A magnetic suspension centrifugal heat pump compressor variable load adaptive control method and system

Abstract

The application discloses a kind of magnetic suspension centrifugal heat pump compressor variable load adaptive control method and system, it is related to compressor control technical field, including the construction offline calibration dynamic effective surge margin lookup table, optimal variable speed trajectory lookup table and inlet guide vane-speed path lookup table;In each control cycle, the current operating state parameters of magnetic suspension centrifugal heat pump compressor are obtained;The application obtains dynamic effective surge margin by unit operating state parameters in each control cycle by offline calibration multiple lookup tables, obtains optimal variable speed setting value by table lookup interpolation when load is increased and matches guide vane-speed regulation timing, can adaptively adjust load change rate with the margin of operating point relative to surge boundary changes;When load is reduced, segmented deceleration is carried out with check valve critical rebound rate as constraint, can inhibit the risk of surge, while avoiding check valve rebound impact, improve the stability of magnetic suspension centrifugal heat pump compressor variable load whole process operation.

Description

Technical Field

[0001] This invention relates to the field of compressor control technology, and in particular to a variable load adaptive control method and system for a magnetic levitation centrifugal heat pump compressor. Background Technology

[0002] Magnetic levitation centrifugal heat pump compressors are a new type of compressor that integrates magnetic levitation bearing technology and high-speed motor technology into traditional compressors. They offer advantages such as oil-free operation, frictionless operation, and low operating costs, and are widely used in industries such as pharmaceuticals, textiles, and glass. In the operation and control of magnetic levitation centrifugal heat pump compressors, variable load (load increase, load decrease) control is the core control element that determines whether the unit can safely, efficiently, and quickly respond to load regulation requirements across the entire operating range.

[0003] Existing magnetic levitation centrifugal heat pump compressors mostly use a fixed-rate control method for variable load control. During variable load control, the magnetic levitation centrifugal compressor needs to move from one operating point to another on the pressure ratio-flow performance diagram. Regardless of how far the current operating point is from the surge boundary (surge line, blockage line), the load increase and decrease processes are completed by switching operating conditions at a preset fixed rate.

[0004] However, in the variable load control process of a magnetic levitation centrifugal heat pump compressor, when the operating point is close to the surge boundary, the fixed-rate load increase process will directly exceed the safety margin and trigger surge. When the operating point is far from the surge boundary, the fixed-rate load increase will unnecessarily prolong the response time. Its load increase process cannot adaptively adjust to the change in distance between the operating point and the surge boundary. In addition, due to the rapid drop in outlet pressure when the compressor speed suddenly drops, the pressure difference on both sides of the check valve reverses, causing the valve plate to rebound. The fixed-rate load reduction does not take the critical rebound rate of the check valve as a constraint. If the load reduction rate is too fast, the rebound of the check valve will easily cause violent periodic metal impacts, making the unit operation unstable. Its load reduction process does not consider the dynamic characteristics of the check valve. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a variable load adaptive control method and system for a magnetic levitation centrifugal heat pump compressor, which solves the problems of existing magnetic levitation centrifugal heat pump compressor variable load control technology being unable to adaptively adjust to changes in the distance between the operating point and the surge boundary during the load increase process, and not considering the dynamic characteristics of the check valve during the load reduction process.

[0006] The technical means employed in this invention are as follows: In a first aspect, the present invention provides a variable load adaptive control method for a magnetically levitated centrifugal heat pump compressor, comprising: Construct a lookup table for the dynamic effective surge margin, the optimal variable speed trajectory, and the inlet guide vane-rotation speed path for offline calibration; In each control cycle, the current operating status parameters of the magnetic levitation centrifugal heat pump compressor are acquired; Determine the direction of load change based on the target load and the current load; Based on the current operating status parameters, the dynamic effective surge margin between the current operating point and the surge boundary is obtained by querying the dynamic effective surge margin lookup table; If the variable load direction is the load increase direction, the optimal variable rate trajectory lookup table is queried and interpolated using the current operating point, target load, and dynamic effective surge margin as indexes to obtain the optimal variable rate setpoint for the current control cycle; the inlet guide vane-speed path lookup table is queried using the dual-dimensional proximity determination result as indexes to obtain the inlet guide vane-speed adjustment sequence; based on the optimal variable rate setpoint and the inlet guide vane-speed adjustment sequence, the current load of the magnetic levitation centrifugal heat pump compressor is increased to the target load; If the load change direction is the load reduction direction, the current load of the magnetic levitation centrifugal heat pump compressor is reduced to the target load using a segmented speed reduction strategy, with the critical rebound rate of the check valve as a constraint.

[0007] Furthermore, based on the operating condition range in which the dynamic effective surge margin is located, the optimal variable rate setpoint is determined; The operating condition range includes a free operating condition range, a buffer operating condition range, and a safe operating condition range; If the current dynamic effective surge margin is in the free operating range, the optimal variable rate setting value is the maximum allowable rate; If the current dynamic effective surge margin is in the buffer operating range, the optimal variable rate setting value decreases monotonically as the dynamic effective surge margin decreases. If the current dynamic effective surge margin is within the safe operating range, the optimal variable rate setting value is the minimum allowable rate.

[0008] Furthermore, the index dimensions of the dynamic effective surge margin lookup table include the static surge margin range, condensation temperature range, variable load direction, and system thermal inertia level.

[0009] Further, based on the current operating state parameters, the dynamic effective surge margin between the current operating point and the surge boundary is retrieved from the dynamic effective surge margin lookup table, including: Calculate the static surge margin based on the current mass flow rate and the surge flow rate determined by referring to the surge line based on the current rotational speed and inlet guide vane opening. Using the static surge margin range, the current condensing temperature range, the current load change direction, and the system thermal inertia level as indexes, the dynamic effective surge margin lookup table is queried and interpolated between adjacent index points to obtain the dynamic effective surge margin.

[0010] Furthermore, based on the target load and the current load, the direction of load change is determined, including: If the target load is greater than the current load, then change the load direction to the load increase direction; If the target load is less than the current load, then change the load direction to the load reduction direction; If the difference between the target load and the current load is within the dead zone, then steady state is maintained.

[0011] Furthermore, the dual-dimensional proximity determination result is the approach direction between the operating point and the surge boundary, and the approach direction is determined based on the head margin and flow margin.

[0012] Furthermore, the pressure head margin is the relative deviation between the current pressure head and the pressure head corresponding to the surge line, and the flow margin is the relative deviation between the current flow rate and the flow rate corresponding to the surge line; If both the pressure head margin and the flow margin are less than a preset proximity threshold, then the condition is determined to be a comprehensive approximation. Otherwise, if the pressure head margin is less than the flow margin and the pressure head margin is less than a preset proximity threshold, it is determined that the pressure head is close to the surge boundary in the pressure head dimension direction. Otherwise, if the flow margin is less than the pressure head margin and the flow margin is less than a preset proximity threshold, it is determined that the flow dimension is close to the surge boundary.

[0013] Furthermore, at each trajectory point, explicit model predictive control is used for steady-state optimization of surge prevention and load matching; The explicit model predictive control is as follows: the steady-state optimization problem of anti-surge and load matching is solved offline as a piecewise affine explicit control law and stored as an explicit control law lookup table. After determining the critical region of the current operating condition online, the explicit control law lookup table is searched to obtain the optimal control quantity.

[0014] Furthermore, the segmented speed reduction strategy is as follows: During the first time period, the current speed of the magnetic levitation centrifugal heat pump compressor is reduced to the first speed at a first rate; During the second time period, maintain the first rotational speed; During the third time period, the first rotational speed is reduced to the target rotational speed at a second rate; The absolute value of the first rate is greater than the absolute value of the second rate.

[0015] Secondly, the present invention provides a variable load adaptive control system for a magnetic levitation centrifugal heat pump compressor, comprising: The offline calibration module is configured to build a dynamic effective surge margin lookup table, an optimal variable speed trajectory lookup table, and an inlet guide vane-rotational speed path lookup table for offline calibration. The parameter acquisition module is configured to acquire the current operating status parameters of the magnetic levitation centrifugal heat pump compressor in each control cycle. The variable load direction determination module is configured to determine the variable load direction based on the target load and the current load; The dynamic effective surge margin calculation module is configured to query the dynamic effective surge margin between the current operating point and the surge boundary based on the current operating state parameters in the dynamic effective surge margin lookup table; The load adjustment module is configured to, if the variable load direction is the load increase direction, use the current operating point, target load, and dynamic effective surge margin as indexes to query the optimal variable rate trajectory lookup table and interpolate to obtain the optimal variable rate setpoint for the current control cycle; use the dual-dimensional proximity determination result as an index to query the inlet guide vane-speed path lookup table to obtain the inlet guide vane-speed adjustment timing; and based on the optimal variable rate setpoint and the inlet guide vane-speed adjustment timing, increase the current load of the magnetic levitation centrifugal heat pump compressor to the target load. The load reduction adjustment module is configured to, if the variable load direction is the load reduction direction, use the critical rebound rate of the check valve as a constraint and adopt a segmented speed reduction strategy to reduce the current load of the magnetic levitation centrifugal heat pump compressor to the target load.

[0016] Compared with the prior art, the present invention has the following advantages: This invention obtains the dynamic effective surge margin through offline calibration of multiple lookup tables and the unit's operating status parameters in each control cycle. When increasing the load, it combines lookup table interpolation to obtain the optimal variable rate setpoint and matches the guide vane-speed adjustment sequence. It can adaptively adjust the load change rate according to the margin change of the operating point relative to the surge boundary. When reducing the load, it uses the critical rebound rate of the check valve as a constraint to perform segmented speed reduction, which can suppress the surge risk and avoid the rebound impact of the check valve, thereby improving the operational stability of the magnetic levitation centrifugal heat pump compressor throughout the variable load process. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is an overall architecture diagram of a variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating the dynamic effective surge margin evaluation model in an embodiment of the present invention. Figure 4 This is a comparison diagram of the running point trajectories of the embodiments of the present invention and the fixed rate scheme during the load increase process; Figure 5 This is a comparison chart of the rate curves of the embodiment of the present invention and the fixed rate scheme during the load increase process; Figure 6 This is a diagram showing the segmented speed reduction curve during the unloading process in an embodiment of the present invention; Figure 7 This is a measured bearing air gap waveform diagram within the surge-sensitive speed range in an embodiment of the present invention; wherein, Figure 7 (a) is a waveform diagram of radial air gap fluctuation under constant rate loading. Figure 7 (b) is a waveform diagram of radial air gap fluctuation of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "comprising" and "having" and any variations thereof in this invention 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.

[0021] The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0022] Please see Figure 1 and Figure 2 This invention provides a variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor, comprising the following steps: Step 101: Construct the offline calibrated dynamic effective surge margin lookup table, optimal variable speed trajectory lookup table, and inlet guide vane-rotational speed path lookup table.

[0023] When the magnetic levitation centrifugal heat pump compressor (hereinafter referred to as the compressor) is deployed for the first time, offline calibration is performed on the dynamic effective surge margin lookup table, the optimal variable speed trajectory lookup table, and the inlet guide vane-speed path lookup table.

[0024] Specifically, surge boundary data across the entire operating condition range is obtained through compressor performance testing and system identification. The static surge margin range, condensing temperature range, variable load direction (load increase or decrease), and system thermal inertia level are used as index dimensions. The dynamic effective surge margin value under each combination of operating conditions is calculated through offline calibration, and a dynamic effective surge margin lookup table is generated. The specific calibration process is as follows: (1) The surge line under the entire operating condition is determined through bench performance testing, and the surge flow distribution and local slope distribution of the surge line with speed and inlet guide vane opening are obtained; (2) The system water-side thermal inertia time constant is determined and classified through system identification; (3) Variable load disturbance test or mechanism simulation is carried out under each index combination of operating conditions, and the minimum proximity of the operating point to the surge boundary during the dynamic process is measured. The dynamic correction amount is obtained by comparing it with the static surge margin, and the dynamic correction amount is filled into the dynamic effective surge margin lookup table.

[0025] The system's thermal inertia rating is determined by the total heat capacity of the entire water circulation loop on both the heating and heat source sides, not just the water stored in the heat exchanger tubes. Taking a unit with a rated heating capacity of 2000kW as an example, the condenser water flow rate is approximately 96kg / s (supply and return water temperature difference 5K). The water stored in the heat exchanger tubes alone reaches several hundred liters, while the system water volume, comprised of piping in the machine room, terminals, and buffer tanks, is typically tens to hundreds of cubic meters. The system-level water-side time constant τ (τ≈ total system water volume ÷ circulation flow rate) is divided into three levels: low (τ<3 minutes, system water volume approximately 20m³ or less), medium (τ=3-10 minutes, system water volume approximately 20-60m³), and high (τ>10 minutes, system water volume approximately 60m³ or more, large-scale regional energy systems can reach several hundred cubic meters). In winter heating operation, the engineering practice of operating at low pressure ratio in the range of 3300-5000 rpm and establishing pressure difference for 8-10 minutes covers about 1-2 system-level time constants and can be used as an indirect on-site calibration basis for τ.

[0026] The offline calibration method for the optimal variable-rate trajectory lookup table is as follows: for different combinations of initial operating conditions, target loads, and boundary conditions, explicit model predictive control is used to solve for the optimal variable-rate trajectory offline, and the results are stored as an optimal variable-rate trajectory lookup table. The optimization objective is a comprehensive evaluation of variable load time and energy consumption. Constraints include a dynamic effective surge margin not lower than the safety threshold, a blockage boundary, a critical rebound rate of the check valve, and an upper limit for motor current.

[0027] The offline calibration method for the inlet guide vane-speed path lookup table is as follows: Based on the compressor performance diagram, for different approaches (head dimension approach, flow dimension approach, and combined head and flow approach), the inlet guide vane-speed adjustment timing path during the load increase process is optimized offline and stored as an inlet guide vane-speed path lookup table.

[0028] Step 102: In each control cycle, obtain the current operating status parameters of the magnetic levitation centrifugal heat pump compressor.

[0029] At the beginning of each control cycle, the current operating status parameters of the magnetic levitation centrifugal heat pump compressor can be obtained. These parameters include rotational speed, inlet guide vane opening, mass flow rate, exhaust pressure, suction pressure, and condensing temperature.

[0030] Step 103: Determine the direction of load change based on the target load and the current load.

[0031] In some embodiments, determining the load change direction based on the target load and the current load includes: if the target load is greater than the current load, the load change direction is the load increase direction; if the target load is less than the current load, the load change direction is the load decrease direction; if the difference between the target load and the current load is within the dead zone, the steady state is maintained.

[0032] Step 104: Based on the current operating status parameters, query the dynamic effective surge margin between the current operating point and the surge boundary in the dynamic effective surge margin lookup table.

[0033] Please see Figure 3 Based on the current operating parameters, a dynamic effective surge margin assessment model is constructed to calculate the dynamic effective surge margin.

[0034] The dynamic effective surge margin assessment model differs from the static surge margin in that it comprehensively considers the local slope of the surge line (different safety margin changes correspond to different operating condition zones for the same distance change), condensing temperature zones (the surge line is steeper at high condensing temperatures), variable load direction (the dynamic margin is smaller than the static margin when increasing load, and vice versa when decreasing load), and the system thermal inertia time constant (the dynamic offset amplitude is determined by the evaporator / condenser water volume and flow rate). Notably, the local slope of the surge line does not appear as a separate index dimension in the lookup table; its role is internalized in offline calibration: when calibrating the dynamic correction, measurements are taken separately for each operating condition zone. Steep surge line zones calibrate with larger correction values. Therefore, the dynamic effective surge margin obtained from the same static surge margin differs in different slope zones.

[0035] The dynamic effective surge margin is obtained by querying the dynamic effective surge margin lookup table. The index dimensions of the dynamic effective surge margin lookup table include the static surge margin range, condensing temperature range, variable load direction, and system thermal inertia level. The dynamic effective surge margin lookup table stores the dynamic effective surge margin values ​​obtained through offline calibration for each combination of operating conditions.

[0036] In some embodiments, based on the current operating state parameters, the dynamic effective surge margin between the current operating point and the surge boundary is obtained by querying the dynamic effective surge margin lookup table. This includes: calculating the static surge margin based on the current mass flow rate and the surge flow rate determined by referring to the surge line according to the current rotational speed and inlet guide vane opening; and obtaining the dynamic effective surge margin by querying the dynamic effective surge margin lookup table using the interval where the static surge margin is located, the current condensing temperature interval, the current variable load direction, and the system thermal inertia level as indexes and interpolating between adjacent index points.

[0037] Specifically, the calculation process for dynamic effective surge margin is as follows: Based on the current measured flow rate and the surge flow rate corresponding to the current rotational speed and inlet guide vane opening, calculate the static surge margin according to the definition of static surge margin (current mass flow rate minus surge flow rate, then divided by surge flow rate); using the interval of static surge margin, the current condensing temperature interval, the current load change direction, and the system thermal inertia level as indexes, look up the dynamic effective surge margin lookup table and interpolate between adjacent index points to obtain the dynamic effective surge margin. The above process is equivalent to: Dynamic effective surge margin = Static surge margin - Dynamic correction amount; the dynamic correction amount is determined by offline calibration and stored in the lookup table, requiring only one table lookup and interpolation operation online, and the dynamic correction amount is negative during load reduction.

[0038] Engineering calibration reference for system thermal inertia time constant: Taking a 2000kW magnetic levitation centrifugal heat pump as an example, under winter heating start-up conditions, running at a low pressure ratio for 8-10 minutes in the 3300-5000rpm range to establish the suction and exhaust pressure difference before slowly increasing the load can effectively improve the dynamic effective surge margin. This residence time covers approximately 1-2 system-level water-side time constants (τ≈ total system water volume ÷ circulation flow rate, approximately 5-10 minutes for medium-sized commercial water systems), and can serve as an indirect on-site calibration basis for the system thermal inertia time constant.

[0039] Step 105: If the load change direction is the load increase direction, use the current operating point, target load, and dynamic effective surge margin as indexes to query the optimal variable rate trajectory lookup table and interpolate to obtain the optimal variable rate setpoint for the current control cycle; use the dual-dimensional proximity judgment result as an index to query the inlet guide vane-speed path lookup table to obtain the inlet guide vane-speed adjustment timing; based on the optimal variable rate setpoint and the inlet guide vane-speed adjustment timing, increase the current load of the magnetic levitation centrifugal heat pump compressor to the target load.

[0040] By using table lookup and interpolation to obtain the optimal variable rate setpoint adapted to the current operating conditions, and simultaneously acquiring the inlet guide vane-speed adjustment sequence, coordinated timing adjustment of speed and inlet guide vane is achieved during the load increase process of the magnetic levitation centrifugal heat pump compressor. This method can constrain the rate of change during the load increase process, avoid drastic fluctuations in the unit's aerodynamic conditions caused by asynchronous speed and guide vane adjustment actions, effectively maintain the dynamic effective surge margin throughout the load increase process, suppress surge risk, and improve the stability of the variable load operation process while ensuring the unit quickly follows the target load, thus protecting the safe operation of the magnetic levitation centrifugal heat pump compressor.

[0041] Specifically, using the current operating point, target load, and current dynamic effective surge margin as indexes, the nearest neighbor trajectory is located in the optimal variable speed trajectory lookup table and linearly interpolated dimension by dimension to obtain the speed regulation rate setpoint for the current control cycle.

[0042] In some embodiments, the optimal variable rate setting is determined based on the operating condition range in which the dynamic effective surge margin is located; the operating condition range includes a free operating condition range, a buffer operating condition range, and a safe operating condition range; if the current dynamic effective surge margin is in the free operating condition range, the optimal variable rate setting is the maximum permissible rate; if the current dynamic effective surge margin is in the buffer operating condition range, the optimal variable rate setting monotonically decays as the dynamic effective surge margin decreases; if the current dynamic effective surge margin is in the safe operating condition range, the optimal variable rate setting is the minimum permissible rate.

[0043] The safe operating range is the aerodynamic instability region rather than the rotor dynamics critical speed region; the boundary threshold of the safe operating range is determined by offline calibration based on the local slope of the surge line and the system thermal inertia time constant with respect to the operating range.

[0044] By adaptively matching the optimal variable rate setpoint within the operating condition range of the dynamic effective surge margin, rapid adjustment is performed when the margin is sufficient, and the adjustment rate is gradually suppressed when the margin decreases, thus balancing the unit's load following performance and aerodynamic stability constraints and effectively avoiding surge risks during load changes.

[0045] In some embodiments, the two-dimensional proximity determination result is the approach direction between the operating point and the surge boundary, and the approach direction is determined based on the head margin and flow margin. The head margin is the relative deviation between the current head and the head corresponding to the surge line, and the flow margin is the relative deviation between the current flow and the flow corresponding to the surge line. If both the head margin and the flow margin are less than a preset proximity threshold, it is determined to be a comprehensive proximity; otherwise, if the head margin is less than the flow margin and the head margin is less than the preset proximity threshold, it is determined to be close to the surge boundary in the head dimension direction; otherwise, if the flow margin is less than the head margin and the flow margin is less than the preset proximity threshold, it is determined to be close to the surge boundary in the flow dimension direction. The preset proximity threshold ranges from 5% to 10%.

[0046] The inlet guide vane-speed adjustment timing is obtained by querying the inlet guide vane-speed path lookup table based on the two-dimensional proximity judgment results; the rate setpoint and the inlet guide vane-speed adjustment timing are executed point by point in the lower steady-state optimization layer.

[0047] Dual-dimensional proximity refers to the proximity of head margin and flow margin; where head margin = (current head - head corresponding to surge line) / head corresponding to surge line, and flow margin = (current flow rate - flow rate corresponding to surge line) / flow rate corresponding to surge line. The direction of proximity is determined based on the dual-dimensional proximity: small head margin, large flow margin: close to the surge boundary in the head dimension; large head margin, small flow margin: close to the surge boundary in the flow dimension; both margins are small: overall close.

[0048] The optimal inlet guide vane-speed timing path under different approach directions is optimized offline and stored as an inlet guide vane-speed path lookup table. Online, the path lookup table is queried based on the two-dimensional proximity determination results to execute the process. Specific path parameters are determined offline based on the compressor performance diagram.

[0049] The path selection does not hardcode a fixed inlet guide vane lifting / lowering direction, but instead determines the optimal timing based on the actual performance diagram using a path lookup table. The path lookup table is organized according to the following directional rules: If the flow rate is determined to be similar, a timing sequence of "increase speed first, then open inlet guide vanes" is adopted. That is, the pressure head and flow rate are first increased synchronously by increasing the speed, and the inlet guide vanes are gradually opened after the flow margin is restored, avoiding transient flow-pressure head mismatch caused by directly opening the inlet guide vanes at constant speed. If the pressure head is determined to be similar, a timing sequence of "open inlet guide vanes slowly first, then increase speed" is adopted. That is, the flow rate is first introduced with a small opening increment to exchange for pressure head space, and then the speed is increased to restore the pressure head. It is worth noting that, based on actual test results from units on the same platform, fully opening the inlet guide vanes in the high-speed range actually makes surge more likely to be triggered. The specific rate and opening parameters of the above directional rules are calibrated offline from the compressor performance diagram.

[0050] Step 106: If the load change direction is the load reduction direction, the current load of the magnetic levitation centrifugal heat pump compressor is reduced to the target load using a segmented speed reduction strategy, with the critical rebound rate of the check valve as a constraint.

[0051] The dynamic effective surge margin obtained from the table during load reduction is used for process monitoring and verification. Using the critical rebound rate of the check valve as a constraint and adopting a segmented speed reduction strategy for load reduction adjustment, the degree of speed drop during the load reduction process is limited, preventing rebound impact of the check valve. This can weaken the aerodynamic disturbance caused by pipeline backflow, maintain the aerodynamic stability of the compression system, and balance the efficiency of load reduction adjustment with the safety of equipment operation.

[0052] In some embodiments, the segmented speed reduction strategy is as follows: during a first time period, the current speed of the magnetic levitation centrifugal heat pump compressor is reduced to a first speed at a first rate; during a second time period, the first speed is maintained; during a third time period, the first speed is reduced to a target speed at a second rate; the absolute value of the first rate is greater than the absolute value of the second rate.

[0053] The critical rebound rate of the check valve is calculated based on the check valve dynamics model, which is constructed based on the valve plate mass, spring stiffness, damping coefficient, compressor outlet volume, and current discharge pressure.

[0054] The critical rebound rate of the check valve is incorporated as a hard constraint in the unloading process into the upper-level trajectory planning. A segmented deceleration strategy is adopted: In the first time period, the first time period is a rapid deceleration phase, reducing the current speed to a first speed, which is a safe value determined according to the compressor performance chart; in the second time period, the second time period is a buffer pause phase, maintaining the first speed to balance the system pressure; in the third time period, the third time period is a slow deceleration phase, reducing the speed to the target speed at a second rate, which is lower than the critical rebound rate.

[0055] In some embodiments, at each trajectory point, explicit model predictive control is used for steady-state optimization of anti-surge and load matching. The explicit model predictive control is as follows: the steady-state optimization problem of anti-surge and load matching is solved offline as a piecewise affine explicit control law and stored as an explicit control law lookup table. After determining the critical region to which the current operating condition belongs online, the explicit control law lookup table is searched to obtain the optimal control quantity.

[0056] Specifically, a two-layer planning-control architecture is adopted, consisting of an upper trajectory planning layer and a lower steady-state optimization layer.

[0057] In the upper-level trajectory planning layer (slow logic layer, execution cycle in seconds), the optimal variable-rate trajectory for different initial operating conditions, target loads, and boundary conditions is pre-calculated offline and stored as an optimal variable-rate trajectory lookup table. Online execution: Based on the current operating conditions and target load, the nearest neighbor trajectory is located in the optimal variable-rate trajectory lookup table, and the rate setpoint for the current control cycle is obtained through multi-dimensional interpolation; the inlet guide vane-speed adjustment timing is obtained by querying the inlet guide vane-speed path lookup table based on the two-dimensional proximity judgment result. Each table lookup and interpolation calculation takes no more than 5 milliseconds. Optimization objective: A comprehensive trade-off between variable load time and energy consumption (multi-objective optimization). Constraints: Dynamic effective surge margin not lower than the safety threshold, blockage boundary, check valve critical rebound rate, and motor current upper limit.

[0058] The lower-level steady-state optimization layer (fast logic layer, 50 milliseconds): employs an explicit model predictive control method (solving the steady-state optimization problem of surge prevention and load matching offline as an explicit control law lookup table, then executing the lookup table online), performing steady-state optimization of surge prevention and load matching at each trajectory point. The explicit control law is in piecewise affine form: the optimal control quantity equals the state quantity multiplied by the gain matrix of the corresponding critical zone plus the bias vector, i.e., u = Ki·x + gi, where x is the state quantity (including current mass flow rate, pressure ratio, speed, and target load deviation), and i is the critical zone number; the division of each critical zone and the corresponding gain Ki and bias gi are obtained offline by multi-parameter quadratic programming. During online execution, only the critical zone of the current operating condition needs to be determined and one affine calculation needs to be performed, so the online calculation time per cycle does not exceed 5 milliseconds.

[0059] This invention also provides a variable load adaptive control system for a magnetic levitation centrifugal heat pump compressor, comprising: an offline calibration module configured to construct an offline calibration dynamic effective surge margin lookup table, an optimal variable speed trajectory lookup table, and an inlet guide vane-speed path lookup table; a parameter acquisition module configured to acquire the current operating state parameters of the magnetic levitation centrifugal heat pump compressor in each control cycle; a variable load direction determination module configured to determine the variable load direction based on the target load and the current load; and a dynamic effective surge margin calculation module configured to query the dynamic effective surge margin lookup table based on the current operating state parameters to obtain the current operating point and surge margin. The system includes: a dynamic effective surge margin at the vibration boundary; a load increase adjustment module configured to, if the variable load direction is the load increase direction, query the optimal variable rate trajectory lookup table and interpolate to obtain the optimal variable rate setpoint for the current control cycle; query the inlet guide vane-speed path lookup table to obtain the inlet guide vane-speed adjustment timing; based on the optimal variable rate setpoint and the inlet guide vane-speed adjustment timing, increase the current load of the magnetic levitation centrifugal heat pump compressor to the target load; and a load decrease adjustment module configured to, if the variable load direction is the load decrease direction, use the critical rebound rate of the check valve as a constraint and adopt a segmented speed reduction strategy to decrease the current load of the magnetic levitation centrifugal heat pump compressor to the target load.

[0060] The same or similar parts between the various embodiments in this specification can be referred to each other, and will not be repeated here.

[0061] This invention obtains the dynamic effective surge margin through offline calibration of multiple lookup tables and the unit's operating status parameters in each control cycle. When increasing the load, it combines lookup table interpolation to obtain the optimal variable rate setpoint and matches the guide vane-speed adjustment sequence. It can adaptively adjust the load change rate according to the margin change of the operating point relative to the surge boundary. When reducing the load, it uses the critical rebound rate of the check valve as a constraint to perform segmented speed reduction, which can suppress the surge risk and avoid the rebound impact of the check valve, thereby improving the operational stability of the magnetic levitation centrifugal heat pump compressor throughout the variable load process.

[0062] The technical solution of the present invention will be described below with reference to specific embodiments.

[0063] Example 1: Increasing the load of a magnetic levitation centrifugal heat pump from 50% to 100% during winter heating conditions. This embodiment applies to a magnetic levitation centrifugal heat pump for commercial building heating. It employs an oil-free magnetic levitation bearing drive and a combined adjustment structure with inlet guide vanes, with a direct-drive motor and gearless speed increase. The unit has a rated heating capacity of 2000kW, a rated speed of 11000rpm, a speed adjustment range of 30%~105% (3300~11550rpm), an inlet guide vane opening adjustment range of 20%~100%, and an energy level adjustment range of 30%~100% (30% is the minimum levitation speed requirement for the magnetic levitation bearing; during low-load operation, system circulation is maintained through hot gas bypass (used to maintain the minimum circulating flow of the compressor, distinct from anti-surge bypass, and is not considered the full-load heating operation point). The designed heating outlet water temperature is 45℃ (condensing temperature approximately 47℃), and the designed heat source side inlet water temperature is 15℃ (evaporating temperature approximately 8℃). In winter heating conditions, this embodiment demonstrates the adaptive control process of the unit increasing from 50% load (1000kW) to 100% load (2000kW).

[0064] The hardware configuration is as follows: the control system uses an industrial-grade PLC (such as the Siemens S7-1500 series), with a control cycle of 50 milliseconds. The upper-level trajectory planning layer has an execution cycle of 1 second. The dynamic effective surge margin lookup table, the optimal variable speed trajectory lookup table, and the inlet guide vane-speed path lookup table are stored in the PLC memory card. It should be noted that the control cycle mentioned in this invention refers to the execution cycle of the upper-level trajectory planning layer (in seconds, 1 second in this embodiment), while the lower-level steady-state optimization layer performs steady-state optimization at each trajectory point with a control cycle of 50 milliseconds.

[0065] The dynamic effective surge margin is calculated as follows: Under the current operating conditions: 50% load (1000kW), speed 7000rpm, inlet guide vane opening 60%, mass flow rate approximately 6.0kg / s, exhaust pressure approximately 1230kPa (saturation pressure corresponding to R134a refrigerant and condensation temperature of 47℃), suction pressure approximately 386kPa (evaporation temperature approximately 8℃). Target load: 2000kW (full-load refrigerant flow rate approximately 12kg / s).

[0066] According to the dynamic effective surge margin lookup table, the current static surge margin is approximately 25% (current mass flow rate is approximately 6.0 kg / s, and surge flow rate under this condition is approximately 4.8 kg / s). After dynamic correction, the dynamic effective surge margin is approximately 14%. It is determined that the system is in the buffer operating range.

[0067] The direction of load change is determined cycle by cycle as follows: if the target load is greater than the current load, it is determined as an increase in load; if it is less than the current load, it is determined as a decrease in load. If the difference between the two is within the dead zone, it is determined as a steady state. The determination result serves as one of the index dimensions of the dynamic effective surge margin lookup table, and also determines whether to call the increase in load trajectory or the decrease in load trajectory. This embodiment is for increase in load: using the current operating point, target load, and current dynamic effective surge margin as indexes, the nearest neighbor trajectory is located in the optimal variable speed trajectory lookup table, and linear interpolation is performed dimension by dimension to obtain the speed regulation rate setpoint for the current control cycle; please refer to [link to relevant documentation]. Figure 5 The optimal variable rate setting value is determined according to the current dynamic effective surge margin range. The maximum allowable rate is taken in the free operating range, the rate is reduced according to the margin in the buffer operating range, and the minimum rate is taken in the safe operating range. The inlet guide vane-speed adjustment timing is obtained by querying the inlet guide vane-speed path lookup table according to the dual-dimensional proximity judgment result. The above rate setting value and timing command are executed point by point by the lower steady-state optimization layer.

[0068] Approach direction determination and path selection: During the load increase process, the head margin and flow margin are continuously calculated. If it is determined that the pressure head is close to the surge boundary, the path lookup table is consulted to select path A ("gradual opening of the inlet guide vane first, then increase the speed" timing: first, a small opening increment is used to introduce flow to exchange for head space, and then the speed is increased to restore the head). The execution unit adjusts the inlet guide vane and speed according to the timing instructions of path A. The specific rate and opening parameters of path A are determined according to the actual performance chart.

[0069] If the flow rate is determined to be close to the surge boundary, select path B ("increase speed first, then open inlet guide vane" timing) from the path lookup table: first increase speed to synchronously raise the pressure head and flow rate, and then gradually open the inlet guide vane after the flow margin is restored. The execution unit adjusts according to the timing instructions of path B. The "fully open inlet guide vane first" timing is prohibited in the high-speed range.

[0070] Control Performance: Simulation prediction based on the mechanistic model during the load increase process shows a smooth dynamic transition without triggering surge warnings. The total time for load increase from 50% to 100% is approximately 95 seconds (approximately 40 seconds for the free section, 40 seconds for the buffer section, and 15 seconds for the safe operating range, based on the rate setting simulation of the offline calibrated trajectory). Compared to a fixed slow-speed scheme that increases load at the rate within the safe operating range (total time approximately 180 seconds), the response time is reduced by approximately 47%. The anti-surge bypass valve remains open throughout the entire process, resulting in no backflow loss. The minimum rate within the safe operating range is determined by a lookup table, with even lower rates in steep surge areas.

[0071] Please see Figure 4Compared to the fixed-rate load increase scheme: using the exact same hardware platform and test conditions as Example 1, the load increase rate was fixed at 50 revolutions per second. During the dynamic process, the fixed rate could not adaptively adjust according to the actual surge margin. When the dynamic effective surge margin was insufficient, a surge warning was triggered, and the anti-surge bypass valve was forced to open for protection. Simulation prediction: Approximately 20 seconds after entering the corresponding operating range of the buffer section, a surge warning was triggered, the bypass valve opened, and the load increase continued after the system stabilized. The total time was approximately 210 seconds (including recovery time), accompanied by significant backflow loss.

[0072] Example 2: Winter Heating Start-up of Magnetic Levitation Centrifugal Heat Pump - Safe Transition to Dangerous Speed ​​Zone This embodiment addresses a specific scenario where, during winter heating operation, the compressor starts at low speed and increases load, traversing the dangerous speed range of 6000-6600 rpm. The following data is derived from actual test conditions of the magnetic levitation centrifugal compressor product (see appendix). Figure 7 (a)). Actual measurements show that aerodynamic instability is prone to occur in this speed range when the intake and exhaust pressure difference is small or the loading is fast.

[0073] Operating Context: During winter heating startup, the compressor starts at its lowest suspension speed of 3300 rpm. In the initial stage, the pressure difference between the suction and discharge is small, requiring 8-10 minutes of operation within the 3300-5000 rpm range to establish the pressure difference (the system thermal inertia time constant is approximately 15-20 seconds, only the water-side time constant of the heat exchanger; the system-level water-side time constant is approximately 5-10 minutes; this residence time is an empirical value, including control loop stability and safety margin).

[0074] Assessment of dynamic effective surge margin in dangerous speed range: When the speed increases to 6000 rpm, the dynamic effective surge margin lookup table is consulted: This speed range corresponds to a steeper surge line area and the intake and exhaust pressure difference is small. The dynamic effective surge margin is about 7%, and it is determined that the safe operating condition range has been entered.

[0075] Fine-tuning of the safe operating range: The speed increase rate is limited to no more than 30-50 revolutions per minute (far lower than the normal rate in the buffer section). The inlet guide vane is adjusted slowly according to the safe path in the path lookup table, allowing sufficient time for the flow field to stabilize and waiting for the dynamic effective surge margin to recover.

[0076] It should be noted that this speed range is the aerodynamic instability zone rather than the rotor dynamics critical speed zone: the steady-state operation in this region is stable (there is already measured data), and instability is only triggered during transient loading due to insufficient dynamic effective surge margin. Therefore, this invention adopts a strategy of deceleration rather than acceleration to pass through the critical speed zone, which is the opposite of the conventional practice of "rapidly passing through the critical speed" in rotor dynamics. The two have different mechanisms.

[0077] Continuously monitor the dynamic effective surge margin. When the margin recovers to more than 10%, gradually enter the buffer phase, and the rate increases according to the margin decay law.

[0078] High-speed recovery: After the speed reaches 7000 rpm or more, the dynamic effective surge margin recovers to more than 20%, and the system has been loaded at a normal rate in the buffer section, gradually increasing to the target speed.

[0079] Compared to a fixed-rate loading method traversing the dangerous speed range: Using a fixed rate (50 rpm increase per second) to traverse the 6000-6600 rpm range, the inability to identify insufficient dynamic effective surge margin directly leads to aerodynamic instability due to rapid loading: noticeable exhaust pipe vibration and sharp fluctuations in the bearing air gap waveform (see...). Figure 7 (a) Accompanied by rhythmic airflow sounds. In severe cases, it triggers a surge warning, forcing the anti-surge bypass valve to open for protection, and may even cause the compressor to shut down due to severe fluctuations.

[0080] Under the same operating conditions, after applying the fine adjustment strategy of the safe operating range of this invention, the bearing air gap fluctuation returned to stability, and the dangerous speed range was successfully passed (see...). Figure 7 (b)).

[0081] Example 3 This embodiment addresses the check valve rebound constraint during load reduction. It uses the same unit and winter heating conditions as Embodiment 1, demonstrating the process of reducing the unit's load from 100% (2000kW) to 50% (1000kW). The end point of the load reduction is the same operating condition point as the start point of the load increase in Embodiment 1.

[0082] Calculation of critical rebound rate of check valve: Based on valve plate mass, spring stiffness, damping coefficient, compressor outlet volume and current discharge pressure, a dynamic model of check valve is constructed to calculate the critical rebound rate under the current operating conditions.

[0083] Please see Figure 6 The speed reduction is performed in stages: the offline calibrated load reduction trajectory is obtained by looking up the table: the first time period is rapid speed reduction, the speed is reduced from 11,000 rpm to 8,000 rpm; the second time period is buffer pause, maintaining 8,000 rpm to wait for pressure balance; the third time period is slow speed reduction, the speed is reduced from 8,000 rpm to 7,000 rpm, and the speed reduction rate is lower than the critical rebound rate.

[0084] Compared with the fixed-rate unloading scheme: When a fixed unloading rate is used, the pressure difference on both sides of the check valve reverses during the sudden drop in speed. The valve plate rebounds and generates a violent periodic metallic impact noise. The rebound amplitude exceeds the allowable value, and the system is seriously unstable.

[0085] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0086] 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.

Claims

1. A variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor, characterized in that, include: Construct a lookup table for the dynamic effective surge margin, the optimal variable speed trajectory, and the inlet guide vane-rotation speed path for offline calibration; In each control cycle, the current operating status parameters of the magnetic levitation centrifugal heat pump compressor are acquired; Determine the direction of load change based on the target load and the current load; Based on the current operating status parameters, the dynamic effective surge margin between the current operating point and the surge boundary is obtained by querying the dynamic effective surge margin lookup table; If the variable load direction is the load increase direction, the optimal variable rate trajectory lookup table is queried and interpolated using the current operating point, target load, and dynamic effective surge margin as indexes to obtain the optimal variable rate setpoint for the current control cycle; the inlet guide vane-speed path lookup table is queried using the dual-dimensional proximity determination result as indexes to obtain the inlet guide vane-speed adjustment sequence. Based on the optimal variable speed setpoint and the inlet guide vane-speed adjustment timing, the current load of the magnetic levitation centrifugal heat pump compressor is increased to the target load; If the load change direction is the load reduction direction, the current load of the magnetic levitation centrifugal heat pump compressor is reduced to the target load using a segmented speed reduction strategy, with the critical rebound rate of the check valve as a constraint.

2. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, Determine the optimal variable rate setting value based on the operating condition range in which the dynamic effective surge margin is located; The operating condition range includes a free operating condition range, a buffer operating condition range, and a safe operating condition range; If the current dynamic effective surge margin is in the free operating range, the optimal variable rate setting value is the maximum allowable rate; If the current dynamic effective surge margin is in the buffer operating range, the optimal variable rate setting value decreases monotonically as the dynamic effective surge margin decreases. If the current dynamic effective surge margin is within the safe operating range, the optimal variable rate setting value is the minimum allowable rate.

3. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, The index dimensions of the dynamic effective surge margin lookup table include static surge margin range, condensation temperature range, variable load direction, and system thermal inertia level.

4. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, Based on the current operating state parameters, the dynamic effective surge margin between the current operating point and the surge boundary is retrieved from the dynamic effective surge margin lookup table, including: Calculate the static surge margin based on the current mass flow rate and the surge flow rate determined by referring to the surge line based on the current rotational speed and inlet guide vane opening. Using the static surge margin range, the current condensing temperature range, the current load change direction, and the system thermal inertia level as indexes, the dynamic effective surge margin lookup table is queried and interpolated between adjacent index points to obtain the dynamic effective surge margin.

5. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, Based on the target load and the current load, determine the direction of the load change, including: If the target load is greater than the current load, then change the load direction to the load increase direction; If the target load is less than the current load, then change the load direction to the load reduction direction; If the difference between the target load and the current load is within the dead zone, then steady state is maintained.

6. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, The dual-dimensional proximity determination result is the approach direction between the operating point and the surge boundary, and the approach direction is determined based on the head margin and flow margin.

7. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 6, characterized in that, The pressure head margin is the relative deviation between the current pressure head and the pressure head corresponding to the surge line, and the flow margin is the relative deviation between the current flow rate and the flow rate corresponding to the surge line; If both the pressure head margin and the flow margin are less than a preset proximity threshold, then the condition is determined to be a comprehensive approximation. Otherwise, if the pressure head margin is less than the flow margin and the pressure head margin is less than a preset proximity threshold, it is determined that the pressure head is close to the surge boundary in the pressure head dimension direction. Otherwise, if the flow margin is less than the pressure head margin and the flow margin is less than a preset proximity threshold, it is determined that the flow dimension is close to the surge boundary.

8. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, At each trajectory point, explicit model predictive control is used for steady-state optimization of surge prevention and load matching; The explicit model predictive control is as follows: the steady-state optimization problem of anti-surge and load matching is solved offline as a piecewise affine explicit control law and stored as an explicit control law lookup table. After determining the critical zone of the current operating condition online, the explicit control law lookup table is searched to obtain the optimal control quantity.

9. The variable load adaptive control method for a magnetic levitation centrifugal heat pump compressor according to claim 1, characterized in that, The segmented speed reduction strategy is as follows: During the first time period, the current speed of the magnetic levitation centrifugal heat pump compressor is reduced to the first speed at a first rate; During the second time period, maintain the first rotational speed; During the third time period, the first rotational speed is reduced to the target rotational speed at a second rate; The absolute value of the first rate is greater than the absolute value of the second rate.

10. A variable load adaptive control system for a magnetic levitation centrifugal heat pump compressor, characterized in that, include: The offline calibration module is configured to build a dynamic effective surge margin lookup table, an optimal variable speed trajectory lookup table, and an inlet guide vane-rotational speed path lookup table for offline calibration. The parameter acquisition module is configured to acquire the current operating status parameters of the magnetic levitation centrifugal heat pump compressor in each control cycle. The variable load direction determination module is configured to determine the variable load direction based on the target load and the current load; The dynamic effective surge margin calculation module is configured to query the dynamic effective surge margin between the current operating point and the surge boundary based on the current operating state parameters in the dynamic effective surge margin lookup table; The load adjustment module is configured to, if the variable load direction is the load increase direction, use the current operating point, target load and the dynamic effective surge margin as indexes to query the optimal variable rate trajectory lookup table and interpolate to obtain the optimal variable rate setpoint for the current control cycle; and use the dual-dimensional proximity determination result as an index to query the inlet guide vane-speed path lookup table to obtain the inlet guide vane-speed adjustment timing. Based on the optimal variable speed setpoint and the inlet guide vane-speed adjustment timing, the current load of the magnetic levitation centrifugal heat pump compressor is increased to the target load; The load reduction adjustment module is configured to, if the variable load direction is the load reduction direction, use the critical rebound rate of the check valve as a constraint and adopt a segmented speed reduction strategy to reduce the current load of the magnetic levitation centrifugal heat pump compressor to the target load.

Images