Air-cooled modular magnetic suspension refrigeration unit dynamic control system
By assessing the health status of the fans in real time and dynamically grouping and scheduling them, the problem of heat dissipation capacity degradation and surge in the fan array of the air-cooled modular magnetic levitation chiller unit was solved, realizing balanced aging of the fan array and improving the stability and self-adaptive capability of the chiller unit.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- QINGSHU ZHIYUAN SYSTEM INTEGRATION (JIANGSU) CO LTD
- Filing Date
- 2026-04-14
- Publication Date
- 2026-05-29
Smart Images
Figure CN122107608A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of refrigeration-related technologies, and in particular to a dynamic control system for an air-cooled modular magnetic levitation refrigeration unit. Background Technology
[0002] Air-cooled modular magnetic levitation chillers have been widely used in data centers, commercial buildings, industrial cooling, and medical purification due to their advantages such as being oil-free, highly efficient, low-noise, and having excellent partial load efficiency ratio. The heat dissipation performance on the condenser side directly determines the chiller's operating efficiency and reliability, and this heat dissipation function is mainly accomplished by an array of multiple independent fans within the air-cooled module.
[0003] Currently, control methods for air-cooled modular fan arrays mainly fall into the following categories: First, simple feedback control based on condensing pressure or exhaust temperature, where all fans are uniformly speed-regulated or started / stopped in stages; second, fans are divided into several groups according to fixed physical locations, and then sequentially activated or deactivated based on load demand. For example, Chinese invention patent CN110332620B discloses a control method for the condenser fans of an air-cooled air conditioning unit. By grouping the condenser fans, fans in the same group share a single control node, and the groups are controlled in stages based on condensing pressure, allowing fans to be activated or deactivated sequentially. Another example is Chinese invention patent application CN118102677A, which discloses an air-cooled heat dissipation system, control method, and terminal. This system divides the fans into a first fan group and a second fan group, using relays and fuses to achieve independent power supply control for each group. While these technical solutions achieve basic fan scheduling functions within a fixed grouping framework, they still have the following shortcomings: The above solutions all employ a fixed grouping structure, with fans within each group operating uniformly under the same command, making it impossible to differentiate scheduling based on the individual health status of each fan. Fans in good health are used in the same way as those with deteriorating performance, resulting in a non-linear, accelerated decline in overall heat dissipation capacity, and the aging process cannot be actively slowed down. Furthermore, when the performance of a fan deteriorates, it is often not discovered until it completely fails. At this point, a sudden shutdown will cause a sharp drop in heat dissipation capacity, easily triggering compressor surge or excessive condensing pressure protection. Summary of the Invention
[0004] The core of this invention lies in the control system of the air-cooled modular magnetic levitation chiller unit that evaluates the health status of the fan in real time, so as to extend the overall life of the fan array, avoid sudden interruption of heat dissipation, reduce the risk of surge and improve the system's adaptability.
[0005] To solve the above problems, the present invention adopts the following technical solution.
[0006] A dynamic control system for an air-cooled modular magnetic levitation chiller unit includes an air-cooling module, a health assessment module, a group management unit, a scheduling execution unit, a compressor-fan coupling optimizer, a floating regrouping module, a load reduction protection module, and an online learning and updating module. The air-cooled module is an array of multiple independently controllable fans, each of which is equipped with a group of status monitoring sensors. The health assessment module communicates with the condition monitoring sensor group to calculate the comprehensive health index (HI) of each wind turbine in real time. The group management unit is connected to the health assessment module. The group management unit dynamically divides the wind turbine array into at least four conventional groups based on HI: high performance group, medium performance group, low performance group, and emergency standby group. The scheduling execution unit is connected to the group management unit and the control terminal of each wind turbine, and is used to activate the wind turbines in the priority order of high performance group → medium performance group → low performance group during daily operation; The compressor-fan coupling optimizer communicates with the controller of the magnetic levitation compressor to receive information on the compressor's surge risk level, bearing power consumption status, and refrigerant subcooling. It dynamically adjusts the start-stop strategy and speed command of the scheduling execution unit to achieve coordinated control between the compressor side and the condenser side. The floating regrouping module is signal-connected to the group management unit. It is used to reassess the performance ranking of all wind turbines according to a preset period or when the HI change exceeds a threshold, and dynamically adjust their respective conventional groups. During floating regrouping, a buffer zone is set based on the lower limit HI value of each conventional group. This buffer zone is used to raise the upgrade threshold and lower the downgrade threshold. Specifically, the lower limit threshold for the high-performance group is 0.85, but the downgrade threshold is set to 0.82; the upgrade threshold for the medium-performance group is 0.87, and the downgrade threshold is 0.58; the upgrade threshold for the low-performance group is 0.63, and the downgrade threshold is 0.28. A minimum dwell time of 10 minutes is also set, and wind turbines are not allowed to be regrouped again within 10 minutes of being regrouped. The load reduction protection module is signal-connected to the floating regrouping module and the scheduling execution unit, and is configured to gradually reduce the workload of a wind turbine instead of shutting it down directly when the turbine floats from a higher performance group to a lower performance group. The online learning update module is used to update the calculation model of the Health Index (HI) and the parameters of the coupling optimizer based on actual operating data.
[0007] Furthermore, the comprehensive health index HI is calculated by weighting at least two parameters among the following: the ratio of actual air volume to rated air volume, the ratio of motor current to rated current, vibration amplitude, and bearing temperature, and HI∈[0,1]. Among them, the health index HI of the high-performance group is ≥0.85, the health index of the medium-performance group is 0.60≤HI<0.85, the health index of the low-performance group is 0.30≤HI<0.60, and the health index of the emergency backup group is HI<0.30 and has not yet failed.
[0008] Furthermore, surge risk level R surge Based on the exhaust pressure change rate dP / dt and the current pressure ratio P ratio =P dis / P suc calculate: When dP / dt>TH dP And P ratio <TH PR At that time, R surge =2 indicates a high risk of surge; When dP / dt>TH dP or P ratio <TH PR At that time, R surge =1 indicates a mild risk of surge; Otherwise R surge =0 indicates no risk of surge; Among them, TH dP The threshold value for the rate of change of exhaust pressure is 0.05-0.15 MPa / s; TH PR The pressure ratio threshold is 1.2-1.5.
[0009] Furthermore, the compressor-fan coupling optimizer executes the following objective function: minJ = W1·P comp +W2·ΣP fan +W3·Σ(C i ·α i )+W4·R surge_penalty ; Among them, P comp P represents the real-time power consumption of the compressor. fan For the power consumption of each wind turbine, C i Let α be the normalized replacement cost of the i-th wind turbine. i R is a degradation penalty factor based on the health index HI. surge_penalty This is a surge risk penalty item; the constraints include the condensing temperature range, the upper limit of the subcooling change rate, and the fan speed limit.
[0010] Optionally, the emergency standby group includes an elite emergency subgroup and a last-place emergency subgroup; the elite emergency subgroup contains at least one wind turbine with the best health index selected from the high-performance group and keeps it in hot standby status, while the last-place emergency subgroup contains several wind turbines with the lowest health index selected from the low-performance group but which have not yet failed and keep them in cold standby status. The elite emergency subgroup and the last emergency subgroup do not participate in daily scheduling. When an emergency condition is triggered, the elite emergency subgroup is activated first. If the heat dissipation requirements are still not met, the last emergency subgroup is activated in order of health index from high to low. The start-up time of the last emergency subgroup fan is short-term and the time of use is counted in seconds.
[0011] Furthermore, the ratio of the number of elite emergency subgroups to the number of last emergency subgroups is 1:1-3, and the total number of fans in the elite emergency subgroups and the last emergency subgroups does not exceed 20% of the total number of fans in the air-cooled module.
[0012] Furthermore, it also includes a periodic rotation module, which is configured to swap the roles of the wind turbines in the elite emergency subgroup with the wind turbines in the high-performance group that have the best health index and the longest cumulative running time every preset running time or time period, so as to balance wear and prevent the wind turbines in the elite emergency subgroup from developing hidden faults due to long-term idleness.
[0013] Furthermore, it also includes an emergency reconfiguration module, configured as follows: When the wind turbines of the elite emergency subgroup are actually called up, the wind turbines are removed from the emergency standby group and reassigned to the high-performance group or medium-performance group according to the health index after the call. At the same time, the wind turbines with the highest health index are selected from the current high-performance group or medium-performance group, and after a short self-check, they are added to the elite emergency subgroup. When the fan in the last emergency subgroup is actually called up, the fan is removed from the emergency standby group and marked as to be replaced. At the same time, the fan with the highest health index and that has not yet failed is selected from the current low performance group and, after a short self-check, is added as the new fan in the last emergency subgroup. If no available wind turbines are available in the corresponding source group, a maintenance alarm will be triggered, and wind turbines from another emergency subgroup will temporarily take over the missing role.
[0014] Furthermore, the fans in the last emergency subgroup are equipped with a periodic self-check logic: every 72 hours, they are run at a speed not exceeding 20% of the rated speed for no more than 3 minutes to monitor the starting current, vibration amplitude and air volume; fans that fail the self-check are immediately removed from the emergency group and marked as pending maintenance or replacement.
[0015] Furthermore, the emergency response conditions for triggering the emergency standby group are as follows: Condition 1: Any one of the wind turbines in the high-performance group or the medium-performance group experiences a sudden failure, including loss of monitoring signal, overcurrent, or stall. Condition 2: The total heat dissipation requirement exceeds the maximum heat dissipation capacity that the high-performance group, medium-performance group, and low-performance group can currently output; Condition 3: The average health index of the high-performance group decreases by more than a preset threshold within a unit of time. When any of these conditions are detected, the emergency standby group will be activated.
[0016] Compared with the prior art, the advantages of this invention are: (1) This solution calculates the comprehensive health index of each fan in real time through the health assessment module, and dynamically divides the fans into high, medium and low performance groups and emergency standby groups based on the index. This achieves differentiated scheduling based on individual health status, avoids the accelerated decline of overall heat dissipation capacity caused by the equal use of healthy fans and declining fans, and can dynamically adjust the fan affiliation according to the changes in the health index, thereby actively delaying the aging process of the fans.
[0017] (2) In addition, the emergency backup group is further divided into elite emergency subgroups (selecting the best hot backup from the high-performance group) and last emergency subgroups (taking the lowest cold backup from the low-performance group), and is equipped with regular rotation and post-emergency reorganization, realizing the hierarchical storage and precise scheduling of emergency resources. While ensuring continuous and stable cooling, it significantly improves the resource utilization rate and reliability of the fan array. Attached Figure Description
[0018] Figure 1 This is a logic flowchart of the present invention; Figure 2 This is a schematic diagram of the coupling between the fan and the compressor of the present invention; Figure 3 This is a schematic diagram illustrating the grouping based on health indices according to the present invention; Figure 4 This is a schematic diagram of the fan composition of the emergency standby group during the initial grouping of the present invention; Figure 5 This is a schematic diagram of the fan composition of the emergency standby group during regrouping according to the present invention; Figure 6 This is a flowchart illustrating the floating control method of the present invention. Detailed Implementation
[0019] The technical solutions will now be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention.
[0020] First implementation method: like Figures 1-2 A dynamic control system for an air-cooled modular magnetic levitation chiller unit includes an air-cooling module, a health assessment module, a group management unit, a scheduling execution unit, a compressor-fan coupling optimizer, a floating regrouping module, a load reduction protection module, and an online learning and updating module. The air-cooled module is an array of multiple independently controllable fans. Each fan is equipped with a condition monitoring sensor group, which includes, but is not limited to, electrical parameter sensors (current sensors, voltage sensors), vibration sensors, and temperature sensors. The health assessment module is connected to the condition monitoring sensor group to calculate the comprehensive health index HI of each fan in real time. The comprehensive health index HI is calculated by weighting at least two parameters among the following: the ratio of actual air volume to rated air volume, the ratio of motor current to rated current, vibration amplitude, and bearing temperature, and HI∈[0,1].
[0021] The compressor-fan coupling optimizer communicates with the controller of the magnetic levitation compressor to receive information on the compressor's surge risk level, bearing power consumption status, and refrigerant subcooling. It dynamically adjusts the start-stop strategy and speed command of the scheduling execution unit to achieve coordinated control between the compressor side and the condenser side. Surge risk level R surge Based on the exhaust pressure change rate dP / dt and the current pressure ratio P ratio =P dis / P suc calculate: When dP / dt>TH dP And P ratio <TH PR At that time, R surge =2 indicates a high risk of surge; When dP / dt>TH dP or P ratio <TH PR At that time, R surge =1 indicates a mild risk of surge; Otherwise R surge =0 indicates no risk of surge; Among them, TH dP The threshold value for the rate of change of exhaust pressure is 0.05-0.15 MPa / s; TH PR The pressure ratio threshold is 1.2-1.5.
[0022] The compressor-fan coupling optimizer executes the following objective function: minJ = W1·P comp +W2·ΣP fan +W3·Σ(C i ·α i )+W4·R surge_penalty ; Among them, P comp P represents the real-time power consumption of the compressor. fan For the power consumption of each wind turbine, C i Let α be the normalized replacement cost of the i-th wind turbine. i R is a degradation penalty factor based on the health index HI. surge_penalty This is a surge risk penalty item; the constraints include the condensing temperature range, the upper limit of the rate of change of subcooling, and the fan speed limit. W1 represents the weighting coefficient for compressor power consumption, used to adjust the importance of compressor power consumption (Pcomp) in the overall optimization objective. A larger W1 indicates a greater tendency for the system to reduce compressor power consumption. W2 represents the weighting coefficient for total fan power consumption, used to adjust the importance of the total fan power consumption (ΣPfan) in the optimization objective. A larger W2 indicates a greater tendency for the system to reduce total fan energy consumption. W3 represents the weighting coefficient for fan degradation cost, used to adjust the weighting coefficient for fan life degradation cost (Σ(C...). i ·α i The importance of W3 in the optimization objective is that the larger the W3, the more the system tends to protect the poorly healthy fans and delay their aging; W4 represents the weight coefficient of surge risk, which is used to adjust the importance of the surge risk penalty (Rsurge_penalty) in the optimization objective. The larger the W4, the more the system tends to avoid the compressor from entering the surge state and prioritize operational safety.
[0023] These four weighting coefficients (W1-W4) are adjustable parameters used to balance three conflicting goals: energy saving (reducing compressor and fan power consumption), life extension (reducing fan degradation costs), and safety (avoiding surge risk).
[0024] In high-reliability scenarios such as data centers, W4 can be increased to prioritize ensuring no surge and uninterrupted heat dissipation; In conventional commercial buildings, W1 and W2 can be appropriately increased to pursue optimal energy efficiency; When old units are in short supply or spare parts are scarce, W3 can be increased to extend the service life of existing fans as much as possible.
[0025] Through online learning and update modules, W1-W4 can automatically optimize and adjust based on actual operating data, achieving adaptive balance in long-term operation.
[0026] The group management unit is signal-connected to the health assessment module. The group management unit dynamically divides the fan array into at least four regular groups based on the Health Index (HI): high-performance group, medium-performance group, low-performance group, and emergency standby group. The health index HI of the high-performance group is ≥ 0.85, the health index of the medium-performance group is 0.60 ≤ HI < 0.85, the health index of the low-performance group is 0.30 ≤ HI < 0.60, and the health index of the emergency standby group is HI < 0.30 and has not yet failed. In this embodiment, the fans in the emergency standby group are all low-performance fans that have not yet failed. Under normal circumstances, they do not participate in daily scheduling. They are only activated in an emergency when the fans in the high-performance group, medium-performance group, and low-performance group cannot meet the total heat dissipation demand, or when some fans suddenly fail, in order to avoid a sudden paralysis or suspension of heat dissipation and ensure the stable operation of the refrigeration unit.
[0027] It is worth noting that, in order to ensure the service life of the motor and maintain the stability of the refrigeration unit, the motors of the high-performance group can be started at full speed during use; the fans of the medium-performance group are limited to 80% speed during use; the fans of the low-performance group are limited to 50% speed during use; the emergency standby group is only allowed to be put into use for a short time when the surge risk level is ≥2 or the first three levels of fans fail, and the time of use is measured in seconds. The scheduling execution unit is connected to the group management unit and the control terminal of each wind turbine, and is used to activate the wind turbines in the priority order of high performance group → medium performance group → low performance group during daily operation; The floating regrouping module is signal-connected to the group management unit and is used to re-evaluate the performance ranking of all wind turbines according to a preset period or when the HI change exceeds a threshold, and dynamically adjust their regular group affiliation. Figure 3 During floating regrouping, buffer zones are set based on the lower limit of the Health Index (HI) for each regular group. These buffer zones raise the upgrade threshold and lower the downgrade threshold. Specifically, the lower limit threshold for the high-performance group is 0.85, but the downgrade threshold is set to 0.82; the upgrade threshold for the medium-performance group is 0.87, and the downgrade threshold is 0.58; the upgrade threshold for the low-performance group is 0.63, and the downgrade threshold is 0.28. A minimum dwell time of 10 minutes is also set, prohibiting further group changes within 10 minutes of a fan being regrouped. These buffer zones effectively prevent fans from repeatedly switching between groups due to slight fluctuations in the HI, thus avoiding group oscillations and ensuring system stability, reliability, and infrequent operation.
[0028] The load reduction protection module is signal-connected to the floating regrouping module and the scheduling execution unit. It is configured to gradually reduce the workload of a fan when it floats from a higher performance group to a lower performance group instead of shutting it down directly. This effectively avoids a sudden drop in heat dissipation capacity caused by sudden shutdown and prevents sudden changes in exhaust temperature / condensing pressure of the chiller unit. It also smoothly transfers the load to other fans, effectively maintaining system stability. At the same time, it effectively avoids electrical or mechanical shocks that may be caused by sudden fan stoppage, extending the fan's lifespan.
[0029] The online learning and update module is used to update the calculation model of the health index HI and the parameters of the coupled optimizer based on actual operating data. This enables the health assessment model to adapt to different operating conditions, environmental changes, and individual differences, thereby improving grouping accuracy and predictive ability. The online update of the coupled optimizer parameters can continuously optimize the scheduling strategy, achieve a balance between energy saving and reliability in long-term operation, avoid performance deviations caused by model solidification, and ensure that the basis for the next grouping and scheduling is a real and reliable health status.
[0030] In summary, the health assessment module calculates the comprehensive health index (HI) of the fans in real time, and combines this with the compressor-fan coupling optimizer for coordinated control, thereby reducing surge risk and total power consumption while meeting heat dissipation requirements. The group management unit divides the fans into four groups based on HI: high-performance, medium-performance, low-performance, and emergency standby, and matches them with differentiated speed limits (high-performance full speed, medium-performance 80% speed limit, low-performance 50% speed limit), maximizing the capacity of healthy fans while preventing overload. The emergency standby group is activated briefly in case of surge or failure of a conventional fan to avoid interruption of heat dissipation. The floating regrouping module introduces buffer zones and minimum dwell time to effectively suppress frequent oscillations between groups. The load reduction protection module gradually reduces the workload of fans when they float across groups instead of shutting them down directly, reducing electrical and mechanical shocks and extending fan life. The online learning and update module continuously optimizes the health index model and scheduling parameters, enabling the system to adapt to changes in environment and operating conditions. This implementation achieves balanced aging of the fan array, intelligent fault tolerance, and energy efficiency optimization, comprehensively improving the reliability, stability, and adaptability of the chiller unit.
[0031] Second implementation method: This embodiment further improves the emergency backup group based on the first embodiment, while the rest remains the same as the first embodiment.
[0032] like Figure 4 The emergency standby group further includes an elite emergency subgroup and a last-place emergency subgroup. The elite emergency subgroup contains at least one fan with the best health index selected from the high-performance group and keeps it in hot standby status. The last-place emergency subgroup contains several fans with the lowest health index selected from the low-performance group but which have not yet failed and keeps them in cold standby status. That is, the fan health index HI of the elite emergency subgroup is ≥0.85, while the fan health index HI of the last-place emergency subgroup is consistent with the first implementation method <0.30 and has not yet failed.
[0033] The elite emergency subgroup and the last emergency subgroup do not participate in daily scheduling. When an emergency condition is triggered, the elite emergency subgroup is activated first. If the heat dissipation requirements are still not met, the last emergency subgroup is activated in order of health index from high to low. The start-up time of the last emergency subgroup fan is short-term and the time of use is counted in seconds.
[0034] The emergency standby group is triggered under the following conditions. Activation of the emergency standby group will be triggered if any one of these conditions is detected: Condition 1: Any one of the wind turbines in the high-performance group or the medium-performance group experiences a sudden failure, including loss of monitoring signal, overcurrent, or stall. Condition 2: The total heat dissipation requirement exceeds the maximum heat dissipation capacity that the high-performance group, medium-performance group, and low-performance group can currently output; Condition 3: The average health index of the high-performance group decreases by more than a preset threshold within a unit of time.
[0035] The ratio of the number of elite emergency subgroups to the number of last emergency subgroups is 1:1-3, and the total number of fans in the elite emergency subgroups and the last emergency subgroups does not exceed 20% of the total number of fans in the air-cooled module. The number of fans in the elite emergency subgroups is only one. Too many fans will affect the heat dissipation capacity of the high, medium and low performance group fans for the refrigeration unit.
[0036] This system also includes a periodic rotation module matched with the emergency standby group. Its configuration is as follows: every preset running time or time period, the fans in the elite emergency subgroup and the fans in the high-performance group with the best health index and the longest cumulative running time are swapped. This allows the fans in the high-performance group with the longest cumulative running time to be properly maintained, and also prevents the fans in the elite emergency subgroup from developing hidden faults due to long-term idleness. At the same time, it can also even out the wear and tear between the fans in the high-performance group and the fans in the elite emergency subgroup.
[0037] This system further includes an emergency reconfiguration module, configured as follows: When the wind turbines of the elite emergency subgroup are actually called up, the wind turbines are removed from the emergency standby group and reassigned to the high-performance group or medium-performance group according to the health index after the call. At the same time, the wind turbines with the highest health index are selected from the current high-performance group or medium-performance group, and after a short self-check, they are added to the elite emergency subgroup. like Figure 5 When a fan in the last emergency subgroup is actually called up, it is removed from the emergency standby group and marked as awaiting replacement. Simultaneously, a fan with the highest health index that is not yet failed is selected from the current low-performance group. After a short self-check, it is added as the new fan in the last emergency subgroup. If the fan with the lowest health index is always selected from the low-performance group after each call-up, the added fan is likely to fail quickly, leading to frequent replacement alarms and manual intervention. Therefore, when regrouping after an emergency, selecting the fan with the highest health index from the low-performance group can extend the replacement cycle and reduce maintenance frequency.
[0038] If no available wind turbines are available in the corresponding source group, a maintenance alarm will be triggered, and wind turbines from another emergency subgroup will temporarily take over the missing role.
[0039] The fans in the last emergency subgroup are equipped with a periodic self-check logic: every 72 hours, they are run at a speed not exceeding 20% of the rated speed for no more than 3 minutes to monitor the starting current, vibration amplitude and air volume; fans that fail the self-check are immediately removed from the emergency group and marked as pending maintenance or replacement.
[0040] like Figure 6 The dynamic control method for the fan in the dynamic control system of the air-cooled modular magnetic levitation chiller unit includes the following steps: Step S1: Initialize the grouping. Based on the initial health index HI of each wind turbine, divide the wind turbine array into high-performance group, medium-performance group, low-performance group, elite emergency subgroup and last-place emergency subgroup. Step S2: Collect the operating status data of each wind turbine in real time and update the comprehensive health index (HI) of each wind turbine; Step S3: Perform floating regrouping according to the real-time sorting of HI based on the preset cycle or trigger conditions, and dynamically adjust the regular group to which each fan belongs; Step S4: In daily operation, the fans are activated in order of priority from high performance group to medium performance group to low performance group according to the target cooling load. The elite emergency subgroup and the last emergency subgroup do not participate in daily scheduling. Step S5: When an emergency trigger condition is detected, the fans in the elite emergency subgroup are started first within the set time threshold. If the heat dissipation requirements are still not met, the fans in the last emergency subgroup are started in order of health index from high to low. Step S6: When the health index of any fan drops and triggers cross-group floating, start load reduction protection: linearly or exponentially reduce the upper limit of the fan speed within a preset time window, and smoothly transfer the reduced load to other fans in the same group or a higher performance group. Step S7: After the emergency is lifted, perform post-emergency reorganization, replenish the fans of the elite emergency subgroups and the fans of the last emergency subgroup, and restore the complete configuration of the emergency group; Step S8: Use the online learning update module to update the calculation model of the health index HI and the parameters of the coupling optimizer based on the actual running data.
[0041] By subdividing the emergency backup group into elite emergency subgroups and last-place emergency subgroups, tiered reserve and precise scheduling of emergency resources are achieved: the elite emergency subgroup selects the fan with the best health index from the high-performance group as a hot backup, ensuring a rapid and high-quality response in emergencies; the last-place emergency subgroup selects the fan with the lowest health index but not yet failed from the low-performance group as a cold backup, making full use of the remaining value of fans nearing the end of their service life. When an emergency is triggered, the elite subgroup is activated first, and the last-place subgroup is activated in sequence if insufficient, avoiding both interruption of cooling services and excessive idleness of healthy fans. The periodic rotation module swaps the fans with the longest cumulative operating time in the elite subgroup and the high-performance group, balancing wear and preventing latent faults; the post-emergency reconfiguration module selects the fan with the highest health index from the high-performance / medium-performance group or the low-performance group to supplement the elite or last-place subgroup, effectively extending the replenishment cycle and reducing maintenance frequency; the periodic self-check logic of the last-place subgroup ensures its basic availability. By combining the load reduction protection, floating regrouping, and online learning updates in the first embodiment, this embodiment significantly improves the resource utilization, reliability, and intelligent maintenance level of the fan array while ensuring continuous and stable heat dissipation of the chiller unit.
[0042] The above description is merely a preferred embodiment of the present invention; it encompasses all the protection scope of the present invention. Any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solutions and improved concepts of the present invention, should be covered within the protection scope of the present invention.
Claims
1. A dynamic control system for an air-cooled modular magnetic levitation refrigeration unit, characterized in that, It includes an air-cooling module, a health assessment module, a group management unit, a scheduling execution unit, a compressor-fan coupling optimizer, a floating regrouping module, a load reduction protection module, and an online learning and update module; The air-cooled module is an array composed of multiple independently controllable fans, each equipped with a status monitoring sensor group. The health assessment module is communicatively connected to the status monitoring sensor group to calculate the comprehensive health index (HI) of each fan in real time. The group management unit is signal-connected to the health assessment module and dynamically divides the fan array into at least four regular groups based on the HI: high-performance group, medium-performance group, low-performance group, and emergency standby group. The scheduling execution unit is signal-connected to the group management unit and the control terminal of each fan to schedule the operation in daily operation, from high-performance group to medium-performance group. The priority order for activating fans is from performance group to low performance group; the compressor-fan coupling optimizer is communicatively connected to the controller of the magnetic levitation compressor to receive information on the compressor's surge risk level, bearing power consumption status, and refrigerant subcooling; the floating regrouping module is signal-connected to the group management unit to re-evaluate the performance ranking of all fans according to a preset period or when the HI change exceeds a threshold and dynamically adjust their respective regular groups. During floating regrouping, a buffer zone is set according to the lower limit value of the HI of each regular group; the load reduction protection module is signal-connected to the floating regrouping module and the scheduling execution unit.
2. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 1, characterized in that, The comprehensive health index HI is calculated by weighting at least two parameters among the following: the ratio of actual air volume to rated air volume, the ratio of motor current to rated current, vibration amplitude, and bearing temperature, and HI∈[0,1]. Among them, the health index HI of the high-performance group is ≥0.85, the health index of the medium-performance group is 0.60≤HI<0.85, the health index of the low-performance group is 0.30≤HI<0.60, and the health index of the emergency backup group is HI<0.30 and has not yet failed.
3. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 1, characterized in that, The surge risk level R surge Based on the exhaust pressure change rate dP / dt and the current pressure ratio P ratio =P dis / P suc calculate: When dP / dt>TH dP And P ratio <TH PR At that time, R surge =2 indicates a high risk of surge; When dP / dt>TH dP or P ratio <TH PR At that time, R surge =1 indicates a mild risk of surge; Otherwise R surge =0 indicates no risk of surge; Among them, TH dP The threshold value for the rate of change of exhaust pressure is 0.05-0.15 MPa / s; TH PR The pressure ratio threshold is 1.2-1.
5.
4. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 1, characterized in that, The compressor-fan coupling optimizer executes the following objective function: minJ=W1·P comp +W2·ΣP fan +W3·Σ(C i ·α i )+W4·R surge_penalty ; Among them, P comp P represents the real-time power consumption of the compressor. fan For the power consumption of each wind turbine, C i Let α be the normalized replacement cost of the i-th wind turbine. i R is a degradation penalty factor based on the health index HI. surge_penalty This is a surge risk penalty item; the constraints include the condensing temperature range, the upper limit of the subcooling change rate, and the fan speed limit.
5. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 1, characterized in that, The emergency standby group includes an elite emergency subgroup and a last-place emergency subgroup; the elite emergency subgroup contains at least one fan with the best health index selected from the high-performance group and keeps it in hot standby status, and the last-place emergency subgroup contains several fans with the lowest health index selected from the low-performance group but which have not yet failed and keep them in cold standby status. The elite emergency subgroup and the last emergency subgroup do not participate in daily scheduling. When an emergency condition is triggered, the elite emergency subgroup is activated first. If the heat dissipation requirements are still not met, the last emergency subgroup is activated in order of health index from high to low. The start-up time of the fan in the last emergency subgroup is short-term and the time of use is measured in seconds.
6. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 5, characterized in that, The ratio of the number of elite emergency subgroups to the number of last emergency subgroups is 1:1-3, and the total number of fans in the elite emergency subgroups and the last emergency subgroups does not exceed 20% of the total number of fans in the air-cooled module.
7. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 6, characterized in that, It also includes a periodic rotation module, which is configured to swap the roles of the wind turbines in the elite emergency subgroup with the wind turbines in the high-performance group with the best health index and the longest cumulative running time every preset running time or time period, so as to balance wear and prevent the wind turbines in the elite emergency subgroup from developing hidden faults due to long-term idleness.
8. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 7, characterized in that, It also includes an emergency reconfiguration module, configured as follows: When the fan in the elite emergency subgroup is actually called up, the fan is removed from the emergency standby group and reassigned to the high-performance group or medium-performance group according to the health index after the call. At the same time, the fan with the highest health index is selected from the current high-performance group or medium-performance group, and after a short self-check, it is added to the elite emergency subgroup. When the fan in the last emergency subgroup is actually called up, the fan is removed from the emergency standby group and marked as to be replaced. At the same time, the fan with the highest health index and that has not yet failed is selected from the current low performance group and, after a short self-check, is added as the new fan in the last emergency subgroup.
9. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 8, characterized in that, The fans in the last emergency subgroup are equipped with a periodic self-check logic: every 72 hours, they are run at a speed not exceeding 20% of the rated speed for no more than 3 minutes to monitor the starting current, vibration amplitude and air volume; fans that fail the self-check are immediately removed from the emergency group and marked as pending maintenance or replacement.
10. The dynamic control system for the air-cooled modular magnetic levitation refrigeration unit according to claim 8, characterized in that, The emergency conditions for triggering the emergency standby group are as follows: Condition 1: A sudden failure occurs in any of the wind turbines in the high-performance group or the medium-performance group. The sudden failure includes loss of monitoring signal, overcurrent, or stall. Condition 2: The total heat dissipation requirement exceeds the maximum heat dissipation capacity that the high-performance group, medium-performance group, and low-performance group can currently output; Condition 3: The average health index of the high-performance group decreases by more than a preset threshold within a unit of time. The emergency standby group will be activated when any of the conditions are detected.