Machine room temperature monitoring method and device, equipment, storage medium and program product

Abstract

The invention relates to the technical field of temperature monitoring, and discloses a machine room temperature monitoring method, device and equipment, a storage medium and a program product, and the system comprises a plurality of fixed sensors, a plurality of mobile sensors, a temperature field simulation module and a temperature field prediction module. The multiple fixed sensors are arranged at preset positions of a machine room and used for collecting first temperature data of the preset positions. The plurality of mobile sensors are used for collecting second temperature data of the target area according to the control instruction; the temperature field simulation module is used for simulating according to a preset working condition range of the machine room to obtain first temperature field simulation data of the machine room under a plurality of working conditions; and the temperature field prediction module is used for constructing a three-dimensional temperature field prediction model according to the working condition parameters, the first temperature field simulation data, the first temperature data and the second temperature data, and predicting a three-dimensional temperature field of the machine room according to the target working condition parameters of the machine room and the three-dimensional temperature field prediction model, so that the temperature field data are ensured to be comprehensive, flexible and reliable.

Description

Technical Field

[0001] This invention relates to the field of temperature monitoring technology, specifically to methods, devices, equipment, storage media, and software products for monitoring computer room temperature. Background Technology

[0002] Servers in data center server rooms release a large amount of heat during operation, requiring air conditioning systems to continuously dissipate it to maintain stable server operation. However, the entire process of heat generation and output is dynamic; localized overheating can affect the safe operation of servers, while excessively cold environments will result in energy waste. Therefore, real-time monitoring of the temperature distribution within the server room and adjusting it to a reasonable range is crucial.

[0003] In related technologies, the current method for monitoring computer room temperature relies on manual experience, involving the even distribution of fixed temperature sensors in some server rack locations, and additional fixed sensors at locations such as the air conditioning vents. This method is limited by factors such as manual experience and the number of sensors, resulting in incomplete monitoring of the three-dimensional temperature field of the computer room and potential omissions. Furthermore, under the trend of energy conservation and carbon reduction in data centers, it cannot monitor the temperature distribution in any location within the computer room in real time, nor can it adjust the airflow organization in a timely manner to reduce energy consumption. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, storage medium, and program product for monitoring computer room temperature, in order to solve the problem that computer room temperature monitoring methods are limited by human experience and the number of sensors, making it impossible to perform comprehensive monitoring.

[0005] In a first aspect, the present invention provides a computer room temperature monitoring system, comprising: multiple fixed sensors, multiple mobile sensors, a temperature field simulation module, and a temperature field prediction module; multiple fixed sensors are arranged at preset positions in the computer room to collect first temperature data at the preset positions; multiple mobile sensors are used to collect second temperature data of a target area according to control commands; the temperature field simulation module is used to simulate the first temperature field simulation data of the computer room under multiple operating conditions according to a preset operating condition range of the computer room; the temperature field prediction module is used to construct a three-dimensional temperature field prediction model based on the operating condition parameters of multiple operating conditions, the first temperature field simulation data, and the first and second temperature data under multiple first operating conditions, the three-dimensional temperature field prediction model being used to characterize the correlation between the operating condition parameters of the computer room and the three-dimensional temperature field data, the multiple first operating conditions being included in multiple operating conditions; the temperature field prediction module is also used to obtain the target operating condition parameters of the computer room when receiving a temperature prediction request, and predict the three-dimensional temperature field data of the computer room based on the target operating condition parameters and the three-dimensional temperature field prediction model.

[0006] The data center temperature monitoring system provided by this invention includes multiple fixed sensors, multiple mobile sensors, a temperature field simulation module, and a temperature field prediction module. Fixed sensors collect first temperature data at preset core locations within the data center, providing stable benchmark anchors for monitoring and avoiding data loss in critical areas due to human experience. Mobile sensors can flexibly enter blind spots not covered by fixed sensors according to control commands, collecting second temperature data for target areas, essentially equipping the monitoring system with dynamic probes and filling physical coverage gaps such as cabinet gaps and the area around newly added equipment. Simultaneously, the temperature field simulation module performs temperature field simulation calculations within the operating range required by the temperature field prediction algorithm, compensating for operating ranges that cannot be provided by physical sensors. Based on this, the temperature field prediction module constructs a three-dimensional temperature field prediction model of the data center using data center operating data, first temperature data, second temperature data, and first temperature field simulation data, achieving complete deduction of the data center temperature field. This upgrades discrete-point monitoring of related technologies to continuous monitoring throughout the entire space, completely eliminating monitoring omissions in the three-dimensional temperature field and allowing maintenance personnel to clearly understand the temperature distribution of every space in real time.

[0007] In some optional implementations, the system further includes a data transmission module connected to multiple fixed sensors, multiple mobile sensors, and a temperature field prediction module, for synchronously acquiring first temperature data collected by each fixed sensor and second temperature data collected by each mobile sensor, and synchronously transmitting the acquired first temperature data and second temperature data to the temperature field prediction module.

[0008] Secondly, the present invention provides a method for monitoring the temperature of a computer room, which is applied to the computer room temperature monitoring system of the first aspect or its corresponding embodiment. The method includes: acquiring target operating parameters of the computer room under target operating conditions; and inputting the target operating parameters into a pre-constructed three-dimensional temperature field prediction model so that the three-dimensional temperature field prediction model outputs three-dimensional temperature field data of the computer room under the target operating conditions.

[0009] The data center temperature monitoring method provided by this invention first acquires the target operating condition parameters of the data center, and then inputs the target operating condition parameters into a preset three-dimensional temperature field prediction model. The model can deduce continuous three-dimensional temperature field data of the entire data center based on the operating condition parameters, so that operation and maintenance no longer rely on experience to infer blind spot temperature, realize full-space monitoring without dead angles, and effectively avoid the risk of missing local overheating. Moreover, the data can be updated in real time, and the model can synchronously output the latest temperature field according to the dynamic changes of the room's thermal environment, supporting advance prediction of trends and timely adjustments, solving the problem of lag in traditional monitoring.

[0010] In one alternative implementation, the three-dimensional temperature field prediction model is constructed through the following steps: The process involves acquiring operating parameters for multiple operating conditions, simulation data of the first temperature field of the computer room under each operating condition, and temperature data collected by multiple sensors under multiple first operating conditions. Each operating condition consists of multiple first operating conditions and at least multiple second operating conditions. The simulation data of the first temperature field under each operating condition is corrected to obtain the first target three-dimensional temperature field data for that operating condition. The temperature data collected by multiple sensors under each first operating condition is used to replace the temperature data at corresponding positions in the first target three-dimensional temperature field data to obtain the second target three-dimensional temperature field data for that first operating condition. The operating parameters of the multiple operating conditions are used as first training data, and the second target three-dimensional temperature field data for each first operating condition and the first target three-dimensional temperature field data for each second operating condition are used as second training data. The first and second training data are correlated to obtain a correlated dataset. A preset neural network model is trained using the correlated dataset until the model accuracy meets preset conditions, resulting in a three-dimensional temperature field prediction model.

[0011] In one optional implementation, the temperature field simulation data for each operating condition is corrected to obtain the first target three-dimensional temperature field data for the corresponding operating condition. This includes: aligning the first temperature field simulation data for each operating condition with temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding operating condition; inputting the second temperature field simulation data for each operating condition into a pre-built data correction model so that the data correction model outputs the temperature deviation data for the corresponding operating condition; and correcting the second temperature field simulation data based on the temperature deviation data for each operating condition to obtain the first target three-dimensional temperature field data for the corresponding operating condition.

[0012] In one optional implementation, the first temperature field simulation data for each operating condition is aligned with the temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding operating condition. This includes: using a preset random sampling consensus algorithm to coarsely match the first temperature field simulation data for each operating condition with the temperature data to obtain the coarsely matched first temperature field simulation data; and using an iterative nearest point algorithm to finely match the coarsely matched first temperature field simulation data for each operating condition with the temperature data to obtain the second temperature field simulation data for the corresponding operating condition.

[0013] Thirdly, the present invention provides a computer room temperature monitoring device for performing the method of the second aspect above. The device includes: an acquisition module for acquiring target operating condition parameters of the computer room under target operating conditions; and a determination module for inputting the target operating condition parameters into a pre-constructed three-dimensional temperature field prediction model, so that the three-dimensional temperature field prediction model outputs three-dimensional temperature field data of the computer room under target operating conditions.

[0014] Fourthly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the computer room temperature monitoring method of the first aspect or any corresponding embodiment described above.

[0015] Fifthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the computer room temperature monitoring method described in the first aspect or any corresponding embodiment thereof.

[0016] In a sixth aspect, the present invention provides a computer program product, including computer instructions for causing a computer to execute the computer room temperature monitoring method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

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

[0018] Figure 1 This is a structural block diagram of a computer room temperature monitoring system according to an embodiment of the present invention; Figure 2 This is a schematic flowchart of a first method for monitoring computer room temperature according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a second process for a computer room temperature monitoring method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the overall process of data acquisition, prediction model construction, and temperature prediction result output of the computer room temperature monitoring method according to an embodiment of the present invention. Figure 5 This is a structural block diagram of a computer room temperature monitoring device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

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

[0020] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0021] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0022] In related technologies, the current method for monitoring computer room temperature relies on manual experience, involving the even distribution of fixed temperature sensors in some server rack locations, and additional fixed sensors at locations such as the air conditioning vents. This method is limited by factors such as manual experience and the number of sensors, resulting in incomplete monitoring of the three-dimensional temperature field of the computer room and potential omissions. Furthermore, under the trend of energy conservation and carbon reduction in data centers, it cannot monitor the temperature distribution in any location within the computer room in real time, nor can it adjust the airflow organization in a timely manner to reduce energy consumption.

[0023] In view of this, this application provides a computer room temperature monitoring system, such as... Figure 1 As shown, the system includes: multiple fixed sensors 101, multiple moving sensors 102, a temperature field simulation module 103, and a temperature field prediction module 104. Multiple fixed sensors are arranged at preset locations in the computer room to collect the first temperature data at the preset locations.

[0024] For example, the fixed sensor 101 can be a high-precision temperature sensor, and its preset location may include, but is not limited to, the air outlet and return air outlet of the computer room air conditioner, the top, middle, and bottom of the cabinet near the air conditioner's cold air outlet and hot air inlet. The mobile sensor 102 can be composed of a mobile temperature sensor mounted on an intelligent inspection robot. In this embodiment, high-precision temperature sensors are installed at the air outlet and return air outlet of the computer room air conditioner to achieve real-time monitoring of airflow temperature; temperature sensors are evenly distributed at the top, middle, and bottom of the cabinet to ensure that the vertical temperature distribution of each cabinet is completely captured; redundant sensors are set near key cabinets near the air conditioner's cold air outlet and hot air inlet to improve monitoring reliability. Each cabinet is typically equipped with more than 6 sensors to ensure the accuracy and comprehensiveness of temperature distribution.

[0025] Multiple motion sensors 102 are used to acquire second temperature data of the target area according to control commands.

[0026] For example, in this embodiment, the mobile sensor 102 consists of an intelligent inspection robot equipped with a mobile temperature sensor. It can move autonomously within the computer room and, according to system instructions, travel to specific areas to supplement temperature field data collection. Here, the entire computer room space is segmented and coded to facilitate system instruction issuance and the intelligent inspection robot's location-based temperature data collection. The mobile sensor employs a low-power design and is equipped with a high-precision temperature measurement module to ensure accurate temperature data is acquired during movement.

[0027] The temperature field simulation module 103 is used to simulate and obtain the first temperature field simulation data of the computer room under multiple operating conditions according to the preset operating conditions range of the computer room. For example, the first temperature field simulation data is obtained based on computational fluid dynamics simulation (CFD) and is used to characterize the temperature information at different locations in the computer room. In this embodiment, a three-dimensional geometric model is first constructed based on the actual structure of the computer room (such as server racks, air conditioners, and wall locations), and then the model is divided into a large number of small computational grids (dissolving continuous space into discrete units); then boundary conditions and a physical model are set; subsequently, CFD software is used to solve the fluid dynamics and thermal equations using numerical calculation methods to obtain the temperature, airflow velocity, and other data of each grid unit, forming the first temperature field simulation data covering the entire space of the computer room.

[0028] The temperature field prediction module 104 is used to construct a three-dimensional temperature field prediction model based on the operating parameters of multiple operating conditions, the simulation data of the first temperature field, and the first and second temperature data under multiple first operating conditions. The three-dimensional temperature field prediction model is used to characterize the correlation between the operating parameters of the computer room and the three-dimensional temperature field data. Multiple first operating conditions are included in multiple operating conditions. The temperature field prediction module 104 is also used to obtain the target operating parameters of the computer room and predict the three-dimensional temperature field data of the computer room based on the target operating parameters and the three-dimensional temperature field prediction model.

[0029] For example, in this embodiment, the operating parameters are the core operating parameters affecting the generation and changes of the thermal environment of the computer room. These mainly include air conditioning system operating parameters, such as supply and return air temperature, wind speed, and operating mode, which directly determine the cooling supply and airflow direction; server and rack load parameters (such as real-time rack power, which is the main heat source of the computer room); computer room space and structural parameters (such as the length, width, and height of the computer room, and the location of racks and air conditioning vents, affecting airflow diffusion and heat transfer paths); and external and environmental auxiliary parameters (such as outdoor temperature and humidity, and the status of the fresh air system, which indirectly interfere with the thermal balance of the computer room). These data together constitute a complete driving system for the thermal environment, providing a macroscopic basis for three-dimensional temperature field prediction. The temperature field prediction module 104 predicts the three-dimensional temperature field data of the computer room under the target operating conditions based on the target operating parameters and a pre-built neural network model. The target operating condition can be any operating condition of the computer room; this embodiment does not limit the specific content of the target operating condition, which can be determined by those skilled in the art according to their needs.

[0030] In some optional embodiments, the system further includes a data transmission module, which is connected to a plurality of fixed sensors 101, a plurality of mobile sensors 102 and a temperature field prediction module 104, respectively, for synchronously acquiring first temperature data collected by each fixed sensor and second temperature data collected by each mobile sensor, and synchronously transmitting the acquired first temperature data and second temperature data to the temperature field prediction module.

[0031] For example, in this embodiment, the data transmission module is used for the synchronous acquisition and transmission of multi-source sensor data, ensuring seamless integration of data from fixed and mobile sensor networks. The fixed sensor network uses RS485 bus communication, transmitting temperature data to the central processing unit via the Modbus protocol. Due to its strong anti-interference capability and long transmission distance, the RS485 protocol is well-suited for deployment in data center server rooms with complex electromagnetic environments. To address compatibility issues between devices using different protocols, a serial port server is introduced to convert between RS485 and Ethernet protocols, enabling sensor data to be transmitted to the processing layer via TCP / IP. The mobile sensor network uses 5G communication technology, leveraging its low latency and high bandwidth to achieve real-time data interaction between mobile devices and the central processing unit. The mobile device uses visual SLAM technology to obtain its three-dimensional coordinates within the server room, transmitting temperature data and location information together to the central processing unit.

[0032] To ensure data synchronization, the system employs the TPSN time synchronization protocol, which uses a hierarchical structure to achieve time synchronization between nodes within the sensor network. This protocol is similar to the NTP protocol in traditional networks, but it has been optimized for sensor networks, effectively eliminating errors caused by access time and improving the synchronization of data acquisition.

[0033] The data center temperature monitoring system provided in this application uses fixed sensors at preset core locations within the data center to collect primary temperature data, providing a stable benchmark and avoiding data loss in critical areas due to human experience. Simultaneously, mobile sensors can flexibly enter blind spots not covered by the fixed sensors according to control commands, collecting secondary temperature data for the target area. This effectively equips the monitoring system with dynamic probes, filling physical coverage gaps such as cabinet gaps and areas around newly added equipment. Furthermore, the temperature field simulation module performs simulation calculations of the data center temperature field within the required operating range of the temperature field prediction algorithm, compensating for operating ranges not provided by physical sensors. Based on this, the temperature field prediction module integrates data center operating data, primary temperature data, secondary temperature data, and primary temperature field simulation data to construct a prediction model. Through model deduction, it generates complete three-dimensional temperature field data for the data center, upgrading discrete point monitoring to continuous monitoring throughout the entire space. This completely eliminates monitoring omissions in the three-dimensional temperature field, allowing maintenance personnel to clearly understand the temperature distribution of every space in real time.

[0034] According to an embodiment of the present invention, a method for monitoring computer room temperature is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] This embodiment provides a method for monitoring computer room temperature, which can be used in the temperature field prediction module of the aforementioned computer room temperature monitoring system. Figure 2 This is a flowchart of a computer room temperature monitoring method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain the target operating condition parameters of the computer room under the target operating conditions.

[0036] For example, target operating condition parameters collected during the period to be monitored are obtained. The specific content of the monitoring period is not limited in this embodiment, and those skilled in the art can determine it according to their needs. For details of the relevant data, please refer to the descriptions in the above embodiments, which will not be repeated here.

[0037] Step S202: Input the target operating condition parameters into the pre-built three-dimensional temperature field prediction model so that the three-dimensional temperature field prediction model outputs the three-dimensional temperature field data of the computer room.

[0038] For example, in the embodiments of this application, the three-dimensional temperature field prediction model is constructed based on a neural network model.

[0039] The data center temperature monitoring method provided in this embodiment first obtains the target operating condition parameters of the data center, and then inputs the target operating condition parameters into a preset three-dimensional temperature field prediction model. The model can deduce continuous three-dimensional temperature field data of the entire data center based on the operating condition parameters, so that operation and maintenance no longer rely on experience to infer blind spot temperature, realize full-space monitoring without dead angles, and effectively avoid the risk of missing local overheating. Moreover, the data can be updated in real time, and the model can synchronously output the latest temperature field according to the dynamic changes of the room's thermal environment, supporting advance prediction of trends and timely adjustments, solving the problem of lag in traditional monitoring.

[0040] This embodiment provides a method for monitoring computer room temperature, which can be used in the temperature field prediction module of a computer room temperature monitoring system. Figure 3 This is a flowchart of a computer room temperature monitoring method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: Obtain the target operating condition parameters of the computer room under the target operating conditions. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0041] Step S302: Input the target operating condition parameters into the pre-built three-dimensional temperature field prediction model so that the three-dimensional temperature field prediction model outputs the three-dimensional temperature field data of the computer room.

[0042] In some alternative implementations, the three-dimensional temperature field prediction model is constructed through the following steps: Step a1: Obtain operating parameters for multiple operating conditions, simulation data of the first temperature field of the computer room under each operating condition, and temperature data collected by multiple sensors under multiple first operating conditions. The multiple operating conditions consist of multiple first operating conditions and at least multiple second operating conditions.

[0043] For example, the temperature data collected by multiple sensors refers to the temperature data obtained by fusing high-frequency global data (first temperature data) from a fixed sensor network with low-frequency hotspot data (second temperature data) from a mobile sensor network through coordinate mapping and time alignment methods.

[0044] Step a2 involves correcting the simulation data of the first temperature field for each working condition to obtain the simulation data of the target temperature field for the corresponding working condition.

[0045] Specifically, step a2 above includes: Step a21: Align the first temperature field simulation data for each working condition with the temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding working condition.

[0046] For example, firstly, a pre-defined Random Sample Consensus (RANSAC) algorithm is used to coarsely match the first temperature field simulation data for each operating condition with the temperature data collected by multiple sensors, resulting in coarsely matched first temperature field simulation data. The pre-defined RANSAC algorithm can include, but is not limited to, improved RANSAC algorithms, and the specific processing steps include: (1) Data preprocessing and format standardization, including: Sensor data: Extract the spatial coordinates (x, y, z) and corresponding temperature value T_sensor of each sensor; construct the sensor point cloud: P_sensor = {(x_i, y_i, z_i, T_i)}.

[0047] Simulation data: The simulated temperature field is regarded as a regular grid point cloud, and each grid point has coordinates and temperature value T_sim; downsampling processing is used to form the simulation point cloud: P_sim = {(x_j, y_j, z_j, T_j)}.

[0048] Normalization: Normalize the coordinates to avoid the impact of dimensional differences on distance calculation.

[0049] (2) Feature extraction, including: Since temperature is a scalar, it cannot be directly used for geometric matching in RANSAC. Therefore, a joint space-temperature feature vector needs to be constructed: For each point, construct a feature vector: Geometric features: coordinates (x, y, z); Temperature features: temperature value T; Temperature gradient features: calculate the local temperature gradient (…). T), which reflects the "direction of heat flow".

[0050] (3) Coarse registration using the improved RANSAC, including: Random sampling: Randomly select 3 or more points from the sensor point cloud. Find the nearest neighbor matching point in the simulated point cloud based on feature distance.

[0051] Estimating the transformation matrix: Using the least squares method, estimate the rigid body transformation T = [R|t] (rotation + translation) from the simulation point to the sensor point.

[0052] Interior point determination: After transforming all simulation points with T, obtain p_sim', and calculate the weighted error of its spatial distance and temperature difference with the nearest sensor point: error = α·||p_sim' - p_sensor||+ β·|T_sim - T_sensor|.

[0053] If the error is less than the threshold, it is an interior point.

[0054] Iterative optimization: Repeat the above process and select the transformation T_initial with the most interior points as the coarse registration result.

[0055] Second, the Iterative Closest Point (ICP) algorithm is used to finely match the coarsely matched first temperature field simulation data under each operating condition with temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding operating condition, specifically including: (1) Initialization: Transform the simulated point cloud using the transformation T_initial obtained from RANSAC.

[0056] (2) Iterative nearest neighbor matching: For each transformed simulation point, find the nearest neighbor in the sensor point cloud based on spatial distance.

[0057] Optimize the objective function: Minimize the following weighted error: E = Σw_i · [ ||p_sim_i' - p_sensor_i|| 2 + λ |T_sim_i - T_sensor_i| 2 w_i can be weighted based on spatial location, and λ controls the weight of temperature error.

[0058] (3) Solve for the optimal transformation: Use nonlinear optimization to solve for the new R and t.

[0059] (4) Iterative convergence: Repeat matching and optimization until the transformation change is less than the threshold or the maximum number of iterations is reached.

[0060] (5) Result fusion: The original simulation grid points are mapped to the sensor coordinate system using the final transformation T_final. At this time, the simulated temperature field and the sensor data are aligned in space, and the aligned simulation data is obtained.

[0061] Step a22: Input the second temperature field simulation data of each working condition into the pre-built data correction model so that the data correction model outputs the temperature deviation data of the corresponding working condition.

[0062] For example, in this embodiment of the application, a neural network model is trained, with the input being simulated second temperature field data and the output being the temperature deviation from that of the actual sensor. After the model training is completed, a data-corrected model is obtained. The temperature deviation data refers to the difference between the simulated second temperature field data and the actual sensor temperature data.

[0063] Step a23: Based on the temperature deviation data of each working condition, the simulation data of the second temperature field is corrected to obtain the first target three-dimensional temperature field data of the corresponding working condition.

[0064] For example, the embodiments of this application do not specifically limit the specific modification process, as long as it is reasonable.

[0065] Step a3: Replace the temperature data at the corresponding position in the three-dimensional temperature field data of the first target with the temperature data collected by multiple sensors under each first working condition to obtain the three-dimensional temperature field data of the second target under the corresponding first working condition.

[0066] For example, in the embodiments of this application, the temperature value collected by the sensor is used at the sensor location; in the sensorless area, the corrected simulated temperature value is used to obtain the three-dimensional temperature field data of the second target.

[0067] Step a4: Use the operating parameters of multiple operating conditions as the first training data, use the second target three-dimensional temperature field data of each first operating condition and the first target three-dimensional temperature field data of each second operating condition as the second training data, and associate the first training data and the second training data to obtain the associated dataset.

[0068] For example, in the embodiments of this application, the specific association method is not specifically limited, and those skilled in the art can determine it according to their needs.

[0069] Step a5: Train the preset neural network model using the associated dataset until the model accuracy meets the preset conditions, and obtain the three-dimensional temperature field prediction model.

[0070] For example, the preset neural network model may include, but is not limited to, the U-net neural network. In this embodiment, 60% of the data in the associated dataset is selected as the U-net neural network training set data for learning model weights and biases; 20% of the data is selected as the U-net neural network validation set data for determining the optimal model structure and adjusting the model's hyperparameters; and the remaining 20% ​​of the fused data is used as the U-net neural network test set data for testing the model's accuracy.

[0071] In this process, creating the CFD simulation dataset requires a significant amount of simulation time. Therefore, after designing the required simulation range through orthogonal experiments to meet the initial training of the U-net neural network, no further CFD simulation datasets will be added. The dataset collected by temperature sensors is generated in real time, thus continuously inputting data throughout the entire temperature field prediction model (U-net neural network) operation for dynamic model optimization, ensuring model accuracy. The system introduces a dynamic correction method, dynamically adjusting the model's prediction results based on the deviation between real-time sensor data and prediction results. Specifically, when the temperature deviation exceeds 0.5℃, the model correction mechanism is triggered, importing the temperature sensor data from the most recent 5 minutes into the U-net neural network using a temperature sliding window method for model fine-tuning, improving prediction reliability. Furthermore, the system periodically updates the model at time intervals (e.g., every 5 minutes) to ensure the long-term stability of the prediction results. A flowchart of a specific example of the data center temperature monitoring method provided in this application embodiment is shown below. Figure 4 As shown.

[0072] After triggering the correction, the system first collects sensor data to construct a sliding window dataset; then, it uses this data to fine-tune the model; finally, it verifies the performance of the corrected model to ensure that the prediction accuracy meets the requirements. The system's dynamic correction and update mechanism ensures the real-time performance and accuracy of the temperature field prediction results.

[0073] Step S303: Send the three-dimensional temperature field data of the computer room to the display terminal for display.

[0074] For example, the display terminal may include, but is not limited to, a display screen. In this embodiment, the system provides an intuitive 3D visualization interface, allowing users to monitor the temperature field distribution in the computer room in real time. The visualization interface uses digital twin technology to map the 3D temperature field model onto the physical space of the computer room, visually displaying the temperature distribution through different colors. The interface also supports hotspot tracking; when an abnormal temperature area is detected, the view is automatically focused on that area, improving operational efficiency. The system can also locate the source of the problem through correlation analysis (such as correlation between temperature anomalies and air conditioning malfunctions), providing root cause analysis.

[0075] Furthermore, the implementation steps for monitoring the computer room temperature include: (1) System deployment: First, based on the size and layout of the computer room, design a deployment scheme for the fixed sensor network and determine the number and location of sensors in each cabinet; then, configure the mobile sensor network, including the number of inspection robots, movement path planning and communication protocol settings; finally, build a central processing platform to realize data acquisition, fusion, prediction and visualization.

[0076] (2) Model training: Collect CFD simulation data and historical sensor data to build a hybrid training set; design a neural network learning model architecture based on U-net; and use a hybrid training strategy to train the model.

[0077] (3) System integration: Develop data acquisition and transmission interfaces to ensure that data from fixed and mobile sensor networks can be acquired and transmitted synchronously; realize the fusion processing of multi-source data, and realize real-time prediction and dynamic correction of temperature field based on prediction models; develop a three-dimensional visualization interface and intelligent alarm system.

[0078] (4) Testing and optimization: Conduct system testing in an actual computer room environment to verify the accuracy and real-time performance of temperature field monitoring; optimize sensor deployment scheme, path planning algorithm and prediction model based on test results; conduct long-term operation testing to ensure the stability and reliability of the system.

[0079] The method provided in this application has the following beneficial effects: (1) Comprehensive coverage: The combination of fixed sensor network, mobile sensor network and prediction model ensures comprehensive coverage of the temperature field in the computer room and solves the problem of blind spots in traditional methods.

[0080] (2) Hotspot tracking: Mobile sensor networks can accurately locate and continuously monitor areas with abnormal temperatures, improving the efficiency and accuracy of hotspot detection.

[0081] (3) Real-time prediction: The U-net-based neural network model compresses the prediction time to the second level, and the global average prediction accuracy error is controlled within 0.5℃, which meets the requirements of real-time monitoring.

[0082] (4) Dynamic correction: The system can dynamically adjust the prediction results based on real-time sensor data to improve the adaptability and reliability of the model.

[0083] (5) Economic benefits: Through precise environmental control and fault early warning, the system can reduce the energy consumption of the data center and improve energy utilization efficiency. It reduces the PUE value of the data center and saves electricity costs. At the same time, the system can reduce unnecessary equipment maintenance and replacement costs and improve the overall operating efficiency of the computer room.

[0084] (6) Safety benefits: By improving the reliability of the data center, the system reduces equipment failures and data loss caused by abnormal temperatures, thus ensuring the continuity of critical business and data security.

[0085] This embodiment also provides a computer room temperature monitoring device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0086] This embodiment provides a computer room temperature monitoring device, such as... Figure 5 As shown, it includes: The acquisition module 501 is used to acquire the target operating condition parameters of the computer room under the target operating conditions. The determination module 502 is used to input the target operating condition parameters into the pre-built three-dimensional temperature field prediction model so that the three-dimensional temperature field prediction model outputs the three-dimensional temperature field data of the computer room under the target operating condition.

[0087] In some alternative implementations, the three-dimensional temperature field prediction model is constructed through the following steps: The system acquires operating parameters for multiple operating conditions, simulation data of the first temperature field of the computer room under each operating condition, and temperature data collected by multiple sensors under multiple first operating conditions. The multiple operating conditions consist of multiple first operating conditions and at least one second operating condition. The simulation data of the first temperature field under each working condition are corrected to obtain the first target three-dimensional temperature field data of the corresponding working condition. The temperature data at the corresponding position in the three-dimensional temperature field data of the first target is replaced by the temperature data collected by multiple sensors under each first working condition to obtain the three-dimensional temperature field data of the second target under the corresponding first working condition. The working parameters of multiple working conditions are used as the first training data, and the three-dimensional temperature field data of the second target of each first working condition and the three-dimensional temperature field data of the first target of each second working condition are used as the second training data. The first training data and the second training data are associated to obtain the associated dataset. The pre-defined neural network model is trained using the associated dataset until the model accuracy meets the pre-defined conditions, thus obtaining a three-dimensional temperature field prediction model.

[0088] In some optional implementations, the temperature field simulation data for each operating condition is corrected to obtain the first target three-dimensional temperature field data for the corresponding operating condition, including: The first temperature field simulation data for each working condition is aligned with the temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding working condition. The simulation data of the second temperature field under each working condition is input into the pre-built data correction model so that the data correction model outputs the temperature deviation data of the corresponding working condition. The simulation data of the second temperature field is corrected based on the temperature deviation data of each working condition to obtain the first target three-dimensional temperature field data of the corresponding working condition.

[0089] In some optional implementations, the first temperature field simulation data for each operating condition is aligned with temperature data collected by multiple sensors to obtain the second temperature field simulation data for the corresponding operating condition, including: The first temperature field simulation data of each working condition is coarsely matched with the temperature data using a preset random sampling consistency algorithm to obtain the coarsely matched first temperature field simulation data. The iterative nearest point algorithm is used to finely match the first temperature field simulation data after coarse matching under each working condition with the temperature data to obtain the second temperature field simulation data for the corresponding working condition.

[0090] The computer room temperature monitoring device provided in this embodiment of the invention can execute the computer room temperature monitoring method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0091] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0092] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0093] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0094] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the computer room temperature monitoring method of the embodiments of the present invention.

[0095] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0096] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the computer room temperature monitoring method shown in the above embodiments is implemented.

[0097] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0098] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A machine room temperature monitoring system, characterized by, The system comprises: a plurality of fixed sensors, a plurality of mobile sensors, a temperature field simulation module, and a temperature field prediction module; The plurality of fixed sensors are arranged at preset positions of the machine room and are used to collect first temperature data at the preset positions; The plurality of mobile sensors are used to collect second temperature data of a target region according to a control instruction; The temperature field simulation module is used to simulate first temperature field simulation data of the machine room under a plurality of working conditions according to a preset working condition range of the machine room; The temperature field prediction module is used to construct a three-dimensional temperature field prediction model according to working condition parameters of the plurality of working conditions, the first temperature field simulation data, and the first temperature data and the second temperature data under a plurality of first working conditions, the three-dimensional temperature field prediction model is used to represent a correlation between the working condition parameters and three-dimensional temperature field data of the machine room, and the plurality of first working conditions are included in the plurality of working conditions; The temperature field prediction module is further used to obtain target working condition parameters of the machine room when a temperature prediction request is received, and predict three-dimensional temperature field data of the machine room according to the target working condition parameters and the three-dimensional temperature field prediction model.

2. The system of claim 1, wherein, The system further comprises a data transmission module connected with the plurality of fixed sensors, the plurality of mobile sensors, and the temperature field prediction module, and used to synchronously obtain the first temperature data collected by each fixed sensor and the second temperature data collected by each mobile sensor, and synchronously transmit the obtained first temperature data and second temperature data to the temperature field prediction module.

3. A method of monitoring the temperature of a machine room, characterized by, The method applied to the machine room temperature monitoring system of claim 1 or 2 comprises: obtaining target working condition parameters of the machine room under a target working condition; inputting the target working condition parameters into a pre-constructed three-dimensional temperature field prediction model, so that the three-dimensional temperature field prediction model outputs three-dimensional temperature field data of the machine room under the target working condition.

4. The method of claim 3, wherein, The three-dimensional temperature field prediction model is constructed by the following steps: obtaining working condition parameters of a plurality of working conditions, first temperature field simulation data of the machine room under each working condition, and temperature data collected by a plurality of sensors under a plurality of first working conditions, the plurality of working conditions are composed of the plurality of first working conditions and at least one second working condition; performing correction processing on the first temperature field simulation data of each working condition to obtain first target three-dimensional temperature field data corresponding to the working condition; replacing temperature data at corresponding positions in the first target three-dimensional temperature field data with the temperature data collected by the plurality of sensors under each first working condition to obtain second target three-dimensional temperature field data under the corresponding first working condition; using the working condition parameters of the plurality of working conditions as first training data, using the second target three-dimensional temperature field data of each first working condition and the first target three-dimensional temperature field data of each second working condition as second training data, and associating the first training data and the second training data to obtain an association data set; training a preset neural network model by using the association data set until the model accuracy meets a preset condition, and obtaining a three-dimensional temperature field prediction model.

5. The method of claim 4, wherein, The correction processing on the first temperature field simulation data of each working condition to obtain the first target three-dimensional temperature field data corresponding to the working condition comprises: The first temperature field simulation data of each working condition is aligned with the temperature data collected by the plurality of sensors to obtain second temperature field simulation data corresponding to the working condition; The second temperature field simulation data of each working condition is input into a pre-constructed data correction model to enable the data correction model to output temperature deviation data corresponding to the working condition; The second temperature field simulation data is corrected based on the temperature deviation data of each working condition to obtain first target three-dimensional temperature field data corresponding to the working condition.

6. The method of claim 5, wherein, The first temperature field simulation data of each working condition is aligned with the temperature data collected by the plurality of sensors to obtain second temperature field simulation data corresponding to the working condition, including: The first temperature field simulation data of each working condition is coarsely matched with the temperature data by using a preset random sampling consistency algorithm to obtain the first temperature field simulation data after coarse matching; The first temperature field simulation data after coarse matching under each working condition is finely matched with the temperature data by using an iterative closest point algorithm to obtain the second temperature field simulation data corresponding to the working condition.

7. A machine room temperature monitoring device, characterized by, The device for executing the method of claim 3 comprises: An acquisition module configured to acquire target working condition parameters of the machine room under a target working condition; A determination module configured to input the target working condition parameters into a pre-constructed three-dimensional temperature field prediction model to enable the three-dimensional temperature field prediction model to output three-dimensional temperature field data of the machine room under the target working condition.

8. An electronic device, comprising: It comprises: A memory and a processor, which are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the machine room temperature monitoring method of any one of claims 3 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing a computer to perform the machine room temperature monitoring method of any one of claims 3 to 6.

10. A computer program product, characterised in that, The computer instructions are used to cause a computer to perform the machine room temperature monitoring method of any one of claims 3 to 6.

Images