A method, system, program product and storage medium for synchronously bridging a gas turbine simulation and a cycle control

Abstract

The application discloses a kind of gas turbine simulation and period control synchronous bridge method, system, program product and storage medium, it is related to gas turbine control technical field.The method includes: establishing the task table including reference period and additional task period;From variable step simulation model acquisition time-stamped data frame;Through simulation model, OPC interface, communication station and the closed loop link transmission data and link information acquisition formed by controller;Link delay and jitter are estimated online;According to task table, variable is grouped and carries out hierarchical resampling and reconstruction;To key closed loop variable, semantic consistency and effective period check are carried out;Mode switching and stability constraint are carried out based on risk value;Priority mechanism is used to inject data to controller and return controller output;Record reproduction log is used for scene reproduction.The application can effectively solve the problem of multi-rate mismatch and uncontrollable link time delay jitter, improve closed loop commissioning stability and reproducibility.

Description

Technical Field

[0001] This invention relates to the field of gas turbine control technology, specifically to a method, system, program product, and storage medium for synchronous bridging of gas turbine simulation and periodic control. Background Technology

[0002] In the integrated commissioning of gas turbine control systems, it is typically necessary to establish a closed loop between the variable-step excitation simulation model and the actual controller to verify the control logic, protection logic, and operating condition switching strategies. The actual controller operates on a fixed cycle, generally employing a scheduling structure that combines a base cycle with multiple integer multiples of the task cycle in parallel. However, the simulation model output, due to its variable integral step size, results in uneven data generation time. Current common commissioning methods rely on general interfaces such as OPC for data exchange. OPC refresh cycles are typically on the order of seconds, and after the communication station comprehensively manages data from the gas turbine main control unit and other equipment, the forwarding point table can reach 20,000 to 30,000 points.

[0003] Under the aforementioned conditions, the core technical challenge lies in the uncontrollable multi-rate mismatch and link delay jitter. Specifically, this manifests as: simulation data refreshed at the second level directly driving millisecond-level closed-loop control tasks, resulting in equivalent sample-and-hold errors and discrete jumps, inducing oscillations in temperature control or load control loops; simultaneously, link delay fluctuates with network load, queue length, and batch arrivals, lacking online estimation and constraint mechanisms, causing key closed-loop variables to be driven by outdated data. This problem is particularly prominent in large-scale data table scenarios with 20,000 to 30,000 points, and currently lacks an engineered synchronous bridging method capable of collaboratively managing multi-rate mismatch and link delay jitter. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention proposes a method, system, program product, and storage medium for synchronous bridging of gas turbine simulation and periodic control.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows; Firstly, a method for synchronously bridging gas turbine simulation and periodic control includes the following steps: Step S1: Establish a task table, set the baseline period T0 and the additional task period Ti=ni·T0. The task table shall include at least the task identifier, ni, variable set, variable grouping, reconstruction strategy, validity period parameter, and risk threshold parameter. Step S2: Data acquisition and labeling. Obtain data frames from the variable step size simulation model, including at least timestamp ts, variable identifier, variable value, and quality identifier. Step S3: Link closing and link information acquisition. Transmit the data frame of S2 through the closed-loop link of simulation model-OPC-communication station-controller-communication station-OPC-simulation model, and acquire link information. Step S4: Online estimation of latency and jitter. Based on the link information in S3, estimate the latency τk and jitter jk in each baseline or task cycle. Step S5: Hierarchical resampling and reconstruction: Group and hierarchically resample the variables of S2 according to the task table of S1, and reconstruct them into a fixed periodic sequence aligned with T0 or Ti by combining τk and jk from S4. The reconstruction includes preservation, interpolation or prediction, and the strategy is switched online with τk and jk. Step S6: Semantic Consistency and Validity Verification. Perform semantic consistency processing on the key closed-loop variables after S5 reconstruction, including unit / range mapping, validity verification, and nonlinear consistency. Step S7: Risk monitoring and stability constraints. Calculate the risk value Rk based on the variables processed in S6 and τk and jk in S4. When Rk meets the triggering condition, switch to conservative mode. Simultaneously, adjust the reconstruction strategy and injection priority in S5, and apply stability constraints to the output after reconstruction in S5. Step S8: Priority injection and feedback. The data after S5 reconstruction, S6 unification, and S7 constraint are injected into the controller using a priority mechanism, and the controller output is fed back to the simulation model. Step S9: Record the reproduction log, including τk and jk of S4, the task table version of S1, the mode switching event of S7, key variable window data, and injection statistics, for scene reproduction.

[0006] As a further technical solution, the quality identifier in step S2 includes at least one or more of the following: a data validity identifier, a simulation model running status identifier, and a timestamp reliability identifier; when the quality identifier indicates that the data is invalid or the simulation model is in an unstable state, the corresponding data frame does not participate in the reconstruction of S5 or triggers the conservative mode. As a further technical solution, the link information collected in step S3 for delay estimation includes at least one of the following: simulation side transmission timestamp, OPC interface reception / forward timestamp, communication station reception / forward timestamp, controller reception timestamp, queue length information from communication station to controller, and network round-trip time measurement value from OPC interface to communication station.

[0007] As a further technical solution, the hierarchical processing in step S5 includes: assigning variables to base period tasks or additional period tasks of different multiples according to the task table, and adopting independent reconstruction strategies for the same variable under different task periods; wherein the reconstruction strategy corresponding to the base period task is controlled by the online switching of τk and jk, and the reconstruction strategy corresponding to the additional period task adopts either maintain reconstruction or interpolation reconstruction and does not switch with τk and jk.

[0008] As a further technical solution, the calculation of the risk value Rk in step S7 further introduces the delay τk and jitter jk estimated in S4: when τk exceeds the first threshold or jk exceeds the second threshold, the risk value Rk is increased by an additional risk term on the basis of the original value; the additional risk term increases monotonically with the degree of τk and jk exceeding the limit, and the additional risk term decays in a hysteretic manner after τk and jk return to the normal range.

[0009] As a further technical solution, the stability constraint in step S7 includes at least one of limiting the rate of change of the output variable, limiting the acceleration of change, and integral separation; and in the conservative mode, the amplitude limit or rate limit value of the stability constraint is further narrowed compared with the normal mode.

[0010] As a further technical solution, the reproduction log in step S9 also includes trigger snapshot information: when the risk value Rk exceeds the trigger condition or a mode switch occurs, the waveforms of key closed-loop variables, link delay sequences, queue length change curves, and injection action records within at least one control cycle window before and after the switch time are automatically recorded for subsequent reproduction of the delay jitter and strategy response process during the working condition switch.

[0011] Secondly, the present invention also provides a gas turbine combustion switching zone no-stay crossing control and protection linkage system, comprising: The first module is configured to create a task table, set a baseline period T0 and an additional task period Ti=ni·T0, and the task table includes at least a task identifier, ni, a set of variables, variable grouping, a reconstruction strategy, a validity period parameter, and a risk threshold parameter. The second module is configured for data acquisition and labeling, which obtains data frames from the variable step size simulation model, including at least timestamp ts, variable identifier, variable value, and quality identifier. The third module, link closure and link information acquisition, transmits S2 data frames through a closed-loop link of simulation model-OPC-communication station-controller-communication station-OPC-simulation model and acquires link information. The fourth module is configured for online latency and jitter estimation. Based on the link information of S3, it estimates the latency τk and jitter jk in each baseline or task cycle. The fifth module is configured for hierarchical resampling and reconstruction: the variables of S2 are grouped and hierarchically according to the task table of S1, and combined with τk and jk of S4, the reconstructed into a fixed periodic sequence aligned with T0 or Ti. The reconstruction includes preservation, interpolation or prediction, and the strategy is switched online with τk and jk. The sixth module is configured for semantic consistency and validity period verification. It performs semantic consistency processing on the key closed-loop variables after S5 reconstruction, including unit / range mapping, validity period verification, and nonlinear consistency. The seventh module is configured with risk monitoring and stability constraints. It calculates the risk value Rk based on the variables processed by S6 and τk and jk of S4. When Rk meets the triggering conditions, it switches to conservative mode. At the same time, Rk adjusts the reconstruction strategy and injection priority of S5 and applies stability constraints to the output after reconstruction of S5. The eighth module is configured for priority injection and feedback. It injects the data after S5 reconstruction, S6 unification, and S7 constraint into the controller using a priority mechanism, and feeds the controller output back to the simulation model. The ninth module is configured to record reproduction logs, including τk and jk of S4, the task table version of S1, the mode switching event of S7, key variable window data, and injection statistics, for the purpose of scene reproduction.

[0012] Thirdly, the present invention also provides a computer program product, the computer program product including a computer program stored on a non-transitory storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer executes the gas turbine simulation and periodic control synchronous bridging method.

[0013] Fourthly, the present invention also provides a storage medium storing a computer program thereon, which, when executed, implements the aforementioned gas turbine simulation and periodic control synchronous bridging method.

[0014] The beneficial effects of this invention are as follows: This invention addresses the problems of multi-rate mismatch and uncontrollable link delay jitter by integrating online delay and jitter estimation, hierarchical resampling reconstruction, semantic consistency, priority injection, and risk monitoring into a closed-loop link of "simulation model—OPC interface—communication station—controller," forming a "risk-strategy-injection" closed-loop anti-divergence mechanism. The output of online delay and jitter estimation directly drives the strategy selection for hierarchical resampling reconstruction: in normal mode, hold-and-reconstruction is used, while in conservative mode, interpolation or prediction reconstruction is used. This adaptively reconstructs second-level variable-step data into a sequence aligned with the controller's reference period and multiple integer multiples of the task period, effectively suppressing discrete jumps and equivalent sample-and-hold errors, and eliminating oscillations in temperature control and load control loops. Online estimation, combined with validity period verification and over-limit actions, ensures that key closed-loop variables receive controllable worst-case injection delays, preventing outdated data from driving critical loops. By applying nonlinear consistency processing such as unit / range mapping, dead zone / saturation / rate limiting, and filtering to the fuel valve, VIGV, and measurement link, the risks of valve position jitter, integral saturation, and protection false triggering are reduced. With a forwarding scale of 20,000 to 30,000 points, priority injection and sharded incremental updates ensure the injection stability of 200 to 300 key closed-loop points, and a worst-case injection latency cap is set. Risk values ​​constructed based on error energy, instruction increment, saturation ratio, and congestion intensity simultaneously influence the reconstruction strategy, injection priority, and stability constraints. When the risk exceeds the limit, it automatically switches to a conservative mode and applies stricter slope limits and anti-integral saturation constraints, forming a risk-driven anti-divergence link. Solidified logs record latency jitter, task table version, mode switching, and injection statistics, significantly improving the reproducibility of the operating condition switching process and the efficiency of fault location, thereby improving the overall stability and repeatability of closed-loop joint debugging, and reducing debugging costs and iteration cycles. Attached Figure Description

[0015] Figure 1 The flowchart of the synchronous bridging method for gas turbine simulation and periodic control proposed in this invention is shown below; Detailed Implementation Example 1 As described in the background section, the direct driving of millisecond-level closed-loop control tasks by existing simulation data refreshed at the second level results in equivalent sample-and-hold errors and discrete jumps, inducing oscillations in temperature control or load control loops. Simultaneously, link latency fluctuates with network load, queue length, and batch arrivals, lacking online estimation and constraint mechanisms, leading to key closed-loop variables being driven by outdated data. This problem is particularly prominent in large-scale data tables of 20,000 to 30,000 points, and currently lacks an engineered synchronous bridging method capable of collaboratively managing multi-rate mismatch and link latency jitter.

[0016] To address this issue, this embodiment proposes a synchronous bridging method for gas turbine simulation and periodic control. The core innovation lies in constructing a functional interlocking and collaborative closed loop among three elements—online delay and jitter estimation, hierarchical resampling and reconstruction, and risk monitoring and stability constraints—within the closed-loop link of "variable step size simulation model—OPC interface—communication station—controller—communication station—OPC interface—simulation model." This forms a "risk-strategy-injection" closed-loop anti-divergence mechanism, thereby addressing the problems of multi-rate mismatch and uncontrollable link delay jitter.

[0017] This embodiment provides a method for synchronously bridging gas turbine simulation and periodic control, referring to... Figure 1 Specifically, it includes the following steps: Step S1: Establish a task table, set the gas turbine controller base period T0 = 50ms, and configure at least one additional task period Ti = ni·T0, where ni is a positive integer. The additional task period in S1 satisfies... ,and This is a configurable integer.

[0018] Specifically, the task table should include at least a task identifier, a cycle multiple ni, a variable set, variable grouping, a set of reconstruction strategies, variable validity parameters, and risk threshold parameters; among which, the variable grouping in S1 should include at least: G1 key closed-loop group, G2 slow variable group, and G3 monitoring group; the target injection cycle for the G1 key closed-loop group is... The target injection period for the G2 slow variable group is The injection cycle of the G3 monitoring group shall not be less than an integer multiple thereof. The OPC interface refresh cycle is ; when At that time, S5 to G1 critical closed-loop group at two adjacent OPC refresh times and Construct a 50ms reconstruction sequence between them The construction method includes at least one of the following: 1. Maintaining reconstruction: Internal extraction ;2) Interpolation reconstruction: based on , Perform linear or cubic interpolation; 3) Predict reconstruction: perform one-step prediction based on local slope or local model; and apply stability constraints to the reconstructed sequence.

[0019] Specifically, stability constraints include at least a slope constraint, satisfying: fuel valve commands : ; Regarding the VIGV angle : ;in It is a configurable parameter and allows different values ​​to be used depending on the startup phase.

[0020] Specifically, the G1 critical closed-loop group includes at least two of the following: fuel valve command, VIGV command, temperature control related variables, and load control related variables, and the number of G1 critical closed-loop points is 200–300; the G2 slow variable group includes at least the ambient temperature variable and configures it as a low-priority task cycle variable.

[0021] In conservative mode, at least two of the following should be performed on the critical closed-loop group of G1: maintain reconstruction, limit slope / acceleration, anti-integral saturation processing, and limit critical quantity amplitude processing.

[0022] Step S2: Data Acquisition and Labeling: Obtain simulation data frames from the variable step size gas turbine excitation simulation model. The simulation data frames shall include at least the simulation timestamp ts, variable identifier, variable value, and quality identifier. As a further technical solution, the quality identifier in step S2 includes at least one or more of the following: data validity identifier, simulation model running status identifier, and timestamp trust identifier; when the quality identifier indicates that the data is invalid or the simulation model is in an unstable state, the corresponding data frame does not participate in the reconstruction of S5 or trigger the conservative mode.

[0023] Step S3: Link closure and link information acquisition. The simulation data frames acquired in step S2 are transmitted through the closed-loop link of “simulation model - OPC interface - communication station - controller - communication station - OPC interface - simulation model”, and link information for delay estimation is acquired. As a further technical solution, the link information collected in step S3 for delay estimation includes at least one of the following: simulation side transmission timestamp, OPC interface reception / forward timestamp, communication station reception / forward timestamp, controller reception timestamp, queue length information from communication station to controller, and network round-trip time measurement from OPC interface to communication station.

[0024] Step S4: Online Delay and Jitter Estimation: Based on the link information collected in Step S3, estimate the equivalent link delay τk and jitter jk in each baseline period or each task period. The equivalent delay is calculated in S4. The estimation is performed using one or a combination of the following methods: 1) Timestamp Alignment: Attach send / receive timestamps to data on the simulation side, OPC interface, or communication station, and calculate... ; 2) Correlation estimation: in a length of The cross-correlation between input and response is calculated within a sliding window, and the peak lag is taken as... ; 3) Minimize prediction error: Construct a local prediction model to minimize the prediction error. smallest As .

[0025] The above , The calculation uses a sliding window, with a window length of... It is a configurable positive integer, where jitter Use the standard deviation of the delayed sequence or the range of the delayed sequence.

[0026] Step S5: Hierarchical resampling and reconstruction. The variables in the simulation data frame obtained in Step S2 are grouped and hierarchically processed according to the task table established in Step S1. Combined with τk and jk estimated in Step S4, the variable step-size data is reconstructed into a fixed-period sequence aligned with T0 or Ti. Reconstruction includes at least one of preservation, interpolation, and prediction, and the reconstruction strategy is switched online based on τk and jk. The strategy switching in Step S5 satisfies the following condition: when jitter... Exceeding the threshold or delay When the threshold is exceeded, the system switches from interpolation / prediction to hold; when... , Once the value drops to the recovery threshold and the dwell time requirement is met, the value is restored to interpolation / prediction.

[0027] As a further technical solution, the layered processing in step S5 includes: assigning variables to the baseline periodic task or additional periodic tasks of different multiples according to the task table, and adopting independent reconstruction strategies for the same variable under different task periods; wherein the reconstruction strategy corresponding to the baseline periodic task is controlled by the online switching of τk and jk, and the reconstruction strategy corresponding to the additional periodic task adopts the maintenance reconstruction or interpolation reconstruction and does not switch with τk and jk.

[0028] Step S6: Semantic Consistency and Validity Verification. Semantic consistency processing is performed on at least one type of key closed-loop variable after reconstruction in Step S5. Semantic consistency processing includes at least one of unit / range mapping, timestamp validity verification, and actuator / measurement link nonlinearity consistency. The timestamp validity verification in Step S6 satisfies the following: when Data is judged to be expired at that time, among which Configurable validity period, validity period Configure by group: G1 key closed-loop configuration G2 slow variable group configuration G3 monitoring group configuration Furthermore, all three are independently configurable.

[0029] Specifically, expired data will be downgraded: For the critical closed-loop group G1, maintain the previous effective value and apply slope or acceleration constraints; for the slow variable group G2, use interpolation correction or delayed injection; for the monitoring group G3, discard or reduce the injection frequency.

[0030] It should be noted that the nonlinear uniformity in S6 includes at least one of the following: 1) Apply dead zone to fuel valve command ,saturation Rate limiting; 2) Apply saturation to the VIGV angle With rate limiting; 3) Apply first-order filtering to the temperature measurement variable. .

[0031] Step S7: Risk Monitoring and Stability Constraints. Based on the key closed-loop variables processed in Step S6, calculate the risk value Rk, and introduce τk and jk estimated in Step S4 as additional inputs for risk calculation. When Rk meets the triggering condition, switch the synchronization bridging from normal mode to conservative mode. The risk value Rk is also used to adjust the reconstruction strategy and subsequent injection priority in Step S5, and applies stability constraints to the output variables after reconstruction in Step S5. satisfy:

[0032] in To maintain a saturated and continuous proportion, For batch arrival / congestion intensity indicators, Configurable weights.

[0033] As a further technical solution, the calculation of the risk value Rk in step S7 further incorporates the delay τk and jitter jk estimated in S4: when τk exceeds the first threshold or jk exceeds the second threshold, an additional risk term is added to the risk value Rk based on the original value; the additional risk term monotonically increases with the degree of τk and jk exceeding the limits, and the additional risk term decays hysteretly after τk and jk return to the normal range; the risk value in S7 It consists of at least two types of items: window energy based on temperature control error or load control error; window energy based on fuel valve or VIGV command increment; and optionally includes saturation percentage item or batch arrival congestion item.

[0034] Step S7: The triggering condition includes at least one of the following: 1) Continuous satisfaction 1) Baseline period; 2) Risk growth rate Continuous satisfaction One benchmark period; among which This is the average of the previous window. Configurable parameters Mode switching employs a hysteresis and dwell mechanism: Entering conservative mode requires meeting an entry threshold condition; exiting conservative mode requires meeting a recovery threshold condition and a dwell time. The recovery threshold is less than the entry threshold.

[0035] Furthermore, the stability constraints in step S7 include at least one of limiting the rate of change of the output variable, limiting the acceleration of change, and integral separation; and in the conservative mode, the amplitude limit or rate limit value of the stability constraints is further narrowed compared to the normal mode.

[0036] Step S8: Priority Injection and Feedback. Group the variables in the task table of Step S1, and use the priority mechanism to inject the data after reconstruction in Step S5, semantic consistency in Step S6, and stability constraint in Step S7 into the controller, and feed back the controller output to the simulation model to form a virtual-real closed loop. S8 employs a priority queue injection mechanism and sets an upper limit constraint on the injection delay; the injection delay is defined as the timestamp of the variable data frame. Until the time the variable is written to the controller input area Time difference ; Set a worst-case injection delay cap for the critical closed-loop group of G1 (200–300 points). ,in ; Set a worst-case injection delay cap for the G2 slow variable set. ,in ; Set a worst-case injection delay cap for the G3 monitoring group. ,in .

[0037] When the injection delay of any group is detected to exceed the corresponding worst-case injection delay limit, an over-limit action is executed. The over-limit action includes at least one of the following: 1) triggering conservative mode for the G1 critical closed-loop group and forcing the adoption of maintain reconstruction and limit slope constraints; 2) reducing the injection frequency or delaying the injection for the G2 slow variable group; 3) performing drop, frequency reduction or fragmented injection limit contraction for the G3 monitoring group.

[0038] When the communication station forwarding point table has a size of 20,000 to 30,000 points, the low-priority queue uses fragmented injection: a maximum of [number] injections are allowed per base period. A low-priority point, and ensure that the injection is completed in each baseline cycle of the G1 critical closed loop group.

[0039] The sharded injection includes an incremental update strategy: updates are only injected when the variable changes exceed a threshold or the quality indicator changes, and a fallback update cycle is set. .

[0040] Step S9: Record the reproduction log, which includes at least the estimated τk and jk from step S4, the version of the task table established in step S1, the mode switching event in step S7, key variable window data, and point table injection statistics, for the reproduction experiments of startup, grid connection, load shedding, and fuel switching scenarios. Congestion intensity index The calculation is based on the burst level of arrival points within the OPC refresh cycle. When the burst level exceeds the threshold, a conservative mode is triggered or the injection rate of low-priority injections is reduced.

[0041] When the communication station forwarding point table has a size of 20,000 to 30,000 points and queue congestion occurs, the priority queue ensures that the G1 critical closed loop group completes at least one injection attempt in each baseline period and ensures that the drop rate of the G1 critical closed loop group does not exceed the configurable upper limit.

[0042] As a further technical solution, the log reproduction in step S9 also includes trigger snapshot information: when the risk value Rk exceeds the trigger condition or a mode switch occurs, the waveforms of key closed-loop variables, link delay sequences, queue length change curves, and injection action records within at least one control cycle window before and after the switch time are automatically recorded for subsequent reproduction of delay jitter and strategy response processes during the working condition switch.

[0043] Example 2 Based on the fault diagnosis method for the davit mechanical system disclosed in Embodiment 1, this embodiment discloses a synchronous bridging system for gas turbine simulation and periodic control, including: The first module is configured to create a task table, set a baseline period T0 and an additional task period Ti=ni·T0. The task table includes at least a task identifier, ni, a set of variables, variable grouping, a reconstruction strategy, an expiration period parameter, and a risk threshold parameter. The second module is configured for data acquisition and labeling, which obtains data frames from the variable step size simulation model, including at least timestamp ts, variable identifier, variable value, and quality identifier. The third module is configured for closed-loop link and link information acquisition. It transmits S2 data frames through a closed-loop link of simulation model-OPC-communication station-controller-communication station-OPC-simulation model and collects link information. This task table module divides variables into three groups: G1 critical closed-loop group, G2 slow variable group, and G3 monitoring group. The G1 critical closed-loop group has 200-300 points, and the G1 target injection cycle is... The injection cycle for targets G2 and G3 is no less than the OPC refresh cycle. , The fourth module is configured for online latency and jitter estimation. Based on the link information of S3, it estimates the latency τk and jitter jk in each baseline or task cycle. The fifth module is configured for hierarchical resampling and reconstruction: the variables of S2 are grouped and hierarchically according to the task table of S1, and combined with τk and jk of S4, the reconstructed into a fixed periodic sequence aligned with T0 or Ti. The reconstruction includes preservation, interpolation or prediction, and the strategy is switched online with τk and jk. The sixth module is configured for semantic consistency and validity period verification. It performs semantic consistency processing on the key closed-loop variables after S5 reconstruction, including unit / range mapping, validity period verification, and nonlinear consistency. The seventh module, configured with risk monitoring and stability constraints, calculates the risk value Rk based on the variables processed by S6 and τk and jk from S4. When Rk meets the triggering condition, it switches to conservative mode. Simultaneously, Rk adjusts the reconstruction strategy and injection priority of S5 and applies stability constraints to the reconstructed output of S5. The risk monitoring and strategy switching module employs a mechanism with hysteresis and dwell time. A state machine includes at least a normal state, a conservative state, and a recovery state.

[0044] The eighth module, configured for priority injection and feedback, injects the data reconstructed by S5, standardized by S6, and constrained by S7 into the controller using a priority mechanism, and feeds the controller output back to the simulation model. The priority injection and feedback module sets a worst-case injection delay upper limit for the G1 critical closed-loop group. When the injection delay exceeds the upper limit, a conservative mode is triggered and a retain-reconstruction and slope limit constraint is enforced. When the communication station forwarding point table size is 20,000 to 30,000 points, the priority injection and backhaul module sets the maximum number of injection points per base period for low priority points. and only when the variable changes beyond a threshold An update is injected when the quality label changes, and a fallback update cycle is set. .

[0045] The ninth module is configured to record reproduction logs, including τk and jk of S4, the task table version of S1, the mode switching event of S7, key variable window data, and injection statistics, for the purpose of scene reproduction.

[0046] The configurable parameters are listed below:

[0047] Example 3 Thirdly, the present invention also provides a storage medium storing a computer program, which, when executed, implements a method for synchronously bridging gas turbine simulation and periodic control.

[0048] Example 4 Fourthly, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory storage medium. The computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer executes a method for synchronously bridging gas turbine simulation and periodic control.

[0049] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0050] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0051] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0052] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated modules described above can be implemented in hardware or as software functional modules.

[0053] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for synchronously bridging gas turbine simulation and periodic control, characterized in that, Includes the following steps: Step S1: Establish a task table, set the baseline period T0 and the additional task period Ti=ni·T0. The task table shall include at least the task identifier, ni, variable set, variable grouping, reconstruction strategy, validity period parameter, and risk threshold parameter. Step S2: Data acquisition and labeling. Obtain data frames from the variable step size simulation model, including at least timestamp ts, variable identifier, variable value, and quality identifier. Step S3: Link closing and link information acquisition. Transmit the data frame of S2 through the closed-loop link of simulation model-OPC-communication station-controller-communication station-OPC-simulation model, and acquire link information. Step S4: Online estimation of latency and jitter. Based on the link information in S3, estimate the latency τk and jitter jk in each baseline or task cycle. Step S5: Hierarchical resampling and reconstruction: Group and hierarchically resample the variables of S2 according to the task table of S1, and reconstruct them into a fixed periodic sequence aligned with T0 or Ti by combining τk and jk from S4. The reconstruction includes preservation, interpolation or prediction, and the strategy is switched online with τk and jk. Step S6: Semantic Consistency and Validity Verification. Perform semantic consistency processing on the key closed-loop variables after S5 reconstruction, including unit / range mapping, validity verification, and nonlinear consistency. Step S7: Risk monitoring and stability constraints. Calculate the risk value Rk based on the variables processed in S6 and τk and jk in S4. When Rk meets the triggering condition, switch to conservative mode. Simultaneously, adjust the reconstruction strategy and injection priority in S5, and apply stability constraints to the output after reconstruction in S5. Step S8: Priority injection and feedback. The data after S5 reconstruction, S6 unification, and S7 constraint are injected into the controller using a priority mechanism, and the controller output is fed back to the simulation model. Step S9: Record the reproduction log, including τk and jk of S4, the task table version of S1, the mode switching event of S7, key variable window data, and injection statistics, for scene reproduction.

2. The method according to claim 1, characterized in that, The quality identifier in step S2 includes at least one or more of the following: data validity flag, simulation model running status flag, and timestamp trust flag; when the quality identifier indicates that the data is invalid or the simulation model is in an unstable state, the corresponding data frame does not participate in the reconstruction of S5 or trigger the conservative mode.

3. The method according to claim 1, characterized in that, The link information collected in step S3 for delay estimation includes at least one of the following: simulation side transmission timestamp, OPC interface reception / forward timestamp, communication station reception / forward timestamp, controller reception timestamp, queue length information from communication station to controller, and network round-trip time measurement from OPC interface to communication station.

4. The method according to claim 1, characterized in that, The hierarchical processing in step S5 includes: assigning variables to base period tasks or additional period tasks of different multiples according to the task table, and adopting independent reconstruction strategies for the same variable under different task periods; wherein the reconstruction strategy corresponding to the base period task is controlled by the online switching of τk and jk, and the reconstruction strategy corresponding to the additional period task adopts the maintenance reconstruction or interpolation reconstruction and does not switch with τk and jk.

5. The method according to claim 1, characterized in that, The calculation of the risk value Rk in step S7 further introduces the delay τk and jitter jk estimated in S4: when τk exceeds the first threshold or jk exceeds the second threshold, the risk value Rk is increased by an additional risk term on the basis of the original value; the additional risk term increases monotonically with the degree of τk and jk exceeding the limit, and the additional risk term decays in a hysteretic manner after τk and jk return to the normal range.

6. The method according to claim 1, characterized in that, The stability constraints in step S7 include at least one of limiting the rate of change of the output variable, limiting the acceleration of change, and integral separation; and in conservative mode, the amplitude limit or rate limit value of the stability constraints is further narrowed compared to the normal mode.

7. The method according to claim 1, characterized in that, The reproduction log in step S9 also includes trigger snapshot information: when the risk value Rk exceeds the trigger condition or a mode switch occurs, the key closed-loop variable waveforms, link delay sequences, queue length change curves, and injection action records are automatically recorded within at least one control cycle window before and after the switch time, for the purpose of reproducing the delay jitter and strategy response process during the working condition switch process afterward.

8. A synchronous bridging system for gas turbine simulation and periodic control, characterized in that, include: The first module is configured to create a task table, set a baseline period T0 and an additional task period Ti=ni·T0, and the task table includes at least a task identifier, ni, a set of variables, variable grouping, a reconstruction strategy, a validity period parameter, and a risk threshold parameter. The second module is configured for data acquisition and labeling, which obtains data frames from the variable step size simulation model, including at least timestamp ts, variable identifier, variable value, and quality identifier. The third module is configured to perform closed-loop link and link information acquisition. It transmits S2 data frames through the closed-loop link of simulation model—OPC—communication station—controller—communication station—OPC—simulation model and acquires link information. The fourth module is configured for online latency and jitter estimation. Based on the link information of S3, it estimates the latency τk and jitter jk in each baseline or task cycle. The fifth module is configured for hierarchical resampling and reconstruction: the variables of S2 are grouped and hierarchically according to the task table of S1, and combined with τk and jk of S4, the reconstructed into a fixed periodic sequence aligned with T0 or Ti. The reconstruction includes preservation, interpolation or prediction, and the strategy is switched online with τk and jk. The sixth module is configured for semantic consistency and validity period verification. It performs semantic consistency processing on the key closed-loop variables after S5 reconstruction, including unit / range mapping, validity period verification, and nonlinear consistency. The seventh module is configured with risk monitoring and stability constraints. It calculates the risk value Rk based on the variables processed by S6 and τk and jk of S4. When Rk meets the triggering conditions, it switches to conservative mode. At the same time, Rk adjusts the reconstruction strategy and injection priority of S5 and applies stability constraints to the output after reconstruction of S5. The eighth module is configured for priority injection and feedback. It injects the data after S5 reconstruction, S6 unification, and S7 constraint into the controller using a priority mechanism, and feeds the controller output back to the simulation model. The ninth module is configured to record reproduction logs, including τk and jk of S4, the task table version of S1, the mode switching event of S7, key variable window data, and injection statistics, for the purpose of scene reproduction.

9. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer performs the gas turbine simulation and periodic control synchronous bridging method as described in any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the synchronous bridging method for gas turbine simulation and periodic control as described in any one of claims 1-7.

Images