LNG gasification efficiency optimization control method and system

Abstract

The application discloses an LNG gasification efficiency optimization control method and system, relates to the technical field of energy chemical process control and industrial automation, and converges a disturbance source originally acting simultaneously in multiple branches into a single-source disturbance with a main branch as the main source by dynamically specifying a main adjusting branch and a slave adjusting branch in a target branch set and by the main adjusting branch preferentially bearing a first share of a change amount, reduces superimposed disturbance intensity under parallel coupling conditions, makes a propagation chain of total pipe pressure disturbance shorter and controllability stronger, and further binds the intervention time and intervention intensity of the slave adjusting branch to a global pressure state by acquiring real-time pressure of an LNG liquid phase total pipe, taking deviation and total pipe pressure change rate as constraint inputs, performing delay start and control amount change rate limiting follow-up adjustment on the slave adjusting branch, and automatically inhibiting rapid action of the slave adjusting branch when pressure deviation or pressure change rate is abnormal, so that feedback amplification caused by mutual occupation of flow among branches is reduced.

Description

Technical Field

[0001] This invention relates to the fields of energy and chemical process control and industrial automation technology, and in particular to a method and system for optimizing LNG gasification efficiency control. Background Technology

[0002] In large LNG receiving terminals, the export gasification system mostly consists of storage tanks, low-pressure / high-pressure LNG pumps, recondensation and BOG processing units, as well as open-frame gasifiers, submerged combustion gasifiers, or intermediate medium gasifiers. To meet the export capacity and equipment availability requirements under different seasons and gas consumption fluctuations, the export gasification section is often arranged as multiple parallel gasification branches. Each branch obtains LNG from a common liquid phase main pipe and, after gasification, merges into the gas phase main pipe to supply gas to the export pipeline network.

[0003] In existing projects, the automated control of parallel gasification branches often adopts an architecture that combines discrete program control and fixed-value PID regulation. Discrete program control is used for the start-up and shutdown sequence of the branch and its auxiliary equipment, valve interlocking, pressure balancing and flow switching, etc. The fixed-value PID loop is used to maintain the branch outlet pressure, temperature and related process parameters. This architecture can realize branch switching and basic parameter control in steady-state or slow adjustment scenarios.

[0004] When downstream load commands require rapid changes in the total gas output within a short period, parallel branches are prone to dynamic coupling amplification under shared main pipe conditions. Due to inherent differences in heat exchange capacity, valve characteristics, pump performance, and pipeline resistance among parallel branches, and the formation of a coupled fluid network between the liquid and gas phase main pipes, when multiple operating branches simultaneously receive similar adjustment commands, branches with faster responses will preferentially change valve openings or equivalent resistances, leading to transient pressure drop and flow redistribution in the main pipe. The main pipe disturbance is further fed back to other branches, causing their control loops to overshoot in the setpoint tracking process, thus creating low-frequency pressure and flow oscillations in the main pipe-branch network. Such oscillations cause continuous fluctuations in the gas output and key parameters, and keep pumps, valves, and heat exchange equipment in alternating operating conditions for extended periods, resulting in frequent valve actuation, pump drift, and equipment fatigue.

[0005] To suppress the aforementioned oscillations, manual intervention is often used in field operations. This includes staggering the adjustment sequence of each branch, limiting the adjustment slope, using trial operations with large amplitude and small step size, or keeping more branches in operation during peak shaving to reduce the adjustment amplitude of a single branch. These methods rely on experience and are difficult to maintain consistent control quality across different personnel and operating conditions. At the same time, it is difficult to achieve coordinated optimization based on energy consumption and efficiency indicators in the selection of parallel system combinations, load allocation, and start-up and shutdown strategies, which can easily lead to redundant operation and high energy consumption. In scenarios where rapid peak shaving and high-reliability gas supply coexist, there is still room for improvement in coordinating and controlling parallel gasification branches while taking into account gasification efficiency. Summary of the Invention

[0006] This application provides a method and system for optimizing LNG gasification efficiency control, which solves the problems of main pipe pressure resonance, flow fluctuation and equipment fatigue caused by synchronous adjustment of parallel gasification branches.

[0007] In a first aspect, embodiments of the present invention provide an LNG gasification efficiency optimization control method, applied to an export gasification system. The export gasification system includes an LNG liquid phase main pipe and multiple gasification branches connected in parallel with the LNG liquid phase main pipe. Each gasification branch is equipped with an adjustable actuator for regulating the LNG flow rate of the branch. The method includes: Step S1: Receive the total external gas volume command, determine the target branch set to be regulated based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determine the main regulating branch and at least one secondary regulating branch from the target branch set. Step S2: Control the main adjustment branch to change the adjustable actuator to take on the first share of the change before the secondary adjustment branch; Step S3: Obtain the real-time pressure of the LNG liquid phase main pipe, and based on the deviation between the real-time pressure and the main pipe pressure set value, perform delayed start and change rate limit adjustment on the remaining share borne by the regulating branch, so that the real-time pressure is kept within the preset pressure control band.

[0008] In some embodiments, the efficiency index is the energy consumption index per unit of external gas output, which is calculated by the measured power value of the target branch and the measured gas output of the target branch within a preset statistical window. The determination of the target branch set includes: under the condition of satisfying the minimum number of operating branches, prioritizing the selection of gasification branches with lower energy consumption per unit of external gas transmission volume into the target branch set.

[0009] In some embodiments, determining the main regulating branch includes: For each gasification branch within the target branch set, acquire the equipment runtime, health status indicators, and historical regulation oscillation indicators; According to the preset priority, the gasification branch that meets the operating conditions and has a small historical regulation oscillation index is selected as the main regulation branch, and the remaining target branches are determined as the secondary regulation branches.

[0010] In some embodiments, the delayed start-up and rate of change limit follow-up adjustment includes: After the main regulating branch starts to perform load regulation, the load regulation of the slave regulating branch is started after a delay period. During the adjustment process from the regulating branch, the absolute value of the deviation and the rate of change of the real-time pressure are used as limiting inputs to dynamically limit the rate of change of the control quantity of the adjustable actuator. The delay time is adaptively generated by the controller based on real-time operating conditions, which include at least the LNG liquid phase main pipe pressure change rate, the control quantity change rate of the adjustable actuator of the main regulating branch, the inertial time constant of the slave regulating branch, and the number of currently operating branches. Furthermore, hysteresis and minimum dwell time are introduced in the determination of slave regulating branch start-up and exit-up, so that the slave regulating branch maintains its existing state before the dwell time is reached, and is not allowed to re-enter the following mode within the cooling time after exiting the following mode, thereby reducing the frequent start-up and stop-up of slave regulating branch.

[0011] In some embodiments, the dynamic limitations include: When the absolute value of the deviation is greater than the first threshold, the rate of change of the control quantity of the adjustable actuator is limited to the first upper limit; When the absolute value of the deviation is not greater than the first threshold and the rate of change of the real-time pressure is greater than the second threshold, the rate of change of the control quantity of the adjustable actuator is limited to the second upper limit, where the first upper limit is less than the second upper limit.

[0012] In some embodiments, the pressure oscillation characteristics of the LNG liquid phase manifold are continuously monitored; If the pressure oscillation characteristic meets the preset resonance judgment condition, a cooperative vibration suppression command is output; in response to the cooperative vibration suppression command, all load regulation of the control branch is frozen and its adjustable actuator control amount is kept unchanged, while the main control branch is controlled to perform single-branch regulation at a vibration suppression change rate lower than the preset change rate upper limit, until the pressure oscillation characteristic exits the preset resonance judgment condition and the freeze is released. The preset resonance determination conditions are composed of a pressure oscillation amplitude threshold, a period stability threshold, and a duration period number threshold. The resonance determination is achieved by using time-domain autocorrelation or peak-valley detection to obtain the oscillation period and amplitude, or by using frequency-domain bandpass filtering combined with an energy threshold to obtain the resonance intensity and thereby determine the period stability and duration period number. Freezing and unfreezing are performed by a state machine with hysteresis. When the entry criterion is met, the slave adjustment branch is frozen and adjusted by the main adjustment branch at the vibration suppression rate. When the exit criterion is met, the freeze is lifted and the slave adjustment branch is gradually restored to follow the oscillation using a soft-start recovery method to cover repeated entry and exit scenarios and noise misjudgment scenarios.

[0013] Secondly, embodiments of the present invention provide an LNG gasification efficiency optimization and control system, which is applied to an external gasification system, the external gasification system including an LNG liquid phase main pipe and multiple gasification branches connected in parallel with the LNG liquid phase main pipe; The control system includes a controller configured to perform: receiving a total external gas volume command, determining a target branch set based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determining a main regulating branch and at least one secondary regulating branch; The main control branch performs load regulation before the secondary control branch; the real-time pressure of the LNG liquid phase main pipe is obtained, and based on the deviation between the real-time pressure and the main pipe pressure set value, the secondary control branch performs delayed start and rate of change limit follow-up regulation to keep the real-time pressure within the preset pressure control range.

[0014] In some embodiments, the system further includes an efficiency calculation unit, which is configured to acquire the electrical power measurement value and the external gas output measurement value of each gasification branch, calculate the energy consumption index per unit external gas output within a preset statistical window, and output the index to the controller.

[0015] In some embodiments, a resonance monitoring unit is further included, which is configured to monitor the pressure oscillation characteristics of the LNG liquid phase manifold and determine whether a preset resonance determination condition is met; in response to meeting the preset resonance determination condition, the controller freezes the secondary regulation branch and controls the main regulation branch to perform single-branch regulation with the vibration suppression rate.

[0016] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the LNG gasification efficiency optimization control method as described in the first aspect of the present invention.

[0017] Through the above technical solution, the present invention can achieve at least the following beneficial effects: 1) To address the issues of flow grabbing, pressure disturbance, reverse over-adjustment, and low-frequency oscillation that occur in parallel branches during rapid peak shaving, the main regulating branch and the secondary regulating branch are dynamically designated within the target branch set. The main regulating branch is given priority to bear the first share of the change, thereby converging the disturbance source that originally operated by multiple branches simultaneously into a single-source disturbance dominated by the main branch. This reduces the intensity of superimposed disturbances under parallel coupling conditions, making the propagation chain of the main pipe pressure disturbance shorter and more controllable.

[0018] 2) To address the issue of oscillation amplification caused by the intervention of the control branch during pressure disturbances in the main pipe, the real-time pressure of the LNG liquid phase main pipe is obtained and the deviation and the main pipe pressure change rate are used as constraint inputs. The control branch is then subjected to delayed start-up and control quantity change rate limit adjustment. This binds the intervention timing and intensity of the control branch to the global pressure state. When the pressure deviation or pressure change rate is abnormal, the rapid action of the control branch is automatically suppressed, thereby reducing the feedback amplification caused by the mutual flow competition between branches.

[0019] 3) To address the mismatch between fixed delay or fixed amplitude limiting strategies caused by changes in system inertia and coupling degree under different operating conditions, the delay time is correlated with the rate of change of the main pipe pressure, the rate of change of the control quantity of the main regulating branch, the inertial time constant of the slave regulating branch, and the number of operating branches. This makes the delay and amplitude limiting constraints change with the operating conditions. Furthermore, hysteresis, minimum dwell time, and cooling time are introduced to decouple the entry / exit judgment of the slave regulating branch from small fluctuations near the boundary, reduce valve reciprocation and load jitter caused by frequent start-stop following, and improve the continuity of the regulation process.

[0020] 4) To address the problem of persistent periodic pressure fluctuations being difficult to suppress in a timely manner and relying heavily on manual intervention, this paper continuously monitors the characteristics of pressure oscillations and uses amplitude, periodic stability, and the number of periods as resonance criteria. When persistent periodic fluctuations are detected, a coordinated vibration suppression command is generated, freezing the control branch and keeping its control quantity unchanged. At the same time, the main control branch performs single-branch regulation with the vibration suppression change rate, so that the system enters a single-branch, limited-rate vibration suppression mode during the resonance phase, reducing the resonance maintenance conditions caused by the interaction of multiple branches.

[0021] 5) To address the issue of secondary excitation during the resonance release process, a state machine with hysteresis and a soft start for recovery are introduced during the freeze release process. This allows the control branch after the freeze to resume following the control in a phased manner with a limited rate of change. During the recovery phase, the entry criterion is monitored, forming a closed-loop switching logic for freeze-recovery. This reduces the risk of secondary pressure disturbance caused by the simultaneous recovery of multiple branches at the moment of release.

[0022] 6) To address the challenge of balancing the stability of gas output volume with the long-term reliability of equipment, the action rate of the regulating branch is linked to the pressure deviation and change rate of the main pipe. The regulating branch is frozen during the resonance phase, which suppresses the high-frequency reciprocating motion of pumps, valves, and other actuators, reduces the frequency of unexpected alternating conditions, and alleviates the accumulation of alternating stress on the equipment under oscillating conditions. This improves the stable operation characteristics of the gas output process and the manageability of the equipment's operating status.

[0023] 7) To address the problem that it is difficult to balance energy consumption efficiency in the selection of parallel branch combinations and load allocation, an energy consumption index per unit of external gas output is introduced and used as input for the selection of target branch sets and share allocation. This provides a quantifiable efficiency basis for the selection and allocation process of branches participating in regulation. When the measurement point is abnormal, the efficiency index is marked as unusable and the set is reconstructed, forming an efficiency participation mechanism consistent with the validity of the field signal. This allows the efficiency target and stability constraints to work together within the same control framework.

[0024] 8) To address the challenges of reviewing operational processes and tracing control strategies, operational data structures such as target branch sets, share allocation tables, oscillation characteristic records, resonance event records, and state machine records are set up. This enables a timestamp-linked index to be generated throughout the entire process of a single adjustment task, from instruction changes, branch selection, amplitude limiting triggering to freezing / unfreezing. This supports subsequent tracing and attribution of adjustment behavior, abnormal operating conditions, and parameter tuning. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation on the scope of this application.

[0026] Figure 1 This is a flowchart of the LNG gasification efficiency optimization control method in the embodiment.

[0027] Figure 2 This is a framework diagram of the LNG gasification efficiency optimization control system in the embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] All terms used in this application (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0030] Furthermore, for ease of understanding, the relevant terms and concepts involved in the embodiments of this application will be introduced below: The external vaporization system is a process system in the LNG receiving terminal that pumps and vaporizes LNG before supplying it to the downstream external pipeline network. The LNG liquid phase main is a common pipeline that distributes LNG liquid to multiple vaporization branches, and its pressure is the real-time pressure of the LNG liquid phase main. Each vaporization branch is a single process channel connected in parallel to the LNG liquid phase main, and includes a supply pipe section, a heat exchange vaporization section, and a branch outlet pipe section connected to the external transmission side. The adjustable actuator is an actuating component that regulates the LNG flow rate or equivalent resistance of the vaporization branch. The control quantity is the target value output by the controller to the adjustable actuator, and the control quantity can correspond to valve opening, pump frequency, or a combination thereof. The controller is a control device used to receive measurement point signals and output the control quantity to drive the adjustable actuator.

[0031] The total external gas output command is the target total external gas output signal issued by the external dispatching or station control system. The change is the difference between the total external gas output command in adjacent control cycles or the target increment under a preset ramp function. The real-time pressure is the current pressure value collected from the LNG liquid phase main pipeline pressure measurement point. The main pipeline pressure setpoint is the target pressure value used to constrain the LNG liquid phase main pipeline pressure. The deviation is the difference between the real-time pressure and the main pipeline pressure setpoint. The pressure control band is the allowable pressure fluctuation range set around the main pipeline pressure setpoint.

[0032] The load margin is the available adjustment amount that the gasification branch can adjust upwards or downwards without triggering capacity boundary constraints. The first share is the target increment of the change that is borne by the main regulating branch, and the remaining share is the target increment of the change that is borne by the secondary regulating branch. The unfulfilled share is the temporary amount of the remaining share that could not be allocated under the limiting, stopping, or freezing state, and it will continue to be released after the real-time pressure meets the conditions for entering the pressure control zone.

[0033] The delay time is the time interval from when the main control branch begins load regulation to when the secondary control branch is allowed to initiate follow-up regulation. Hysteresis is a paired threshold set for entering and exiting the follow-up state from the secondary control branch. Minimum dwell time is the shortest duration for which the secondary control branch maintains a certain state after entering it. Cooling time is the time interval during which the secondary control branch is not allowed to re-enter the follow-up state after exiting it. Stability threshold is the threshold value used to determine whether the rate of change of the control quantity in the main control branch or the rate of change of the main pipe pressure has entered the stable range.

[0034] The statistical window is the time window used to calculate the energy consumption index per unit of external gas transmission. The monitoring window is the time window used to extract pressure oscillation characteristics. The main pipe pressure change rate is the rate of change obtained by the difference between the real-time LNG liquid phase main pipe pressure and adjacent sampling times. The control quantity change rate is the rate of change obtained by the difference between the control quantity output by the controller and adjacent sampling times. The periodic stability index is a quantitative indicator used to characterize the stability of the dominant pressure oscillation period within the monitoring window.

[0035] The anti-integral saturation processing involves freezing or recalculating the integral state when the control quantity is subject to amplitude or limit constraints, ensuring a continuous recovery process of the control quantity after the constraints are lifted. The damping change rate is the upper limit of the control quantity change rate allowed in the main control branch during the freezing period from the control branch.

[0036] Availability constraints are a set of constraints used to determine whether a gasification branch is allowed to participate in regulation. This set includes at least the following: interlock status is valid, key measuring point signals are valid, adjustable actuators are in a controllable state, and the branch is in an adjustable operating state. Capacity boundary constraints are a set of constraints used to limit the adjustable range of a gasification branch. This set includes at least the branch's minimum stable operating load, maximum allowable load, valve opening limits, pump operating range, and heat exchange capacity limits. Minimum number of operating branches constraint is the lower limit for the number of parallel operating branches used to maintain output continuity and redundancy.

[0037] The target branch set is the set of gasification branches included in load sharing and coordinated control in the current regulation task. The primary regulating branch is the gasification branch within the target branch set designated as having priority to bear the first share of changes. The secondary regulating branch is the gasification branch within the target branch set excluding the primary regulating branch. Efficiency indicators are a set of indicators used to characterize the energy consumption level or heat source consumption level corresponding to the unit export capacity of the gasification branch, with the energy consumption per unit export volume being one type of efficiency indicator.

[0038] Pressure oscillation characteristics are a set of features characterizing the periodic fluctuations in LNG liquid phase manifold pressure in the time or frequency domain. Resonance criteria are a set of combined thresholds and persistence constraints used to determine whether pressure oscillations have reached a sustained periodic fluctuation state. Coordinated oscillation suppression commands are a set of control commands generated after the resonance criteria are met. Freezing is a control state that maintains the control quantity from the regulating branch without changing with the regulating command during the freezing period; unfreezing is the transition from the frozen state to the state following the regulating command.

[0039] Example 1: like Figure 1As shown, this embodiment proposes an LNG gasification efficiency optimization control method, applied to an external gasification system. The external gasification system includes an LNG liquid phase main pipe and multiple gasification branches connected in parallel with the LNG liquid phase main pipe. Each gasification branch is equipped with an adjustable actuator for regulating the LNG flow rate of the branch. The method includes: Step S1: Receive the total external gas volume command, determine the target branch set to be regulated based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determine the main regulating branch and at least one secondary regulating branch from the target branch set. In one embodiment, the determination of the target branch set is based on availability constraints, capacity boundary constraints, and minimum number of operating branches constraints. Availability constraints include: interlocks in the gasification branch are not triggered, key measurement point signals are valid, adjustable actuators are controllable, and the branch is in an adjustable operating state. Capacity boundary constraints include: the minimum stable operating load and maximum allowable load of the gasification branch under current operating conditions, with boundaries jointly defined by the branch flow limit, pump operating range, valve opening limit, and heat exchange capacity limit. The minimum number of operating branches constraint is the minimum number of parallel branches required to maintain output continuity and redundancy.

[0040] Under the premise of satisfying the above constraints, the controller determines the number of branches that need to participate in this adjustment based on the direction and magnitude of the change. When the change is a positive increment, the target branch set includes gasification branches that can increase the load and have a load margin; when the change is a negative increment, the target branch set includes gasification branches that can decrease the load and do not trigger the lower limit constraint. During the formation of the target branch set, the availability flag, load margin, and boundary triggering reason of each gasification branch are recorded for subsequent share allocation and reallocation.

[0041] Step S2: The main control branch changes the adjustable actuator to take the first share of the change before the secondary control branch. In one embodiment, the first share is the target increment of the change borne by the primary regulating branch, and the remaining share is the target increment of the change borne by the secondary regulating branch. The allocation of the first share and the remaining share is determined based on both load margin and efficiency index. Load margin is the available regulation amount that the gasification branch can adjust upwards or downwards without triggering boundary constraints. The controller generates a share allocation table for each gasification branch within the target branch set. The field set of the share allocation table includes: branch identifier, load margin, efficiency index, weight, first share, remaining share, boundary constraint flag, and number of reallocations.

[0042] For example, the weights are generated by the efficiency index and the load margin and written into the share allocation table. During the generation process, the efficiency index and the load margin are normalized separately and combined at the same timestamp, so that the weights, the first share, and the remaining shares are aligned with the same key. The boundary constraint flag is used to identify the cause type of valve opening limit, pump operating range trigger, or heat exchange capacity limit trigger. The cause type and trigger time are written into the share allocation table. The redistribution number is incremented each time and written synchronously into the operation record table. When the redistribution number is updated, the incomplete share field and the trigger cause field are written at the same time, so that the process of transferring the incomplete first share to the remaining share and the temporary storage of the uncompleted share is traceable in the record layer.

[0043] When the main regulating branch triggers a control limit or insufficient load margin during regulation, the controller transfers the unfinished first share to the remaining share and performs redistribution. During redistribution, the secondary regulating branch with better efficiency and no boundary constraints is given priority to undertake the transferred share. When the secondary regulating branch triggers a limit during follow-up regulation, causing the remaining share to be unable to be completed within a preset time, the controller temporarily stores the unfinished portion as a pending share and releases it after the real-time pressure returns to the pressure control band.

[0044] Step S3: Obtain the real-time pressure of the LNG liquid phase main pipe, and based on the deviation between the real-time pressure and the main pipe pressure set value, perform delayed start and change rate limit adjustment on the remaining share borne by the regulating branch to keep the real-time pressure within the preset pressure control band. In this embodiment, the efficiency index is the energy consumption index per unit of external gas output, which is calculated by the measured power value of the target branch and the measured gas output value of the target branch within a preset statistical window. In one embodiment, the electrical power measurement value is the sum of power collection values ​​from the electrical equipment related to the gasification branch. This equipment includes at least one of the following: LNG pump drive equipment directly related to load changes in the gasification branch, and auxiliary power equipment related to the branch's heat exchange process. The outgoing gas volume measurement value is the measurement value from the outgoing flow meter of the gasification branch or the equivalent outgoing gas volume of the branch obtained by fusing multiple measurement points. A preset statistical window is a time window used to suppress transient noise; within this window, consistent alignment, noise reduction, and validity checks are performed on the electrical power measurement value and the outgoing gas volume measurement value.

[0045] In one implementation, alignment uses timestamps as alignment keys. When the sampling frequencies of power and gas volume signals are inconsistent, the low-frequency signal is resampled to the time grid corresponding to the sampling period. During the resampling process, the original timestamp and resampling flag fields are retained for traceability. Denoising is performed on power and gas volume simultaneously using the same caliber. During the denoising process, abrupt changes are marked as anomaly removal flags and anomaly type fields are recorded. The anomaly type fields include at least missing measurements, jumps, and saturation. When the gas volume measurement values ​​inside and outside the window are continuously missing or close to zero, making the energy consumption index uncalculate, the calculation results of that window are set to invalid and the most recent valid result is retained as a historical value for display only. At the same time, the efficiency index is marked as unsuitable for sorting.

[0046] The calculation criteria for the unit external gas transmission volume energy consumption index include the following set of fields: statistical window identifier, power summary caliber identifier, gas volume measurement point identifier, sampling period, percentage of valid data, anomaly rejection flag, calculation result, and timestamp. The anomaly rejection flag is used to identify situations such as measurement point failure, flow meter jump, or missing power signal within the statistical window. When the anomaly rejection flag is triggered, the controller marks the efficiency index of that gasification branch as unsuitable for sorting and re-forms the target branch set based on availability constraints.

[0047] Determining the target branch set includes: under the condition of satisfying the minimum number of operating branches, prioritizing the selection of gasification branches with lower energy consumption per unit of external gas transmission volume into the target branch set; In this embodiment, determining the main regulating branch includes: For each gasification branch within the target branch set, acquire the equipment runtime, health status indicators, and historical regulation oscillation indicators; According to the preset priority, the gasification branch that meets the operating conditions and has a small historical regulation oscillation index is selected as the main regulation branch, and the remaining target branches are determined as the secondary regulation branches. In one embodiment, the health status identifier is a discrete identifier of the operational health level of the gasification branch, generated by combining the status variables of pumps, valves, and key measuring points. The field set of the health status identifier includes: equipment alarm level, interlock warning flag, vibration and temperature over-limit flag, valve jamming / hysteresis flag, adjustable actuator saturation count statistics, measuring point validity flag, and the timestamp of the most recent maintenance record.

[0048] The historical regulation oscillation index is a quantification of the degree to which the gasification branch introduces or amplifies the main pipe pressure fluctuation during historical regulation. The index consists of the following set of fields: statistical period identifier, number of regulation participations, peak-to-peak pressure deviation statistics, peak pressure change rate statistics, number of control reversals statistics, peak control change rate statistics, and decay time statistics. These fields are written to the operation record after each regulation task and used as sorting input when selecting a new main regulation branch.

[0049] The preset priority is used to specify the selection order of the main control branch among multiple target branches. The order is based on the premise that the health status indicator meets the operating conditions. First, the historical regulation oscillation index is compared, and then the equipment running time is compared. When the ranking results are the same or within the preset difference range, the controller switches the main control branch among the candidate branches according to the rotation rule to balance the distribution of equipment running time.

[0050] In this embodiment, the following adjustment of delayed start and rate of change limit includes: After the main regulating branch starts to perform load regulation, the load regulation of the slave regulating branch is started after a delay period. During the adjustment process from the regulating branch, the absolute value of the deviation and the rate of change of the real-time pressure are used as limiting inputs to dynamically limit the rate of change of the control quantity of the adjustable actuator. In one embodiment, a preset delay time is used to activate the slave regulation branch only after the main regulation branch has generated an initial response to the change. The preset delay time is consistent with or an integer multiple of the control cycle. The rate of change of real-time pressure is obtained by the real-time pressure difference between adjacent control cycles and then denoised. The rate of change limit takes the absolute value of the deviation and the rate of change of real-time pressure as inputs, and outputs the upper limit of the rate of change of the control quantity of the slave regulation branch.

[0051] The controller uses the remaining share as the target input for the follow-up regulation of the control branch. After ramping the remaining share, it generates the target increment for the control branch and updates the control quantity under the rate of change limit constraint. The update process of the control quantity includes anti-integral saturation processing: when the control quantity cannot continue to change due to the limit or stop, the integral state used to generate the control quantity is frozen or recalculated to ensure that the recovery process of the control quantity after the limit is lifted remains continuous.

[0052] The start conditions for the control branch include: the delay expires, the real-time pressure is within the pressure control band or the absolute value of the deviation is lower than the preset threshold, and the rate of change of the control quantity of the main control branch is lower than the preset stability threshold; the pause conditions for the control branch include: the absolute value of the deviation exceeds the first threshold, the rate of change of the real-time pressure exceeds the second threshold, or the resonance judgment condition is met, resulting in a freeze trigger.

[0053] The delay time is adaptively generated by the controller based on real-time operating conditions, which include at least the LNG liquid phase main pipe pressure change rate, the control quantity change rate of the adjustable actuator of the main regulating branch, the inertial time constant of the slave regulating branch, and the number of currently running branches. Furthermore, hysteresis and minimum dwell time are introduced in the determination of slave regulating branch start-up and exit-up, so that the slave regulating branch maintains its existing state before the dwell time is reached, and is not allowed to re-enter the follow-up during the cooling time after exiting, thereby reducing the frequent start-up and stop-up of slave regulating branch. Preset delay time: The delay time is adaptively generated by real-time operating conditions (the rate of change of the main pipe pressure, the rate of change of the main branch control quantity, the branch inertia time constant, and the number of currently running branches), and hysteresis and minimum dwell time are introduced to avoid frequent start-stop following from the branches.

[0054] In one implementation of delayed start, the preset delay time is adaptively generated by the controller from the trigger moment when the main regulating branch begins to execute load regulation, combined with real-time operating conditions, and acts together with the subsequent rate of change limit to follow the regulation of the slave regulating branch. This adaptive generation process includes a comprehensive evaluation of the pressure change rate, the control quantity change rate of the adjustable actuator of the main regulating branch, the inertial time constant of the slave regulating branch, and the number of currently operating branches.

[0055] After the main control branch begins regulation, the controller calculates the rate of change of the real-time pressure in the LNG liquid phase main pipe and the rate of change of the control quantity in the main control branch within the sampling period, for example, using a discrete differential form: , in, Indicates the first Rate of change of main pipe pressure at each sampling time; Indicates the first Real-time pressure of the LNG liquid phase manifold at each sampling time; Indicates the sampling period; Indicates the first The rate of change of the control quantity of the adjustable actuator in the main control branch at each sampling time; Indicates the first The control quantity of the adjustable actuator in the main control branch at each sampling time; Indicates the sampling sequence number.

[0056] After obtaining the aforementioned rate of change, the preset delay time is set to an adaptive delay time that varies with the operating conditions and is limited to an allowable range. This is used to determine the earliest time to initiate follow-up regulation from the regulating branch, for example: , in, Indicates the first The adaptive delay time is calculated at each sampling moment; Indicates the lower limit of the delay time; Indicates the maximum delay time; Indicates the reference delay time; This represents the weighting coefficient for the rate of change of pressure. This represents the normalized baseline for the rate of change of pressure; This represents the absolute value of the rate of change of the main pipe pressure; This represents the weighting coefficient of the rate of change of the control quantity in the main regulating branch; This represents the normalized baseline for the rate of change of the control quantity; This represents the absolute value of the rate of change of the control quantity in the main regulating branch; This represents the weighting coefficient of the inertia term; This represents the representative value of the inertial time constant of the regulating branch; This represents the weighting coefficient for the number of running branches; This indicates the number of gasification branches currently in operation.

[0057] in, The maximum value of the inertial time constant of each slave regulating branch within the target branch set, or the average value obtained by weighting each slave regulating branch according to its rated capacity, can be used to reflect the degree of lag in the response of the slave regulating branch to changes in the control quantity. Used to characterize the degree of coupling introduced by changes in the number of parallel branches. When there are many operating branches, the delay can be appropriately increased to reduce the superimposed disturbance of the main pipe pressure caused by the simultaneous operation of multiple branches.

[0058] In addition to the adaptive delay time, hysteresis and minimum dwell time are introduced to suppress frequent start-stop following from the regulation branch: , in, Indicates the first The deviation of the main pipe pressure at each sampling time; Indicates the first Real-time pressure of the LNG liquid phase manifold at each sampling time; This indicates the main pipe pressure setting value.

[0059] The hysteresis logic is implemented using a pair of pressure deviation thresholds. The follow-up start threshold of the control branch is set to a larger deviation threshold, and the exit threshold of the control branch is set to a smaller deviation threshold. When the pressure deviation fluctuates between the two thresholds, the existing state is maintained, making the start-stop decision insensitive to small fluctuations.

[0060] The minimum dwell time logic starts timing after entering the follow adjustment from the adjustment branch, and maintains the follow state without exiting before the dwell time is reached; after exiting the follow from the adjustment branch, it enters the cooling timing, and is not allowed to re-enter the follow state before the cooling time is reached, thus limiting the start and stop frequency to a controllable range.

[0061] The determination of the hysteresis and dwell time can be carried out in parallel with the freezing logic. When the resonance determination condition triggers freezing, the freezing logic takes priority and directly freezes the control quantity from the regulating branch and is then regulated by the main regulating branch with the single branch regulation of the vibration suppression rate.

[0062] In this implementation, the delay setting is expanded from a fixed constant to an adaptive quantity that varies with operating conditions. When the main branch's operation causes large fluctuations in the main pipe pressure, rapid changes in the main branch's valve position, significant lag in the response of the slave branch, or a large number of parallel-operating branches, the delay is increased accordingly. This allows the slave branch to intervene and follow up only after the main branch has completed its major share adjustment and exhibited a more stable pressure change trend, thereby reducing the pressure superposition disturbance caused by the simultaneous operation of multiple branches. Under operating conditions with small pressure deviations and slow changes, the delay can be reduced, allowing the slave branch to take on the remaining share earlier to improve the tracking of the total command. The hysteresis and minimum dwell time decouple the slave branch start / stop determination from small pressure fluctuations, reducing valve position reciprocation and load jitter caused by frequent start / stop operations. This makes the following adjustment closer to a continuous process and complements the resonance freezing mechanism, maintaining the consistency and feasibility of the branch coordination strategy under different oscillation intensities.

[0063] In this embodiment, the dynamic limitations include: When the absolute value of the deviation is greater than the first threshold, the rate of change of the control quantity of the adjustable actuator is limited to the first upper limit. When the absolute value of the deviation is not greater than the first threshold and the rate of change of the real-time pressure is greater than the second threshold, the rate of change of the control quantity of the adjustable actuator is limited to the second upper limit, and the first upper limit is less than the second upper limit. In this embodiment, the pressure oscillation characteristics of the LNG liquid phase manifold are continuously monitored; If the pressure oscillation characteristics meet the preset resonance judgment conditions, a coordinated vibration suppression command is output. In response to the coordinated vibration suppression command, all load regulation of the control branch is frozen and its adjustable actuator control amount is kept unchanged. At the same time, the main control branch is controlled to perform single-branch regulation at a vibration suppression change rate lower than the preset change rate upper limit until the pressure oscillation characteristics exit the preset resonance judgment conditions and then the freeze is lifted. In one embodiment, the pressure oscillation characteristics are obtained by windowing a real-time pressure sequence. Windowing extracts periodic fluctuation characteristics within a preset monitoring window. The set of fields for the pressure oscillation characteristics includes: monitoring window identifier, pressure fluctuation peak-to-peak value, dominant period estimate, period stability index, bandpass energy index, rate of change of amplitude between adjacent periods, and percentage of valid data. The period stability index characterizes the stability of the dominant period within the monitoring window, and the bandpass energy index characterizes the pressure fluctuation energy level within a preset frequency band.

[0064] For example, before windowing, the real-time pressure sequence is detrended and smoothed. Detrending can be done using a moving average, and the time constant can be 5~20 s. The window is updated using a sliding method, and the window can be updated with 50% overlap to improve continuity. The amplitude correlation field is obtained from the peak-valley difference within the window, and the period correlation field is obtained from the period sequence statistics obtained by peak-valley detection or autocorrelation. The period stability index is characterized by the dispersion of the period sequence. When the number of identifiable complete periods within the monitoring window is less than 3, the period stability index is marked as unavailable and the entry criterion is not triggered to avoid misjudgment caused by short windows or noise.

[0065] The resonance determination condition is a combination criterion, which includes at least: the peak-to-peak value of the pressure fluctuation exceeds the amplitude threshold, the periodic stability index meets the stability threshold, and the above states continuously satisfy the persistence constraint; the persistence constraint is limited by the duration or the number of continuous periods. The field set of the coordinated vibration suppression command includes: trigger timestamp, trigger reason identifier, list of frozen objects, vibration suppression change rate, minimum freeze dwell time, release threshold, and release hysteresis flag.

[0066] During the freeze period, the control quantity remains unchanged from before the freeze in the regulating branch; the main regulating branch performs single-branch regulation of the change quantity using the vibration suppression rate, and marks the unfinished remaining portion as the unfinished portion. Unfreezing is triggered after the resonance judgment condition is exited and the minimum freeze dwell time is met; after unfreezing, the regulating branch resumes follow regulation by soft-start ramp, so that the control quantity change rate gradually transitions from the lower upper limit to the normal upper limit.

[0067] The preset resonance judgment conditions are composed of pressure oscillation amplitude threshold, periodic stability threshold, and duration period number threshold. Among them, the resonance judgment is to obtain the oscillation period and amplitude by time domain autocorrelation or peak-valley detection, or to obtain the resonance intensity by frequency domain bandpass filtering and energy threshold, and to form the periodic stability and duration period number judgment based on this. Freezing and unfreezing are executed by a state machine with hysteresis. When the entry criterion is met, the slave regulation branch is frozen and the main regulation branch is adjusted by a single branch with the vibration suppression change rate. When the exit criterion is met, the freeze is lifted and the slave regulation branch is gradually restored to follow by a soft start recovery method to cover repeated entry and exit and noise misjudgment scenarios. In one implementation of the preset resonance judgment condition, the resonance judgment condition is composed of a combination of amplitude threshold, periodic stability and duration period number, which is used to distinguish the random fluctuation of the main pipe pressure from the oscillation with a stable period. The resonance monitoring unit provides two types of implementation paths, and the controller uses the same set of frozen state machine with hysteresis to perform freezing, unfreezing and recovery soft start on the judgment results output by the two types of paths.

[0068] I. The Indicator Composition of Combined Criteria The resonance monitoring unit extracts the pressure oscillation amplitude index within a sliding statistical window. The amplitude index is obtained based on the peak-to-valley difference within the window: , in, Indicates the first The pressure oscillation amplitude index at each sampling time; Indicates the first Real-time pressure of the LNG liquid phase manifold at each sampling time Same as the definition above; Indicates the first Each sampling time is the set of sampling sequence numbers within the statistical window at the end; Indicates the sampling sequence number within the window.

[0069] Periodic stability is used to describe whether the oscillation period is stable. It is obtained by statistically analyzing the periodic sequence obtained from peak-valley detection or autocorrelation within a window, for example, using the coefficient of variation. , in, Indicates the first Periodic stability index at each sampling time; This represents the standard deviation of a periodic series within a statistical window. It represents the mean of the periodic sequence within the statistical window; the periodic sequence is obtained by the time interval between adjacent feature points of the same type within the window, and the feature points are peak points or valley points.

[0070] The duration of cycles is used to suppress noise misjudgment. It can be set to a threshold of consecutive cycles that meet both the amplitude threshold and the period stability threshold before freezing is initiated. This duration of cycles is updated by the resonance monitoring unit in each detection cycle and output to the controller.

[0071] II. Two types of implementation paths 1) Time-domain implementation path (autocorrelation / peak-valley detection) Resonance monitoring unit After detrending and low-pass smoothing, peak and valley detection is used to obtain peak and valley sequences, and the amplitude index is output. Periodic sequences and calculation of periodic stability. Alternatively, the autocorrelation method can be used to find the main peak position within the candidate delay range as the main period estimate, and the main period fluctuation can be statistically analyzed within a continuous window to obtain the result. When the peak-valley detection path is sensitive to spike noise, minimum peak spacing and minimum peak height constraints can be added to the feature point determination process to ensure that the periodic sequence is generated only from feature points that satisfy the constraints.

[0072] 2) Frequency domain implementation path (bandpass and energy threshold) The resonance monitoring unit will The input is a bandpass filter, and the bandpass frequency band covers the frequency range where the main pipe pressure in the external gasification system is prone to resonance, thus obtaining a bandpass signal. The bandpass signal energy is calculated within the statistical window as an indicator of resonance intensity. , in, Indicates the first Bandpass energy index at each sampling time; Indicates the first The main pipe pressure bandpass signal at each sampling moment; This indicates the number of sample points in the statistical window; and As defined above. In the frequency domain, periodic stability can be calculated by constructing a periodic sequence from the zero-crossing intervals or envelope peak intervals of the bandpass signals. This ensures that the amplitude / energy-period stability-duration number of periods remains isomorphic under both types of paths.

[0073] III. Freeze-state machine with hysteresis and recovery soft boot Freezing / unfreezing is implemented using a discrete state machine, with states including running state, frozen state, and unfrozen state: In operational mode, when the amplitude index reaches the entry threshold and the period stability is not greater than the entry threshold, and the number of consecutive cycles that meet the above conditions reaches the entry duration threshold, a coordinated vibration suppression command is output and the system switches to a frozen state. The entry and exit thresholds are paired using hysteresis, with the amplitude entry threshold being greater than the amplitude exit threshold and the period stability entry threshold being less than the period stability exit threshold, making the determination insensitive to boundary jitter.

[0074] In one implementation, the state transition of the state machine is based on the event record timestamp. The entry criterion and exit criterion correspond to the transition conditions from the running state to the frozen state and from the frozen state to the recovery state, respectively. The minimum frozen dwell time for the frozen state is an implementation parameter, which can be 30~120 s, to ensure that the frozen state has an observable vibration damping effect. In the recovery state, the follower control branch is gradually restored by a soft start method. The ramp time of the soft start is an implementation parameter, which can be 60~300 s. If the entry criterion is met again in the frozen state, the count is refreshed and the frozen state is maintained. If the entry criterion is met again in the recovery state, the state machine returns directly to the frozen state and keeps the control quantity of the follower control branch unchanged, so that the state machine maintains consistent execution logic under repeated entry and exit scenarios and noise disturbance scenarios.

[0075] In the frozen state, all load adjustments of the control branches are frozen and their adjustable actuator control values ​​remain unchanged. The main control branch performs single-branch adjustment at a vibration-damping rate lower than the preset upper limit of the rate of change. When the amplitude index is lower than the exit threshold and the risk beyond the exit threshold of periodic stability deterioration subsides, and the number of consecutive periods reaches the release duration threshold, the system switches to the recovery state. If the entry criterion is met again during the frozen state, the frozen state is maintained and the release count is refreshed to cover repeated entry and exit scenarios.

[0076] In the recovery state, the unfreezing of the slave control branch does not directly restore full-scale following, but adopts a soft-start strategy: a recovery delay and a smaller upper limit for the rate of change of control quantity are set for each slave control branch, so that it recovers following in a phased ramp manner; when the pressure oscillation characteristic does not trigger the entry criterion again within a preset time, it returns to the running state. If the entry criterion is triggered again during the recovery state, it immediately returns to the frozen state and refreezes, covering the re-excitation situation during the recovery process.

[0077] In this implementation, resonance determination constrains pressure oscillations in three dimensions: intensity, regularity, and duration. Amplitude thresholds filter low-energy disturbances, periodic stability eliminates non-periodic fluctuations, and duration period count separates incidental noise from genuine oscillations. Combined, these factors ensure that freeze triggering and de-triggering have consistent physical meanings. The time-domain path forms a periodic sequence through peaks and valleys or autocorrelation, suitable for scenarios with high real-time requirements and clear oscillation patterns. The frequency-domain path focuses on specific frequency bands through bandpass and energy extraction, suitable for scenarios with multi-frequency interference or requiring locking the resonance frequency band. Freeze and de-freeze utilize a state machine with hysteresis, providing different entry and exit thresholds to reduce repeated switching near boundaries. After de-freezing, a soft-start recovery is implemented, allowing the secondary branches to gradually intervene with a limited rate of change, preventing simultaneous recovery of multiple branches from causing secondary pressure excitation, thus forming a coordinated closed loop with the main branch's vibration suppression regulation.

[0078] Example 2: Based on Example 1, such as Figure 2 As shown, this embodiment provides an LNG gasification efficiency optimization control system, which is applied to an external gasification system. The external gasification system includes an LNG liquid phase main pipe and multiple gasification branches connected in parallel with the LNG liquid phase main pipe. The control system includes a controller configured to perform: receiving a total external gas volume command, determining a target branch set based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determining the main regulating branch and at least one slave regulating branch; In one embodiment, the controller is deployed in a station control system, DCS, or PLC, and connects to measuring points and actuators via I / O or industrial communication interfaces. The set of fields for measuring point input includes: total external gas output command, real-time pressure of the LNG liquid phase main pipe, main pipe pressure setpoint, measured external gas output of each gasification branch, measured electrical power of each gasification branch, feedback on the opening degree of key valves in each gasification branch, feedback on the pump frequency of each gasification branch, and status variables and interlock signals of each gasification branch. The set of fields for actuator output includes: control values ​​of adjustable actuators in each gasification branch, freeze / unfreeze control flags, and main regulating branch identifier.

[0079] The controller performs consistent sampling cycle management and validity checks on the measurement point signals. When the controller detects that the main pipe pressure measurement point or the critical branch flow measurement point is invalid, it marks the efficiency index and resonance judgment input as unavailable and switches the control mode to perform degraded control with limited ramp adjustment only on the main regulating branch until the measurement point is restored to validity.

[0080] In this embodiment, the invalidity of a measuring point is indicated by the measuring point validity flag field. During the period when a measuring point is invalid, the resonance judgment input is marked as unavailable and the pressure oscillation characteristic record is stopped from being updated. At the same time, the measuring point invalidity flag field in the event record table is kept valid. The efficiency index and share allocation table are stopped from being updated and the efficiency index is marked as unavailable for sorting. The target branch set is determined only based on availability constraints and capacity boundary constraints. When the controller output is restricted by ramp adjustment, the ramp activation flag, the current main regulating branch identifier, and the control quantity limit trigger flag are written into the operation record table so that after the measuring point is restored, the boundary conditions for the start and end of the degradation can be traced back from the operation record table.

[0081] In this embodiment, the sampling period is an implementation parameter used to perform time alignment and differential calculation on the total external gas output command, the real-time pressure of the LNG liquid phase main pipe, the measured values ​​of the external gas output of each gasification branch, the measured values ​​of the electrical power of each gasification branch, and the feedback of the adjustable actuators of each gasification branch. The sampling period can be 1~2 s. The preset statistical window and the preset monitoring window are both implementation parameters, used for calculating the energy consumption index per unit external gas output and extracting pressure oscillation features, respectively. The statistical window can be 60~300 s, and the monitoring window can be 120~600 s. The percentage of valid data in the field set is used to determine the availability of data within the window. Its threshold is an implementation parameter and can be 0.8. The minimum peak spacing in peak-valley detection is an implementation parameter used to suppress false feature points caused by noise and can be 10 s.

[0082] The main control branch performs load regulation before the secondary control branch; the real-time pressure of the LNG liquid phase main pipe is obtained, and based on the deviation between the real-time pressure and the main pipe pressure set value, the secondary control branch performs delayed start and rate of change limit follow-up regulation to keep the real-time pressure within the preset pressure control range; In this embodiment, an efficiency calculation unit is also included. The efficiency calculation unit is configured to obtain the electrical power measurement value and the external gas volume measurement value of each gasification branch, calculate the energy consumption index per unit external gas volume within a preset statistical window, and output it to the controller. In this embodiment, a resonance monitoring unit is also included. The resonance monitoring unit is configured to monitor the pressure oscillation characteristics of the LNG liquid phase main pipe and determine whether the preset resonance determination condition is met. In response to the preset resonance determination condition being met, the controller freezes the secondary regulation branch and controls the main regulation branch to perform single-branch regulation with the vibration suppression rate. In one embodiment, the resonance monitoring unit and the controller share the same real-time pressure data source and employ consistent monitoring windows and data validity rules. The resonance monitoring unit generates event records based on the determination results of pressure oscillation characteristics and resonance criteria. The event record's field set includes: event type, event timestamp, pressure fluctuation peak-to-peak value, dominant period estimate, period stability index, trigger threshold identifier, and current main regulating branch identifier. The event records and the share allocation table are jointly written into the operation record for tracing the control process of a single regulation task.

[0083] Freeze control is managed using a state machine. The state machine's state set includes: normal following, freeze dwell, soft start release, and normal recovery. In the freeze dwell state, the state machine maintains a minimum freeze dwell time. In the soft start release state, it performs a step-wise increase on the upper limit of the control rate of change of the slave regulating branch, allowing the slave regulating branch to gradually resume bearing the remaining share and keeping the main pressure within the pressure control band.

[0084] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements an LNG gasification efficiency optimization control method as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0085] The runtime data structure corresponding to the computer program includes: target branch set, share allocation table, efficiency index record, health status identifier record, historical regulation oscillation index record, pressure oscillation characteristic record, resonance monitoring event record, and controller state machine record. The runtime data structure establishes a relational index using timestamps as the primary key, ensuring that the total external gas volume command change, main regulation branch selection, follow-up regulation by secondary regulation branches, amplitude limiting triggering, and freeze / unfreeze processes within a single regulation task are traceable records under the same index link.

[0086] Example 3: This embodiment provides an application in a large LNG receiving terminal's export peak-shaving scenario. The export gasification system includes one LNG liquid phase main pipe and multiple gasification branches connected in parallel to the LNG liquid phase main pipe. Each gasification branch operates in parallel and jointly supplies gas to the downstream export pipeline network. In this scenario, the total export gas volume command issued by the external dispatcher may have a step or ramp-like change, requiring export tracking while being constrained by the LNG liquid phase main pipe pressure. The LNG liquid phase main pipe pressure must be maintained within the pressure control band corresponding to the main pipe pressure setpoint. The gasification branches operate under capacity boundary constraints, which are jointly limited by the branch's minimum stable operating load, the branch's maximum allowable load, valve opening limits, pump operating range, and heat exchange capacity limitations. The minimum number of operating branches constraint is used to maintain export continuity and redundancy.

[0087] In this embodiment, the controller is deployed in the station control system, DCS, or PLC. The set of measurement point input fields accessed by the controller includes: total external gas output command, real-time pressure of the LNG liquid phase main pipe, main pipe pressure setpoint, external gas output measurement value of each gasification branch, electrical power measurement value of each gasification branch, opening feedback of key valves of each gasification branch, pump frequency feedback of each gasification branch, status quantity and interlock signal of each gasification branch. The set of execution output fields of the controller includes: control quantity of adjustable actuators of each gasification branch, freeze / unfreeze control flag and main regulating branch identifier. Time alignment uses the timestamp generated by the controller clock as the alignment key, and the sampling period is the implementation parameter, which can be 1~2 s. When there is an inconsistency in the sampling frequency of the measurement point input, the low frequency signal is resampled to the time grid corresponding to the sampling period. The resampling process retains the original timestamp and resampling flag field. When a single sample is missing, the previous valid value is retained and the missing measurement flag field is set. The missing measurement flag field and the timestamp are written together into the operation record table to support traceability and deduplication.

[0088] A complete operation begins with initialization. Upon startup, the controller performs validity checks on key measurement point signals and loads the most recently valid efficiency index record, health status record, and historical regulation oscillation index record. If records are missing or a window has not yet been formed, the efficiency index is marked as unsuitable for sorting, and the target branch set is determined solely based on availability constraints and capacity boundary constraints. Subsequently, the controller establishes preset statistical windows and preset monitoring windows. The statistical window, a parameter for implementation, can be 60–300 s and is used to calculate the energy consumption per unit of external gas output. The monitoring window, also a parameter for implementation, can be 120–600 s and is used to extract pressure oscillation characteristics. Window updates use a sliding method, and the threshold for the percentage of valid data within the window, a parameter for implementation, can be 0.8, used to determine the availability of data within the window.

[0089] After entering continuous operation, upon receiving the total external gas output command and detecting a change, the controller forms a target branch set based on the direction and magnitude of the change. The determination of the target branch set is based on availability constraints, capacity boundary constraints, and minimum number of operating branches constraints. Availability constraints include interlocks not being triggered, key measuring point signals being valid, adjustable actuators being controllable, and the branch being in an adjustable operating state. When the change is a positive increment, the target branch set includes gasification branches capable of increasing load and having load margin; when the change is a negative increment, the target branch set includes gasification branches capable of decreasing load and not triggering the lower limit constraint. The controller determines the primary and secondary regulating branches within the target branch set. The determination of the primary regulating branch is based on a combination of equipment runtime, health status indicators, and historical regulation oscillation indicators. Health status indicators are generated by combining pump, valve, and key measuring point status quantities. Historical regulation oscillation indicators consist of the number of times regulation was participated in, peak-to-peak pressure deviation statistics, peak pressure change rate statistics, number of times control quantities reversed, peak control change rate statistics, and decay time statistics. The controller writes the availability flag, load margin, and boundary triggering reason of this adjustment into the target branch set record and generates a share allocation table. The share allocation table field set includes branch identifier, load margin, efficiency index, weight, first share, remaining share, boundary constraint flag, and redistribution number. The weight is generated by combining the efficiency index and load margin after normalization at the same timestamp, so that the weight, first share, and remaining share are aligned with the same key.

[0090] After the main and secondary control branches are determined, the controller first issues a control update to the main control branch, enabling it to bear the first share of the change. The allocation of the first share and the remaining share is limited by load margin and capacity boundary constraints. When the main control branch triggers the control limit or the load margin is insufficient, the unfinished first share is transferred to the remaining share and triggers redistribution. During redistribution, the secondary control branch with better efficiency and no boundary constraints is preferentially selected to bear the transferred share. After the main control branch starts load regulation, the controller performs delayed start and rate-of-change limit follow regulation on the secondary control branch. The delay time is an implementation parameter and is adaptively generated by real-time operating conditions. Real-time operating conditions include at least the LNG liquid phase main pipe pressure change rate, the control quantity change rate of the adjustable actuator of the main control branch, the inertial time constant of the secondary control branch, and the number of currently operating branches. The determination of the secondary control branch's start and stop of follow is based on hysteresis and minimum dwell time, and it is not allowed to re-enter follow within the cooling time after exiting, in order to reduce the frequent start and stop of follow by the secondary control branch. After the delay expires, the controller uses the remaining fraction as the target input from the regulating branch. It then performs ramping processing on the remaining fraction to generate the target increment for the regulating branch. The absolute value of the deviation and the rate of change of real-time pressure are used as the limiting input to dynamically restrict the rate of change of the control quantity. When the control quantity cannot continue to change due to limiting or stopping, anti-integral saturation processing is performed on the integral state to freeze or backcalculate, ensuring continuous recovery of the control quantity after the limiting is lifted. If the regulating branch triggers a limiting during the follow-up regulation process, causing the remaining fraction to fail to complete within the preset time, the uncompleted portion is temporarily stored as a pending fraction, and released after the real-time pressure returns to the pressure control zone. The pending fraction and the trigger reason field are simultaneously written to the operation log table.

[0091] For resonance monitoring, the controller continuously monitors the real-time pressure of the LNG liquid phase manifold and extracts pressure oscillation characteristics. The pressure oscillation characteristic field set includes the monitoring window identifier, pressure fluctuation peak-to-peak value, dominant period estimate, period stability index, bandpass energy index, adjacent period amplitude change rate, and effective data ratio. The resonance determination criteria are composed of the pressure oscillation amplitude threshold, period stability threshold, and duration period number threshold. The determination path can use time-domain autocorrelation or peak-valley detection to obtain the oscillation period and amplitude, or use frequency-domain bandpass filtering combined with the energy threshold to obtain the resonance intensity and form the period stability and duration period number determination accordingly. When the number of identifiable complete periods within the monitoring window is less than 3, the period stability index is marked as unavailable and the entry criterion is not triggered. If the pressure oscillation characteristics meet the resonance judgment condition, the controller outputs a coordinated vibration suppression command and freezes the load regulation of all slave regulation branches, keeping the control quantity of the slave regulation branches unchanged before freezing. At the same time, the main regulation branch performs single-branch regulation at a vibration suppression change rate lower than the preset upper limit of the change rate. Freezing and unfreezing are performed through a state machine with hysteresis. The state machine includes at least a running state, a frozen state, and a recovery state. The minimum freeze dwell time and the recovery soft-start ramp time are implementation parameters, which can be set to 30~120 s and 60~300 s, respectively. If the entry criterion is met again during the frozen state, the unfreezing count is refreshed and the freeze is maintained. If the entry criterion is met again during the recovery state, the system returns to the frozen state. When the exit criterion is met and the unfreezing duration period threshold is reached, the system switches to the recovery state. The slave regulation branches are gradually restored to follow using the recovery soft-start method, so that the change rate of the control quantity gradually transitions from the lower upper limit to the normal upper limit. During the recovery phase, the entry criterion is continuously monitored to cover repeated entry and exit scenarios and noise misjudgment scenarios. The set of event record fields related to resonance monitoring includes event type, event timestamp, peak-to-peak pressure fluctuation, estimated dominant cycle, cycle stability index, trigger threshold identifier, and current main regulating branch identifier. Event records and share allocation table are written together into the operation record table.

[0092] During the abnormal phase, if an invalid pressure measurement point in the main pipe or an invalid flow measurement point in a key branch is detected, the measurement point validity flag field is used to determine the measurement point invalidity and the invalid measurement point flag field is set. During the measurement point invalidity period, the pressure oscillation characteristic record is stopped from being updated and the resonance judgment input is marked as unavailable. The efficiency index and share allocation table are stopped from being updated and the efficiency index is marked as unavailable for sorting. The target branch set is determined only based on availability constraints and capacity boundary constraints. The controller switches to degraded control that only performs limited ramp adjustment on the main regulating branch. The freeze / unfreeze control flag maintains its existing state and does not trigger new freeze entry criteria. The ramp activation flag, the current main regulating branch identifier, and the control quantity limit trigger flag are written to the operation record table. During the recovery phase, when the measurement point validity flag field is restored to validity and the proportion of continuous valid data meets the threshold, the controller reinitializes the statistics window and monitoring window, clears the measurement point invalidity flag, and resumes efficiency index calculation and pressure oscillation characteristic extraction. The operation record table is deduplicated by timestamp and the flag fields of the missing measurement intervals are merged by alignment key to keep the boundary conditions for the start and end of the degradation traceable at the record layer.

[0093] Through the above-described operational process, the gasification system can achieve priority response of the main regulating branch and follow-up regulation of the slave regulating branch under the constraint of the main pipe pressure when the total gas output command changes rapidly. This delays and limits the simultaneous action of parallel branches, suppresses the amplification of main pipe pressure disturbance caused by fast-responding branches competing for flow, and binds the intervention timing and intensity of slave regulating branches to the main pipe pressure status, thereby maintaining the real-time pressure within the pressure control band and ensuring the continuity of gas output tracking.

[0094] Furthermore, upon identifying persistent periodic pressure fluctuations, the system freezes the regulation branch and then adjusts it single-path by the main regulation branch at a rate of oscillation suppression. After unfreezing, a state machine with hysteresis and a soft-start recovery mechanism are used to gradually restore the system, suppressing the sustained maintenance of low-frequency pressure oscillations and secondary excitation during the recovery phase. When measurement points are missing or invalid, a degraded control with limited ramp adjustment of the main regulation branch is employed while maintaining the integrity of the recording link, ensuring consistent and traceable control behavior during anomalies and recovery. By using the energy consumption index per unit of external gas output to participate in the selection and allocation of target branch sets, a reproducible branch selection and allocation caliber can be formed under the premise of meeting availability and capacity boundary constraints. Through the allocation table, event log, and operation log, the command changes, branch selection, amplitude limiting triggers, and freeze / unfreeze processes of a single regulation task are linked by timestamps, reducing reliance on manual peak-shifting operations and maintaining the reproducibility and verifiability of the operation process.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0096] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this application and form different embodiments. For example, all the embodiments above can be used in any combination. The information disclosed in this background section is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art.

Claims

1. A method for optimizing and controlling LNG gasification efficiency, characterized in that, The method is applied to an external gasification system, which includes an LNG liquid phase main pipe and multiple gasification branches connected in parallel with the LNG liquid phase main pipe. Each gasification branch is equipped with an adjustable actuator for regulating the LNG flow rate in the branch. Step S1: Receive the total external gas volume command, determine the target branch set to be regulated based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determine the main regulating branch and at least one secondary regulating branch from the target branch set. Step S2: Control the main adjustment branch to change the adjustable actuator to take on the first share of the change before the secondary adjustment branch; Step S3: Obtain the real-time pressure of the LNG liquid phase main pipe, and based on the deviation between the real-time pressure and the main pipe pressure set value, perform delayed start and change rate limit adjustment on the remaining share borne by the regulating branch, so that the real-time pressure is kept within the preset pressure control band.

2. The LNG gasification efficiency optimization and control method as described in claim 1, characterized in that, The efficiency index is the energy consumption index per unit of external gas output, which is calculated by the measured power value of the target branch and the measured gas output value of the target branch within a preset statistical window. The method for determining the target branch set includes: under the condition of satisfying the minimum number of operating branches, prioritizing the selection of gasification branches with lower energy consumption per unit of external gas transmission volume into the target branch set.

3. The LNG gasification efficiency optimization control method as described in claim 1, characterized in that, The determination of the main regulating branch includes: For each gasification branch within the target branch set, acquire the equipment runtime, health status indicators, and historical regulation oscillation indicators; According to the preset priority, the gasification branch that meets the operating conditions and has a small historical regulation oscillation index is selected as the main regulation branch, and the remaining target branches are determined as the secondary regulation branches.

4. The LNG gasification efficiency optimization and control method as described in claim 1, characterized in that, The delayed start and rate of change limit adjustment include: After the main regulating branch starts to perform load regulation, the load regulation of the slave regulating branch is started after a delay period. During the adjustment process from the regulating branch, the absolute value of the deviation and the rate of change of the real-time pressure are used as limiting inputs to dynamically limit the rate of change of the control quantity of the adjustable actuator. The delay time is adaptively generated by the controller based on real-time operating conditions, which include at least the LNG liquid phase main pipe pressure change rate, the control quantity change rate of the adjustable actuator of the main regulating branch, the inertial time constant of the slave regulating branch, and the number of currently operating branches. Furthermore, hysteresis and minimum dwell time are introduced in the determination of slave regulating branch start-up and exit-up, so that the slave regulating branch maintains its existing state before the dwell time is reached, and is not allowed to re-enter the following mode within the cooling time after exiting the following mode, thereby reducing the frequent start-up and stop-up of slave regulating branch.

5. The LNG gasification efficiency optimization and control method as described in claim 4, characterized in that, The dynamic restrictions include: When the absolute value of the deviation is greater than the first threshold, the rate of change of the control quantity of the adjustable actuator is limited to the first upper limit; When the absolute value of the deviation is not greater than the first threshold and the rate of change of the real-time pressure is greater than the second threshold, the rate of change of the control quantity of the adjustable actuator is limited to the second upper limit, where the first upper limit is less than the second upper limit.

6. The LNG gasification efficiency optimization control method as described in claim 1, characterized in that, Continuously monitor the pressure oscillation characteristics of the LNG liquid phase manifold; If the pressure oscillation characteristic meets the preset resonance judgment condition, a cooperative vibration suppression command is output; in response to the cooperative vibration suppression command, all load regulation of the control branch is frozen and its adjustable actuator control amount is kept unchanged, while the main control branch is controlled to perform single-branch regulation at a vibration suppression change rate lower than the preset change rate upper limit, until the pressure oscillation characteristic exits the preset resonance judgment condition and the freeze is released. The preset resonance determination conditions are composed of a pressure oscillation amplitude threshold, a period stability threshold, and a duration period number threshold. The resonance determination is achieved by using time-domain autocorrelation or peak-valley detection to obtain the oscillation period and amplitude, or by using frequency-domain bandpass filtering combined with an energy threshold to obtain the resonance intensity and thereby determine the period stability and duration period number. Freezing and unfreezing are performed by a state machine with hysteresis. When the entry criterion is met, the slave adjustment branch is frozen and adjusted by the main adjustment branch at the vibration suppression rate. When the exit criterion is met, the freeze is lifted and the slave adjustment branch is gradually restored to follow the oscillation using a soft-start recovery method to cover repeated entry and exit scenarios and noise misjudgment scenarios.

7. An LNG gasification efficiency optimization and control system, applied to an external gasification system, the external gasification system comprising an LNG liquid phase main and multiple gasification branches connected in parallel with the LNG liquid phase main; characterized in that, The control system includes a controller configured to perform: receiving a total external gas volume command, determining a target branch set based on the change in the total external gas volume command and the efficiency index of each gasification branch, and determining a main regulating branch and at least one secondary regulating branch; The main control branch performs load regulation before the secondary control branch; the real-time pressure of the LNG liquid phase main pipe is obtained, and based on the deviation between the real-time pressure and the main pipe pressure set value, the secondary control branch performs delayed start and rate of change limit follow-up regulation to keep the real-time pressure within the preset pressure control range.

8. The LNG gasification efficiency optimization control system as described in claim 7, characterized in that, It also includes an efficiency calculation unit, which is configured to acquire the electrical power measurement value and the external gas output measurement value of each gasification branch, calculate the energy consumption index per unit external gas output within a preset statistical window, and output it to the controller.

9. The LNG gasification efficiency optimization control system as described in claim 7, characterized in that, It also includes a resonance monitoring unit, which is configured to monitor the pressure oscillation characteristics of the LNG liquid phase main pipe and determine whether a preset resonance determination condition is met; the controller, in response to meeting the preset resonance determination condition, freezes the secondary regulation branch and controls the main regulation branch to perform single-branch regulation with the vibration suppression rate.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the LNG gasification efficiency optimization control method according to any one of claims 1 to 6.

Images