Anti-surge control method and system for centrifugal compressor and storage medium

Abstract

The invention discloses an anti-surge control method and system of a centrifugal compressor and a storage medium, and relates to the technical field of anti-surge control of centrifugal compressors, a reference prediction operation point is predicted through a compressor physical model, a future residual error is predicted through a data driving residual error compensation model, and the anti-surge control of the centrifugal compressor is achieved. And the reference prediction operation point and the future residual error are fused to obtain a high-precision fusion prediction result, so that high-precision dynamic prediction of the surge boundary is realized. The dynamic surge margin is calculated in advance, the hierarchical intervention control strategy is executed based on the dynamic surge margin, conversion from passive response to active prevention is achieved, and surge prevention timeliness and safety are remarkably improved. And through dynamic and accurate margin management, the compressor can run closer to a high-efficiency area on the premise of ensuring safety, so that the overall energy efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of anti-surge control technology for centrifugal compressors, and in particular to an anti-surge control method, system, and storage medium for centrifugal compressors. Background Technology

[0002] Magnetic levitation centrifugal refrigeration compressors, with their advantages of being oil-free, highly efficient, and low-vibration, are widely used in large-scale central air conditioning and data center cooling. The core component of this type of compressor is the high-speed rotating impeller; however, under low-flow conditions, the impeller is highly susceptible to surge. Surge is a severe unsteady flow instability phenomenon, accompanied by periodic backflow of airflow and violent pulsations in system pressure and flow, which can induce strong mechanical vibrations. In severe cases, it can lead to failure of the magnetic levitation bearing, impeller damage, or even complete unit failure, posing a significant threat to the safety and reliability of the system.

[0003] Existing anti-surge technologies mainly employ a static margin control method based on compressor characteristic diagrams. The core logic is as follows: by experimentally determining the compressor's surge line, a safety margin of 10%-15% is shifted towards higher flow rates to form an anti-surge control line. During compressor operation, the control system monitors its operating point in real time. Once the operating point touches or crosses this control line, a bypass valve or hot gas bypass valve is quickly opened to return some high-pressure gas to the compressor inlet, thereby increasing the flow rate and moving the operating point away from the surge zone.

[0004] However, the existing anti-surge technology has the following significant drawbacks: 1. Severe efficiency loss: To ensure absolute safety, the fixed safety margin is usually set very conservatively, which means that the compressor cannot operate in the high-efficiency zone closest to the surge boundary most of the time, resulting in significant energy waste.

[0005] 2. Poor adaptability: The surge boundary of a compressor is not constant; it drifts with factors such as intake air temperature / pressure, condensing pressure, equipment aging and wear, and blade fouling. Fixed control lines cannot adapt to this dynamic change, potentially resulting in excessive margin (wasting energy) under some operating conditions, while insufficient margin (posing safety risks) under others.

[0006] 3. Response lag: This method is a passive response control; the control action is only triggered when the operating point actually reaches the control line. For dynamic disturbances, the system may momentarily enter surge before the control action fully takes effect, making early prevention impossible.

[0007] Therefore, a predictive anti-surge technology is needed that can accurately predict surge boundaries, dynamically adjust safety margins, and intervene in advance to address the shortcomings of existing technologies. Summary of the Invention

[0008] In order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a method, system and storage medium for anti-surge control of centrifugal compressors, aiming to solve the problems of severe efficiency loss, poor adaptability and lag in the existing anti-surge technology.

[0009] The technical solution adopted by this invention to solve its technical problem is: a method for anti-surge control of a centrifugal compressor, comprising the following steps: S1: Real-time acquisition of compressor operating status parameters; S2: Input the operating status parameters into the compressor physical model and output the baseline predicted operating point; S3: Calculate the historical residual sequence and predict future residuals using a data-driven residual compensation model; S4: Fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; S5: Based on the high-precision fusion prediction results, deduce the dynamic surge boundary and calculate the dynamic surge margin; S6: Implement a graded intervention control strategy based on the dynamic surge margin.

[0010] As a further improvement of the present invention: In step S1, the operating status parameters include compressor speed, inlet guide vane opening, inlet and outlet pressure, inlet and outlet temperature, flow rate, motor power, and shaft vibration signal. By real-time acquisition of multi-dimensional parameters including compressor speed, inlet guide vane opening, inlet and outlet pressure, inlet and outlet temperature, flow rate, motor power, and shaft vibration signal, a comprehensive and synchronous perception of the compressor's operating status is achieved. This provides an accurate and rich input foundation for subsequent models, making the calculation of the compressor physical model more closely match actual working conditions, and the data-driven residual compensation model more sensitive to error capture.

[0011] As a further improvement of the present invention: in step S2, the physical model of the compressor is constructed based on the first principles of fluid dynamics and thermodynamics.

[0012] As a further improvement of the present invention: In step S3, the historical residual sequence is obtained by comparing the historical baseline predicted output value of the compressor physical model with the actual measurement value of the sensor during the same period; the data-driven residual compensation model is a Long Short-Term Memory (LSTM) network or a Gated Recurrent Unit (GRU). By comparing the historical baseline predicted output value of the compressor physical model with the actual measurement value of the sensor during the same period to obtain the historical residual sequence, the residual data truly reflects the systematic errors and dynamic disturbances that the compressor physical model fails to cover. By selecting neural networks with temporal memory capabilities, such as LSTM networks or GRUs, as the data-driven residual compensation model, it can effectively learn and predict the complex patterns of residual evolution over time. The combination of the two enables the data-driven residual compensation model to accurately correct model mismatch caused by equipment degradation, operating condition drift, etc., thereby significantly improving the tracking accuracy and adaptive capability of the actual compressor dynamic characteristics.

[0013] As a further improvement of the present invention: In step S4, the fusion is a linear addition, and the fusion formula is: High-precision fusion prediction result = Baseline prediction running point + Future residual. By using the linear addition of the baseline prediction running point and the future residual for fusion, the entire prediction process has extremely high computational efficiency and system real-time performance, significantly reducing the computational load on the controller and ensuring rapid response in prediction and decision-making in industrial real-time control systems.

[0014] As a further improvement of the present invention: In step S5, the calculation formula for the dynamic surge margin is: Dynamic surge margin = (High-precision fusion prediction result - Dynamic surge boundary) / Dynamic surge boundary × 100%. By defining the dynamic surge margin as (High-precision fusion prediction result - Dynamic surge boundary) / Dynamic surge boundary × 100%, a standardized and quantitative assessment of safety risks is achieved. This percentage-based dynamic surge margin index is intuitive and uniform. By using the dynamic surge boundary as the calculation benchmark, the dynamic surge margin can reflect the true safe distance under the current operating conditions in real time, overcoming the rigidity of the traditional fixed margin method.

[0015] As a further improvement of the present invention: In step S6, the hierarchical intervention control strategy includes: no intervention when the dynamic surge margin is greater than the first threshold; fine-tuning control when the dynamic surge margin is less than or equal to the first threshold and greater than the second threshold; and fast protection control when the dynamic surge margin is less than or equal to the second threshold. By setting a hierarchical intervention control strategy, a leap from "one-size-fits-all" to "refined" anti-surge control is achieved. Through the division of the first and second thresholds, the system can ensure efficient operation without intervention when the surge margin is sufficient, perform fine-tuning control in advance to prevent risks when the margin narrows, and decisively perform fast protection control to ensure safety when the margin is severely insufficient. The adoption of a tiered response logic avoids the frequent malfunctions or response delays that may be caused by traditional single-threshold control. Furthermore, through the buffering of the warning zone and early intervention, the probability of the system entering the danger zone and triggering severe protection is significantly reduced, thereby achieving overall synergistic optimization of operational safety, control stability, and work efficiency.

[0016] As a further improvement of the present invention, the fine-tuning control involves slowly pre-opening the bypass valve with extremely small steps, fine-tuning the speed, or fine-tuning the guide vanes to smoothly push the future operating trajectory away from the dynamic surge boundary. By employing fine-tuning control in the early warning zone by slowly pre-opening the bypass valve with extremely small steps or fine-tuning the speed / guide vanes, proactive and smooth intervention against potential surge risks is achieved. This "small and proactive" control action avoids the system pressure and flow shocks caused by the sudden large opening and closing of the bypass valve in traditional control, significantly improving the operational stability of the refrigeration circuit and the magnetic levitation bearing.

[0017] As a further improvement of the present invention, the rapid protection control is to rapidly increase the opening of the bypass valve to ensure system safety.

[0018] By setting a protective control that triggers a rapid increase in the bypass valve opening in the danger zone, a reliable and decisive final safety barrier is built for the entire anti-surge method. When the dynamic surge margin is extremely low, this control strategy can bypass the conventional regulation logic, establish bypass flow at the fastest speed, and forcibly pull the operating point away from the surge zone.

[0019] As a further improvement of this invention: the first threshold is 15%, and the second threshold is 8%. By setting the first threshold to 15% and the second threshold to 8%, a clear, reasonable, and engineering-verified quantitative basis is provided for the graded intervention control strategy. The 15% first threshold ensures that the safety zone has ample buffer space, allowing the compressor to operate without any intervention under most normal operating conditions, thereby ensuring continuous and efficient operation. The 8% second threshold defines the bottom line where decisive protective actions must be taken, ensuring the ultimate safety of the system. The warning zone formed between the two (15% to 8%) is designed to allow the system sufficient time to mitigate risks through smooth fine-tuning, while avoiding premature or late intervention, thus achieving a delicate balance between safety, stability, and operational economy.

[0020] The present invention also provides an anti-surge control system for a centrifugal compressor, comprising: The data acquisition module is used to collect the compressor's operating status parameters in real time. The baseline prediction module is used to input the operating status parameters into the compressor physical model and output the baseline prediction operating point. The residual processing module is used to calculate the historical residual sequence and predict future residuals through a data-driven residual compensation model. The fusion prediction module is used to fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; The surge margin prediction module is used to deduce the dynamic surge boundary based on high-precision fusion prediction results and calculate the dynamic surge margin. A hierarchical control module is used to execute hierarchical intervention control strategies based on the dynamic surge margin. Through a modular system design, the data acquisition module, benchmark prediction module, residual processing module, fusion prediction module, surge margin prediction module, and hierarchical control module are integrated into a collaborative functional module, achieving a high degree of integration and automated operation of the anti-surge control process. By collaborating with the benchmark prediction module and the residual processing module, the system both follows physical laws and corrects model errors, achieving high-precision dynamic prediction of the surge boundary. Furthermore, the surge margin prediction module outputs the dynamic surge boundary in real time and calculates the dynamic surge margin, which is then executed by the hierarchical control module to implement the hierarchical intervention control strategy. This enables the entire anti-surge control system to achieve closed-loop intelligent control from state perception and risk prediction to precise intervention, significantly improving the initiative and reliability of surge protection, and also significantly optimizing the compressor's operating efficiency and overall stability through smooth and proactive control actions.

[0021] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned anti-surge control method for a centrifugal compressor.

[0022] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention predicts the baseline operating point using a compressor physical model, predicts future residuals using a data-driven residual compensation model, and fuses the baseline predicted operating point with the future residuals to obtain a high-precision fusion prediction result, achieving high-precision dynamic prediction of surge boundaries. By pre-calculating the dynamic surge margin and implementing a graded intervention control strategy based on the dynamic surge margin, a shift from "passive response" to "active prevention" is achieved, significantly improving the timeliness and safety of surge prevention. Through dynamic and precise margin management, the compressor can operate closer to its high-efficiency range while ensuring safety, thereby improving overall energy efficiency.

[0023] 2. This invention achieves a leap from a "one-size-fits-all" approach to "refined" surge control by setting a tiered intervention control strategy. Through the division of first and second thresholds, the system can ensure efficient operation without intervention when the surge margin is sufficient, perform fine-tuning control in advance to prevent risks when the margin narrows, and decisively execute rapid protection control to ensure safety when the margin is severely insufficient. The tiered response logic avoids the frequent malfunctions or response delays that may result from traditional single-threshold control. Furthermore, the buffering and early intervention of the warning zone significantly reduces the probability of the system entering the danger zone and triggering severe protection, thus achieving synergistic optimization of operational safety, control stability, and work efficiency. Attached Figure Description

[0024] Figure 1 This is a flowchart of an anti-surge control method for a centrifugal compressor according to the present invention.

[0025] Figure 2 This is a structural block diagram of an anti-surge control system for a centrifugal compressor according to the present invention. Detailed Implementation

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0027] It should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0028] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.

[0029] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0030] It should be understood that specific details are provided in the following description to facilitate a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, the system may be shown in block diagrams to avoid obscuring the example with unnecessary details. In other instances, well-known processes, structures, and techniques may be shown without unnecessary details to avoid obscuring the exemplary embodiments.

[0031] Implementation Case 1: Please see Figure 1 A method for preventing surge control in a centrifugal compressor includes the following steps: S1: Real-time acquisition of compressor operating status parameters; S2: Input the operating status parameters into the compressor physical model and output the baseline predicted operating point; S3: Calculate the historical residual sequence and predict future residuals using a data-driven residual compensation model; S4: Fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; S5: Based on the high-precision fusion prediction results, deduce the dynamic surge boundary and calculate the dynamic surge margin; S6: Implement a graded intervention control strategy based on the dynamic surge margin.

[0032] By predicting the baseline operating point using a compressor physical model and forecasting future residuals using a data-driven residual compensation model, a high-precision fusion prediction result is obtained by fusing the baseline predicted operating point with the future residuals. This achieves high-precision dynamic prediction of surge boundaries. By pre-calculating the dynamic surge margin and implementing a tiered intervention control strategy based on the dynamic surge margin, a shift from "passive response" to "active prevention" is achieved, significantly improving the timeliness and safety of surge prevention. Through dynamic and precise margin management, the compressor can operate closer to its high-efficiency range while ensuring safety, thereby improving overall energy efficiency.

[0033] In some implementations, in step S1, the operating status parameters include compressor speed, inlet guide vane opening, inlet and outlet pressure, inlet and outlet temperature, flow rate, motor power, and shaft vibration signal.

[0034] By collecting multi-dimensional parameters in real time, including compressor speed, inlet guide vane opening, inlet and outlet pressure, inlet and outlet temperature, flow rate, motor power, and shaft vibration signals, a comprehensive and synchronous perception of the compressor's operating status is achieved. This provides a precise and rich input foundation for subsequent models, making the calculations of the compressor's physical model more closely resemble actual operating conditions, and enabling the data-driven residual compensation model to capture errors more sensitively. In particular, the introduction of shaft vibration signals enhances the monitoring capability of early correlation characteristics between mechanical dynamics and flow instability, thereby improving the timeliness and reliability of surge risk identification as a whole.

[0035] In some implementations, in step S2, the compressor physical model is constructed based on the first principles of fluid dynamics and thermodynamics.

[0036] In some implementations, in step S3, the historical residual sequence is obtained by comparing the historical baseline predicted output value of the compressor physical model with the actual measurement value of the sensor during the same period; the data-driven residual compensation model is a long short-term memory network (LSTM) or a gated recurrent unit (GRU).

[0037] By comparing the historical baseline predicted output values ​​of the compressor physical model with the actual sensor measurements during the same period, a historical residual sequence is obtained. This allows the residual data to accurately reflect the systematic errors and dynamic disturbances that the compressor physical model fails to cover. By selecting neural networks with temporal memory capabilities, such as Long Short-Term Memory networks or gated recurrent units, as the data-driven residual compensation model, it can effectively learn and predict the complex patterns of residual evolution over time. The combination of these two approaches enables the data-driven residual compensation model to accurately correct model mismatches caused by equipment degradation, operating condition drift, etc., thereby significantly improving the tracking accuracy and adaptive capability of the actual compressor's dynamic characteristics.

[0038] In some implementations, in step S4, the fusion is a linear addition, and the fusion formula is: high-precision fusion prediction result = baseline prediction running point + future residual.

[0039] By employing a method of linearly adding the baseline predicted operating point to the future residuals for fusion, the entire prediction process achieves extremely high computational efficiency and system real-time performance, significantly reducing the computational load on the controller and ensuring rapid response in prediction and decision-making within industrial real-time control systems. Simultaneously, this transparent linear relationship ensures the high-precision fusion prediction results have excellent interpretability, allowing engineers to clearly trace the contributions of the compressor physical model and the data-driven residual compensation model, facilitating system debugging, verification, and trust establishment.

[0040] In some implementations, in step S5, the dynamic surge margin is calculated as follows: Dynamic surge margin = (high-precision fusion prediction result - dynamic surge boundary) / dynamic surge boundary × 100%.

[0041] By defining the dynamic surge margin as (high-precision fusion prediction result - dynamic surge boundary) / dynamic surge boundary × 100%, a standardized and quantitative assessment of safety risks is achieved. This percentage-based dynamic surge margin index is intuitive and uniform. Using the dynamic surge boundary as the calculation benchmark, the dynamic surge margin can reflect the true safe distance under the current operating conditions in real time, overcoming the rigidity of the traditional fixed margin method.

[0042] In some implementations, in step S6, the tiered intervention control strategy includes: Safe zone: No intervention is performed when the dynamic surge margin exceeds the first threshold; Warning zone: When the dynamic surge margin is less than or equal to the first threshold and greater than the second threshold, fine-tuning control is executed; Danger Zone: When the dynamic surge margin is less than or equal to the second threshold, fast protection control is executed.

[0043] By setting up a tiered intervention control strategy, a leap from "one-size-fits-all" to "refined" anti-surge control has been achieved. Through the division of first and second thresholds, the system can ensure efficient operation without intervention when the surge margin is sufficient, perform fine-tuning control in advance to prevent risks when the margin narrows, and decisively execute rapid protection control to ensure safety when the margin is severely insufficient. The tiered response logic avoids the frequent malfunctions or response delays that may result from traditional single-threshold control. Furthermore, the buffering and early intervention of the warning zone significantly reduces the probability of the system entering the danger zone and triggering severe protection, thus achieving synergistic optimization of operational safety, control stability, and work efficiency.

[0044] In some implementations, the fine-tuning control involves slowly pre-opening the bypass valve, fine-tuning the rotational speed, or fine-tuning the guide vanes in very small steps to smoothly push the future operating trajectory away from the dynamic surge boundary.

[0045] By employing fine-tuning control in the early warning zone—using extremely small steps to slowly pre-open the bypass valve or to fine-tune the speed / guide vane—proactive and smooth intervention against potential surge risks is achieved. This "small and proactive" control action avoids the system pressure and flow shocks caused by sudden large opening and closing of the bypass valve in traditional control, significantly improving the operational stability of the refrigeration circuit and magnetic levitation bearing. Simultaneously, the fine-tuning operation can slowly and steadily push the future operating trajectory away from the surge boundary, effectively preventing risks while minimizing unnecessary efficiency losses or process disturbances caused by overly aggressive control actions, achieving a fine balance between safety protection and operational efficiency.

[0046] In some implementations, the rapid protection control is to rapidly increase the opening of the bypass valve to ensure system safety.

[0047] By setting up a protective control that triggers a rapid increase in the bypass valve opening within the danger zone, a reliable and decisive final safety barrier is constructed for the entire anti-surge method. When the dynamic surge margin is extremely low, this control strategy can bypass conventional regulation logic and establish bypass flow at the fastest speed, forcibly pulling the operating point away from the surge zone. This design ensures that, under the most critical operating conditions, the control system can always prioritize the physical safety of the equipment, effectively preventing surge that may occur due to prediction uncertainties or extreme disturbances, thereby greatly enhancing the operational resilience and reliability of the entire unit under complex or harsh conditions.

[0048] In some implementations, the first threshold is 15% and the second threshold is 8%.

[0049] By setting the first threshold at 15% and the second threshold at 8%, a clear, reasonable, and engineering-verified quantitative basis is provided for the tiered intervention control strategy. The 15% first threshold ensures ample buffer space in the safety zone, allowing the compressor to operate without intervention under most normal operating conditions, thus guaranteeing continuous high-efficiency operation. The 8% second threshold defines the bottom line for decisive protective actions, ensuring the ultimate safety of the system. The warning zone (15% to 8%) formed between the two thresholds is designed to allow the system sufficient time to mitigate risks through smooth fine-tuning, while avoiding premature or late intervention, thus achieving a delicate balance between safety, stability, and operational economy.

[0050] Implementation Case 2: Please see Figure 2 An anti-surge control system for a centrifugal compressor, comprising: The data acquisition module is used to collect the compressor's operating status parameters in real time. The baseline prediction module is used to input the operating status parameters into the compressor physical model and output the baseline prediction operating point. The residual processing module is used to calculate the historical residual sequence and predict future residuals through a data-driven residual compensation model. The fusion prediction module is used to fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; The surge margin prediction module is used to deduce the dynamic surge boundary based on high-precision fusion prediction results and calculate the dynamic surge margin. A graded control module is used to execute a graded intervention control strategy based on the dynamic surge margin.

[0051] By adopting a modular system design, the data acquisition module, benchmark prediction module, residual processing module, fusion prediction module, surge margin prediction module, and hierarchical control module are integrated into collaborative functional modules, achieving a high degree of integration and automated operation of the anti-surge control process. Through the collaboration of the benchmark prediction module and the residual processing module, the system both follows physical laws and corrects model errors, achieving high-precision dynamic prediction of the surge boundary. Furthermore, the surge margin prediction module outputs the dynamic surge boundary in real time and calculates the dynamic surge margin, and finally, the hierarchical control module executes a hierarchical intervention control strategy. This enables the entire anti-surge control system to achieve closed-loop intelligent control from state perception and risk prediction to precise intervention, significantly improving the initiative and reliability of surge protection, and also significantly optimizing the compressor's operating efficiency and overall stability through smooth and proactive control actions.

[0052] Implementation Case 3: A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned anti-surge control method for a centrifugal compressor.

[0053] By embedding an anti-surge control method for centrifugal compressors into an executable program stored on a storage medium, the standardized packaging and convenient deployment of this advanced control strategy have been achieved. By having the processor call and run this program, it is ensured that the control method can be completely and accurately repeated each time, eliminating deviations and uncertainties that may arise from manual operation or ad-hoc configuration. This reliably reproduces all the technical advantages of the method in improving safety, efficiency, and stability. Simultaneously, this also provides a foundation for subsequent remote program updates and continuous algorithm optimization.

[0054] The functions described herein can be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions can be stored as one or more instructions or codes on or transmitted via a computer-readable medium. Other examples and embodiments are within the scope and spirit of this invention and the appended claims. For example, due to the nature of software, the functions described above can be implemented using software executed by a processor, hardware, firmware, hardwired, or any combination thereof. Furthermore, the functional units can be integrated into a single processing unit, or each unit can exist physically separately, or two or more units can be integrated into a single unit.

[0055] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0056] The units described as separate components may or may not be physically separate. Similarly, the components of the control device may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0057] 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, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0058] 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 or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for anti-surge control of a centrifugal compressor, characterized in that: Includes the following steps: S1: Real-time acquisition of compressor operating status parameters; S2: Input the operating status parameters into the compressor physical model and output the baseline predicted operating point; S3: Calculate the historical residual sequence and predict future residuals using a data-driven residual compensation model; S4: Fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; S5: Based on the high-precision fusion prediction results, deduce the dynamic surge boundary and calculate the dynamic surge margin; S6: Implement a graded intervention control strategy based on the dynamic surge margin.

2. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In step S1, the operating status parameters include compressor speed, inlet guide vane opening, inlet and outlet pressure, inlet and outlet temperature, flow rate, motor power, and shaft vibration signal.

3. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In step S2, the compressor physical model is constructed based on the first principles of fluid dynamics and thermodynamics.

4. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In step S3, the historical residual sequence is obtained by comparing the historical baseline predicted output value of the compressor physical model with the actual measurement value of the sensor during the same period; the data-driven residual compensation model is a long short-term memory network or a gated recurrent unit.

5. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In step S4, the fusion is a linear addition, and the fusion formula is: high-precision fusion prediction result = baseline prediction running point + future residual.

6. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In step S6, the tiered intervention control strategy includes: No intervention is performed when the dynamic surge margin exceeds the first threshold. When the dynamic surge margin is less than or equal to the first threshold and greater than the second threshold, fine-tuning control is executed. When the dynamic surge margin is less than or equal to the second threshold, fast protection control is executed.

7. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: The fine-tuning control involves slowly pre-opening the bypass valve with extremely small steps, fine-tuning the rotational speed, or fine-tuning the guide vanes; the rapid protection control involves rapidly increasing the opening degree of the bypass valve.

8. The anti-surge control method for a centrifugal compressor according to claim 1, characterized in that: In some implementations, the first threshold is 15% and the second threshold is 8%.

9. A surge prevention control system for a centrifugal compressor, characterized in that: include: The data acquisition module is used to collect the compressor's operating status parameters in real time. The baseline prediction module is used to input the operating status parameters into the compressor physical model and output the baseline prediction operating point. The residual processing module is used to calculate the historical residual sequence and predict future residuals through a data-driven residual compensation model. The fusion prediction module is used to fuse the baseline prediction running point with the future residual to obtain a high-precision fusion prediction result; The surge margin prediction module is used to deduce the dynamic surge boundary based on high-precision fusion prediction results and calculate the dynamic surge margin. A graded control module is used to execute a graded intervention control strategy based on the dynamic surge margin.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When executed by a processor, the computer program implements a surge control method for a centrifugal compressor as described in any one of claims 1-8.

Images