A data center wue water consumption efficiency monitoring system

Abstract

The application discloses a data center WUE water consumption efficiency monitoring system and relates to the technical field of data center energy efficiency management.The system comprises a scene monitoring node unit, a distributed peer-to-peer communication unit, a global collaborative intelligent middle station unit and a system integration and linkage control unit.The application adopts a distributed node collaborative architecture, sinks data processing and decision-making capabilities to the monitoring node, realizes local rapid response of water consumption anomalies, effectively avoids the single point failure risk and data transmission delay problem of the traditional centralized architecture, improves the accuracy of WUE calculation through dynamic data calibration based on the equipment feature library, guarantees the consistency and accuracy of global data through a distributed consistency algorithm, realizes efficient scheduling of cross-node computing power resources, forms a complete closed loop of water consumption monitoring, anomaly early warning and automatic disposal through deep integration with existing management systems and equipment linkage control, and reduces the water resource consumption of the data center.

Description

Technical Field

[0001] This invention relates to the field of data center energy efficiency management technology, specifically a data center WUE water consumption efficiency monitoring system. Background Technology

[0002] Data centers are infrastructures that centrally deploy computing, storage, and networking equipment to provide computing power and data storage services for various digital businesses, forming the core foundation for the development of the digital economy. With the widespread application of digital technologies, the data center industry continues to expand, and its energy and water consumption has gradually become a focus of industry attention. Water Use Efficiency (WUE) is a core indicator for evaluating the water resource utilization level of data centers, reflecting the water consumption per unit of electrical energy consumed by a unit of information equipment. Real-time monitoring of data center WUE helps operators comprehensively understand the distribution of water consumption, identify inefficient water use, and is an important way to achieve water conservation and energy saving in data centers, meet green and low-carbon policy requirements, and reduce operating costs.

[0003] However, in existing technologies, data center WUE monitoring mostly adopts a centralized hierarchical architecture. Data needs to be aggregated at multiple levels before being processed uniformly by the central platform. This results in a high risk of single point of failure and large data transmission delays, making it impossible to respond promptly to water consumption anomalies. Some systems rely on manual meter reading and calculation, leading to delayed data updates and significant errors, which are insufficient to meet real-time monitoring requirements. Furthermore, existing monitoring systems mostly only collect basic water and electricity parameters, lacking targeted monitoring of core water-using aspects such as cooling systems. Their integration with existing data center operation and maintenance and energy management systems is low, failing to form a closed-loop management system for anomaly early warning and equipment control. Consequently, the problem of water waste is difficult to solve effectively. Therefore, developing a data center WUE water consumption efficiency monitoring system is of great significance. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a data center WUE water consumption efficiency monitoring system. This system can solve the problems of high single-point failure risk, large data transmission delay, low WUE calculation accuracy, untimely anomaly response, and poor system integration in existing technologies.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a data center WUE water consumption efficiency monitoring system, the system comprising: a scenario-based monitoring node unit, a distributed peer-to-peer communication unit, a global collaborative intelligent middleware unit, and a system integration and linkage control unit; The scenario-based monitoring node unit deploys multiple independent nodes according to the functional areas of the data center, and completes the collection, cleaning and calibration of water consumption data, electricity consumption data and equipment operation status data in the corresponding area, local WUE calculation and local anomaly decision-making, and transmits the processed local data to the global collaborative intelligent platform unit through the distributed peer-to-peer communication unit. The distributed peer-to-peer communication unit constructs a peer-to-peer communication network without a central node, which is used to realize data interaction between each scenario-based monitoring node and the global collaborative intelligent platform, and forward the control commands issued by the global collaborative intelligent platform to the corresponding scenario-based monitoring node. The global collaborative intelligent platform unit integrates the local data uploaded by each node, completes global WUE calculation, data consistency verification and cross-node resource scheduling, and generates control commands and optimization suggestions; The system integration and linkage control unit interacts with the global collaborative intelligent platform unit to connect with the existing management system of the data center, thereby enabling device linkage control and compliant data export.

[0006] Furthermore, the scenario-based monitoring node unit includes a data acquisition module, an edge computing module, a local decision-making module, and a device control module. The edge computing module performs the following operations during data calibration: The system receives raw water consumption data, raw electricity consumption data, and raw equipment operating status data transmitted from the data acquisition module, and uses a sliding window filtering algorithm to remove noise interference from the raw data. The node's built-in device feature library is invoked to compare the currently collected data with the rated operating parameters of the corresponding device; The system extracts operational data from historical data collected during the same period under the same equipment load conditions, corrects the currently collected data, and generates calibrated valid data. The edge computing module then calculates the local WUE based on the calibrated data. The calculation formula is as follows: ,in For the efficiency of water resource use in a single area. This represents the total water consumption after calibration within a continuous statistical period for a single region. The power consumption of calibrated information equipment in a single area within the same statistical period is calculated and the results are corrected for deviations by the metering accuracy coefficient of the built-in equipment at the node. The metering accuracy coefficient is determined by the sensor's factory calibration parameters and the field calibration data.

[0007] Furthermore, the data acquisition module is deployed at key equipment nodes in the data center cooling area, server room area, and clean water area. The collected water usage data includes water flow rate, liquid level, and water quality parameters. The collected power consumption data includes real-time power consumption of information equipment and operating current of cooling equipment. The collected equipment operating status data includes water pump speed, cooling tower fan frequency, and valve opening.

[0008] Furthermore, the local decision module stores preset anomaly judgment rules, which include local WUE value fluctuation thresholds and normal range of device operating parameters. The local decision module compares the real-time local WUE value output by the edge computing module and the device operating parameters with the anomaly judgment rules respectively. When the real-time data exceeds the limit of the anomaly judgment rules, a local control command is generated and sent to the device control module, and an anomaly reporting message is generated and sent to the distributed peer-to-peer communication unit.

[0009] Furthermore, the distributed peer-to-peer communication unit includes a dynamic routing module, a data encryption and fragmentation module, and a QoS priority scheduling module. The dynamic routing module performs the following operations when maintaining the communication topology: It periodically sends heartbeat probe messages to surrounding nodes to obtain their online status and link quality information. The local communication topology table is constructed and updated based on the obtained online status and link quality information; When a direct link to the target node is detected to be interrupted, the nearest node with the best link quality is selected from the local communication topology table as a relay node, and a new communication link is established through multi-hop transmission.

[0010] Furthermore, the data encryption and fragmentation module uses the AES-256 encryption algorithm to encrypt all transmitted data, fragments data packets whose size exceeds a set threshold, adds a unique identifier and sequence number to each fragment data packet, and transmits them to the target node through different communication paths. After receiving all fragment data packets, the target node reassembles them according to the sequence number information to obtain a complete data packet.

[0011] Furthermore, the QoS priority scheduling module divides the transmitted data into three priority levels: the first priority is abnormal alarm data and device control command data, the second priority is real-time WUE data and device operating status data, and the third priority is historical data and report data. The QoS priority scheduling module prioritizes the allocation of communication bandwidth to the first priority data and uses a preemptive transmission method to process the first priority data.

[0012] Furthermore, the global collaborative intelligent platform unit includes a data consistency verification module, a global WUE calculation module, a cross-node resource scheduling module, and an energy efficiency analysis and optimization module. The global WUE calculation module performs the following operations when performing global WUE calculation: Send data synchronization requests to all scenario-based monitoring nodes, and receive local WUE data and corresponding area information device load data uploaded by each node; The Raft algorithm is used to verify the consistency of data uploaded by all nodes. The weighted average of the verified local WUE data is calculated based on the load ratio of information equipment in each region to obtain the global WUE value for the entire data center. The formula for calculating the weighted global WUE is as follows: ,in To improve the overall water resource utilization efficiency of the data center, Let i be the local water resource utilization efficiency of the i-th monitoring area. Let n be the load weighting coefficient for the i-th monitoring area, and n be the total number of monitoring areas in the data center. The load weighting coefficient is determined by the ratio of the power consumption of information equipment in the i-th region to the total power consumption of information equipment in the data center, and satisfies the following conditions: and .

[0013] Furthermore, the data consistency verification module compares the data uploaded by each node, and the data deviation verification calculation formula is as follows: Where δ is the deviation rate of single-node monitoring data, For the water or electricity consumption monitoring data uploaded by the i-th node, The data deviation rate threshold is the arithmetic mean of the monitoring data of all normal nodes of the same type. The data deviation rate judgment threshold is calibrated by the rated measurement accuracy of the monitoring sensor. When the deviation rate of the data uploaded by a certain node exceeds the deviation rate judgment threshold, the data tracing process is triggered to extract the historical interaction data between the node and the neighboring nodes, locate the cause of the data anomaly, generate a data anomaly record and send it to the energy efficiency analysis and optimization module.

[0014] Furthermore, the system integration and linkage control unit includes a standardized interface module, an equipment linkage control module, and a compliance data export module. The standardized interface module uses a RESTful API interface to interact with the data center energy management system, operation and maintenance management system, and equipment management system. The equipment linkage control module receives control commands issued by the global collaborative intelligent platform and coordinates the collaborative operation of water-using equipment in multiple areas. The compliance data export module generates tamper-proof water-using data records and compliance reports in accordance with industry standard formats.

[0015] Compared with existing technologies, this data center WUE water consumption efficiency monitoring system has the following advantages: This invention employs a distributed node collaborative architecture, decentralizing data processing and decision-making capabilities to monitoring nodes. This enables rapid local response to water usage anomalies, effectively avoiding the single-point-of-failure risk and data transmission latency issues inherent in traditional centralized architectures. Dynamic data calibration based on a device feature library improves the accuracy of WUE calculations. A distributed consensus algorithm ensures the consistency and accuracy of global data, enabling efficient scheduling of cross-node computing resources. Deep integration with existing management systems and device linkage control forms a complete closed loop of water usage monitoring, anomaly early warning, and automatic handling, reducing water consumption in data centers and improving operational efficiency and compliance.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

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

[0018] Figure 1 This is a schematic diagram of a data center WUE water consumption efficiency monitoring system. Figure 2 A flowchart of a data center WUE water consumption efficiency monitoring system; Figure 3 This is a flowchart of the edge computing module performing data calibration operations. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] This invention provides a data center water consumption efficiency (WUE) monitoring system, comprising four core components: scenario-based monitoring node units, distributed peer-to-peer communication units, a global collaborative intelligent platform unit, and a system integration and linkage control unit. The system adopts a distributed node collaborative architecture, decentralizing data processing and decision-making capabilities to monitoring nodes in various functional areas. It leverages edge computing to clean and calibrate raw data and perform local WUE calculations, combined with a local decision-making module to achieve rapid response to water consumption anomalies, avoiding the single-point-of-failure and transmission delay problems of traditional centralized architectures. The distributed peer-to-peer communication unit constructs a decentralized peer-to-peer communication network, ensuring the stability, security, and real-time performance of data transmission through dynamic routing, data encryption and sharding, and QoS priority scheduling. The global collaborative intelligent platform unit integrates data from various nodes, performing data verification and global WUE weighted calculations through a distributed consensus algorithm, enabling cross-node computing resource scheduling and energy efficiency optimization. The system integration and linkage control unit connects to the existing data center management system through standardized interfaces, forming a complete closed loop of water consumption monitoring, anomaly early warning, and automatic handling. Through precise data calibration and efficient collaborative control, it improves WUE calculation accuracy, reduces data center water consumption, and meets the requirements of green, low-carbon operation and compliance management. The following is a detailed description with reference to specific embodiments.

[0021] This embodiment selects a medium-sized data center built by a large internet company as the application scenario. The data center has a total building area of ​​18,000 square meters and primarily hosts cloud computing and big data storage services. It operates 1,200 server racks daily and is divided into three core functional areas: a cooling area, a server room area, and a clean water area. The cooling area is equipped with 8 crossflow cooling towers, 12 cooling water pumps, and 6 plate heat exchangers, and is the core water-consuming area of ​​the data center. The server room area houses servers, switches, UPS power supplies, and other information equipment, and is the core area for power consumption and heat dissipation. The clean water area is equipped with water purification equipment and water terminals for offices and restrooms, handling daily cleaning and domestic water supply. This embodiment, based on the water and electricity consumption characteristics of this data center, completes the full-process deployment, debugging, and operation of the data center's WUE (Water Usage Efficiency) monitoring system, achieving real-time monitoring of water efficiency across the entire area, rapid handling of anomalies, and overall energy efficiency optimization.

[0022] like Figure 1As shown, this embodiment strictly follows the data center's functional zoning and business requirements to complete the physical deployment and logical networking of the four core units. Scenario-based monitoring node units deploy independent edge monitoring nodes in the cooling zone, server room zone, and clean water zone. Each node collects water usage data, electricity consumption data, and equipment operating status data for its corresponding area. Nodes are physically isolated but logically interconnected to prevent a single node failure from affecting overall monitoring. Distributed peer-to-peer communication units are deployed in the data center's core network server room, using a combination of industrial-grade switches and wireless sensor modules to construct a peer-to-peer communication network without a central node. This eliminates the traditional master-slave structure of centralized communication, enabling bidirectional peer-to-peer data interaction between each monitoring node and the global collaborative intelligent platform.

[0023] The global collaborative intelligent middleware unit is deployed on the core server of the data center operation and maintenance management system. Equipped with high-performance computing chips and distributed storage components, it undertakes core tasks such as data integration across the entire data center, global WUE calculation, resource scheduling, and energy efficiency analysis. The system integration and linkage control unit is deployed on the terminal of the data center operation and maintenance management platform. It connects to the existing energy management system, operation and maintenance management system, and equipment management system through standardized interfaces, breaking down data barriers and enabling equipment linkage control and compliant data export.

[0024] The scenario-based monitoring node unit is the core of the system's front-end perception and edge computing. It includes a data acquisition module, an edge computing module, a local decision-making module, and a device control module. These four modules work together to complete regional-level data acquisition, cleaning and calibration, local WUE calculation, and local anomaly handling, which is the foundation for achieving distributed rapid response.

[0025] The data acquisition module is precisely deployed according to the key equipment nodes of the data center's functional areas. In the cooling area, acquisition terminals are deployed at the inlet and outlet pipes of the cooling tower, the outlet of the cooling water pump, the heat exchange end of the plate heat exchanger, and the liquid level monitoring point of the storage tank. In the computer room area, acquisition terminals are deployed at the power supply circuit of the server rack, the water circuit of the computer room precision air conditioning, and the power supply end of the rack cooling fan. In the clean water area, acquisition terminals are deployed at the water outlet of the water purification equipment, the main public water pipeline, and the branch water pipeline of the restroom. The acquisition terminals collect data in real time with a cycle of 1 second. The collected water data includes water flow, liquid level, water quality conductivity, and pH value parameters; the collected power data includes real-time power consumption of information equipment and operating current and voltage parameters of cooling equipment; the collected equipment operating status data includes water pump speed, cooling tower fan frequency, and water valve opening parameters. After completing the raw data acquisition, the acquisition module directly transmits the data to the edge computing module of the corresponding node, without going through multiple relays, reducing data transmission loss.

[0026] like Figure 3As shown, after receiving the raw data, the edge computing module completes the data calibration process in three steps. The first step executes a sliding window filtering algorithm with a sliding window duration of 10 seconds to remove noise and abnormal spikes caused by electromagnetic interference, equipment vibration, and voltage fluctuations, retaining stable and valid raw monitoring data. The second step calls the node's built-in device feature library to compare the currently collected data with the rated operating parameters of the corresponding equipment. For example, the rated speed of a cooling water pump is 1450 rpm, and the rated frequency of a cooling tower fan is 50 Hz. If the collected data deviates from the rated parameters by more than 5%, it is marked as data to be corrected. The third step extracts operating data from the same historical period under the same equipment load conditions and dynamically corrects the collected data based on the current ambient temperature and equipment load rate, ultimately generating calibrated and valid data.

[0027] In the specific implementation of this embodiment, the edge computing module calculates the local WUE of a single region based on the calibrated data. The calculation formula is as follows: , in the formula It represents the efficiency of water resource use in a single region. Represents the total water consumption after calibration within a continuous statistical period in a single region. This represents the power consumption of calibrated information equipment in a single area within the same statistical period. In this embodiment, the statistical period is set to 1 hour, with a local WUE calculation performed every hour. After the local WUE calculation is completed, the final deviation is corrected using the device's built-in measurement accuracy coefficient. This coefficient is determined by both the sensor's factory calibration parameters and the on-site calibration data. The sensor's factory calibration parameters are the manufacturer-provided 0.2% accuracy level, and the on-site calibration data are the measurement deviation values ​​verified monthly by data center maintenance personnel. The weighted average of these two values ​​yields the final measurement accuracy coefficient, ensuring that the local WUE calculation accuracy is controlled within 0.1%.

[0028] The local decision-making module has a built-in pre-set anomaly judgment rule library, which includes local WUE value fluctuation thresholds and normal ranges for equipment operating parameters. In this embodiment, based on historical data from the data center, the local WUE value fluctuation threshold is set at 10%, the normal range for cooling water pump speed is 90-110% of rated speed, the normal range for cooling tower fan frequency is 30-50 Hz, and the normal range for water valve opening is 20-100%. The local decision-making module receives the local WUE value and equipment operating parameters output by the edge computing module in real time, comparing the real-time data with the anomaly judgment rules one by one. When the real-time data exceeds the rule limits, a local control command is immediately generated and sent to the equipment control module, and an anomaly reporting message is simultaneously generated and sent to the distributed peer-to-peer communication unit. For example, if the speed of a cooling water pump in the cooling zone suddenly increases to 1600 rpm, exceeding the normal operating range, the equipment control module directly adjusts the pump inverter to reduce the speed to the rated value, without waiting for instructions from the central platform, achieving a local second-level anomaly response.

[0029] The distributed peer-to-peer communication unit is a bridge connecting the front-end monitoring nodes and the back-end intelligent platform. It includes a dynamic routing module, a data encryption and sharding module, and a QoS priority scheduling module, which solves the problems of link interruption, data leakage, and transmission congestion in traditional communication architectures.

[0030] The dynamic routing module sends heartbeat probe messages to neighboring monitoring nodes and the global platform every 5 seconds. It obtains link quality information such as the online status of surrounding nodes, data transmission latency, and packet loss rate through these messages. Based on the real-time probe results, the module automatically builds and updates its local communication topology table, which records the location, link quality, and communication path of all online nodes in real time. When a direct link to the target node is detected to be interrupted due to line damage or network failure, the module selects the nearest node with the best link quality from the local communication topology table as a relay node. It then quickly rebuilds a new communication link using multi-hop transmission, ensuring uninterrupted data transmission with a link switching time controlled within 1 second.

[0031] The data encryption and fragmentation module employs the AES-256 encryption algorithm to encrypt all transmitted data throughout the entire process. The encryption key is automatically updated every 24 hours to prevent data theft or tampering during transmission. For large data packets exceeding 1MB, the module automatically performs fragmentation, adding a unique identifier and sequential sequence number to each fragment, which is then distributed and transmitted to the target node via different communication paths. After receiving all fragmented data packets, the target node automatically reassembles them according to the sequence number information to restore the complete data packet, avoiding issues such as transmission delays and data loss when transmitting large data packets.

[0032] The QoS priority scheduling module divides all transmitted data into three priority levels: first priority is abnormal alarm data and device control command data; second priority is real-time WUE data and device operating status data; and third priority is historical data and report data. The scheduling module prioritizes allocating communication bandwidth to first-priority data and uses a preemptive transmission method to process it. Even if third-priority data is being transmitted, it will be immediately paused and given priority to first-priority data, ensuring that abnormal alarms and device control commands can be transmitted in real time without bandwidth being consumed by other data, thus guaranteeing the timeliness of anomaly handling.

[0033] The global collaborative intelligent middleware unit is the core brain of the system, which includes a data consistency verification module, a global WUE calculation module, a cross-node resource scheduling module, and an energy efficiency analysis and optimization module, to complete global data integration, accurate calculation, and intelligent scheduling.

[0034] After receiving water and electricity consumption monitoring data uploaded by each monitoring node, the data consistency verification module performs a horizontal comparison of similar data and eliminates abnormal deviation data. In the specific implementation of this embodiment, the data deviation verification calculation formula is as follows: In the formula, δ represents the deviation rate of single-node monitoring data. This represents the water or electricity consumption monitoring data uploaded by the i-th node. This represents the arithmetic mean of similar monitoring data from all normal nodes. The data deviation rate threshold is determined by the rated measurement accuracy of the monitoring sensor. In this embodiment, the rated measurement accuracy of the sensor is 0.5%, therefore, the deviation rate threshold is set to 0.5%. When the data deviation rate uploaded by a node exceeds 0.5%, the system automatically triggers the data tracing process, extracts the historical interaction data between that node and its neighboring nodes, and the equipment operation logs, locates the cause of the data anomaly, generates a data anomaly record, and sends it to the energy efficiency analysis and optimization module, providing a basis for subsequent operation and maintenance.

[0035] The global WUE calculation module first sends a data synchronization request to all scenario-based monitoring nodes, uniformly collecting local WUE data and corresponding regional information device load data from each node. It then uses the Raft algorithm to perform consistency verification on the data from all nodes, ensuring the accuracy of the data used in the calculation. In this specific implementation, the verified local WUE data is weighted and averaged according to the load ratio of information devices in each region to obtain the global WUE value for the entire data center. The global WUE weighted calculation formula is: , in the formula Represents the overall water resource utilization efficiency of the data center. This represents the local water resource utilization efficiency of the i-th monitoring area. represents the load weighting coefficient for the i-th monitoring area, and n represents the total number of monitoring areas within the data center. Load weighting coefficient The power consumption of information equipment in the i-th region is determined by the ratio of its power consumption to the total power consumption of information equipment in the data center. In this embodiment, the power consumption ratios of information equipment in the cooling zone, server room zone, and clean water zone are 20%, 75%, and 5%, respectively. Therefore, the corresponding load weighting coefficients are 0.2, 0.75, and 0.05, respectively. All load weighting coefficients satisfy... Furthermore, the sum of all coefficients is 1, ensuring the rationality and accuracy of the global WUE weighted calculation.

[0036] The cross-node resource scheduling module intelligently schedules the coordinated operation of water-using equipment in various areas based on the global WUE calculation results and the operating status of each node's equipment. When the global WUE value is high, the scheduling module issues optimization instructions to the cooling zone, reducing the frequency of cooling tower fans, adjusting the number of operating cooling water pumps, and optimizing the opening of water valves to reduce water consumption while meeting the heat dissipation needs of the computer room. When the WUE in a local area is abnormally low, the scheduling module coordinates with equipment in neighboring areas to supplement water supply, balancing the water efficiency of the entire data center. The energy efficiency analysis and optimization module integrates data such as data anomaly records, global WUE change trends, and equipment runtime to generate daily, weekly, and monthly energy efficiency analysis reports, and proposes optimization suggestions such as cooling tower cleaning, water pump energy-saving retrofits, and water system leak detection, providing data support for long-term water conservation and consumption reduction in the data center.

[0037] The system integration and linkage control unit is the core of the integration of the system with the existing management system of the data center, and includes a standardized interface module, a device linkage control module, and a compliant data export module.

[0038] The standardized interface module uses a RESTful API to seamlessly integrate with the data center energy management system, operation and maintenance management system, and equipment management system. It synchronizes water usage data, electricity consumption data, WUE metrics, and equipment operating status in real time, breaking down data silos between systems. The equipment linkage control module receives control commands from the global collaborative intelligent platform, coordinating the operation of water-using equipment in multiple areas. For example, when the temperature in the server room area rises, it coordinates with the cooling area to increase the speed of cooling water pumps and the flow rate of cooling water, ensuring effective heat dissipation while controlling water usage efficiency. When the liquid level in the clean water area is low, it coordinates with the water purification equipment to increase water production to meet daily water needs. The compliance data export module generates tamper-proof water usage data records, WUE index reports, and water-saving compliance reports according to the industry standard format for data center energy efficiency management. It supports export in PDF, Excel, and other formats, meeting the on-site inspection and data reporting requirements of environmental protection departments and industry regulatory agencies.

[0039] like Figure 2As shown, after system startup, it enters a fully automated operation state. First, the scenario-based monitoring nodes complete the raw data collection for each functional area. The edge computing module performs data cleaning and calibration, and local WUE calculation. The local decision-making module determines in real time whether there are any water usage anomalies. If no water usage anomalies are found, the nodes directly upload the calibrated local data to the distributed peer-to-peer communication unit. If water usage anomalies are found, the nodes first execute local automatic handling through the device control module, and simultaneously generate an anomaly message which is uploaded to the communication unit. The distributed peer-to-peer communication unit encrypts and transmits the data to the global collaborative intelligent platform. The platform completes data consistency verification, global WUE calculation, cross-node resource scheduling, and energy efficiency analysis, and then pushes the global WUE data and optimization instructions to the system integration and linkage control unit. Finally, the linkage control unit completes data integration with the external management system, collaborative device control, and compliance data export, forming a complete closed loop of water usage monitoring, anomaly warning, automatic handling, energy efficiency optimization, and compliance management.

[0040] In summary, this embodiment, by fully deploying the WUE (Water Usage Effectiveness) monitoring system in a medium-sized data center, completely solves the core problems of traditional centralized monitoring architectures, such as high single-point failure risk, large data transmission latency, low WUE calculation accuracy, untimely anomaly response, and poor system integration. The system adopts a distributed node collaborative architecture, decentralizing data processing and decision-making capabilities to edge nodes, reducing water usage anomaly response time from minutes to seconds in traditional architectures, and lowering data transmission latency. Dynamic calibration based on device feature libraries and historical data improves WUE calculation accuracy, with data errors far lower than traditional manual meter reading and centralized calculation methods. A distributed peer-to-peer communication network ensures the stability and security of data transmission, while QoS priority scheduling ensures priority transmission of core alarm and control data, eliminating transmission congestion. The global collaborative intelligent platform, through the Raft algorithm and weighted average calculation, achieves global data consistency verification and accurate global WUE calculation, and cross-node resource scheduling reduces water resource consumption in the data center.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A data center WUE water consumption efficiency monitoring system, characterized in that, The system includes: a scenario-based monitoring node unit, a distributed peer-to-peer communication unit, a global collaborative intelligent platform unit, and a system integration and linkage control unit; The scenario-based monitoring node unit deploys multiple independent nodes according to the functional areas of the data center, and completes the collection, cleaning and calibration of water consumption data, electricity consumption data and equipment operation status data in the corresponding area, local WUE calculation and local anomaly decision-making, and transmits the processed local data to the global collaborative intelligent platform unit through the distributed peer-to-peer communication unit. The distributed peer-to-peer communication unit constructs a peer-to-peer communication network without a central node, which is used to realize data interaction between each scenario-based monitoring node and the global collaborative intelligent platform, and forward the control commands issued by the global collaborative intelligent platform to the corresponding scenario-based monitoring node. The global collaborative intelligent platform unit integrates the local data uploaded by each node, completes global WUE calculation, data consistency verification and cross-node resource scheduling, and generates control commands and optimization suggestions; The system integration and linkage control unit interacts with the global collaborative intelligent platform unit to connect with the existing management system of the data center, thereby enabling device linkage control and compliant data export.

2. The data center WUE water consumption efficiency monitoring system according to claim 1, characterized in that, The scenario-based monitoring node unit includes a data acquisition module, an edge computing module, a local decision-making module, and a device control module. The edge computing module performs the following operations during data calibration: The system receives raw water consumption data, raw electricity consumption data, and raw equipment operating status data transmitted from the data acquisition module, and uses a sliding window filtering algorithm to remove noise interference from the raw data. The node's built-in device feature library is invoked to compare the currently collected data with the rated operating parameters of the corresponding device; The system extracts operational data from historical data collected during the same period under the same equipment load conditions, corrects the currently collected data, and generates calibrated valid data. The edge computing module then calculates the local WUE based on the calibrated data. The calculation formula is as follows: ,in For the efficiency of water resource use in a single area. This represents the total water consumption after calibration within a continuous statistical period for a single region. This refers to the power consumption of calibrated information equipment in a single region within the same statistical period.

3. The data center WUE water consumption efficiency monitoring system according to claim 2, characterized in that, The data acquisition module is deployed at key equipment nodes in the data center cooling area, computer room area, and clean water area. The collected water usage data includes water flow rate, liquid level, and water quality parameters. The collected power consumption data includes real-time power consumption of information equipment and operating current of cooling equipment. The collected equipment operating status data includes water pump speed, cooling tower fan frequency, and valve opening.

4. The data center WUE water consumption efficiency monitoring system according to claim 2, characterized in that, The local decision module stores preset anomaly judgment rules, which include local WUE value fluctuation thresholds and normal range of device operating parameters. The local decision module compares the real-time local WUE value output by the edge computing module and the device operating parameters with the anomaly judgment rules respectively. When the real-time data exceeds the limit of the anomaly judgment rules, a local control command is generated and sent to the device control module, and an anomaly reporting message is generated and sent to the distributed peer-to-peer communication unit.

5. A data center WUE water consumption efficiency monitoring system according to claim 1, characterized in that, The distributed peer-to-peer communication unit includes a dynamic routing module, a data encryption and fragmentation module, and a QoS priority scheduling module. The dynamic routing module performs the following operations when maintaining the communication topology: It periodically sends heartbeat probe messages to surrounding nodes to obtain their online status and link quality information. The local communication topology table is constructed and updated based on the obtained online status and link quality information; When a direct link to the target node is detected to be interrupted, the nearest node with the best link quality is selected from the local communication topology table as a relay node, and a new communication link is established through multi-hop transmission.

6. A data center WUE water consumption efficiency monitoring system according to claim 5, characterized in that, The data encryption and fragmentation module uses the AES-256 encryption algorithm to encrypt all transmitted data. Data packets exceeding a set threshold are fragmented, and each fragment is given a unique identifier and sequence number. They are then transmitted to the target node through different communication paths. After receiving all fragmented data packets, the target node reassembles them into a complete data packet based on the sequence number information.

7. A data center WUE water consumption efficiency monitoring system according to claim 5, characterized in that, The QoS priority scheduling module divides the transmitted data into three priority levels: the first priority is abnormal alarm data and device control command data, the second priority is real-time WUE data and device operating status data, and the third priority is historical data and report data. The QoS priority scheduling module prioritizes the allocation of communication bandwidth to the first priority data and uses a preemptive transmission method to process the first priority data.

8. A data center WUE water consumption efficiency monitoring system according to claim 1, characterized in that, The global collaborative intelligent platform unit includes a data consistency verification module, a global WUE calculation module, a cross-node resource scheduling module, and an energy efficiency analysis and optimization module. The global WUE calculation module performs the following operations when performing global WUE calculation: Send data synchronization requests to all scenario-based monitoring nodes, and receive local WUE data and corresponding area information device load data uploaded by each node; The Raft algorithm is used to verify the consistency of data uploaded by all nodes. The weighted average of the verified local WUE data is calculated based on the load ratio of information equipment in each region to obtain the global WUE value for the entire data center. The formula for calculating the weighted global WUE is as follows: ,in To improve the overall water resource utilization efficiency of the data center, Let i be the local water resource utilization efficiency of the i-th monitoring area. Let be the load weight coefficient for the i-th monitoring area, and n be the total number of monitoring areas in the data center.

9. A data center WUE water consumption efficiency monitoring system according to claim 8, characterized in that, The data consistency verification module compares the data uploaded by each node, and the data deviation verification calculation formula is as follows: Where δ is the deviation rate of single-node monitoring data, For the water or electricity consumption monitoring data uploaded by the i-th node, The arithmetic mean of the monitoring data of the same type for all normal nodes is used. When the deviation rate of the data uploaded by a certain node exceeds the deviation rate judgment threshold, the data tracing process is triggered to extract the historical interaction data between the node and its neighboring nodes, locate the cause of the data anomaly, generate a data anomaly record, and send it to the energy efficiency analysis and optimization module.

10. A data center WUE water consumption efficiency monitoring system according to claim 1, characterized in that, The system integration and linkage control unit includes a standardized interface module, an equipment linkage control module, and a compliance data export module. The standardized interface module uses a RESTful API interface to interact with the data center energy management system, operation and maintenance management system, and equipment management system. The equipment linkage control module receives control commands issued by the global collaborative intelligent platform and coordinates the collaborative operation of water-using equipment in multiple areas. The compliance data export module generates tamper-proof water-using data records and compliance reports in accordance with industry standard formats.

Images