A waste heat recovery scheduling method and system for two-phase cold plates in data centers

Abstract

This invention discloses a waste heat recovery scheduling method and system for two-phase cold plates in data centers, relating to the field of refrigeration and thermal management technology. The waste heat recovery scheduling method and system for two-phase cold plates in data centers includes the following steps: S1: Collecting key thermal parameters and terminal state variables through various thermal monitoring devices, acquiring and preprocessing thermal state monitoring data; S2: Based on the thermal state monitoring data and combined with SLA configuration files, prioritizing heat-using terminals and constructing a scheduling deviation judgment mechanism to identify changes in thermal state; S3: Under the drive of control commands, accessing multi-source waste heat and distributing it to various terminal scenarios, evaluating the matching between waste heat and terminal demand; S4: Based on changes in heat distribution status and thermal state monitoring data, identifying overload risks through a multi-parameter linkage mechanism. This solves the problems of insufficient waste heat recovery and limited energy utilization under light load and multi-heat-source collaborative operating conditions.

Description

Technical Field

[0001] This invention relates to the field of refrigeration and heat management technology, specifically to a waste heat recovery scheduling method and system for two-phase cold plates in a data center. Background Technology

[0002] With the widespread application of liquid cooling technology in data centers, two-phase cold plates, as core heat dissipation components for high heat density equipment, not only ensure the stable operation of computing equipment but also generate a large amount of underutilized low- and medium-grade heat energy. During operation, the industry generally attempts to extract the heat from the cold plate outlet and condensation area for heating, warming, and other energy applications, forming a preliminary framework for waste heat recovery.

[0003] For example, the invention patent with announcement number CN116164437B discloses an outdoor unit of a heat pump unit, including: a finned heat exchanger, including a left end plate, a right end plate and fins, with a first heat exchange component and a second heat exchange component arranged sequentially from top to bottom inside the fins; the first heat exchange component includes m first heat exchange copper tubes stacked from top to bottom, and the second heat exchange component includes n second heat exchange copper tubes stacked from top to bottom; a gas collection pipe, with the first ends of the m first heat exchange copper tubes and the n second heat exchange copper tubes all connected to the gas collection pipe; a liquid distribution component, including a first liquid distribution pipe, a second liquid distribution pipe and a first one-way valve, the second liquid distribution pipe being connected to the first liquid distribution pipe through the first one-way valve, the flow direction of the first one-way valve being from the second liquid distribution pipe to the first liquid distribution pipe; the first liquid distribution pipe including a brass distributor, the brass distributor being connected to the second ends of the m first heat exchange copper tubes and the second liquid distribution pipe through a pipeline, and the second liquid distribution pipe being connected to the second ends of the n second heat exchange copper tubes through a pipeline, which has a good defrosting effect.

[0004] For example, invention patent CN118932814B discloses a roadbed and pavement constant temperature system, relating to the field of road temperature control technology. It includes a ground heat exchange component located beneath the pavement; a buried underground pipe heat exchange device; a pipeline assembly; a heat pump unit connected to the buried pipe heat exchange device; a heat pump unit connected to the ground heat exchange component; a solar energy device connected to the heat pump unit; a cooling water device connected to the heat pump unit; and a phase change energy storage device connected to both the solar energy device and the cooling water device. The phase change energy storage device can input hot water from the solar energy device for heat exchange and energy storage, and output the cooled water after heat exchange to the cooling water device for cold storage. The phase change energy storage device can also input cooling water from the cooling water device to heat it, and output the heated water through the pipeline assembly to the heat pump unit for heating. This invention comprehensively utilizes multiple forms of energy to achieve roadbed and pavement thermal balance regulation in different seasons.

[0005] Existing waste heat scheduling methods mostly rely on preset priorities and fixed channels for heat allocation, which is difficult to adapt to the supply and demand imbalance caused by dynamic load changes. In terms of thermal status identification, there are generally problems such as insufficient monitoring dimensions, low sampling granularity, and slow response, which cannot achieve accurate control based on real-time data. Many types of terminal status parameters have not yet been included in the scheduling basis, resulting in a low degree of matching between heat allocation results and actual needs, and there are risks of low utilization efficiency and local heat redundancy.

[0006] To address the above issues, there is an urgent need for a waste heat recovery scheduling method and system for two-phase cold plates in data centers. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a waste heat recovery scheduling method and system for two-phase cold plates in data centers, solving the problems of insufficient waste heat recovery and limited energy utilization under light load and multi-heat source collaborative operating conditions.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a waste heat recovery scheduling method for two-phase cold plates in a data center, comprising the following steps: S1: collecting key thermal parameters and terminal state variables through various thermal monitoring devices, acquiring and preprocessing thermal state monitoring data; S2: prioritizing heat-using terminals based on thermal state monitoring data and in conjunction with SLA configuration files, constructing a scheduling deviation judgment mechanism to identify changes in thermal state, triggering the execution of control commands, and completing priority switching, flow restriction, and path adjustment; S3: under the drive of control commands, accessing multi-source waste heat and distributing it to various terminal scenarios, evaluating the matching of waste heat with terminal demand, adjusting heating intensity and frequency in stages, and updating the heat allocation list; S4: based on changes in heat allocation status and thermal state monitoring data, identifying overload risks in conjunction with a multi-parameter linkage mechanism, constructing response judgment criteria, and triggering the heat channel switching and operation status recording process.

[0009] Furthermore, the specific steps for acquiring and preprocessing thermal state monitoring data by collecting key thermal parameters and terminal state variables through multiple thermal monitoring devices are as follows: Thermal state monitoring data is acquired by collecting key thermal parameters and terminal state variables through multiple thermal monitoring devices. This thermal state monitoring data includes: cold plate outlet dryness, chip temperature, heat exchanger inlet temperature, heat exchanger outlet temperature, heat exchange flow rate, terminal flow rate, terminal inlet temperature, terminal outlet temperature, total number of terminals, water tank temperature, and return water temperature; component structural parameters are obtained by reading equipment factory data, measured data during installation, and control interface data; the dryness detection device integrates capacitive sensors, microwave sensors, and Venturi tube differential pressure feedback to analyze the dryness signal and calculate the cold plate outlet dryness; thermocouples attached to the chip surface continuously sense temperature changes and convert them into real-time numerical records of the chip temperature; a digital temperature probe is embedded in the inlet section of the plate heat exchanger, combined with the temperature gradient at the moment the fluid enters. The system extracts the inlet temperature of the heat exchanger and measures the outlet temperature using a thermistor at the end of the outlet section. Flow sensors installed on the inlet and outlet pipes of the heat exchanger collect the heat exchange flow rate in real time and dynamically adjust the flow rate in conjunction with flow control valves. Flow sensors installed on the terminal inlet pipes collect the flow rate through the terminal in real time. Platinum resistance thermometers embedded in the inlet of each terminal device periodically sample and update the terminal inlet temperature, while temperature-sensing patches on the terminal return water section synchronously output the terminal outlet temperature. The total number of terminals is maintained in real time based on the total number of registered heat-receiving unit nodes during operation. Immersion temperature sensors are deployed on the upper and lower layers of the water storage tank to extract the tank temperature by monitoring temperature difference changes. Digital temperature sensors installed on the return water branches of the heating network continuously upload sampling results to obtain the return water temperature. The thermal status monitoring data is standardized and normalized, and time alignment and interpolation are performed. Sudden changes in cold plate outlet dryness, abnormal increases in chip temperature, and sudden drops in heat exchange flow rate are marked and processed.

[0010] Furthermore, the specific steps for prioritizing heat-using terminals based on thermal condition monitoring data and combined with SLA configuration files are as follows: Input component structural parameters and thermal condition monitoring data under typical load scenarios; construct a waste heat scheduling simulation model through thermal calculation and operating condition simulation methods; output efficiency and safety datasets and waste heat allocation ratios; input thermal condition monitoring data and simulation model calculation results; construct a digital twin model through error fitting and deviation source correction algorithms; output closed-loop calibrated twin model data; based on the efficiency and safety dataset, waste heat allocation ratio, and twin model data, read the cold plate outlet dryness, chip temperature, heat exchanger inlet temperature, heat exchanger outlet temperature, and heat exchange flow rate, and... The system synchronously retrieves the availability level and energy efficiency requirements recorded in the SLA configuration file; determines the initial scheduling weights based on the SLA configuration file, and sets constraints while meeting safety margins and waste heat utilization; when abnormal deviations are detected in the dryness of the cold plate outlet, chip temperature, or recovered heat, the scheduling weights are readjusted, and the configuration switch is completed within the maximum allowable switchover time; based on the current scheduling weights, the operating status of the plate heat exchanger, shell-and-tube heat exchanger, and water pump is controlled, and the heat exchange medium flow rate is adjusted to achieve heat allocation; combining the current load level, recovered heat intensity, seasonal characteristics, and waste heat type, different heat-using terminals are prioritized, and the waste heat scheduling strategy is determined based on the priority ranking and terminal demand status.

[0011] Furthermore, the specific steps for constructing the scheduling deviation judgment mechanism to identify changes in thermal state are as follows: Read the dryness safety threshold and chip temperature threshold from the SLA configuration file; call the heat exchanger inlet temperature, heat exchanger outlet temperature, and heat exchange flow rate, and calculate the current recovered heat by combining the temperature difference and flow rate; obtain the cold plate outlet dryness and chip temperature; subtract the dryness safety threshold from the cold plate outlet dryness to obtain the deviation value between the two, and then square the deviation value to obtain the first dryness deviation square value; subtract the chip temperature threshold from the chip temperature, calculate the difference between the two, and then square it to obtain the second temperature deviation square value; add the above two deviation square values ​​together as the numerator; then take the product of the current recovered heat and the scheduling intervention coefficient, and add it to a constant to form the denominator; finally, divide the numerator by the denominator to obtain the strategy deviation judgment value.

[0012] Furthermore, the specific steps for executing the trigger control command to complete priority switching, flow restriction, and path adjustment are as follows: Real-time comparison of the strategy offset judgment value and the offset threshold, where the offset threshold includes a primary offset threshold and a secondary offset threshold; when the strategy offset judgment value is greater than or equal to the primary offset threshold, priority switching to chip cooling is executed, limiting the flow output of the plate heat exchanger, and suspending the heat path to residential water heating and office buildings, retaining only the main heat flow to the small organic Rankine cycle equipment; when the strategy offset judgment value is greater than the secondary offset threshold but less than the primary offset threshold, the current weight configuration and heat change trend are periodically read, and the allocation ratio between terminals is dynamically adjusted based on continuous comparison results; when the strategy offset judgment value is less than or equal to the secondary offset threshold, the existing waste heat scheduling strategy remains unchanged, and the cold plate outlet dryness, chip temperature, and current recovered heat are recorded as the basis for subsequent fitting.

[0013] Furthermore, the specific steps for assessing the matching between waste heat and terminal demand under the control command are as follows: Under the control command, plate heat exchangers at the cold plate outlet and shell-and-tube heat exchangers at the condenser outlet are connected to achieve the separation and utilization of sensible heat and latent heat, and the recovered waste heat is distributed to different heat-using terminals; the temperature difference between the terminal inlet temperature and the terminal outlet temperature is extracted, and combined with the flow rate through the terminal, the actual heat absorbed by the heat medium in the terminal is calculated to obtain the heat required by the terminal; the terminal reference temperature is extracted from the SLA configuration file, corresponding to the target heating temperature determined for different terminal types; the current recovered heat and the total number of terminals are obtained; for each terminal, the heat required by the terminal is extracted and multiplied by an adjustment factor. The adjustment factor is calculated by taking the absolute value of the difference between the terminal inlet temperature and the terminal reference temperature, adding it to a constant, and finally taking the reciprocal of the result; the product results corresponding to all terminals are summed and used as the denominator; the current recovered heat is divided by the summation result to obtain the global heat matching value.

[0014] Further, the specific steps for adjusting the heating intensity and frequency and updating the heat allocation list in a tiered manner are as follows: Real-time comparison of the global heat compatibility value and the compatibility threshold, where the compatibility threshold includes a primary compatibility threshold and a secondary compatibility threshold; when the global heat compatibility value is greater than or equal to the primary threshold, maintain heating for high-priority terminals, increase the heating frequency of medium-priority terminals, and adjust the scheduling priority field to prioritize recycling rate; when the global heat compatibility value is greater than the secondary compatibility threshold but less than the primary compatibility threshold, reduce the heating frequency of medium and low-priority terminals, reorder the heat allocation list, and switch the opening and closing of some terminal heating valves according to the monitoring data cycle; when the terminal inlet temperature is higher than the terminal reference temperature, the current allocation can be skipped; when the global heat compatibility value is less than or equal to the secondary compatibility threshold, suspend heating to low-compatibility terminals, clear the heat allocation list and record the abnormal terminal number, adjust the scheduling priority field to prioritize chip cooling; if the global heat compatibility value is lower than the lower limit of the global heat compatibility value for three or more consecutive scheduling cycles, call redundant heat sources from other nodes in the park.

[0015] Furthermore, the specific steps for identifying overload risks and constructing response judgment criteria based on changes in heat distribution status and thermal operating status monitoring data, combined with a multi-parameter linkage mechanism, are as follows: Based on changes in heat distribution status and thermal operating status monitoring data, determine in real time whether the overload trigger condition has been reached; acquire the water tank temperature, return water temperature, chip temperature, and cold plate outlet dryness; multiply the water tank temperature by the water tank weighting coefficient to obtain the water tank temperature rise influence term, and add it to the return water temperature to form the numerator; add the chip temperature to the temperature tolerance constant to form the denominator; divide the numerator by the denominator to obtain the temperature-related factor; subtract the cold plate outlet dryness from one to obtain the difference, and multiply the difference by the temperature-related factor to obtain the final abnormal response trigger value.

[0016] Furthermore, the specific steps of the process for triggering heat channel switching and recording operating status are as follows: real-time comparison of the abnormal response trigger value and the trigger threshold; when the abnormal response trigger value is greater than or equal to the trigger threshold, the main heat flow channel for domestic hot water and residential heating is closed, the bypass valve is opened to guide excess heat to the heat exhaust branch, while the heat path to the small organic Rankine cycle equipment is retained, and high-frequency data acquisition is started to record the maximum values ​​of cold plate outlet dryness and chip temperature every second; when the abnormal response trigger value is less than the trigger threshold, the existing heat channel remains unchanged, the water tank temperature, return water temperature and chip temperature are written into the cache, and the temperature change trend is continuously tracked.

[0017] The second aspect of this invention provides a waste heat recovery scheduling system for two-phase cold plates in a data center, comprising: a data acquisition and monitoring module, a waste heat scheduling and control module, a heat energy transmission and utilization module, and a safety response and linkage module. The data acquisition and monitoring module is used to collect key thermal parameters and terminal state variables through various thermal monitoring devices, and to acquire and preprocess thermal state monitoring data. The waste heat scheduling and control module is used to prioritize heat-using terminals based on the thermal state monitoring data and in conjunction with an SLA configuration file, construct a scheduling deviation judgment mechanism to identify changes in thermal state, trigger the execution of control commands, and complete priority switching, flow restriction, and path adjustment. The heat energy transmission and utilization module is used to access multi-source waste heat and distribute it to various terminal scenarios under the drive of control commands, assess the matching between waste heat and terminal demand, adjust the heating intensity and frequency in stages, and update the allocation list. The safety response and linkage module is used to identify overload risks based on changes in heat allocation status and thermal state monitoring data, combined with a multi-parameter linkage mechanism, construct response judgment criteria, and trigger the heat channel switching and operation status recording process.

[0018] The present invention has the following beneficial effects:

[0019] (1) This invention collects cold plate outlet dryness, chip temperature, heat exchanger temperature difference, terminal temperature and flow rate and other parameters in real time through multiple thermal monitoring devices, and performs standardization, normalization and time alignment processing, so that the thermal status data can maintain high accuracy and consistency before entering the scheduling calculation, providing a reliable real-time data foundation for subsequent waste heat distribution and safety judgment.

[0020] (2) Based on the collected thermal state data and SLA configuration file, the present invention dynamically calculates the scheduling weight, and combines the strategy offset judgment value to trigger priority switching, flow restriction and path adjustment, thereby realizing real-time control of multiple heat sources and multiple terminals, so that the heat distribution can automatically adapt to different operating conditions as the load changes.

[0021] (3) Under the control command, the present invention calculates the heat required by the terminal by combining the terminal inlet temperature, outlet temperature and flow rate, generates a global heat matching value, and makes hierarchical adjustments to the heating intensity, execution frequency and heat distribution list, thereby forming a dynamic and variable heat distribution strategy among different terminals and improving the matching of waste heat utilization.

[0022] (4) The present invention calculates the abnormal response trigger value through a multi-parameter linkage mechanism, and automatically performs heat channel switching, bypass heat dissipation and operation status recording operations when the threshold is exceeded, so that the system can quickly perform safety protection before the risk of overload occurs while efficiently recovering and utilizing waste heat, and maintain overall operational stability.

[0023] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0024] Figure 1 This is a flowchart of a waste heat recovery scheduling method for a two-phase cold plate in a data center according to the present invention.

[0025] Figure 2 This is a structural diagram of a waste heat recovery scheduling system for a two-phase cold plate in a data center according to the present invention;

[0026] Figure 3 This is a trend chart of the strategy offset judgment value of the present invention;

[0027] Figure 4 This is a schematic diagram illustrating the working principle of the present invention;

[0028] Figure 5 This is a flowchart of the waste heat distribution process of the present invention.

[0029] In the diagram, 1 is a heat exchanger; 2 is an office building; 3 is a residential building; 4 is a generator; 5 is a server; 6 is a two-phase cold plate; 7 is a capacitive sensor; 8 is a microwave sensor; and 9 is a venturi tube. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please see Figures 1-5 This invention provides a technical solution: a waste heat recovery scheduling method for two-phase cold plates in a data center, comprising the following steps: S1: Collecting key thermal parameters and terminal state variables through multiple thermal monitoring devices, acquiring and preprocessing thermal state monitoring data; S2: Based on the thermal state monitoring data and combined with the SLA configuration file, prioritizing heat-using terminals, constructing a scheduling deviation judgment mechanism to identify changes in thermal state, triggering the execution of control commands, and completing priority switching, flow restriction, and path adjustment; S3: Driven by control commands, accessing multi-source waste heat and distributing it to various terminal scenarios, evaluating the matching of waste heat with terminal demand, adjusting heating intensity and frequency in stages, and updating the heat allocation list; S4: Based on changes in heat allocation status and thermal state monitoring data, identifying overload risks through a multi-parameter linkage mechanism, constructing response judgment criteria, and triggering the heat channel switching and operation status recording process.

[0032] Specifically, key thermal parameters and terminal state variables are collected through multiple types of thermal monitoring devices to acquire and preprocess thermal state monitoring data. The specific steps are as follows: Multiple thermal monitoring devices distributed at various nodes are used to simultaneously collect data on cold plate outlet dryness, chip temperature, heat exchanger inlet temperature, heat exchanger outlet temperature, heat exchange flow rate, terminal flow rate, terminal inlet temperature, terminal outlet temperature, total number of terminals, water tank temperature, return water temperature, and other core parameters. Simultaneously, combined with equipment factory data, measured data from the installation phase, and control interface information, corresponding component structural parameters are extracted. The dryness detection device integrates capacitive sensor 7, microwave sensor 8, and venturi tube 9 for differential pressure feedback, analyzing and calculating the signal to obtain the cold plate outlet dryness. Thermocouples attached to the chip surface continuously sense temperature changes and convert them into real-time numerical records of the chip temperature. A digital temperature probe is embedded in the inlet section of the plate heat exchanger to extract the inlet temperature based on the temperature gradient at the moment the fluid enters. A thermistor at the end of the outlet section measures the outlet temperature. Flow sensors installed on the inlet and outlet pipes of the heat exchanger collect flow data in real time and are linked with flow control valves to achieve dynamic adjustment. The flow sensor in the terminal inlet pipe collects the flow rate in real time, the platinum resistance thermometer at each terminal inlet periodically samples and updates the inlet temperature, and the temperature sensing patch in the return water section synchronously outputs the outlet temperature; the total number of registered heating unit nodes during operation maintains the total number of terminals in real time. Immersion temperature sensors on the upper and lower layers of the water storage tank extract the tank temperature by monitoring the temperature difference, and digital temperature sensors on the return water branches of the heating network continuously upload sampling results to obtain the return water temperature. All collected data undergoes standardized and normalized processing, time alignment, and interpolation completion. Trends in detected cold plate outlet dryness, abnormal increases in chip temperature, and sudden drops in heat exchange flow are labeled to provide a stable and accurate data foundation for subsequent scheduling and control.

[0033] In this implementation plan, various thermal monitoring devices are used to collect dryness, temperature, flow rate, and other key parameters of cold plates, heat exchangers, terminals, and thermal storage units. Component structural parameters are obtained by combining equipment information with measured data. Multi-source detection and real-time monitoring ensure coverage of the entire operational chain. The collected results are standardized, normalized, and time-aligned, and anomalies are marked to provide an accurate and reliable data foundation for waste heat distribution and safety control.

[0034] Specifically, based on thermal condition monitoring data and combined with SLA configuration files, heat-using terminals are prioritized. The specific steps are as follows: Input component structural parameters and real-time thermal condition monitoring data obtained under typical load scenarios; construct a waste heat scheduling simulation model through thermal calculation and operating condition simulation methods; generate an efficiency and safety dataset containing efficiency, safety margin, and other indicators, as well as the corresponding waste heat allocation ratio; then compare the thermal condition monitoring data with the simulation model output results; construct a digital twin model using error fitting and deviation source correction algorithms; and output the twin model data after closed-loop calibration. Based on the efficiency and safety dataset, waste heat allocation ratio, and twin model data, read the cold plate outlet dryness, chip temperature, heat exchanger inlet and outlet temperatures, and heat exchange flow rate; and simultaneously retrieve the availability level and energy efficiency requirements recorded in the SLA configuration file; determine the initial scheduling weight based on this information, and set the constraint range under the premise of meeting the safety margin and waste heat utilization rate limits. When abnormal deviations are detected in the dryness of the cold plate outlet, the chip temperature, or the recovered heat, the scheduling weights are adjusted in a timely manner, and the configuration is updated within the specified maximum switching time. Based on the updated scheduling weights, the operating status of the plate heat exchanger, the shell-and-tube heat exchanger, and the water pump are controlled, and the flow rate of the heat exchange medium is dynamically adjusted to complete the heat distribution. Combining the current load level, the intensity of recovered heat, seasonal characteristics, and the type of waste heat, different heat-using terminals are prioritized, and the final waste heat scheduling strategy is determined based on the priority order and the actual demand status of the terminals.

[0035] In this implementation plan, terminal priorities are dynamically assigned based on thermal condition monitoring data and SLA configuration files. Scheduling weights are calculated through simulation and digital twin models to maintain a balance between safety margin and waste heat utilization. When parameters are abnormal, the configuration is adjusted in a timely manner and the equipment operation is controlled so that heat distribution can quickly adapt to load changes and generate a scheduling strategy with a high degree of matching.

[0036] Specifically, a scheduling deviation judgment mechanism is constructed to identify changes in thermal state. The specific steps are as follows: The preset dryness safety threshold and chip temperature threshold in the SLA configuration file are read as scheduling safety reference baselines; the inlet temperature, outlet temperature, and heat exchange flow rate data of the heat exchanger are retrieved, and the current recovered heat is calculated by combining the temperature difference and flow rate; the cold plate outlet dryness and chip temperature are simultaneously acquired, and the deviation value is obtained by subtracting the dryness safety threshold from the cold plate outlet dryness and squaring it to form the first dryness deviation squared value; the difference between the chip temperature and the chip temperature threshold is calculated and squared to obtain the second temperature deviation squared value; the two deviation squared values ​​are added together as the numerator; the product of the current recovered heat and the scheduling intervention coefficient is taken and added to a constant to form the denominator. The scheduling intervention coefficient is obtained by fitting the coupling relationship between heat recovery intensity and strategy adjustment amplitude during the scheduling strategy training process, with a value range of 0.005-0.05; finally, the numerator is divided by the denominator to obtain the strategy deviation judgment value reflecting the degree of deviation of the current thermal state, providing a quantitative basis for subsequent scheduling switching and strategy adjustment.

[0037] The specific calculation method for the strategy offset judgment value is as follows:

[0038] ;

[0039] In the formula, Indicates the policy offset judgment value. Indicates the dryness of the cold plate outlet. This indicates the critical value for dryness safety. Indicates the current heat recovery. Indicates chip temperature. This indicates the critical temperature value of the chip. This represents the scheduling intervention coefficient.

[0040] Table 1 shows the strategy offset judgment value data table provided in the embodiments of this application. In this embodiment, the cold plate outlet dryness of record 1 is set to 0.80, the dryness safety threshold is set to 0.65, the chip temperature is set to 46.00, the chip temperature threshold is set to 45.00, and the current recovered heat is set to 10; the cold plate outlet dryness of record 2 is set to 0.95, the dryness safety threshold is set to 0.65, the chip temperature is set to 46.10, the chip temperature threshold is set to 45.00, and the current recovered heat is set to 15; the cold plate outlet dryness of record 3 is set to 0.90, and the dryness safety threshold is set to 0.65. The chip temperature is set to 46.20, the chip temperature critical value is set to 45.00, and the current recovered heat is set to 15. The dryness of the cold plate outlet in record 4 is set to 0.67, the dryness safety critical value is set to 0.65, the chip temperature is set to 46.30, the chip temperature critical value is set to 45.00, and the current recovered heat is set to 15. The dryness of the cold plate outlet in record 5 is set to 0.70, the dryness safety critical value is set to 0.65, the chip temperature is set to 45.90, the chip temperature critical value is set to 45.00, and the current recovered heat is set to 8.

[0041] Table 1. Strategy Offset Judgment Value Data Table

[0042]

[0043] like Figure 3 The figure shows a trend chart of strategy offset judgment values ​​provided in the embodiments of this application. According to the data in the image and table, the set first-level offset threshold is 1.1, the second-level offset threshold is 0.9, and the strategy offset judgment values ​​of the five sets of records fluctuate between 0.70 and 1.30, showing an overall trend of first increasing and then decreasing. Record 4 has a strategy offset judgment value of 1.30, exceeding the first-level offset threshold, indicating that the current thermal state has significantly deviated from the scheduling balance, requiring priority switching and flow restriction operations. Record 3 has a strategy offset judgment value of 1.16, also higher than the first-level offset threshold, indicating a need for scheduling adjustment. Record 2 has a strategy offset judgment value of 1.00, between the second-level and first-level offset thresholds, requiring dynamic adjustment of the allocation ratio based on the trend. Records 1 and 5 have strategy offset judgment values ​​of 0.85 and 0.70 respectively, both lower than the second-level offset threshold, indicating a relatively stable operating state. This figure can intuitively reflect the degree of offset in the scheduling state, providing a reference for triggering priority switching, adjusting heat allocation, and implementing protection measures.

[0044] In this implementation scheme, the deviation of the thermal state from the safety baseline is quantified by calculating the strategy offset judgment value, so that the system can accurately identify the changes in scheduling status based on the changes in dryness, temperature and recovered heat, and provide a basis for subsequent strategy switching and heat allocation adjustment.

[0045] Specifically, the process of triggering control commands involves priority switching, flow restriction, and path adjustment. The specific steps are as follows: Real-time monitoring of the strategy offset judgment value and comparison with preset first-level and second-level offset thresholds; When the judgment value is greater than or equal to the first-level threshold, the scheduling priority is immediately switched to chip cooling mode, limiting the flow output of the plate heat exchanger to within a safe range, and suspending the heat path to residential water heating and office building 2, retaining only the main heat flow to the small organic Rankine cycle equipment; When the judgment value is between the second-level threshold and the first-level threshold, the current scheduling weight configuration and heat change trend are periodically read, and the heat allocation ratio of different terminals is dynamically adjusted based on the continuous comparison results; When the judgment value is less than or equal to the second-level threshold, the existing waste heat scheduling strategy remains unchanged, and the dryness of the cold plate outlet, the chip temperature, and the current recovered heat are recorded as reference data for subsequent fitting and optimization.

[0046] In this implementation plan, by comparing the strategy offset judgment value with the preset threshold, it is determined whether it is necessary to switch priorities, limit and adjust heat allocation, and dynamically modify the output path and allocation ratio when the conditions are met; when no adjustment is needed, the original strategy is maintained, and key data such as dryness, temperature and recovered heat are recorded to provide support for subsequent allocation optimization.

[0047] Specifically, driven by control commands, multi-source waste heat is accessed and distributed to various terminal scenarios to assess the matching between waste heat and terminal demand. The specific steps are as follows: Under the action of commands, plate heat exchangers at the cold plate outlet and shell-and-tube heat exchangers at the condenser outlet are connected to achieve separation and recovery of sensible heat and latent heat, and the obtained waste heat is distributed to different heat-using terminals according to type and demand; the temperature difference between the inlet and outlet of each terminal is extracted, combined with the real-time flow rate, and substituted into the heat calculation formula to obtain the heat currently absorbed by the terminal, which is the heat required by the terminal; the terminal reference temperature is retrieved from the SLA configuration file, and the target heating temperature is determined according to the terminal type; the current total recovered heat and the total number of terminals are obtained simultaneously; for each terminal, its required heat is multiplied by the adjustment factor, which is calculated as the absolute value of the difference between the terminal inlet temperature and the reference temperature plus one and then the reciprocal; the sum of the product results of all terminals is used as the denominator, and the current recovered heat is divided by this value to obtain the global heat matching value reflecting the overall supply and demand matching degree.

[0048] The specific calculation method for the global thermal fit value is as follows:

[0049] ;

[0050] In the formula, Indicates the global thermal fit value. Indicates the current heat recovery. This indicates the heat required by the terminal. Indicates the terminal inlet temperature. Indicates the terminal reference temperature. This indicates the total number of terminals.

[0051] In this implementation plan, under the action of control commands, the multi-source waste heat recovered by the cold plate and condenser is integrated, and the actual absorbed heat of each terminal based on temperature difference and flow rate is calculated. An adjustment factor is generated by combining the reference temperature. The results are summarized to calculate the global heat matching value, thereby quantifying the overall matching degree between the current waste heat supply and the demand of each terminal, and providing a basis for subsequent heat allocation adjustment.

[0052] Specifically, the heating intensity and frequency are adjusted in stages, and the heat allocation list is updated. The specific steps are as follows: Real-time monitoring and comparison of the global heat matching value with the preset first-level and second-level matching thresholds; When the matching value is greater than or equal to the first-level threshold, stable heating is maintained for high-priority terminals, while the heating frequency of medium-priority terminals is increased, and the scheduling priority field is switched to recycling rate priority; When the matching value is between the second-level and first-level thresholds, the heating frequency of medium- and low-priority terminals is reduced, the heat allocation list is reordered, and the opening and closing of heating valves of some terminals are switched periodically according to the monitoring data. When the terminal inlet temperature is detected to be higher than the reference temperature, the current allocation can be skipped; when the matching value is less than or equal to the secondary threshold, heat supply to the low-matching terminal is stopped, the heat allocation list is cleared and its number is recorded, and the scheduling priority field is adjusted to prioritize chip cooling; if the global heat matching value is lower than the lower limit of the global heat matching value in three or more consecutive scheduling cycles, redundant heat sources of other nodes in the park are called to supplement it. The lower limit of the global heat matching value refers to the critical reference value used to calibrate the lowest stable matching level among the global heat matching values ​​extracted based on historical supply and demand balance conditions.

[0053] In this implementation plan, based on the comparison results of the global heat matching value and the threshold, the heating intensity and execution frequency of terminals with different priorities are adjusted in stages, and the heat allocation list is reordered and cleared when necessary. When the matching degree is high, high priority heating is maintained and the frequency of medium priority is increased. When the matching degree is medium, the heating of medium and low priority is reduced and the allocation order is optimized. When the matching degree is low, the heating of low matching terminals is suspended and the chip cooling is prioritized. In the state of continuous low value, redundant heat sources in the park can be called to supplement the supply, thereby ensuring the adaptability of heat allocation to terminal needs.

[0054] Specifically, based on changes in heat distribution status and thermal state monitoring data, and combined with a multi-parameter linkage mechanism, overload risks are identified and response judgment criteria are constructed. The specific steps are as follows: First, based on the real-time changes in heat distribution status and various thermal state monitoring data, it is determined whether the preset overload triggering conditions have been met. Then, the water tank temperature, return water temperature, chip temperature, cold plate outlet dryness, and other key parameters are acquired. The water tank temperature is multiplied by the water tank weighting coefficient to obtain the water tank temperature rise influence term, which is then added to the return water temperature to form the numerator. The rate of increase in domestic hot water temperature is then fitted through regression. The water tank weighting coefficient is obtained by analyzing the functional relationship between the rate and the water tank temperature change amplitude, with a value range of 0.3-1.5. The chip temperature and the temperature tolerance constant are added to form the denominator. The temperature tolerance constant is obtained by fitting the historical minimum stable fluctuation range of the chip temperature under low load conditions, with a value range of 0.5-2.0. The temperature correlation factor is calculated by dividing the numerator by the denominator. The difference is obtained by subtracting the dryness of the cold plate outlet from one. The difference is multiplied by the temperature correlation factor to obtain the final abnormal response trigger value, which provides a quantitative judgment basis for the execution of subsequent overload protection actions.

[0055] The specific calculation method for the abnormal response trigger value is as follows:

[0056] ;

[0057] In the formula, Indicates the value that triggers an abnormal response. Indicates the water tank temperature. This represents the weighting factor for the water tank. Indicates the return water temperature. Indicates chip temperature. Indicates the dryness of the cold plate outlet. This represents the temperature tolerance constant.

[0058] In this implementation plan, the abnormal response trigger value is calculated by multi-parameter linkage. The water tank temperature, return water temperature, chip temperature, cold plate outlet dryness and other key variables are comprehensively quantified to accurately reflect the overload risk level under the current operating state, and provide a reliable basis for triggering overload protection and executing corresponding control actions.

[0059] Specifically, the process of triggering heat channel switching and recording operating status involves the following steps: Real-time monitoring and comparison of abnormal response trigger values ​​with preset trigger thresholds; when the trigger value is greater than or equal to the threshold, immediately shutting off the main heat flow channels for domestic hot water and residential heating, opening the bypass valve to direct excess heat into the exhaust branch, while maintaining the normal operation of the heat path leading to the small organic Rankine cycle equipment; and in this state, initiating high-frequency data acquisition to record the maximum values ​​of cold plate outlet dryness and chip temperature once per second for subsequent analysis and judgment. When the trigger value is lower than the threshold, maintaining the existing heat channel, storing the water tank temperature, return water temperature, and chip temperature in a cache, and continuously monitoring and recording their changing trends to dynamically understand the operating status.

[0060] In this implementation plan, the system determines whether to switch the heat channel by comparing the abnormal response trigger value with the threshold in real time. When the threshold is exceeded, the system shuts off the domestic hot water and residential heating channels and opens the bypass for heat dissipation, while maintaining the heating supply of the small organic Rankine cycle equipment and recording key temperature parameters. When the threshold is not reached, the system keeps the channels unchanged and continuously tracks temperature changes to provide data support for subsequent operation analysis and control.

[0061] like Figure 2 The diagram shown is a structural schematic of a waste heat recovery scheduling system for a two-phase cold plate in a data center, provided in an embodiment of this application. This system utilizes a waste heat recovery scheduling method for two-phase cold plates in a data center, comprising: a data acquisition and monitoring module, a waste heat scheduling control module, a heat energy transfer and utilization module, and a safety response linkage module. The data acquisition and monitoring module collects real-time data on cold plate outlet dryness, chip temperature, heat exchanger temperature difference, flow rate, terminal status, and other key parameters using various thermal monitoring devices, and combines this data with standardization, normalization, and time alignment processing to form usable thermal status monitoring data. The waste heat scheduling control module... Based on the above data and combined with the SLA configuration file, dynamic priority is assigned to each heat-using terminal, a scheduling deviation judgment mechanism is constructed to identify changes in operating status, and control commands are triggered to complete priority switching, flow restriction and path adjustment; the heat energy transmission and utilization module, under the action of control commands, accesses multi-source waste heat from cold plates and condensers, distributes it to different terminal scenarios, and adjusts the heating intensity and execution frequency according to the matching degree between waste heat and terminal demand, while updating the allocation list; the safety response linkage module, based on changes in heat allocation status and thermal operating status monitoring data, combined with a multi-parameter linkage mechanism, identifies overload risks, forms a response judgment basis, and performs heat channel switching and operating status recording operations when necessary.

[0062] In this implementation plan, through the synergistic effect of data acquisition and monitoring, waste heat scheduling and control, heat energy transmission and utilization, and safety response linkage, the entire process management from thermal parameter acquisition, terminal priority determination, waste heat matching and allocation to overload risk protection is realized. This enables waste heat allocation to dynamically adapt to changes in operating status and improves heat utilization efficiency while ensuring heat dissipation safety.

[0063] like Figure 4 The diagram shown illustrates the working principle of an embodiment of this application. As can be seen from the image, the two-phase cold plate 6 is connected to the server 5 to absorb the heat generated during chip operation. The dryness detection stage is completed collaboratively by a capacitive sensor 7, a microwave sensor 8, and a venturi tube 9 to acquire and evaluate the working fluid state at the cold plate outlet. After heat conversion via the heat exchanger 1, the waste heat can be distributed to different terminals, including office building 2, residential building 3, and generator 4, enabling multi-scenario utilization for building heating, domestic hot water, and low-voltage power generation. This diagram visually demonstrates the overall process of waste heat collection, detection, heat exchange, and multi-terminal distribution, providing a structural basis for subsequent heat regulation and energy efficiency assessment.

[0064] like Figure 5 The diagram shown is a flow chart of waste heat distribution provided in an embodiment of this application. As can be seen from the image, the heat from the steam generated at the cold plate outlet is transferred to the office building via a high-temperature heat exchange unit, while the heat released from the condenser is transferred to residential water / heating facilities and small-scale organic Rankine cycle equipment via a low-temperature heat exchange unit, thus realizing the distribution and utilization of waste heat across multiple terminals, including offices, residential areas, and power generation. This diagram visually illustrates the flow of waste heat from its source to different usage scenarios, providing a reference for formulating heat allocation strategies.

[0065] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0066] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A waste heat recovery scheduling method for two-phase cold plates in a data center, characterized in that, Includes the following steps: S1: Collect key thermal parameters and terminal state variables through various thermal monitoring devices, and obtain and preprocess thermal state monitoring data; S2: Based on thermal condition monitoring data and combined with SLA configuration files, priority is assigned to heat-using terminals, and a scheduling deviation judgment mechanism is constructed to identify changes in thermal condition, trigger the execution of control commands, and complete priority switching, flow restriction and path adjustment. The specific steps for constructing the scheduling deviation judgment mechanism to identify changes in thermal state are as follows: Read the dryness safety threshold and chip temperature threshold from the SLA configuration file; retrieve the heat exchanger inlet temperature, heat exchanger outlet temperature, and heat exchange flow rate, and calculate the current recovered heat based on the temperature difference and flow rate; obtain the cold plate outlet dryness and chip temperature; subtract the dryness safety threshold from the cold plate outlet dryness to obtain the deviation between the two, and then square the deviation to obtain the first dryness offset square value; subtract the chip temperature threshold from the chip temperature, calculate the difference between the two, and then square it to obtain the second temperature offset square value; add the two offset square values ​​together as the numerator. Then, take the product of the current recovered heat and the scheduling intervention coefficient, add it to the constant to form the denominator; finally, divide the numerator by the denominator to obtain the strategy offset judgment value; The specific steps involved in executing the trigger control command, including priority switching, flow limiting, and path adjustment, are as follows: The strategy offset judgment value is compared with the offset threshold in real time. The offset threshold includes a primary offset threshold and a secondary offset threshold. When the strategy offset judgment value is greater than or equal to the primary offset threshold, the priority is switched to chip cooling, the flow output of the plate heat exchanger is limited, and the heat path to residential water heating and office buildings is suspended, while only the main heat flow to the small organic Rankine cycle equipment is retained. When the strategy offset judgment value is greater than the secondary offset threshold but less than the primary offset threshold, the current weight configuration and heat change trend are read periodically, and the allocation ratio between terminals is dynamically adjusted according to the continuous comparison results. When the strategy offset judgment value is less than or equal to the secondary offset threshold, the existing waste heat scheduling strategy remains unchanged, and the dryness of the cold plate outlet, the chip temperature, and the current recovered heat are recorded as the basis for subsequent fitting. S3: Driven by control commands, it accesses multi-source waste heat and distributes it to various terminal scenarios, assesses the matching between waste heat and terminal demand, adjusts the heating intensity and frequency in stages, and updates the heat distribution list. S4: Based on changes in heat distribution status and thermal status monitoring data, and combined with a multi-parameter linkage mechanism, identify overload risks, construct response judgment criteria, and trigger the process of heat channel switching and operation status recording.

2. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps for acquiring and preprocessing thermal condition monitoring data by collecting key thermal parameters and terminal state variables through multiple types of thermal monitoring devices are as follows: Thermal condition monitoring data is obtained by collecting key thermal parameters and terminal state variables through various thermal monitoring devices. The thermal condition monitoring data includes: cold plate outlet dryness, chip temperature, heat exchanger inlet temperature, heat exchanger outlet temperature, heat exchange flow rate, terminal flow rate, terminal inlet temperature, terminal outlet temperature, total number of terminals, water tank temperature, and return water temperature. Component structural parameters are obtained by reading equipment factory data, measured data during installation, and control interface data. The dryness detection device integrates capacitive sensors, microwave sensors, and Venturi tube differential pressure feedback to analyze the dryness signal and calculate the dryness of the cold plate outlet. A thermocouple attached to the chip surface continuously senses temperature changes and converts them into real-time numerical records of the chip temperature. A digital temperature probe is embedded in the inlet section of the plate heat exchanger to extract the inlet temperature of the heat exchanger based on the temperature gradient at the moment the fluid enters. A thermistor at the end of the outlet section measures the outlet temperature of the heat exchanger. Flow sensors installed on the inlet and outlet pipes of the heat exchanger collect heat exchange flow in real time and dynamically adjust the flow rate in conjunction with flow control valves. Flow sensors deployed on the terminal inlet pipes collect the flow rate through the terminal in real time. Platinum resistance thermometers embedded in the inlet of each terminal device periodically sample and update the terminal inlet temperature, while temperature-sensing patches in the terminal return section synchronously output the terminal outlet temperature. The total number of terminals is maintained in real time based on the total number of registered heat-receiving unit nodes during operation. Immersion temperature sensors are deployed on the upper and lower layers of the water storage tank to extract the tank temperature by monitoring temperature difference changes. Digital temperature sensors deployed in the return water branches of the heating network continuously upload sampling results to obtain the return water temperature. The thermal condition monitoring data is standardized and normalized, and time alignment and interpolation are performed. The characteristics of sudden changes in cold plate outlet dryness, abnormal rise in chip temperature, and sudden drop in heat exchange flow are marked.

3. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps for prioritizing heat-using terminals based on thermal condition monitoring data and SLA configuration files are as follows: Input component structural parameters and thermal condition monitoring data under typical load scenarios, construct a waste heat scheduling simulation model through thermal calculation and operating condition simulation methods, and output efficiency and safety datasets and waste heat allocation ratios. Input thermal condition monitoring data and simulation model calculation results, construct a digital twin model through error fitting and deviation source correction algorithms, and output twin model data after closed-loop calibration; Based on efficiency and safety datasets, waste heat allocation ratios, and twin model data, the system reads the cold plate outlet dryness, chip temperature, heat exchanger inlet temperature, heat exchanger outlet temperature, and heat exchange flow rate, and simultaneously retrieves the availability level and energy efficiency requirements recorded in the SLA configuration file. Initial scheduling weights are determined according to the SLA configuration file, and constraints are set based on meeting safety margins and waste heat utilization. When abnormal deviations are detected in the cold plate outlet dryness, chip temperature, or recovered heat, the scheduling weights are readjusted, and the configuration switch is completed within the maximum allowable switchover time. Based on the current scheduling weights, the operating status of the plate heat exchanger, shell-and-tube heat exchanger, and water pumps is controlled, and the heat exchange medium flow rate is adjusted to achieve heat allocation. Combining the current load level, recovered heat intensity, seasonal characteristics, and waste heat type, different heat-using terminals are prioritized, and a waste heat scheduling strategy is determined based on the priority ranking and terminal demand status.

4. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps for assessing the matching between waste heat and terminal demand, under the drive of control commands, involve accessing multi-source waste heat and distributing it to various terminal scenarios: Driven by control commands, plate heat exchangers connected to the cold plate outlet and shell-and-tube heat exchangers connected to the condenser outlet are used to separate and utilize sensible heat and latent heat, and distribute the recovered waste heat to different heat-using terminals. The temperature difference between the terminal inlet temperature and the terminal outlet temperature is extracted, and combined with the flow rate through the terminal, the actual heat absorbed by the heat medium in the terminal is calculated to obtain the heat required by the terminal. The terminal reference temperature is extracted from the SLA configuration file and corresponds to the target heating temperature determined for different terminal types; the current recovered heat and the total number of terminals are obtained. For each terminal, the required heat is extracted and multiplied by an adjustment factor. The adjustment factor is calculated as follows: take the absolute value of the difference between the terminal inlet temperature and the terminal reference temperature, add it to a constant, and finally take the reciprocal of the result. Sum the product results corresponding to all terminals and use it as the denominator. Divide the current recovered heat by the summation result to obtain the global heat fit value.

5. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps for adjusting the heating intensity and frequency in stages and updating the heat distribution list are as follows: The global heat matching value is compared with the matching threshold in real time. The matching threshold includes a first-level matching threshold and a second-level matching threshold. When the global heat matching value is greater than or equal to the first-level threshold, the heating of high-priority terminals is maintained, the heating frequency of medium-priority terminals is increased, and the scheduling priority field is adjusted to prioritize the recycling rate. When the global heat compatibility value is greater than the secondary compatibility threshold but less than the primary compatibility threshold, the heating frequency of low- and medium-priority terminals is reduced, the heat allocation list is reordered, and the opening and closing of some terminal heating valves are switched according to the monitoring data cycle. When the terminal inlet temperature is higher than the terminal reference temperature, the current allocation can be skipped. When the global heat compatibility value is less than or equal to the secondary compatibility threshold, heating to low-compatibility terminals is suspended, the heat allocation list is cleared and the abnormal terminal number is recorded, and the scheduling priority field is adjusted to prioritize chip cooling. If the global heat compatibility value is lower than the lower limit of the global heat compatibility value for three or more consecutive scheduling cycles, redundant heat sources from other nodes in the park are called.

6. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps for identifying overload risks and constructing response judgment criteria based on changes in heat distribution status and thermal operating status monitoring data, combined with a multi-parameter linkage mechanism, are as follows: Based on changes in heat distribution and thermal state monitoring data, it is determined in real time whether the overload trigger condition has been reached; the water tank temperature, return water temperature, chip temperature, and cold plate outlet dryness are acquired; the water tank temperature is multiplied by the water tank weighting coefficient to obtain the water tank temperature rise influence term, which is added to the return water temperature to form the numerator; the chip temperature is added to the temperature tolerance constant to form the denominator; the numerator is divided by the denominator to obtain the temperature correlation factor; the cold plate outlet dryness is subtracted from one to obtain the difference, and the difference is multiplied by the temperature correlation factor to obtain the final abnormal response trigger value.

7. The waste heat recovery scheduling method for a two-phase cold plate in a data center according to claim 1, characterized in that: The specific steps of the process for triggering heat channel switching and recording operating status are as follows: The system compares the abnormal response trigger value with the trigger threshold in real time. When the abnormal response trigger value is greater than or equal to the trigger threshold, it closes the main heat flow channels for domestic hot water and residential heating, opens the bypass valve to guide excess heat to the heat exhaust branch, and simultaneously maintains the heat path to the small organic Rankine cycle equipment. It also starts high-frequency data acquisition to record the maximum values ​​of cold plate outlet dryness and chip temperature every second. When the abnormal response trigger value is less than the trigger threshold, it keeps the existing heat channels unchanged, writes the water tank temperature, return water temperature, and chip temperature into the cache, and continuously tracks the temperature change trend.

8. A waste heat recovery scheduling system for a two-phase cold plate in a data center, employing the waste heat recovery scheduling method for a two-phase cold plate in a data center as described in any one of claims 1-7, comprising: The data acquisition and monitoring module, waste heat dispatch and control module, heat energy transmission and utilization module, and safety response and linkage module are characterized by: The data acquisition and monitoring module is used to collect key thermal parameters and terminal state variables through various thermal monitoring devices, and to acquire and preprocess thermal state monitoring data. The waste heat scheduling and control module is used to prioritize heat-using terminals based on thermal status monitoring data and in conjunction with SLA configuration files, construct a scheduling deviation judgment mechanism to identify changes in thermal status, trigger the execution process of control commands, and complete priority switching, flow restriction and path adjustment. The heat energy transmission and utilization module is used to access multi-source waste heat and distribute it to various terminal scenarios under the drive of control commands, evaluate the matching of waste heat with terminal demand, adjust the heating intensity and frequency in stages, and update the heat distribution list. The safety response linkage module is used to identify overload risks based on changes in heat distribution status and thermal state monitoring data, combined with a multi-parameter linkage mechanism, to construct response judgment criteria and trigger the heat channel switching and operation status recording process.

Images