A method for multi-device group control and energy saving optimization of a heating and ventilation machine room

Abstract

The application discloses a kind of HVAC equipment room multi-device group control energy-saving optimization method, it is related to HVAC control technical field, including, using water balance and heat balance constraint bayesian inference and markov chain monte carlo sampling, correct flow measuring point and determine flow credible range, heat exchange demand credible range and equipment switching permission, generate equipment switching boundary;According to equipment switching boundary, the candidate state that meets heat exchange capacity, flow, stable output, start-stop time limit and interlock condition is screened, the transfer path that switching energy consumption is lowest and heat exchange is continuous is retained, and target equipment state and transfer sequence are obtained;According to target equipment state and transfer sequence, switch equipment, in succession to cooling tower fan, circulating water pump and main machine load in equipment switching boundary, generate forward low-power operating point.The application uses water balance and heat balance constraint bayesian inference and markov chain monte carlo sampling, reduce the possibility of frequent equipment switching.

Description

Technical Field

[0001] This invention relates to the field of HVAC control technology, and in particular to an energy-saving optimization method for group control of multiple devices in an HVAC computer room. Background Technology

[0002] With the large-scale development of public buildings, industrial plants, and regional energy stations, HVAC equipment rooms are gradually forming a multi-equipment collaborative operation structure consisting of multiple main units, circulating water pumps, cooling tower fans, electric valves, and bypass pipelines. Existing HVAC equipment room group control technology typically determines the number of main units and the operating status of supporting equipment based on operating parameters such as supply and return water temperatures, main flow rates, branch flow rates, pipeline pressure differentials, main unit loads, and equipment power, combined with the equipment's rated heat exchange capacity, allowable flow range, start-up and shutdown time limits, and interlocking relationships. After the equipment combination is determined, the overall energy consumption of the main units, water pumps, and cooling tower fans is usually reduced through frequency conversion regulation, load distribution, and equipment start-up and shutdown sequencing, while the supply water temperature and terminal hydraulic conditions are maintained through operational feedback.

[0003] However, flow measurement points in HVAC equipment rooms are affected by instrument accuracy, installation location, hydraulic disturbances, and changes in equipment status. A single flow measurement value cannot accurately reflect the actual flow rate and heat exchange demand. Existing methods directly determine the number of equipment to start and stop based on this, which is prone to frequent additions and subtractions of equipment or deviations in equipment capacity judgment when heat exchange demand approaches the equipment switching boundary. At the same time, there is a significant power coupling relationship between the main unit, circulating water pump, and cooling tower fan. Local adjustment of only a single device or a single control variable is difficult to take into account the energy consumption of equipment switching, heat exchange continuity, and the total power of the equipment room. When equipment malfunctions, conventional backup connection methods are also difficult to use the original thermo-hydraulic constraints to determine the executable connection state, resulting in a lack of consistent control boundaries in group control results under measurement uncertainty, equipment switching process, and mutual influence of multiple devices. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an energy-saving optimization method for multi-equipment group control in HVAC computer rooms to solve the problem that it is difficult to balance heat exchange continuity and total energy consumption of computer rooms when switching and jointly regulating multiple equipment under uncertain flow measurement conditions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides a method for energy-saving optimization of multi-equipment group control in HVAC computer rooms, comprising,

[0008] By using water balance and heat balance constrained Bayesian inference and Markov chain Monte Carlo sampling, flow measurement points are corrected and the flow confidence range, heat exchange demand confidence range and equipment switching permission are determined, and equipment switching boundaries are generated.

[0009] Candidate states that meet the requirements of heat exchange capacity, flow rate, stable output, start-stop time limit and interlocking conditions are selected according to the equipment switching boundary. The transfer path with the lowest switching energy consumption and continuous heat exchange is retained to obtain the target equipment state and transfer sequence.

[0010] Based on the target equipment status and transfer sequence, the equipment is switched, and within the equipment switching boundary, the cooling tower fan, circulating water pump and main unit load are sequentially adjusted by relay-type extreme value regulation to generate a positive low-power operating point.

[0011] By adjusting the main unit load, circulating water pump and cooling tower fan in reverse order from the positive low power operating point, and checking the water supply temperature, pipeline pressure difference and main unit flow, a group control operating point with the lowest total power is formed.

[0012] By issuing commands for host load, frequency conversion and valve position from the group control operation point, verifying the status and transfer sequence of target equipment, isolating abnormal equipment and connecting backup equipment, and generating energy-saving operation results for the computer room.

[0013] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC room described in this invention, the water balance refers to the flow conservation relationship of the main supply water flow, the main return water flow, the bypass flow, and the operating branch flow, and the heat balance refers to the heat conservation relationship of the sum of the heat exchange of the operating host and the heat exchange of the main supply and return water flow.

[0014] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC room described in this invention, the calibration of the flow measurement point specifically includes:

[0015] Select the flow meter calibration accuracy and the effective flow rate confirmed in the previous control cycle to establish the prior distribution of each flow measurement point. Then, use the current flow measurement value and the corresponding pipeline temperature change to correct the prior distribution and obtain the posterior distribution of each flow measurement point.

[0016] Markov chain Monte Carlo sampling was performed according to the posterior distribution to obtain candidate flow combinations for each flow measurement point. The candidate flow combinations were then tested successively using water balance and heat balance. Candidate flow combinations that did not simultaneously satisfy both balance relationships were eliminated to obtain an effective flow sample set.

[0017] The current measured values ​​of each flow measurement point are corrected based on the effective flow sample set, and the flow measurement point correction results are generated.

[0018] As a preferred embodiment of the energy-saving optimization method for multi-device group control in HVAC computer rooms described in this invention, the generation of equipment switching boundaries specifically includes:

[0019] Extract the lower and upper boundaries of the flow rate of each pipeline from the effective flow rate sample set in the flow measurement point correction results, and determine the reliable range of the flow rate of each pipeline.

[0020] The heat exchange demand is calculated item by item by combining the reliable range of flow rate and the corresponding pipeline temperature change. The heat exchange demand value that simultaneously satisfies water balance and heat balance is retained to determine the reliable range of heat exchange demand.

[0021] The reliable range of heat exchange demand is divided according to the minimum load adjustment allowed by the host controller, and the continuous demand segment is obtained.

[0022] The maximum heat exchange capacity, minimum stable output and allowable flow rate of the current equipment combination are checked segment by segment, and control conclusions are obtained for adding equipment, maintaining equipment and reducing equipment, respectively.

[0023] When each consecutive demand segment corresponds to the same control conclusion, confirm the equipment switching permission that is consistent with the control conclusion; when each consecutive demand segment corresponds to different control conclusions, confirm the maintenance of the equipment permission and limit the current number of equipment to remain unchanged.

[0024] The equipment operating flow rate is limited according to the flow rate confidence range, the equipment heat exchange capacity is limited according to the heat exchange demand confidence range, and the direction of equipment status change is limited by the equipment switching permission to generate the equipment switching boundary.

[0025] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC computer rooms described in this invention, the step of screening candidate states that meet the requirements of heat exchange capacity, flow rate, stable output, start-stop time limit, and interlocking conditions specifically includes:

[0026] Based on the direction of device status change defined by the device switching boundary, establish a licensed device status that maintains the current device combination, adds a set of host and supporting equipment, and removes a set of host and supporting equipment. When the licensed device status contains abnormal devices, replace the abnormal devices with healthy standby devices to obtain the device status to be verified.

[0027] The total heat exchange capacity of the equipment under test is verified by using the upper boundary of the reliable range of heat exchange demand in the equipment switching boundary, and the total allowable flow rate of the equipment under test is verified by using the reliable range of flow rate in the equipment switching boundary, so as to obtain the thermal and hydraulic qualified state.

[0028] Verify the stable output range, start-up and shutdown time limits, and equipment interlocking conditions of each host unit within the qualified thermal and hydraulic conditions, eliminate equipment states that do not meet any of the conditions, and obtain executable candidate states.

[0029] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC computer rooms described in this invention, the step of obtaining the target equipment state and transition sequence specifically includes:

[0030] Starting from the current equipment status, establish transfer paths corresponding to each executable candidate status according to the stage sequence of valve opening, water pump flow establishment, main unit startup, load transfer between main units, original main unit shutdown, and auxiliary equipment shutdown;

[0031] Each state transition phase only changes one set of interlocked equipment. The heat exchange capacity of the remaining operating equipment is checked using the upper boundary of the reliable range of heat exchange demand. Transition paths that have experienced effective heat exchange interruption are eliminated. The energy consumption of path switching is accumulated based on the actual power and duration of the equipment in each state transition phase.

[0032] For transfer paths that reach the same executable candidate state, retain the transfer path with the lowest switching energy consumption, and when the switching energy consumption is the same, retain the transfer path with the fewest device start-stop times, thus obtaining the candidate transfer path;

[0033] The switching energy consumption and equipment operation energy consumption of the candidate transfer paths are calculated. The executable candidate state with the lowest total energy consumption and effective heat exchange is selected as the target equipment state, and the corresponding candidate transfer path is used as the transfer sequence.

[0034] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC computer rooms described in this invention, the generation of a positive low-power operating point specifically includes:

[0035] The current equipment state is switched to the target equipment state according to the transfer sequence. While keeping the target equipment state unchanged, the cooling tower fan, circulating water pump and main unit load are selected as control variables in sequence, and only one control variable is adjusted at a time.

[0036] Change the minimum stable adjustment of the current control variable along the initial search direction. After the water supply temperature, circulation flow, pipeline pressure difference and equipment power have completed the response, compare the total power of the machine room before and after the adjustment.

[0037] When the total power of the computer room decreases, continue to adjust along the original search direction. When the total power of the computer room increases, restore the previous control quantity, change the search direction and reduce the adjustment quantity. When neither search direction produces a power decrease that exceeds the uncertainty range of power metering, fix the current control quantity and switch to the next control quantity.

[0038] If any of the following situations occur during the adjustment process: water supply temperature deviation, insufficient pressure difference in the most unfavorable pipeline, insufficient main unit flow, equipment protection action, flow exceeding the equipment switching boundary, or heat exchange capacity exceeding the equipment switching boundary, the control quantity before adjustment shall be restored.

[0039] The cooling tower fan, circulating water pump, and main unit load are adjusted sequentially to generate a positive low-power operating point.

[0040] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC computer rooms described in this invention, the reverse-order adjustment of the main unit load, circulating water pump, and cooling tower fan specifically includes:

[0041] The control quantities at the forward low-power operating point are arranged in the order of main unit load, circulating water pump, and cooling tower fan to obtain the reverse order adjustment sequence.

[0042] Adjust the host load according to the reverse order of adjustment, keep the equipment combination, total heat exchange and total circulation flow unchanged, transfer the host load adjustment from the host with a unit heat exchange power consumption greater than another operating host to another operating host, compare the total power of the machine room before and after the load transfer, and obtain the host load reverse order adjustment result;

[0043] The circulating water pump is adjusted according to the reverse order of adjustment to maintain the most unfavorable pipeline pressure difference, the minimum flow rate of the main unit and the reliable flow range. The frequency of the circulating water pump is adjusted and the total power of the main unit, water pump and supporting equipment is compared to obtain the reverse order adjustment result of the circulating water pump.

[0044] Adjust the cooling tower fan according to the reverse order of adjustment, adjust the frequency of the cooling tower fan, and compare the total power of the main unit, cooling water pump and cooling tower fan to obtain the reverse order adjustment result of the cooling tower fan;

[0045] The reverse order adjustment results are fixed sequentially to generate the reverse order running point.

[0046] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC room described in this invention, the step of forming the group control operating point with the lowest total power specifically includes:

[0047] The water supply temperature, the most unfavorable pipeline pressure difference, and the main unit flow rate corresponding to the positive low power operation point and the reverse sequence operation point are checked respectively. The operation points corresponding to any of the states of water supply temperature deviation, insufficient most unfavorable pipeline pressure difference, and insufficient main unit flow rate are eliminated to obtain the qualified operation points.

[0048] Compare the total power of the computer room at each qualified operating point, and select the operating point with the lowest total power among the forward low-power operating points and the reverse-sequence operating points to form a group control operating point.

[0049] As a preferred embodiment of the energy-saving optimization method for multi-equipment group control in HVAC computer rooms described in this invention, the generation of energy-saving operation results for the computer room specifically includes:

[0050] The group control operation point is converted into host load command, variable frequency command and valve position command, and each command is issued in sequence according to the transfer sequence. The actual valve position, actual pump flow, actual pipeline pressure difference, actual heat exchange of the host, actual water supply temperature and actual equipment power corresponding to each command are verified in sequence to obtain the equipment execution confirmation result.

[0051] If the equipment execution confirmation result of the current transfer phase matches the target equipment status, the next transfer phase continues until the verification of the target equipment status and transfer sequence is completed;

[0052] When the equipment fails to complete the corresponding instruction and a protection action is triggered, stop the subsequent adjustment of the equipment, maintain the operating status of the remaining confirmed equipment, isolate the abnormal equipment and determine the remaining heat exchange capacity;

[0053] The heat exchange gap is determined by using the credible range of heat exchange demand and the remaining heat exchange capacity. A standby equipment state covering the heat exchange gap is selected from the executable candidate states, and the standby equipment takeover is completed according to the corresponding transfer sequence.

[0054] Based on the feedback values ​​after the backup equipment is taken over, verify the actual operating equipment, actual heat exchange, actual circulation flow, and actual equipment power to generate the data center energy-saving operation results.

[0055] The beneficial effects of this invention are as follows: By utilizing water balance and heat balance constrained Bayesian inference and Markov chain Monte Carlo sampling, the flowmeter calibration error and actual operating measurement values ​​are converted into a reliable range for flow rate and a reliable range for heat exchange demand. Based on this, a device switching boundary is formed, so that the addition or reduction of equipment no longer depends on a single flow measurement value, reducing the possibility of frequent device switching when heat exchange demand is in the critical range. Within the device switching boundary, the heat exchange capacity, allowable flow rate, stable output, start-stop time limit, and interlocking conditions of the equipment are further verified, and the target equipment state with lower switching and operating energy consumption is selected from the transfer path that maintains continuous heat exchange, which can take into account the energy consumption during the equipment combination change process and the stable operation phase. By sequentially executing forward relay extreme value regulation and reverse sequence regulation, the power coupling between the main load, circulating water pump, and cooling tower fan is incorporated into the comparison of the total power of the machine room, avoiding the increase of the power of other equipment caused by reducing the power of a single device. When a faulty device triggers a protection action, the standby device status is selected and the takeover is completed based on the reliable range of heat exchange demand and the remaining heat exchange capacity. This ensures that flow correction, equipment switching, power regulation, and fault handling are always constrained by the same thermo-hydraulic boundary, thereby improving the stability, executability, and continuous power supply capability of the HVAC room group control energy-saving results. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart for an energy-saving optimization method for multi-device group control in a HVAC room.

[0058] Figure 2 A schematic diagram is generated for the boundary between flow correction and device switching.

[0059] Figure 3 A schematic diagram for candidate device status screening and transfer sequence determination.

[0060] Figure 4 This is a schematic diagram of group control operation point adjustment and abnormal equipment takeover. Detailed Implementation

[0061] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0062] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0063] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0064] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides an energy-saving optimization method for multi-equipment group control in a HVAC room, comprising the following steps:

[0065] S1. Using water balance and heat balance constrained Bayesian inference and Markov chain Monte Carlo sampling, the flow measurement points are calibrated and the flow confidence range, heat exchange demand confidence range and equipment switching permission are determined, generating the equipment switching boundary.

[0066] S1.1. The IoT controller acquires the flow rate of the main water supply pipe, the main return water pipe, the bypass flow rate, the flow rate of each operating branch, the flow rate of the operating host, the supply and return water temperatures, the calibration accuracy of the flow meter, and the calibration accuracy of the temperature measuring point according to the water flow direction in the HVAC room, and converts the flow rate into cubic meters per second and the temperature into degrees Celsius.

[0067] The flow rates of the main water supply line, the main return water line, the bypass flow rate, the flow rates of each operating branch, the flow rate of the main operating unit, and the supply and return water temperatures are all measured using the average values ​​within the same stable data collection period. The stable data collection period is determined according to the stability determination duration recorded in the on-site commissioning record. When the equipment is in the process of starting up, stopping, or load transfer, the average value of the most recent stable data collection period before the start of the equipment state transfer is used.

[0068] Assume the flow rate of the main water supply pipe is The return water main flow rate is Bypass traffic is , No. The flow rate of each operating branch is The number of running branches is The water balance is derived from the mass conservation law in a closed waterway, expressed as:

[0069] ;

[0070] in, For the running branch sequence number, This represents the total number of running branches.

[0071] It should be noted that the operating branch refers to the branch, excluding the bypass pipe, that supplies water to the water-using equipment from the main water supply pipe and flows into the return water main pipe.

[0072] To further clarify, the calibration accuracy of the flowmeter uses the values ​​recorded in the corresponding flowmeter calibration certificate. When the calibration accuracy is expressed as a percentage of the measured value, the absolute error is calculated by multiplying the current flow measurement value by the calibration accuracy. When the calibration accuracy is expressed as a percentage of full scale, the absolute error is calculated by multiplying the flowmeter's full scale by the calibration accuracy. The calibration accuracy of the temperature measuring point is converted into an absolute temperature error according to the values ​​recorded in the corresponding temperature measuring point calibration certificate. The current temperature measurement value minus the absolute temperature error is used as the lower temperature boundary, and the current temperature measurement value plus the absolute temperature error is used as the upper temperature boundary.

[0073] Full scale of a flow meter refers to the upper limit of the measurement range specified by the flow meter manufacturer and confirmed by a valid calibration certificate. It corresponds to the maximum volumetric flow rate that the flow meter can maintain with the specified measurement accuracy. If the upper limit of the measurement range recorded on the flow meter nameplate, technical documents and calibration certificate are inconsistent, the upper limit of the measurement range recorded in the most recent valid calibration certificate shall be adopted.

[0074] If the difference between the main water supply flow and the main return flow does not exceed the sum of their absolute errors, and the difference between the main water supply flow and the sum of the bypass flow and the flow of each operating branch does not exceed the sum of the absolute errors of the main water supply flow measurement point, the bypass flow measurement point, and the flow measurement point of each operating branch, then the candidate flow combination is deemed to satisfy water balance; if any difference exceeds the sum of the corresponding absolute errors, then the candidate flow combination is deemed not to satisfy water balance.

[0075] The heat exchange capacity of the main operating unit and the heat exchange capacity of the supply and return water mains are calculated based on the sensible heat transfer relationship of water. Let the first... The host's traffic is , No. The operating unit's water supply temperature is , No. The return water temperature of the main unit is The number of running hosts is The temperature of the main water supply pipe is The temperature of the return water main is The sum of the heat exchange capacity of the operating main unit and the heat exchange capacity of the supply and return water mains are respectively:

[0076] ;

[0077] ;

[0078] in, The sum of heat exchanged by the operating host. Heat exchange for the main supply and return water pipes; The density of water; The specific heat capacity of water at constant pressure; For the running host serial number, This represents the total number of running hosts.

[0079] The water density and specific heat capacity at constant pressure are selected based on the water physical properties corresponding to the average temperature of the supply and return water. If no water physical properties are available, the water density is taken as... The specific heat capacity of water at constant pressure is taken as .

[0080] The minimum and maximum sums of heat exchange of the main operating unit, the minimum and maximum sums of heat exchange of the main operating unit, and the minimum and maximum heat exchange of the main supply and return water pipes are calculated using the upper and lower error boundaries of the measured values ​​of the flow rate and temperature measurement points, respectively. When the upper boundary of the heat exchange interval of the main operating unit is not less than the lower boundary of the heat exchange interval of the main supply and return water pipes, and the upper boundary of the heat exchange interval of the main supply and return water pipes is not less than the lower boundary of the heat exchange interval of the main operating unit, it is determined that the two heat exchange intervals overlap and the candidate flow combination satisfies the heat balance. If the above interval overlap condition is not met, it is determined that the candidate flow combination does not satisfy the heat balance.

[0081] S1.2. For each flow measurement point, the effective flow confirmed in the previous control cycle is taken as the prior distribution center, and the absolute error obtained by converting the flow meter calibration accuracy is taken as the standard deviation of the prior distribution. A truncated normal prior distribution with a value not less than zero and not greater than the full scale of the flow meter is established. When there is no effective flow confirmed in the previous control cycle in the first control cycle, the current flow measurement value is taken as the prior distribution center.

[0082] Let the first The prior distribution center of each flow measurement point is: The standard deviation of the prior distribution is The full-scale flow meter is , will candidate real traffic Substituting the probability density of the truncated normal distribution, we obtain the... The prior probability density function of a flow measurement point is expressed as:

[0083] ;

[0084] when or hour, .

[0085] in, For the first Candidate true flow at each flow measurement point The prior probability density of the distribution; For the first The prior distribution center of each flow measurement point is used. If there is a confirmed valid flow in the previous control cycle, the confirmed valid flow in the previous control cycle is taken. In the first control cycle, the current flow measurement value is taken. For the first The standard deviation of the prior distribution of each flow measurement point is taken as the absolute error obtained by converting the flow meter calibration accuracy. For the first The full scale of the flow meter corresponding to each flow measurement point; This is the cumulative distribution function of the standard normal distribution.

[0086] It should be noted that when the flowmeter calibration certificate records the standard uncertainty, the standard uncertainty is used as the prior distribution standard deviation; when the flowmeter calibration certificate records the expanded uncertainty and the coverage factor, the ratio of the expanded uncertainty to the coverage factor is used as the standard uncertainty; when the flowmeter calibration certificate only records the calibration accuracy, the absolute error obtained by converting the calibration accuracy is used as the conservative prior distribution standard deviation.

[0087] Using the current flow measurement as the observation, the likelihood probability between the candidate true flow and the current flow measurement is calculated using the normal distribution probability density function. The expression is as follows:

[0088] ;

[0089] in, For the first The likelihood probability density corresponding to the candidate true flow at each flow measurement point is as follows: the closer the candidate true flow is to the current flow measurement value, the greater the likelihood probability density. For flow measurement point serial numbers; For the first Candidate real traffic at each traffic measurement point; For the first Current flow measurement value at each flow measurement point; For the first The observation standard deviation of each flow measurement point.

[0090] It should be noted that the observation standard deviation is obtained according to the standard uncertainty of the flowmeter corresponding to the current flow measurement value; when the flowmeter calibration certificate records the expanded uncertainty, it is converted into the standard uncertainty according to the corresponding coverage factor; when the flowmeter calibration certificate only records the calibration accuracy, the absolute error obtained by converting the calibration accuracy is used as the conservative observation standard deviation.

[0091] For each candidate flow combination, the candidate true flow rates are substituted into the prior probability density and likelihood probability density of the corresponding flow measurement point. The prior probability density and likelihood probability density of the same flow measurement point are multiplied, and then the products of all flow measurement points are multiplied to obtain the unnormalized posterior probability of the candidate flow combination. The heat exchange corresponding to the candidate flow combination is calculated using the supply and return water temperature difference of the corresponding pipeline. The posterior probability of candidate flow combinations that do not meet the requirements of water balance, heat balance, non-negative flow rate, and full-scale flowmeter is set to zero, resulting in the joint posterior distribution of the candidate flow combinations. The corresponding candidate true flow rates are extracted from the joint posterior distribution for each flow measurement point to obtain the marginal posterior distribution of each flow measurement point. The corresponding pipeline temperature change only participates in the heat exchange calculation and heat balance verification and is not considered as a repeated observation for the flow likelihood probability.

[0092] Markov chain Monte Carlo sampling is carried out simultaneously according to the posterior distribution of each flow measurement point. The next candidate flow combination is generated by using a symmetric normal random walk. The center of the symmetric normal random walk is taken as the current flow combination, and the standard deviation of the symmetric normal random walk is taken as the absolute error of the corresponding flow meter. If the candidate flow combination has a negative flow or exceeds the full range of the flow meter, the candidate flow combination is rejected. If the candidate flow combination does not pass the water balance and heat balance at the same time, the candidate flow combination is also rejected.

[0093] When a candidate flow combination simultaneously passes through both water and heat balance, the ratio of the posterior probability of the candidate flow combination to the posterior probability of the current flow combination is calculated. If the posterior probability ratio is not less than 1, the candidate flow combination is accepted. If the posterior probability ratio is less than 1, a uniformly random number between 0 and 1 is generated. If the uniformly random number is not greater than the posterior probability ratio, the candidate flow combination is accepted. If the uniformly random number is greater than the posterior probability ratio, the current flow combination is retained. Symmetric normal random walks have the same proposal probability in both forward and backward directions; therefore, the posterior probability ratio does not require additional proposal probability correction.

[0094] In one embodiment, four Markov chains are run in parallel, each performing 2000 samplings. The first 1000 samplings are discarded, and the remaining sampling results are used for convergence verification. When the latent scaling statistic for each flow measurement point is no greater than 1.05, the Markov chains are confirmed to have converged, and the flow combinations retained by the four Markov chains are merged into a valid flow sample set. When the latent scaling statistic is greater than 1.05, each Markov chain continues to perform 1000 more samplings until the latent scaling statistic is no greater than 1.05 or a single Markov chain has accumulated 8000 samplings. The latent scaling statistic is obtained by comparing the sample differences between multiple Markov chains with the sample differences within a single Markov chain. A latent scaling statistic close to 1 indicates that all Markov chains have converged to the same posterior distribution.

[0095] If a single Markov chain fails to converge after 8000 samplings, the flow rate confidence range and heat exchange demand confidence range from the previous control cycle are used in this control cycle, and the equipment switching permission is set to maintain equipment permission. If the first control cycle does not have the results of the previous control cycle, a temporary flow rate confidence range is formed by adding or subtracting the flow meter absolute error from the current flow rate measurement value, and the equipment switching permission is set to maintain equipment permission. The lower boundary of the temporary flow rate confidence range is taken as zero when it is less than zero, and the upper boundary of the temporary flow rate confidence range is taken as the flow meter full scale when it exceeds the full scale of the flow meter. The upper and lower boundaries of the temporary flow rate confidence range are substituted into the heat exchange calculation relationship of the supply and return water mains recorded in S1.1, and combined with the upper and lower error boundaries of the supply and return water temperatures, a temporary heat exchange demand confidence range is formed. The temporary flow rate confidence range and the temporary heat exchange demand confidence range are only used to maintain the current number of equipment and are not used to add or reduce equipment.

[0096] S1.3. For the converged valid flow sample set, arrange the valid flow samples for each flow measurement point separately. Use the median of the valid flow samples as the corrected flow value for the corresponding flow measurement point. Use the 2.5 percentile of the valid flow samples as the lower boundary of the flow confidence range, and the 97.5 percentile of the valid flow samples as the upper boundary of the flow confidence range. If the lower boundary of the flow confidence range is less than zero, use zero; if the upper boundary of the flow confidence range exceeds the full-scale range of the flow meter, use the full-scale range. The corrected flow value for each flow measurement point, the valid flow sample set, and the flow confidence range together constitute the flow measurement point correction result. The 2.5 percentile to the 97.5 percentile correspond to the middle 95% of the valid flow sample set and are used to exclude a small number of extreme sampling results located at both ends of the sample distribution.

[0097] The flow confidence range preserves the range of flow variation caused by the combined effects of flow meter calibration error, water balance deviation, and heat balance deviation. Compared to directly using a single flow measurement value for equipment switching judgment, the flow confidence range can clearly define the flow measurement uncertainties faced by equipment control judgment.

[0098] Each candidate flow combination within the effective flow sample set is selected sequentially. The candidate water supply main flow rate in the candidate flow combination is substituted into the heat exchange calculation relationship of the supply and return water mains recorded in S1.1. The supply and return water main temperatures within the same stable acquisition period of this control cycle are used to obtain the heat exchange demand sample value corresponding to each candidate flow combination. The heat exchange demand sample value refers to the instantaneous heat exchange demand corresponding to the candidate water supply main flow rate under the temperature difference between the supply and return water mains in this control cycle. The heat exchange demand sample value follows the sample number of the corresponding candidate flow combination, so that the heat exchange demand sample value and the candidate flow rate of each pipeline are in one-to-one correspondence.

[0099] When the candidate flow combinations corresponding to the heat exchange demand sample values ​​simultaneously satisfy both water balance and heat balance, the heat exchange demand sample values ​​are retained; when the candidate flow combinations do not simultaneously satisfy both water balance and heat balance, the corresponding heat exchange demand sample values ​​are discarded. The retained heat exchange demand sample values ​​are arranged in ascending order, and the 2.5 percentile is used as the lower boundary of the heat exchange demand confidence range, and the 97.5 percentile is used as the upper boundary of the heat exchange demand confidence range, thus obtaining the confidence range of the heat exchange demand.

[0100] Each heat demand value within the reliable range of heat demand can be correlated with a candidate flow combination within the reliable range of flow rate, and all are verified by heat balance on the operating host side and the main supply and return water pipe side, thereby avoiding the masking of flow measurement errors by using a single point to calculate heat demand with a corrected flow rate value.

[0101] S1.4. Obtain the minimum allowable load adjustment amount of the host controller from the host controller technical documents, take the lower boundary of the heat exchange demand confidence range as the starting point of the first continuous demand segment, and continuously divide the heat exchange demand confidence range according to the minimum allowable load adjustment amount of the host controller; if the last continuous demand segment is less than one minimum load adjustment amount, take the upper boundary of the heat exchange demand confidence range as the end point of the last continuous demand segment.

[0102] Obtain the maximum heat exchange capacity, minimum stable output, minimum allowable flow rate, and maximum allowable flow rate of the current equipment combination from the operating host technical documents and on-site commissioning records, and obtain control conclusions segment by segment according to the following rules:

[0103] For each continuous demand segment, select all valid flow samples whose heat exchange demand sample values ​​fall within the continuous demand segment, and extract the maximum value of the main water supply flow from the corresponding candidate flow combination as the upper boundary of the flow confidence range corresponding to the continuous demand segment; when there are no valid flow samples within the continuous demand segment, select the valid flow sample with the smallest difference between the heat exchange demand sample value and the midpoint of the continuous demand segment, and take the main water supply flow from the corresponding candidate flow combination as the flow corresponding to the continuous demand segment.

[0104] When the lower boundary of the continuous demand segment is greater than the maximum heat exchange capacity of the current equipment combination, or the upper boundary of the flow confidence range corresponding to the continuous demand segment is greater than the maximum allowable flow of the current equipment combination, the corresponding equipment control conclusion for the continuous demand segment is increased.

[0105] When the upper boundary of the continuous demand segment is less than the minimum stable output of the current equipment combination, and the upper boundary of the flow confidence range corresponding to the continuous demand segment is less than the minimum allowable flow of the current equipment combination, the continuous demand segment corresponds to a reduction in equipment control conclusion.

[0106] When the conclusions for increasing or decreasing equipment control are not met, the corresponding equipment control conclusions for continuous demand sections shall be maintained.

[0107] When the heat exchange capacity verification result and the allowable flow rate verification result of the same continuous demand section point to different control conclusions, the equipment control conclusion should be maintained for the continuous demand section.

[0108] When all continuous demand segments correspond to an increase in equipment control conclusions, the equipment switching permission is an increase in equipment permission; when all continuous demand segments correspond to a decrease in equipment control conclusions, the equipment switching permission is a decrease in equipment permission; when all continuous demand segments correspond to maintaining equipment control conclusions, the equipment switching permission is a maintain in equipment permission; when there are different control conclusions between continuous demand segments, the equipment switching permission is a maintain in equipment permission, and the current number of equipment remains unchanged. Equipment switching permission only allows changes in the number of equipment when the control conclusions within the reliable range of heat exchange demand remain consistent.

[0109] The reliable flow range of each pipeline is used as the equipment operating flow boundary, the reliable heat exchange demand range is used as the equipment heat exchange capacity boundary, and the equipment switching permission is used as the direction of equipment status change. When the equipment switching permission is to add equipment, the direction of equipment status change is limited to maintaining the current number of equipment and adding equipment; when the equipment switching permission is to reduce equipment, the direction of equipment status change is limited to maintaining the current number of equipment and reducing equipment; when the equipment switching permission is to maintain equipment, the direction of equipment status change is limited to maintaining the current number of equipment. A correspondence is established based on the equipment operating flow boundary, the equipment heat exchange capacity boundary, and the direction of equipment status change to generate the equipment switching boundary.

[0110] S2. Filter candidate states that meet the requirements of heat exchange capacity, flow rate, stable output, start-stop time limit and interlocking conditions according to the equipment switching boundary, retain the transfer path with the lowest switching energy consumption and continuous heat exchange, and obtain the target equipment state and transfer sequence.

[0111] S2.1. The IoT controller uses device switching boundaries to confirm the current device status based on the operational feedback from the host, water pump, cooling tower fan, electric valve, and bypass valve. The current device status includes the operating status, stop status, host load, water pump frequency, cooling tower fan frequency, and valve position of each device. If any of the following occurs in the device operation feedback: communication interruption, protection action, or continuous inconsistency between control commands and device feedback within a control cycle, the corresponding device will be marked as an abnormal device. Standby devices with normal communication, no protection action, and start-up permission are healthy standby devices.

[0112] Extract the main unit and supporting equipment currently performing heat exchange tasks from the current equipment status to obtain the current equipment combination. The main unit and supporting equipment refer to a single main unit and the electric valves, bypass valves, water pumps, and cooling tower fans that provide the necessary flow, heat exchange, and heat dissipation conditions for that main unit according to the actual piping connections and start-stop interlocking relationships in the HVAC room. When multiple main units share water pumps, cooling tower fans, or valves, the shared equipment is included in the relevant equipment combination according to the actual interlocking relationships; it is not required that each main unit correspond one-to-one with each supporting device. During cooling operation, supporting equipment includes the actually put into operation chilled water pumps, cooling water pumps, and cooling tower fans; during heating operation, supporting equipment includes the actually put into operation hot water pumps.

[0113] The licensed device state is established according to the direction of device state change defined by the device switching boundary: when the direction of device state change is to maintain the current number of devices, the current device combination is used as the licensed device state; when the direction of device state change includes increasing the number of devices, a healthy standby host and supporting device are added to the current device combination, and a corresponding licensed device state is obtained each time a host and supporting device is added; when the direction of device state change includes decreasing the number of devices, a running host and supporting device are removed from the current device combination, and a corresponding licensed device state is obtained each time a host and supporting device is removed.

[0114] When the licensed device status includes abnormal devices, a set of healthy backup host and supporting equipment is used to replace the abnormal devices, while keeping the number of devices in the licensed device status and the direction of device status change unchanged; if there are no healthy backup devices that meet the replacement conditions, the licensed device status containing abnormal devices is removed, and the device status to be verified is obtained.

[0115] S2.2. For each device status to be verified, obtain the maximum permissible output, minimum stable output, minimum permissible flow rate, maximum permissible flow rate, minimum continuous operating time, and minimum continuous downtime of each operating host from the operating host technical documents and on-site commissioning records; the maximum permissible output and minimum stable output are obtained according to the values ​​in the technical documents under the supply and return water temperatures and corresponding permissible flow ranges of this control cycle.

[0116] Based on the operating host included in the status of the equipment to be verified, the upper boundary of the total heat exchange capacity is obtained by accumulating the maximum allowable output, and the lower boundary of the total heat exchange capacity is obtained by accumulating the minimum stable output. The total allowable flow range is determined according to the current pipeline connection relationship: the minimum allowable flow and the maximum allowable flow are accumulated for parallel branches respectively, and the overlapping intervals of the allowable flow ranges of the relevant equipment are used for shared main pipes, series equipment and shared water pumps; if there are no overlapping intervals, the corresponding equipment status to be verified is eliminated.

[0117] When the upper boundary of the total heat exchange capacity is not less than the upper boundary of the reliable range of heat exchange demand, the lower boundary of the total heat exchange capacity is not greater than the lower boundary of the reliable range of heat exchange demand, and the total allowable flow range covers the reliable flow range of each pipeline in the equipment switching boundary, the status of the equipment to be verified is retained and marked as a qualified thermal and hydraulic status; if any condition is not met, the corresponding status of the equipment to be verified is removed.

[0118] For each qualified thermal and hydraulic status, the continuous operating time of each operating host is compared with the corresponding minimum continuous operating time, and the continuous shutdown time of each stopped host is compared with the corresponding minimum continuous shutdown time. When adding a host and supporting equipment, if the continuous shutdown time of the host and supporting equipment to be put into operation reaches the corresponding minimum continuous shutdown time, the start-up and shutdown time limit is deemed qualified. When removing a host and supporting equipment, if the continuous operating time of the host and supporting equipment to be removed reaches the corresponding minimum continuous operating time, the start-up and shutdown time limit is deemed qualified. When maintaining the current number of equipment, the status of equipment that does not meet the start-up and shutdown time limit is not changed.

[0119] Verify the interlocking conditions according to the equipment controller technical documents and on-site commissioning records. When adding a main unit and supporting equipment, verify in sequence that the electric valve has the permission to open, the water pump has the permission to start, the cooling tower fan has the permission to start, and the main unit has the permission to start; when removing a main unit and supporting equipment, verify that the remaining operating main unit can handle the current load of the removed main unit, and confirm that the main unit and supporting equipment to be removed have the permission to stop. When both the start / stop time limit and the interlocking conditions are qualified, retain the qualified thermal and hydraulic status as an executable candidate status; if any start / stop time limit or interlocking condition is unqualified, remove the corresponding qualified thermal and hydraulic status.

[0120] If no executable candidate state is formed, maintain the current device state and the confirmed device control variables, do not perform subsequent device switching and relay-type extreme value adjustment, and regenerate the device switching boundary in the next control cycle.

[0121] S2.3. Using the current device state as the starting state of each state transition path, compare the device differences between the current device state and each executable candidate state. According to the stage sequence of valve opening, water pump establishing flow, host starting, load transfer between hosts, original host stopping, and auxiliary equipment exiting, retain the state transition stages required to complete the corresponding device differences, and obtain the transition path corresponding to each executable candidate state.

[0122] The valve opening phase is completed when the electric valve reaches the valve position command. The water pump flow establishment phase is completed when the water pump operation feedback is normal and the actual flow rate enters the reliable flow range within the equipment switching boundary. The main unit startup phase is completed when the main unit operation feedback is normal and the main unit output is not lower than the minimum stable output. The load transfer phase between main units is completed when the sum of the heat exchange of the operating main units covers the upper boundary of the reliable heat exchange demand range. The original main unit shutdown phase is completed when the load of the main unit to be shut down drops to the stop permission range and a stop feedback is obtained. The auxiliary equipment decommissioning phase is completed when the water pump, cooling tower fan, and valve complete the corresponding stop and close commands. After the previous state transition phase meets the completion conditions, the next state transition phase continues, forming a continuous intermediate equipment state.

[0123] Except for the load transfer phase between mainframes, each state transfer phase only changes one operating state corresponding to a set of interlocked equipment; the load transfer phase between mainframes is a coordinated transfer phase, which simultaneously reduces the load transferred out of the mainframe and increases the load transferred into the mainframe, while the sum of the loads of the operating mainframes remains unchanged.

[0124] The upper boundary of the total heat exchange capacity and the total allowable flow range of the remaining operating equipment are recalculated based on the intermediate equipment status. If either the upper boundary of the total heat exchange capacity is less than the upper boundary of the reliable range of heat exchange demand or the total allowable flow range does not cover the equipment operating flow boundary, it is determined that an effective heat exchange interruption has occurred in the corresponding state transition stage, and the transition path containing the corresponding state transition stage is removed. If both checks pass, the corresponding intermediate equipment status is retained and the next state transition stage is established.

[0125] S2.4. For each retained state transition phase, obtain the actual power and phase duration of each device from the operation record of the most recent state transition of the same device; if there is no operation record of the same device's state transition, obtain the actual power and phase duration of each device from the field commissioning record. The actual power is the average value of the device power measurement within the state transition phase, and the phase duration is calculated from the time the state transition command is issued until the corresponding phase completion condition is met.

[0126] Based on the relationship that electrical energy equals the product of power and operating time, the actual power of the equipment and the duration of each state transition phase are accumulated to obtain the switching energy consumption of the transition path, expressed as:

[0127] ;

[0128] in, Energy consumption for switching transfer paths; This refers to the state transition stage number; This represents the total number of state transition stages. For the first The device serial number within each state transition phase; For the first The number of devices that are in operation and consume electrical energy during each state transition phase; For the first Within the state transition phase, the first The average actual power of the operating equipment; For the first The duration of each state transition phase.

[0129] The transition paths are processed step-by-step according to the state transition stages. For each intermediate device state formed, the energy consumption of the path switching from the initial state to the intermediate device state is accumulated. When multiple transition paths reach the same intermediate device state, only the transition path with the lowest accumulated switching energy consumption is retained. If the accumulated switching energy consumption is the same, the transition path with the fewest device start-stop cycles is retained, resulting in candidate transition paths corresponding to each executable candidate state. Step-by-step deletion of transition paths that have experienced effective heat exchange interruptions and have high accumulated switching energy consumption avoids a complete enumeration of all device start-stop sequences, while retaining low-switching-energy-constraint paths that meet thermo-hydraulic constraints.

[0130] S2.5. The control cycle duration recorded by the IoT controller is used as the unified energy consumption calculation period. For each candidate transfer path, the duration of each state transfer stage is accumulated, and the accumulated duration is subtracted from the control cycle duration to obtain the runtime after the device switchover is completed; if the accumulated duration is not less than the control cycle duration, the corresponding candidate transfer path is eliminated.

[0131] The operating power of the executable candidate state is obtained from the most recent operating record of the same equipment state where the heat exchange demand is within the reliable range of heat exchange demand. If no operating record of the same equipment state exists, the main unit power, water pump power, and cooling tower fan power corresponding to the upper boundary of the reliable range of heat exchange demand are obtained according to the power curve recorded in the main unit's technical documents, and these are accumulated to obtain the operating power of the executable candidate state. The operating power of the executable candidate state is multiplied by the running time after the equipment switchover is completed to obtain the equipment operating energy consumption. The equipment operating energy consumption is then added to the switching energy consumption of the candidate transfer path to obtain the total energy consumption of the executable candidate state, expressed as:

[0132]

[0133] in, The total energy consumption of the executable candidate state; Energy consumption for switching candidate transfer paths; The device operating power for the executable candidate state; To control cycle duration; For the first The duration of each state transition phase.

[0134] The current equipment combination's state is included in the executable candidate states. The switching energy consumption corresponding to the current equipment combination is kept at zero, and the equipment operating energy consumption is calculated using the current total power of the data center for the current equipment combination. The total energy consumption of each executable candidate state is compared, and the executable candidate state with the lowest total energy consumption and which maintains effective heat exchange throughout the corresponding candidate transfer path is selected as the target equipment state. If the total energy consumption of multiple executable candidate states is the same, the executable candidate state with the lowest switching energy consumption and the fewest equipment start-stop times is retained in that order. If the switching energy consumption and the number of equipment start-stop times are still the same, the current equipment state is retained.

[0135] The target device state and transition sequence are formed according to the state transition stages and their order contained in the candidate transition paths corresponding to the target device state.

[0136] S3. Based on the target equipment status and transfer sequence, switch equipment and sequentially perform relay-type extreme value adjustment on the cooling tower fan, circulating water pump and main unit load within the equipment switching boundary to generate a positive low-power operating point.

[0137] S3.1. In accordance with the transfer sequence, the valve opening, water pump flow establishment, main unit start-up, load transfer between main units, original main unit shutdown, and auxiliary equipment shutdown are executed in sequence, and each state transfer stage is verified by stage completion conditions; when the previous state transfer stage meets the corresponding stage completion conditions, the next state transfer stage is executed; when the previous state transfer stage does not meet the corresponding stage completion conditions, the completed equipment state is maintained and the subsequent state transfer stages are stopped.

[0138] When all state transition stages meet the corresponding stage completion conditions, verify the actual operating equipment, main unit load, water pump frequency, cooling tower fan frequency, and valve position; when the verification result is consistent with the target equipment state, complete the switch from the current equipment state to the target equipment state, and use the control quantity under the target equipment state as the starting control quantity for relay-type extreme value regulation; when the verification result is inconsistent with the target equipment state, do not execute relay-type extreme value regulation.

[0139] S3.2. Keep the number and combination of equipment as defined by the target equipment status unchanged, and arrange the control quantities in the order of cooling tower fan, circulating water pump and main unit load; when cooling, the circulating water pump is arranged in the order of cooling water pump and chilled water pump, and when heating, the circulating water pump corresponds to the hot water pump; the main unit load refers to the load set value of each operating main unit, and when adjusting the main unit load, keep the sum of the operating main unit loads within the reliable range of heat exchange demand.

[0140] For each control variable, the instruction resolution is obtained from the corresponding device controller technical document, and the minimum change that can form stable operation feedback is obtained from the field commissioning record. The larger value between the instruction resolution and the minimum change is taken as the minimum stable adjustment value of the corresponding control variable. If the minimum change value is not recorded in the field commissioning record, the instruction resolution is taken as the minimum stable adjustment value.

[0141] The initial adjustment value is obtained from the last effective adjustment record of the current control quantity in the same target equipment state of the previous control cycle. If no effective adjustment record exists, the commonly used stable adjustment value is obtained from the field commissioning record as the initial adjustment value. If the field commissioning record does not record the commonly used stable adjustment value, the minimum stable adjustment value is used as the initial adjustment value. The initial adjustment value must not be less than the minimum stable adjustment value, and must not cause the control quantity to exceed the allowable range of the corresponding equipment.

[0142] The cooling tower fan frequency, circulating water pump frequency, and main unit load are adjusted using their respective current adjustment values. Only one control value is changed at a time, while the other control values ​​remain at the previous confirmed value. This way, the change in the total power of the computer room is mapped to the current control value, avoiding the inability to distinguish the source of the change in the total power of the computer room due to the simultaneous change of multiple control values.

[0143] S3.3. When there is a relay-type extreme value adjustment record with the same target equipment state in the previous control cycle, the last adjustment direction that can reduce the total power of the machine room in the previous control cycle shall be selected as the initial search direction of the current control quantity; when there is no relay-type extreme value adjustment record with the same target equipment state in the previous control cycle, the cooling tower fan and circulating water pump shall adopt the direction of reducing frequency as the initial search direction, and the main load shall adopt the direction of reducing the total load of the main unit as the initial search direction, provided that the sum of the operating main unit loads is not lower than the upper boundary of the reliable range of heat exchange demand.

[0144] Change the current adjustment value of the current control variable along the initial search direction to obtain the water supply temperature, circulation flow rate, pipeline pressure difference, main unit flow rate, equipment power, and total power of the machine room before and after adjustment. The total power of the machine room before adjustment is the average total power of the machine room during the stable operation period before the adjustment command is issued, and the total power of the machine room after adjustment is the average total power of the machine room during the stable operation period after the equipment completes the response.

[0145] Equipment completion response means that after the adjustment command is issued, the response time recorded in the equipment controller technical document is reached, and the difference between the maximum and minimum values ​​of each parameter, such as water supply temperature, circulation flow rate, pipeline pressure difference, and equipment power, does not exceed the corresponding stable fluctuation range recorded in the on-site commissioning record within the stable judgment time recorded in the on-site commissioning record; if the on-site commissioning record does not record the stable fluctuation range, the absolute measurement error of the corresponding measuring point shall be used as the stable fluctuation range.

[0146] When the water supply temperature, circulation flow rate, pipeline pressure difference, and equipment power all meet the above conditions, the stability judgment condition is determined to have been met, and the measured values ​​within the corresponding stability judgment period are used for subsequent comparison of the total power of the equipment room. If any parameter does not meet the above conditions, the current control quantity is maintained, and the water supply temperature, circulation flow rate, pipeline pressure difference, and equipment power are continuously collected. If the stability judgment condition is still not met when the remaining time of this control cycle is less than one stability judgment period, the control quantity before adjustment is restored, the current search direction is marked as an unexecutable direction, and subsequent adjustments in the current search direction are stopped.

[0147] S3.4. The measurement uncertainty range of the main power meter in the computer room shall be based on the absolute error of power measurement recorded in the most recent valid calibration certificate. If the difference between the total power of the computer room before adjustment and the total power of the computer room after adjustment is greater than the sum of the absolute errors of the two power measurements before and after adjustment, it shall be determined that the total power of the computer room has decreased, the adjusted control quantity shall be retained, and the current adjustment quantity shall be changed again along the original search direction.

[0148] If the difference between the total power of the computer room after adjustment and the total power of the computer room before adjustment is greater than the sum of the absolute errors of the two power measurements before and after adjustment, it is determined that the total power of the computer room has increased. The control quantity before adjustment is restored, the search direction is changed to the opposite direction, and the current adjustment quantity is reduced by a minimum stable adjustment quantity. If the current adjustment quantity after reduction is less than the minimum stable adjustment quantity, the minimum stable adjustment quantity is taken.

[0149] If the difference between the total power of the computer room before and after adjustment does not exceed the sum of the absolute errors of the two power measurements, it is determined that no identifiable change in the total power of the computer room has occurred in the current search direction. The control quantity before adjustment is restored and the search is changed to the opposite direction. If no identifiable decrease in the total power of the computer room has occurred in either of the two search directions, the control quantity corresponding to the lowest total power of the computer room is fixed as the confirmed value of the current control quantity, and the next control quantity is entered according to the control quantity sequence arranged in S3.2.

[0150] S3.5. After each adjustment of the current control quantity is completed and the stability judgment condition is reached, the actual flow rate of each pipeline is compared with the corresponding equipment operating flow rate boundary in the equipment switching boundary item by item. The sum of the heat exchange of the operating host calculated according to the sensible heat exchange relationship recorded in S1.1 is compared with the equipment heat exchange capacity boundary in the equipment switching boundary. At the same time, the water supply temperature, the most unfavorable pipeline pressure difference, the host flow rate and the equipment protection status are verified.

[0151] When the water supply temperature is within the allowable range of the main controller's technical documents and on-site commissioning records, the water supply temperature verification is passed; when the water supply temperature is lower than the lower boundary of the allowable range or higher than the upper boundary of the allowable range, the water supply temperature is determined to be deviated.

[0152] The most unfavorable pipeline pressure difference refers to the minimum supply and return water pressure difference at the end of each operating branch connected to the current equipment combination. When the most unfavorable pipeline pressure difference is not lower than the minimum allowable pressure difference recorded in the on-site commissioning record, the pipeline pressure difference verification is passed. When the most unfavorable pipeline pressure difference is lower than the minimum allowable pressure difference, the most unfavorable pipeline pressure difference is determined to be insufficient.

[0153] The host traffic verification is passed when the actual traffic of each running host is not less than the corresponding minimum allowed traffic obtained by S2.2; if the actual traffic of any running host is less than the corresponding minimum allowed traffic, the host traffic is determined to be insufficient.

[0154] When the actual flow rate of each pipeline is within the operating flow rate boundary of the corresponding equipment, and the sum of the heat exchange of the operating main unit is within the heat exchange capacity boundary of the equipment, the equipment switching boundary verification is passed; when the actual flow rate of any pipeline leaves the operating flow rate boundary of the corresponding equipment, it is determined that the flow rate exceeds the equipment switching boundary, and when the sum of the heat exchange of the operating main unit leaves the heat exchange capacity boundary of the equipment, it is determined that the heat exchange capacity exceeds the equipment switching boundary.

[0155] If any of the following situations occur during the adjustment process: water supply temperature deviation, insufficient pressure difference in the most unfavorable pipeline, insufficient main unit flow, equipment protection action, flow exceeding the equipment switching boundary, or heat exchange capacity exceeding the equipment switching boundary, the control quantity before adjustment is restored, and the current search direction is marked as an unexecutable direction; if the opposite search direction has not yet been executed, the opposite search direction is switched to. If any of the above situations also occur in the opposite search direction, the restored control quantity is fixed as the confirmed value of the current control quantity.

[0156] If no cooling tower fan is running in the target equipment status, the relay-type extreme value regulation of the cooling tower fan is not performed, and the regulation is directly performed according to the order of the circulating water pump and the main unit load.

[0157] After the cooling tower fan completes relay-type extreme value adjustment in two search directions, the confirmed value of the cooling tower fan is used as the fixed value during the adjustment period of the circulating water pump; after the circulating water pump completes relay-type extreme value adjustment in two search directions, the confirmed values ​​of the cooling tower fan and the circulating water pump are used as the fixed values ​​during the adjustment period of the main unit load; after the main unit load completes relay-type extreme value adjustment in two search directions, the positive adjustment of all control quantities is completed.

[0158] The target equipment status, the confirmed value of the cooling tower fan, the confirmed value of the circulating water pump, the confirmed value of the main unit load, and the corresponding total power of the machine room are combined to generate a positive low-power operating point; when there is no operating cooling tower fan in the target equipment status, the positive low-power operating point does not include the confirmed value of the cooling tower fan frequency.

[0159] S4. Adjust the main unit load, circulating water pump and cooling tower fan in reverse order from the positive low power operating point, check the water supply temperature, pipeline pressure difference and main unit flow, and form the group control operating point with the lowest total power.

[0160] S4.1. Extract the target equipment status, main unit load confirmation value, circulating water pump frequency confirmation value, cooling tower fan frequency confirmation value, and total power of the machine room from the positive low-power operating point. Keep the number and combination of equipment defined by the target equipment status unchanged, and arrange the control quantities in the order of main unit load, circulating water pump, and cooling tower fan to obtain the reverse order adjustment sequence. When there is no operating cooling tower fan in the target equipment status, the reverse order adjustment sequence does not include the cooling tower fan.

[0161] During reverse-order adjustment, only one control variable is changed each time. The control variable that has not yet entered the adjustment maintains the confirmed value corresponding to the positive low power operating point, and the control variable that has completed reverse-order adjustment maintains the corresponding reverse-order adjustment result, so that each change in the power of the machine room can correspond to the current control variable.

[0162] Before the reverse-order adjustment begins, the initial adjustment directions of the circulating water pump and the cooling tower fan are determined respectively: if there is a final frequency adjustment direction that causes the corresponding total power to decrease during the formation of the positive low-power operating point, the final frequency adjustment direction is taken as the initial adjustment direction of the corresponding control quantity; if there is no final frequency adjustment direction that causes the corresponding total power to decrease, the frequency reduction direction is taken as the initial adjustment direction of the corresponding control quantity.

[0163] The host load is selected according to the reverse order of adjustment. The ratio of the average actual power to the average actual heat exchange during the stable determination period of the positive low power operation point of each host is used as the unit heat exchange power consumption of each host. The average actual heat exchange is obtained according to the heat exchange calculation relationship of the host recorded in S1.1. The average actual power is the average value of the power measurement value of the corresponding host. The unit heat exchange power consumption is calculated only for the host whose average actual heat exchange is not lower than the minimum stable output of the corresponding host.

[0164] The operating units are arranged in descending order of unit heat exchange power consumption. The operating unit with the largest unit heat exchange power consumption is selected as the load transfer-out unit, and the operating unit with the smallest unit heat exchange power consumption is selected as the load transfer-in unit. When there are fewer than 2 operating units, the unit load is not changed, and the unit load confirmation value at the positive low power operating point is used as the result of the reverse order adjustment of the unit load.

[0165] When there are no fewer than two operating main units, the minimum stable adjustment of the main unit load is uniformly converted into heat exchange power. The minimum stable adjustment of the main unit load is reduced by one from the load-out main unit, while the same amount of main unit load adjustment is added to the load-in main unit. The sum of the operating main unit loads remains unchanged, and the frequency of the circulating water pump and the frequency of the cooling tower fan are kept at the confirmed values ​​corresponding to the positive low power operating point.

[0166] After the host load completes the response, the sum of the heat exchange of the operating host before and after the load transfer is compared, and the flow rate of the main water supply pipe before and after the load transfer is compared. If the difference in the sum of the heat exchange of the operating host does not exceed the error range corresponding to the heat exchange interval recorded in S1.1, and the difference in the flow rate of the main water supply pipe does not exceed the absolute error of the flow rate measurement point of the main water supply pipe, it is determined that the total heat exchange and the total circulation flow remain unchanged; otherwise, the host load before the load transfer is restored.

[0167] When the total heat exchange and total circulation flow remain constant, compare the total power of the computer room before and after the load transfer; if the decrease in the total power of the computer room is greater than the sum of the absolute errors of the two power measurements, retain the host load after the load transfer and continue to transfer one host load minimum stability adjustment in the same direction; if the increase in the total power of the computer room is greater than the sum of the absolute errors of the two power measurements, or if the change in the total power of the computer room does not exceed the sum of the absolute errors of the two power measurements, restore the previous host load.

[0168] The host load transfer is performed sequentially until the next load transfer causes the load transfer-out host to fall below the minimum stable output, the load transfer-in host to exceed the maximum allowable output, the total heat exchange changes, the total circulation flow changes, or the total power of the computer room does not continue to decrease. The load of each operating host corresponding to the lowest total power of the computer room is taken as the result of the host load reverse sequence adjustment.

[0169] S4.2. Select circulating water pumps according to the reverse order of adjustment, take the result of the reverse order adjustment of the main load as the fixed value of the main load, and adjust the frequency of each circulating water pump in reverse order of the circulating water pump sequence listed in S3.2; each time, only change the current circulating water pump frequency by a minimum stable adjustment amount, and keep the frequencies of the other circulating water pumps and cooling tower fans at the previously confirmed value.

[0170] Change the current circulating water pump frequency by a minimum stable adjustment amount along the initial adjustment direction of the circulating water pump. After the water supply temperature, circulating flow rate, pipeline pressure difference and equipment power reach the stability judgment conditions recorded in S3.3, verify the reliable range of the most unfavorable pipeline pressure difference, the actual flow rate of each operating host and the flow rate in the equipment switching boundary.

[0171] When the most unfavorable pipeline pressure difference is not lower than the minimum allowable pressure difference recorded in S3.5, the actual flow rate of each operating host is not lower than the corresponding minimum allowable flow rate, and the actual flow rate of each pipeline is within the corresponding reliable flow rate range, the average actual power of the operating host, circulating water pump and supporting equipment during the stable judgment period is accumulated to obtain the total power of the circulating water pump after adjustment; if any verification condition is not met, the frequency of the circulating water pump before adjustment is restored, and the current adjustment direction is marked as an unexecutable direction.

[0172] If the total power of the circulating water pump after adjustment is lower than the total power before adjustment, and the power decrease is greater than the sum of the absolute errors of the two power measurements, the adjusted circulating water pump frequency is retained, and the minimum stable adjustment amount is changed along the current adjustment direction. If the total power of the circulating water pump after adjustment is higher than the total power before adjustment, or the power change does not exceed the sum of the absolute errors of the two power measurements, the circulating water pump frequency before adjustment is restored, and the adjustment direction is switched to the opposite direction.

[0173] When the current adjustment direction is marked as an unexecutable direction, the frequency of the circulating water pump before adjustment is restored; when the opposite adjustment direction has not yet been executed, switch to the opposite adjustment direction; when the opposite adjustment direction has been marked as an unexecutable direction, end the frequency adjustment of the current circulating water pump.

[0174] If neither adjustment direction yields a decrease in total power, or if both adjustment directions are marked as unexecutable, fix the circulating water pump frequency that has the lowest total power and meets the most unfavorable pipeline pressure difference, minimum main unit flow rate, and flow rate confidence range; sequentially perform reverse-order adjustment of each circulating water pump to obtain the reverse-order adjustment result of the circulating water pump.

[0175] S4.3. When there are operating cooling tower fans in the target equipment status, select the cooling tower fans according to the reverse order of adjustment, keep the reverse order adjustment results of the main load and the reverse order adjustment results of the circulating water pump unchanged, and adjust the frequency of the cooling tower fans one by one according to the minimum stable adjustment amount of the cooling tower fan frequency, changing the frequency of only one cooling tower fan at a time.

[0176] Change the current cooling tower fan frequency by a minimum stable adjustment amount along the initial adjustment direction of the cooling tower fan. After the water supply temperature, circulation flow rate, pipeline pressure difference, and equipment power reach the stability judgment conditions, accumulate the average actual power of the main unit, cooling water pump, and cooling tower fan during the stability judgment period to obtain the total power of the cooling tower fan after adjustment. Then check the water supply temperature, the most unfavorable pipeline pressure difference, the main unit flow rate, and the equipment protection status according to the verification method recorded in S3.5.

[0177] If the total power of the cooling tower fan decreases after adjustment, and the decrease is greater than the sum of the absolute errors of the two power measurements, the adjusted cooling tower fan frequency is retained, and the minimum stable adjustment amount is changed along the current adjustment direction. If the total power of the cooling tower fan increases after adjustment, the power change does not exceed the sum of the absolute errors of the two power measurements, or any operating condition fails verification, the cooling tower fan frequency before adjustment is restored, and the adjustment direction is switched to the opposite direction.

[0178] If any operating condition fails verification, the current adjustment direction is marked as an unexecutable direction; if the opposite adjustment direction has not yet been executed, switch to the opposite adjustment direction; if the opposite adjustment direction has been marked as an unexecutable direction, end the current frequency adjustment of the cooling tower fan.

[0179] If neither adjustment direction results in a decrease in total power, the frequency of the cooling tower fan with the lowest total power and which passes the operating condition verification is fixed. The reverse-sequence adjustment of each cooling tower fan is then performed sequentially. The reverse-sequence adjustment results of the main unit load, the circulating water pump, and the cooling tower fan are combined with the corresponding total power of the machine room to generate a reverse-sequence operating point. If no cooling tower fan is operating in the target equipment state, the reverse-sequence adjustment results of the main unit load, the circulating water pump, and the corresponding total power of the machine room are combined to generate a reverse-sequence operating point.

[0180] The cooling tower fan adjustment direction is determined by the total power of the main unit, cooling water pump, and cooling tower fan. This can simultaneously reflect the changes in main unit power caused by changes in cooling tower fan power and changes in cooling conditions, thus avoiding frequency selection based solely on the cooling tower fan's own power.

[0181] S4.4. Extract the water supply temperature, supply and return water pressure difference at the end of each operating branch, actual flow rate of each operating host and total power of the machine room corresponding to the forward low power operation point and the reverse sequence operation point, and verify them according to the allowable range of water supply temperature, minimum allowable pressure difference and minimum allowable flow rate of the host as recorded in S3.5.

[0182] When the water supply temperature is within the allowable range, the pressure difference in the most unfavorable pipeline is not lower than the minimum allowable pressure difference, and the actual flow rate of each operating unit is not lower than the corresponding minimum allowable flow rate, the corresponding operating point will be retained as a qualified operating point; if any condition is not met, the corresponding operating point will be removed.

[0183] If only one of the forward low-power operating points and the reverse-sequence operating points is qualified, the corresponding qualified operating point will be used as the group control operating point; if neither the forward low-power operating point nor the reverse-sequence operating point is qualified, the last operating point that simultaneously passed the verification of water supply temperature, the most unfavorable pipeline pressure difference and the main unit flow during the S3 and S4 adjustment process will be restored, and the restored operating point will be used as the group control operating point.

[0184] Compare the average total power of the computer room at each qualified operating point during the stability judgment period; if the difference between the average total power of the two computer rooms is greater than the sum of the absolute errors of the two power measurements, select the qualified operating point with the smaller average total power; if the difference between the average total power of the two computer rooms does not exceed the sum of the absolute errors of the two power measurements, retain the positive low power operating point.

[0185] The selected qualified operating points are combined with the target equipment status, main unit load, circulating water pump frequency, cooling tower fan frequency, valve position, and total power of the machine room to form the group control operating point with the lowest total power among the forward low-power operating points and the reverse sequence operating points.

[0186] S5. Issue commands for host load, frequency conversion and valve position through the group control operation point, verify the status and transfer sequence of target equipment, isolate abnormal equipment and connect to backup equipment, and generate energy-saving operation results for the computer room.

[0187] S5.1. Extract the target equipment status, host load, circulating water pump frequency, cooling tower fan frequency, and valve position contained in the group control operation point. Convert the host load corresponding to each operating host into host load command, convert the circulating water pump frequency and cooling tower fan frequency into variable frequency command that can be recognized by the corresponding frequency converter, and convert the valve position into valve position command that can be recognized by the electric valve controller.

[0188] The unit and value range of the host load command are determined according to the host controller technical documents, the unit and value range of the frequency conversion command are determined according to the corresponding frequency converter technical documents, and the valve position command adopts the valve opening degree expression method recorded in the electric valve controller technical documents.

[0189] When the control quantity in the group control operation point cannot be converted into an executable instruction within the allowable range of the corresponding equipment, the corresponding control quantity is not issued; when the control quantity in the group control operation point needs to be rounded due to the instruction resolution, the executable value closest to the group control operation point is selected, and the equipment switching boundary, water supply temperature, most unfavorable pipeline pressure difference and host flow are re-verified in accordance with the method described in S3.5. When all re-verifications pass, the executable value is issued; when any verification fails, the previous confirmed value of the corresponding control quantity is restored.

[0190] Commands are issued sequentially according to the transfer sequence, including valve opening, pump flow establishment, main unit startup, load transfer between main units, shutdown of the original main unit, and deactivation of auxiliary equipment. If the current equipment feedback indicates compliance with the requirements of the current transfer stage, the corresponding equipment status is not changed repeatedly, and the current transfer stage is directly verified. If the current equipment feedback indicates non-compliance with the requirements of the current transfer stage, the main unit load command, frequency converter frequency command, and valve position command required to complete the current transfer stage are issued. Commands for the next transfer stage are issued only after the previous transfer stage has been verified, preventing multiple transfer stages from executing simultaneously, which could lead to the main unit starting before the pump has established flow or the original main unit deactivating before the load has been transferred.

[0191] S5.2. After the instruction for each transfer stage is issued, continuously acquire the actual valve position, actual pump flow rate, actual pipeline pressure difference, actual heat exchange of the main unit, actual water supply temperature and actual equipment power according to the response time recorded in the corresponding equipment controller technical document, and confirm the completion of the operation feedback response according to the stability judgment conditions recorded in S3.3.

[0192] When the actual valve position reaches the position corresponding to the valve position command, and the difference between the actual valve position and the valve position command does not exceed the command resolution of the electric valve controller, the valve position execution is deemed qualified; when the valve position does not reach the position corresponding to the valve position command or the difference exceeds the command resolution, the valve position execution is deemed unqualified.

[0193] When the pump operation feedback is normal, the actual pump flow rate enters the corresponding equipment operation flow rate boundary in the equipment switching boundary, and the actual pipeline pressure difference is not lower than the minimum allowable pressure difference recorded in S3.5, the pump operation is deemed qualified; when any of the following occurs: the pump operation feedback is abnormal, the actual pump flow rate leaves the corresponding equipment operation flow rate boundary, or the actual pipeline pressure difference is lower than the minimum allowable pressure difference, the pump operation is deemed unqualified.

[0194] If the host machine is operating normally, the actual heat exchange of the host machine is between the minimum stable output and the maximum allowable output of the corresponding host machine, the sum of the actual heat exchange of all operating hosts covers the upper boundary of the reliable range of heat exchange demand, and the actual water supply temperature is within the allowable range of water supply temperature, then the host machine is deemed to be performing successfully; if any verification condition is not met, then the host machine is deemed to be performing unsuccessfully.

[0195] The actual power of the equipment is the average of the power measurements taken within the stable judgment period; the equipment power feedback is deemed qualified when the actual power of the equipment completes the response and no protection action occurs. The verification results of valves, pumps, main units, and auxiliary equipment in the current transfer phase are combined to obtain the equipment execution confirmation result for the current transfer phase.

[0196] S5.3. If all equipment execution confirmation results for the current transfer phase are qualified, and the actual operating equipment, main unit load, water pump frequency, cooling tower fan frequency, and valve position conform to the content corresponding to the current transfer phase in the target equipment status, then the current transfer phase is confirmed to be completed, and the next transfer phase in the transfer sequence is executed.

[0197] If the equipment execution confirmation result of the current transfer phase contains unqualified content, but the corresponding equipment does not take protective action, the confirmed equipment status is maintained, the next transfer phase is not executed, and operational feedback is continuously obtained from the time the current transfer phase instruction is issued; if the operational feedback meets the execution qualification conditions before the response time recorded in the corresponding equipment controller technical document expires, the current transfer phase is confirmed to be completed; if the execution qualification conditions are not met after the response time expires, the instructions for the current transfer phase and subsequent transfer phases are stopped.

[0198] After all transfer phases are completed, recheck the actual operating equipment, main unit load, water pump frequency, cooling tower fan frequency, and valve positions. If all checked items match the target equipment status, the verification of the target equipment status and transfer sequence is completed; if any checked item does not match the target equipment status, maintain the already confirmed equipment status and do not proceed to subsequent group control adjustments.

[0199] S5.4. When the device fails to complete the corresponding instruction and perform a protection action within the corresponding response time, stop issuing subsequent adjustment instructions to the corresponding device, maintain the operating status of the remaining confirmed devices, and identify the corresponding device as an abnormal device.

[0200] Stop the malfunctioning equipment according to the protection interlock sequence recorded in the corresponding equipment controller's technical documents; after the malfunctioning equipment receives stop feedback and the flow rate of the branch where the malfunctioning equipment is located enters the stop confirmation range recorded in the field commissioning record, close the electric valve corresponding to the malfunctioning equipment, and remove the malfunctioning equipment from the actual operating equipment and subsequent command objects to complete the malfunctioning equipment isolation. If the malfunctioning equipment does not receive stop feedback, do not close the common valves that affect the water supply of other operating equipment, maintain the malfunctioning equipment in the protection state, and stop the switching of subsequent equipment.

[0201] The remaining heat exchange capacity is obtained by summing the maximum allowable output of the remaining confirmed operating hosts after the isolation of the abnormal equipment, using the upper boundary of the total heat exchange capacity. At the same time, the total allowable flow of the remaining confirmed equipment is verified by the flow confidence range in the equipment switching boundary. The status of equipment that does not cover the corresponding flow confidence range is not used as the basis for the takeover of standby equipment.

[0202] The upper boundary of the credible range of heat exchange demand is compared with the remaining heat exchange capacity. When the upper boundary of the credible range of heat exchange demand is not greater than the remaining heat exchange capacity, the heat exchange gap is set to zero, and the operating status of the other confirmed equipment is maintained; when the upper boundary of the credible range of heat exchange demand is greater than the remaining heat exchange capacity, the difference between the two is the heat exchange gap.

[0203] From the executable candidate states, those containing faulty equipment are removed. Executable candidate states that retain the following standby equipment states: those whose total heat exchange capacity upper boundary covers the upper boundary of the reliable heat exchange demand range; whose total allowable flow range covers the reliable flow range within the equipment switching boundary; those with a qualified stable output range; and those with qualified start / stop time limits and qualified equipment interlocking conditions. If multiple standby equipment states exist, the standby equipment state with the lowest total energy consumption is selected according to the comparison order of switching energy consumption and total energy consumption recorded in S2.4 and S2.5.

[0204] According to the transfer sequence corresponding to the selected standby equipment status, the valve opening, water pump flow establishment, main unit startup, and load transfer between main units are completed in sequence. Each transfer stage uses the verification method described in S5.2 to form equipment execution confirmation results. After the current transfer stage is verified and qualified, the next transfer stage is executed until the actual operating equipment and corresponding control quantities meet the selected standby equipment status, and the standby equipment takeover is completed.

[0205] If there is no standby equipment available to cover the heat exchange gap, the remaining confirmed equipment shall be kept operating within the limits of minimum stable output, maximum allowable output, and equipment operating flow rate. No further instructions shall be issued to increase load, reduce equipment, or change equipment combinations, and the heat exchange gap results shall be retained.

[0206] S5.5. When the target equipment status and transfer sequence verification is completed and no abnormal equipment occurs, obtain the actual operating equipment, flow rate of each operating host, flow rate of the main water supply pipe, temperature of the main water supply pipe, temperature of the return water pipe, and actual power of each equipment within the target equipment status stability judgment period; when abnormal equipment occurs and the standby equipment takeover is completed, obtain the above-mentioned operating feedback within the standby equipment status stability judgment period.

[0207] The actual heat exchange of each operating unit is calculated and accumulated according to the heat exchange calculation relationship of the operating unit to obtain the actual heat exchange; the actual flow rate of the water supply main is used as the actual circulation flow rate, and the flow rate of the water supply main, return main, bypass and each operating branch are verified by water balance; the actual power of each equipment is the average value of the power measurement value within the stable judgment period.

[0208] When no standby equipment takeover occurs, if the actual operating equipment meets the target equipment status, the actual heat exchange is within the reliable range of heat exchange requirements, the actual circulating flow is within the reliable flow range of the equipment switching boundary, and the actual power of each device achieves a stable response, then the target equipment status execution feedback is confirmed as valid. When standby equipment takeover occurs, if the actual operating equipment meets the selected standby equipment status, the actual heat exchange is within the reliable range of heat exchange requirements, the actual circulating flow is within the reliable flow range of the equipment switching boundary, and the actual power of each device achieves a stable response, then the standby equipment takeover feedback is confirmed as valid. If any of these conditions are not met, the confirmed equipment status is maintained, and the corresponding verification content is recorded as incomplete.

[0209] The data center energy-saving operation results are generated by combining the verification results of actual operating equipment, actual heat exchange, actual circulation flow, actual equipment power, target equipment status and transfer sequence, abnormal equipment isolation results and standby equipment takeover results.

[0210] In summary, this invention utilizes Bayesian inference constrained by water and heat balance, along with Markov chain Monte Carlo sampling, to convert flowmeter calibration errors and actual operating measurements into reliable flow ranges and reliable heat exchange demand ranges. Based on this, a device switching boundary is established, ensuring that device additions and reductions no longer depend on a single flow measurement value, thus reducing the likelihood of frequent device switching when heat exchange demand is within critical ranges. Within the device switching boundary, the heat exchange capacity, allowable flow rate, stable output, start-stop time limits, and interlocking conditions of the equipment are further verified. The target device state with lower switching and operating energy consumption is selected from the transfer paths that maintain continuous heat exchange, taking into account both the energy consumption during equipment combination changes and stable operation phases. By sequentially executing forward relay-type extreme value regulation and reverse-sequence regulation, the power coupling between the main unit load, circulating water pumps, and cooling tower fans is incorporated into the total power comparison of the machine room, avoiding the situation where reducing the power of one device alone would increase the power of other devices. When a faulty device triggers a protection action, the standby device status is selected and the takeover is completed based on the reliable range of heat exchange demand and the remaining heat exchange capacity. This ensures that flow correction, equipment switching, power regulation, and fault handling are always constrained by the same thermo-hydraulic boundary, thereby improving the stability, executability, and continuous power supply capability of the HVAC room group control energy-saving results.

[0211] It should be noted that 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing energy conservation through group control of multiple devices in a HVAC room, characterized in that: include, By using water balance and heat balance constrained Bayesian inference and Markov chain Monte Carlo sampling, flow measurement points are corrected and the flow confidence range, heat exchange demand confidence range and equipment switching permission are determined, and equipment switching boundaries are generated. Candidate states that meet the requirements of heat exchange capacity, flow rate, stable output, start-stop time limit and interlocking conditions are selected according to the equipment switching boundary. The transfer path with the lowest switching energy consumption and continuous heat exchange is retained to obtain the target equipment state and transfer sequence. Based on the target equipment status and transfer sequence, the equipment is switched, and within the equipment switching boundary, the cooling tower fan, circulating water pump and main unit load are sequentially adjusted by relay-type extreme value regulation to generate a positive low-power operating point. By adjusting the main unit load, circulating water pump and cooling tower fan in reverse order from the positive low power operating point, and checking the water supply temperature, pipeline pressure difference and main unit flow, a group control operating point with the lowest total power is formed. By issuing commands for host load, frequency conversion and valve position from the group control operation point, verifying the status and transfer sequence of target equipment, isolating abnormal equipment and connecting backup equipment, and generating energy-saving operation results for the computer room.

2. The energy-saving optimization method for multi-equipment group control in HVAC computer rooms as described in claim 1, characterized in that: The water balance refers to the flow conservation relationship between the main supply water flow, the main return water flow, the bypass flow, and the operating branch flow. The heat balance refers to the heat conservation relationship between the sum of the heat exchange of the operating unit and the heat exchange of the main supply and return water flow.

3. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 2, characterized in that: The calibration flow measurement points specifically include: Select the flow meter calibration accuracy and the effective flow rate confirmed in the previous control cycle to establish the prior distribution of each flow measurement point. Then, use the current flow measurement value and the corresponding pipeline temperature change to correct the prior distribution and obtain the posterior distribution of each flow measurement point. Markov chain Monte Carlo sampling was performed according to the posterior distribution to obtain candidate flow combinations for each flow measurement point. The candidate flow combinations were then tested successively using water balance and heat balance. Candidate flow combinations that did not simultaneously satisfy both balance relationships were eliminated to obtain an effective flow sample set. The current measured values ​​of each flow measurement point are corrected based on the effective flow sample set, and the flow measurement point correction results are generated.

4. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 3, characterized in that: The generation device switching boundary specifically includes: Extract the lower and upper boundaries of the flow rate of each pipeline from the effective flow rate sample set in the flow measurement point correction results, and determine the reliable range of the flow rate of each pipeline. The heat exchange demand is calculated item by item by combining the reliable range of flow rate and the corresponding pipeline temperature change. The heat exchange demand value that simultaneously satisfies water balance and heat balance is retained to determine the reliable range of heat exchange demand. The reliable range of heat exchange demand is divided according to the minimum load adjustment allowed by the host controller, and the continuous demand segment is obtained. The maximum heat exchange capacity, minimum stable output and allowable flow rate of the current equipment combination are checked segment by segment, and control conclusions are obtained for adding equipment, maintaining equipment and reducing equipment, respectively. When each consecutive demand segment corresponds to the same control conclusion, confirm the equipment switching permission that is consistent with the control conclusion; when each consecutive demand segment corresponds to different control conclusions, confirm the maintenance of the equipment permission and limit the current number of equipment to remain unchanged. The equipment operating flow rate is limited according to the flow rate confidence range, the equipment heat exchange capacity is limited according to the heat exchange demand confidence range, and the direction of equipment status change is limited by the equipment switching permission to generate the equipment switching boundary.

5. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 4, characterized in that: The screening process selects candidate states that meet the requirements of heat exchange capacity, flow rate, stable output, start / stop time limit, and interlocking conditions. Specifically, it includes: Based on the direction of device status change defined by the device switching boundary, establish a licensed device status that maintains the current device combination, adds a set of host and supporting equipment, and removes a set of host and supporting equipment. When the licensed device status contains abnormal devices, replace the abnormal devices with healthy standby devices to obtain the device status to be verified. The total heat exchange capacity of the equipment under test is verified by using the upper boundary of the reliable range of heat exchange demand in the equipment switching boundary, and the total allowable flow rate of the equipment under test is verified by using the reliable range of flow rate in the equipment switching boundary, so as to obtain the thermal and hydraulic qualified state. Verify the stable output range, start-up and shutdown time limits, and equipment interlocking conditions of each host unit within the qualified thermal and hydraulic conditions, eliminate equipment states that do not meet any of the conditions, and obtain executable candidate states.

6. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 5, characterized in that: The process of obtaining the target device state and transition sequence specifically includes: Starting from the current equipment status, establish transfer paths corresponding to each executable candidate status according to the stage sequence of valve opening, water pump flow establishment, main unit startup, load transfer between main units, original main unit shutdown, and auxiliary equipment shutdown; Each state transition phase only changes one set of interlocked equipment. The heat exchange capacity of the remaining operating equipment is checked using the upper boundary of the reliable range of heat exchange demand. Transition paths that have experienced effective heat exchange interruption are eliminated. The energy consumption of path switching is accumulated based on the actual power and duration of the equipment in each state transition phase. For transfer paths that reach the same executable candidate state, retain the transfer path with the lowest switching energy consumption, and when the switching energy consumption is the same, retain the transfer path with the fewest device start-stop times, thus obtaining the candidate transfer path; The switching energy consumption and equipment operation energy consumption of the candidate transfer paths are calculated. The executable candidate state with the lowest total energy consumption and effective heat exchange is selected as the target equipment state, and the corresponding candidate transfer path is used as the transfer sequence.

7. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 6, characterized in that: The generation of a positive low-power operating point specifically includes: The current equipment state is switched to the target equipment state according to the transfer sequence. While keeping the target equipment state unchanged, the cooling tower fan, circulating water pump and main unit load are selected as control variables in sequence, and only one control variable is adjusted at a time. Change the minimum stable adjustment of the current control variable along the initial search direction. After the water supply temperature, circulation flow, pipeline pressure difference and equipment power have completed the response, compare the total power of the machine room before and after the adjustment. When the total power of the computer room decreases, continue to adjust along the original search direction. When the total power of the computer room increases, restore the previous control quantity, change the search direction and reduce the adjustment quantity. When neither search direction produces a power decrease that exceeds the uncertainty range of power metering, fix the current control quantity and switch to the next control quantity. If any of the following situations occur during the adjustment process: water supply temperature deviation, insufficient pressure difference in the most unfavorable pipeline, insufficient main unit flow, equipment protection action, flow exceeding the equipment switching boundary, or heat exchange capacity exceeding the equipment switching boundary, the control quantity before adjustment shall be restored. The cooling tower fan, circulating water pump, and main unit load are adjusted sequentially to generate a positive low-power operating point.

8. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 7, characterized in that: The reverse-order regulation of the host load, circulating water pump, and cooling tower fan specifically includes: The control quantities at the forward low-power operating point are arranged in the order of main unit load, circulating water pump, and cooling tower fan to obtain the reverse order adjustment sequence. Adjust the host load according to the reverse order of adjustment, keep the equipment combination, total heat exchange and total circulation flow unchanged, transfer the host load adjustment from the host with a unit heat exchange power consumption greater than another operating host to another operating host, compare the total power of the machine room before and after the load transfer, and obtain the host load reverse order adjustment result; The circulating water pump is adjusted according to the reverse order of adjustment to maintain the most unfavorable pipeline pressure difference, the minimum flow rate of the main unit and the reliable flow range. The frequency of the circulating water pump is adjusted and the total power of the main unit, water pump and supporting equipment is compared to obtain the reverse order adjustment result of the circulating water pump. Adjust the cooling tower fan according to the reverse order of adjustment, adjust the frequency of the cooling tower fan, and compare the total power of the main unit, cooling water pump and cooling tower fan to obtain the reverse order adjustment result of the cooling tower fan; The reverse order adjustment results are fixed sequentially to generate the reverse order running point.

9. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 8, characterized in that: The group control operating point that forms the lowest total power specifically includes: The water supply temperature, the most unfavorable pipeline pressure difference, and the main unit flow rate corresponding to the positive low power operation point and the reverse sequence operation point are checked respectively. The operation points corresponding to any of the states of water supply temperature deviation, insufficient most unfavorable pipeline pressure difference, and insufficient main unit flow rate are eliminated to obtain the qualified operation points. Compare the total power of the computer room at each qualified operating point, and select the operating point with the lowest total power among the forward low-power operating points and the reverse-order operating points to form a group control operating point.

10. The energy-saving optimization method for multi-equipment group control in HVAC room as described in claim 9, characterized in that: The generated data center energy-saving operation results specifically include: The group control operation point is converted into host load command, variable frequency command and valve position command, and each command is issued in sequence according to the transfer sequence. The actual valve position, actual pump flow, actual pipeline pressure difference, actual heat exchange of the host, actual water supply temperature and actual equipment power corresponding to each command are verified in sequence to obtain the equipment execution confirmation result. If the equipment execution confirmation result of the current transfer phase matches the target equipment status, the next transfer phase continues until the verification of the target equipment status and transfer sequence is completed; When the equipment fails to complete the corresponding instruction and a protection action is triggered, stop the subsequent adjustment of the equipment, maintain the operating status of the remaining confirmed equipment, isolate the abnormal equipment and determine the remaining heat exchange capacity; The heat exchange gap is determined by using the credible range of heat exchange demand and the remaining heat exchange capacity. A standby equipment state covering the heat exchange gap is selected from the executable candidate states, and the standby equipment takeover is completed according to the corresponding transfer sequence. Based on the feedback values ​​after the backup equipment is taken over, verify the actual operating equipment, actual heat exchange, actual circulation flow, and actual equipment power to generate the data center energy-saving operation results.

Images