Automobile fastener production energy consumption optimization control method and system

Abstract

The application discloses a method and system for optimizing and controlling energy consumption in automobile fastener production, and relates to the technical field of fastener production. The method comprises the following steps: obtaining real-time current waveform signals and power grid fundamental voltage signals of an inverter unit in an energy feedback link, establishing a phase dynamic drift model by using a time difference analysis algorithm, and forming a continuous mapping of energy flow direction changes on a time axis; extracting a continuous change section of an offset rate between the current waveform and the power grid fundamental based on the established phase dynamic drift model, constructing a phase trend prediction function by using a sliding window algorithm, and converting the time characteristics of energy flow direction changes into a phase prediction signal trajectory. The application realizes the synchronization of feedback current and power grid voltage by phase dynamic drift modeling and advance compensation control, reduces ripple current and component stress, extracts energy density change rate and dynamically filters and regulates, makes energy flow smooth and continuous, and realizes efficient and energy-saving control of fastener production.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fastener production, in particular to a method and system for optimizing energy consumption in automobile fastener production. BACKGROUND

[0002] Optimizing energy consumption in automobile fastener production refers to dynamically monitoring, data modeling and intelligently controlling energy consumption links in the production line during the manufacturing process of automobile fasteners (such as bolts, nuts, screws, etc.), so as to minimize the consumption of energy such as electricity, heat, compressed air and cooling water under the premise of ensuring product quality and production capacity. The core idea is to use sensors and industrial Internet of Things technology to collect energy consumption data and process parameters (such as temperature, torque, speed, heating time, etc.) of each device in real time, combine energy consumption distribution model and optimization algorithm, and implement adaptive energy distribution and load balancing control for key links such as heating, forming, heat treatment, electroplating and testing. For example, when the heat treatment furnace is in a low load state, the system can automatically adjust the heating power or delay the start time to avoid energy waste; in the forming process, the system dynamically matches the driving energy according to the material deformation resistance to realize on-demand energy supply. This method can effectively improve the overall energy efficiency by 10%~30%, prolong the service life of the equipment, and realize green and intelligent production.

[0003] The prior art has the following disadvantages:

[0004] In the prior art, fastener production lines generally use energy feedback mode to improve power utilization, but this mode has potential risks of phase detection response lag in the running environment of multiple processes and high-frequency start-stop. When the phase detection of the inverter unit is delayed beyond the set threshold, the current phase in the energy feedback link will deviate from the power grid fundamental wave, causing the feedback energy flow direction to be reversed, resulting in a large amount of ripple current in a short time. Such abnormal ripples not only make the filter capacitor, power tube and other components of the inverter unit bear transient stress exceeding the design limit, but also may cause overheating of the energy recovery module, out-of-sync of the drive circuit, even equipment burning, which seriously affects the stability and safety of energy consumption optimization control.

[0005] Therefore, how to effectively suppress the energy flow reversal and abnormal ripple current caused by phase detection delay in the energy feedback process has become a key technical problem that the prior art needs to solve.

[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 automobile fastener production energy consumption optimization control method and system can solve the problems in the background art.

[0008] To achieve the above-mentioned purpose, the present application provides the following technical solutions: an automobile fastener production energy consumption optimization control method, comprising the following steps:

[0009] Step one, obtaining the real-time current waveform signal of the inverter unit in the energy feedback link and the grid fundamental voltage signal, using the time difference analysis algorithm to establish a phase dynamic drift model, forming a continuous mapping of the energy flow direction change on the time axis, which is used to provide a stable reference baseline for phase prediction;

[0010] Step two, based on the established phase dynamic drift model, extracting the continuous change section of the offset rate between the current waveform and the grid fundamental, using the sliding window algorithm to construct a phase trend prediction function, converting the time characteristics of the energy flow direction change into a phase prediction signal trajectory, to realize the early identification of the instantaneous segment that may occur phase delay;

[0011] Step three, according to the phase prediction signal trajectory generated by the phase trend prediction function, using the phase advance compensation algorithm to dynamically adjust the pulse width modulation trigger time of the inverter unit, so that the phase of the feedback current and the phase of the grid voltage keep synchronous flow, thereby weakening the formation conditions of the reverse impact current in the feedback link;

[0012] Step four, based on the feedback current signal adjusted by the phase advance compensation algorithm, extracting the energy density change rate, and using the dynamic filtering gating algorithm to reduce the high-frequency oscillation components of the feedback energy signal, to form a smooth and stable energy transmission path, ensuring the continuity of the energy flow under different load fluctuations;

[0013] Step five, according to the smooth energy signal output by the dynamic filtering gating algorithm, real-time updating the parameters of the phase dynamic drift model, realizing the adaptive adjustment of the energy feedback link, so that the system continuously maintains the stability and safety of the energy flow direction under complex working conditions, thereby realizing efficient energy feedback control.

[0014] Preferably, the step of obtaining the real-time current waveform signal of the inverter unit in the energy feedback link and the grid fundamental voltage signal comprises:

[0015] Current detection points and voltage detection points are arranged at the output side and the grid connection side of the inverter unit in the energy feedback link, respectively, to synchronously obtain the real-time current waveform signal of the inverter unit output end and the grid side fundamental voltage signal, and ensure that the sampling period is triggered at the same time under the same time reference;

[0016] The phase difference at each sampling time is determined by point-by-point comparison of the current waveform signal and the grid fundamental voltage signal on the time axis, and a phase difference time sequence is formed with time as the horizontal axis and the phase shift angle as the vertical axis;

[0017] The phase difference time sequence is mapped to the time axis to form a continuous curve of energy flow direction change, and the instantaneous trend of energy flow direction switching is identified according to the phase difference change rate;

[0018] A phase dynamic drift model is established based on the continuous curve of energy flow direction change, and a continuously updated phase dynamic drift model is formed by comparing the phase change reference point with the local extreme point.

[0019] Preferably, the step of extracting the continuous change section of the offset rate between the current waveform and the grid fundamental based on the phase dynamic drift model comprises:

[0020] The phase offset between the current waveform and the grid fundamental voltage is continuously extracted in the established phase dynamic drift model, and the section with a higher phase change rate on the time axis is identified to form a phase offset rate change identification sequence;

[0021] The phase change curve in the offset rate change section is continuously segmented and analyzed to extract the phase change directionality and stability characteristics, and to exclude unstable factors caused by grid fluctuations or current jitter;

[0022] The phase change directionality curve is mapped to the time axis to convert the time law of energy flow direction change into a continuous trend signal trajectory to identify the acceleration interval and the stable interval of the phase change;

[0023] The trend signal trajectory is analyzed and dynamically monitored in real time, and when the trajectory slope sharply rises in a short time, the time period is marked as a potential phase delay interval to realize early identification of the possible phase delay instantaneous segment.

[0024] Preferably, the step of dynamically adjusting the pulse width modulation trigger time of the inverter unit according to the phase prediction signal trajectory generated by the phase trend prediction function comprises:

[0025] The phase synchronization state of the feedback current and the grid voltage is analyzed in real time according to the phase prediction signal trajectory to determine the deviation direction and degree of the current phase relative to the grid fundamental voltage phase;

[0026] After the analysis of the phase deviation direction and degree is completed, the pulse width modulation trigger time of the inverter unit is preliminarily adjusted and prepared, and the best advance amount of the conduction trigger time is determined according to the position of the phase change inflection point in the phase prediction signal trajectory;

[0027] After determining the adjustment range of the trigger time, the pulse width modulation signal of the inverter unit is continuously and dynamically adjusted to make the phase of the feedback current waveform and the grid voltage waveform time-synchronized and achieve real-time calibration;

[0028] After completing the dynamic adjustment of the pulse width modulation trigger time, the phase difference, peak value corresponding relationship and waveform symmetry of the adjusted feedback current waveform and the grid voltage waveform are verified in real time to ensure that the phase synchronization effect of the energy feedback process is continuously stable.

[0029] Preferably, during the continuous and dynamic adjustment of the pulse width modulation trigger time of the inverter unit, the trigger time changes of multiple driving cycles are smoothed to make the trigger adjustment change slowly to avoid waveform oscillation. When the phase difference fluctuation exceeds the preset range, the phase prediction signal trajectory is updated to calculate a new trigger advance amount, thereby ensuring the phase synchronization stability of the feedback current and the grid voltage.

[0030] Preferably, the step of extracting the energy density change rate based on the feedback current signal adjusted by the phase lead compensation algorithm and performing high-frequency oscillation component reduction processing includes:

[0031] The instantaneous energy density characteristics are extracted from the feedback current signal adjusted by the phase lead compensation, and a distribution curve of energy density change is established on the time axis to reflect the transmission state of the energy flow in the feedback link;

[0032] The energy density change distribution curve is continuously differentiated to extract the energy density change rate time sequence and identify the dynamic trend of the energy flow intensity change over time;

[0033] The energy density change rate time sequence is dynamically smoothed to reduce high-frequency oscillation components and form a gentle trend curve of energy change, so that the energy flow curve remains in a continuous transition state;

[0034] The smoothed energy signal is remapped on the time axis to construct a stable energy transmission path, and the continuity of the path is verified in real time to ensure the smoothness and continuity of the energy flow under different load fluctuations.

[0035] Preferably, in the step of dynamically smoothing the energy density change rate time sequence, the smoothing process is gradual, which retains the low-frequency energy change trend while reducing high-frequency oscillation components, so that the energy transmission path remains in a continuous transition on the time axis, and the time scale of the smoothing interval is automatically adjusted when the energy flow direction changes, to ensure that the energy feedback link still maintains a stable flow state under load mutation conditions.

[0036] Preferably, the step of updating the phase dynamic drift model parameters in real time according to the smoothed energy signal output by the dynamic filtering gating algorithm includes:

[0037] Based on the smooth energy signal output after dynamic filtering and gating processing, the running state of the current energy feedback link is identified and feature extracted to obtain energy flow amplitude, energy flow direction change rate, energy flow stable interval and energy flow mutation point position;

[0038] The energy flow feature parameters extracted from the smooth energy signal are compared with the historical parameters in the phase dynamic drift model to identify the deviation between the energy flow state and the phase change model, and the time difference between the energy flow feature change and the phase offset change is calculated;

[0039] After identifying the deviation, the parameters of the phase dynamic drift model are updated step by step, and the energy flow intensity, direction change rate and stable interval are continuously fed back to the model to realize adaptive adjustment of the model parameters and maintain smooth transition;

[0040] After completing real-time parameter updating, the synchronization relationship between the phase dynamic drift model output before and after updating and the smooth energy signal is verified to ensure the stability of the energy flow direction and the safety of the electrical operation.

[0041] Preferably, during the step-by-step updating process of the phase dynamic drift model parameters, when the energy flow direction change rate exceeds the preset threshold, the phase drift rate and direction correction coefficient in the model are adjusted first, and the parameter smooth transition time is automatically extended at the energy flow mutation point to prevent phase fluctuations caused by excessive model response, thereby ensuring the stable operation of the energy feedback link under high-frequency start-stop working conditions.

[0042] The automobile fastener production energy consumption optimization control system comprises a phase dynamic modeling module, a phase trend prediction module, a phase synchronization compensation module, an energy signal smoothing module and an adaptive adjustment module.

[0043] The phase dynamic modeling module acquires real-time current waveform signals and grid fundamental voltage signals of an inverter unit in the energy feedback link, and establishes a phase dynamic drift model using a time difference analysis algorithm to form a continuous mapping of energy flow direction change on the time axis.

[0044] The phase trend prediction module extracts the offset rate continuous change section between the current waveform and the grid fundamental based on the established phase dynamic drift model, and uses a sliding window algorithm to construct a phase trend prediction function to convert the time characteristics of energy flow direction change into a phase prediction signal trajectory.

[0045] The phase synchronization compensation module uses a phase lead compensation algorithm to dynamically adjust the pulse width modulation trigger time of the inverter unit according to the phase prediction signal trajectory generated by the phase trend prediction function, so that the phase of the feedback current and the phase of the grid voltage remain in synchronous flow.

[0046] The energy signal smoothing module extracts the energy density change rate based on the feedback current signal adjusted by the phase lead compensation algorithm, and uses a dynamic filter gating algorithm to reduce high-frequency oscillation components of the feedback energy signal to form an energy transmission path.

[0047] The adaptive adjustment module updates the parameters of the phase dynamic drift model in real time according to the smoothed energy signal output by the dynamic filter gating algorithm, and performs adaptive adjustment on the energy feedback link.

[0048] In the above technical solution, the present application provides technical effects and advantages:

[0049] The present application establishes a phase dynamic drift model and combines phase trend prediction and lead compensation control to keep the current phase in the energy feedback link and the grid voltage in continuous synchronization, effectively avoiding the energy flow reverse impact phenomenon caused by phase response lag. This control method can correct the phase offset in real time in a complex production environment with multiple processes and high-frequency start-stop, so that the feedback current waveform is always in a stable conduction state, thereby significantly reducing the ripple current amplitude, reducing the transient stress of the filter capacitor and power tube of the inverter unit, and improving the stability of energy feedback operation and the service life of the equipment.

[0050] The present application introduces energy density change rate extraction and dynamic filter gating mechanism in the feedback current signal processing process to keep the energy transmission path smooth and continuous under different load fluctuations, and realizes real-time adaptive adjustment of energy flow. This method can dynamically balance the energy flow between energy feedback and equipment load, significantly improve energy utilization and feedback efficiency, and avoid heat loss and oscillation risk caused by energy accumulation or mutation, so that the entire fastener production line reaches a dynamic optimal state in terms of energy-saving control and operation safety. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced. 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.

[0052] Figure 1 The method flow chart of the automobile fastener production energy consumption optimization control method of the present application.

[0053] Figure 2 The module schematic diagram of the automobile fastener production energy consumption optimization control system of the present application. DETAILED DESCRIPTION

[0054] Example implementations are now described with reference to the drawings. Example implementations can, however, 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 inventive gist to those skilled in the art.

[0055] The application provides an automobile fastener production energy consumption optimization control method as shown in the drawings, comprising the following steps: Figure 1 The application provides an automobile fastener production energy consumption optimization control method as shown in the drawings, comprising the following steps:

[0056] Step one, obtaining the real-time current waveform signal of the inverter unit in the energy feedback link and the fundamental voltage signal of the power grid, using a time difference analysis algorithm to establish a phase dynamic drift model, forming a continuous mapping of energy flow direction change on the time axis, which is used to provide a stable reference baseline for phase prediction;

[0057] The specific implementation of this step is as follows:

[0058] Current detection points and voltage detection points are arranged at the output side and the power grid connection side of the inverter unit in the energy feedback link, respectively, to synchronously obtain the real-time current waveform signal of the inverter unit output end and the fundamental voltage signal of the power grid side. To ensure the authenticity and time consistency of the signal data, the current detection point should be arranged close to the inverter unit output bus, and the voltage detection point should be directly connected to the power grid input node, so that the collected signals cover the complete energy feedback path from the inverter unit output end to the power grid input end. During data collection, the sampling period of the current signal and the voltage signal is strictly set to be triggered at the same time, so as to avoid phase errors caused by asynchronous sampling. To ensure that subtle changes in the current waveform can be accurately recorded, the sampling resolution should be sufficient to reflect the small amplitude changes and transition edge characteristics of the waveform. For example, when the inverter unit is rapidly switched between light load and full load, the current waveform will have a sudden change in slope and waveform distortion in a short time, and these transition details can be completely captured through high-resolution sampling. At the same time, the collection of voltage signals should ensure the integrity of the waveform, avoiding the distortion of the fundamental waveform form caused by electromagnetic interference or sampling noise. Through the full-time domain synchronous collection of current signals and voltage signals, two groups of data sequences with unified time stamps can be obtained, providing a stable data foundation for the subsequent establishment of the time mapping of energy flow direction.

[0059] After the synchronization acquisition of the current signal and the voltage signal, the corresponding relationship of the two on the time axis is compared point by point to determine the phase difference at each sampling time. In this process, the voltage zero-crossing point of the current waveform is time-aligned with the zero-crossing point of the grid voltage waveform, and the degree of advance or lag of the current waveform relative to the voltage waveform is calculated within each sampling interval. Through this time alignment, an initial phase difference sequence is obtained, with time as the horizontal axis and the phase shift angle as the vertical axis. The energy feedback device in the fastener production line frequently starts and stops, and the phase difference between the current waveform and the voltage waveform will change significantly in a very short time. This change is related to multiple factors such as load switching, heating of the heat treatment furnace, and starting of the forming machine. In order to ensure the continuity of the sequence, the phase difference in each sampling period needs to be smoothly transitioned with the result of the previous period to eliminate the phase jump error caused by signal mutation or noise. The final phase difference time sequence not only reflects the current response lag characteristics in the energy feedback process, but also provides an accurate time reference for the subsequent analysis of the energy flow direction change trend.

[0060] After obtaining the phase difference time sequence of the current waveform and the voltage waveform, the sequence is mapped one-to-one with the time axis to form a continuous curve of energy flow direction change. This mapping process needs to match each time sampling point of the current waveform with the corresponding voltage waveform sampling point, and determine the energy flow direction according to the positive and negative direction of the phase difference. When the phase of the current waveform leads the phase of the voltage waveform, it indicates that energy flows from the inverter unit to the grid, which belongs to forward energy feedback; when the phase of the current waveform lags behind the phase of the voltage waveform, it indicates that energy flows from the grid into the inverter unit, which belongs to reverse energy absorption. On the fastener production line with multiple processes running in parallel, these two energy flow states will frequently alternate, especially during the rapid start-stop of the forming machine or the heating stage of the heat treatment equipment. The switching of energy flow direction often occurs within milliseconds. In order to accurately depict this rapid switching process, the boundary points of energy flow direction change need to be identified on the mapping curve, and the instantaneous trend of energy flow direction switching is calculated according to the phase difference change rate of these boundary points. When the slope of the mapping curve changes from positive to negative, it indicates that the energy flow direction changes from output to absorption; when the slope changes from negative to positive, it indicates that the energy feedback restarts. In this way, a continuous mapping trajectory of energy flow direction change is formed on the time axis, which can intuitively reflect the energy flow law in the entire feedback process. The mapping curve not only reflects the stability of energy flow, but also reveals the dynamic response characteristics of energy flow under different load changes.

[0061] After obtaining the continuous mapping of the energy flow direction change, the phase difference change characteristics on the entire time axis are modeled to form a phase dynamic drift model. The model is based on the continuous mapping results of the energy flow direction, and the phase difference value, phase change rate and energy flow direction identification at each time point are associated in multiple dimensions, thereby establishing a time domain model that can dynamically describe the energy flow direction change trend. In the model establishment process, first, the reference points of phase change on the time axis are determined, which usually correspond to the critical time of energy flow direction change. By comparing these reference points with the local extreme points in the continuous mapping curve, the key time window of the change of energy flow direction from positive feedback to reverse absorption can be identified. Then, the phase change rate in these time windows is continuously tracked, so that the model can reflect the inertia and delay characteristics of the phase change in the energy feedback link. When the energy feedback link is affected by load disturbance, power grid fluctuation or temperature rise of the inverter unit itself, the phase dynamic drift model will automatically adjust its drift trend on the time axis to maintain the real-time correspondence between the phase difference change and the energy flow direction. Finally, by time-calibrating the phase drift characteristics of each update with the mapping results of the previous period, a continuously updated phase dynamic drift model can be formed, so that the mapping curve of the energy flow direction remains continuous, smooth and high-precision in long-term operation.

[0062] It should be noted that:

[0063] The time difference analysis algorithm refers to the calculation method of accurately aligning and calculating the difference between the relative time sequence of the inverter output current waveform and the grid fundamental voltage waveform in the time domain, so as to determine the phase offset at any sampling time, and to establish the calculation method of the phase dynamic change relationship. The core idea is to compare the time difference of the zero-crossing points, peak points and transition edge positions of the two sets of synchronous sampling signals, and convert the time offset into a phase difference value, thereby forming a continuous and traceable phase difference curve on the time axis.

[0064] For example, during the operation of the energy feedback link, when the inverter unit outputs energy to the power grid, the current waveform should theoretically be in phase with the power grid voltage waveform (i.e., the peaks of the two should appear at the same time). If the peak of the current signal appears later than the peak of the voltage signal, for example, the voltage peak appears at 10 ms and the current peak appears at 10.5 ms in a 50 Hz power grid cycle, there is a time difference of 0.5 ms between them. Through the time difference analysis algorithm, the 0.5 ms time difference is converted into a phase angle difference, i.e., the phase lag angle is about 9° (the calculation formula is: phase angle = time difference / cycle × 360°). This phase lag angle reflects the degree of dynamic delay of the current response relative to the voltage change. If the time difference further increases to 0.8 ms in the subsequent sampling period, it indicates that the phase lag trend of the energy feedback link is increasing, and the system may have the risk of energy flow reversal or increased current ripple.

[0065] In practical applications, the time difference analysis algorithm not only calculates the phase difference at a single time, but also tracks the time difference change rate in consecutive sampling periods to form a dynamic trajectory of the phase change over time, which is used to identify the trend of the slight drift of the energy flow direction. For example, in the fastener heat treatment process, when the heating load suddenly switches, the waveform of the energy feedback current will be temporarily mismatched, and the time difference analysis algorithm can immediately capture the transient change from 0.3 ms to 0.7 ms and record this phase drift into the phase dynamic drift model. In this way, the system can accurately depict the dynamic change law of the energy flow direction and achieve high-precision monitoring and modeling of the energy feedback phase state.

[0066] Through the above continuous implementation steps, the phase relationship between the output current of the inverter unit and the grid voltage in the energy feedback link can be comprehensively perceived, and an accurate mapping of the energy flow direction on the time axis can be formed. This process not only reveals the actual flow state of energy in the feedback link, but also stably tracks the phase change trend under the complex working conditions of high-frequency start-stop of the production line, thereby providing a solid foundation for subsequent phase synchronization and energy optimization. Through the established phase dynamic drift model, the energy feedback link can maintain a stable correspondence between the current and voltage phases during operation, avoiding sudden changes in the energy flow direction and current reverse impact problems caused by phase detection delay, making the energy feedback control in the production process of automotive fasteners more secure, smooth and efficient.

[0067] Step two, based on the established phase dynamic drift model, extract the continuous change section of the offset rate between the current waveform and the grid fundamental wave, use the sliding window algorithm to construct the phase trend prediction function, and convert the time characteristics of the energy flow direction change into a phase prediction signal trajectory to realize the early identification of the instantaneous segment that may have a phase delay;

[0068] The specific implementation of this step is as follows:

[0069] In the established phase dynamic drift model, the phase offset between the current waveform and the grid fundamental voltage is continuously extracted, and the section with high phase change rate is identified on the time axis. By analyzing the continuous mapping results of the energy flow direction change in the phase dynamic drift model, it can be clearly observed that the phase offset shows a periodic or non-periodic change trend within a certain period of time. When the energy feedback link is in a stable energy flow state, the phase difference between the current waveform and the grid voltage waveform remains in a relatively constant range, and the slope of the offset rate change curve is small; when the load state in the production line suddenly changes, for example, the forming machine enters the stamping or the heat treatment furnace switches from constant temperature to heating stage, the phase difference change rate will suddenly increase, and the offset rate curve will appear obvious steep rise or steep drop. At this time, these time sections with rapid changes need to be marked on the time axis of the phase dynamic drift model to form a preliminary phase offset rate change identification sequence, providing a data basis for subsequent trend identification. In order to ensure the continuity of identification, the sampling time interval needs to be accurately controlled, so that the time interval between each sampling point is less than the phase change response time, so as to ensure that the complete phase change trajectory can still be captured when the energy flow direction is quickly switched.

[0070] After completing the preliminary identification of the offset rate change section, the phase change curve in these sections is continuously analyzed and segmented to extract the directionality and stability characteristics of the phase change. Specifically, the phase change curve is divided into multiple isochronous sections in chronological order, and the direction of the phase offset change is calculated in each section. When the phase offset increases with time, it means that the current waveform gradually lags behind the grid voltage, and the energy flow direction may change from positive feedback to reverse absorption; when the phase offset decreases with time, it means that the current waveform gradually leads, and the energy flow direction is in the transition process from absorption to feedback. By continuously analyzing the direction of phase change in each section, a time curve describing the phase change trend can be formed, which can reflect the directionality characteristics of the phase difference change with time in the phase dynamic drift model. In addition, the smoothness of the phase change rate should also be analyzed in each time section. When the change rate curve appears irregular fluctuations, it means that there are transient disturbances in this time period, such as short-term fluctuations of grid voltage or current jitter of inverter output. Such unstable factors need to be identified and excluded to ensure that the subsequent trend prediction is based on smooth phase change data.

[0071] After obtaining the stable phase change directionality curve, the time characteristics of these continuous change segments are mapped to convert the time regularity of energy flow direction change into a predictable signal trajectory. In this process, the phase shift rate at each time point in the phase change curve needs to be compared with the change trend in the time interval before and after it to determine the phase change trend stage in which the time point is located. For example, when the phase change rate continues to rise and the phase shift amount approaches the stable threshold set in the model, it indicates that the system is about to enter the phase delay stage; when the phase change rate begins to decline and the shift amount falls back, it indicates that the system is recovering the phase synchronization state. In this way, the time characteristics of the phase change can be converted into a continuous trend signal trajectory that reflects the acceleration interval, deceleration interval and stable interval of the phase change on the time axis. Especially in the fast switching stage of the energy feedback process, this trend signal trajectory can clearly identify the moment when the phase change is about to reach the lag critical point, providing a direct basis for subsequent advance recognition. In order to ensure the continuity of the trajectory, dynamic smoothing processing needs to be performed on the time sequence to make the phase trend change form a natural transition on the time axis without sudden faults.

[0072] After obtaining the phase trend signal trajectory, real-time analysis and dynamic monitoring are performed on the trajectory to identify the instantaneous segments that may occur phase delay. By continuously observing the slope change of the trend trajectory, when the slope changes from slow rise to steep increase in a short time, it indicates that the change speed of energy flow direction has obviously accelerated, and the system enters the phase response lag risk area. At this time, this time period can be marked as a potential phase delay interval and recorded in the model. When the energy feedback link continues to run, these marked time intervals form a set of dynamically updated time segment collection, each segment corresponding to a potential risk of phase change anomaly. By continuously tracking these time segments, the moment when phase synchronization may occur can be identified in advance, thereby providing a time warning signal for the regulation of the energy feedback link. In order to make the identification process have continuity and self-adaptability, the trend trajectory needs to be updated with the latest current waveform and voltage waveform data during the operation of the energy feedback link, so that the timeliness of phase trend prediction can be maintained. When the load working condition, energy flow intensity or power grid fluctuation state changes, the phase trend signal trajectory can automatically extend or shrink to reflect the latest change trend of energy flow direction.

[0073] It should be noted that:

[0074] The phase trend prediction function is constructed by using a sliding window algorithm, which means that on the basis of the phase dynamic drift model, the phase offset data of the continuously collected current waveform and the power grid fundamental voltage signal are divided into multiple time windows with fixed length and mutual partial overlap according to the time sequence, the phase change rate and directionality in each window are dynamically calculated and trend fitting is performed, so that a continuously predictable phase change trend function is formed in the time dimension. The core of the method is that through the segment-by-segment analysis and continuous updating of the sliding window, the transformation from “discrete detection” to “continuous prediction” of the phase change trend is realized, so that the system can identify the formation process of phase delay or phase drift in advance.

[0075] For example, in the operation of an automobile fastener production line, when the heat treatment furnace switches from the constant temperature state to the heating stage, the phase of the feedback current will show a slow lagging trend. If the full-time domain data is directly analyzed, it is easy to be disturbed by local mutations and cause distortion of the trend judgment; while the sliding window algorithm divides the time axis into multiple overlapping windows, for example, each window has a length of 10 ms and an overlapping area of 5 ms. In each window, the average value, change rate and direction feature of the phase offset rate are calculated, and a local trend curve of the phase change is fitted in a linear or nonlinear way. When the next window slides in, the system automatically discards the old data of the previous window and adds new sampling values, so that the phase trend prediction function remains continuous and real-time in time. If the phase change rate continues to rise in adjacent windows, the prediction function will show that the phase delay trend is intensifying, and the system can determine in advance that the feedback current will lag behind the grid voltage.

[0076] The trend prediction mechanism constructed by the sliding window can smooth short-time noise and retain long-term trends, so that the energy feedback control can dynamically identify the phase abnormal interval within milliseconds of response time. For example, when it is detected that the phase offset rate in three consecutive windows exceeds the set threshold, the system can trigger the compensation control in advance to avoid the reversal of energy flow direction or the sudden increase of ripple current, thereby realizing real-time prediction and active defense of phase response lag.

[0077] Through the above specific implementation steps, the change of the offset rate between the current waveform and the power grid fundamental voltage can be continuously identified and trend extracted on the basis of the established phase dynamic drift model, and the time law of the change of the energy flow direction can be converted into an intuitive and predictable signal trajectory. This process not only realizes the early identification of the phase change trend in the energy feedback link, but also effectively predicts the time period when the phase response lag may occur, so that the energy feedback control has the ability of predictive regulation under dynamic working conditions, thereby avoiding the phenomenon of energy reverse impact caused by phase asynchronization, and ensuring the stability and safety of energy flow in the complex operation environment of the automobile fastener production line with multiple processes and high frequency start-stop.

[0078] Step three, according to the phase prediction signal trajectory generated by the phase trend prediction function, the pulse width modulation trigger time of the inverter unit is dynamically adjusted by using the phase lead compensation algorithm, so that the phase of the feedback current and the phase of the grid voltage keep synchronous flow, thereby weakening the forming conditions of the reverse impact current in the feedback link;

[0079] The specific implementation of this step is as follows:

[0080] According to the phase prediction signal trajectory obtained in the foregoing, the phase synchronization state of the feedback current and the grid voltage is analyzed in real time. By reading the time characteristics of the phase change in the signal trajectory, the deviation direction and the deviation degree of the current phase relative to the grid fundamental voltage phase can be determined. When the phase prediction signal trajectory shows that the phase deviation rate of the current waveform is in the rising stage, it means that the phase of the feedback current gradually lags behind the grid voltage, and if it is not adjusted, it will tend to reverse the direction of energy flow. When the trajectory shows that the phase deviation rate is in the falling stage, it means that the phase of the feedback current gradually leads, and the system has the risk of excessive energy feedback. By continuously monitoring the directional change of the phase deviation, the time when the feedback current and the grid voltage will be out of synchronization can be determined on the time axis. In order to ensure the accuracy of the analysis result, it is necessary to ensure that the data of the phase prediction signal trajectory is strictly synchronized with the real-time waveform of the energy feedback link, and each sampling point must correspond to the current actual voltage and current state, so as to avoid misjudgment caused by time drift.

[0081] After completing the analysis of the phase deviation direction and degree, the pulse width modulation trigger time of the inverter unit is preliminarily adjusted. The core of the adjustment is to determine the optimal trigger point of each inverter drive cycle, so that the conduction time of the inverter unit is time-corresponding to the peak position of the grid voltage. Specifically, the time sequence of the phase prediction signal trajectory is analyzed, and the inflection point position of the phase change in each cycle is extracted. These inflection points reflect the critical boundary of the synchronization state of the current phase and the voltage phase. By matching the inflection point time with the drive trigger cycle of the inverter unit, the optimal trigger advance can be determined under the current energy flow direction. When the energy flow direction changes quickly and the phase lag trend is obvious, the conduction trigger time of the inverter unit should be appropriately advanced, so that the rising edge of the output current is coincided with the peak value of the positive half cycle waveform of the grid voltage, thereby realizing the synchronous following of the current phase; on the contrary, when the energy flow direction appears a slight leading trend, the trigger time should be appropriately delayed to avoid the over-compensation effect caused by the current waveform exceeding the voltage waveform. In this process, each adjustment of the trigger time relies on the time information of the phase prediction signal trajectory, so that the adjustment process has foresight and dynamic response.

[0082] After determining the adjustment range of the trigger moment, the pulse width modulation signal of the inverter unit is continuously and dynamically adjusted, so that the phase of the feedback current waveform gradually tends to be consistent with the grid voltage waveform. In this process, each on period of the inverter unit needs to be synchronously corrected according to the real-time changes of the phase prediction signal trajectory, so that the rising edge of the pulse width modulation signal maintains a constant time difference with the zero-crossing point of the grid voltage waveform. When it is detected that the current phase lag trend is intensified, the adjustment logic will trigger the inverter unit's drive signal in advance, so as to make the feedback current waveform move forward as a whole and reduce the current lag angle; when it is detected that the current phase leads, the trigger signal is appropriately delayed, so that the peak value of the feedback current is re-aligned with the grid voltage waveform. Through this continuous and dynamic trigger correction method, the phase of the feedback current and the grid voltage can be real-time calibrated during operation. In order to avoid waveform oscillation in the trigger adjustment process, the trigger moment changes of continuous multiple drive periods are smoothed, so that the trigger adjustment presents a slow change characteristic, ensuring the stability of the energy feedback link. Through this process, the feedback current waveform can accurately follow the fluctuations of the grid voltage on the time axis, realizing the true phase synchronization flow.

[0083] After completing the dynamic adjustment of the pulse width modulation trigger moment, the adjusted feedback current waveform is verified in real time to ensure that the phase synchronization effect in the energy feedback process is maintained. The verification process includes comparing the phase difference, peak value correspondence and waveform symmetry of the current waveform and the grid voltage waveform before and after adjustment. When the verification result shows that the phase difference between the current waveform and the voltage waveform remains constant and consistent in the whole period, it means that the feedback link is in a stable synchronization state; when it is detected that the phase difference fluctuates beyond the preset range within a short time, it means that the trigger adjustment has not completely offset the influence of load disturbance or grid fluctuation, and the phase prediction signal trajectory needs to be updated to recalculate the trigger advance amount of the next period. Through this continuous verification and feedback method, the synchronization relationship between the feedback current and the grid voltage can be continuously corrected during operation, thereby maintaining long-term stability. After a long time of operation of the energy feedback link, the dynamic adjustment mechanism can automatically adapt to changes in different working conditions in the production line, such as frequent start-stop of the forming machine, load fluctuation of the heat treatment equipment, slight imbalance of the grid voltage, and other complex situations, so that the feedback current and the grid voltage always maintain the same phase, avoiding the phenomenon of energy reverse impact caused by phase delay.

[0084] It should be noted that:

[0085] The phase advance compensation algorithm refers to predicting and pre-correcting the real-time deviation trend of the current phase relative to the grid voltage phase, adjusting the pulse width modulation (PWM) trigger time of the inverter unit in advance, making the phase of the feedback current slightly ahead of the grid voltage phase in time, and thus compensating in advance when the energy flow is about to lag, ensuring that the current and voltage flow synchronously. Its essence is a dynamic advance conduction control mechanism based on time prediction and phase correction, aiming to weaken the phase delay and reverse impact in the energy feedback process.

[0086] For example, during the operation of an automobile fastener production line, when the heat treatment equipment suddenly starts, causing the load power to rise sharply, the output current of the inverter unit often lags behind the grid voltage by about 5° to 10°. At this time, the system detects that the deviation rate in the phase prediction signal trajectory is rapidly rising, i.e., the current phase is about to further lag. The phase advance compensation algorithm will fine-tune the PWM trigger time in advance, so that the inverter unit starts to conduct about 0.5 ms before the grid voltage peak arrives. In this way, the rising edge of the feedback current can be aligned in advance with the peak position of the grid voltage, thereby reducing the phase lag angle from 10° to about 2°. If the grid voltage subsequently shows a downward trend, the algorithm automatically delays the trigger time according to the trend prediction result, avoiding over-compensation caused by the feedback current waveform exceeding the voltage waveform in advance.

[0087] Through this dynamic "pre-correction-real-time verification-continuous smoothing" method, the system can automatically maintain phase synchronization within milliseconds, making the current waveform accurately follow the changes in the grid voltage, effectively avoiding the reverse energy impact and current ripple amplification caused by phase delay, and maintaining the smooth operation of the energy feedback link in complex scenarios such as multi-process start-stop and rapid load fluctuations.

[0088] Through the above continuous implementation steps, the phase prediction signal trajectory and the pulse width modulation trigger logic of the inverter unit form a real-time closed loop, thereby maintaining the high synchronization of the feedback current and grid voltage phase during the energy feedback process. This process not only effectively weakens the formation conditions of reverse impact current in the energy feedback link, but also significantly reduces the transient stress on the power devices of the inverter unit, improving the stability of the energy feedback process and the service life of the equipment. Through this implementation, the automobile fastener production line can achieve continuous and stable energy feedback control in a multi-process high-frequency start-stop operation environment, achieving a dynamic balance between energy saving and safety in the production process.

[0089] Step four, based on the feedback current signal adjusted by the phase advance compensation algorithm, extract the energy density change rate, and use the dynamic filter gating algorithm to reduce the high-frequency oscillation components of the feedback energy signal, to form a smooth and stable energy transmission path, ensuring the continuity of energy flow under different load fluctuations;

[0090] The specific implementation of this step is as follows:

[0091] The instantaneous energy density feature is extracted from the feedback current signal after phase lead compensation adjustment, and a preliminary distribution curve of energy density change is established on the time axis. Since the phase lead compensation has made the feedback current waveform and the grid voltage waveform achieve synchronous flow, the feedback current signal collected at this stage has high time accuracy and phase consistency, and can truly reflect the energy transmission state in the feedback link. In the specific implementation, the feedback current signal needs to be point-by-point corresponding to the grid voltage signal at the corresponding time, and the energy transient change trend at each time is calculated, so as to obtain the energy density sequence reflecting the time change of energy flow intensity. In the multi-process operation environment of the production line, for example, in the pressurization stage of the forming equipment, the heating stage of the heat treatment equipment or the stable stage of the electroplating process, the energy density will fluctuate with the change of load, so when establishing the energy density curve, it is necessary to ensure that the sampling frequency is high enough to capture these short-time change processes. By continuously arranging the energy density data at each time on the time axis, a complete energy density change distribution curve can be formed, and the fluctuation amplitude and change slope of the curve reflect the stability of energy flow in different time periods, providing a basis for subsequent change rate extraction.

[0092] The established energy density change distribution curve is subjected to continuous difference analysis to extract the time sequence of energy density change rate. The key of this step is to capture the dynamic trend of energy density change, not just the static value. By analyzing the difference between adjacent time points in the energy density curve, the rate at which energy density increases or decreases over time can be identified. When the energy density change rate is positive and the value is large, it means that energy is rapidly increasing in a short period of time, and there may be a risk of energy accumulation in the feedback link; when the energy density change rate is negative and the change speed is fast, it means that energy is being rapidly released, and there may be energy flow attenuation phenomenon in the link; when the change rate is close to zero, it means that the energy flow is in a balanced state, which is an ideal steady-state operation stage. In order to ensure the accuracy of the change rate, it is necessary to exclude sudden noise or short-term disturbance in the calculation process to prevent it from affecting the overall trend judgment. In the case of alternating operation of multiple processes, the fastener production line will frequently experience the alternating process of energy release and energy absorption, so the energy density change rate curve will usually present periodic fluctuations, and by continuously tracking the curve, the unstable interval of energy flow can be accurately identified.

[0093] After obtaining the time series of energy density change rate, dynamic smoothing is performed on the series to reduce the high-frequency oscillation components therein, forming a smooth trend curve of energy change. This step is the core of energy signal optimization, and its purpose is to eliminate energy fluctuations caused by device start-stop, power grid disturbance, or load mutation. In specific implementation, the energy density change rate curve is divided into time periods, and time intervals with abnormally increased change rate are identified. These intervals often correspond to the stages of oscillation in energy flow. For example, when the fastener forming equipment is frequently started in the high-speed stamping state, the rising and falling edges of the current waveform will cause rapid changes in energy density, resulting in high-frequency energy oscillation. By detecting these mutation segments on the time axis and performing smoothing processing, the peak parts of energy density change can be weakened, and the energy flow curve can present a continuous transition state. In addition, in order to maintain the timeliness and balance of the energy signal, the smoothing process should be gradual, that is, while eliminating high-frequency oscillation, the low-frequency energy trend is preserved, so that the macroscopic change characteristics of energy flow are retained. The energy change curve after smoothing processing can clearly show the overall change rule of energy flow under different working conditions, providing stable signal input for the construction of energy transmission path.

[0094] Finally, after completing the smoothing processing of the energy density change rate, the processed energy signal is used to construct a stable energy transmission path, and its continuity is verified. By remapping the smoothed energy signal on the time axis, a stable trajectory of energy flow can be formed, which reflects the transmission law of energy between the output end of the inverter unit and the input end of the power grid. When the slope of the energy flow trajectory is stable and the direction is consistent, it indicates that the energy transmission in the link is smooth and there is no reverse impact; when the slope of the trajectory appears short-term fluctuation or reverse jump, it indicates that the energy flow is disturbed and needs to be adjusted again. In order to ensure the continuity of the energy transmission path, it is necessary to continuously track the smoothed energy signal in time to ensure that it can still maintain continuous flow under different load fluctuations. Especially in the state of multi-process parallel operation of the fastener production line, the load changes of heat treatment, electroplating and forming equipment will cause the energy flow to fluctuate in multiple directions. By keeping the stability of the energy transmission path, energy flow interruption or local energy accumulation in the energy feedback link can be avoided. Finally, through continuous smoothing signal correction and path tracking, the energy flow in the energy feedback link can be kept in a stable and controllable state, so that the entire energy feedback process has high stability and high continuity under different working conditions.

[0095] It should be noted that:

[0096] The dynamic filter gating algorithm refers to a signal smoothing method in energy feedback signal processing, which adaptively adjusts the filtering range and threshold control conditions according to the real-time fluctuation characteristics of the energy density change rate. The core principle is to identify and dynamically suppress the high-frequency oscillation components in the feedback current signal, so that the energy flow remains continuous and smooth in the time axis, while avoiding the energy delay or signal distortion caused by traditional fixed filtering. The algorithm has the core characteristics of "dynamic response, threshold control, and gradual smoothing", and can flexibly switch the filtering strength between energy mutation and steady state.

[0097] For example, in the automobile fastener production line, when the forming machine enters the high-speed stamping stage, the output current of the inverter unit will have high-amplitude oscillation in a very short time, and a sharp peak fluctuation will be formed in the energy density change rate curve. If fixed parameter filtering is used, the system will either react too slowly and fail to weaken the high-frequency oscillation in time, or react too quickly and cause the actual trend of the energy signal to be weakened too much. The dynamic filter gating algorithm dynamically sets the filtering threshold by monitoring the amplitude and change rate of the energy density change rate in real time: when the change rate exceeds the set threshold (such as the energy density change rate exceeds 3 times the steady state interval), the strong filter gating is automatically started to suppress the high-frequency components with high intensity; when the change rate falls back to the stable interval, the algorithm automatically reduces the filtering strength to retain the low-frequency energy components, so as to maintain the trend integrity of the energy flow.

[0098] For example, when the heat treatment equipment switches from the heating stage to the constant temperature stage, the energy feedback current waveform changes from unstable oscillation state to smooth periodic signal. At this time, the dynamic filter gating algorithm identifies the downward trend of the energy density change rate and gradually weakens the filter threshold to restore the natural flow of the energy signal. After this process, the high-frequency peaks in the feedback energy curve are effectively reduced, and the overall energy flow trend is preserved, forming a smooth and continuous energy transmission path. The advantage of this algorithm is that it can adaptively adjust the signal smoothing degree under different working conditions, achieving intelligent energy signal optimization control that neither filters key information nor effectively suppresses high-frequency disturbances.

[0099] Through the above implementation steps, the quality of the energy flow signal can be further optimized based on phase lead compensation, thereby reducing the energy feedback instability problem caused by high-frequency oscillation from the root. After processing by this process, the energy flow not only remains continuous in time, but also is more balanced in space, thereby effectively improving the feedback energy utilization rate, reducing the electrical stress of the inverter unit, and enhancing the energy consumption control stability and safety of the entire automobile fastener production line under the condition of multi-process high-frequency start-stop.

[0100] Step five, according to the smooth energy signal output by the dynamic filtering gating algorithm, the parameters of the phase dynamic drift model are updated in real time, the adaptive adjustment of the energy feedback link is realized, the stability and safety of the energy flow direction are continuously maintained under complex working conditions, and thus the efficient energy feedback control is realized;

[0101] The specific implementation of this step is as follows:

[0102] Based on the smooth energy signal output after dynamic filtering gating processing, the running state of the current energy feedback link is identified and characteristic is extracted. The smooth energy signal can truly reflect the stable change characteristics of the feedback energy on the time axis after the high-frequency oscillation component is reduced, so it is necessary to extract the key characteristic parameters representing the energy flow state from the smooth signal in this stage, including the energy flow amplitude, the energy flow direction change rate, the energy flow stable interval, and the position of the energy flow mutation point. Through continuous collection of these characteristics, the real-time running state atlas of the energy feedback link can be formed. In order to ensure the accuracy of state identification, it is necessary to keep the consistency of signal sampling and time reference, so that each energy signal segment can be accurately corresponding to the corresponding current waveform and grid voltage signal in time. Especially under the high-frequency start-stop working condition of the fastener production line, the energy flow state often presents periodic fluctuations, and by continuously extracting the time characteristics of these fluctuations, the current energy flow trend and its stability can be accurately reflected, providing basic data for the parameter update of the phase dynamic drift model.

[0103] The energy flow characteristic parameters extracted from the smooth energy signal are compared with the historical parameters in the existing phase dynamic drift model to identify the deviation between the current energy flow state and the phase change model. Specifically, the phase dynamic drift model describes the dynamic change rule of the feedback current phase relative to the grid voltage phase, while the smooth energy signal reflects the actual state of energy transmission intensity and direction. When there is consistency between the two, it means that the system is running stably; when there is a difference between the two, it means that there is a phase mapping drift in the current energy feedback link. At this time, the time difference between the energy flow characteristic change and the phase shift change is calculated to judge the response lag degree of the model. For example, when the energy flow direction has changed but the model still maintains the old phase relationship, it means that the dynamic response of the model parameter lags, and the phase drift rate and direction correction coefficient in the model need to be adjusted immediately. In this way, the energy signal and the phase model can be dynamically corresponded in the time dimension, so that the phase dynamic drift model can immediately perceive the change of the energy flow state and identify the deviation trend of the phase drift.

[0104] After identifying the deviation between the energy flow state and the phase model, the parameters of the phase dynamic drift model are updated step by step, so that the model has the ability to adaptively adjust. The updating process is driven by the real-time characteristic changes of the energy flow, and the energy flow intensity, direction change rate and stable interval are continuously fed back to the model to correct the time constant and drift slope of the phase change in the model. Specifically, when it is detected that the energy flow rises rapidly in a short time, it means that the phase change speed of the feedback link should be accelerated to prevent the energy flow direction from responding with delay; when the energy flow is stable and the change rate is low, the phase drift rate of the model should be slowed down accordingly to maintain the balance between the energy flow and the phase change. In addition, near the energy flow mutation point, the phase parameters of the model should be automatically adjusted to make the phase response advance or delay for a certain time to offset the mutation effect of the energy flow. Through this real-time updating method, the model parameters are no longer fixed, but are continuously corrected with the dynamic changes of the energy flow, so that the phase dynamic drift model can maintain self-adaptive characteristics under different load conditions. In order to prevent frequent adjustment from causing model instability, a smooth transition should be performed after each parameter update to ensure that the phase change curve output by the model is continuous and smooth on the time axis, avoiding excessive correction or response oscillation.

[0105] After completing the real-time updating of the parameters of the phase dynamic drift model, the overall operation effect of the energy feedback link is dynamically verified to ensure that the stability and safety of the energy flow direction are maintained after the model is updated. The verification process includes comparing the coupling relationship between the phase dynamic drift model output before and after updating and the smoothed energy signal, and analyzing the synchronization of the two on the time axis to determine the effectiveness of the model update. When the phase change curve output by the model is completely consistent with the energy flow direction of the smoothed energy signal and the synchronization delay time is within the preset threshold, it means that the model updating is successful and the energy feedback link has achieved adaptive adjustment; when it is detected that the phase response of the model output has a slight delay or the energy flow direction deviates instantaneously, it means that the model updating amplitude is insufficient or the parameter response is incomplete, and it needs to be recalibrated. In the long-term operation process, the parameter updating of the model should form a continuous self-circulation mechanism, so that every change of the energy signal can trigger the fine tuning of the phase model. Especially in the fastener production process, when multiple devices such as forming machines, heat treatment furnaces and electroplating tanks are operated alternately, the working conditions of the energy feedback link are complex and variable. Only through continuous parameter updating can the direction of the energy flow be ensured not to be impacted, and the phenomenon of excessive electrical stress of the inverter unit caused by instantaneous energy backflow can be avoided. Ultimately, through continuous real-time updating and verification, the energy feedback link can still maintain the stability of the energy flow direction and the safety of the electrical operation under complex load conditions, achieving truly efficient energy feedback control.

[0106] Through the above specific implementation steps, the smooth energy signal can be combined with the phase dynamic drift model, the model has self-adaptive updating capability in the running process, so that the energy feedback link can maintain stable energy flow under different loads, different power grid states and different working tempos. The process not only improves the control accuracy and dynamic response speed of energy feedback, but also significantly enhances the anti-disturbance ability and safety of the energy feedback process, so that the automobile fastener production line can still maintain the maximization of energy utilization and the long-term stability of equipment operation in the high-frequency start-stop and multi-process collaborative operation environment.

[0107] The present application establishes a phase dynamic drift model and combines phase trend prediction and advance compensation control, so that the current phase in the energy feedback link and the grid voltage are kept in continuous synchronization, effectively avoiding the energy flow reverse impact phenomenon caused by phase response lag. The control method can real-time correct the phase offset in the complex production environment of multi-process and high-frequency start-stop, so that the feedback current waveform is always in a stable conduction state, thereby significantly reducing the ripple current amplitude, reducing the transient stress of the filter capacitor and power tube of the inverter unit, and improving the stability of energy feedback operation and the service life of the equipment.

[0108] The present application introduces energy density change rate extraction and dynamic filtering gate mechanism in the feedback current signal processing process, so that the energy transmission path remains smooth and continuous under different load fluctuations, realizing real-time adaptive adjustment of energy flow. The method can dynamically balance the energy flow between energy feedback and equipment load, significantly improve energy utilization and feedback efficiency, while avoiding the risk of heat loss and oscillation caused by energy accumulation or mutation, so that the entire fastener production line reaches a dynamic optimal state in terms of energy-saving control and operation safety.

[0109] The present application provides an automobile fastener production energy consumption optimization control system as shown in Figure 2 The automobile fastener production energy consumption optimization control system comprises a phase dynamic modeling module, a phase trend prediction module, a phase synchronization compensation module, an energy signal smoothing module and an adaptive adjustment module.

[0110] The phase dynamic modeling module acquires real-time current waveform signals and grid fundamental voltage signals of the inverter unit in the energy feedback link, and establishes a phase dynamic drift model using a time difference analysis algorithm to form a continuous mapping of energy flow direction change on the time axis.

[0111] The phase trend prediction module extracts the continuous change section of the offset rate between the current waveform and the grid fundamental based on the established phase dynamic drift model, and uses a sliding window algorithm to construct a phase trend prediction function to convert the time characteristics of energy flow direction change into a phase prediction signal trajectory.

[0112] The phase synchronization compensation module generates a phase prediction signal track according to a phase trend prediction function, and dynamically adjusts the pulse width modulation trigger time of the inverter unit by using a phase lead compensation algorithm, so that the phase of the feedback current and the phase of the grid voltage keep synchronous flow.

[0113] The energy signal smoothing module extracts the energy density change rate based on the feedback current signal adjusted by the phase lead compensation algorithm, and uses a dynamic filter gating algorithm to reduce the high-frequency oscillation component of the feedback energy signal to form an energy transmission path.

[0114] The adaptive adjustment module updates the parameters of the phase dynamic drift model in real time according to the smoothed energy signal output by the dynamic filter gating algorithm, and performs adaptive adjustment on the energy feedback link.

[0115] The automobile fastener production energy consumption optimization control method provided by the embodiment of the application is implemented through the automobile fastener production energy consumption optimization control system, and the specific method and process of the automobile fastener production energy consumption optimization control system are described in the embodiment of the automobile fastener production energy consumption optimization control method, which will not be described here.

[0116] The above only describes some exemplary embodiments of the 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 departing from the spirit and scope of the application. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the application.

Claims

1. A method for optimizing energy consumption control in automobile fastener production, characterized by, The method comprises the following steps: Step one, obtaining the real-time current waveform signal of the inverter unit in the energy feedback link and the grid fundamental voltage signal, using the time difference analysis algorithm to establish a phase dynamic drift model, and forming a continuous mapping of the energy flow direction change on the time axis; Step two, based on the established phase dynamic drift model, extracting the continuous change section of the offset rate between the current waveform and the grid fundamental, using the sliding window algorithm to construct a phase trend prediction function, and converting the time characteristics of the energy flow direction change into a phase prediction signal trajectory; Step three, according to the phase prediction signal trajectory generated by the phase trend prediction function, using the phase lead compensation algorithm to dynamically adjust the pulse width modulation trigger time of the inverter unit, so that the phase of the feedback current and the phase of the grid voltage keep synchronous flow; Step four, based on the feedback current signal adjusted by the phase lead compensation algorithm, extracting the energy density change rate, and using the dynamic filtering gating algorithm to reduce the high-frequency oscillation components of the feedback energy signal, forming an energy transmission path; Step five, according to the smooth energy signal output by the dynamic filtering gating algorithm, real-time update the parameters of the phase dynamic drift model, and adaptively adjust the energy feedback link; After obtaining the continuous mapping of the energy flow direction change, the phase difference change characteristics on the whole time axis are modeled to form a phase dynamic drift model. The phase dynamic drift model is based on the continuous mapping result of the energy flow direction, and is a multidimensional correlation of the phase difference value, the phase change rate and the energy flow direction identifier at each time point. A time domain model that can dynamically describe the energy flow direction change trend is established. In the model establishment process, first, the reference points of phase change on the time axis are determined, and by comparing these reference points with the local extreme points in the continuous mapping curve, the key time window in which the energy flow direction changes from positive feedback to reverse absorption is identified. Then, the phase change rate in these key time windows is continuously tracked, so that the phase dynamic drift model can reflect the inertia and delay characteristics of the phase change in the energy feedback link. Finally, by time calibration of the updated phase drift characteristics and the mapping result of the previous period, a phase dynamic drift model that is continuously updated with the running state is formed. The sliding window algorithm is used to construct a phase trend prediction function, which means that the phase offset data of the continuously collected current waveform and grid fundamental voltage signal are divided into multiple time windows with fixed length and mutual partial overlap according to time sequence. The phase change rate and directionality in each window are dynamically calculated and trend fitted to form a continuously predictable phase change trend function in time dimension. The dynamic filtering gating algorithm is a signal smoothing method that adaptively adjusts the filtering range and threshold control conditions according to the real-time fluctuation characteristics of the energy density change rate in the energy feedback signal processing.

2. The automobile fastener production energy consumption optimization control method according to claim 1, characterized by, The step of obtaining the real-time current waveform signal of the inverter unit in the energy feedback link and the grid fundamental voltage signal comprises: Current detection points and voltage detection points are arranged at the output side and the grid connection side of an inverter unit of the energy feedback link respectively, and real-time current waveform signals of the output end of the inverter unit and fundamental voltage signals of the grid side are synchronously acquired; The current waveform signals and the grid fundamental voltage signals are compared point by point on a time axis to determine a phase difference at each sampling time, and a phase difference time sequence with time as the horizontal axis and the phase offset angle as the vertical axis is formed; The phase difference time sequence is mapped to the time axis to form a continuous curve of energy flow direction change, and an instantaneous trend of energy flow direction switching is identified according to a phase difference change rate; A phase dynamic drift model is established based on the continuous curve of energy flow direction change, and the phase dynamic drift model which can be continuously updated according to the running state is formed by comparing the phase change reference points and the local extreme points.

3. The automobile fastener production energy consumption optimization control method according to claim 2, characterized by, The steps of extracting the continuous change section of the offset rate between the current waveform and the grid fundamental wave based on the phase dynamic drift model include: The phase offset amount between the current waveform and the grid fundamental voltage is continuously extracted in the established phase dynamic drift model, and the section with a high phase change rate on the time axis is identified to form a phase offset rate change identification sequence; The phase change curve in the offset rate change section is continuously segmented and analyzed to extract the phase change directionality and stability characteristics, and unstable factors caused by grid fluctuations or current jitter are excluded; The phase change directionality curve is mapped to the time axis to convert the time law of energy flow direction change into a continuous trend signal trajectory, and the acceleration interval and the stable interval of the phase change are identified; The trend signal trajectory is analyzed and dynamically monitored in real time, and when the trajectory slope sharply rises in a short time, the time period corresponding to the sharply rising trajectory slope in a short time is marked as a potential phase delay interval.

4. The automobile fastener production energy consumption optimization control method according to claim 3, characterized by, The steps that the phase prediction signal trajectory generated according to the phase trend prediction function dynamically adjusts the pulse width modulation trigger time of the inverter unit include: The phase prediction signal trajectory is used to compare and analyze the phase synchronization state of the feedback current and the grid voltage in real time to determine the deviation direction and degree of the current phase relative to the grid fundamental voltage phase; After the analysis of the phase deviation direction and degree is completed, the pulse width modulation trigger time of the inverter unit is preliminarily adjusted, and the best advance amount of the conduction trigger time is determined according to the position of the phase change inflection point in the phase prediction signal trajectory; After the adjustment range of the trigger time is determined, the pulse width modulation signal of the inverter unit is continuously and dynamically adjusted to make the phase of the feedback current waveform and the grid voltage waveform time-synchronized and real-time calibrated; After the dynamic adjustment of the pulse width modulation trigger time is completed, the phase difference, the peak value corresponding relationship and the waveform symmetry of the adjusted feedback current waveform and the grid voltage waveform are verified in real time.

5. The automobile fastener production energy consumption optimization control method according to claim 4, characterized by, During the continuous dynamic adjustment of the pulse width modulation trigger time of the inverter unit, the trigger time changes of continuous multiple driving periods are smoothed to make the trigger adjustment present a slow change characteristic, and when the phase difference fluctuation exceeds the preset range, the phase prediction signal trajectory is updated to calculate a new trigger advance amount.

6. The automobile fastener production energy consumption optimization control method according to claim 4, characterized by, The step of extracting the energy density change rate based on the feedback current signal adjusted by the phase lead compensation algorithm and performing high-frequency oscillation component reduction processing includes: Extracting the instantaneous energy density feature in the feedback current signal adjusted by the phase lead compensation, and establishing the distribution curve of the energy density change on the time axis to reflect the transmission state of the energy flow in the feedback link; Performing continuous difference analysis on the energy density change distribution curve, extracting the energy density change rate time sequence, and identifying the dynamic trend of the energy flow intensity change over time; Performing dynamic smoothing processing on the energy density change rate time sequence, reducing the high-frequency oscillation component, and forming a smooth trend curve of the energy change, so that the energy flow curve remains in a continuous transition state; Re-mapping the smoothed energy signal on the time axis to construct a stable energy transmission path, and verifying the continuity of the path in real time.

7. The automobile fastener production energy consumption optimization control method according to claim 6, characterized by, In the step of performing dynamic smoothing processing on the energy density change rate time sequence, the energy transmission path is kept in a continuous transition on the time axis by retaining the low-frequency energy change trend while reducing the high-frequency oscillation component.

8. The automobile fastener production energy consumption optimization control method according to claim 6, characterized by, The step of updating the phase dynamic drift model parameters in real time based on the smoothed energy signal output by the dynamic filtering gating algorithm includes: Based on the smoothed energy signal output after dynamic filtering gating processing, the running state of the current energy feedback link is identified and the feature is extracted to obtain the energy flow amplitude, the energy flow direction change rate, the energy flow stable interval and the position of the energy flow sudden change point; Compare the energy flow feature parameters extracted from the smoothed energy signal with the historical parameters in the phase dynamic drift model, identify the deviation between the energy flow state and the phase change model, and calculate the time difference between the energy flow feature change and the phase shift change; After identifying the deviation, the parameters of the phase dynamic drift model are updated step by step, and the energy flow intensity, direction change rate and stable interval are continuously fed back to the model to realize adaptive adjustment of the model parameters and keep smooth transition; After completing the real-time updating of the parameters, the synchronization relationship between the output of the phase dynamic drift model before and after updating and the smoothed energy signal is verified.

9. The automobile fastener production energy consumption optimization control method according to claim 8, characterized by, During the step-by-step updating of the phase dynamic drift model parameters, when the energy flow direction change rate exceeds the preset threshold, the phase drift rate and direction correction coefficient in the model are adjusted first, and the parameter smoothing transition time is automatically extended at the energy flow sudden change point to prevent phase fluctuations caused by excessive model response.

10. The system for optimizing energy consumption control in the production of automotive fasteners, for implementing the method for optimizing energy consumption control in the production of automotive fasteners according to any one of claims 1 to 9, characterized in that, It includes a phase dynamic modeling module, a phase trend prediction module, a phase synchronization compensation module, an energy signal smoothing module, and an adaptive adjustment module. The phase dynamic modeling module acquires real-time current waveform signals and grid fundamental voltage signals of the inverter unit in the energy feedback link, establishes a phase dynamic drift model using a time difference analysis algorithm, and forms a continuous mapping of energy flow direction change on the time axis. The phase trend prediction module extracts the continuous change section of the deviation rate between the current waveform and the power grid fundamental wave based on the established phase dynamic drift model, uses a sliding window algorithm to construct a phase trend prediction function, and converts the time characteristics of the energy flow direction change into a phase prediction signal trajectory. The phase synchronization compensation module uses a phase lead compensation algorithm to dynamically adjust the pulse width modulation trigger time of the inverter unit according to the phase prediction signal trajectory generated by the phase trend prediction function, so that the phase of the feedback current and the phase of the power grid voltage remain synchronous. The energy signal smoothing module extracts the energy density change rate based on the feedback current signal adjusted by the phase lead compensation algorithm, and uses a dynamic filtering gating algorithm to reduce high-frequency oscillation components of the feedback energy signal to form an energy transmission path. The adaptive adjustment module updates the parameters of the phase dynamic drift model in real time according to the smoothed energy signal output by the dynamic filtering gating algorithm, and adjusts the energy feedback link adaptively.

Images