Ventilation valve control system with pressure regulation function for generator set

Abstract

The invention relates to the field of control systems, and discloses a vent valve control system with a pressure regulation function for a generator set, which is used for breaking through the limitation of static state, single target and passive response of a traditional vent valve control method. Comprising the following steps: acquiring pressure, flow velocity and valve opening data of a ventilation system in real time, generating a dimensionless parameter set for dynamic calibration, integrating parameter uncertainty disturbance terms to form a hybrid prediction model, solving an optimal valve opening instruction sequence by adopting a multi-target robust optimization algorithm, and designing a layered safety control mechanism. The system safety is guaranteed through dual-threshold monitoring and emergency intervention, redundant sensor data fusion and parameter reverse optimization are introduced, and fault tolerance and model self-adaption are achieved. According to the method, through collaborative innovation of dynamic modeling, robust optimization and intelligent fault tolerance, the control precision, the anti-interference capability, the safety and the reliability of the ventilation system of the generator set are remarkably improved, and the method is suitable for complex and changeable industrial operation scenes.

Description

Technical Field

[0001] This invention relates to the field of control systems, and more particularly to a ventilation valve control system for generator sets with pressure regulation. Background Technology

[0002] With the rapid development of modern industrial technology, generator sets, as core equipment for energy conversion, directly affect the stability and economy of power supply through their operational safety and efficiency. The ventilation system, as a crucial component of the generator set, plays a vital role in regulating the internal temperature and pressure of the unit, ensuring that the equipment operates under suitable conditions. Especially under complex conditions of high pressure, high temperature, and high flow rate, precise control of ventilation valves is of great significance for maintaining stable system pressure, preventing equipment overheating and damage, and ensuring personnel safety.

[0003] Traditional ventilation valve control methods are mostly based on simple PID (proportional-integral-derivative) control strategies, relying on fixed parameter settings. This makes them ill-suited to adapting to dynamic changes caused by factors such as load variations, ambient temperature fluctuations, and equipment aging during generator operation. Furthermore, traditional methods often neglect in-depth analysis of pressure wave propagation characteristics and lack effective modeling and handling of system uncertainties. This results in poor control performance when facing sudden disturbances or extreme operating conditions, and may even lead to system instability or safety accidents.

[0004] Therefore, we propose a pressure-regulated ventilation valve control system for generator sets to solve the above problems. Summary of the Invention

[0005] This invention provides a ventilation valve control system for generator sets with pressure regulation, which overcomes the limitations of traditional ventilation valve control methods that are static, single-objective, and passive in response.

[0006] The first aspect of this invention provides a pressure-regulated ventilation valve control system for generator sets, comprising: an acquisition module for acquiring actual pressure values, gas velocity values, and valve opening values ​​of a ventilation system; dividing the actual pressure value of the ventilation system by a preset rated pressure value to generate dimensionless pressure parameters; dividing the gas velocity value by a preset sound velocity value to generate dimensionless velocity parameters; and dividing the valve opening value by a preset maximum valve opening value to generate dimensionless opening parameters, thereby obtaining a real-time dimensionless parameter set; and a construction module for constructing a one-dimensional nonlinear wave model describing pressure wave propagation based on the real-time dimensionless parameter set; generating a set of disturbance terms with well-defined boundaries according to a preset uncertainty range of model parameters; and combining the nonlinear wave model with the disturbance... The set of terms is combined into a hybrid prediction model; the intervention module is used to perform spatiotemporal discretization processing on the hybrid prediction model, predict the pressure distribution of the ventilation system in the future period, with the goal of minimizing pressure tracking deviation and valve action amplitude, and solve the valve opening command sequence under the following constraints: the predicted pressure value is always within the pressure safety threshold range; the valve opening change rate does not exceed the preset upper limit value; all possible disturbance terms in the hybrid prediction model are valid; and the optimal valve opening command sequence is obtained; the allocation module is used to extract the first opening command value of the optimal valve opening command sequence, drive the valve actuator to act, collect the actual pressure value of the ventilation system in real time, and if it exceeds the preset physical pressure safety range, trigger the emergency control protocol to forcibly switch the valve opening to the preset safe opening value.

[0007] Optionally, in the first implementation of the first aspect of the present invention, the method includes: matching the corresponding rated pressure value from the historical operating database according to the current generator load rate as the dynamic rated pressure value; calculating the sound velocity value under the current operating condition according to real-time data from the gas temperature sensor as the dynamic sound velocity value; correcting the preset maximum valve opening value according to the valve aging coefficient model to generate the dynamic maximum valve opening value; obtaining a calibrated set of dynamic reference values; dividing the actual pressure value by the dynamic rated pressure value to generate a dimensionless pressure parameter; dividing the gas velocity value by the dynamic sound velocity value to generate a dimensionless velocity parameter; dividing the valve opening value by the dynamic maximum valve opening value to generate a dimensionless opening parameter; generating a real-time dimensionless parameter set; verifying whether the dimensionless pressure parameter is within the preset theoretical range; verifying whether the dimensionless velocity parameter is less than the preset critical value; if any parameter exceeds the limit, the current value is replaced by the effective parameter value of the previous control cycle.

[0008] Optionally, in the second implementation of the first aspect of the present invention, the method includes: dividing the ventilation duct into a preset number of discrete units; establishing a pressure-velocity coupled dynamic equation in each unit based on the law of conservation of gas mass and momentum; initializing the pressure and velocity states of each unit according to a real-time dimensionless parameter set to generate a spatially discretized pressure fluctuation model; mapping the independent uncertainty range of each parameter to a bounded closed convex set in a multidimensional space according to preset nominal values ​​of model parameters and preset uncertainty ranges of parameters; dynamically scaling the boundary of the convex set according to the current operating stage of the generator set to generate a multidimensional disturbance boundary set; embedding the multidimensional disturbance boundary set into the state equation of the pressure fluctuation model as an additional term; adding a robust satisfaction verification interface for pressure safety constraints to the model output layer to generate a hybrid prediction model.

[0009] Optionally, in the third implementation of the first aspect of the present invention, the method includes: dividing the future prediction period into equally spaced time steps based on the hybrid prediction model and the real-time dimensionless parameter set; solving the spatial discrete unit state of the hybrid prediction model within each time step to generate a spatiotemporal discrete pressure distribution dataset; constructing a multi-disturbance robust optimization problem based on the spatiotemporal discrete pressure distribution dataset, a preset pressure safety threshold range, and a preset upper limit value for the valve opening change rate: the optimization objective is to minimize the cumulative deviation between the pressure prediction value and the target pressure plus the weighted penalty of the valve opening change amplitude, thereby generating a robust feasible valve opening sequence; based on the robust feasible valve opening sequence, extracting the opening command of the first control cycle in the sequence as the current execution value, caching the remaining sequence for optimization hot start of the next control cycle, and generating the optimal valve opening command sequence.

[0010] Optionally, in the fourth implementation of the first aspect of the present invention, the method includes: based on the first opening command value of the optimal valve opening command sequence, converting the opening command value into a drive current command according to the dynamic response model of the valve actuator, superimposing a preset mechanical hysteresis compensation amount, and generating a compensated valve drive signal; and based on the real-time collected actual pressure value of the ventilation system and a preset pressure safety warning threshold range. Preset pressure safety emergency threshold range If the actual pressure value exceeds the warning threshold but does not reach the emergency threshold, a pressure warning sign is generated; if the actual pressure value exceeds the emergency threshold, a pressure over-limit emergency sign is generated, and a safety monitoring status sign is obtained. Based on the safety monitoring status sign and the compensated valve drive signal, when the sign is in a warning state: while maintaining the current valve action, a pressure approaching the threshold alarm is fed back, requesting a re-optimization of the instruction sequence; when the sign is in an emergency state: the compensated valve drive signal is immediately interrupted, switched to the preset safe opening value, and the valve is locked to the safe mode; the final valve control signal is generated.

[0011] Optionally, in the fifth implementation of the first aspect of the present invention, a feedback module is further included for the fault tolerance and parameter adaptation steps: based on the redundant pressure values ​​detected by the pressure sensor group and the redundant flow velocity values ​​detected by the flow meter group, the standard deviation of the data of the same type of sensor is calculated. If it exceeds the preset tolerance threshold, it is marked as a suspicious sensor; the consensus pressure value and consensus flow velocity value are extracted from the effective sensors using the median filtering method; verified reliable data is generated; based on the verified reliable data and the hybrid prediction model, the actual pressure / flow velocity dynamic response is compared with the model prediction value to generate a residual sequence; based on the residual minimization criterion, the parameter correction amount of friction coefficient and specific heat ratio is optimized in reverse; updated nominal values ​​of model parameters are generated; based on the suspicious sensor marking and residual sequence, if the proportion of suspicious sensors is ≥50% or the residual continues to exceed the limit, the system switches to conservative control mode: the disturbance boundary is expanded to 150% to reduce the pressure safety threshold range; otherwise, the standard control mode is maintained, and an adaptive control mode instruction is generated.

[0012] Beneficial effects: By dynamically calibrating rated pressure, sound velocity, and maximum valve opening, this invention can automatically adjust control parameters according to the real-time operating status of the generator set, significantly improving the dynamic adaptability and accuracy of the control system. Compared with traditional fixed parameter control methods, this invention can more effectively cope with complex and changing operating conditions, ensuring the stable operation of the ventilation system. By incorporating model parameter uncertainty into the control strategy and constructing a hybrid prediction model by generating a set of disturbance terms with well-defined boundaries, the system's ability to handle uncertainty is effectively improved. This enables the control system to maintain stable control performance when facing uncertainties such as parameter fluctuations and external disturbances, thereby reducing the risk of system instability. By constructing a one-dimensional nonlinear wave model and performing spatiotemporal discretization, the propagation characteristics of pressure waves in the ventilation system can be accurately predicted, enabling accurate prediction of pressure distribution in future time periods. This provides a scientific basis for optimizing valve opening command sequences, helps to adjust control strategies in advance, and avoids the adverse effects of sudden pressure changes on the system. When the actual pressure value exceeds the safety threshold or the sensor data is abnormal, it can respond quickly and take corresponding measures to ensure system safety, significantly improve the reliability and safety of the control system, and reduce the probability of accidents. By using a parameter adaptive adjustment mechanism based on the residual minimization criterion, the model parameters can be continuously optimized to ensure the long-term stable operation of the control system. This allows the control system to automatically adjust to factors such as generator aging and changes in environmental conditions, maintaining optimal control performance and extending the service life of the equipment. Attached Figure Description

[0013] Figure 1This is a schematic diagram of an embodiment of a generator set ventilation valve control system with pressure regulation according to the present invention; Figure 2 This is a schematic diagram of an embodiment of a generator set ventilation valve control device with pressure regulation according to an embodiment of the present invention. Detailed Implementation

[0014] This invention provides a pressure-regulated ventilation valve control system for generator sets, overcoming the limitations of traditional ventilation valve control methods that are static, single-target, and passively responsive. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 One embodiment of the generator set ventilation valve control system with pressure regulation in this invention includes: 101. Real-time data acquisition and dimensionless parameter generation: Input the actual pressure value of the ventilation system detected by the pressure sensor, the gas velocity value detected by the flow meter, and the valve opening value detected by the valve opening sensor; divide the actual pressure value by the preset rated pressure value to generate a dimensionless pressure parameter; divide the gas velocity value by the preset sound velocity value to generate a dimensionless velocity parameter; divide the valve opening value by the preset maximum valve opening value to generate a dimensionless opening parameter; generate a real-time dimensionless parameter set consisting of the dimensionless pressure parameter, dimensionless velocity parameter, and dimensionless opening parameter.

[0016] It is understood that the executing entity of this invention can be a ventilation valve control system for a generator set with pressure regulation, or it can be a terminal or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0017] It should be noted that the input includes the rated pressure value, sound velocity value, and maximum valve opening value stored in the historical operation database; based on the current generator load rate, the corresponding rated pressure value is matched from the historical operation database as the dynamic rated pressure value; based on real-time data from the gas temperature sensor, the sound velocity value under the current operating condition is calculated as the dynamic sound velocity value; based on the valve aging coefficient model, the preset maximum valve opening value is corrected to generate the dynamic maximum valve opening value; and a calibrated set of dynamic reference values ​​(dynamic rated pressure value, dynamic sound velocity value, and dynamic maximum valve opening value) is generated.

[0018] The system inputs the actual pressure value of the ventilation system detected in real time by the pressure sensor; the gas velocity value detected in real time by the flow meter; the valve opening value detected in real time by the valve opening sensor; and a dynamic reference value set. It then divides the actual pressure value by the dynamic rated pressure value to generate a dimensionless pressure parameter; divides the gas velocity value by the dynamic sound velocity value to generate a dimensionless velocity parameter; divides the valve opening value by the dynamic maximum valve opening value to generate a dimensionless opening parameter; and finally generates a real-time dimensionless parameter set consisting of the dimensionless pressure parameter, dimensionless velocity parameter, and dimensionless opening parameter.

[0019] Input the real-time dimensionless parameter set; verify whether the dimensionless pressure parameter is within the preset theoretical range [0.5, 1.5]; verify whether the dimensionless flow velocity parameter is less than the preset critical value of 0.8; if any parameter exceeds the limit, replace the current value with the effective parameter value of the previous control cycle; generate the verified effective real-time dimensionless parameter set.

[0020] 102. Construct a hybrid prediction model by inputting the generated real-time dimensionless parameter set; constructing a one-dimensional nonlinear wave model describing the propagation of pressure waves based on the laws of conservation of gas mass and momentum; generating a set of disturbance terms with well-defined boundaries according to the preset uncertainty range of model parameters; combining the nonlinear wave model and the set of disturbance terms into a unified prediction model; and generating a hybrid prediction model with uncertainty boundaries.

[0021] It should be noted that the input includes a real-time dimensionless parameter set (including dimensionless pressure parameters and dimensionless velocity parameters); ventilation system duct topology data; dividing the ventilation duct into a preset number of discrete units; establishing pressure-velocity coupled dynamic equations in each unit based on the law of conservation of gas mass and momentum; initializing the pressure and velocity states of each unit according to the real-time dimensionless parameter set; and generating a spatially discretized pressure fluctuation model (including unit state equations and initial conditions).

[0022] Input the preset nominal values ​​of model parameters (friction coefficient, gas specific heat ratio); preset parameter uncertainty range (friction coefficient deviation ±15%, specific heat ratio deviation ±5%); map the independent uncertainty range of each parameter into a bounded closed convex set in multidimensional space; dynamically scale the convex set boundary according to the current generator set operation stage (start-up / steady state / load change); generate a multidimensional disturbance boundary set (covering all possible parameter disturbance scenarios).

[0023] Input a spatially discretized pressure fluctuation model; a multi-dimensional set of perturbation boundaries; embed the set of perturbation boundaries into the state equation of the pressure fluctuation model as an additional term; add a robust satisfaction verification interface for pressure safety constraints to the model output layer; generate a hybrid prediction model (outputting pressure prediction values ​​with uncertain boundaries).

[0024] 103. Robust optimization control command generation: Input the generated hybrid prediction model; the generated real-time dimensionless parameter set; the preset pressure safety threshold range of the ventilation system; the preset upper limit of the valve opening change rate; perform spatiotemporal discretization processing on the hybrid prediction model to predict the pressure distribution of the ventilation system in future time periods; with the goal of minimizing pressure tracking deviation and valve action amplitude, solve the valve opening command sequence under the following constraints: the predicted pressure value is always within the pressure safety threshold range; the valve opening change rate does not exceed the preset upper limit; all possible disturbance terms in the hybrid prediction model are valid; generate the optimal valve opening command sequence containing N time steps.

[0025] It should be noted that the input is a hybrid prediction model; a real-time dimensionless parameter set; the future prediction period is divided into equally spaced time steps; within each time step, the spatial discrete unit state of the hybrid prediction model is solved to generate pressure distribution prediction data; and a spatiotemporal discrete pressure distribution dataset (including pressure prediction values ​​for each time step and spatial unit) is generated. Input a spatiotemporal discrete pressure distribution dataset; a preset pressure safety threshold range; a preset upper limit for the valve opening change rate; construct a multi-perturbation robust optimization problem: optimization objective: minimize the cumulative deviation between the pressure prediction value and the target pressure + a weighted penalty for the valve opening change amplitude; constraints: the pressure prediction values ​​of all time steps and spatial units are within the pressure safety threshold range; the valve opening change rate of adjacent time steps does not exceed the preset upper limit; all possible perturbation terms in the hybrid prediction model satisfy the above constraints; use a convex optimization algorithm to solve for the globally optimal solution that satisfies the constraints; generate a robust feasible valve opening sequence (covering the entire prediction period). Input a robust and feasible valve opening sequence; extract the opening command of the first control cycle in the sequence as the current execution value; cache the remaining sequence for optimized hot start in the next control cycle; generate the optimal valve opening command sequence (the first item is used for current execution, and the remaining items are cached).

[0026] 104. Valve execution and safety monitoring: Input the optimal valve opening command sequence; extract the first opening command value of the command sequence and drive the valve actuator to act; collect the actual pressure value of the ventilation system in real time. If it exceeds the preset physical pressure safety range, trigger the emergency control protocol and force the valve opening to switch to the preset safe opening value.

[0027] It should be noted that the first opening command value of the optimal valve opening command sequence is input; the opening command value is converted into a drive current command according to the dynamic response model of the valve actuator; a preset mechanical hysteresis compensation is superimposed to generate an anti-hysteresis valve drive signal; and a compensated valve drive signal is generated (sent to the actuator). Input the actual pressure value of the ventilation system collected in real time by the pressure sensor; Preset pressure safety warning threshold range ; Preset pressure safety emergency threshold range ; If the actual pressure value exceeds the warning threshold but does not reach the emergency threshold, a pressure warning sign is generated; if the actual pressure value exceeds the emergency threshold, a pressure over-limit emergency sign is generated; a safety monitoring status sign is generated (including warning / emergency status and over-limit range). Input safety monitoring status indicator; compensated valve drive signal; When the warning status is indicated: while maintaining the current valve operation, a pressure approaching the threshold alarm is fed back, requesting a re-optimization of the instruction sequence; When an emergency state is identified: immediately interrupt the compensated valve drive signal, switch to the preset safe opening value (fully open / fully closed), and lock the valve to the safe mode; generate the final valve control signal (normal drive signal or safe opening value).

[0028] 105. Fault tolerance and parameter adaptation steps; It should be noted that the input pressure sensor group detects redundant pressure values; the flow meter group detects redundant flow rate values. Calculate the standard deviation of data from similar sensors. If the standard deviation exceeds the preset tolerance threshold, the sensor is marked as a suspicious sensor. Use the median filtering method to extract consensus pressure and consensus flow rate values ​​from valid sensors. Generate reliable data (pressure value, flow rate value) after verification; Input validated reliable data; hybrid prediction model; The actual pressure / flow velocity dynamic response is compared with the model prediction to generate a residual sequence; Based on the residual minimization criterion, the parameter correction amounts for friction coefficient and specific heat ratio are optimized in reverse. Generate updated nominal values ​​for model parameters (input reconstructed model).

[0029] Input suspicious sensor tags; residual sequences; If the number of suspicious sensors accounts for ≥50% or the residual continues to exceed the limit, switch to conservative control mode: expand the disturbance boundary to 150%; reduce the pressure safety threshold range to ±90%; otherwise, maintain the standard control mode; generate adaptive control mode instructions.

[0030] In this embodiment of the invention, a dynamic reference value set is generated by dynamically calibrating the rated pressure value, sound velocity value, and maximum valve opening value, combined with real-time load rate and gas temperature, to replace traditional fixed parameters. This enables the system to automatically adapt to different operating conditions, significantly improving control accuracy and stability, and reducing the risk of control failure due to parameter mismatch. A one-dimensional nonlinear fluctuation model is constructed to describe pressure wave propagation, and parameter uncertainty disturbance terms are integrated to form a hybrid prediction model covering all possible disturbance scenarios. Through uncertainty quantification and robust modeling, the system can accurately predict pressure distribution and cope with parameter fluctuations, significantly improving anti-interference capability and ensuring the effectiveness of the control strategy under complex operating conditions. The goal is to minimize pressure tracking deviation and valve action amplitude, while simultaneously meeting... By satisfying the constraints of pressure safety threshold and valve opening change rate, the globally optimal valve opening sequence is solved. Under the premise of ensuring system safety, control costs are optimized, equipment lifespan is extended, energy consumption is reduced, and a balance between control performance and economy is achieved. A dual-threshold monitoring mechanism is designed, combined with an emergency intervention protocol, to form a hierarchical safety control system. Timely response to pressure anomalies prevents system instability or equipment damage, significantly improving safety and reliability and reducing accident risks. Fault tolerance and model adaptation are achieved through redundant sensor data fusion and parameter reverse optimization. Control modes are dynamically switched based on sensor status and residual sequences. Even in the event of sensor failure or model mismatch, the system can maintain stable operation, reducing downtime and maintenance time, and improving the long-term reliability and robustness of the system. Dynamic dimensionless and hybrid predictive models enable the system to accurately match changes in operating conditions; hierarchical safety control and emergency intervention mechanisms effectively avoid the risk of pressure exceeding limits; adaptive fault tolerance and parameter optimization reduce equipment failure and manual intervention requirements; multi-objective optimization control reduces frequent valve actions and mechanical wear.

[0031] above Figure 1 The pressure-regulated ventilation valve control system for generator sets in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The pressure-regulated ventilation valve control device for generator sets in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0032] Figure 2This is a schematic diagram of a pressure-regulated generator set ventilation valve control device 200 provided in an embodiment of the present invention. The pressure-regulated generator set ventilation valve control device 200 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 210 (e.g., one or more processors) and a memory 220, and one or more storage media 230 (e.g., one or more mass storage devices) storing application programs 233 or data 232. The memory 220 and storage media 230 can be temporary or persistent storage. The program stored in the storage media 230 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the pressure-regulated generator set ventilation valve control device 200. Furthermore, the processor 210 may be configured to communicate with the storage media 230 and execute the series of instruction operations in the storage media 230 on the pressure-regulated generator set ventilation valve control device 200.

[0033] The generator set ventilation valve control device 200 with pressure regulation may also include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input / output interfaces 260, and / or one or more operating systems 231, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 2 The illustrated structure of a pressure-regulated generator set ventilation valve control device does not constitute a limitation on pressure-regulated generator set ventilation valve control devices, which may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0034] The present invention also provides a pressure-regulated ventilation valve control device for generator sets, the pressure-regulated ventilation valve control device for generator sets including a memory and a processor, the memory storing computer-readable instructions, and when the computer-readable instructions are executed by the processor, causing the processor to perform the steps of the pressure-regulated ventilation valve control system for generator sets described in the above embodiments.

[0035] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the pressure-regulated generator set ventilation valve control system.

[0036] 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.

[0037] If the integrated unit is implemented as a software functional unit 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 the present 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 described in the various embodiments of the present 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.

[0038] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A ventilation valve control system for a generator set with pressure regulation, characterized in that, The generator set ventilation valve control system with pressure regulation includes: The acquisition module is used to acquire the actual pressure value, gas velocity value, and valve opening value of the ventilation system. It divides the actual pressure value of the ventilation system by the preset rated pressure value to generate dimensionless pressure parameters, divides the gas velocity value by the preset sound velocity value to generate dimensionless velocity parameters, and divides the valve opening value by the preset maximum valve opening value to generate dimensionless opening parameters, thus obtaining a real-time dimensionless parameter set. The construction module is used to construct a one-dimensional nonlinear wave model describing the propagation of pressure waves based on a real-time dimensionless parameter set, generate a set of disturbance terms with clear boundaries according to a preset uncertainty range of model parameters, and combine the nonlinear wave model and the set of disturbance terms into a hybrid prediction model. The intervention module is used to perform spatiotemporal discretization processing on the hybrid prediction model, predict the pressure distribution of the ventilation system in the future period, and minimize the pressure tracking deviation and valve action amplitude. Under the following constraints, the valve opening command sequence is solved: the predicted pressure value is always within the pressure safety threshold range; the valve opening change rate does not exceed the preset upper limit value; all possible disturbance terms in the hybrid prediction model are valid; and the optimal valve opening command sequence is obtained. The allocation module is used to extract the first opening command value of the optimal valve opening command sequence, drive the valve actuator to act, and collect the actual pressure value of the ventilation system in real time. If it exceeds the preset physical pressure safety range, the emergency control protocol is triggered to forcibly switch the valve opening to the preset safe opening value.

2. The generator set ventilation valve control system with pressure regulation according to claim 1, characterized in that, include: Based on the current generator load rate, the corresponding rated pressure value is matched from the historical operation database as the dynamic rated pressure value. Based on the real-time data of the gas temperature sensor, the sound velocity value under the current operating condition is calculated as the dynamic sound velocity value. Based on the valve aging coefficient model, the preset maximum valve opening value is corrected to generate the dynamic maximum valve opening value. The calibrated dynamic reference value set is obtained. Divide the actual pressure value by the dynamic rated pressure value to generate a dimensionless pressure parameter; divide the gas velocity value by the dynamic sound velocity value to generate a dimensionless velocity parameter; divide the valve opening value by the dynamic maximum valve opening value to generate a dimensionless opening parameter; and generate a real-time dimensionless parameter set. Verify whether the dimensionless pressure parameter is within the preset theoretical range, and verify whether the dimensionless flow velocity parameter is less than the preset critical value. If either parameter exceeds the limit, replace the current value with the effective parameter value from the previous control cycle.

3. The generator set ventilation valve control system with pressure regulation according to claim 2, characterized in that, include: The ventilation duct is divided into a predetermined number of discrete units. Based on the law of conservation of gas mass and momentum, a pressure-velocity coupled dynamic equation is established in each unit. The pressure and velocity states of each unit are initialized according to the real-time dimensionless parameter set to generate a spatially discretized pressure fluctuation model. Based on the preset nominal values ​​of model parameters and the preset uncertainty range of parameters, the independent uncertainty range of each parameter is mapped to a bounded closed convex set in a multidimensional space. According to the current operating stage of the generator set, the boundary of the convex set is dynamically scaled to generate a multidimensional disturbance boundary set. The multidimensional perturbation boundary set is embedded as an additional term into the state equation of the pressure fluctuation model, and a robust satisfaction verification interface for pressure safety constraints is added to the model output layer to generate a hybrid prediction model.

4. The generator set ventilation valve control system with pressure regulation according to claim 3, characterized in that, include: Based on the hybrid prediction model and the real-time dimensionless parameter set, the future prediction period is divided into equally spaced time steps. Within each time step, the spatial discrete unit state of the hybrid prediction model is solved to generate a spatiotemporal discrete pressure distribution dataset. Based on the spatiotemporal discrete pressure distribution dataset, the preset pressure safety threshold range, and the preset upper limit of valve opening change rate, a multi-perturbation robust optimization problem is constructed: the optimization objective is to minimize the cumulative deviation between the pressure prediction value and the target pressure plus the weighted penalty of the valve opening change rate, and generate a robust feasible valve opening sequence. Based on the robust feasible valve opening sequence, the opening command of the first control cycle in the sequence is extracted as the current execution value, and the remaining sequence is cached for optimized hot start in the next control cycle, thereby generating the optimal valve opening command sequence.

5. The generator set ventilation valve control system with pressure regulation according to claim 4, characterized in that, include: Based on the first opening command value of the optimal valve opening command sequence, and according to the dynamic response model of the valve actuator, the opening command value is converted into a drive current command, and a preset mechanical hysteresis compensation is added to generate a compensated valve drive signal. Based on the real-time collected actual pressure value of the ventilation system and the preset pressure safety warning threshold range Preset pressure safety emergency threshold range ; If the actual pressure value exceeds the warning threshold but does not reach the emergency threshold, a pressure warning sign is generated; if the actual pressure value exceeds the emergency threshold, a pressure over-limit emergency sign is generated, and a safety monitoring status sign is obtained. Based on the safety monitoring status indicator and the compensated valve drive signal, when the indicator is in the early warning state: while maintaining the current valve action, a pressure approaching threshold alarm is fed back, requesting re-optimization of the instruction sequence; when the indicator is in the emergency state: the compensated valve drive signal is immediately interrupted, switched to the preset safe opening value, and the valve is locked to the safe mode; the final valve control signal is generated.

6. The generator set ventilation valve control system with pressure regulation according to claim 5, characterized in that, It also includes a feedback module for fault tolerance and parameter adaptation steps: Based on the redundant pressure values ​​detected by the pressure sensor group and the redundant flow velocity values ​​detected by the flow meter group, the standard deviation of the data of the same type of sensor is calculated. If it exceeds the preset tolerance threshold, it is marked as a suspicious sensor. The median filtering method is used to extract the consensus pressure value and consensus flow velocity value from the valid sensors. Reliable data after verification is generated. Based on validated reliable data and a hybrid prediction model, the actual pressure / flow velocity dynamic response is compared with the model prediction values ​​to generate a residual sequence; based on the residual minimization criterion, the parameter correction amounts for friction coefficient and specific heat ratio are optimized in reverse; and updated nominal values ​​of model parameters are generated. Based on the suspicious sensor markers and residual sequences, if the number of suspicious sensors accounts for ≥50% or the residuals continue to exceed the limit, switch to conservative control mode: expand the disturbance boundary to 150% and reduce the pressure safety threshold range; otherwise, maintain the standard control mode and generate adaptive control mode instructions.

Images