Hierarchical control method for unit one-key start-stop

Abstract

The application discloses a unit one-key start-stop hierarchical control method and relates to the technical field of industrial automation control, and comprises the following steps: continuously collecting load signals and self-checking beat signals in the unit start-stop process, superimposing processing along a time axis, and generating a self-checking rhythm disturbance trajectory band reflecting beat compression and stretching changes; comparing the self-checking rhythm disturbance trajectory band with the start and end moments of the self-checking in sections, separating beat segments affected by load sudden rise, reconstructing the opening and closing boundary sequences of the safety window and embedding beat disturbance information. Through multi-signal superposition and rhythm disturbance identification, the application realizes adaptive identification of the start-stop rhythm and dynamic reconstruction of the safety window, guarantees the continuity and stability of the unit start-stop process under load sudden rise, and through breathing type time gap rearrangement control based on risk levels, dynamically matches the oil discharge stage and the pressure release process, realizes adaptive pressure reduction and collaborative improvement of start-stop safety.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, specifically to a hierarchical control method for one-button start / stop of generating units. Background Technology

[0002] One-button start / stop hierarchical control of the generating unit refers to breaking down the multi-stage, multi-device, and multi-state linkage process, which was originally operated manually in sequence, into several logically progressive control levels during the unit's start-up and shutdown. Upon receiving a single start / stop command, the system automatically completes status determination, condition checks, execution sequence arrangement, and safety linkage response according to the functional division of each control level. Specifically, the upper level is responsible for the unified analysis of start / stop objectives and overall rhythm planning; the middle level is responsible for confirming the operating conditions of key subsystems and organizing the action sequence; and the lower level is responsible for the precise response and safety protection actions of specific execution units. Through this top-down hierarchical structure, the start / stop process no longer relies on manual item-by-item confirmation, but is automatically coordinated by the system, enabling the unit to complete start-up or shutdown actions at a controlled pace even under complex operating conditions. This ensures consistency of actions, reduces the risk of misoperation, and improves the operational safety and response efficiency of the entire start / stop process.

[0003] The existing technology has the following shortcomings: In the existing one-button start-stop control mode, the unit start-stop sequence relies on a preset self-check rhythm to confirm key operating states in stages, ensuring that each action is completed within a safe window. However, in scenarios of sudden load increases, the self-check rhythm is easily affected by transient impacts, causing rhythmic jitter, which may lead the system to mistakenly determine that the self-check window has ended before the necessary state confirmation is completed. Such misjudgments will directly trigger the protection freeze mechanism of the start-stop link, causing subsequent actions to remain in a frozen state. In order to maintain the start-stop rhythm, the upper-level scheduling module often performs forced state refresh, causing the incomplete critical oil draining step to be skipped. Since the oil draining stage is irreplaceable in the internal pressure release process of the unit, skipping this step will cause residual pressure to accumulate continuously in the sealed cavity. If the pressure accumulation cannot be released in time, it will cause sudden stress concentration on the pressure-bearing structural components, thereby leading to a serious risk of shell rupture or even overall failure.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a one-button start-stop hierarchical control method for generator units to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a one-button start / stop hierarchical control method for generator units, comprising the following steps: During the start-up and shutdown of the unit, load signals and self-test beat signals are continuously collected and superimposed along the time axis to generate a self-test rhythm disturbance trajectory band that reflects the changes in beat compression and stretching. Based on the self-test rhythm disturbance trajectory band, the start and end times of the self-test are compared segment by segment to separate the beat segments affected by the sudden increase in load, reconstruct the opening and closing boundary sequence of the safety window and embed the beat disturbance information. Expand the safety window boundary sequence along the time axis and extend it to the oil discharge start point to form an expanded interval containing self-check end information, and lock the time gap of the key oil discharge stage in the expanded interval. Within the continuous window of the time gap during the critical oil discharge stage, the cavity pressure signal and the oil discharge valve command status are simultaneously acquired, and the two types of signals are superimposed along the time axis to generate a residual pressure evolution indicator curve. Based on the residual pressure evolution indicator curve, three features are extracted: pressure ramp-up slope, peak duration, and self-test recovery delay. These three features are then mapped to the load surge start-up and shutdown risk level. Using the risk level of sudden load increase and start-stop as the control driving quantity, breathing time slot rearrangement control is performed on the time slot of the critical oil discharge stage, so that the oil discharge command slides back and forth along the time axis, and the time slot is expanded under high risk level, forming an adaptive decompression rhythm with dynamic decompression capability.

[0007] Preferably, the steps for generating the self-checking rhythm perturbation trajectory band are as follows: After the unit enters the start-up and shutdown process, load signals and self-test cycle signals are continuously collected. Load signals include motor output power, shaft torque, fluid flow, exhaust pressure, temperature and vibration signals. Self-test cycle signals are used to mark the start and end times of each stage of the start-up and shutdown procedure and are synchronously sampled with the start-up and shutdown control master clock as the time reference. The collected load signal and the self-test beat signal are superimposed along a unified time axis, and the timestamp is used as the corresponding basis to make the load change curve and the beat pulse synchronized in time. After superposition, time expansion is performed in units of self-test stage to make the change trend of load signal form a continuous response distribution with clock compression and clock stretching. After time expansion, a normalized mapping is performed based on the total duration of the start-stop process to generate a self-checking rhythm disturbance trajectory band that is continuously distributed along the time direction, which is used to characterize the dynamic coupling relationship between load changes and beat rhythm.

[0008] Preferably, the steps for reconstructing the opening and closing boundary sequence of the safety window are as follows: After obtaining the self-test rhythm disturbance trajectory band, determine the time boundary range of each self-test stage in the entire start-stop process, and take the rising edge and falling edge of the self-test beat signal as the start time and end time of the self-test, respectively. Based on the analysis of the self-check rhythm disturbance trajectory, each self-check stage is compared segment by segment to identify the beat segments in the trajectory affected by the sudden increase in load. Based on the comparison results, the disturbed beat segments are separated from the normal beats, and the start time of the self-test, the point of sudden rise in the load signal, and the end time of the self-test are recorded. Based on the beat separation results, the opening and closing boundary sequence of the safety window is reconstructed, and the beat disturbance information is embedded into the time boundary of the safety window, so that the opening and closing time of the safety window can be dynamically adjusted according to the load.

[0009] Preferably, when reconstructing the safety window opening and closing boundary sequence, if the clock disturbance occurs in the early stage of the self-test phase, the safety window opening time is extended forward to compensate for the time deviation caused by clock compression; if the clock disturbance occurs in the later stage of the self-test phase, the safety window closing time is extended backward to match the load recovery process, thereby ensuring that the self-test phase maintains a complete execution cycle after a sudden load surge and achieves continuous connection of start and stop timing.

[0010] Preferably, the steps for expanding and locking the time gap of the critical oil discharge stage along the time axis in the opening and closing boundary sequence of the safety window are as follows: The opening and closing boundary sequence of the safety window is expanded along the time axis. Taking the total start and stop time as a reference, each safety window is arranged in sequence and adjacent boundaries are continuously connected by time translation. Starting from the end time of the self-check of each safety window, the time is extended to the start time of the oil discharge action, forming an extended interval covering the two. Divide the extended range into stages from the end of self-inspection to the preparation for oil draining, and determine that the oil draining action should be performed from the end of the load stabilization stage to the beginning of the oil draining preparation stage. Based on the phase division results, the start and end boundaries of the key oil drainage phase time gaps are marked and fixed on the time axis, so as to serve as the execution time window for the oil drainage action during the start-up and shutdown process.

[0011] Preferably, the steps for generating the residual pressure evolution indicator curve are as follows: A time duration window is established based on the time gap of the critical oil discharge stage, and time extension is performed at both ends of the time gap to cover the entire process from the end of the self-test to the complete balance of the cavity pressure. Within the time duration window, the cavity pressure signal and the oil discharge valve command status signal are simultaneously acquired, and the timestamps of the two types of signals are recorded with a unified time reference. The chamber pressure signal and the oil drain valve command status are superimposed along the time axis, and the two are arranged correspondingly at the same time point with time as the horizontal axis. Based on the superposition results, a continuous curve of cavity pressure changing with time is plotted, and key moments of valve opening, full opening and closing are marked on the curve to generate a residual pressure evolution indicator curve reflecting the relationship between pressure changes and valve command changes.

[0012] Preferably, the cavity pressure signal is collected simultaneously at the main pressure detection point, the oil passage pressure detection point, and the exhaust detection point in the oil discharge cavity. The oil discharge valve command status signal records the entire process of valve opening, response, full opening, and closing. During superposition processing, a unified time reference is used for smooth alignment so that the pressure change curve and the valve action curve form a continuous correspondence on the time axis.

[0013] Preferably, the steps for mapping the residual pressure evolution indicator curve to the load surge start-stop risk level are as follows: The residual pressure evolution indicator curve is identified in the time domain and divided into stages, distinguishing the pressure response stage, the rapid pressure decrease stage, the stable pressure release stage, and the pressure decay convergence stage. In each stage, three features were extracted: pressure ramp rate, peak duration, and self-test recovery delay, which were used to characterize the pressure change rate, pressure duration, and system recovery time. Establish the time correspondence between pressure ramp-up slope, peak duration, self-check recovery delay, and oil drain valve command status to form a continuous feature chain describing the entire oil draining process; Based on the combined state of three characteristics—pressure ramp-up slope, peak duration, and self-test recovery delay—the three characteristics are comprehensively mapped to a load surge start-stop risk level, enabling the risk level to have a dynamic response capability to the evolution of residual pressure.

[0014] Preferably, the steps for implementing breathing-type time slot rearrangement control based on the load surge start-stop risk level are as follows: The risk level of sudden load increase in start-stop is used as the time management logic of the start-stop control link to input the control drive signal, forming a risk drive signal sequence of drive oil discharge timing changes; Establish a time response benchmark for the time gap in the critical oil drainage stage, and determine the center benchmark point of the time gap based on the standard execution rhythm of the oil drainage stage. Based on risk-driven signals, breathing-type time-slip control is implemented within the time gap of the critical oil discharge stage, so that the oil discharge command generates a back-and-forth slip on the time axis. When the risk level exceeds the set threshold, the time gap of the critical oil discharge stage will be extended backward. An adaptive decompression rhythm with dynamic decompression capability is formed through the linkage of risk-driven signals, breathing-style slip control, and time slot extension.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention introduces time-axis-based multi-signal superposition and rhythm disturbance recognition during start-up and shutdown processes, transforming the determination of unit start-up and shutdown states from a single logical timing control to continuous dynamic correlation control. By time fusion and disturbance stripping of load signals and self-test beat signals, adaptive recognition of start-up and shutdown rhythms and dynamic reconstruction of safety windows are achieved, enabling the system to maintain temporal continuity and state consistency during start-up and shutdown even under sudden load surges. This approach effectively avoids self-test misjudgments and action freezes caused by transient interference during start-up and shutdown, significantly improving the stability and reliability of the unit start-up and shutdown link.

[0016] This invention introduces a risk-level-driven, breathing-style time-slot rearrangement control, enabling the execution time of the oil draining phase to adaptively adjust with pressure evolution trends. This achieves dynamic coordination between the oil draining action and the pressure release process in start-up and shutdown control. This control method allows the oil draining rhythm to automatically slide or expand according to the risk level, ensuring sufficient release of residual pressure and thus avoiding structural stress concentration caused by pressure accumulation within the chamber. Through this dynamic pressure reduction mechanism, the unit's pressure response during start-up and shutdown is more stable, and overall operational safety and start-up and shutdown efficiency are simultaneously improved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a flowchart of the hierarchical control method for one-button start / stop of the unit according to the present invention. Detailed Implementation

[0019] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0020] This invention provides, for example Figure 1 The one-button start / stop hierarchical control method for the generator units shown includes the following steps: During the start-up and shutdown of the unit, load signals and self-test beat signals are continuously collected, and the load signals and self-test beat signals are superimposed along the time axis to make the beat compression characteristics and beat stretching characteristics continuously distributed on the time axis, generating a self-test rhythm disturbance trajectory band. To achieve high-precision coupled control of load changes and self-check cycle status during unit start-up and shutdown, and to fully identify the patterns of cycle compression and stretching and form a continuous response distribution over time, a process of continuous data acquisition, time overlay, feature expansion, and trajectory generation is employed to create the self-check rhythm disturbance trajectory band. The specific implementation process is as follows: After the unit enters the start-up and shutdown process, a continuous acquisition process of load signals and self-test cycle signals is initiated. The acquired load signals are derived from the real-time operating parameters of key actuators during unit operation, including physical quantities that characterize changes in operating load, such as motor output power, shaft torque, fluid flow rate, exhaust pressure, temperature, and vibration signals. The self-test cycle signal is generated by the self-test process in the start-up and shutdown control logic and is used to mark the start and end times of each stage of the unit's start-up and shutdown procedure. To ensure consistency of the two types of signals in the time dimension, sampling is performed using a unified time reference, provided by the start-up and shutdown control master clock. The sampling period is determined based on the unit's inertial characteristics, ensuring that the sampling frequency is sufficient to cover the dynamic changes during all state transitions in the start-up and shutdown phases. Signal acquisition adopts a continuous recording method, meaning that from the moment the unit enters the start-up preparation phase from a static state until it completes stable operation or a complete shutdown, the load signals and self-test cycle signals are continuously acquired throughout the entire process. Each load signal and self-test cycle signal carries a corresponding timestamp to ensure accurate correspondence between the two types of signals in subsequent steps. In this way, it is ensured that the transient process of load change and the logical change of self-test beat maintain a one-to-one correspondence in time, providing basic data for subsequent signal superposition and trajectory construction.

[0021] After signal acquisition, the acquired load signals and self-test cycle signals are superimposed using a unified time axis as a reference. The superposition process uses timestamps as the core identifier, arranging the two types of signals corresponding to the same time point, ensuring that load changes and cycle rhythm changes are synchronized in the time dimension. The self-test cycle signal typically appears as a periodic pulse, with each pulse corresponding to an execution cycle of a self-test phase; the load signal appears as a continuously changing curve, reflecting the unit's physical response during its respective self-test phase. During superposition, the start and end times of each self-test cycle are used as the boundaries of a time window. Within this window, the load signal change curve is time-normalized, ensuring that the load change process completely covers the corresponding cycle period. To avoid time drift during signal superposition, all time window boundaries are interpolated and smoothed to maintain continuity in the transition zone between cycles. This processing method allows the impact of load changes on the self-test cycle rhythm to be expressed in a time-continuous form. The superimposed signal exhibits obvious fluctuation characteristics on the time axis, with compression of the beat interval during periods of sudden load increases and stretching of the beat interval during periods of stable load. By superimposing the time axis, a correspondence between load signal changes and self-test beat changes is established, ensuring that the temporal changes during the start-stop phase are completely recorded within the same time distribution space, providing the original structure for forming a continuous disturbance trajectory.

[0022] After signal superposition, the superimposed signal undergoes time unfolding and structuring to form a continuous perturbation distribution reflecting the interaction between load changes and rhythm. During the unfolding process, the superimposed signal is segmented along the time axis, with each self-test phase's time interval serving as the unfolding unit, and the load signal change trend within that interval as the unfolding content. Specifically, in intervals where the load signal rises rapidly, the signal curve exhibits a steep slope, corresponding to the compression interval of the self-test rhythm; in intervals where the load signal changes slowly, the signal curve has a smaller slope, corresponding to the stretching interval of the self-test rhythm. This time unfolding creates a continuous response distribution for different rhythm phases. To ensure the continuity of each phase, the boundaries of adjacent rhythms are time-fused during the unfolding process, eliminating gaps or overlaps between adjacent rhythms. After unfolding, a continuous signal band spanning the entire start-stop process is obtained on the time axis, reflecting the coupling characteristics between load changes and rhythm. This signal band not only reveals the temporal distribution of the cycle time changes during the start-up and shutdown process, but also shows the dynamic influence of load changes on the cycle time rhythm. Thus, it can clearly reflect the temporal transition pattern between the area where the cycle time is compressed when the load suddenly increases and the area where the cycle time is stretched when the load gradually decreases.

[0023] After time unrolling, the resulting continuous signal band is normalized and continuously mapped to generate a self-checking rhythm disturbance trajectory band. During normalization, the total duration of the entire start-stop phase is used as the standard time reference, and the durations of different beat phases are uniformly mapped to relative time coordinates, making the relative changes between each beat comparable on the same time scale. Subsequently, using the amplitude change of the load signal curve as the longitudinal distribution feature, the signal on the time axis is continuously mapped into a disturbance trajectory extending along the time direction. This trajectory band forms a continuous distribution on the time axis, with its compressed sections corresponding to the beat contraction response during the load surge phase and its stretched sections corresponding to the beat delay response during the load stabilization phase. The formation process of the trajectory band not only preserves the temporal correspondence between the load signal and the self-checking beat signal but also integrates the change characteristics of the two types of signals into a dynamic curve with temporal continuity. Through this trajectory band, the disturbance pattern of the beat rhythm during start-stop as the load changes can be intuitively observed in the time dimension, thus providing a basis for subsequent safety window reconstruction and start-stop risk identification.

[0024] Using the self-check rhythm disturbance trajectory as input, the start and end times of each self-check are compared segment by segment. The beat segments affected by the sudden load increase in the trajectory are separated from the normal beats. Based on the separation results, the opening and closing boundary sequence of the safety window is reconstructed, so that the beat disturbance information is embedded in the safety window boundary sequence. To ensure that the unit can accurately identify the rhythm changes caused by sudden load increases during start-up and shutdown, and to fully reflect these changes in the opening and closing time structure of the safety window, thereby enabling the safety window to have adaptive adjustment capabilities and achieving rhythm coordination and operational safety during start-up and shutdown, the specific implementation steps are as follows: After obtaining the self-test rhythm disturbance trajectory, the time boundary range of each self-test stage in the entire start-up and shutdown process is determined. Each self-test stage corresponds to a specific physical action in the unit start-up and shutdown chain, such as establishing the fuel supply path, filling the coolant circulation, establishing the lubricating oil pressure, opening the drain valve, starting the main drive shaft, maintaining stable speed, and closing the exhaust valve. The self-test beat signal of each stage is represented as an independent time period on the time axis, with its start time indicating the start point of the action in that stage and its end time indicating the end point of the self-test. To ensure the accurate correspondence between the trajectory and the self-test timing, the start and end times of each stage are calibrated in this step. During the calibration process, the rising and falling edges of the self-test beat signal are used as the references for the start and end of the self-test, respectively, while the response change of the load signal at the same time point is recorded. This forms a set of one-to-one corresponding self-test timing segments on the time axis, each segment containing both the load response curve and the beat change trajectory, providing the basic time unit for subsequent segment-by-segment comparison.

[0025] After time segmentation, the trajectory pattern of each self-check stage is compared segment by segment based on the self-check rhythm disturbance trajectory band. This process uses the time axis as a reference, comparing the trajectory pattern within each self-check segment with adjacent segments to analyze the impact of load changes on the beat rhythm. During a sudden load increase, the trajectory band exhibits dense curve aggregation on the time axis, representing beat compression, meaning the duration of the self-check stage is significantly shortened. Conversely, during periods of stable or decreasing load, the trajectory curve shows a widening gap, representing beat stretching, meaning the duration of the self-check stage is extended. To identify these characteristics, this step continuously compares the time span and trajectory pattern of each self-check stage to determine if the stage is affected by a sudden load increase. When the trajectory band shows a rapid rise in the middle or end of a stage accompanied by a significant time compression trend, that stage is identified as a beat segment affected by a sudden load increase. During comparison, to avoid time misalignment between different stages, the start and end boundaries of each stage are time-aligned to ensure seamless temporal connection between consecutive stages. By comparing each segment in this way, it is possible to clearly identify on the timeline which beat segments are directly affected by the sudden increase in load and which beat segments maintain a stable rhythm, providing a basis for subsequent beat separation operations.

[0026] After identifying the beat segments affected by load surges, these segments are precisely separated. The core purpose of beat separation is to isolate the disturbed beat segments from the normal beats in the track band, thus providing a clear time boundary for the reconstruction of the safety window. Specifically, using the disturbed intervals determined in the segment-by-segment comparison as the start and end points, the corresponding time periods in the track band are completely extracted and their original order is maintained on the time axis. Each extracted beat segment contains three key nodes: the start time of the self-test, the point of load signal surge, and the end time of the self-test. The load change trend of each beat segment is continuously tracked, recording the rise rate, peak duration, and recovery time of the load signal within that time period. This allows for determining the opening and closing adjustment range of the safety window based on the actual load response changes during the reconstruction process. Simultaneously, for normal beat segments that are not disturbed, their original length and position are maintained. After separation, two types of segments are formed on the time axis: stable beat intervals and disturbed beat intervals. These two types of intervals are continuously connected but logically distinct. This approach presents a clear temporal structure for the entire start-stop timeline, laying the foundation for the reconstruction of the safety window boundaries.

[0027] After separating the beat segments, the opening and closing boundary sequence of the safety window is reconstructed based on the separation results. This ensures that the temporal structure of the safety window accurately reflects the rhythm shift caused by load surges, and that beat disturbance information is embedded into the boundary changes of the safety window. During reconstruction, the time intervals of disturbed beat segments are inserted into the boundary sequence of the safety window, using the original start and end times of each self-test phase as a reference. When a disturbed beat appears in the early stage of the self-test phase, the opening time of the safety window is extended forward, with the extension corresponding to the degree of beat compression during load surges, thus ensuring sufficient self-test preparation time under load surge conditions. When a disturbed beat appears in the middle or late stage of the self-test phase, the closing time of the safety window is correspondingly delayed, with the delay length determined by the duration of load recovery, ensuring that the self-test process does not prematurely end before the load stabilizes again. For undisturbed beat intervals, the original opening and closing times remain unchanged, maintaining the overall continuity of the entire safety window boundary sequence. After reconstruction, the opening and closing boundaries of the safety window are no longer fixed times but become a dynamic temporal structure associated with load changes. Each safety window in this boundary sequence carries clock perturbation information, enabling the start / stop control logic to automatically adjust the execution rhythm of the self-test phase based on load fluctuations. In this way, the safety window achieves adaptive adjustment capability to load perturbations in the time dimension, thereby preventing link freezing and critical action skipping caused by clock misjudgment.

[0028] Expand the opening and closing boundary sequence of the safety window along the time axis, extend the self-test end point in the safety window to the start point of the oil draining action, form an expanded interval containing self-test end information, and lock the time gap of the key oil draining stage in the expanded interval. To ensure the logical timing between the self-check phase and the oil draining phase during unit start-up and shutdown, and to ensure that the oil draining action is performed when the internal pressure, temperature, and fluid state of the unit are all within a controllable range, thereby avoiding abnormal cavity pressure and oil residue problems caused by premature or delayed oil draining, the specific implementation steps are as follows: After obtaining the opening and closing boundary sequence of the safety windows, all safety windows in the sequence are expanded along a time axis to ensure the continuity of the time sequence of the entire start-up and shutdown process. The opening and closing boundary sequence of the safety windows is generated from the preceding stages, where each safety window includes the start and end times of the self-test stage. These time points reflect the operational state boundaries of the unit in different self-test stages. In actual operation, the duration of different self-test stages varies, and direct superposition may cause uneven time intervals or overlapping time sequences. Therefore, during the time axis expansion process, the total start-up and shutdown time is used as a reference benchmark, and each safety window is expanded sequentially along the time axis according to the natural execution sequence from unit start-up to shutdown, so that the boundaries between adjacent safety windows remain continuous in the time dimension. During expansion, for safety windows that have been dynamically adjusted due to load surges or control response delays, their ending boundaries are precisely aligned with the starting boundaries of the next safety window through time translation to avoid time discontinuities. Once expanded, the time distribution of the safety window is no longer a discrete set of intervals, but a time band arranged continuously along the time axis. This time band runs through the entire start-up and shutdown process, forming a continuously traceable time sequence chain, providing a complete time structure basis for subsequent interval extension and key stage identification.

[0029] After the safety window timeline is expanded, time extension processing is performed on the end boundary of each safety window to establish a time correlation between the self-check phase and the oil draining phase. Specifically, the end time of the self-check for each safety window is used as the extension starting point, and the extension ending point is set at the start time of the oil draining action, forming an extension interval covering the complete time period between the two. The extension duration is determined based on the load change characteristics of the unit during start-up and shutdown. That is, if there is still internal medium flow, pressure fluctuation, or temperature gradient imbalance after the unit finishes the self-check, the duration of the extension interval is increased accordingly; if the unit has entered a stable state after the self-check, the duration of the extension interval is appropriately shortened. Through this extension processing, it is ensured that the end information of the self-check phase can completely cover the time range within which the oil draining action is about to be triggered, thus forming a continuous transition relationship on the timeline. The extended interval not only retains the end time information of the self-check phase but also includes the time position of the oil draining start, enabling the unit to logically connect the self-check completion state with the oil draining action during start-up and shutdown. This time extension enables the two originally separate stages to be continuously coupled in the time dimension, providing a traceable time frame for identifying key oil discharge time gaps.

[0030] After forming extended intervals containing self-test completion information, the time periods within each extended interval are characterized to determine the time range of the critical oil draining stage. Each extended interval starts from the self-test completion time and ends at the oil draining initiation time, containing multiple continuous physical processes. Specifically, it is divided into three stages: The first stage is the transition stage from the end of the self-test to the load stabilization stage. During this stage, the flow of oil, gas, or steam inside the unit still has a certain inertia, and the pressure and temperature are fluctuating. The second stage is the load stabilization stage. During this stage, the operating parameters of each actuator of the unit gradually return to the normal range, and the flow rate, pressure gradient, and heat exchange rate of the internal fluid gradually stabilize. The third stage is the oil draining preparation stage. During this stage, the unit control logic enters the state of waiting for the oil draining command, the oil draining valve enters the pre-charge state, and the internal oil is in a static pressure balance process. When dividing the extended intervals into time periods, by continuously observing the timing relationship of each stage, it is determined that the oil draining action should be performed from the end of the second stage to the beginning of the third stage. Within this time range, the internal fluid flow of the unit is basically stable, the cavity pressure is within the release range, and the execution of the oil draining action will not disrupt the overall balance. By defining this time range, the optimal execution period for the oil drainage action can be clearly identified within the extended interval; this period is the critical oil drainage phase time gap. In this way, the most suitable window for performing the oil drainage action can be continuously marked on the time axis, allowing subsequent oil drainage control to operate based on this time gap.

[0031] After identifying the critical oil draining phase time gap, this time gap is calibrated and fixed so that it can be used as a time reference during start-up and shutdown control. During calibration, the start and end points of the extended interval are used as references to precisely mark the start and end boundaries of the critical oil draining phase time gap on the time axis, and the relative time positions are recorded. After time calibration, the duration of the time gap is confirmed to fully cover the entire process required for the oil draining action. The execution time of the oil draining action includes valve response time, oil flow start-up time, oil circuit emptying time, and pressure release stabilization time; the length of the time gap must simultaneously cover these processes. During fixing, the start and end times of the time gap are adjusted to form a continuous connection with the end time of the self-check of the previous stage, avoiding delays or advances in the oil draining action due to timing gaps. After the time gap is fixed, the resulting critical oil draining phase time interval is regarded as the execution time window for the oil draining action during start-up and shutdown. The unit's control logic will use this time gap as a trigger condition, initiating the oil draining command when the arrival of this period is detected. In this way, the oil draining action and the self-check process are closely linked in time, ensuring the logical continuity of each action during start-up and shutdown.

[0032] Based on the time gap of the key oil discharge stage, the cavity pressure signal and the oil discharge valve command status are simultaneously collected within the continuous window of the time gap. The cavity pressure signal and the oil discharge valve command status are superimposed along the time axis to generate a residual pressure evolution indicator curve that reflects the pressure change and command change. To achieve precise control during the critical oil discharge phase of the unit's start-up and shutdown, ensuring a strict time correspondence between the oil discharge action and changes in internal unit pressure, and guaranteeing efficient and safe completion of the oil discharge operation during pressure release, this implementation method uses the time gap of the critical oil discharge phase as a benchmark. Within the continuous window of this time gap, it simultaneously acquires the unit pressure signal and the oil discharge valve command status, and continuously superimposes these two signals along the time axis to generate a residual pressure evolution indicator curve reflecting the relationship between pressure changes and valve command changes. The specific steps are as follows: After determining the time slot for the critical oil drainage phase, a time duration window for signal acquisition is established using this time slot as a reference. The critical oil drainage phase time slot is a time interval locked in the previous stage through time extension and phase division methods. This time interval lies between the oil drainage preparation stage after the self-test and the formal start stage of the oil drainage action. To ensure complete capture of the entire process of cavity pressure changes before and after the oil drainage action, time extension is required at both ends of the original time slot. Specifically, a fixed duration is extended forward from the starting position of the time slot so that the time window covers the initial stage where the cavity pressure begins to change after the self-test; simultaneously, the time slot is extended backward from the ending position to a stable period after the oil drainage action is completely completed to ensure that the decay process of cavity pressure after the oil drainage action is completed can be acquired. The extended time duration window covers the entire process from the end of the self-test to the complete equilibrium of cavity pressure. To ensure the integrity of the time window, the master clock of the start-stop control system is used as a unified time reference during the establishment process, and the absolute timestamps of the start and end points of the window are recorded to form a time acquisition range with defined boundaries. This time duration window provides strict time constraints for subsequent synchronous signal acquisition, ensuring that all signal acquisitions are performed on the same time scale.

[0033] After the time duration window is established, synchronous acquisition of the cavity pressure signal and the drain valve command status is initiated. The cavity pressure signal reflects the pressure change process of the medium inside the unit during the drain process and is an important parameter characterizing the dynamics of pressure release. The signal is acquired from multiple monitoring points inside the unit, including the main pressure detection point in the drain cavity, the oil passage pressure detection point connected to the drain valve, and the exhaust detection point at the cavity outlet. Each monitoring point continuously acquires data within the time duration window. The acquisition frequency should be selected to ensure that instantaneous pressure changes can be recorded throughout the entire drain operation, typically requiring millisecond-level time resolution to ensure the identification of subtle pressure fluctuations. The drain valve command status signal originates from the control command execution link, and its recorded content includes the time of the drain valve opening command, the time of valve response, the time of valve reaching the fully open state, the time of valve closing command, and the time of valve complete closure. To achieve synchronous acquisition of the two types of signals, synchronous recording is performed using the same time base during the acquisition process, ensuring that the cavity pressure signal and the valve command status signal are acquired and stored at the same timestamp. This ensures point-to-point correspondence between the two types of signals during subsequent time-overlay processing, avoiding information misalignment due to time shifts. The result of synchronous acquisition is a set of time-series data, in which the cavity pressure signal reflects the actual process of medium flow and discharge, while the valve command status signal reflects the execution process of control actions. Together, they constitute the basic dataset describing the dynamic behavior during the oil discharge phase.

[0034] After synchronous acquisition, the cavity pressure signal and the drain valve command status are superimposed along the time axis, using time as the core coordinate. The superposition process uses a time duration window as the boundary, mapping the pressure signal and valve status signal at the same timestamp one-to-one. Specifically, the curves of the cavity pressure signal and the valve command status are superimposed on the same time axis, creating a synchronous mapping between them. During the superposition process, the instant the drain valve opens corresponds to the starting point of the pressure curve's descent; the period from when the valve gradually opens to the fully open state corresponds to the interval with the largest rate of pressure curve descent; and the period after the valve closing command is issued corresponds to the stage where the rate of pressure curve descent gradually slows down and stabilizes. To maintain the continuity of the superimposed curves, the sampling intervals of signals from different sources are time-smoothed during the superposition process, ensuring that the transition between the pressure signal and the valve status signal remains time-continuous without jumps. This time superposition method clearly demonstrates the correspondence between valve command changes and cavity pressure responses on the same time scale. After the superposition is completed, two synchronously changing signal curves are formed on the time axis: one is the curve of the cavity pressure changing with time, and the other is the curve of the oil discharge valve command status changing. The two maintain a strict correspondence in time distribution, providing a complete time matching basis for the subsequent generation of pressure evolution curves.

[0035] After superposition, a residual pressure evolution indicator curve is generated based on the superimposed time distribution results. This curve reflects the changing trend of cavity pressure during the oil discharge phase and its correlation with valve command actions. During generation, a continuous trajectory of pressure change is plotted along the time axis, with time as the horizontal axis and the cavity pressure change amplitude as the vertical axis. Key moments in the valve command state are marked on the curve. Specifically, after the oil discharge valve opening command is issued, the curve begins to show a pressure decrease, forming the initial downward segment of the curve. As the valve gradually opens, the rate of oil and gas discharge from the cavity increases, and the slope of the curve gradually increases, forming a rapid pressure release segment. When the valve reaches the fully open state, the oil discharge flow tends to stabilize, and the downward speed of the curve slows down, forming a stable pressure release segment. When the valve closing command is issued, the cavity pressure enters a slow decay stage, and the curve gradually flattens. The entire pressure change process forms a continuous distribution on the time axis. The residual pressure evolution indicator curve plotted in this way can clearly reflect the pressure response at each key moment in the oil discharge process. The curve's shape reveals the dynamic impact of the drain valve's action on the chamber pressure release, providing a direct indicator of whether pressure balance has been achieved during the draining phase. This curve can not only characterize pressure change trends but also serve as a reference signal for subsequent risk identification and timing adjustments, enabling the unit to dynamically perceive the real-time pressure release status during start-up and shutdown.

[0036] Based on the residual pressure evolution indicator curve, three features are extracted: pressure ramp-up slope, peak duration, and self-test recovery delay. These three features are then mapped to the load surge start-stop risk level, enabling the risk level to respond to the residual pressure evolution. To achieve dynamic identification and risk response control of pressure changes during the oil draining phase of the unit during start-up and shutdown, and to ensure that the oil draining action is coordinated with the internal pressure release process of the unit, thereby avoiding the risks of structural shocks, seal failures, or start-up / shutdown logic out-of-synchronization caused by abnormal residual pressure during start-up and shutdown, this step, based on the generated residual pressure evolution indicator curve, extracts three features—pressure rise slope, peak duration, and self-check recovery delay—from the entire pressure change process reflected in the curve. These three features are then mapped in chronological order to a load surge start-up / shutdown risk level, enabling this risk level to respond to the residual pressure evolution process and achieving adaptive risk control during start-up and shutdown. The specific process is as follows: After generating the residual pressure evolution indicator curve, the curve is subjected to continuous time domain identification and stage characteristic division to clearly define the range of pressure state changes on the time axis. The residual pressure evolution indicator curve is a continuous change curve obtained by superimposing the cavity pressure signal and the oil discharge valve command status along the time axis during the time gap of the critical oil discharge stage. The vertical axis of the curve represents the pressure change amplitude, and the horizontal axis represents the time duration, completely recording the entire process from the initiation of the oil discharge action to the complete pressure equilibrium. To facilitate the analysis of the curve, it needs to be divided into four stages according to the curve shape and time distribution. The first stage is the pressure response stage. After the drain valve is initially opened, the fluid inside the cavity begins to flow, and the pressure transitions from a static value to a dynamic value, with the curve showing a downward trend from a stable state. The second stage is the rapid pressure drop stage. During this stage, the medium inside the cavity is discharged more rapidly, the fluid kinetic energy increases, the pressure drop rate reaches its maximum, and the curve slope is steepest. The third stage is the stable pressure release stage. During this stage, the valve opening remains stable, the drain flow rate tends to stabilize, the pressure change rate gradually decreases, and the curve changes from a steep drop to a gentle drop. The fourth stage is the pressure decay and convergence stage. During this stage, the valve gradually closes, the residual fluid flow slows down, the pressure eventually returns to an equilibrium state, and the curve tends to be horizontal. This stage division clearly defines each physical stage of the drain process, providing a temporal and state basis for subsequent pressure characteristic extraction.

[0037] After completing the phase division, three key physical characteristics were extracted from the residual pressure evolution indicator curve for each stage: pressure ramp-up slope, peak duration, and self-check recovery delay. These three characteristics reflect the rate characteristics of pressure response, the duration of pressure bearing, and the system recovery characteristics during the oil discharge phase, respectively. First, the pressure ramp-up slope describes the rate characteristics of the cavity pressure changing from low to high. Although the overall trend of oil discharge is pressure decrease, the internal pressure may experience a momentary ramp-up when the oil discharge valve is first opened due to disturbances in the oil-gas mixture flow or a lag in the valve opening and closing response. By analyzing the rising section of the curve in the initial stage of the pressure response phase, the proportional relationship between pressure change and time in this stage is determined, thus obtaining the pressure ramp-up rate characteristics. Second, the peak duration reflects the length of time the pressure is maintained near the highest point. This period usually occurs between the rapid pressure drop phase and the stable release phase. The curve shows a relatively flat high-pressure plateau state in this interval, indicating that there is still a high residual pressure inside the cavity. This feature can be used to assess the duration of the high-pressure state the unit bears during the oil discharge process, which is of great significance for judging the structural pressure risk. Third, the self-check recovery delay describes the time span required for the pressure to recover to the safe threshold after the oil discharge operation ends. This feature can be determined by analyzing the time difference between the gradual decrease of the curve to a stable level during the pressure decay convergence phase, reflecting the time required for the system to return to a stable state after the oil discharge ends. By extracting the above three features, dynamic response information during the pressure change process can be comprehensively obtained, providing input parameters for risk level calculation.

[0038] After extracting the three features, a temporal and physical correspondence between them is constructed to form a comprehensive feature chain that describes the pressure evolution during the oil discharge phase. This feature chain is built around temporal continuity and based on the physical logic of pressure changes. Specifically, the pressure ramp-up slope is used as the starting point feature of the pressure response during the oil discharge phase, corresponding to the brief pressure disturbance period after the valve's initial opening; the peak duration is used as the intermediate feature, corresponding to the pressure-bearing phase where the pressure stabilizes at a high level; and the self-check recovery delay is used as the ending feature, corresponding to the system equilibrium recovery phase after the oil discharge action. These three features are arranged sequentially on the time axis, forming a complete response chain. To ensure that this feature chain accurately reflects the dynamic changes in cavity pressure, the time intervals of the three features are mapped one-to-one with the time sequence of the oil discharge valve command state when establishing the feature correspondence. For example, the occurrence time of the pressure ramp-up slope corresponds to the initial valve opening change time; the peak duration corresponds to the duration of the valve in its fully open state; and the self-check recovery delay corresponds to the residual fluid decay phase after the valve closes. Through this correspondence, a temporally continuous and physically interconnected sequence structure is formed among the three features. This structure can not only describe the evolution of pressure throughout the oil discharge process, but also reflect the feedback effect of pressure changes on control actions, providing a reliable logical chain for subsequent risk level mapping.

[0039] After establishing a continuous correspondence among the three characteristics, the pressure ramp-up slope, peak duration, and self-check recovery delay are comprehensively mapped to the load surge start-up / shutdown risk level to achieve risk-level response control during start-up and shutdown. The risk level is set based on the impact of pressure change trends on unit operational safety. By comprehensively analyzing the combined states of the three characteristics, the risk level corresponding to different pressure evolution modes is determined. Specifically, a large pressure ramp-up slope indicates drastic pressure changes and strong pressure fluctuations in the cavity during the initial oil discharge phase; a long peak duration indicates prolonged high-pressure bearing in the cavity, increasing the risk of internal structural pressure; and a long self-check recovery delay indicates slow system recovery after oil discharge and high residual internal pressure. In this case, the combined state is mapped to a higher risk level. Conversely, a small pressure ramp-up slope, short peak duration, and short self-check recovery delay indicate smooth oil discharge, sufficient pressure release, and rapid recovery, which is mapped to a lower risk level. Through this mapping relationship, the current start-up / shutdown risk level can be automatically determined based on the real-time extracted three characteristic values ​​during start-up and shutdown control. This risk level is not only a static assessment of the current pressure state, but also possesses dynamic response capabilities, adjusting in real time as pressure changes occur. When the risk level is high, the start-stop control logic automatically extends the safety window time and postpones the triggering of high-load actions in subsequent stages, thereby ensuring sufficient pressure release. When the risk level is low, the start-stop chain continues to execute according to the predetermined rhythm. In this way, the unit can automatically adjust its start-stop control strategy based on the trend of residual pressure changes under complex operating conditions such as sudden load increases, fluid disturbances, or temperature changes, achieving a balance between safety and efficiency.

[0040] Using the risk level of sudden load increase during start-stop as the control driving quantity, breathing time slot rearrangement control is performed on the time slot of the critical oil discharge stage in the start-stop control link, so that the oil discharge command generates reciprocating sliding on the time axis, and the time slot of the critical oil discharge stage is expanded under high risk level, forming an adaptive decompression rhythm with dynamic decompression capability. To enable the unit to automatically adjust the timing of oil discharge operations during start-up and shutdown based on dynamic changes in the start-up and shutdown risk level caused by sudden load increases, and to provide the unit with adaptive decompression capability under high load fluctuations, thereby avoiding the risks of stress concentration, structural deformation, or control logic out-of-synchronization caused by the inability to release residual pressure in the cavity in a timely manner, the start-up and shutdown risk level caused by sudden load increases is used as the control driving variable. A breathing-style time slot rearrangement control is implemented on the time slots of the critical oil discharge stage in the start-up and shutdown control link, causing the oil discharge command to slide back and forth on the time axis. Under high risk levels, the time slots of the critical oil discharge stage are automatically expanded, thus forming a decompression rhythm that can dynamically adjust with changes in risk. The specific implementation method is as follows: After obtaining the risk level of load surge start-stop, this risk level is input as a control drive signal into the time management logic of the start-stop control link to form the basis signal for driving changes in the oil discharge timing. The load surge start-stop risk level is a dynamic result obtained in the previous stage by extracting three characteristics—pressure climb slope, peak duration, and self-check recovery delay—from the residual pressure evolution indicator curve. This risk level reflects the safety level of cavity pressure release during the start-stop phase. When the risk level is low, it indicates stable internal pressure release, a small rate of pressure change, and a safe operating environment during the oil discharge phase; when the risk level increases, it indicates residual pressure accumulation inside the cavity, an increased rate of pressure change, and a higher risk of oil discharge. In this step, the risk level numerical signal is mapped onto the start-stop control time axis in chronological order to form a continuous risk drive signal sequence, enabling the start-stop control logic to perceive the risk level's changing trend over time in real time. Each risk signal value corresponds to a specific moment in the start-stop process and serves as the control basis for subsequent time slot adjustments and oil discharge command slippage. Through this step, the start-stop control link acquires dynamic timing response capability based on changes in risk level, enabling start-stop control to shift from passive execution to proactive adjustment.

[0041] After the risk-driven signal is generated, a response benchmark is established for the critical oil draining phase time slot, providing a definite time reference point for subsequent time sliding and extension adjustments. The critical oil draining phase time slot is a specific time interval determined by the time axis extension method after the unit's self-inspection ends and before the oil draining action is initiated. Its starting point is the self-inspection end time, and its ending point is the oil draining valve action end time. To enable reasonable adjustment of the time slot according to the risk level, a time response benchmark needs to be established. In specific implementation, based on the standard execution rhythm of the oil draining phase, the center moment of the time slot is defined as the initial response benchmark point. When the risk level is low, the center point of the time slot remains unchanged, and the oil draining command is executed at the originally scheduled time. When the risk level rises to a medium level, the center benchmark point of the time slot moves backward along the time axis, slightly delaying the oil draining action to allow more time for pressure release in the chamber. When the risk level further rises to a high level, the center point of the time slot slides backward again, with the delay proportional to the increase in risk level. To avoid control fluctuations caused by excessively rapid time adjustments, smoothing buffers are set before and after the baseline point to ensure the continuity and gradualness of the time glide process. Through this step, the time gap acquires a time response mechanism that automatically adjusts according to changes in risk level, laying the foundation for breathing-style glide.

[0042] After establishing the time slot response benchmark, based on the periodic changes in the risk-driven signal, a breathing-style time-slip control is implemented within the time slots of the critical oil drainage phase. This causes the oil drainage command to slide back and forth on the time axis, thereby achieving flexible rhythm adjustment of the oil drainage action. Breathing-style time slip refers to the start and end boundaries of the time slot moving back and forth along the time axis driven by the periodic changes in risk level, thus continuously fine-tuning the trigger time of the oil drainage action within a small range. In this step, when the risk level is moderate, the center point of the time slot slides periodically along the time axis with a fixed amplitude, forming a dynamic change of "contraction-recovery" in time. When the risk level fluctuates significantly, the slip amplitude increases with the change in risk level, and the slip period lengthens accordingly, forming a slower breathing rhythm, causing a controllable delay in the execution time of the oil drainage command on the time axis. In this way, the oil drainage action is no longer triggered at a fixed time, but dynamically adjusted on the time axis according to changes in risk level. During the slip process, the start and end points of the time slot maintain a constant relative proportion to ensure the overall rhythm coordination of the oil drainage action. This breathing-style sliding control allows for a smoother transition in the chamber pressure release process, preventing sudden pressure drops or residual pressure rebounds caused by a fixed execution rhythm during oil discharge.

[0043] Based on the stable operation of the breathing-type time-slide control, when the risk level continuously rises and exceeds the set high-risk threshold, the time gap of the critical oil draining stage is extended, stretching it backward on the time axis to form a dynamically delayed pressure reduction zone. Specifically, when the risk level rises to the high-risk zone and the residual pressure curve shows a continuous, slow decline and significant pressure recovery delay, the start-stop control logic automatically extends the duration of the critical oil draining stage. During the extension process, the starting point of the time gap remains unchanged, while the ending point is moved backward, extending the time the drain valve remains open and thus increasing the draining duration. Simultaneously, within the extended time gap, the drain valve command remains open, allowing the cavity pressure to be released over a longer period until the pressure curve approaches the safety threshold before triggering the valve closing command. To prevent excessive time extension from causing subsequent start-stop actions to accumulate, an upper limit constraint is set during the time extension process, ensuring that the maximum extension of the time gap does not exceed a predetermined proportion. Through this step, the time structure of the oil draining stage can be automatically stretched under high-risk conditions, ensuring sufficient release of residual pressure and avoiding stress concentration or valve oscillation caused by pressure accumulation.

[0044] After adjusting the time slot expansion, an adaptive decompression rhythm with dynamic decompression capability is formed through the integrated linkage of risk-driven signals, breathing-style slip control, and time slot extension. The adaptive decompression rhythm refers to the real-time matching of the oil discharge action's time distribution with the unit's internal pressure changes, allowing the oil discharge rhythm to automatically adjust with pressure variations. Specifically, when the risk level is in the low range, the time slot maintains a standard length, and the oil discharge action is executed at a fixed rhythm. When the risk level fluctuates, the time slot undergoes reciprocating fine adjustments under the breathing-style slip control, keeping the oil discharge action dynamically synchronized with pressure fluctuations. When the risk level rises to the high range, the time slot automatically expands, the oil discharge action is delayed, and its duration is extended, allowing residual pressure to be gradually released within the extended time, ultimately causing the cavity pressure to smoothly return to below the safe threshold. Through the formation of this adaptive pressure reduction rhythm, the unit's oil discharge phase is no longer subject to fixed time logic, but automatically adjusts the rhythm according to the risk status. Thus, even under conditions of drastic pressure changes or sudden load increases, it can still achieve coordinated and unified pressure release and start-up / shutdown rhythm, ensuring the safety, continuity, and stability of the start-up / shutdown process.

[0045] This invention introduces time-axis-based multi-signal superposition and rhythm disturbance recognition during start-up and shutdown processes, transforming the determination of unit start-up and shutdown states from a single logical timing control to continuous dynamic correlation control. By time fusion and disturbance stripping of load signals and self-test beat signals, adaptive recognition of start-up and shutdown rhythms and dynamic reconstruction of safety windows are achieved, enabling the system to maintain temporal continuity and state consistency during start-up and shutdown even under sudden load surges. This approach effectively avoids self-test misjudgments and action freezes caused by transient interference during start-up and shutdown, significantly improving the stability and reliability of the unit start-up and shutdown link.

[0046] This invention introduces a risk-level-driven, breathing-style time-slot rearrangement control, enabling the execution time of the oil draining phase to adaptively adjust with pressure evolution trends. This achieves dynamic coordination between the oil draining action and the pressure release process in start-up and shutdown control. This control method allows the oil draining rhythm to automatically slide or expand according to the risk level, ensuring sufficient release of residual pressure and thus avoiding structural stress concentration caused by pressure accumulation within the chamber. Through this dynamic pressure reduction mechanism, the unit's pressure response during start-up and shutdown is more stable, and overall operational safety and start-up and shutdown efficiency are simultaneously improved.

[0047] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A hierarchical control method for one-button start / stop of generating units, characterized in that, Includes the following steps: During the start-up and shutdown of the unit, load signals and self-test beat signals are continuously collected and superimposed along the time axis to generate a self-test rhythm disturbance trajectory band that reflects the changes in beat compression and stretching. Based on the self-test rhythm disturbance trajectory band, the start and end times of the self-test are compared segment by segment to separate the beat segments affected by the sudden increase in load, reconstruct the opening and closing boundary sequence of the safety window and embed the beat disturbance information. Expand the safety window boundary sequence along the time axis and extend it to the oil discharge start point to form an expanded interval containing self-check end information, and lock the time gap of the key oil discharge stage in the expanded interval. Within the continuous window of the time gap during the critical oil discharge stage, the cavity pressure signal and the oil discharge valve command status are simultaneously acquired, and the two types of signals are superimposed along the time axis to generate a residual pressure evolution indicator curve. Based on the residual pressure evolution indicator curve, three features are extracted: pressure ramp-up slope, peak duration, and self-test recovery delay. These three features are then mapped to the load surge start-stop risk level. Using the load surge start-stop risk level as the control driver, a breathing-type time slot rearrangement control is implemented for the critical oil discharge phase time slots. This causes the oil discharge command to slide back and forth along the time axis, and expands the time slots under high-risk levels, forming an adaptive decompression rhythm with dynamic decompression capability. The steps for implementing the breathing-type time slot rearrangement control using the load surge start-stop risk level as the control driver are as follows: The risk level of sudden load increase in start-stop is used as the time management logic of the start-stop control link to input the control drive signal, forming a risk drive signal sequence of drive oil discharge timing changes; Establish a time response benchmark for the time gap in the critical oil drainage stage, and determine the center benchmark point of the time gap based on the standard execution rhythm of the oil drainage stage. Based on risk-driven signals, breathing-type time-slip control is implemented within the time gap of the critical oil discharge stage, so that the oil discharge command generates a back-and-forth slip on the time axis. Among them: Breathing time slip refers to the start and end boundaries of the time gap moving back and forth along the time axis under the drive of the cyclical change of risk level, so that the triggering time of the oil discharge action is continuously fine-tuned within a small range; When the risk level exceeds the set threshold, the time gap of the critical oil discharge stage will be extended backward. An adaptive decompression rhythm with dynamic decompression capability is formed through the linkage of risk-driven signals, breathing-style slip control, and time slot extension.

2. The unit one-button start / stop hierarchical control method according to claim 1, characterized in that, The steps for generating the self-checking rhythm disturbance trajectory band are as follows: After the unit enters the start-up and shutdown process, load signals and self-test cycle signals are continuously collected and synchronously sampled using the start-up and shutdown control master clock as the time reference. The collected load signal and the self-test beat signal are superimposed along a unified time axis, and the timestamp is used as the corresponding basis to make the load change curve and the beat pulse synchronized in time. After superposition, time expansion is performed in units of self-test stage to make the change trend of load signal form a continuous response distribution with clock compression and clock stretching. After time expansion, a normalized mapping is performed based on the total duration of the start-stop process to generate a self-checking rhythm disturbance trajectory band that is continuously distributed along the time direction, which is used to characterize the dynamic coupling relationship between load changes and beat rhythm.

3. The unit one-button start / stop hierarchical control method according to claim 2, characterized in that, The steps to reconstruct the opening and closing boundary sequence of the safety window are as follows: After obtaining the self-test rhythm disturbance trajectory band, determine the time boundary range of each self-test stage in the entire start-stop process, and take the rising edge and falling edge of the self-test beat signal as the start time and end time of the self-test, respectively. Based on the analysis of the self-check rhythm disturbance trajectory, each self-check stage is compared segment by segment to identify the beat segments in the trajectory affected by the sudden increase in load. Based on the comparison results, the disturbed beat segments are separated from the normal beats, and the start time of the self-test, the point of sudden rise in the load signal, and the end time of the self-test are recorded. Based on the beat separation results, the opening and closing boundary sequence of the safety window is reconstructed, and the beat disturbance information is embedded into the time boundary of the safety window, so that the opening and closing time of the safety window can be dynamically adjusted according to the load.

4. The unit one-button start / stop hierarchical control method according to claim 3, characterized in that, When reconstructing the safety window opening and closing boundary sequence, if the clock disturbance occurs in the early stage of the self-test phase, the safety window opening time is extended forward to compensate for the time deviation caused by clock compression; if the clock disturbance occurs in the later stage of the self-test phase, the safety window closing time is extended backward to match the load recovery process, thereby ensuring that the self-test phase maintains a complete execution cycle after a load surge and achieves continuous connection of start and stop timing.

5. The unit one-button start / stop hierarchical control method according to claim 3, characterized in that, The steps for unfolding the opening and closing boundary sequence of the safety window along the time axis and locking the time gap of the critical oil discharge stage are as follows: The opening and closing boundary sequence of the safety window is expanded along the time axis. Taking the total start and stop time as a reference, each safety window is arranged in sequence and adjacent boundaries are continuously connected by time translation. Starting from the end time of the self-check of each safety window, the time is extended to the start time of the oil discharge action, forming an extended interval covering the two. Divide the extended range into stages from the end of self-inspection to the preparation for oil draining, and determine that the oil draining action should be performed from the end of the load stabilization stage to the beginning of the oil draining preparation stage. Based on the phase division results, the start and end boundaries of the key oil drainage phase time gaps are marked and fixed on the time axis, so as to serve as the execution time window for the oil drainage action during the start-up and shutdown process.

6. The unit one-button start / stop hierarchical control method according to claim 5, characterized in that, The steps for generating the residual pressure evolution indicator curve are as follows: A time duration window is established based on the time gap of the critical oil discharge stage, and time extension is performed at both ends of the time gap to cover the entire process from the end of the self-test to the complete equilibrium of the cavity pressure. Within the time duration window, the cavity pressure signal and the oil discharge valve command status signal are simultaneously acquired, and the timestamps of the two types of signals are recorded with a unified time reference. The chamber pressure signal and the oil drain valve command status are superimposed along the time axis, and the two are arranged correspondingly at the same time point with time as the horizontal axis. Based on the superposition results, a continuous curve of cavity pressure changing with time is plotted, and key moments of valve opening, full opening and closing are marked on the curve to generate a residual pressure evolution indicator curve reflecting the relationship between pressure changes and valve command changes.

7. The unit one-button start / stop hierarchical control method according to claim 6, characterized in that, The chamber pressure signal is acquired simultaneously at the main pressure detection point, oil passage pressure detection point, and exhaust detection point in the oil discharge chamber. The oil discharge valve command status signal records the entire process of valve opening, response, full opening, and closing. During superposition processing, a unified time reference is used for smooth alignment, so that the pressure change curve and the valve action curve form a continuous correspondence on the time axis.

8. The hierarchical control method for one-button start / stop of a generating unit according to claim 6, characterized in that, The steps to map the residual pressure evolution indicator curve to the load surge start-stop risk level are as follows: The residual pressure evolution indicator curve is identified in the time domain and divided into stages, distinguishing the pressure response stage, the rapid pressure decrease stage, the stable pressure release stage, and the pressure decay convergence stage. In each stage, three features were extracted: pressure ramp rate, peak duration, and self-test recovery delay, which were used to characterize the pressure change rate, pressure duration, and system recovery time. Establish the time correspondence between pressure ramp-up slope, peak duration, self-check recovery delay, and oil drain valve command status to form a continuous feature chain describing the entire oil draining process; Based on the combined state of three characteristics—pressure ramp-up slope, peak duration, and self-test recovery delay—the three characteristics are comprehensively mapped to a load surge start-stop risk level, enabling the risk level to have a dynamic response capability to the evolution of residual pressure.

Images