An intelligent monitoring and automatic adjusting system and method for operation of engineering equipment

Abstract

The application discloses an engineering equipment operation intelligent monitoring and automatic adjusting system and method, which collects multi-dimensional operation data of equipment in real time, constructs a three-dimensional operation state evaluation model by using an intelligent algorithm, accurately identifies a stable operation area, generates a dynamic control benchmark and a feedback adjusting loop, adopts control channel decoupling and selective reconstruction technology to eliminate harmful coupling while retaining beneficial synergistic effect, identifies potential risk points such as resonance through spectrum analysis, establishes a perfect safety operation boundary of a, analyzes equipment response dead zone characteristics, converts the same into an energy-saving working area through energy analysis, formulates an optimized power distribution strategy and establishes a multi-level emergency reserve system, establishes a fault early warning mechanism based on vibration characteristics, realizes rapid detection of abnormal states and power compensation, constructs an accurate real-time control instruction through equipment response matrix construction and time delay compensation analysis, and comprehensively improves operation efficiency, stability and energy-saving level of engineering equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control, in particular to an engineering equipment operation intelligent monitoring and automatic adjustment system and method. BACKGROUND

[0002] With the continuous expansion of modern industrial production scale, the running environment of engineering equipment is becoming more and more complex, and higher requirements are put forward for the stability, efficiency and reliability of equipment operation. However, the existing equipment monitoring system mainly relies on simple threshold alarm and manual intervention, lacks deep understanding and prediction ability of equipment running state, and is difficult to realize real intelligent management.

[0003] The main problems existing in the current engineering equipment control field include: the mutual influence relationship between various parameters in the equipment operation process is complex, and the traditional control method is difficult to realize multi-parameter collaborative optimization; the running characteristics of the equipment under different working conditions are quite different, and the fixed control strategy cannot adapt to the dynamically changing running environment; the energy loss problem is serious, especially under low load and transition working conditions, a large amount of energy is wasted in the form of heat energy; the equipment failure often has the characteristics of gradualness, but the existing system lacks effective early warning and adaptive compensation mechanism, which leads to small problems evolving into big failures. Therefore, a method is needed to solve at least one of the above problems. SUMMARY

[0004] The present application discloses an engineering equipment operation intelligent monitoring and automatic adjustment system and method, which aims to realize comprehensive perception of equipment running state through multi-dimensional data acquisition and fusion analysis; use intelligent algorithm to identify running mode and potential risk, and build adaptive control strategy; through energy optimization and fault compensation mechanism, improve the running efficiency and reliability of the equipment, and finally realize the intelligent monitoring and automatic adjustment of the engineering equipment.

[0005] The first aspect of the present application proposes an engineering equipment operation intelligent monitoring and automatic adjustment method, comprising the following steps:

[0006] Collecting physical running parameters and working condition signals of engineering equipment, constructing an equipment running situation map based on the physical running parameters, identifying a stable domain of the running situation map to generate a control reference line, and forming a feedback regulation loop based on the control reference line and the working condition signals;

[0007] Based on the feedback regulation loop, control channel decoupling analysis is performed to generate independent control subrings, selective recoupling processing is performed on the independent control subrings to form enhanced coupling channels, resonance risk points are identified through the enhanced coupling channels, and a safe running envelope is constructed by avoiding the resonance risk points;

[0008] A dead zone characteristic parameter is generated by performing response boundary analysis on the safe operation envelope, available dead zones are identified by performing dead zone energy analysis using the dead zone characteristic parameter, the available dead zones are converted into energy-saving working areas to form a low-power consumption operation trajectory, and a power allocation scheme is generated along the low-power consumption operation trajectory;

[0009] The power allocation scheme is decomposed into a main power output and an auxiliary power output, a margin evaluation is performed on the auxiliary power output to extract a standby power pool, and energy aggregation is performed on the standby power pool to form an emergency power reserve;

[0010] An abnormal vibration signal is obtained by performing fault symptom detection on the main power output, a power compensation demand is generated based on vibration amplitude analysis of the abnormal vibration signal, and a corresponding power is retrieved from the emergency power reserve according to the power compensation demand to form a compensation power flow, and the compensation power flow is switched with the main power output to generate a stable output network;

[0011] A device response matrix is constructed based on the stable output network, and a device control instruction is generated by performing time delay compensation analysis on the device response matrix.

[0012] The second aspect of the application provides an engineering equipment operation intelligent monitoring and automatic adjustment system, comprising:

[0013] A parameter acquisition module is configured to acquire physical operation parameters and working condition signals of the engineering equipment, construct a device operation situation map based on the physical operation parameters, identify a stable domain of the operation situation map to generate a control reference line, and form a feedback regulation loop based on the control reference line and the working condition signals;

[0014] A channel decoupling module is configured to perform control channel decoupling analysis based on the feedback regulation loop to generate independent control sub-rings, perform selective recoupling processing on the independent control sub-rings to form enhanced coupling channels, identify resonance risk points through the enhanced coupling channels, and construct a safe operation envelope by avoiding the resonance risk points;

[0015] A dead zone analysis module is configured to generate a dead zone characteristic parameter by performing response boundary analysis on the safe operation envelope, identify available dead zones by performing dead zone energy analysis using the dead zone characteristic parameter, convert the available dead zones into energy-saving working areas to form a low-power consumption operation trajectory, and generate a power allocation scheme along the low-power consumption operation trajectory;

[0016] A power distribution module is configured to decompose the power allocation scheme into a main power output and an auxiliary power output, perform margin evaluation on the auxiliary power output to extract a standby power pool, and perform energy aggregation on the standby power pool to form an emergency power reserve;

[0017] A fault compensation module is configured to detect a fault symptom of the main power output to obtain an abnormal vibration signal, analyze a vibration amplitude generation power compensation demand based on the abnormal vibration signal, and form compensation power flow from the emergency power reserve according to the power compensation demand to generate a stable output network by switching the compensation power flow with the main power output.

[0018] A control output module is configured to construct a device response matrix based on the stable output network, analyze time delay compensation of the device response matrix, and generate a device control instruction.

[0019] The beneficial effects of the present application are embodied in the following aspects: first, through multi-dimensional operation data acquisition and situation map construction technology, the overall perception and stable domain accurate identification of the device operation state are realized, combined with control channel decoupling analysis and selective recoupling strategy, the system resonance risk point is accurately identified and the safe operation envelope is constructed, which improves the stability and safety of the device operation. Second, through dead zone characteristic parameter analysis and energy flow blocking identification technology, the device response dead zone is converted into an energy-saving working area to form a low-power consumption operation trajectory, and through the main and auxiliary decomposition of the power allocation scheme and the extraction of the standby power pool, a hierarchical emergency power reserve system is established to effectively reduce the device energy consumption and optimize the power resource allocation. Finally, through abnormal vibration signal detection and power compensation demand analysis, rapid response and non-disturbance switching in fault state are realized, combined with time delay compensation analysis of the device response matrix, accurate real-time control instructions are generated to ensure stable operation and rapid response capability of the device under various working conditions.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0021] The drawings herein show specific examples of the technical solutions described in the present application, and constitute part of the specification together with the specific embodiments, for explaining the technical solutions, principles and effects of the present application.

[0022] Unless specifically stated or otherwise, the same reference signs in different drawings represent the same or similar technical features, and different reference signs may also be used to represent the same or similar technical features.

[0023] Figure 1 is a flow diagram of an engineering device operation intelligent monitoring and automatic adjustment method of the present application.

[0024] Figure 2 is a structural block diagram of an engineering device operation intelligent monitoring and automatic adjustment system of the present application. DETAILED DESCRIPTION

[0025] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0026] It is to be understood that the terminology "including", "comprising", "consisting" and "consisting essentially of" used in the specification and the appended claims, are used in the sense of making an inclusion of the features, integers, steps, operations, elements, and / or components listed thereafter, but does not preclude the addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0027] In the present specification, the reference to "one embodiment" or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrase "in one embodiment" or "in some embodiments" in various places in the specification are not necessarily all referring to the same embodiment, although it can. The terms "including", "comprising", "having" and variations thereof as used in the specification are intended to be broad and encompass the terms "consisting of" and "consisting essentially of", unless otherwise indicated. The use of the term "or" in the claims is used to mean "and / or" unless otherwise indicated.

[0028] The technical solutions of the embodiments of the present application are described below.

[0029] As shown in Figure 1 The embodiments of the present application provide an engineering equipment operation intelligent monitoring and automatic adjustment method, which comprises the following steps S110-S160:

[0030] In step S110, the physical operation parameters and working condition signals of the engineering equipment are collected, the equipment operation situation map is constructed based on the physical operation parameters, the control reference line is generated by identifying the stable domain of the operation situation map, and the feedback adjustment loop is formed based on the control reference line and the working condition signals.

[0031] Specifically, the physical operation parameters and working condition signals of the engineering equipment are collected; a multi-type sensor network is arranged at key positions of the engineering equipment to collect the physical operation parameters in real time during the operation of the equipment; the physical operation parameters include multi-dimensional data such as rotation speed, vibration amplitude, temperature distribution, pressure change, current and voltage, torque output, etc.; the sensors are designed with high precision in the industrial grade, the vibration sensors use piezoelectric accelerometers, the temperature sensors use platinum resistance temperature probes, and the pressure sensors use strain gauge pressure transmitters; the data collection system synchronously samples according to a unified time reference, and the sampling frequency is set according to the dynamic characteristics of different parameters, the vibration signal sampling frequency reaches the level of kilohertz, and the temperature signal sampling frequency is at the level of seconds; meanwhile, the working condition signals are collected, including external input information such as load change, start-stop state, operation mode, environmental conditions, etc.; the working condition signals are obtained through the communication interface of the equipment control system, transmitted by using standard industrial protocols, and the real-time and accuracy of the data are ensured; the time correlation index of the physical operation parameters and the working condition signals is established to realize accurate alignment and synchronous storage of multi-source heterogeneous data.

[0032] The device operation situation map is constructed based on the physical operation parameters; the collected physical operation parameters are organized into a multi-dimensional data matrix according to the time sequence and parameter type, the rows of the matrix represent time sampling points, and the columns represent different physical parameters; the physical operation parameters are subjected to data standardization processing to eliminate the influence of different parameter dimensions, and each parameter is mapped to a unified scale; a correlation matrix between the parameters is constructed based on the standardized physical operation parameters, the correlation coefficients between different parameters are calculated, and the coupling relationship between the physical operation parameters is reflected; principal component analysis is performed on the correlation matrix of the physical operation parameters, the first three principal components are extracted as the main characteristic dimensions of the operation situation, and the cumulative variance contribution rate is ensured to reach a high enough level; the principal components of the physical operation parameters are projected into a three-dimensional space to form a device operation situation map; each point in the operation situation map represents the device operation state at a time, the position of the point is determined by the three principal component values extracted from the physical operation parameters, and the color of the point represents the device operation mode; the dynamic evolution process of the device operation state is exhibited through the trajectory of the continuous time points, the density of the trajectory reflects the speed of the change of the physical operation parameters, and the morphological characteristics of the trajectory reveal the regularity mode of the device operation.

[0033] The stable region identification is performed on the operation situation graph to generate the control reference line. The clustering method is applied to the constructed operation situation graph to identify the stable operation region. The high-density region in the operation situation graph is found through density analysis. The trajectory points in the operation situation graph are divided into different clusters by setting a proper neighborhood parameter, and each cluster represents a stable operation mode. The center position and distribution range of each cluster in the operation situation graph are calculated. The center position represents the typical state of the stable mode, and the distribution range describes the fluctuation degree of the state. The stable region boundary is defined based on the statistical characteristics of the operation situation graph. The points within the boundary have similar operation characteristics. The control reference line is fitted by using the geometric center of the stable region of the operation situation graph. The least square method is used to process the trajectory data in the situation graph. The control reference line is represented in a parameterized form. The trajectory of the reference line in the three-dimensional space of the operation situation graph is described by a parameter equation. The reference line represents the ideal operation trajectory of the equipment under the stable working condition and reflects the evolution path of the normal state in the operation situation graph. The control reference line is smoothed to eliminate the local fluctuation and noise influence and ensure the continuity and smoothness of the reference line.

[0034] In some embodiments, the feedback regulation loop is formed based on the control reference line and the working condition signal, including: generating a reference deviation band according to the control reference line; forming a signal-deviation correlation characteristic by combining the working condition signal and the reference deviation band; extracting a stable control interval within the signal-deviation correlation characteristic; and forming a feedback regulation loop based on the gain value of the stable control interval.

[0035] The reference deviation band is generated according to the control reference line. Based on the determined control reference line, the minimum distance from the actual operation trajectory point to the control reference line is calculated to form a deviation sequence. Statistical analysis is performed on the deviation sequence of the control reference line to calculate the mean μ and the standard deviation σ of the deviation. The mean μ reflects the system deviation relative to the control reference line, and the standard deviation σ reflects the dispersion degree of the deviation. The upper and lower boundaries of the reference deviation band are constructed around the control reference line. The statistical control limit method is used. The upper boundary and the lower boundary are set at positions with the control reference line plus or minus several times the standard deviation. The width of the reference deviation band is adjusted according to the characteristics of different sections of the control reference line. The deviation tolerance is appropriately relaxed in the section where the state of the control reference line changes dramatically, and the control requirement is tightened in the stable operation section. The interpolation method is used to smooth the boundaries of the reference deviation band to ensure the continuity of the boundary surface of the deviation band. The reference deviation band forms a tubular region around the control reference line. The normal operation state should fall within this tubular region. The mathematical description of the reference deviation band is established. The spatial range of the deviation band relative to the control reference line is defined by using a geometric equation.

[0036] Signal-deviation correlation characteristics are formed in combination with the working condition signals and the reference deviation band; the working condition signals collected in real time are correlated and analyzed with the position of the operating state in the reference deviation band, a mapping relationship between the working condition signal conditions and the deviation size in the reference deviation band is established; a signal-deviation correlation characteristics matrix is constructed, the matrix elements representing the correlation strength between different working condition signals and each deviation dimension of the reference deviation band; the correlation degree of the working condition signals and the deviation in the reference deviation band is quantified using an information theory method, the non-linear correlation is measured by calculating mutual information I(X;Y) = ΣΣp(x,y)log(p(x,y) / (p(x)p(y))), where X is the working condition signal variable, Y is the deviation value variable in the reference deviation band, p(x) and p(y) are the marginal probability distributions, and p(x,y) is the joint probability distribution; the key working condition factors having the greatest impact on the deviation are identified based on the signal-deviation correlation characteristics, the influence weights of the working condition parameters are determined by analyzing the correlation characteristics matrix; a transmission relationship of the working condition signals to the deviation in the reference deviation band is established, how the working condition changes affect the operating deviation relative to the reference deviation band is described; the signal-deviation correlation characteristics reveal the influence mechanism of the external working condition signals on the operating state of the equipment in the reference deviation band.

[0037] Stable control intervals are extracted in the signal-deviation correlation characteristics; based on the established signal-deviation correlation characteristics, the stability of the system in the correlation characteristics space under different working condition combinations is analyzed; the boundaries of the stable region in the parameter space are searched through numerical analysis of the signal-deviation correlation characteristics; the stable control interval appears as a connected region in the parameter space in the signal-deviation correlation characteristics, and the parameter combinations in the region can ensure the stability of the system; the dynamic behavior of the signal-deviation correlation characteristics under different initial conditions is studied using the phase trajectory analysis method, and the stable equilibrium point and its attractor domain in the correlation characteristics space are identified; a stability margin index is calculated based on the signal-deviation correlation characteristics, which quantifies the safety distance of the current operating point to the stable boundary in the correlation characteristics space; a common stable control interval that can adapt to various conditions is extracted by comprehensively considering the various working condition conditions of the signal-deviation correlation characteristics; the boundary of the stable control interval is represented by a hyper-surface in the signal-deviation correlation characteristics space, which provides a constraint condition for parameter selection in controller design; a stable interval monitoring mechanism based on the signal-deviation correlation characteristics is established, and a warning signal is sent when the operating parameters approach the stable boundary.

[0038] The gain value based on the stable control interval forms a feedback regulation loop; within the determined stable control interval, the gain parameter of the feedback controller is designed to ensure the stability and performance of the closed-loop system within the stable control interval; a state feedback control strategy is adopted, and the control law is expressed as u=-Kx+r, where u is the control input vector, K is the feedback gain matrix designed based on the stable control interval, x is the state deviation vector relative to the control reference line, and r is the reference input vector; the value of the feedback gain matrix K is constrained by the stable control interval, and a suitable gain value needs to be selected within the interval; according to the change of the working condition signal, a gain scheduling strategy is implemented within the stable control interval to establish the mapping relationship between the working condition parameters and the gain values within the stable control interval; an anti-disturbance mechanism based on the stable control interval is designed to improve the robustness of the system to external disturbances; by calculating the state deviation relative to the control reference line in real time and applying the determined control action within the stable control interval, the device operating state is adjusted to within the reference deviation band; the feedback regulation loop uses the gain value of the stable control interval to realize closed-loop control, automatically compensates for the influence of various disturbances, and ensures that the device operates in a stable state near the control reference line.

[0039] In step S120, control channel decoupling analysis is performed based on the feedback regulation loop to generate independent control sub-loops, and selective recoupling processing is performed on the independent control sub-loops to form enhanced coupling channels. The resonance risk points are identified through the enhanced coupling channels, and the safety operation envelope is constructed by avoiding the resonance risk points.

[0040] Specifically, control channel decoupling analysis is performed based on the feedback regulation loop to generate independent control sub-loops; the constructed feedback regulation loop is represented as a multiple-input multiple-output system, and multiple control channels are coupled in the loop; the transfer function matrix of the feedback regulation loop is decomposed, and the matrix element G_ij represents the influence of the jth input on the ith output; the relative gain array RGA of the feedback regulation loop is calculated as RGA=G×(G^(-1))^T, where G is the open-loop transfer function matrix, and × represents element-wise multiplication; the diagonal elements of the RGA matrix reflect the degree of independent control of each channel, and the non-diagonal elements represent the coupling strength between channels; based on the RGA analysis result of the feedback regulation loop, strong coupling channel pairs and weak coupling channels are identified; decoupling compensation is performed on the strong coupling channels in the feedback regulation loop, and a feedforward decoupling matrix D is designed such that G×D becomes a diagonal dominant matrix; the feedback regulation loop is decomposed into independent control modes by singular value decomposition, and each mode corresponds to an independent control sub-loop; the independent control sub-loop has a single-input single-output characteristic, eliminating the cross-influence in the original feedback regulation loop; a dedicated controller is assigned to each independent control sub-loop to achieve complete decoupling control between channels; the frequency response characteristics and stability margin parameters of the independent control sub-loops are recorded.

[0041] The independent control sub-rings are selectively recoupled to form an enhanced coupling channel; dynamic characteristics of each independent control sub-ring are analyzed to identify a sub-ring combination with synergistic enhancement potential; a coherence function γ_ij(f)=|S_ij(f)|² / (S_ii(f)×S_jj(f)) between the independent control sub-rings is calculated, where S is a power spectral density, f is a frequency, and γ_ij reflects a correlation degree of two sub-rings in a frequency domain; the independent control sub-rings with a coherence function exceeding a threshold are selected for recoupling to form a synergistic control structure; a recoupling matrix C is designed to form the enhanced coupling channel by cross-connection of the selected independent control sub-rings; the enhanced coupling channel retains the stability advantage of the independent control sub-rings while obtaining better dynamic performance through coupling; a coupling gain K_c is introduced to the independent control sub-rings to adjust the coupling strength and optimize system response; a state space description of the enhanced coupling channel is established, including state variables of the original independent control sub-rings and newly added coupling terms; coordinated action of multiple independent control sub-rings is realized to improve overall control effect of the system; key parameters such as transfer characteristics, bandwidth, and phase margin of the enhanced coupling channel are recorded, which describe the system characteristics after recoupling.

[0042] In some embodiments, the resonance risk point is identified through the enhanced coupling channel, including: dividing the enhanced coupling channel into a core frequency band and an edge frequency band; emitting a frequency scanning path from the core frequency band to the edge frequency band; recording the positions of amplitude sudden increase points on the frequency scanning path to form a sudden increase point set; marking the point with the largest amplitude in the sudden increase point set as the resonance risk point.

[0043] The enhanced coupling channel is divided into a core frequency band and an edge frequency band; the frequency response characteristics of the enhanced coupling channel are analyzed to determine a frequency range where the main energy is concentrated; the cumulative distribution of the power spectral density of the enhanced coupling channel is calculated, and the frequency range containing the proportion of the main energy is defined as the core frequency band; the core frequency band usually corresponds to the main working frequency and low-order modes of the system, and is the key control area of the enhanced coupling channel; the frequency range outside the core frequency band is divided into the edge frequency band, which contains high-frequency components and secondary modes; the frequency range of the core frequency band is recorded as [f_core_min, f_core_max], and the edge frequency band is divided into a low-frequency edge [0, f_core_min] and a high-frequency edge [f_core_max, f_max]; the dividing line between the core frequency band and the edge frequency band is marked on the frequency spectrum of the enhanced coupling channel to form a clear frequency band division; the dividing limit between the core frequency band and the edge frequency band is dynamically determined according to the energy distribution characteristics of the enhanced coupling channel; the response characteristic differences of the enhanced coupling channel in different frequency bands are analyzed, the core frequency band has high energy density and strong coupling characteristics, and the edge frequency band shows low energy and weak coupling; the frequency band division result is used to guide the subsequent scanning strategy to ensure that the key attention area is fully detected.

[0044] The frequency scanning path extends from the core band to the edge band; starting from the center frequency of the core band, the scanning path is designed to extend to the edge band; the scanning path adopts a logarithmic step manner, and the frequency step size Δf = f × α, where f is the current frequency, and α is the step coefficient, which ensures uniform scanning density in the entire frequency range; first, scan from the core band to the high-frequency edge band, and the path covers the high-frequency characteristic region of the enhanced coupling channel; then, scan from the core band to the low-frequency edge band to detect the possibility of low-frequency resonance; at each frequency point, a test signal is injected through the enhanced coupling channel, the signal amplitude remains constant, and the phase continuously changes; the response characteristics of the enhanced coupling channel at each scanning frequency are recorded, including amplitude, phase, and response time; the design of the scanning path takes into account the bandwidth limitation and dynamic range of the enhanced coupling channel, avoiding missing potential resonance frequencies; the bidirectional scanning strategy ensures complete coverage from the core band to the edge band; during the scanning process, the response of the enhanced coupling channel is monitored in real time, and the scanning step size is refined when the amplitude increases sharply; the frequency scanning path forms a detection trajectory that penetrates the entire frequency domain, fully exploring the resonance characteristics of the enhanced coupling channel.

[0045] For example, the recording of the positions of the amplitude sudden increase points on the frequency scanning path forms a sudden increase point set, including: determining an analysis window according to the amplitude jump characteristic of the frequency scanning path, the amplitude jump characteristic including the rising slope, the peak duration and the decay rate; forming an amplitude trajectory along the analysis window to track the amplitude change process; extracting the frequency coordinates of each sudden increase point in the amplitude trajectory; arranging the sudden increase point set according to the amplitude of the frequency coordinates.

[0046] According to the amplitude jump characteristic of the frequency scanning path, an analysis window is determined; in the response data of the frequency scanning path, the amplitude change rate dA / df of adjacent frequency points is calculated, where A is the amplitude and f is the frequency; when the amplitude change rate exceeds a set threshold, it is marked as the starting point of the amplitude jump; three key parameters of the amplitude jump characteristic are analyzed: the rising slope reflects the speed of resonance establishment, which is obtained by linear fitting of the initial part of the jump section; the peak duration indicates the stability of the resonance, and the length of time that the amplitude maintains around the peak value is counted; the decay rate describes the characteristics of the resonance decay, and the frequency change required to drop from the peak value to a certain proportion is calculated; based on the jump characteristic parameters on the frequency scanning path, the range [f_start, f_end] of the analysis window is determined, and the window width covers the complete amplitude jump process; the starting point of the analysis window is set in the stable section before the jump starts, and the end point extends to the recovery section after the jump ends; statistical analysis of multiple jump characteristics helps to optimize the parameter setting of the analysis window, so that the window contains key information and maintains a reasonable size; the analysis window clearly defines the frequency range for subsequent fine analysis.

[0047] The amplitude trajectory is formed by tracking the amplitude variation along the analysis window; the amplitude values of each frequency point on the frequency scanning path within the determined analysis window are extracted to form an amplitude-frequency sequence; the amplitude sequence is smoothed to eliminate the influence of measurement noise and retain the main trend of amplitude variation; the amplitude trajectory is plotted with the frequency within the analysis window as the horizontal axis and the corresponding amplitude value as the vertical axis; the amplitude trajectory clearly shows the variation of amplitude with frequency within the analysis window, and the shape of the curve reflects the occurrence and development process of resonance; the local maximum points of amplitude are marked on the trajectory, which correspond to the potential resonance frequencies; the first and second derivatives of the amplitude trajectory are calculated to determine the precise position of the extreme points; the amplitude trajectory displays key information such as the position, intensity and influence range of the sudden increase point; there may be multiple amplitude peaks within the analysis window, indicating that the system has multiple resonance modes; the slope change of the trajectory reflects the sensitivity of the system to frequency changes, and a sharp rising edge indicates sharp resonance characteristics.

[0048] The frequency coordinates of each sudden increase point are extracted within the amplitude trajectory; all local maximum points in the amplitude trajectory are scanned, and their corresponding frequency values are recorded; for each maximum point, it is verified whether the amplitude exceeds a certain multiple of the background amplitude to confirm it as a true sudden increase point; the precise frequency coordinate f_peak of the sudden increase point is extracted, and the frequency positioning accuracy is improved through interpolation method; the amplitude value A_peak and the quality factor Q=f_peak / Δf_3dB of each sudden increase point are recorded, where Δf_3dB is the half-power bandwidth; all identified sudden increase points are marked on the amplitude trajectory to form a visual distribution of sudden increase points; the number and distribution density of sudden increase points in the amplitude trajectory are counted to evaluate the complexity of the resonance characteristics of the system; the feature vector of the sudden increase point is established, including frequency coordinates, amplitude peaks, bandwidth and other multi-dimensional information; the sudden increase points extracted in the amplitude trajectory represent the resonance response of the system at different frequencies, and each point is a potential risk source; the frequency coordinate accuracy of the sudden increase point directly affects the effectiveness of subsequent risk assessment and avoidance measures.

[0049] The amplitude trajectory diagram is generated, and a set of surge points is generated according to the amplitude size of the frequency coordinates; all the surge points extracted from the amplitude trajectory diagram are sorted in descending order according to the amplitude peak value; a set of surge points S={(f_i,A_i,Q_i)|i=1,2,...,n} is constructed, where f_i is the frequency coordinate of the i th surge point, A_i is the amplitude, Q_i is the quality factor, and n is the total number of surge points; the first element of the set of surge points corresponds to the resonance point with the largest amplitude, and has the highest risk level; the frequency interval between adjacent surge points is calculated, and the distribution density and mutual influence of the resonance frequencies are evaluated; the set of surge points is subjected to cluster analysis, and surge point groups with similar frequencies are identified, which may be caused by the same physical mechanism or coupling effect; an index structure of the set of surge points is established to support fast query according to the frequency range or amplitude threshold; the set of surge points completely records the resonance characteristic distribution of the system, and each element in the set contains key parameters required for positioning and evaluating resonance risk; the statistical characteristics of the set of surge points, such as amplitude distribution and frequency distribution, are analyzed to reveal the overall law of the resonance behavior of the system.

[0050] The point with the largest amplitude in the set of surge points is marked as a resonance risk point; the first element in the sorted set of surge points, i.e. the surge point with the largest amplitude, is extracted; the frequency f_risk of the point is marked as the main resonance risk point, which is the frequency at which the system is most likely to have serious resonance; the complete characteristic parameters of the resonance risk point are recorded, including the exact frequency, peak amplitude, quality factor, and influence bandwidth; the physical cause of the resonance risk point in the enhanced coupling channel is analyzed, and traced back to the specific structural mode or control link; the potential impact of the resonance risk point on the system performance is evaluated, including the vibration amplification factor, stability deterioration degree, and energy concentration degree; in addition to the main resonance risk point, other points in the set of surge points with amplitudes exceeding the safety threshold are marked as secondary risk points; a characteristic database of resonance risk points is established to record historical resonance events and corresponding system responses; accurate identification of the resonance risk point enables the system to take targeted measures to avoid resonance and minimize the harm caused by resonance; the marked resonance risk point becomes a key constraint condition for subsequent safe operation envelope design.

[0051] The safety operation envelope is constructed by avoiding resonance risk points; based on the identified resonance risk points, an inhibition area is delimited in the operation parameter space to avoid the system working near the resonance frequency; a safety margin band is set around each resonance risk point, and the width of the margin band is determined according to the resonance strength and the system damping; the control parameters are modified to make the system natural frequency away from the resonance risk points, and active vibration avoidance is realized; damping compensation is increased near the resonance risk points to reduce the resonance peak value and the vibration amplitude; a notch filter is designed to suppress the frequency component corresponding to the resonance risk point, and the filter transfer function is H(s) = (s 2 + 2ζ n ×ω n ×s+ω n 2) / (s 2 +2ζ d ×ω n ×s+ω n 2), wherein ω n is the notch frequency, i.e., the resonance risk point frequency, and ζ n and ζ d are the damping ratios of the numerator and the denominator, respectively; the avoidance measures of all resonance risk points are integrated to form a complete safety operation envelope; the safety operation envelope defines the parameter range in which the system can stably operate, including the speed limit, the load range, the frequency constraint, etc.; within the safety operation envelope, the system can avoid all identified resonance risk points and ensure smooth operation.

[0052] In step S130, dead zone characteristic parameters are generated by response boundary analysis of the safety operation envelope, dead zone energy analysis is performed using the dead zone characteristic parameters to identify available dead zones, the available dead zones are converted into energy-saving working areas to form a low-power consumption operation trajectory, and a power allocation scheme is generated along the low-power consumption operation trajectory.

[0053] Specifically, dead zone characteristic parameters are generated by response boundary analysis of the safety operation envelope; within the constructed safety operation envelope, the response characteristics of the system at different operating points are analyzed to identify areas with sluggish response or no response; the boundary of the safety operation envelope is composed of multiple constraint surfaces, each constraint surface corresponding to one operating limit; detection is performed along the boundary of the safety operation envelope to measure the response sensitivity of the system to small-amplitude control inputs; when the control input changes by less than a certain threshold, the system output remains unchanged, and this threshold is defined as the dead zone width; the dead zone width distribution at different positions within the safety operation envelope is calculated to form a dead zone characteristic parameter matrix D = [d_ij], wherein d_ij represents the dead zone width of the i th state variable at the j th operating point; the dead zone characteristic parameters also include attributes such as symmetry, nonlinearity, and temperature dependence; in the stable region of the safety operation envelope, the dead zone width is usually large, and the system is not sensitive to disturbances; near the envelope boundary, the dead zone width decreases, and the system response becomes sensitive; the variation law of the dead zone characteristic parameters with operating conditions is recorded to establish a mapping relationship between operating conditions and dead zones.

[0054] In some embodiments, the dead zone energy analysis using the dead zone characteristic parameter identifies available dead zones, including: energy flow blockage identification of the dead zone characteristic parameter generates flow resistance termination zones; energy saving potential evaluation according to the flow resistance termination zones forms energy saving factors; continuous energy saving distribution is generated by response interpolation through the energy saving factors; and threshold extraction is implemented according to the continuous energy saving distribution to generate available dead zones.

[0055] Energy flow blockage identification of the dead zone characteristic parameter generates flow resistance termination zones; energy transfer characteristics in the dead zone characteristic parameter matrix are analyzed to identify regions where energy flow is blocked; large numerical elements in the dead zone characteristic parameter correspond to positions where energy transfer efficiency is extremely low, forming block points of energy flow; the gradient ∇d of the dead zone characteristic parameter is calculated, and the gradient direction points to the direction in which the energy flow resistance increases; when the gradient of the dead zone characteristic parameter exceeds a critical value, the region is marked as a flow resistance termination zone; the energy transfer path in the flow resistance termination zone is cut off, and input energy cannot effectively reach the output end; the spatial distribution of the dead zone characteristic parameter is analyzed, and the flow resistance termination zone usually appears near the local maximum point of the parameter value; the boundary of the flow resistance termination zone is determined by the contour line d=d_critical, where d_critical is the critical dead zone width; in the flow resistance termination zone, the system exhibits strong energy dissipation characteristics, and most of the input energy is converted into heat energy or other forms of loss; the identified flow resistance termination zone forms a key focus area for energy optimization, and these areas have the largest energy saving improvement space.

[0056] Energy saving potential evaluation according to the flow resistance termination zones forms energy saving factors; in the identified flow resistance termination zone, the amount of recoverable or avoidable energy loss is calculated; the volume V_block of the flow resistance termination zone reflects the scale of energy blockage, and the larger the volume, the higher the energy saving potential; the average energy density ρ_energy in the flow resistance termination zone is evaluated, and high energy density areas are given priority for energy saving reform; the energy saving factor is defined as k_save=(E_waste×η_recover) / E_total, where E_waste is the wasted energy in the flow resistance termination zone, η_recover is the theoretical recovery efficiency, and E_total is the total energy of the system; the energy saving factor of each flow resistance termination zone is calculated to form an energy saving factor distribution map; the numerical range of the energy saving factor is between 0 and 1, and the larger the value, the greater the energy saving potential of the area; the position and shape of the flow resistance termination zone affect the size of the energy saving factor, and the energy saving factor of the central region is usually higher than that of the edge region; the flow resistance termination zones are sorted by energy saving factor size, and areas with high energy saving factors are processed first; the energy saving factor not only considers the absolute amount of energy loss, but also considers the feasibility and cost-effectiveness of reform.

[0057] The continuous energy-saving distribution is generated by interpolation of the energy-saving factors; a continuous energy-saving potential distribution field is generated using an interpolation method based on discrete energy-saving factor data points; an interpolation algorithm suitable for the characteristics of the energy-saving factors is selected, such as a radial basis function interpolation or a thin plate spline interpolation; the interpolation function f interp(x, y, z) =∑w i×φ(||r-r i||) is used, wherein w i is a weight coefficient, φ is a basis function, r is a spatial position, and r i is the spatial position of the i th energy-saving factor sampling point; the interpolation process of the energy-saving factors retains the main characteristics of the original data and smoothes local discontinuity; in the area where the energy-saving factor changes dramatically, the density of the interpolation nodes is increased, and the resolution of the distribution field is improved; the continuous energy-saving distribution covers the entire operating space, and each point has a corresponding energy-saving potential value; the equal-value surface of the distribution field shows the spatial variation law of the energy-saving potential, and the high-value area forms an energy-saving "hot spot"; the continuous distribution obtained by interpolation of the energy-saving factors eliminates the information loss caused by discrete sampling; the gradient information of the continuous energy-saving distribution indicates the direction in which the energy-saving potential grows fastest.

[0058] A threshold extraction is performed according to the continuous energy-saving distribution to generate a usable dead zone; a threshold k threshold of the energy-saving potential is set in the continuous energy-saving distribution field, and an area exceeding the threshold is extracted; the selection of the threshold comprehensively considers the technical feasibility and economic benefits, and is usually a certain proportion of the maximum value of the energy-saving factor; a threshold segmentation operation is performed to binarize the continuous energy-saving distribution: if k save>k threshold, the area is marked as a usable area; a connectedness analysis is performed to merge adjacent usable areas to form a complete usable dead zone; each usable dead zone is screened by the continuous energy-saving distribution to ensure that it has sufficient energy-saving value; the geometric characteristics of the usable dead zone are calculated, including volume, surface area, centroid position, and other parameters; a larger usable dead zone usually corresponds to the main energy-saving opportunity of the system and should be developed and utilized preferentially; the boundary of the usable dead zone is accurately determined by the contour line of the continuous energy-saving distribution, avoiding the subjectivity of artificial division; the number and total volume of the usable dead zones are counted to evaluate the overall energy-saving potential of the system; the generated usable dead zone becomes the spatial range of the subsequent energy-saving work area design, ensuring the pertinence and effectiveness of the energy-saving measures.

[0059] The available dead zone is converted into an energy-saving working area to form a low-power running track; the boundary of the identified available dead zone is reconstructed to convert the original "dead zone" into an actively used energy-saving working area; a special control strategy is designed in the available dead zone to make the system run at the lowest energy consumption in the area; the running principle of the energy-saving working area is to utilize the low response characteristics of the dead zone to reduce unnecessary control actions and energy consumption; according to the distribution of the available dead zone, the running path connecting each energy-saving working area is planned; the path planning follows the principle of minimum energy consumption and preferentially selects the track passing through the available dead zone; the parameter equation of the low-power running track is constructed: x(t) = x0 + ∫v_min(τ)dτ, where x is the state variable and v_min is the minimum energy consumption speed; the low-power running track forms a curve in the state space, and the curve passes through multiple energy-saving working areas; the design of the track considers the transition between the available dead zones to ensure the smoothness of state switching; in the energy-saving working area, the system maintains the minimum necessary control input, greatly reducing the power consumption; the implementation of the low-power running track requires accurate state monitoring and switching control to ensure that the system always runs on the predetermined track.

[0060] A power allocation scheme is generated along the low-power running track; based on the designed low-power running track, a detailed power allocation strategy is developed; the low-power running track is divided into multiple sections, each corresponding to different power requirements; in the energy-saving working area section, the power allocation scheme reduces the system power to the minimum level required for operation; the power requirement P(s) = f(v(s), a(s), F_load(s)) at each point on the low-power running track is calculated, where s is the track parameter, v is the speed, a is the acceleration, and F_load is the load; the power allocation scheme includes basic power allocation and dynamic power adjustment; the basic power ensures the basic operation requirements of the system, and the dynamic power is adjusted according to the real-time position of the low-power running track; in the transition section of the track, the power allocation scheme arranges a short-term power boost to complete the state switching; a power allocation timing table is established to clearly define the power set value and adjustment strategy at each time; the power allocation scheme also considers the energy recovery mechanism to recover kinetic and potential energy in deceleration or descent sections.

[0061] In step S140, the power allocation scheme is decomposed into active power output and auxiliary power output, the auxiliary power output is evaluated to extract a standby power pool, and the standby power pool is aggregated to form an emergency power reserve.

[0062] Specifically, the power allocation scheme is decomposed into main power output and auxiliary power output; based on the generated power allocation scheme, the functional attributes and importance levels of each power component are analyzed; the power allocation scheme contains multiple power requirements, which are classified according to the degree of influence on system operation; the main power output is defined as the power part necessary to maintain the core function of the system, including the power to drive the main shaft, the basic power to maintain the operation of the control system, and the power to ensure the operation of the safety device; the auxiliary power output covers the power requirements of non-critical functions, such as lighting, ventilation, auxiliary monitoring, data recording, environmental regulation, and power consumption of secondary systems; the decomposition process uses power flow analysis method to track the destination and use characteristics of each power branch in the power allocation scheme; a power decomposition matrix P_total=P_main+P_aux is established, where P_total is the total power matrix, P_main is the main power output matrix, and P_aux is the auxiliary power output matrix; the main power output usually accounts for a large proportion of the total power, with the characteristics of strong continuity, small fluctuation, and non-interruptibility; the auxiliary power output has the characteristics of intermittent power demand, adjustability, and priority difference; the power decomposition process also considers the difference between dynamic load and static load, and the power demand of dynamic load changes with the working condition, while the static load remains relatively constant.

[0063] The margin evaluation of auxiliary power output extracts the backup power pool; the actual use of auxiliary power output is analyzed to identify the power allocation with margin; many devices in auxiliary power output are not running at full capacity, and there is adjustable power space; the power margin M_i=P_rated-P_actual of each auxiliary system is calculated, where M_i is the margin of the i-th auxiliary system, P_rated is the rated power, and P_actual is the actual used power; statistical analysis is performed on the time series data of auxiliary power output to determine the distribution of average margin, peak margin, and minimum margin; the margin evaluation considers the working cycle and load characteristics of auxiliary systems, and the intermittently working systems have greater available margin; a margin availability matrix is established to mark the size and time window of the power margin that can be called without affecting the auxiliary function; the order of margin calling is determined through the priority sorting of auxiliary power output to ensure that important auxiliary functions are not affected; auxiliary systems with low priority preferentially contribute power margin to form a hierarchical backup power pool structure, including an immediately available layer, a conditionally available layer, and an emergency available layer; the capacity of the backup power pool dynamically changes depending on the current running state of the auxiliary power output and external environmental conditions; all available margins are summarized to form a unified management backup power pool, and the total capacity is calculated as C_backup=Σ(M_i×α_i), where α_i is the available coefficient of the i-th system, reflecting the actual callability of the margin of the system.

[0064] In some embodiments, the energy aggregation of the backup power pool forms an emergency power reserve, including: constructing a power time axis according to the backup power pool; mapping power peak points to the power time axis to form peak time markers; dividing the power time axis into charging periods and discharging periods according to the peak time markers; and comparing energy distribution characteristics of the charging periods and the discharging periods to form an emergency power reserve.

[0065] A power time axis is constructed according to the backup power pool; power data of the backup power pool is arranged in time sequence to form a continuous power time sequence; time resolution of the power time axis is determined according to dynamic characteristics of the backup power pool, millisecond-level resolution is used for fast-changing processes, and second-level or minute-level resolution is used for slow-changing processes; on the power time axis, each time point corresponds to an instantaneous available power value of the backup power pool, reflecting total power resources that can be called at the time; power of the backup power pool presents periodic fluctuations over time, reflecting work rhythm and load variation law of the auxiliary system; the constructed power time axis covers a complete operation period, contains multiple power rising and falling processes, and exhibits dynamic availability of power resources; a coordinate system of the time axis uses relative time for expression, facilitating comparative analysis and pattern recognition of different operation periods; key events are marked on the power time axis, such as start and stop of the auxiliary system, load switching, working condition conversion, and the like, which directly affect capacity change of the backup power pool; the power time axis of the backup power pool presents obvious time-varying characteristics, with significant peak-valley difference, reflecting power surplus degree at different times; through construction of the power time axis, dynamic change process of the backup power pool is quantitatively represented.

[0066] The power peak point is mapped to the power time axis to form a peak time marker; on the constructed power time axis, local power maximum points are searched, which correspond to the peak time of the standby power pool capacity; the peak value detection is realized by comparing the power values of adjacent time points; when the power value at a certain time is greater than the power values before and after the time, the time is marked as a peak point; a peak threshold is set to screen valid peaks; only the maximum value exceeding a certain proportion of the average power is identified as a significant peak to avoid misjudgment caused by small fluctuations; the distribution of the power peak point on the time axis presents a certain rule, reflecting the charging and discharging characteristics of the standby power pool and the system operation mode; the peak time and the corresponding peak power size are recorded to form a peak feature sequence, which is used to analyze the availability mode of the power resource; the peak time marker not only contains time information, but also contains multiple attributes such as peak amplitude, rise time, duration, and fall rate; through the peak distribution analysis on the power time axis, the active period and the silent period of the standby power pool are identified, and the time distribution characteristics of the power resource are understood; the density of the peak time marker reflects the utilization frequency of the standby power pool, and dense peaks represent frequent power calling and releasing; the identification of the peak mode is helpful for optimizing the power scheduling strategy and reasonably allocating power resources at the peak time.

[0067] The power time axis is divided into charging periods and discharging periods according to the peak time markers; according to the peak time markers on the power time axis, the time axis is divided into alternating charging and discharging intervals; the charging period is defined as the time period from the power valley to the peak, during which the standby power pool accumulates available power resources; the discharging period corresponds to the process from the power peak to the valley, and the standby power pool releases the stored power for emergency supply; the interval between two adjacent peak time markers contains a complete charging-discharging cycle, reflecting the periodic change of the power resource; the feature of the charging period is that the power monotonically increases or maintains a high level, indicating that the auxiliary system load is reduced and more available power is released; the discharging period shows a power decreasing trend, indicating that the standby power is called and the available resource is reduced; through the segmentation of the peak time markers, the power time axis is divided into a sequence of charging periods and discharging periods; the duration and power increment of each charging period are counted to analyze the charging efficiency and charging speed variation law; the discharging depth and power release rate of the discharging period are calculated to evaluate the emergency power supply capability and response characteristics of the standby power pool; the division of the charging period and the discharging period reveals the working mode of the standby power pool, providing a time reference for formulating the charging and discharging strategy.

[0068] The energy distribution characteristics of the charging period and the discharging period are compared to form an emergency power reserve, including: converting the power sequence of the charging period into an energy accumulation sequence; superimposing the power sequence of the discharging period onto the energy accumulation sequence to form an energy difference graph; extracting an energy jump accumulation amount in the energy difference graph; and forming an emergency power reserve according to the distribution density of the energy jump accumulation amount.

[0069] The power sequence of the charging period is converted into an energy accumulation sequence; a power time sequence in the charging period is extracted, including power values at all times in the period; the power sequence is converted into an energy sequence by time integration, with an integral formula of E_acc(t)=∫P_charge(τ)dτ, where E_acc(t) is the accumulated energy to time t, P_charge(τ) is the charging power, and the integral starts from the beginning of the charging period; the energy accumulation sequence represents the total energy storage from the beginning of the charging to the current time, and the slope of the curve is equal to the instantaneous charging power; the power sequence of the charging period usually includes multiple sub-stages, with a rapid power rise in the starting stage, a constant power in the stable stage, and a gradually decreasing power in the saturation stage; the energy accumulation curve of the fast charging stage has a large slope, indicating rapid energy accumulation; the curve of the slow charging stage tends to be flat, with a reduced energy accumulation speed; characteristic parameters of the energy accumulation sequence are calculated, including maximum accumulated energy, average charging rate, charging efficiency, and other indicators; the final value of the energy accumulation sequence represents the total storage capacity of the charging period, which is an important parameter for evaluating the reserve capacity; statistical analysis is performed on the energy accumulation sequences of multiple charging periods to obtain typical accumulation patterns and variation ranges; the energy accumulation sequence retains complete time information of the charging process, reflecting the dynamic charging characteristics of the standby power pool.

[0070] The power sequence of the discharging period is superimposed onto the energy accumulation sequence to form an energy difference graph; the power sequence of the discharging period is extracted, including complete data of the power drop process; the starting point of the discharging power sequence is moved to different times of the charging period through time axis translation, achieving superimposition; the energy difference at each time is calculated after superimposition, which is the difference between the charging accumulated energy and the discharging consumed energy; a series of energy difference curves are generated by systematically changing the superimposition amount, and these curves constitute a two-dimensional energy difference graph; the energy difference graph comprehensively displays the energy supply and demand balance under different time matching conditions, with positive areas representing energy surplus and negative areas representing energy deficiency; the zero contour line in the energy difference graph is the energy balance boundary, dividing the graph into surplus and deficit areas; through superimposition analysis of the discharging power sequence, the optimal charging and discharging time sequence matching scheme can be found; the peak position of the energy difference graph corresponds to the maximum energy surplus state, which is the best opportunity to form an emergency reserve; the gradient information in the graph reflects the sensitivity of energy balance to time sequence changes, and the area with a large gradient requires precise time sequence control.

[0071] Extracting the energy jump accumulation amount in the energy difference graph; scanning the numerical changes in the energy difference graph, identifying the mutation position and mutation amplitude of the energy difference; the energy jump is defined as the significant change of energy difference between adjacent sampling points, which reflects the rapid conversion of system energy state; calculate the energy change of each jump point, positive jump represents rapid energy accumulation, corresponding to charging efficiency improvement or load sudden drop; negative jump represents rapid energy consumption, corresponding to load sudden increase or charging interruption; accumulate all jump amounts along the time axis to form the energy jump accumulation curve, which reflects the cumulative effect of energy fluctuation; the growth rate of energy jump accumulation amount characterizes the stability of system energy state, rapid growth means frequent energy fluctuation; in different regions of the energy difference graph, the jump accumulation amount presents different growth patterns, the stable operation area grows slowly, and the transition area grows rapidly; statistical energy jump accumulation amount of spatial distribution characteristics, identify the area where the jump occurs frequently, these areas are the focus of energy management; the maximum value of jump accumulation amount reflects the maximum energy impact that the system may face, which determines the minimum demand for emergency reserves; through jump pattern analysis, future energy fluctuation trend can be predicted, and reserve strategy can be adjusted in advance.

[0072] Forming emergency power reserve according to the distribution density of energy jump accumulation amount; analyze the distribution density of energy jump accumulation amount in the time-dislocation amount plane, the high density area corresponds to frequent and severe energy fluctuation; according to the distribution density, the reserve configuration level is divided, the higher the density, the more emergency power reserve needs to be configured; the calculation of emergency power reserve capacity considers the amplitude and frequency of jump accumulation amount, the calculation formula is P_reserve=k x p_jump x E_jump_max, where P_reserve is the reserve power, k is the safety factor, p_jump is the jump distribution density, E_jump_max is the maximum jump accumulation amount; in the period of dense distribution of jump, the response speed requirement of emergency power reserve is improved to ensure the rapid supplement of power gap; the sparse distribution area can appropriately reduce the reserve configuration, optimize the resource utilization, and reduce the system operation cost; according to the time evolution law of jump accumulation amount, dynamic reserve adjustment strategy is made to realize the optimal allocation of reserve resources; the emergency power reserve in high density period maintains a high alert state and is ready for use at any time; in low density period, part of the reserve can be converted to other purposes; through the reserve configuration guided by distribution density, the precise management of emergency resources is realized, and the economy and reliability of the system are improved; the formed emergency power reserve scheme contains complete elements such as capacity size, response level, input condition and exit mechanism; the reserve scheme can adaptively adjust according to the real-time energy jump characteristics, and always maintain the matching with system demand.

[0073] Step S150, fault symptom detection is performed on the main power output to obtain abnormal vibration signals, power compensation demand is analyzed based on the abnormal vibration signals, and compensation power is formed by calling corresponding power from the emergency power reserve according to the power compensation demand, the compensation power is switched with the main power output without disturbance to generate a stable output network.

[0074] Specifically, fault symptom detection is performed on the main power output to obtain abnormal vibration signals; high-sensitivity vibration sensors are arranged on key transmission components of the main power output to monitor vibration characteristics in real time during operation; vibration data collected by the sensors include three forms of acceleration, speed and displacement, and a sampling frequency is set according to a speed range of the main power output; during normal operation, vibration signals of the main power output present stable periodic characteristics, and an amplitude and a frequency are kept within a set range; the fault symptom detection adopts a multi-parameter joint judgment method, including vibration amplitude out-of-limit detection, spectrum characteristic abnormal identification and vibration trend analysis; when early faults such as bearing wear, poor gear engagement and dynamic imbalance occur in the main power output, characteristic changes will occur in the vibration signals; identification of the abnormal vibration signals is performed by comparing real-time vibration data with normal vibration benchmarks, and a deviation exceeding a threshold value is determined as abnormal; characteristic parameters of the abnormal vibration signals are extracted, including a peak frequency f_peak, a vibration amplitude A_vib and a spectrum energy distribution; the abnormal vibration signals not only contain fault information, but also reflect power fluctuation characteristics of the main power output; occurrence time, duration and evolution trend of the abnormal vibration signals are recorded, and a corresponding relationship between fault symptoms and vibration characteristics is established.

[0075] Power compensation demand is analyzed based on the abnormal vibration signals; time-domain and frequency-domain characteristics of the vibration amplitude are extracted by in-depth analysis of the obtained abnormal vibration signals; changes of the vibration amplitude A(t) with time reflect power fluctuation of the main power output, and an increase in the amplitude usually accompanies power loss; a relationship model between the vibration amplitude and the power loss is established: P_loss=k_v×A², wherein P_loss is the power loss, k_v is a vibration-power conversion coefficient, and A is the vibration amplitude; power loss caused by the abnormal vibration signals needs to be compensated to maintain output stability, and a compensation amount is equal to the identified power loss plus a safety margin; frequency components of the vibration amplitude are analyzed, different frequencies of the vibration correspond to different power compensation characteristics, low-frequency vibration needs large-power slow compensation, and high-frequency vibration needs small-power fast compensation; the power compensation demand P_comp(t) is time-varying and dynamically changes with the intensity and characteristics of the abnormal vibration signals; a response time requirement of the power compensation is calculated to ensure that the compensation action can timely offset power fluctuation caused by the vibration; the power compensation demand also considers a development trend of the vibration, and a margin is left for growing vibration; the generated power compensation demand contains complete parameters such as compensation power size, response speed and duration.

[0076] According to the power compensation demand, the corresponding power is retrieved from the emergency power reserve to form a compensation power flow; based on the calculated power compensation demand, appropriate power resources are matched in the emergency power reserve; the emergency power reserve contains power resources of different response levels, and the call priority is selected according to the emergency degree of the compensation demand; the power retrieval process follows the principle of minimum disturbance, and the reserve resources with matching response speed and capacity are preferentially used; when the power compensation demand P_comp exceeds the capacity of a single reserve source, a multi-source combination retrieval strategy is adopted; a coordination mechanism for power retrieval is established to ensure that the power is extracted from the emergency power reserve without affecting other emergency functions; the formation of the compensation power flow includes three links of power extraction, transmission and adjustment, and each link has a corresponding efficiency loss; the actual formed compensation power flow P_flow = P_comp / η_total, wherein η_total is the comprehensive efficiency; the compensation power flow has dynamic characteristics and can be adjusted in real time following the change of the power compensation demand; upper and lower limit constraints of the compensation power flow are set to prevent excessive compensation or insufficient compensation; the quality parameters of the compensation power flow include power stability, harmonic content, response delay, etc., which affect the compensation effect.

[0077] In some embodiments, the disturbance-free switching of the compensation power flow and the main power output generates a stable output network, including: converting the compensation power flow into a power vector distribution; finding a balance center point in the power vector distribution; performing power diffusion with the balance center point as a seed to form a preliminary stable region; and performing boundary reinforcement on the preliminary stable region to form a stable output network.

[0078] The compensation power flow is converted into a power vector distribution; the continuous compensation power flow is discretized in space, and each discrete point represents a power output node; the power vector contains two attributes of size and direction, the size represents the power amplitude, and the direction represents the power flow path; the vectorization process of the compensation power flow considers the spatial distribution characteristics of the power, and expands the one-dimensional power flow into a multi-dimensional vector field; the power vector distribution V(x, y, z) = [P_x, P_y, P_z], wherein P_x, P_y, P_z are power components in three directions respectively; through the vector decomposition of the compensation power flow, the distribution of power in different directions can be analyzed; the modulus of the power vector |V| = √(P_x²+P_y²+P_z²) represents the total power intensity of the point; the density of the vector distribution reflects the concentration degree of the compensation power flow, and the high-density area has stronger power output capacity; after the compensation power flow is converted into a vector distribution, it is convenient for power optimization and balance analysis in space; the divergence and curl characteristics of the vector field reveal the source-sink distribution and circulation characteristics of the power; the power vector distribution retains the complete information of the compensation power flow, while increasing the analysis ability of the spatial dimension; through the vectorization representation, the power concentration area and the power sparse area can be intuitively identified.

[0079] Finding the balance center in the power vector distribution; analyzing the spatial characteristics of the power vector distribution to find the balance position of the vector field; the balance center is defined as the position where the directional components of the power vector distribution cancel each other out, and the net power flow at this point is zero; the approximate area of the balance center is determined by calculating the centroid position of the power vector distribution; the precise balance center point is obtained through iterative optimization, and the objective function is min∑|V_i×r_i|, where V_i is the i-th vector and r_i is the position vector to the center point; the balance center point has the power convergence feature, and the power vectors around it point to or surround it; there may be multiple local balance points in the power vector distribution, and the globally optimal balance center needs to be identified; the stability of the balance center point is judged by analyzing the eigenvalues of the Jacobian matrix of the vector field around it; a stable balance center point has the ability to recover from small disturbances and is an ideal seed point for building a stable network; the position of the balance center point is affected by the power vector distribution, and the more uniform the distribution, the more stable the balance center; the found balance center point becomes the starting position for subsequent power diffusion, determining the formation mode of the stable region; the power characteristics of the balance center point include convergence strength, stability margin, etc.

[0080] Power diffusion from the balance center point forms a preliminary stable region; starting from the identified balance center point, simulate the process of power diffusion outward; power diffusion follows the law of conservation of energy, propagating from high power density areas to low power density areas; the control equation of the diffusion process is ∂P / ∂t=D∇²P, where P is the power density, D is the diffusion coefficient, and ∇² is the Laplace operator; the balance center point acts as a diffusion source, continuously releasing power to the surrounding space and forming a stable power distribution; the diffusion speed is proportional to the power gradient, and the larger the gradient, the faster the diffusion, until a new equilibrium state is reached; by adjusting the diffusion coefficient D, the formation speed and final range of the stable region can be controlled; the boundary of the preliminary stable region is defined as the isosurface where the power density drops to a certain proportion of the center value; during the diffusion process, the balance center point maintains power supply to ensure the continuous existence of the stable region; the power distribution in the preliminary stable region presents the characteristics of high center and low edge, consistent with the law of natural diffusion; the shape of the stable region is influenced by the initial power vector distribution, and may present a spherical, ellipsoidal or irregular shape; the power diffusion process takes into account the spatial anisotropy, and the diffusion rates in different directions may be different.

[0081] The boundary of the preliminary stable region is identified, which is the area with the largest power gradient; the boundary reinforcement forms a power barrier by increasing the power density at the boundary to prevent the stable region from shrinking; power enhancement nodes are arranged on the boundary of the preliminary stable region, which obtain additional power from the compensation power flow; the power requirement of the boundary reinforcement is P_boundary=k_b×L×∇P, where k_b is the reinforcement coefficient, L is the boundary length, and ∇P is the power gradient at the boundary; the reinforced boundary forms a closed power loop to enclose and protect the preliminary stable region; the boundary reinforcement also includes improving the power transmission capacity of the boundary and reducing power leakage and loss; through boundary reinforcement, the preliminary stable region is transformed into a stable output network with self-sustaining capability; the power distribution inside the stable output network is more uniform, and the volatility is significantly reduced; the topology of the network is optimized to form a multi-channel redundant power transmission path; the stable output network after boundary reinforcement can resist external disturbances and maintain stable power output inside; the finally formed stable output network has self-adaptive adjustment capability and can automatically adjust the internal power distribution according to the load change.

[0082] In step S160, a device response matrix is constructed based on the stable output network, and time delay compensation analysis is performed on the device response matrix to generate a device control instruction.

[0083] Specifically, a device response matrix is constructed based on the stable output network; the stable output network includes multiple power output nodes and transmission paths, and each node has a specific dynamic response behavior; the device response matrix R=[r_ij] is constructed, where r_ij represents the response coefficient of the jth control input to the ith output node; the response coefficient is obtained by applying a unit step input in the stable output network and measuring the output change of each node; the dimension of the device response matrix is determined by the number of inputs and outputs of the stable output network, and is typically a square matrix or close to a square matrix; the matrix element r_ij contains amplitude response and phase response information, which completely describes the transfer relationship between input and output; the topology of the stable output network directly affects the sparsity of the device response matrix, and a tightly connected network corresponds to a dense matrix; the diagonal elements of the response matrix reflect the response strength of the direct control channel, and the non-diagonal elements represent the cross-coupling effect; through frequency scanning of the stable output network, the variation law of the device response matrix at different frequencies is obtained; the device response matrix not only contains static gain information, but also contains dynamic response characteristics such as rise time and overshoot; the condition number of the matrix reflects the control difficulty of the stable output network, and the smaller the condition number, the better the control performance.

[0084] In some embodiments, the time delay compensation analysis on the device response matrix to generate device control instructions comprises: identifying transmission delay and processing delay by delay source positioning on the device response matrix; evaluating the influence degree of the processing delay by the transmission delay to form a delay matrix; generating a dominant delay vector by eigenvalue decomposition of the delay matrix; and generating device control instructions by compensation correction according to the dominant delay vector.

[0085] Identifying transmission delay and processing delay by delay source positioning on the device response matrix; analyzing the time characteristics of each element in the device response matrix to identify the root cause of the response delay; the delay source is mainly divided into two categories: transmission delay comes from the propagation time of signals in physical media, and processing delay comes from the time-consuming of calculation and decision-making process; the transmission delay τ_trans=L / v, where L is the transmission distance and v is the signal propagation speed, this kind of delay has determinacy and predictability; the processing delay includes the cumulative of multiple links such as data acquisition time, calculation processing time and actuator response time; in the device response matrix, transmission delay mainly affects non-diagonal elements because cross-node communication requires longer transmission path; processing delay exists universally in all matrix elements, and its size depends on the complexity of control algorithm and hardware performance; through step response analysis of the device response matrix, pure transmission delay and processing delay components can be separated; transmission delay is manifested as the translation of response curve, and processing delay leads to slow rise of response; the process of locating the delay source needs to analyze the transmission path of the device response matrix one by one and track the complete process of signals from input to output.

[0086] Evaluating the influence degree of the processing delay by the transmission delay to form a delay matrix; analyzing the interaction relationship between transmission delay and processing delay to evaluate the influence degree of transmission delay on processing; when the transmission delay is large, the processing unit may need to wait for data to arrive, causing an increase in processing delay; the influence degree evaluation is performed by calculating the correlation coefficient ρ=cov(τ_trans,τ_proc) / (σ_trans×σ_proc), where cov is covariance and σ is standard deviation; a delay matrix D=[d_ij] is constructed, and the matrix element d_ij=τ_trans_ij+f(τ_trans_ij)×τ_proc_ij, where f(·) is an influence function; the influence function f(τ_trans) reflects the amplification or inhibition of transmission delay on processing delay, which is usually a nonlinear relationship; the delay matrix integrates all time delay information in the device response matrix, forming a complete delay distribution picture; the rows and columns of the matrix correspond to the input and output channels of the device response matrix, maintaining the consistency of the topology structure; the symmetry of the delay matrix reflects the delay characteristics of bidirectional communication, and the asymmetric part is due to the difference in one-way transmission or processing; through the influence degree analysis of transmission delay on processing delay, the delay bottleneck and critical path in the system can be identified.

[0087] The dominant delay vector is generated by eigenvalue decomposition of the delay matrix; the constructed delay matrix is subjected to eigenvalue decomposition: D = V x Lambda x V^(-1), wherein V is an eigenvector matrix, and Lambda is an eigenvalue diagonal matrix; the size of the eigenvalue lambda_i reflects the delay intensity of the corresponding characteristic mode, and the largest eigenvalue corresponds to the dominant delay mode; the dominant delay vector is defined as the eigenvector v_max corresponding to the largest eigenvalue, which indicates the main direction of system delay; the eigenvalue decomposition reveals the internal structure of the delay matrix, simplifying the complex delay distribution into several main modes; the eigenvector corresponding to the second largest eigenvalue represents a secondary delay mode, which may become important after compensation of the dominant mode; the components of the dominant delay vector represent the contribution of different control channels to the dominant delay mode, and the larger the absolute value of the component, the more significant the contribution; by analyzing the distribution of eigenvalues, the difficulty of delay compensation can be judged, and a concentrated eigenvalue indicates a single delay mode and easy compensation; the dominant delay vector has directionality, and the positive and negative components correspond to the enhancement and inhibition effects of delay, respectively; the orthogonality of the eigenvectors ensures the independence between different delay modes, which can be compensated separately.

[0088] The compensation correction generates equipment control instructions according to the dominant delay vector; based on the identified dominant delay vector, a targeted compensation strategy is designed, and the delay mode with the greatest impact is preferentially compensated; the basic idea of compensation correction is to apply control action in the opposite direction of the dominant delay vector to offset the delay effect; the generation formula of the equipment control instruction is: u_comp(t) = u_base(t) - alpha x v_max x e(t - tau_max), wherein u_base is the basic control instruction, alpha is the compensation gain, v_max is the dominant delay vector, e is the error signal, and tau_max is the dominant delay; the selection of the compensation gain alpha needs to balance the compensation effect and system stability, and too large compensation gain alpha will cause oscillation, and too small compensation gain alpha will result in insufficient compensation; the channels corresponding to the large components in the dominant delay vector obtain stronger compensation effect, achieving precise compensation; the equipment control instructions generated by compensation correction mainly include four types: speed regulation instructions for adjusting the running speed of the main shaft and auxiliary shaft of the engineering equipment, power distribution instructions for realizing optimal distribution between the main power output and auxiliary power output, vibration suppression instructions for actively controlling the detected abnormal vibration, and energy-saving mode switching instructions for switching between normal operation and energy-saving working zones according to the dead zone analysis results; these control instructions are subjected to timing adjustment to ensure that they reach the corresponding actuators at the correct time; the control instructions contain feedforward compensation and feedback correction, the feedforward part predicts the equipment state change based on the dominant delay vector, and the feedback part dynamically adjusts according to the actual running effect; the compensation correction process continuously monitors the delay change, and adjusts the compensation parameters in a timely manner when the system characteristics change, ensuring that the control instructions are always adapted to the actual running state of the engineering equipment, and finally completing the intelligent monitoring and automatic adjustment of the engineering equipment running.

[0089] In order to implement the above-mentioned method embodiment corresponding to an engineering equipment operation intelligent monitoring and automatic adjustment method, to realize the corresponding functions and technical effects; see Figure 2 , Figure 2 The structure block diagram of the engineering equipment operation intelligent monitoring and automatic adjustment system 200 provided by the embodiment of the application is shown; for the convenience of description, only the part related to the embodiment is shown, and the engineering equipment operation intelligent monitoring and automatic adjustment system 200 provided by the embodiment of the application comprises:

[0090] The parameter acquisition module 201 is configured to acquire physical operation parameters and working condition signals of the engineering equipment, construct an equipment operation situation map based on the physical operation parameters, generate a control reference line by performing stable domain identification on the operation situation map, and form a feedback regulation loop based on the control reference line and the working condition signals.

[0091] The channel decoupling module 202 is configured to perform control channel decoupling analysis based on the feedback regulation loop to generate an independent control sub-loop, perform selective recoupling processing on the independent control sub-loop to form an enhanced coupling channel, identify a resonance risk point through the enhanced coupling channel, and construct a safe operation envelope by avoiding the resonance risk point.

[0092] The dead zone analysis module 203 is configured to perform response boundary analysis on the safe operation envelope to generate dead zone characteristic parameters, perform dead zone energy analysis using the dead zone characteristic parameters to identify available dead zones, convert the available dead zones into energy-saving working zones to form a low-power consumption operation trajectory, and generate a power allocation scheme along the low-power consumption operation trajectory.

[0093] The power distribution module 204 is configured to decompose the power allocation scheme into a main power output and an auxiliary power output, perform margin evaluation on the auxiliary power output to extract a standby power pool, and perform energy aggregation on the standby power pool to form an emergency power reserve.

[0094] The fault compensation module 205 is configured to perform fault precursor detection on the main power output to obtain an abnormal vibration signal, analyze vibration amplitude based on the abnormal vibration signal to generate a power compensation demand, and according to the power compensation demand, retrieve corresponding power from the emergency power reserve to form a compensation power flow, and perform non-disturbance switching between the compensation power flow and the main power output to generate a stable output network.

[0095] The control output module 206 is configured to construct a device response matrix based on the stable output network, and perform time delay compensation analysis on the device response matrix to generate a device control instruction.

[0096] The engineering equipment operation intelligent monitoring and automatic adjusting system 200 can implement an engineering equipment operation intelligent monitoring and automatic adjusting method of the method embodiment; the optional items in the method embodiment are also applicable to the present embodiment, which will not be described in detail herein; the remaining content of the present embodiment can be referred to the content of the method embodiment, which will not be described in detail herein.

[0097] The purpose of the above embodiments is to exemplarily reproduce and deduce the technical solutions of the present application, and to completely describe the technical solutions, purposes and effects of the present application, so that the public can understand the disclosure of the present application more thoroughly and comprehensively, and the protection scope of the present application is not limited thereby.

[0098] The above embodiments are not exhaustive enumeration based on the present application, and there can be many other unlisted embodiments outside of this; any replacement and improvement made without violating the concept of the present application is within the protection scope of the present application.

Claims

1. A method for intelligent monitoring and automatic adjustment of engineering equipment operation, characterized in that, include: Collect physical operating parameters and operating condition signals of engineering equipment, construct an equipment operation status diagram based on the physical operating parameters, perform stability domain identification on the operation status diagram to generate a control baseline, and form a feedback adjustment loop based on the control baseline and the operating condition signals; Based on the feedback regulation loop, control channel decoupling analysis is performed to generate independent control sub-loops. Selective recoupling processing is performed on the independent control sub-loops to form enhanced coupling channels. Resonance risk points are identified through the enhanced coupling channels, and the resonance risk points are avoided to construct a safe operation envelope. Response boundary analysis is performed on the safe operation envelope to generate dead zone characteristic parameters. Dead zone energy analysis is performed using the dead zone characteristic parameters to identify usable dead zones. The usable dead zones are transformed into energy-saving working areas to form low-power operation trajectories. Power allocation schemes are generated along the low-power operation trajectories. The power allocation scheme is decomposed into main power output and auxiliary power output. The margin of the auxiliary power output is evaluated to extract a backup power pool. The backup power pool is then used to accumulate energy to form an emergency power reserve. The active power output is subjected to fault symptom detection to obtain abnormal vibration signals. Based on the abnormal vibration signals, the vibration amplitude is analyzed to generate power compensation requirements. According to the power compensation requirements, corresponding power is retrieved from the emergency power reserve to form a compensation power flow. The compensation power flow is seamlessly switched with the active power output to generate a stable output network. Based on the stable output network, a device response matrix is ​​constructed, and time delay compensation analysis is performed on the device response matrix to generate device control commands.

2. The method according to claim 1, characterized in that, The step of forming a feedback adjustment loop based on the control baseline and the operating condition signal includes: A reference deviation band is generated based on the control baseline; The operating condition signal and the reference deviation band are combined to form a signal-deviation correlation characteristic; Extract a stable control range within the signal-deviation correlation characteristics; A feedback adjustment loop is formed based on the gain value within the stable control range.

3. The method according to claim 1, characterized in that, The identification of resonance risk points through the enhanced coupling channel includes: The enhanced coupling channel is divided into a core frequency band and an edge frequency band; Transmit a frequency scanning path from the core frequency band to the edge frequency band; Record the positions of amplitude spikes along the frequency scanning path to form a set of spike points; The point with the largest amplitude in the set of sudden increase points is marked as the resonance risk point.

4. The method according to claim 1, characterized in that, The method of using the dead zone characteristic parameters to perform dead zone energy analysis and identify usable dead zones includes: Energy flow blocking identification is performed on the dead zone characteristic parameters to generate a flow resistance termination region; An energy-saving factor is formed by assessing the energy-saving potential based on the flow resistance termination region. A continuous energy-saving distribution is generated by response interpolation using the energy-saving factor; Threshold extraction is performed based on the continuous energy-saving distribution to generate usable dead zones.

5. The method according to claim 1, characterized in that, The process of accumulating energy in the backup power pool to form an emergency power reserve includes: Construct a power time axis based on the aforementioned backup power pool; Map the power peak point onto the power time axis to form a peak moment marker; The power time axis is divided into a charging cycle and an energy release cycle, with the peak time mark as the boundary. An emergency power reserve is formed by comparing the energy distribution characteristics of the charging cycle and the releasing cycle.

6. The method according to claim 1, characterized in that, The step of seamlessly switching the compensated power flow with the active power output to generate a stable output network includes: The compensated power flow is converted into a power vector distribution; Find the equilibrium center point in the power vector distribution; Using the equilibrium center point as a seed, power diffusion is carried out to form a preliminary stable region; The initial stable region is then reinforced at its boundaries to form a stable output network.

7. The method according to claim 1, characterized in that, The step of performing time-delay compensation analysis on the device response matrix to generate device control commands includes: The device response matrix is ​​used to locate and identify the sources of delay, resulting in transmission delay and processing delay. The impact of the transmission delay on the processing delay is evaluated to form a delay matrix; The dominant delay vector is generated by eigenvalue decomposition of the delay matrix. The device control commands are generated by performing compensation and correction according to the dominant delay vector.

8. The method according to claim 3, characterized in that, The recorded positions of amplitude spikes along the frequency scanning path form a set of spike points, including: The analysis window is determined by identifying amplitude jump characteristics based on the frequency scanning path. The amplitude jump characteristics include the rise slope, peak duration, and decay rate. The amplitude change process is traced along the analysis window to form an amplitude trajectory diagram; Extract the frequency coordinates of each sudden increase point from the amplitude trajectory diagram; A set of abrupt increase points is generated by arranging the amplitudes according to the frequency coordinates.

9. The method according to claim 5, characterized in that, The process of forming an emergency power reserve by comparing the energy distribution characteristics of the charging cycle and the releasing cycle includes: The power sequence of the charging cycle is converted into an energy accumulation sequence; The power sequence of the energy release cycle is misaligned and superimposed onto the energy accumulation sequence to form an energy difference map; Extract the cumulative energy jump within the energy difference map; Emergency power reserves are formed based on the distribution density of the accumulated energy jump.

10. An intelligent monitoring and automatic adjustment system for the operation of engineering equipment, characterized in that, include: The parameter acquisition module is used to acquire the physical operating parameters and operating condition signals of the engineering equipment, construct an equipment operation status diagram based on the physical operating parameters, perform stability domain identification on the operation status diagram to generate a control baseline, and form a feedback adjustment loop based on the control baseline and the operating condition signals. The channel decoupling module is used to perform control channel decoupling analysis based on the feedback adjustment loop to generate independent control sub-loops, selectively recouple the independent control sub-loops to form enhanced coupling channels, identify resonance risk points through the enhanced coupling channels, avoid the resonance risk points, and construct a safe operation envelope. The dead zone analysis module is used to perform response boundary analysis on the safe operation envelope to generate dead zone characteristic parameters, use the dead zone characteristic parameters to perform dead zone energy analysis to identify usable dead zones, convert the usable dead zones into energy-saving working areas to form low-power operation trajectories, and generate power allocation schemes along the low-power operation trajectories. The power allocation module is used to decompose the power allocation scheme into main power output and auxiliary power output, perform margin assessment on the auxiliary power output to extract a backup power pool, and perform energy accumulation on the backup power pool to form an emergency power reserve. The fault compensation module is used to detect fault signs of the main power output to obtain abnormal vibration signals, analyze the vibration amplitude based on the abnormal vibration signals to generate power compensation requirements, retrieve corresponding power from the emergency power reserve according to the power compensation requirements to form a compensation power flow, and seamlessly switch the compensation power flow with the main power output to generate a stable output network. The control output module is used to construct a device response matrix based on the stable output network, perform time delay compensation analysis on the device response matrix, and generate device control commands.

Images