Energy-saving control method, device and equipment for machine room air conditioner group control, storage medium and computer program product

Abstract

The application relates to the technical field of air conditioner temperature regulation, in particular to an energy-saving control method, device and equipment for machine room air conditioner group control, a storage medium and a computer program product. The method comprises the following steps: acquiring temperature and humidity data in a machine room, power consumption data of a load device and operation parameters of an air conditioner device; synchronizing the above data according to timestamps to generate corresponding multi-source feature data matrices; constructing space thermodynamic model parameters based on the multi-source feature data matrices; predicting a temperature evolution trend in a target time domain based on the space thermodynamic model parameters, and generating a control instruction sequence of the air conditioner device based on a prediction result; and if the control instruction sequence meets preset temperature range constraints, preset temperature change rate constraints and preset control time interval constraints, corresponding control instructions are sent to the air conditioner device, and the energy-saving efficiency of air conditioner group control is improved.

Description

Technical Field

[0001] This application relates to the field of air conditioning temperature regulation technology, and in particular to an energy-saving control method, device, equipment, storage medium and computer program product for group control of air conditioning in computer rooms. Background Technology

[0002] With the rapid development of technologies such as 5G communication, cloud computing, and big data, the construction scale of enclosed spaces such as data centers, server rooms, and communication base stations is constantly expanding. The large amount of heat generated by the dense operation of internal load equipment places higher demands on the energy consumption management and dynamic control capabilities of air conditioning systems. Existing data center air conditioning systems typically adopt static control strategies, with multiple air conditioning units operating independently based on their respective temperature detection results. The control logic is mainly based on fixed thresholds, simple PID (proportional-integral-derivative) regulation, or manual intervention. However, in actual operation, the heat load inside the data center has obvious dynamic characteristics, often manifested as sudden increases in power consumption in a short period of time, seasonal climate changes, and local heat island effects. Existing air conditioning group control solutions often have a delayed response due to a lack of comprehensive modeling of spatial heat conduction characteristics and load change trends, making it difficult to achieve precise coordinated control of local and overall environments, resulting in poor energy efficiency of air conditioning group control. Therefore, how to improve the energy efficiency of air conditioning group control has become an urgent technical problem to be solved. Summary of the Invention

[0003] The main purpose of this application is to provide an energy-saving control method, device, equipment, storage medium, and computer program product for group control of air conditioning in computer rooms, aiming to solve the technical problem of how to improve the energy-saving efficiency of group control of air conditioning.

[0004] To achieve the above objectives, this application provides an energy-saving control method for group control of computer room air conditioning systems, the method comprising the following steps:

[0005] Acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment;

[0006] The temperature and humidity data, the power consumption data, and the operating parameters are synchronized according to timestamps to generate a corresponding multi-source feature data matrix;

[0007] Based on the multi-source feature data matrix, space thermodynamic model parameters are constructed, including temperature inertia parameters, air conditioning influence parameters, load heat production parameters, and environmental coupling parameters.

[0008] Based on the parameters of the space thermodynamic model, the temperature evolution trend in the target time domain is predicted, and based on the prediction results, a control command sequence for the air conditioning equipment is generated.

[0009] If the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint, then the corresponding control command is sent to the air conditioning device.

[0010] In one embodiment, after the steps of acquiring temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment, the method further includes:

[0011] The temperature and humidity data, the power consumption data, and the communication protocols corresponding to the operating parameters are parsed to extract valid field information.

[0012] The valid field information is subjected to structural standardization processing to obtain standardized data;

[0013] Based on a preset time period, the standardized data is time-aligned.

[0014] In one embodiment, after the step of sending the corresponding control command to the air conditioning device if the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint, the method further includes:

[0015] Obtain the actual temperature data of each temperature acquisition point within the current cycle;

[0016] The actual temperature data is compared with the predicted temperature value corresponding to the control command to generate a prediction error vector;

[0017] Based on the prediction error vector, the temperature inertia parameter, air conditioning influence parameter, or load heat generation parameter are adjusted.

[0018] In one embodiment, the step of synchronizing the temperature and humidity data, the power consumption data, and the operating parameters by timestamp to generate a corresponding multi-source feature data matrix includes:

[0019] The temperature and humidity data, the power consumption data, and the operating parameters are subjected to timestamp normalization processing with a preset time granularity to obtain data alignment results;

[0020] The data alignment results are matched with the corresponding device identifier and timestamp to obtain multi-channel data frames, wherein each data frame corresponds to a time sampling point;

[0021] The multi-channel data frames are arranged in chronological order to generate the multi-source feature data matrix.

[0022] In one embodiment, the step of constructing space thermodynamic model parameters based on the multi-source feature data matrix includes:

[0023] With the air conditioning equipment turned off, temperature decay data at multiple time points were acquired, and the temperature decay data was fitted based on the least squares method to obtain temperature inertia parameters.

[0024] When controlling the output of a single air conditioner to change the cooling power, the response temperature change data of each temperature collection point in the space is acquired. Based on the response temperature change data, the heat conduction correlation coefficient between the air conditioner and the temperature sensing location is determined, and the air conditioner influence parameters are determined according to the heat conduction correlation coefficient.

[0025] Under different load device power consumption states, the corresponding space temperature change data is obtained, and regression analysis is performed based on the space temperature change data and load power consumption value to obtain load heat generation parameters and environmental coupling parameters.

[0026] The parameters of the space thermodynamic model are determined based on the temperature inertia parameter, the air conditioning influence parameter, the load heat generation parameter, and the environmental coupling parameter.

[0027] In one embodiment, the step of predicting the temperature evolution trend in the target time domain based on the parameters of the space thermodynamic model, and generating a control command sequence for the air conditioning equipment based on the prediction results, includes:

[0028] Based on the current environmental state data and the parameters of the space thermodynamic model, the temperature change sequence of each temperature acquisition point in the target time domain is predicted;

[0029] Based on the prediction results, construct an optimization objective function;

[0030] The control command sequence of the air conditioning equipment is generated by minimizing the optimization objective function as the solution objective of the constrained optimization problem.

[0031] Furthermore, to achieve the above objectives, this application also proposes an energy-saving control device for group control of computer room air conditioners, the energy-saving control device for group control of computer room air conditioners comprising:

[0032] The data acquisition module is used to acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment.

[0033] The synchronization module is used to synchronize the temperature and humidity data, the power consumption data, and the operating parameters according to timestamps, and generate a corresponding multi-source feature data matrix.

[0034] The parameter construction module is used to construct space thermodynamic model parameters based on the multi-source feature data matrix. The space thermodynamic model parameters include temperature inertia parameters, air conditioning influence parameters, load heat production parameters, and environmental coupling parameters.

[0035] The prediction module is used to predict the temperature evolution trend in the target time domain based on the parameters of the space thermodynamic model, and generate the control command sequence of the air conditioning equipment based on the prediction results.

[0036] The target module is used to send the corresponding control command to the air conditioning device if the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint.

[0037] In addition, to achieve the above objectives, this application also proposes an energy-saving control device for group control of computer room air conditioners. The device includes: a memory, a processor, and an energy-saving control program for group control of computer room air conditioners stored in the memory and executable on the processor. The energy-saving control program for group control of computer room air conditioners is configured to implement the steps of the energy-saving control method for group control of computer room air conditioners as described above.

[0038] In addition, to achieve the above objectives, this application also proposes a storage medium storing an energy-saving control program for group control of computer room air conditioners. When the energy-saving control program for group control of computer room air conditioners is executed by a processor, it implements the steps of the energy-saving control method for group control of computer room air conditioners as described above.

[0039] In addition, to achieve the above objectives, this application also proposes a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the energy-saving control method for group control of computer room air conditioning as described above.

[0040] This application acquires temperature and humidity data from the computer room, power consumption data from the load equipment, and operating parameters from the air conditioning equipment. It then synchronizes the temperature and humidity data, power consumption data, and operating parameters by timestamp to generate a corresponding multi-source feature data matrix. Based on this multi-source feature data matrix, it constructs spatial thermodynamic model parameters, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. Based on these spatial thermodynamic model parameters, it predicts the temperature evolution trend within the target time domain and generates a control command sequence for the air conditioning equipment based on the prediction results. If the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints, the corresponding control command is sent to the air conditioning equipment. This application acquires temperature and humidity data, power consumption data of load devices, and operating parameters of air conditioning equipment within the computer room. It then performs time-synchronized processing on this data to construct a unified multi-source feature data matrix. Based on this matrix, a spatial thermodynamic model is built, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. This enables the control system to accurately reflect the heat distribution and dynamic changes within the computer room. By predicting the temperature evolution trend within the target time domain and generating a sequence of air conditioning control commands, and by selecting control commands based on constraints such as temperature range, rate of change, and control interval, feedforward and energy-saving control of the air conditioning equipment is achieved, thereby effectively improving the energy efficiency and control accuracy of air conditioning group control. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the first embodiment of the energy-saving control method for group control of computer room air conditioning in this application;

[0042] Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the energy-saving control method for group control of computer room air conditioning in this application;

[0043] Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the energy-saving control method for group control of computer room air conditioning in this application;

[0044] Figure 4 This is a diagram of the AI ​​energy-saving group control gateway framework in one embodiment of the energy-saving control method for group control of computer room air conditioning in this application;

[0045] Figure 5 This is a network structure diagram of the multimodal sensing and execution network in one embodiment of the energy-saving control method for group control of computer room air conditioning in this application;

[0046] Figure 6 This is a flowchart of a model predictive control method based on multimodal perception in one embodiment of the energy-saving control method for group control of air conditioning in a computer room according to this application;

[0047] Figure 7This is a schematic diagram of the module structure of an energy-saving control device for group control of computer room air conditioning, as described in an embodiment of this application.

[0048] Figure 8 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the energy-saving control method for group control of air conditioning in a computer room, as described in this application embodiment.

[0049] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0050] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0051] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0052] It should be noted that with the rapid development of technologies such as 5G communication, cloud computing, and big data, the construction scale of enclosed spaces such as data centers, server rooms, and communication base stations is constantly expanding. The large amount of heat generated by the dense operation of internal load equipment places higher demands on the energy consumption management and dynamic control capabilities of air conditioning systems. Existing data center air conditioning systems typically adopt static control strategies, with multiple air conditioning units operating independently based on their respective temperature detection results. The control logic is mainly based on fixed thresholds, simple PID (proportional-integral-derivative) regulation, or manual intervention. However, in actual operation, the heat load inside the data center has obvious dynamic characteristics, often manifested as sudden increases in power consumption in a short period of time, seasonal climate changes, and local heat island effects. Existing air conditioning group control solutions often have a delayed response due to a lack of comprehensive modeling of spatial heat conduction characteristics and load change trends, making it difficult to achieve precise coordinated control of local and overall environments, resulting in poor energy efficiency of air conditioning group control. Therefore, how to improve the energy efficiency of air conditioning group control has become an urgent technical problem to be solved.

[0053] The main solution of this application is as follows: First, acquire temperature and humidity data from the computer room, power consumption data from the load equipment, and operating parameters of the air conditioning equipment. Second, synchronize the temperature and humidity data, power consumption data, and operating parameters according to timestamps to generate a corresponding multi-source feature data matrix. Third, based on the multi-source feature data matrix, construct spatial thermodynamic model parameters, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. Fourth, based on the spatial thermodynamic model parameters, predict the temperature evolution trend within the target time domain, and generate a control command sequence for the air conditioning equipment based on the prediction results. Fifth, if the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints, send the corresponding control commands to the air conditioning equipment.

[0054] This application acquires temperature and humidity data, power consumption data of load devices, and operating parameters of air conditioning equipment within the computer room. It then performs time-synchronized processing on this data to construct a unified multi-source feature data matrix. Based on this matrix, a spatial thermodynamic model is built, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. This enables the control system to accurately reflect the heat distribution and dynamic changes within the computer room. By predicting the temperature evolution trend within the target time domain and generating a sequence of air conditioning control commands, and by selecting control commands based on constraints such as temperature range, rate of change, and control interval, feedforward and energy-saving control of the air conditioning equipment is achieved, thereby effectively improving the energy efficiency and control accuracy of air conditioning group control.

[0055] It should be noted that the executing entity of the method in this embodiment can be a computing service device with data processing, network communication, and program execution functions, or it can be the energy-saving control device for group control of computer room air conditioning with the same or similar functions. This embodiment and the following embodiments will be described using an energy-saving control device for group control of computer room air conditioning as an example.

[0056] Based on this, the first embodiment of the energy-saving control method for group control of computer room air conditioning is proposed in this application. Please refer to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the energy-saving control method for group control of air conditioning in a computer room according to this application.

[0057] In this embodiment, the energy-saving control method for group control of computer room air conditioning includes the following steps:

[0058] S1: Acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment;

[0059] It should be noted that temperature and humidity data refer to the ambient temperature and relative humidity values ​​collected by temperature and humidity sensors. Load equipment refers to high-energy-consuming electronic devices such as servers, storage units, and switches within the computer room. Power consumption data refers to the electrical energy consumed by the load equipment per unit of time. Air conditioning equipment includes air-cooled or water-cooled air conditioners used for ambient temperature control in the computer room. Operating parameters refer to the key operating status data of the air conditioning equipment.

[0060] Specifically, real-time data is first acquired through sensing devices deployed inside the server room. Temperature and humidity sensors are typically installed at the air inlets and outlets of the load equipment or at the air conditioning supply / return locations to accurately monitor the local thermal and humidity environment of the space. PDU devices are installed in each rack or at the power distribution level to report the power consumption values ​​of the load equipment in real time. Air conditioning equipment provides operating status information through built-in communication modules or external control devices. For non-intelligent air conditioners, key parameters such as their outlet and return air temperatures can be collected through infrared learning modules.

[0061] Furthermore, this sensed data is connected to the computing unit (such as an AI energy-saving group control gateway) in the energy-saving control system via RS485, IP network, or ZigBee / LoRa. To ensure data availability and subsequent processing efficiency, the system needs to perform preliminary verification on data from different sources, such as timestamp checks and data integrity verification. All collected data is time-stamped for subsequent data fusion and thermal modeling.

[0062] This step, by acquiring temperature and humidity data within the computer room, power consumption data of the load devices, and operating parameters of the air conditioning equipment, enables the system to comprehensively perceive the core elements affecting spatial heat distribution and energy consumption, establishing a foundation for the dynamic relationship between heat sources, temperature zones, and the cooling system. This step provides real, continuous, and multi-dimensional raw data support for subsequent data synchronization, modeling analysis, and predictive control, avoiding control errors caused by incomplete information or data distortion, thereby significantly improving the response accuracy of the group control strategy and the target adaptability of energy-saving control.

[0063] S2: Synchronize the temperature and humidity data, the power consumption data, and the operating parameters according to timestamps to generate a corresponding multi-source feature data matrix;

[0064] It should be noted that the timestamp refers to the time when each piece of data was collected. A multi-source feature data matrix is ​​a two-dimensional or multi-dimensional data structure that aggregates and combines data from different types of sensors (such as temperature, humidity, power consumption, and air conditioning parameters) along a unified time axis. Each row corresponds to a time point, and each column corresponds to a feature dimension.

[0065] Specifically, the system aligns the operating parameters from temperature and humidity sensors, PDUs of load devices, and air conditioning equipment along the timeline. Since the reporting frequencies of each device may be inconsistent, the system first performs timestamp normalization on all raw data, such as mapping the data to a fixed sampling period, and interpolating or filling in missing data to ensure that all data can be aggregated at the same time point.

[0066] Furthermore, based on the device identifier, the normalized data such as temperature, humidity, power consumption, and air conditioning operating status are combined in the form of "feature columns" to construct a complete data frame for each time sampling point. Data frames from multiple time points are stacked in chronological order to form a multi-source feature data matrix with a unified structure and consistent dimensions. This matrix serves as the input basis for subsequent modeling and control and is stored in a time-series database or model cache.

[0067] This step synchronizes temperature and humidity data, power consumption data, and air conditioning operating parameters by timestamp and constructs a multi-source feature data matrix. This ensures that different types of sensing data collaboratively participate in the subsequent modeling process under the same time reference, effectively avoiding prediction biases caused by data delays, mismatches, or dimensional inconsistencies. This matrix provides a unified input format and multi-dimensional information support for algorithms such as space thermodynamics modeling and control optimization, thereby improving the accuracy, stability, and system response efficiency of model training and real-time control.

[0068] S3: Based on the multi-source feature data matrix, construct space thermodynamic model parameters, which include temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters;

[0069] It should be noted that the parameters of the space thermodynamic model are a set of mathematical modeling parameters used to characterize the temperature change mechanism within a closed space. Temperature inertia parameters describe the buffering capacity of a space or structure to temperature changes. Air conditioning impact parameters reflect the degree of influence of each air conditioning unit on different temperature points. Load heat generation parameters quantify the ability of load equipment (such as servers) to generate heat at different power consumption levels. Environmental coupling parameters describe the conduction effect of external climatic conditions (such as outdoor temperature) on the internal temperature through media such as walls, doors, and windows.

[0070] Specifically, based on historical temperature change data and power consumption information in the multi-source feature data matrix, the temperature decay process during the period when the air conditioning equipment is turned off is selected. A fitting algorithm (such as the least squares method) is used to extract the slope and trend of the temperature naturally decreasing over time in each region, thereby generating temperature inertia parameters that reflect the delayed response capability of local areas to heat changes.

[0071] Furthermore, within a specific time period, the cooling intensity of a single air conditioning unit is adjusted in a stepwise manner, while response data from multiple temperature monitoring points in the space are collected simultaneously. Based on the temperature response amplitude and time delay, the thermal coupling weight between the air conditioner and each measuring point is calculated to construct air conditioner influence parameters. Further, combining the power consumption status of different server racks with corresponding temperature change trends, regression analysis is performed on the heat generation capacity of each rack to form load heat generation parameters. Finally, by incorporating outdoor temperature data and combining it with structural thermal resistance and temperature difference response analysis, environmental coupling parameters reflecting the influence of external climate are constructed.

[0072] By constructing a spatial thermodynamic model parameter set based on a multi-source feature data matrix, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters, a comprehensive description of the heat conduction mechanism within the computer room and its coupling relationship with the cooling system, load operating status, and external environment can be achieved. Compared to traditional static or empirical control methods, this modeling approach has the ability to express time-varying and spatial distribution characteristics, providing a high-precision, physically relevant mathematical foundation for subsequent temperature prediction and refined control, thereby significantly improving the accuracy and responsiveness of the energy-saving control system.

[0073] S4: Based on the parameters of the space thermodynamic model, predict the temperature evolution trend in the target time domain, and generate the control command sequence of the air conditioning equipment based on the prediction results;

[0074] It should be noted that the target time domain refers to the future time interval used for temperature prediction. Temperature evolution trend prediction refers to extrapolating the temperature change trajectory of each temperature acquisition point over a future period based on the current environmental conditions and model parameters. The control command sequence is a set of air conditioning control parameters corresponding to multiple moments, generated based on the prediction results.

[0075] Specifically, based on the multi-source environmental state data collected at the current moment (including the current temperature distribution, load power, and air conditioning operation status) and the parameters of the constructed space thermodynamic model, the target time domain length is set, and the current state is used as the initial input. The model is used to extrapolate the temperature changes at each future time point to obtain the temperature evolution trend of each temperature collection point within the prediction interval.

[0076] Furthermore, a multi-objective optimization problem is constructed based on the prediction results. The objective function can simultaneously consider temperature deviation (the difference from the set temperature), control energy consumption (such as air conditioner operating power), and control smoothness (the frequency or amplitude of control actions). By solving this optimization problem, a sequence of air conditioning equipment control commands covering the target time domain is generated. The system uses this sequence as a candidate control scheme for subsequent constraint judgment and actual execution.

[0077] This step predicts the temperature evolution trend within the target time domain based on spatial thermodynamic model parameters and generates a control command sequence based on the prediction results. This enables a shift from passive response to active regulation, allowing the air conditioning system to pre-adjust its operating strategy before environmental changes occur, significantly improving the control system's foresight and energy efficiency. This mechanism not only improves the accuracy of temperature control but also reduces frequent start-ups and shutdowns of air conditioners and sudden increases in energy consumption, thereby optimizing the operational stability and energy efficiency of air conditioning group control.

[0078] S5: If the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint, then the corresponding control command is sent to the air conditioning device.

[0079] It should be noted that temperature range constraints limit the predicted temperature at each temperature acquisition point to between a set minimum and maximum value to prevent overcooling or overheating. Temperature change rate constraints limit the magnitude of temperature change per unit time to avoid rapid temperature fluctuations and ensure equipment and personnel safety. Control time interval constraints limit the minimum interval between two consecutive control operations on the same air conditioning unit to prevent damage from frequent start-stop cycles. Control command transmission refers to transmitting the control commands, which have passed multiple constraint checks, to the air conditioning unit's execution module, enabling it to operate according to the instructions.

[0080] Specifically, after generating the control command sequence within the target time domain, the sequence is first subjected to logical verification and physical constraint judgment. This includes: determining whether the predicted temperature at each moment is within the set safe temperature range; evaluating the temperature change amplitude between consecutive time points to ensure it does not exceed the upper limit of the temperature change rate; and calculating the time interval between adjacent control actions to avoid violating the minimum control interval constraint. Only when all control actions in the control command sequence satisfy the above three types of constraints will the system extract the control command corresponding to the current control cycle and send it to the target air conditioning device through the air conditioning equipment communication interface (such as Modbus, infrared protocol, or network command), causing the air conditioning to execute its operating tasks according to the scheduling strategy.

[0081] By introducing multiple constraints, such as temperature range, temperature change rate, and control time interval, before sending control commands, it is possible to effectively prevent environmental fluctuations, equipment fatigue, or operational conflicts caused by aggressive control strategies or prediction errors. This step, while ensuring the safe and stable operation of the system, improves the practicality and robustness of the control strategy, helps extend the service life of air conditioning equipment, reduces maintenance costs, and achieves a balance between energy-saving control targets and equipment protection.

[0082] This embodiment acquires temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment. The temperature and humidity data, power consumption data, and operating parameters are synchronized by timestamp to generate a corresponding multi-source feature data matrix. Based on the multi-source feature data matrix, spatial thermodynamic model parameters are constructed, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. Based on the spatial thermodynamic model parameters, the temperature evolution trend within the target time domain is predicted, and based on the prediction results, a control command sequence for the air conditioning equipment is generated. If the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints, the corresponding control command is sent to the air conditioning equipment. This embodiment acquires temperature and humidity data, power consumption data of load devices, and operating parameters of air conditioning equipment within the computer room. It then performs time synchronization processing on this data to construct a unified multi-source feature data matrix. Based on this matrix, a spatial thermodynamic model is built, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. This enables the control system to accurately reflect the heat distribution and dynamic changes within the computer room. By predicting the temperature evolution trend within the target time domain and generating an air conditioning control command sequence, and combining constraints such as temperature range, rate of change, and control interval to filter control commands, feedforward and energy-saving control of the air conditioning equipment is achieved, thereby effectively improving the energy efficiency and control accuracy of the air conditioning group control.

[0083] Based on the first embodiment described above, a second embodiment of the energy-saving control method for group control of computer room air conditioning systems is proposed in this application. Please refer to... Figure 2 , Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the energy-saving control method for group control of air conditioning in a computer room according to this application.

[0084] like Figure 2 As shown, in this embodiment, after step S1, the following steps are also included:

[0085] S1a: Parse the temperature and humidity data, the power consumption data, and the communication protocol corresponding to the operating parameters, and extract valid field information;

[0086] S1b: Perform structural standardization processing on the valid field information to obtain standardized data;

[0087] S1c: Based on a preset time period, perform time alignment processing on the standardized data.

[0088] It should be noted that communication protocol parsing refers to parsing the data format uploaded by sensors or devices to extract usable information fields. Valid field information refers to data units with clear meaning in the original protocol data. Structure standardization refers to converting data formats collected by different devices into a unified data structure. Time alignment refers to aligning data from multiple data sources according to a unified time step.

[0089] Specifically, for the various data sources (such as temperature and humidity sensors, PDUs, electricity meters, air conditioning controllers, etc.) using different communication protocols, the corresponding parsing modules are invoked to analyze the protocol frame structure and extract valid field information representing environmental and equipment operating status, such as temperature values, current values, and fan speeds. This step can strip away redundant data and retain key indicators used only for modeling and control.

[0090] Furthermore, the extracted field information is formatted according to a preset unified structure, including unified unit system (e.g., temperature is unified to degrees Celsius), field naming rules, data precision, and field order, thereby generating a standardized data set with a unified structure. To ensure the synchronization of different data sources on the timeline, the system performs time padding and alignment on all standardized data according to a set sampling period (e.g., every 30 seconds), resamples non-isochronous data, and interpolates or marks missing values, so that data of different types can form a complete data frame at the same time node.

[0091] This step, through parsing, standardizing, and aligning the communication protocols of temperature and humidity data, power consumption data, and operating parameters, enables the unified fusion of multi-source heterogeneous data in terms of structure and timing, constructing a data infrastructure that meets modeling requirements. This step effectively solves the problems of data format incompatibility and time synchronization difficulties between different devices, improving the system's usability and consistency of raw data. It provides high-quality, synchronized, multi-dimensional data support for subsequent space thermodynamic modeling, predictive control, and energy-saving decisions, thereby enhancing the system's operational stability and prediction accuracy.

[0092] Based on the first embodiment described above, in this embodiment, after step S5, the method further includes:

[0093] S5a: Obtain the actual temperature data of each temperature acquisition point within the current cycle;

[0094] S5b: Compare the actual temperature data with the predicted temperature value corresponding to the control command to generate a prediction error vector;

[0095] S5c: Based on the prediction error vector, adjust the temperature inertia parameter, air conditioning influence parameter, or load heat generation parameter.

[0096] It should be noted that actual temperature data refers to the actual measured temperature values ​​collected by each temperature sensor within the current control cycle. Temperature prediction values ​​are based on a space thermodynamic model and control commands. The prediction error vector is a vector composed of the difference between the actual temperature and the corresponding predicted temperature at each temperature acquisition point. Parameter adjustment refers to fine-tuning parameters in the model, such as thermal inertia, air conditioning effects, or load heat generation, based on error feedback to correct the predictive capability for the next cycle.

[0097] Specifically, firstly, at the end of the current control cycle, the system collects the actual measured values ​​from all temperature sensors, obtains the actual temperature data for each spatial location within the current cycle, and timestamps it to ensure consistency with the prediction time node. Then, these actual temperature data are compared one-to-one with the previously generated temperature prediction values ​​based on the model, calculating the prediction error for each temperature point to form a complete prediction error vector. This error vector serves as a feedback indicator of model performance, quantifying modeling bias and control error. Finally, parameter feedback correction is performed based on the error vector. If the error shows a persistent deviation trend over multiple cycles, targeted adjustments can be made to temperature inertia parameters (such as temperature decay rate), air conditioning influence parameters (such as conduction coupling coefficient), or load heat generation parameters (such as load thermal coefficient) to achieve online correction of the thermodynamic model and enhance prediction accuracy.

[0098] This step involves collecting actual temperature data within the current cycle and comparing it with the predicted results. The system can then generate a prediction error vector and dynamically adjust key parameters in the thermodynamic model accordingly, effectively constructing a closed-loop learning mechanism of "prediction-execution-feedback-correction." This step continuously optimizes the model's adaptability in real-world scenarios, improves the consistency between prediction and control, reduces energy consumption deviations and control failure risks caused by model drift, thereby further enhancing the energy efficiency and robustness of the air conditioning group control system.

[0099] This embodiment acquires temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment. The temperature and humidity data, power consumption data, and operating parameters are synchronized by timestamp to generate a corresponding multi-source feature data matrix. Based on the multi-source feature data matrix, spatial thermodynamic model parameters are constructed, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. Based on the spatial thermodynamic model parameters, the temperature evolution trend within the target time domain is predicted, and based on the prediction results, a control command sequence for the air conditioning equipment is generated. If the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints, the corresponding control command is sent to the air conditioning equipment. This embodiment acquires temperature and humidity data, power consumption data of load devices, and operating parameters of air conditioning equipment within the computer room. It then performs time synchronization processing on this data to construct a unified multi-source feature data matrix. Based on this matrix, a spatial thermodynamic model is built, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. This enables the control system to accurately reflect the heat distribution and dynamic changes within the computer room. By predicting the temperature evolution trend within the target time domain and generating an air conditioning control command sequence, and combining constraints such as temperature range, rate of change, and control interval to filter control commands, feedforward and energy-saving control of the air conditioning equipment is achieved, thereby effectively improving the energy efficiency and control accuracy of the air conditioning group control.

[0100] Based on the second embodiment described above, a third embodiment of the energy-saving control method for group control of computer room air conditioning systems is proposed in this application. Please refer to... Figure 3 , Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the energy-saving control method for group control of air conditioning in a computer room according to this application.

[0101] In this embodiment, step S2 includes:

[0102] S21: Perform time-granularity-based timestamp normalization on the temperature and humidity data, the power consumption data, and the operating parameters to obtain data alignment results;

[0103] S22: The data alignment results are matched with the corresponding device identifier and timestamp to obtain multi-channel data frames, wherein each data frame corresponds to a time sampling point;

[0104] S23: Arrange the multi-channel data frames in chronological order to generate the multi-source feature data matrix.

[0105] It should be noted that time granularity refers to the uniform sampling interval used during data alignment. The data alignment result refers to the synchronized data set from different data sources at a unified time node, obtained through timestamp normalization. A multi-channel data frame refers to a structured data record generated by aggregating multiple data sources (temperature, humidity, power consumption, operating parameters) at time sampling points. The multi-source feature data matrix is ​​a two-dimensional structure composed of consecutive data frames, with each row corresponding to a time sampling point and each column corresponding to a feature variable.

[0106] Specifically, based on the set time granularity (e.g., 30 seconds per cycle), the timestamps of the raw temperature and humidity data, power consumption data, and air conditioning operating parameters are normalized. This process includes operations such as time format unification, non-sampling point data interpolation, and missing data filling to generate an aligned standard time series, ensuring that various types of data are comparable on the same time base. The alignment results are matched and combined using the timestamp as the primary key and the device identifier as the secondary index. At each time sampling point, the system extracts valid data from various data sources and combines it into multi-channel data frames according to a preset field order. Each frame contains the status values ​​of all devices and sensors at that moment. The data frames corresponding to multiple time sampling points are stacked in chronological order to construct a unified and well-organized multi-source feature data matrix, which supports subsequent thermodynamic modeling and control command generation processes.

[0107] This step, by employing a unified time granularity for timestamp normalization and constructing multi-channel data frames based on this, ultimately generates a structured multi-source feature data matrix. This effectively solves the synchronization problem of multi-source heterogeneous data in the time dimension. This method ensures consistent representation of different data types at the same point in time, providing high-quality, continuous input data support for model training, state prediction, and swarm control strategy optimization, thereby improving the accuracy and efficiency of subsequent spatial thermal modeling and control decisions.

[0108] Based on the second embodiment described above, in this embodiment, step S3 includes:

[0109] S31: With the air conditioning equipment turned off, acquire temperature decay data at multiple time points, and fit the temperature decay data based on the least squares method to obtain temperature inertia parameters.

[0110] S32: When controlling the output of a single air conditioner to change the cooling power, acquire the response temperature change data of each temperature acquisition point in the space, determine the heat conduction correlation coefficient between the air conditioner and the temperature sensing location based on the response temperature change data, and determine the air conditioner influence parameters based on the heat conduction correlation coefficient.

[0111] S33: Under different load device power consumption states, acquire the corresponding space temperature change data, and perform regression analysis based on the space temperature change data and load power consumption value to obtain load heat generation parameters and environmental coupling parameters;

[0112] S34: Determine the parameters of the space thermodynamic model based on the temperature inertia parameter, the air conditioning influence parameter, the load heat generation parameter, and the environmental coupling parameter.

[0113] It should be noted that temperature decay data refers to the data sequence formed by the natural decrease of indoor temperature over time under conditions without external heat source intervention (such as when the air conditioner is off). Least squares is a regression fitting method. The heat conduction correlation coefficient reflects the degree of thermal coupling between the air conditioning equipment and each temperature sampling point. The load heat generation parameter quantifies the heat release capacity caused by the unit power consumption of the load equipment. The environmental coupling parameter characterizes the conduction effect of the external climate environment (such as outdoor temperature) on indoor temperature through the building envelope.

[0114] Specifically, firstly, with the air conditioning system off, the system collects indoor temperature decay data over a continuous period, forming a time series of natural temperature decrease over time. By applying the least squares method to fit this data series, the temperature decay rate or exponential decay coefficient is obtained as a temperature inertia parameter, reflecting the space's heat capacity and its slow-changing response characteristics. Next, using controlled output changes of a single air conditioning unit (e.g., step-wise adjustment of cooling power) as an experimental method, response curves at multiple temperature acquisition points are recorded within the control period. Based on these response data, the heat transfer correlation coefficient between the air conditioner and each measurement point is calculated, thereby quantifying the contribution of air conditioning cooling to the temperature of different areas of the space and constructing air conditioning influence parameters.

[0115] Furthermore, based on this, the response of the rack-mounted equipment to temperature changes within the space under different power consumption states is further analyzed. Regression analysis is performed on the temperature change data and the corresponding power consumption data to extract the contribution rate of unit power consumption to temperature rise, which is used as the load heat generation parameter. Simultaneously, considering the correlation between external air temperature changes and indoor temperature, and combining building structural parameters, environmental coupling parameters are determined through the same regression fitting method. Finally, the above four types of parameters are integrated to construct a spatial thermodynamic model parameter set, providing a dynamic thermal behavior basis for subsequent temperature evolution prediction and control command generation.

[0116] This step involves acquiring temperature response data under different controlled and uncontrolled scenarios, and modeling key elements such as temperature inertia, air conditioning response, load heat release, and external climate disturbances. This allows for the construction of a spatial thermodynamic model parameter set with spatial coupling, temporal correlation, and load adaptability. This modeling approach breaks through the reliance on local static data in traditional temperature control, achieving dynamic modeling oriented towards overall real-world thermal behavior. This helps improve the accuracy of temperature prediction and the model's adaptability to environmental changes, thus providing a solid physical foundation for energy-saving optimization of air conditioning group control.

[0117] Based on the second embodiment described above, in this embodiment, step S4 includes:

[0118] S41: Based on the current environmental state data and the parameters of the space thermodynamic model, predict the temperature change sequence of each temperature acquisition point in the target time domain;

[0119] S42: Based on the prediction results, construct an optimization objective function;

[0120] S43: Generate the control command sequence of the air conditioning equipment with minimizing the optimization objective function as the solution objective of the constrained optimization problem.

[0121] It should be noted that environmental status data refers to the currently collected multi-source feature information, including real-time temperature distribution, air conditioning operating status, load power consumption, etc. Temperature change sequence refers to the sequence of temperature evolution values ​​at each temperature collection point over multiple future time steps, predicted based on a thermodynamic model. The optimization objective function represents the mathematical function of the control optimization objective. Constrained optimization problem refers to the mathematical problem of finding the optimal control variables for the objective function while satisfying constraints (such as temperature boundaries and control intervals). Control command sequence refers to the adjustment commands that the air conditioning equipment should execute over several future control cycles.

[0122] Specifically, based on the current environmental state data (such as the current temperature values, current load power consumption, and air conditioning equipment status) and existing space thermodynamic model parameters, the data is first input into the model prediction module to extrapolate the temperature change sequence of each temperature acquisition point within the target future time domain (e.g., the next 10 minutes). This sequence is used to predict the temperature response trend of the system under different control strategies.

[0123] Furthermore, an optimization objective function is constructed based on the prediction results. This objective function comprehensively considers the following objectives: first, to make all temperature points as close as possible to the target temperature (minimizing temperature tracking error); second, to reduce the total energy consumption of the air conditioning equipment (e.g., represented by compressor power or power consumption); and third, to limit drastic fluctuations in control behavior (avoiding frequent start-stop cycles). The objective function can be weighted and combined according to weight coefficients to form a unified quantitative objective. The objective function and control constraints (such as upper / lower temperature limits, control intervals, and rates of change) are jointly constructed into a constrained optimization problem, which is solved using optimization algorithms (such as quadratic programming, linear programming, mixed integer programming, etc.) to generate a sequence of air conditioning control commands covering the target time domain. The system selects the control command corresponding to the current control cycle in the sequence for issuance and execution.

[0124] This step involves predicting temperature based on current environmental data and space thermodynamic model parameters, and constructing an optimized objective function to guide control command generation, enabling a predictive-driven feedforward control strategy. Compared to traditional passive response control, this method is more proactive and systematic in suppressing temperature fluctuations, controlling energy consumption, and protecting equipment. Utilizing constrained optimization methods to generate control sequences ensures that the control strategy achieves multi-objective trade-offs while meeting system safety and stability requirements, significantly improving the intelligence, energy efficiency, and reliability of air conditioning group control.

[0125] This embodiment acquires temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment. The temperature and humidity data, power consumption data, and operating parameters are synchronized by timestamp to generate a corresponding multi-source feature data matrix. Based on the multi-source feature data matrix, spatial thermodynamic model parameters are constructed, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. Based on the spatial thermodynamic model parameters, the temperature evolution trend within the target time domain is predicted, and based on the prediction results, a control command sequence for the air conditioning equipment is generated. If the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints, the corresponding control command is sent to the air conditioning equipment. This embodiment acquires temperature and humidity data, power consumption data of load devices, and operating parameters of air conditioning equipment within the computer room. It then performs time synchronization processing on this data to construct a unified multi-source feature data matrix. Based on this matrix, a spatial thermodynamic model is built, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters. This enables the control system to accurately reflect the heat distribution and dynamic changes within the computer room. By predicting the temperature evolution trend within the target time domain and generating an air conditioning control command sequence, and combining constraints such as temperature range, rate of change, and control interval to filter control commands, feedforward and energy-saving control of the air conditioning equipment is achieved, thereby effectively improving the energy efficiency and control accuracy of the air conditioning group control.

[0126] Please see Figures 4-6 , Figure 4 This is a diagram of the AI ​​energy-saving group control gateway framework in one embodiment of the energy-saving control method for group control of computer room air conditioning in this application; Figure 5 This is a network structure diagram of the multimodal sensing and execution network in one embodiment of the energy-saving control method for group control of computer room air conditioning in this application; Figure 6 This is a flowchart of model predictive control based on multimodal perception in one embodiment of the energy-saving control method for group control of air conditioning in computer rooms according to this application.

[0127] like Figures 4-6 As shown, in one embodiment, the overall implementation process is as follows.

[0128] (1) Deploy AI energy-saving group control gateway.

[0129] The AI ​​energy-saving group control gateway uses a CPU with a hardware NPU neural network acceleration unit as the main controller, which can perform AI air conditioner group control algorithm inference and prediction locally at high speed. It also integrates a large number of RS232 and RS485 serial ports and network ports, as well as an integrated ZigBee coordinator and LoRa gateway module, which can connect to intelligent devices such as temperature and humidity sensors, PDUs, and air conditioners.

[0130] (2) Deploy a multimodal perception and execution network.

[0131] 1. Deploy a row of temperature and humidity sensors. Deploy temperature and humidity sensors at the air inlet and outlet of each load rack. These sensors can be connected in series via a bus to the AI ​​energy-saving group control gateway; or they can be ZigBee or LoRa wireless temperature and humidity sensor terminals connected to the wireless gateway module of the AI ​​energy-saving group control gateway.

[0132] 2. Deploy PDUs. Deploy PDUs in each rack to collect real-time power consumption data of the rack load. Connect the PDUs to the AI ​​energy-saving group control gateway.

[0133] 3. For air conditioners that cannot provide smart device protocols, install a smart infrared air conditioner controller. The smart infrared air conditioner controller can learn various air conditioner control commands from the air conditioner remote control. Point the remote control at the infrared receiver of the smart infrared air conditioner controller to learn the temperature control commands sequentially. Then, install the infrared transmitter on the air conditioner to achieve air conditioner control via commands. Install a temperature sensor on the air conditioner and connect the sensor cable to the corresponding interface of the smart infrared air conditioner controller to collect the air conditioner's outlet and return air temperatures. Connect all smart air conditioners or smart infrared air conditioner controllers to the AI ​​energy-saving group control gateway.

[0134] (3) In response to the needs of computer room temperature prediction and air conditioning optimization control, this embodiment uses the model predictive control (MPC) framework, and takes the thermodynamic model based on the fusion of physical model and data-driven model as the prediction core.

[0135] Constructing a thermodynamic dynamic model:

[0136] T(t+1)=A·T(t)+B(t)·Q ac (t)+C·P load (t)+D·T env (t)

[0137] in:

[0138] T(t)∈R n : n temperature acquisition point vectors, which are acquired and processed by the AI ​​energy-saving group control gateway from the temperature and humidity sensors;

[0139] Q ac (t)∈R m : m air conditioner cooling capacity vector (0-1 normalized), calculated by the AI ​​energy-saving group control gateway from the real-time power consumption of the corresponding electricity meter at time t of each air conditioner and the maximum power consumption on the air conditioner nameplate;

[0140] P load (t): Business load energy consumption (kW), which is read and processed by the AI ​​energy-saving group control gateway from the rack PDU;

[0141] (): Represents the air temperature outside the building where the computer room is located at time t;

[0142] A-matrix (temperature inertia): calibrated through heat capacity testing experiments, reflecting the temperature decay characteristics of each region;

[0143] B(t) matrix (dynamic air conditioning effect): generated in real time based on the three-dimensional weight tensor W[i,j,k];

[0144] C-coefficient (load heat generation): Based on regression analysis of equipment nameplate power and measured temperature rise data;

[0145] D coefficient (environmental coupling): determined by correlation analysis between external wall temperature sensors and indoor temperature.

[0146] Air conditioning temperature control based on thermodynamic dynamic models utilizes model predictive control (MPC). MPC predicts temperature changes over a future period using a model and solves for the optimal air conditioning control sequence, achieving goals such as temperature stability, energy savings, and reduced equipment operation frequency while satisfying various constraints. The steps are detailed below:

[0147] Set the prediction time domain N p (For example, the next 10 minutes, or 20 time steps, assuming a time step size of 30 seconds); control the time domain N c(Typically shorter than the prediction time domain, e.g., 5 minutes, or 10 time steps), the control variable is solved within the control time domain, after which the control variable remains unchanged. In each control cycle (e.g., every 30 seconds), based on the current state and future disturbance predictions, the future N is solved. c One control quantity (i.e., the air conditioning cooling capacity sequence Q) ac (t), Q ac (t+1), ..., Q ac (t+N c -1)) minimizes the objective function in the prediction time domain while satisfying the constraints.

[0148] The objective function includes:

[0149] Temperature tracking deviation: to make the temperature at each temperature acquisition point as close as possible to the set value (or set range);

[0150] Controlling energy consumption: Minimize the total energy consumption of the air conditioner (related to cooling capacity);

[0151] Controlling fluctuations: Reduce the range of changes in air conditioning cooling capacity and avoid frequent adjustments.

[0152] The objective function is designed as follows:

[0153] in:

[0154] T ref These are the target temperatures at each temperature collection point;

[0155] ΔQ ac (t+k)=Q ac (t+k)-Q ac (t+k-1), t≥1;

[0156] W T W Q W ΔQ It is a diagonal weight matrix used to adjust the importance of each item.

[0157] Constraints

[0158] Temperature constraints: The temperature at each temperature acquisition point must be within a safe range;

[0159] Control interval constraint: The time interval between two control commands cannot be less than the set value (in optimization problems, this can be achieved by limiting the control change to zero in multiple time steps).

[0160] Using a dynamic model, starting from the current state T(t), according to the control sequence Q ac (t), Q ac (t+1), ..., Q ac (t+Nc -1) and known (or predicted) perturbation sequences, predict future P ioad (t+k),T env (t+k), (k=0,...,N) p -1) Temperature of step:

[0161] T(t+1)=A·T(t)+B(t)·Q ac (t)+C·P load (t)+D·T env (t)

[0162] T(t+2)=A·T(t+1)+B(t+1)·Q ac (t+1)+C·P load (t+1)+D

[0163] ·T env (t+1)

[0164] ...

[0165] Optimization Solution: Substitute the above prediction model into the optimization problem to form a quadratic programming (QP) problem (if the objective function is quadratic and the constraints are linear) or a mixed-integer quadratic programming (MIQP) problem (if it contains integer variables, such as the air conditioner's start / stop status). Then use an optimization solver to solve it.

[0166] Rolling execution: After solving the optimization problem, the optimal control sequence is obtained.

[0167] Take only the first control variable This is applied to air conditioning systems. In the next control cycle, the current state is remeasured, the predicted disturbances (load and ambient temperature) are updated, and the above process is repeated.

[0168] Feedback correction: Due to uncertainties in model errors and disturbances, actual temperatures may deviate from predicted temperatures. To enhance robustness, the following methods can be used:

[0169] Feedback correction: In the next control cycle, the actual measured temperature is compared with the predicted temperature, and the deviation is added as a correction term to the prediction at the next moment.

[0170] State estimation: If there is unmeasured state or noise, a Kalman filter can be used for state estimation.

[0171] This application also provides an energy-saving control device for group control of computer room air conditioning systems. Please refer to... Figure 7 , Figure 7This is a schematic diagram of the module structure of an energy-saving control device for group control of computer room air conditioners according to an embodiment of this application. The energy-saving control device for group control of computer room air conditioners includes:

[0172] The data acquisition module 701 is used to acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment.

[0173] Synchronization module 702 is used to synchronize the temperature and humidity data, the power consumption data and the operating parameters according to timestamps to generate a corresponding multi-source feature data matrix;

[0174] The parameter construction module 703 is used to construct space thermodynamic model parameters based on the multi-source feature data matrix. The space thermodynamic model parameters include temperature inertia parameters, air conditioning influence parameters, load heat production parameters, and environmental coupling parameters.

[0175] The prediction module 704 is used to predict the temperature evolution trend in the target time domain based on the parameters of the space thermodynamic model, and generate a control command sequence for the air conditioning equipment based on the prediction results.

[0176] The target module 705 is used to send the corresponding control command to the air conditioning device if the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint.

[0177] The energy-saving control device for group control of air conditioning in computer rooms provided in this application adopts the energy-saving control method for group control of air conditioning in computer rooms described in the above embodiments, and can solve the technical problem of how to improve the energy-saving efficiency of group control of air conditioning. Compared with the prior art, the beneficial effects of the energy-saving control device for group control of air conditioning in computer rooms provided in this application are the same as the beneficial effects of the energy-saving control method for group control of air conditioning in computer rooms provided in the above embodiments, and other technical features in the energy-saving control device for group control of air conditioning in computer rooms are the same as the features disclosed in the methods of the above embodiments, and will not be repeated here.

[0178] This application provides an energy-saving control device for group control of air conditioning in a computer room. The energy-saving control device for group control of air conditioning in a computer room includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the energy-saving control method for group control of air conditioning in a computer room as described in the above embodiments.

[0179] The following is for reference. Figure 8 , Figure 8This is a schematic diagram of the hardware operating environment involved in the energy-saving control method for group control of air conditioning in a computer room according to the embodiments of this application. It shows a schematic diagram of the structure of the energy-saving control equipment suitable for implementing the energy-saving control method for group control of air conditioning in a computer room according to the embodiments of this application. Figure 8 The energy-saving control device for group control of computer room air conditioning shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0180] like Figure 8 As shown, the energy-saving control device for group control of computer room air conditioning systems may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the energy-saving control device for group control of computer room air conditioning systems. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows energy-saving control equipment for group control of computer room air conditioning systems to exchange data wirelessly or via wired communication with other devices. Although the figure shows energy-saving control equipment for group control of computer room air conditioning systems with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented alternatively.

[0181] In particular, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. When the computer program is executed by the processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0182] The energy-saving control device for group control of air conditioning in computer rooms provided in this application adopts the energy-saving control method for group control of air conditioning in computer rooms in the above embodiments, and can solve the technical problem of how to improve the energy-saving efficiency of group control of air conditioning. Compared with the prior art, the beneficial effects of the energy-saving control device for group control of air conditioning in computer rooms provided in this application are the same as the beneficial effects of the energy-saving control method for group control of air conditioning in computer rooms provided in the above embodiments, and other technical features in the energy-saving control device for group control of air conditioning in computer rooms are the same as the features disclosed in the method of the previous embodiment, and will not be repeated here.

[0183] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0184] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0185] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the energy-saving control method for group control of computer room air conditioning in the above embodiments.

[0186] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the energy-saving control equipment for the group control of computer room air conditioning, the energy-saving control equipment for the group control of computer room air conditioning performs the following actions: acquiring temperature and humidity data, power consumption data of load equipment, and operating parameters of air conditioning equipment within the computer room; synchronizing the temperature and humidity data, power consumption data, and operating parameters according to timestamps to generate a corresponding multi-source feature data matrix; constructing spatial thermodynamic model parameters based on the multi-source feature data matrix, including temperature inertia parameters, air conditioning influence parameters, load heat generation parameters, and environmental coupling parameters; predicting the temperature evolution trend within the target time domain based on the spatial thermodynamic model parameters, and generating a control command sequence for the air conditioning equipment based on the prediction results; and sending the corresponding control command to the air conditioning equipment if the control command sequence satisfies preset temperature range constraints, preset temperature change rate constraints, and preset control time interval constraints. Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0187] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0188] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0189] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described energy-saving control method for group control of air conditioning in a computer room, thereby solving the technical problem of how to improve the energy-saving efficiency of group control of air conditioning. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the energy-saving control method for group control of air conditioning in a computer room provided in the above embodiments, and will not be repeated here.

[0190] This application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the energy-saving control method for group control of computer room air conditioning as described above.

[0191] The computer program product provided in this application can solve the technical problem of how to improve the energy efficiency of air conditioning group control. Compared with the prior art, the beneficial effects of the computer program product provided in the embodiments of this application are the same as the beneficial effects of the energy-saving control method for computer room air conditioning group control provided in the above embodiments, and will not be repeated here.

[0192] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent scope of this application.

Claims

1. An energy-saving control method for group control of air conditioning in computer rooms, characterized in that, The method includes: Acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment; The temperature and humidity data, the power consumption data, and the operating parameters are synchronized according to timestamps to generate a corresponding multi-source feature data matrix; Based on the multi-source feature data matrix, space thermodynamic model parameters are constructed, including temperature inertia parameters, air conditioning influence parameters, load heat production parameters, and environmental coupling parameters. Based on the parameters of the space thermodynamic model, the temperature evolution trend in the target time domain is predicted, and based on the prediction results, a control command sequence for the air conditioning equipment is generated. If the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint, then the corresponding control command is sent to the air conditioning device. The step of constructing space thermodynamic model parameters based on the multi-source feature data matrix includes: With the air conditioning equipment turned off, temperature decay data at multiple time points are acquired, and the temperature decay data is fitted based on the least squares method to obtain temperature inertia parameters; wherein, the temperature decay index is a data sequence formed by the natural decrease of temperature in the space over time under the condition of no external heat source intervention, and the temperature inertia coefficient is used to characterize the decay rate or exponential decay coefficient of the temperature in the space. When controlling the output of a single air conditioner to change its cooling power, the response temperature change data of each temperature collection point in the space is acquired. Based on the response temperature change data, the heat conduction correlation coefficient between the air conditioner and the temperature sensing location is determined, and the air conditioner influence parameters are determined according to the heat conduction correlation coefficient. The air conditioner influence parameters are used to quantify the contribution of air conditioner cooling to the temperature of different areas of the space. Under different load device power consumption states, the corresponding space temperature change data is obtained, and regression analysis is performed based on the space temperature change data and load power consumption value to obtain load heat generation parameters and environmental coupling parameters. The parameters of the space thermodynamic model are determined based on the temperature inertia parameter, the air conditioning influence parameter, the load heat generation parameter, and the environmental coupling parameter.

2. The method as described in claim 1, characterized in that, After the steps of acquiring temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment, the following is also included: The temperature and humidity data, the power consumption data, and the communication protocols corresponding to the operating parameters are parsed to extract valid field information. The valid field information is subjected to structural standardization processing to obtain standardized data; Based on a preset time period, the standardized data is time-aligned.

3. The method as described in claim 1, characterized in that, After the step of sending the corresponding control command to the air conditioning device if the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint, the method further includes: Obtain the actual temperature data of each temperature acquisition point within the current cycle; The actual temperature data is compared with the predicted temperature value corresponding to the control command to generate a prediction error vector; Based on the prediction error vector, the temperature inertia parameter, air conditioning influence parameter, or load heat generation parameter are adjusted.

4. The method as described in claim 1, characterized in that, The step of synchronizing the temperature and humidity data, the power consumption data, and the operating parameters according to timestamps to generate a corresponding multi-source feature data matrix includes: The temperature and humidity data, the power consumption data, and the operating parameters are subjected to timestamp normalization processing with a preset time granularity to obtain data alignment results; The data alignment results are matched with the corresponding device identifier and timestamp to obtain multi-channel data frames, wherein each data frame corresponds to a time sampling point; The multi-channel data frames are arranged in chronological order to generate the multi-source feature data matrix.

5. The method as described in claim 1, characterized in that, The step of predicting the temperature evolution trend in the target time domain based on the parameters of the space thermodynamic model, and generating a control command sequence for the air conditioning equipment based on the prediction results, includes: Based on the current environmental state data and the parameters of the space thermodynamic model, the temperature change sequence of each temperature acquisition point in the target time domain is predicted; Based on the prediction results, construct an optimization objective function; The control command sequence of the air conditioning equipment is generated by minimizing the optimization objective function as the solution objective of the constrained optimization problem.

6. An energy-saving control device for group control of air conditioning in computer rooms, characterized in that, The device includes: The data acquisition module is used to acquire temperature and humidity data in the computer room, power consumption data of the load equipment, and operating parameters of the air conditioning equipment. The synchronization module is used to synchronize the temperature and humidity data, the power consumption data, and the operating parameters according to timestamps, and generate a corresponding multi-source feature data matrix. The parameter construction module is used to construct space thermodynamic model parameters based on the multi-source feature data matrix. The space thermodynamic model parameters include temperature inertia parameters, air conditioning influence parameters, load heat production parameters, and environmental coupling parameters. The prediction module is used to predict the temperature evolution trend in the target time domain based on the parameters of the space thermodynamic model, and generate the control command sequence of the air conditioning equipment based on the prediction results. The target module is used to send the corresponding control command to the air conditioning device if the control command sequence satisfies the preset temperature range constraint, the preset temperature change rate constraint, and the preset control time interval constraint. The parameter construction module is further configured to acquire temperature decay data at multiple time points when the air conditioning equipment is turned off, and fit the temperature decay data using the least squares method to obtain temperature inertia parameters. The temperature decay exponent is a data sequence formed by the natural decrease in temperature over time in a space without external heat source intervention, and the temperature inertia coefficient is used to characterize the decay rate or exponential decay coefficient of the temperature in the space. When controlling the changing cooling power output of a single air conditioning unit, the module acquires response temperature change data at each temperature sampling point in the space, determines the heat conduction correlation coefficient between the air conditioner and the temperature sensing location based on the response temperature change data, and determines the air conditioning influence parameters based on the heat conduction correlation coefficient. The air conditioning influence parameters are used to quantify the contribution of air conditioning cooling to the temperature of different areas of the space. Under different load device power consumption states, the module acquires corresponding space temperature change data, and performs regression analysis based on the space temperature change data and load power consumption values ​​to obtain load heat generation parameters and environmental coupling parameters. Based on the temperature inertia parameters, the air conditioning influence parameters, the load heat generation parameters, and the environmental coupling parameters, the space thermodynamic model parameters are determined.

7. A computer device, characterized in that, The device includes: a memory, a processor, and an energy-saving control program for group control of computer room air conditioners stored in the memory and executable on the processor, wherein the energy-saving control program for group control of computer room air conditioners is configured to implement the steps of the energy-saving control method for group control of computer room air conditioners as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores an energy-saving control program for group control of computer room air conditioners. When the processor executes the energy-saving control program for group control of computer room air conditioners, it implements the steps of the energy-saving control method for group control of computer room air conditioners as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the energy-saving control method for group control of computer room air conditioning as described in any one of claims 1 to 5.

Images