complete and accurate basic data are provided for subsequent analysis and model evaluation. Through real-time data acquisition, deep learning evaluation, fuzzy control optimization and error compensation, and adaptive regulation and control of dead zone time, the stability, reliability and energy efficiency of the petroleum drilling machine driving permanent magnet motor are improved, the short circuit risk, energy loss and harmonic damage are reduced, the system performance is optimized, and the operation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of permanent magnet motor state monitoring, specifically to a permanent magnet motor state monitoring control system based on oil rig driving. BACKGROUND

[0002] Permanent magnet motor state monitoring control based on oil rig driving refers to real-time monitoring of the running state of the motor and its driving system when the permanent magnet motor (PMM) drives the oil rig for operation, and effective adjustment and optimization through the control system. Permanent magnet motors are widely used in oil rigs due to their high efficiency, long service life, and high power density. State monitoring collects various operating data of the motor through sensors, such as current, voltage, temperature, vibration, etc., and uses data analysis and algorithms for fault diagnosis, performance evaluation, and trend prediction. Through feedback from the control system, precise control of the motor can be achieved to avoid overload, overheating, and other problems, improving the reliability of the motor and the efficiency of the oil rig. At the same time, dynamic adjustment and optimization of control strategies can extend the service life of the equipment, reduce maintenance costs, and ensure the smooth progress of the drilling process.

[0003] In the permanent magnet motor driving system, the controller is the core command and adjustment unit, which monitors and precisely controls the running state of the motor in real time. Through advanced motor control algorithms, the controller dynamically adjusts the driving signal based on real-time data and uses SVPWM (Space Vector Pulse Width Modulation) technology to convert the optimized voltage signal into a three-phase PWM signal to drive the inverter, thereby precisely controlling the input voltage and current of the permanent magnet motor, ensuring the efficiency, stability, and safety of the system.

[0004] The existing technology has the following shortcomings:

[0005] In the existing technology, SVPWM inverter control is usually set with a dead time to prevent the simultaneous conduction (short circuit) of the upper and lower bridge arms IGBT (Insulated Gate Bipolar Transistor). However, improper setting of the dead time (too short or too long) can cause serious problems. When the dead time is too short, it may cause the simultaneous conduction of the upper and lower bridge arms, resulting in a straight-through short circuit, causing the IGBT to withstand a large impact current, and in severe cases, even instant burnout. If the dead time is too long, it will cause distortion of the output voltage waveform, reduce the efficiency of the permanent magnet motor, and may increase high-order harmonics, thereby causing winding overheating, increased vibration, and even insulation breakdown, seriously affecting the reliability and service life of the permanent magnet motor.

[0006] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0007] The purpose of the present application is to provide a permanent magnet motor state monitoring control system based on oil drilling rig driving, based on real-time data acquisition, deep learning intelligent evaluation, fuzzy control dynamic optimization and error compensation, to improve the stability, reliability and energy efficiency of the permanent magnet motor driven by the oil drilling rig. Through data preprocessing, the model input is ensured to be accurate, the deep learning intelligent evaluation optimizes the dead time setting, the fuzzy control self-adaptively adjusts to avoid short circuit risk and energy loss, the error compensation mechanism corrects the waveform distortion, reduces the harmonic loss and winding overheating, and prolongs the service life of the motor. The present application effectively solves the problem of dead time mismatch, optimizes the performance of the motor driving system, and improves the running stability and efficiency of the oil drilling rig under complex working conditions, to solve the problems in the above background technology.

[0008] In order to achieve the above purpose, the present application provides the following technical scheme: a permanent magnet motor state monitoring control system based on oil drilling rig driving, comprising a data acquisition and construction module, a data preprocessing and feature extraction module, a dynamic monitoring and trend analysis module, a deep learning evaluation module and a fuzzy control and real-time compensation module:

[0009] The data acquisition and construction module first acquires the running data in the running process of the inverter through the data acquisition device, stores the acquired real-time data in time sequence, and establishes a data set, providing complete and accurate basic data for subsequent analysis and model evaluation;

[0010] The data preprocessing and feature extraction module preprocesses the inverter running parameter information in the data set, and extracts the core parameters that intuitively reflect the mismatch of dead time from the preprocessed data as the key indicators for evaluating whether the dead time is matched;

[0011] The dynamic monitoring and trend analysis module performs fine-grained analysis on the extracted key indicators under the detection window, and captures the dynamic change trend caused by the mismatch of dead time;

[0012] The deep learning evaluation module inputs the analyzed key indicators into the pre-learned deep learning model, and intelligently evaluates whether the dead time setting is matched from the complex feature relationships through the deep learning model;

[0013] The fuzzy control and real-time compensation module dynamically adjusts the dead time based on the evaluation results according to the preset fuzzy rules when the evaluation results show that the dead time does not match the actual operation, and simultaneously compensates the output error caused by the dead time in real time through digital signal processing technology to correct the output error caused by the change of the dead time.

[0014] Preferably, the specific steps of acquiring the operating data of the inverter in real time during operation by a data acquisition device and establishing a data set are as follows:

[0015] First, deploy high-precision data acquisition equipment in the inverter system to ensure real-time monitoring of the inverter's input, output, and key component status;

[0016] Secondly, the collected signals are processed synchronously, and the data from all sensors are timestamped using a high-precision clock to ensure the time consistency of the data.

[0017] Then, a high-sampling-rate data recording system is used to store data in a time series format to capture transient changes caused by dead-time mismatch during inverter operation.

[0018] Next, we established data sets for merging and classification management, and labeled the data according to the inverter's operating conditions for subsequent analysis.

[0019] Finally, through data integrity checking, abnormal data points are eliminated and interpolation compensation is performed to ensure the integrity and accuracy of the data set, thereby providing high-quality data support for subsequent feature analysis, deep learning evaluation and dynamic regulation.

[0020] Preferably, core parameters that intuitively reflect the dead time mismatch are extracted from the preprocessed data, and the extracted parameters include the DC bus voltage fluctuation and the actual IGBT turn-off time deviation under the pulse working mode. The DC bus voltage fluctuation and the actual IGBT turn-off time deviation under the pulse working mode are analyzed under the detection window to generate a DC bus feedback abnormality reference value and an IGBT turn-off delay reference value respectively. The dynamic change trend caused by the dead time mismatch is jointly quantified by the DC bus feedback abnormality reference value and the IGBT turn-off delay reference value.

[0021] Preferably, the specific steps of analyzing the DC bus voltage fluctuation in the pulse working mode within the detection window to generate the DC bus feedback abnormality reference value are as follows:

[0022] In pulsed operation mode, the DC bus voltage signal exhibits periodic fluctuations. However, the mismatch in dead time will increase the irregularity of the voltage fluctuation. Therefore, the nonlinear features of the bus voltage signal are first extracted and the accumulated offset of its high-order derivatives is analyzed to quantify the fluctuation trend of the bus voltage. The feature extraction formula is as follows:

[0023]

[0024] Where F1 is the nonlinear high-order characteristic index of the DC bus voltage fluctuation, which is used to measure the severity of the DC bus voltage fluctuation. N is the total number of time points in the detection window. is the third derivative of the DC bus voltage, V dc (k) is the bus voltage value collected at time point k, a is a nonlinear weight coefficient, used to adjust the influence degree of the third derivative term, is the voltage fluctuation index decay term, V dc (k) is the voltage change amplitude between V dc (k-1) at the previous time point, e is the natural base, and β is the exponential decay factor;

[0025] The abnormal fluctuation of the bus voltage not only reflects the suddenness, but also causes the imbalance of the bus energy feedback. Therefore, a nonlinear energy deviation function is introduced to measure the unevenness of the bus voltage energy distribution, and the formula is as follows:

[0026]

[0027] In the formula, F2 is the imbalance index of the DC bus energy feedback, E rec (k) is the stored energy at time point k, E del (k) is the energy transmitted to the load side at time point k, ε is a small value to prevent the denominator from being zero, ensuring the stability of the calculation, and γ is an exponential adjustment coefficient, controlling the contribution degree of the bus feedback imbalance to the abnormal index;

[0028] The nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation and the imbalance index F2 of the DC bus energy feedback are combined, and a nonlinear weight distribution function is introduced to generate a DC bus feedback abnormal reference value, and the generation formula is as follows:

[0029]

[0030] In the formula, λ1 is the weight coefficient of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, λ2 is the weight coefficient of the imbalance index F2 of the DC bus energy feedback, η1 is the exponential adjustment factor of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, and η2 is the exponential adjustment factor of the imbalance index F2 of the DC bus energy feedback, is the exponential decay factor, e is the natural base, and ζ is a dynamic adjustment parameter, used to adjust the sensitivity of the exponential decay factor .

[0031] Preferably, the specific steps of analyzing the IGBT actual turn-off time deviation in the detection window to generate the IGBT turn-off delay reference value are as follows:

[0032] In the detection window, the key signals in the IGBT turn-off process are monitored by a high-speed data acquisition system, including current, voltage and power loss changes. Since the IGBT turn-off process involves rapid current decay, a turn-off transient characteristic function is defined to capture the degree of mismatch between the current drop rate and the voltage rise rate, as follows:

[0033]

[0034] where Φ off is the IGBT turn-off transient characteristic, I ds is the drain-source current during IGBT turn-off, V ce is the collector-emitter voltage during IGBT turn-off, Q g is the gate charge, is the current drop rate during IGBT turn-off, is the voltage rise rate during IGBT turn-off, and ρ is a weight factor to balance the influence weight of the current drop rate and the voltage rise rate.

[0035] During the IGBT turn-off process, the loss characteristic is affected by the dead time. When the dead time is too short, the current does not completely decay, resulting in an increase in the instantaneous power peak value. Therefore, an IGBT turn-off loss offset function is introduced to characterize the energy offset during the turn-off process, as follows:

[0036]

[0037] where Ψ loss is the IGBT turn-off loss offset, Ω off is the time region of IGBT turn-off, P sw (ζ) is the actually measured IGBT instantaneous switching loss power, P ideal (ζ) is the theoretically calculated IGBT turn-off loss power, and μ is a scaling factor to adjust the proportion factor of the theoretical loss reference.

[0038] Finally, the IGBT turn-off delay reference value is generated through the comprehensive analysis of the IGBT turn-off transient characteristic Φ off and the IGBT turn-off loss offset Ψ loss , as follows:

[0039]

[0040] where ITDI is the IGBT turn-off delay reference value, max(Φ off ) is the maximum IGBT turn-off transient characteristic in the detection window, max(Ψ loss ) is the maximum IGBT turn-off loss offset in the detection window, and κ1 is the IGBT turn-off transient characteristic Φoff weight factor, κ2 is the IGBT turn-off loss offset Ψ loss weight factor, is the IGBT turn-off transient characteristic normalization, is the IGBT turn-off energy offset normalization, β1 and β2 are both normalization index adjustment factors, β2 is used to control the normalized IGBT turn-off loss offset Ψ loss nonlinear influence in calculation, β1 is used to control the normalized IGBT turn-off transient characteristic Φ off nonlinear influence in calculation.

[0041] Preferably, the analyzed DC bus feedback abnormal reference value and the IGBT turn-off delay reference value are input into the pre-learned deep learning model, and a dead time mismatch risk coefficient is generated through the deep learning model, and the dead time setting is intelligently evaluated through the dead time mismatch risk coefficient.

[0042] Preferably, the dead time mismatch risk coefficient generated when the deep learning model is used to intelligently evaluate the dead time setting is compared and analyzed with the preset dead time mismatch risk coefficient reference threshold, and the dead time setting is divided, and the division steps are as follows:

[0043] If the dead time mismatch risk coefficient is greater than the dead time mismatch risk coefficient reference threshold, the current dead time setting is divided as not matching the actual operation; if the dead time mismatch risk coefficient is less than or equal to the dead time mismatch risk coefficient reference threshold, the current dead time setting is divided as matching the actual operation.

[0044] Preferably, when the evaluation result shows that the dead time does not match the actual operation, the fuzzy logic controller dynamically adjusts the dead time based on the evaluation result according to the preset fuzzy rules, and simultaneously compensates the output error caused by the dead time in real time through digital signal processing technology. The specific steps are as follows:

[0045] In the dead time adjustment process, first, the fuzzy logic controller is used to calculate the dead time adjustment amount according to the dead time mismatch risk coefficient DTM and the dead time mismatch risk coefficient reference threshold, and the calculation expression is as follows:

[0046] ΔDT=μ f ·(α1·(DTM-TDM ref )+α2·|I spk |+α3·|THD|)

[0047] In the formula, ΔDT is the dead time adjustment amount, μ fis the fuzzy adjustment factor, a1 is the dead-time bias weight factor, TDM ref is the dead-time mismatch risk coefficient reference threshold, I spk is the current spike amplitude, a2 is the current spike weight factor, indicating the influence weight of the current spike amplitude on the dead-time adjustment, THD is the total harmonic distortion, a3 is the total harmonic distortion weight factor, indicating the influence weight of the total harmonic distortion on the dead-time adjustment;

[0048] After determining the dead-time adjustment amount DT, the PWM modulation strategy of the inverter is updated in real time to ensure that the adjustment of the dead-time does not affect the overall stability, and the calculation expression is as follows:

[0049]

[0050] DT new is the updated dead-time, DT old is the original dead-time, e -ω·DTM is the exponential adjustment term, an exponential scaling factor for controlling the exponential scaling factor of the dead-time adjustment rate, e is the natural base,

[0051] ω is the adjustment response factor, controlling the adjustment speed, is the ripple component of the DC bus voltage, is the rated voltage of the DC bus, is the normalized DC bus voltage ripple, is the DC bus voltage ripple weight factor, is the load current commutation dynamic weight factor, is the rate of change of the load current to the commutation voltage;

[0052] Since the dynamic adjustment of the dead-time will affect the output voltage and current waveform of the motor, it is necessary to use digital signal processing technology to compensate for the error caused by the dead-time to ensure that the output quality is not affected, and the calculation expression is as follows:

[0053] In the formula, V comp is the compensation voltage, θ1 is the dead-time change weight factor, θ2 is the voltage error weight factor, V ideal is the ideal output voltage, V real is the actual output voltage, is the second-order voltage change rate, θ3 is the voltage second-order derivative weight factor, θ4 is the current ripple weight factor, which measures the influence degree of the current ripple on the compensation voltage, is the normalized current ripple, I ripple is the current ripple amplitude, I nominal is the rated current, used to normalize the contribution of the current ripple to the error compensation.

[0054] In the above technical solutions, the present application provides technical effects and advantages:

[0055] Based on real-time data acquisition, deep learning intelligent evaluation, fuzzy control dynamic optimization and DSP error compensation, the present application constructs an adaptive dead time regulation mechanism, which can effectively improve the stability, reliability and energy efficiency of the permanent magnet motor system driven by the oil drilling rig. Through data acquisition and preprocessing, the data integrity and accuracy of the input model are ensured, and the accuracy and reliability of the evaluation are improved; the deep learning model is used to mine the complex relationship between the dead time setting and the actual running state, and high-precision intelligent evaluation is realized; combined with the fuzzy logic controller, the dead time is adaptively adjusted under different operating conditions to avoid the risk of short circuit caused by too short dead time, while reducing the voltage distortion, high-order harmonics and energy loss caused by too long dead time, and optimizing the output performance of the motor. In addition, the DSP error compensation mechanism further corrects the influence of dead time adjustment on the output waveform, ensures the stability of the voltage and current waveform of the inverter, reduces the harmonic loss, reduces the winding overheating phenomenon, thereby prolonging the service life of the motor, and improving the overall operation efficiency of the oil drilling rig system. The present application not only effectively solves the safety hazards caused by mismatched dead time, but also optimizes the performance of the motor drive system through intelligent means, significantly improves the running stability and working efficiency of the permanent magnet motor driven by the oil drilling rig under complex operating conditions. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments or prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0057] Figure 1 The module schematic diagram of the oil drilling rig driven permanent magnet motor state monitoring control system based on the present application is shown in the figure. DETAILED DESCRIPTION

[0058] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations, however, can be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art.

[0059] The present application provides as Figure 1The shown permanent magnet motor state monitoring control system based on oil drilling rig driving includes a data acquisition and construction module, a data preprocessing and feature extraction module, a dynamic monitoring and trend analysis module, a deep learning evaluation module, and a fuzzy control and real-time compensation module.

[0060] The data acquisition and construction module first acquires real-time operation data of the inverter during operation through a data acquisition device, stores the acquired real-time data in a time sequence manner, and establishes a data set to provide complete and accurate basic data for subsequent analysis and model evaluation.

[0061] During the inverter operation data acquisition process, in order to ensure the accuracy and completeness of subsequent analysis, high-precision data acquisition equipment (such as voltage, current, temperature, vibration sensors) is needed to collect key parameters of the inverter under different operating conditions in real time. These parameters include three-phase output current, voltage waveform, DC bus voltage, IGBT switching state, switching temperature, motor operating parameters, etc. All collected data are stored in a time sequence manner, i.e. each data point is matched with a time stamp, so that subsequent dynamic analysis can be based on time evolution law. At the same time, all data are organized into a data set to ensure that the operating information under different working conditions is covered, providing complete and high-quality basic data for subsequent steps such as deep learning model training, feature extraction, and dead time evaluation. The key goal of this process is to obtain accurate and comprehensive real-time data to provide reliable input for intelligent analysis and control optimization.

[0062] The specific steps of acquiring real-time operation data of the inverter during operation through a data acquisition device and establishing a data set are as follows: First, deploy high-precision data acquisition equipment (such as current, voltage, temperature, vibration sensors) in the inverter system to ensure real-time monitoring of the inverter's input, output, and key device status. Second, synchronize the collected signals and use a high-precision clock to timestamp align all sensor data to ensure time consistency. Third, store the data in a time sequence manner through a high-sampling-rate data recording system to capture transient changes caused by dead time mismatch during inverter operation, such as current spikes and voltage distortion. Fourth, establish a data set and perform classification management by labeling the data according to the inverter's operating conditions (such as different loads, speeds, and temperature conditions) for subsequent analysis. Finally, perform data integrity checks to eliminate abnormal data points and perform interpolation compensation to ensure the integrity and accuracy of the data set, thereby providing high-quality data support for subsequent feature analysis, deep learning evaluation, and dynamic control.

[0063] The data preprocessing and feature extraction module preprocesses the inverter operating parameter information in the data set, and extracts core parameters that intuitively reflect the mismatch of dead time from the preprocessed data, as key indicators for evaluating whether the dead time is matched.

[0064] Due to the complex operating environment of the inverter, the collected raw data may have noise interference, missing values, inconsistent sampling frequencies, etc. In the preprocessing link, the data needs to be denoised, filtered, interpolated and normalized to improve the accuracy and stability of subsequent analysis. For example, moving average method or wavelet transform can be used to smooth the voltage and current waveforms and remove high-frequency interference; for missing data, linear interpolation or interpolation based on adjacent samples can be used to make the data set more complete. Through preprocessing, useful information can be retained to the greatest extent, while reducing the interference of invalid or erroneous data on judgment.

[0065] The dynamic monitoring and trend analysis module performs fine-grained analysis on the extracted key indicators under the detection window to capture the dynamic change trend caused by the mismatch of dead time.

[0066] From the preprocessed data, core parameters that intuitively reflect the mismatch of dead time are extracted, including DC bus voltage fluctuation under pulse operation mode and IGBT actual turn-off time deviation. The DC bus voltage fluctuation under pulse operation mode and the IGBT actual turn-off time deviation are analyzed under the detection window to generate DC bus feedback abnormal reference value and IGBT turn-off delay reference value respectively, and the dynamic change trend caused by the mismatch of dead time is quantified by the DC bus feedback abnormal reference value and the IGBT turn-off delay reference value.

[0067] Increased DC bus voltage fluctuations in pulsed operation indicate that the inverter dead time setting is mismatched with actual operation. This is primarily due to the dynamic impact of the dead time on the inverter's commutation process, which in turn causes unstable DC bus fluctuations. In pulsed operation, the inverter's switching frequency is high, and the bus voltage must remain stable to support normal inverter operation. However, if the dead time is set too short or too long, abnormal bus voltage fluctuations can occur. When the dead time is too short, the inverter's upper and lower IGBTs may briefly cross-conduct, causing energy loss in the DC bus, resulting in a brief voltage sag and potential bus current surges, causing high-frequency bus voltage oscillations. When the dead time is too long, the inverter output voltage experiences more "neutral" states during commutation, potentially leading to energy from the load side being fed back to the bus, causing abnormal bus voltage increases and even waveform distortion or exacerbated oscillations in the power circuit. Greater bus voltage fluctuations indicate a greater degree of dead-time mismatch, which not only impacts the inverter's power supply stability but can also increase switching losses in power devices, accelerate bus capacitor aging, and reduce system reliability. Therefore, increased DC bus voltage fluctuations in pulsed operation are a key indicator of a mismatch between the inverter's dead-time setting and actual operation, requiring correction through adaptive optimization strategies to ensure system stability and efficient operation.

[0068] The specific steps for analyzing the DC bus voltage fluctuation in the pulse working mode within the detection window to generate the DC bus feedback abnormality reference value are as follows:

[0069] In pulsed operation mode, the DC bus voltage signal exhibits periodic fluctuations. However, the mismatch in dead time will increase the irregularity of the voltage fluctuation. Therefore, the nonlinear features of the bus voltage signal are first extracted and the accumulated offset of its high-order derivatives is analyzed to quantify the fluctuation trend of the bus voltage. The feature extraction formula is as follows:

[0070]

[0071] Where F1 is the nonlinear high-order characteristic index of the DC bus voltage fluctuation, which is used to measure the severity of the DC bus voltage fluctuation. A higher value means a stronger voltage fluctuation, which may indicate a dead time mismatch. N is the total number of time points in the detection window. It is the third-order derivative of the DC bus voltage, which indicates the degree of the signal's sudden change in time. dc (k) is the bus voltage value collected at time point k, α is the nonlinear weight coefficient, which is used to adjust the influence of the third-order derivative term. is the voltage fluctuation exponential decay term, and V is calculated at time point k. dc (k) and the previous moment V dc(k-1) between the voltage change amplitude, and its influence on the final F1 value is controlled by an exponential decay function, e is the natural base, β is the exponential decay factor, used to control the decay rate of the voltage fluctuation term, adjust the sensitivity of the system to large voltage fluctuations;

[0072] This step can capture the severity of the bus voltage transient change by calculating the third-order derivative offset accumulation of the bus voltage. If the dead time is mismatched, the high-order derivative value of the voltage fluctuation will increase significantly, thereby enhancing F1, indicating that the intensity of the abnormal fluctuation of the bus voltage is higher.

[0073] The abnormal fluctuation of the bus voltage not only reflects the suddenness, but also leads to the imbalance of the bus energy feedback, therefore, a nonlinear energy offset function is introduced to measure the unevenness of the bus voltage energy distribution, the formula is as follows:

[0074]

[0075] In the formula, F2 is the imbalance index of the DC bus energy feedback, used to measure the abnormality of the bus energy flow caused by the mismatch of the dead time, E rec (k) is the stored energy at time point k, used to quantify the amount of energy fed back to the bus by the load, E del (k) is the energy transmitted from the bus to the load side at time point k, ε is a small value to prevent the denominator from being zero, ensuring the stability of the calculation, γ is the exponential adjustment coefficient, controlling the contribution degree of the bus feedback imbalance to the abnormal index;

[0076] This step calculates the deviation degree between the bus energy feedback and the load energy transmission. Ideally, energy feedback and load transmission should be close to balance, but if the dead time is too long, it may lead to insufficient load energy supply during commutation, and the bus feedback energy rises, making E rec much larger than E del ; on the contrary, if the dead time is too short, it may lead to cross conduction, causing the load power to increase and disrupting the bus energy balance. Therefore, when F2 value increases, it means that the abnormality of bus feedback is more serious, and the possibility of mismatched dead time setting is higher.

[0077] Combine the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation and the imbalance index F2 of the DC bus energy feedback, and introduce a nonlinear weight distribution function to generate the abnormal reference value of the DC bus feedback, the generation formula is as follows:

[0078]

[0079] wherein λ1 is the weight coefficient of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, λ2 is the weight coefficient of the unbalance index F2 of the DC bus energy feedback, η1 is the index adjustment factor of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, η2 is the index adjustment factor of the unbalance index F2 of the DC bus energy feedback,

[0080] is an exponential decay factor, used to adjust the trade-off relationship between F1 and F2, to prevent a single feature from dominating the final abnormality index, e is the natural base, and ζ is a dynamic adjustment parameter, used to adjust the sensitivity of the exponential decay factor .

[0081] This step fuses F1 and F2 through nonlinear weighting, ensuring that the DC bus feedback abnormality index can comprehensively characterize the voltage fluctuation abnormality caused by the mismatch of the dead-time. When the DBFAI value is large, it indicates that the fluctuation amplitude of the bus voltage is large, or the unbalance of the bus energy feedback is strong, meaning that the dead-time may be set unreasonably and needs to be dynamically optimized and adjusted.

[0082] The larger the DC bus feedback abnormality reference value generated by analyzing the DC bus voltage fluctuation in the pulse operation mode under the detection window, the more it indicates that the dead-time setting of the inverter does not match the actual operation, and vice versa. In the pulse operation mode, the amplitude and frequency of the DC bus voltage fluctuation are directly affected by the dead-time. When the dead-time is too short, the IGBTs of the upper and lower bridge arms of the inverter may be briefly cross-conducted, causing the bus current to fluctuate sharply, and in turn causing the bus voltage to drop momentarily or oscillate at high frequency, resulting in an increase in the DC bus feedback abnormality reference value. On the other hand, a long dead-time will cause the commutation time of the inverter output voltage to lag, making the energy on the load side unable to be effectively transmitted during the dead-time period, resulting in abnormal rise of the bus voltage or instability of the power feedback, also causing abnormal rise of the reference value. Therefore, the larger the DC bus feedback abnormality reference value, the more significant the bus voltage fluctuation, indicating that the dead-time setting does not match the actual working condition and needs to be optimized and adjusted. Conversely, if the reference value remains at a low level, it indicates that the dead-time setting is reasonable, the bus voltage fluctuation is within a normal range, the inverter is running stably as a whole, and the dead-time matching degree is high.

[0083] The actual turn-off time of IGBT appears delay, indicating that the dead-time setting of inverter does not match the actual operation, because the dead-time directly affects the switching dynamic characteristics of IGBT and the stability of commutation process. Under normal circumstances, IGBT needs to complete turn-off within the set dead-time to ensure that the upper and lower bridge arms do not cross-conduct. However, if the dead-time is set too short, the lower bridge IGBT may have turned on in advance before the IGBT turns off, causing a short period of shoot-through short circuit, which may exacerbate the carrier storage effect of IGBT, leading to turn-off delay. On the other hand, if the dead-time is set too long, the load current may lack an effective freewheeling path during commutation, causing additional current oscillation after IGBT turns off, prolonging the turn-off time. In addition, too long dead-time may also cause voltage distortion during commutation, affecting the dynamic behavior of IGBT and making it unable to complete turn-off at the desired time point. Therefore, the delay of IGBT actual turn-off time reflects the non-optimal setting of dead-time, which may cause cross-conduction, current spikes, voltage distortion and other problems, thereby affecting the stability of inverter and the operating efficiency of motor. Therefore, the delay of turn-off time is an important indicator of the mismatch of dead-time, which needs to be monitored and optimized in real time to ensure the reliable operation of inverter system.

[0084] The specific steps of generating IGBT turn-off delay reference value by analyzing IGBT actual turn-off time deviation in the detection window are as follows:

[0085] In the detection window, the key signals in the IGBT turn-off process are monitored by a high-speed data acquisition system, including current, voltage and power loss changes. Since the IGBT turn-off process involves rapid current decay, a turn-off transient characteristic function is defined to capture the mismatch between current decay rate and voltage rise rate, as follows:

[0086]

[0087] In the formula, Φ off is the IGBT turn-off transient characteristic, which measures the mismatch between current decay rate and voltage rise rate during IGBT turn-off, thereby reflecting the influence of dead-time on the dynamic behavior of IGBT, I ds is the drain-source current of IGBT at turn-off, V ce is the collector-emitter voltage of IGBT at turn-off, Q g is the gate charge, representing the amount of charge transferred during the charging and discharging process of IGBT gate, determining the switching speed of IGBT, is the current decay rate during IGBT turn-off, describing the rate at which the drain current I ds of IGBT decreases with the gate charge Qg the rate of change of the voltage, is the voltage rise rate of IGBT during turn-off process, describes the collector voltage V ce the rate of change of the gate charge Q g , p is the weight factor, used to balance the influence weight of current drop rate and voltage rise rate;

[0088] The function of this formula is to measure the mismatch degree of current and voltage change during IGBT turn-off. If Φ off is too large, it means that the current decay is too slow or the voltage rises too fast during IGBT turn-off, which may exist unreasonable dead time.

[0089] During the turn-off process of IGBT, its loss characteristics are affected by the dead time. When the dead time is too short, the current has not completely decayed, resulting in the increase of instantaneous power peak value. Therefore, the IGBT turn-off loss offset function is introduced to describe the energy offset during the turn-off process, and the calculation formula is as follows:

[0090]

[0091] Ψ loss is the IGBT turn-off loss offset, represents the total offset of the actual loss compared with the ideal loss of IGBT during turn-off process, Ω off is the time region of IGBT turn-off, P sw (ζ) is the actual measured IGBT instantaneous switching loss power, represents the actual instantaneous switching loss power of IGBT at time point ζ, P ideal (ζ) is the theoretical calculation of IGBT turn-off loss power (assuming the ideal value when the dead time is completely matched), μ is the adjustment factor, used to adjust the proportion factor of the theoretical loss benchmark, the purpose is to match the theoretical calculation value and the actual loss benchmark under different load, current and voltage conditions;

[0092] The function of this step is to measure the offset of the actual IGBT turn-off loss compared with the ideal loss. If Ψ loss is too large, it means that there is additional energy loss during the IGBT turn-off process, which is usually caused by too short or too long dead time.

[0093] Finally, through the comprehensive analysis of IGBT turn-off transient characteristics Φ off and IGBT turn-off loss offset Ψ loss , the IGBT turn-off delay reference value is generated, and the generation formula is as follows:

[0094]

[0095] In the formula, ITDI is the IGBT turn-off delay reference value, max(Φoff ) is the maximum IGBT turn-off transient characteristic in the detection window, max(Ψ loss ) is the maximum IGBT turn-off loss deviation in the detection window, K1 is the weight factor of the IGBT turn-off transient characteristic Φ off , K2 is the weight factor of the IGBT turn-off loss deviation Ψ loss is the IGBT turn-off transient characteristic normalization, is the IGBT turn-off energy deviation normalization, and β1 and β2 are both the exponential adjustment factors after normalization, and β2 is used to control the non-linear influence of the IGBT turn-off loss deviation Ψ loss in the calculation, and β1 is used to control the non-linear influence of the IGBT turn-off transient characteristic Φ off in the calculation.

[0096] The purpose of this step is to construct a normalization index that comprehensively considers the transient characteristic deviation and energy loss deviation during the IGBT turn-off process. If the ITDI value is high, it indicates that the IGBT turn-off process is adversely affected by the dead time, and there may be commutation abnormalities, cross-conduction risks, or increased high-order harmonics. Conversely, if the ITDI value is low, it indicates that the IGBT turn-off time and the dead time are well matched, the commutation process is stable, and the dead time setting is reasonable.

[0097] The larger the IGBT turn-off delay reference value generated after analyzing the actual IGBT turn-off time deviation in the detection window, the more mismatched the inverter dead time setting is with the actual operation, and vice versa. The IGBT turn-off delay reference value is calculated by monitoring the actual IGBT turn-off time deviation in the monitoring window, and its performance value directly reflects the influence of the dead time on the IGBT turn-off dynamic behavior. When the reference value is high, it indicates that there is a large deviation between the IGBT turn-off time and the expected dead time, which may be caused by the following two situations:

[0098] (1) The dead time is too short, and the IGBT is forced to turn off before the carriers are completely depleted, resulting in a delay in the turn-off process and possibly accompanied by cross-conduction phenomena;

[0099] (2) The dead time is too long, and there is a lack of effective freewheeling path during commutation, causing additional delay due to current impact or parasitic capacitance effects after the IGBT turns off. Both situations indicate that the dead time does not accurately match the actual working condition, which may cause problems such as decreased commutation efficiency, current distortion, or increased switching loss. Conversely, when the IGBT turn-off delay reference value performance is low, it means that the IGBT turn-off time is basically consistent with the expected dead time, and the commutation process is stable, indicating that the current dead time setting is reasonable and matches the actual operation. ​

[0100] a deep learning evaluation module that inputs the analyzed key indicators into a pre-learned deep learning model to intelligently evaluate whether the dead time setting matches through the deep learning model from complex feature relationships;

[0101] The analyzed DC bus feedback abnormal reference value and IGBT shutdown delay reference value are input into a pre-learned deep learning model, and a dead time mismatch risk coefficient is generated through the deep learning model, and the dead time setting is intelligently evaluated through the dead time mismatch risk coefficient.

[0102] The pre-learned deep learning model refers to the use of a large amount of historical data for training before actual application, so that it has the ability to intelligently evaluate the risk of dead time mismatch. The training process of this model usually includes key steps such as data collection, data preprocessing, feature extraction, model construction, optimization training and test verification. Specifically, first, the operating data of the inverter under different working conditions need to be collected, including the DC bus voltage fluctuation under pulse working mode, IGBT shutdown delay and other parameters that may affect the matching degree of dead time, and the data is denoised, normalized and feature engineered to improve the quality of training data and the robustness of the model. Next, use these processed data to build training data sets and test data sets, and use supervised learning or unsupervised learning methods to select appropriate deep learning models for training. For example, if using a convolutional neural network (CNN), the voltage waveform data can be used to extract features and identify subtle distortions caused by dead time mismatch; if using a long short-term memory network (LSTM), the changing trend of time series data can be analyzed to capture the dynamic impact of dead time. In addition, Transformer or self-supervised learning models can be combined to improve the model's generalization ability under different operating conditions. During training, use cross-entropy loss function, mean square error (MSE) and other evaluation indicators for optimization, and constantly adjust hyperparameters (such as learning rate, network depth, activation function, etc.) to ensure that the model can accurately judge the dead time matching situation.

[0103] Once the model training is completed, it has the ability to intelligently evaluate the matching degree of the dead time, that is, it can receive the analyzed DC bus feedback abnormal reference value and IGBT turn-off delay reference value as input in real time during actual operation, calculate the mismatch risk of the current dead time through the forward propagation process of the neural network, and output the dead time mismatch risk coefficient. The risk coefficient quantifies the rationality of the current dead time setting, which can be used to determine whether the dead time needs to be adjusted to optimize the operation state of the inverter. For example, if the risk coefficient is high (such as close to 1), it indicates that the dead time is seriously mismatched, which may cause short circuit, increased harmonics or excessive energy loss, and the system should immediately adjust the dead time; if the risk coefficient is low (such as close to 0), it indicates that the dead time setting is reasonable and does not need to be adjusted. In this way, the pre-learned deep learning model can effectively improve the intelligent level of dead time optimization, avoid the errors caused by traditional experience-based dead time setting, and improve the safety, stability and efficiency of inverter operation.

[0104] The deep learning model is not limited here and can achieve comprehensive analysis of the DC bus feedback abnormal reference value DBFAI and the IGBT turn-off delay reference value ITDI to generate a dead time mismatch risk coefficient DTM.

[0105] The deep learning model for generating the dead time mismatch risk coefficient DTM from the DC bus feedback abnormal reference value DBFAI and the IGBT turn-off delay reference value ITDI can be used to achieve the technical solution of the present application, and the present application provides a specific implementation method.

[0106] The dead time mismatch risk coefficient DTM is generated according to the following formula:

[0107]

[0108] In the formula, m1 and m2 are preset proportion coefficients of the DC bus feedback abnormal reference value DBFAI and the IGBT turn-off delay reference value ITDI, respectively, and m1 and m2 are both greater than 0.

[0109] The preset proportion coefficients m1 and m2 are adjustment parameters for weighing the influence degree of the DC bus feedback abnormal reference value DBFAI and the IGBT turn-off delay reference value ITDI on the dead time mismatch risk coefficient DTM. The role of these two coefficients is to determine the relative contribution proportion of DBFAI and ITDI in calculating DTM, so that DTM can accurately reflect the comprehensive risk caused by the dead time mismatch.

[0110] In actual application, DBFAI reflects the influence of dead time mismatch on DC bus energy feedback, while ITDI reflects the influence of dead time mismatch on IGBT switching dynamic behavior. Different inverter topologies, load characteristics and operating conditions may have different sensitivities to these two factors, so m1 and m2 need to be weighted and adjusted. For example:

[0111] If the system pays more attention to the influence of DC bus voltage fluctuation on inverter stability, m1>m2 can be set to give DBFAI a higher weight.

[0112] m1>m2, which gives DBFAI a higher weight.

[0113] If the system pays more attention to the influence of IGBT turn-off delay on power loss and cross-conduction, m2>m1 can be set to make ITDI account for a larger proportion in the calculation of DTM.

[0114] In addition, the denominator The normalization factor is used to balance the DTM calculation values under different working conditions, so as to have certain dimensional consistency and prevent the calculation results from being excessively amplified due to too large m1 and m2. In general, the preset proportion coefficients m1 and m2 are weight parameters set according to system requirements, which are used to optimize DTM calculation so that it can more accurately represent the comprehensive risk of dead-time mismatch.

[0115] From the dead-time mismatch risk coefficient, the greater the DC bus feedback abnormality reference value generated by analyzing the DC bus voltage fluctuation under the detection window in the pulse operation mode, the greater the IGBT turn-off delay reference value generated by analyzing the IGBT actual turn-off time deviation under the detection window, the greater the dead-time mismatch risk coefficient generated by intelligent evaluation of the dead-time setting by the pre-learned deep learning model, which indicates that the current dead-time setting does not match the actual operation, and vice versa, which indicates that the current dead-time setting matches the actual operation.

[0116] The dead-time mismatch risk coefficient generated by intelligent evaluation of the dead-time setting by the pre-learned deep learning model is compared and analyzed with the preset dead-time mismatch risk coefficient reference threshold, and the dead-time setting is divided, and the division steps are as follows:

[0117] If the dead-time mismatch risk coefficient is greater than the dead-time mismatch risk coefficient reference threshold, the current dead-time setting is divided into mismatching the actual operation; if the dead-time mismatch risk coefficient is less than or equal to the dead-time mismatch risk coefficient reference threshold, the current dead-time setting is divided into matching the actual operation.

[0118] Fuzzy control and real-time compensation module, when the evaluation result shows that the dead time does not match the actual operation, the fuzzy logic controller dynamically adjusts the dead time according to the preset fuzzy rules (such as "if the dead time is insufficient and an abnormal peak current is detected, then appropriately increase the dead time"; "if the dead time is too long and the harmonic distortion rate rises, then shorten the dead time) based on the evaluation result, and simultaneously compensates for the output error caused by the dead time in real time through digital signal processing (DSP) technology to correct the output error caused by the change of the dead time.

[0119] When the evaluation result shows that the dead time does not match the actual operation, the fuzzy logic controller and the digital signal processing (DSP) technology work together, and the core function is to dynamically optimize the setting of the dead time to match the current operating condition of the inverter, and to compensate for the voltage and current error caused by the adjustment in real time to ensure the stability of the system, the safety of the power device and the operation quality of the permanent magnet motor.

[0120] First, the fuzzy logic controller (FLC) determines the adjustment strategy of the dead time based on the evaluation result and the preset fuzzy rules. Since the setting of the dead time cannot be too short (to avoid IGBT cross conduction and cause short circuit) or too long (to avoid output voltage distortion and increase high-order harmonics), the fuzzy controller combines the dead time mismatch risk coefficient (DTM), the harmonic distortion rate (THD), the current peak (I spk ) and other key parameters to intelligently calculate the optimization amount of the dead time and dynamically adjust the dead time in real time. This adjustment method based on fuzzy control is more adaptive than fixed dead time setting, and can maintain the optimal dead time range and improve the stability of the system operation as the load, power grid fluctuation, temperature and other external conditions change.

[0121] However, the adjustment of the dead time will directly affect the output of the PWM signal, thereby causing transient errors in the voltage and current waveforms of the permanent magnet motor, which may cause unstable commutation of the permanent magnet motor, increased torque fluctuation, and even affect the precise control of the permanent magnet motor. Therefore, after adjusting the dead time, real-time error compensation is needed through digital signal processing (DSP) technology.

[0122] The DSP calculates the voltage and current deviation caused by the change of the dead time through high-frequency data acquisition and analysis, and corrects the waveforms based on compensation algorithms (such as PI control, voltage feedforward compensation, etc.) to make the output signal approach the ideal state. This process ensures that even if the dead time is dynamically adjusted, the system output still maintains high precision and high stability, and the operation quality of the permanent magnet motor is not affected by the adjustment of the dead time.

[0123] In summary, the role of this step is to intelligently and adaptively optimize the dead time, and ensure the stability and efficiency of the permanent magnet motor drive system through DSP error compensation, avoid short circuit, power loss increase, permanent magnet motor vibration or harmonic pollution caused by mismatched dead time, and improve the safety, energy efficiency and long-term reliability of the system.

[0124] When the evaluation result shows that the dead time does not match the actual operation, the fuzzy logic controller dynamically adjusts the dead time based on the evaluation result according to the preset fuzzy rules, and at the same time, the specific steps of real-time compensation of the output error caused by the dead time through digital signal processing technology are as follows:

[0125] In the dead time adjustment process, first of all, the fuzzy logic controller (Fuzzy Logic Controller, FLC) is used to calculate the dead time adjustment amount according to the dead time mismatch risk coefficient DTM and the dead time mismatch risk coefficient reference threshold, and the calculation expression is as follows:

[0126] ΔDT=μ f ·(α1·(DTM-TDM ref )+α2·|I spk |+α3·|THD|)

[0127] In the formula, ΔDT is the dead time adjustment amount, which determines the magnitude of the increase or decrease of the dead time, μ f is the fuzzy adjustment factor, which is nonlinearly scaled according to the dead time adjustment requirement to avoid over-adjustment or under-adjustment, α1 is the dead time deviation weight factor, which represents the influence weight of the dead time deviation (DTM-TDM ref ) on the adjustment amount ΔDT, TDM ref is the dead time mismatch risk coefficient reference threshold, I spk is the current peak amplitude, which reflects the transient current impact caused by too short dead time, α2 is the current peak weight factor, which represents the influence weight of the current peak amplitude on the dead time adjustment, THD is the total harmonic distortion, which represents the degree of harmonic pollution caused by too long dead time, and measures the proportion of harmonic components in the output voltage or current, and α3 is the total harmonic distortion weight factor, which represents the influence weight of the total harmonic distortion on the dead time adjustment.

[0128] The above steps first calculate the dead time error (DTM-TDM ref ) to quantify the mismatch degree of the current dead time.

[0129] Then, combined with the peak current and the harmonic distortion, the influence of the mismatch on the motor operation is evaluated.

[0130] Then, the fuzzy logic rules are used to determine the dead-time adjustment direction:

[0131] If DTM > TDM ref and I spk is too high, then the dead-time is appropriately increased;

[0132] If DTM > TDM ref and THD is too high, then the dead-time is appropriately decreased;

[0133] Finally, the dead-time adjustment ΔDT is calculated and sent to the next step for application.

[0134] After determining the dead-time adjustment amount ΔDT, the PWM modulation strategy of the inverter is updated in real time to ensure that the adjustment of the dead-time does not affect the overall stability, and the calculation expression is as follows:

[0135]

[0136] DT new is the updated dead-time, which is optimized by dynamic calculation to reduce the impact of dead-time mismatch on system operation, DT old is the original dead-time, e -ω·DTM is an exponential adjustment term, which is an exponential scaling factor for controlling the adjustment rate of the dead-time, e is the natural base,

[0137] ω is the adjustment response factor, which controls the adjustment speed, is the ripple component of the DC bus voltage, which measures the impact of the dead-time on the stability of the bus, is the rated voltage of the DC bus, which is used to normalize the calculation of the ripple impact, is the normalized DC bus voltage ripple, which measures the deviation of the DC bus voltage ripple (transient change) relative to the rated voltage, is the DC bus voltage ripple weight factor, which measures the impact weight of the DC bus voltage ripple on the dead-time adjustment,

[0138] is the load current commutation dynamic weight factor, which measures the impact of the load current commutation characteristics on the dead-time adjustment, is the rate of change of the load current to the commutation voltage, which evaluates the impact of the dead-time adjustment on the current commutation dynamic response;

[0139] The above steps, first, according to ΔDT calculate the new dead-time DT new , to ensure smooth transition of the dead-time adjustment;

[0140] Then, DT new is applied to the PWM generation unit to recalculate the duty cycle of the three-phase PWM signal;

[0141] Then, monitor the new dead time in real time, if there is abnormal waveform distortion, trigger error compensation mechanism for optimization;

[0142] Finally, ensure that the new dead time can adapt to the system load changes, improve the motor control precision.

[0143] Because the dynamic adjustment of dead time will affect the output voltage and current waveform of motor, it is necessary to use digital signal processing (DSP) technology to compensate the error caused by dead time, to ensure the output quality is not affected, the calculation expression is as follows:

[0144]

[0145] In the formula, V comp is the compensation voltage, used to correct the voltage error caused by dead time adjustment, so that the system output voltage tends to be ideal, θ1 is the dead time change weight factor, which measures the influence degree of dead time adjustment on compensation voltage, θ2 is the voltage error weight factor, which measures the influence degree of output voltage error on compensation voltage, V ideal is the ideal output voltage, V real is the actual output voltage, is the second order voltage rate, which analyzes the dynamic influence of dead time adjustment on voltage fluctuation, θ3 is the voltage second order derivative weight factor, which measures the influence degree of voltage second order derivative on compensation voltage, θ4 is the current ripple weight factor, which measures the influence degree of current ripple on compensation voltage,

[0146] is the normalized current ripple, I ripple is the current ripple amplitude, which measures the output current oscillation degree caused by dead time mismatch, I nominal is the rated current, used to normalize the contribution degree of current ripple to error compensation.

[0147] The above steps, first, calculate the PWM modulation error caused by the change of dead time, calculate the compensation amount through voltage error;

[0148] Then, adjust the PWM waveform, so that the actual output voltage gradually approaches the ideal value, reduces the harmonic pollution and voltage deviation caused by dead time adjustment;

[0149] Finally, monitor the compensation effect in real time, if there is still harmonic distortion, iterative adjustment of compensation parameters.

[0150] The application is based on real-time data acquisition, deep learning intelligent evaluation, fuzzy control dynamic optimization and DSP error compensation, and constructs a self-adaptive dead time regulation mechanism, which can effectively improve the stability, reliability and energy efficiency of the permanent magnet motor system driven by the oil drilling rig. Through data acquisition and preprocessing, the data integrity and accuracy of the input model are ensured, and the accuracy and reliability of the evaluation are improved; the deep learning model is used to mine the complex relationship between the dead time setting and the actual running state, and the intelligent evaluation with high precision is realized; combined with the fuzzy logic controller, the dead time is adaptively adjusted under different operating conditions to avoid the risk of short circuit caused by too short dead time, while reducing the voltage distortion, high-order harmonics and energy loss caused by too long dead time, and optimizing the output performance of the motor. In addition, the DSP error compensation mechanism further corrects the influence of dead time adjustment on the output waveform, ensures the stability of the voltage and current waveform of the inverter, reduces the harmonic loss and reduces the winding overheating phenomenon, thereby prolonging the service life of the motor and improving the overall operating efficiency of the oil drilling rig system. The application not only effectively solves the safety hazards caused by the mismatch of dead time, but also optimizes the performance of the motor drive system through intelligent means, significantly improves the running stability and working efficiency of the permanent magnet motor driven by the oil drilling rig under complex working conditions.

[0151] The above formulas are dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to obtain a formula of the most recent real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0152] The above only describes some exemplary embodiments of the application in a descriptive manner, and it is self-evident that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the application. Therefore, the above figures and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the application.

[0153] It should be noted that in this paper, if there are relationship terms such as first and second, they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or equipment including the element.

[0154] It should be understood that the size of the sequence number of the above processes does not mean the order of execution in various embodiments of the present application, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0155] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0156] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0157] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment of the present application according to actual needs.

[0158] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0159] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0160] The above only describes some exemplary embodiments of the present application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without deviating from the spirit and scope of the present application. Therefore, the above figures and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the present application.

Claims

1. A permanent magnet motor condition monitoring and control system based on oil drilling rig drive, characterized in that: It includes data acquisition and construction module, data preprocessing and feature extraction module, dynamic monitoring and trend analysis module, deep learning evaluation module and fuzzy control and real-time compensation module: The data acquisition and construction module first uses data acquisition equipment to obtain real-time operating data of the inverter during operation. The collected real-time data is stored in a time series format and a data set is established to provide complete and accurate basic data for subsequent analysis and model evaluation. The data preprocessing and feature extraction module preprocesses the inverter operating parameter information in the data set and extracts the core parameters that intuitively reflect the dead time mismatch from the preprocessed data as the key indicators for evaluating whether the dead time is matched; The dynamic monitoring and trend analysis module performs fine-grained analysis on the extracted key indicators within the detection window to capture dynamic change trends caused by dead time mismatch; The deep learning evaluation module inputs the analyzed key indicators into a pre-learned deep learning model, and uses the deep learning model to intelligently evaluate whether the dead time setting matches the complex feature relationship; Fuzzy control and real-time compensation module. When the evaluation results show that the dead time does not match the actual operation, the fuzzy logic controller dynamically adjusts the dead time according to the preset fuzzy rules based on the evaluation results. At the same time, it uses digital signal processing technology to compensate for the output error caused by the dead time in real time, and corrects the output error caused by the dead time change.

2. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 1 is characterized in that: The specific steps for acquiring the inverter's operating data in real time through data acquisition equipment and establishing a data set are as follows: First, deploy high-precision data acquisition equipment in the inverter system to ensure real-time monitoring of the inverter's input, output, and key component status; Secondly, the collected signals are processed synchronously, and the data from all sensors are timestamped using a high-precision clock to ensure the time consistency of the data. Then, a high-sampling-rate data recording system is used to store data in a time series format to capture transient changes caused by dead-time mismatch during inverter operation. Next, we established data sets for merging and classification management, and labeled the data according to the inverter's operating conditions for subsequent analysis. Finally, through data integrity checking, abnormal data points are eliminated and interpolation compensation is performed to ensure the integrity and accuracy of the data set, thereby providing high-quality data support for subsequent feature analysis, deep learning evaluation and dynamic regulation.

3. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 1 is characterized in that: Core parameters that intuitively reflect the dead time mismatch are extracted from the preprocessed data. The extracted parameters include the DC bus voltage fluctuation under the pulse working mode and the actual IGBT turn-off time deviation. The DC bus voltage fluctuation and the actual IGBT turn-off time deviation under the pulse working mode are analyzed within the detection window to generate the DC bus feedback abnormality reference value and the IGBT turn-off delay reference value respectively. The dynamic change trend caused by the dead time mismatch is quantified by the DC bus feedback abnormality reference value and the IGBT turn-off delay reference value.

4. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 3 is characterized in that: The specific steps for analyzing the DC bus voltage fluctuation in the pulse working mode within the detection window to generate the DC bus feedback abnormality reference value are as follows: In pulsed operation mode, the DC bus voltage signal exhibits periodic fluctuations. However, the mismatch in dead time will increase the irregularity of the voltage fluctuation. Therefore, the nonlinear features of the bus voltage signal are first extracted and the accumulated offset of its high-order derivatives is analyzed to quantify the fluctuation trend of the bus voltage. The feature extraction formula is as follows: Where F1 is the nonlinear high-order characteristic index of the DC bus voltage fluctuation, which is used to measure the severity of the DC bus voltage fluctuation. N is the total number of time points in the detection window. is the third-order derivative of the DC bus voltage, V dc (k) is the bus voltage value collected at time point k, α is the nonlinear weight coefficient, which is used to adjust the influence of the third-order derivative term. is the voltage fluctuation exponential decay term, and V is calculated at time point k. dc (k) and the previous moment V dc The voltage variation between (k-1), e is the natural base, and β is the exponential decay factor; Abnormal fluctuations in bus voltage are not only manifested in sudden changes but also lead to uneven bus energy feedback. Therefore, a nonlinear energy offset function is introduced to measure the uneven distribution of bus voltage energy. The formula is as follows: Where F2 is the imbalance index of DC bus energy feedback, E rec (k) is the stored energy at time point k, E del (k) is the energy transmitted from the bus to the load at time k, ε is a small value to prevent the denominator from being zero, ensuring calculation stability, and γ is the exponential adjustment coefficient to control the contribution of bus feedback imbalance to the abnormal index; The nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation and the imbalance index F2 of the DC bus energy feedback are combined, and a nonlinear weight distribution function is introduced to generate the DC bus feedback abnormality reference value. The generation formula is as follows: Where λ1 is the weight coefficient of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, λ2 is the weight coefficient of the imbalance index F2 of the DC bus energy feedback, η1 is the exponential adjustment factor of the nonlinear high-order characteristic index F1 of the DC bus voltage fluctuation, and η2 is the exponential adjustment factor of the imbalance index F2 of the DC bus energy feedback. is the exponential decay factor, e is the natural base, and ζ is a dynamic adjustment parameter used to adjust the exponential decay factor sensitivity.

5. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 3 is characterized in that: The specific steps for analyzing the actual IGBT turn-off time deviation within the detection window to generate the IGBT turn-off delay reference value are as follows: Within the detection window, a high-speed data acquisition system is used to monitor key signals during the IGBT turn-off process, including changes in current, voltage, and power loss. Since the IGBT turn-off process involves rapid current decay, a turn-off transient characteristic function is defined to capture the mismatch between the current drop rate and the voltage rise rate. The formula is as follows: Where, Φ off is the IGBT turn-off transient characteristic, I ds is the drain-source current when the IGBT is turned off, V ce is the collector-emitter voltage when the IGBT is turned off, Q g is the gate charge, is the current drop rate during IGBT turn-off process, is the voltage rise rate during the IGBT turn-off process, and ρ is the weight factor used to balance the influence weights of the current drop rate and the voltage rise rate; During the IGBT turn-off process, its loss characteristics are affected by the dead time. When the dead time is too short, the current does not decay completely, resulting in an increase in the instantaneous power peak. Therefore, the IGBT turn-off loss offset function is introduced to characterize the energy offset during the turn-off process. The calculation formula is as follows: Where, loss is the IGBT turn-off loss offset, Ω off is the IGBT off time region, P sw (ζ) is the actual measured IGBT instantaneous switching loss power, P ideal (ζ) is the theoretically calculated IGBT turn-off loss power, and μ is the adjustment factor used to adjust the proportional factor of the theoretical loss benchmark; Finally, the IGBT turn-off transient characteristic Φ off and IGBT turn-off loss offset ψ loss Comprehensive analysis is performed to generate the IGBT turn-off delay reference value. The generation formula is as follows: Where, ITDI is the IGBT turn-off delay reference value, max(Φ off ) is the maximum IGBT turn-off transient characteristic within the detection window, max(Ψ loss ) is the maximum IGBT turn-off loss offset within the detection window, κ1 is the IGBT turn-off transient characteristic Φ off The weighting factor is κ2, which is the IGBT turn-off loss offset Ψ. loss The weight factor, is the normalized IGBT turn-off transient characteristic, is the normalized IGBT turn-off energy offset. β1 and β2 are both normalized exponential adjustment factors. β2 is used to control the normalized IGBT turn-off loss offset Ψ loss The nonlinear effect in the calculation, β1 is used to control the normalized IGBT turn-off transient characteristic Φ off Nonlinear effects in calculations.

6. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 3 is characterized in that: The analyzed DC bus feedback abnormality reference value and IGBT turn-off delay reference value are input into a pre-learned deep learning model, and a dead time mismatch risk coefficient is generated by the deep learning model. The dead time setting is then intelligently evaluated based on the dead time mismatch risk coefficient.

7. The permanent magnet motor state monitoring and control system based on oil drilling rig drive according to claim 6 is characterized in that: The dead time mismatch risk coefficient generated by the pre-learned deep learning model during the intelligent evaluation of the dead time setting is compared with the preset dead time mismatch risk coefficient reference threshold, and the dead time setting is divided as follows: If the dead time mismatch risk coefficient is greater than the dead time mismatch risk coefficient reference threshold, the current dead time setting is classified as mismatching with actual operation; if the dead time mismatch risk coefficient is less than or equal to the dead time mismatch risk coefficient reference threshold, the current dead time setting is classified as matching with actual operation.

8. The state monitoring and control system for a permanent magnet motor driven by an oil drilling rig according to claim 7, characterized in that: When the evaluation results show that the dead time does not match the actual operation, the fuzzy logic controller dynamically adjusts the dead time according to the preset fuzzy rules based on the evaluation results. At the same time, the output error caused by the dead time is compensated in real time through digital signal processing technology. The specific steps are as follows: In the dead time adjustment process, the fuzzy logic controller is first used to calculate the dead time adjustment amount based on the dead time mismatch risk factor DTM and the dead time mismatch risk factor reference threshold. The calculation expression is as follows: ΔDT=μ f ·(α1·(DTM-TDM ref )+α2·|I spk |+α3·|THD|) Where ΔDT is the dead time adjustment, μ f is the fuzzy adjustment factor, α1 is the dead time deviation weight factor, TDM ref is the dead time mismatch risk factor reference threshold, I spk is the current peak amplitude, α2 is the current peak weight factor, which indicates the influence weight of the current peak amplitude on the dead time adjustment, THD is the total harmonic distortion rate, α3 is the total harmonic distortion weight factor, which indicates the influence weight of the total harmonic distortion rate on the dead time adjustment; After determining the dead time adjustment amount Δ After DT, the PWM modulation strategy of the inverter is updated in real time to ensure that the adjustment of the dead time does not affect the overall stability. The calculation expression is as follows: DT new is the updated dead time, DT old is the original dead time, e -ω·DTM is the exponential adjustment term, which is used to control the exponential scaling factor of the dead time adjustment rate. e is the natural base number. ω is the adjustment response factor, which controls the adjustment speed. is the ripple component of the DC bus voltage, is the rated voltage of the DC bus, is the normalized DC bus voltage ripple, is the DC bus voltage ripple weight factor, is the load current commutation dynamic weighting factor, is the rate of change of load current with respect to commutation voltage; Since the dynamic adjustment of the dead time will affect the output voltage and current waveforms of the motor, it is necessary to use digital signal processing technology to compensate for the error caused by the dead time to ensure that the output quality is not affected. The calculation expression is as follows: Where V comp is the compensation voltage, θ1 is the dead time change weight factor, θ2 is the voltage error weight factor, V ideal is the ideal output voltage, V real is the actual output voltage, is the second-order voltage change rate, θ3 is the voltage second-order derivative weight factor, and θ4 is the current ripple weight factor, which measures the influence of current ripple on the compensation voltage. is the normalized current ripple, I ripple is the current ripple amplitude, I nominal is the rated current, which is used to normalize the contribution of current ripple to error compensation.

Images