Analysis method for energy-saving machine room environment data, monitoring system and storage medium

Abstract

The invention provides an energy-saving machine room environment data analysis method, a monitoring system and a storage medium. The method comprises the steps that energy-saving machine room environment data is collected, a cloud platform sets a threshold value, and a Z-Score algorithm is used for calculating the distance between a feature point and an average value to judge an abnormal value. And constructing a random decision tree to isolate abnormal values, setting an abnormal value standard to judge environmental data, and analyzing the environmental data to obtain three analysis results. Based on the three analysis results, abnormal value information is output if all the analysis results are abnormal, data alarm information is output if two analysis results are abnormal, and data deviation information is output if one analysis result is abnormal. And sending an analysis result to a machine room manager. According to the invention, environmental data is comprehensively analyzed, abnormal values are rapidly identified, real-time feedback is realized, rapid response of managers is ensured, the labor cost is reduced, and the intelligent level is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer room environment data monitoring and analysis, and in particular to an analysis method, a monitoring system and a storage medium for energy-saving computer room environment data. Background Art

[0002] In the current industrial Internet and artificial intelligence big data field, the monitoring and analysis of computer room environmental data plays a vital role, aiming to ensure the stable operation of equipment, optimize resource allocation, and effectively prevent the occurrence of failures. However, traditional methods mainly rely on manual inspections and manual analysis, which has revealed its limitations when faced with a large group of computer room equipment. Frequent data collection and analysis tasks not only consume a lot of human resources, but also come with high costs. In addition, the manual operation process is cumbersome and easily interfered by human factors, which may cause the accuracy of data analysis to be impaired or key information to be missed.

[0003] When an abnormality or fault is detected, the time required for manual processing often becomes a potential factor for the spread of the fault or the obstruction of equipment operation. More seriously, manual inspections are limited in time and cannot achieve uninterrupted monitoring around the clock, leaving blind spots in monitoring. At the same time, due to limited human resources, data analysis is often superficial and difficult to touch the deep-level laws and trends behind the data, which reduces the level of intelligence in computer room operation and maintenance. Summary of the invention

[0004] The present invention aims to at least solve the technical problem in the prior art that manual analysis reduces the intelligence level of computer room operation and maintenance, and particularly innovatively proposes an analysis method, monitoring system and storage medium for energy-saving computer room environmental data.

[0005] In order to achieve the above-mentioned object of the present invention, the present invention provides a method for analyzing environmental data of an energy-saving computer room, the method comprising: S100, collecting environmental data of the energy-saving computer room, classifying the environmental data, and sending the environmental data to a cloud platform, wherein the environmental data includes temperature data, humidity data, voltage data, and current data; S200, the cloud platform sets a threshold, obtains an average value based on the environmental data, calculates the distance between the feature point and the average value by a Z-Score algorithm, and when the distance is greater than the threshold, determines that the feature point is an abnormal value; when the distance is less than or equal to the threshold, determines that the feature point is a normal value; S300, constructing at least two random decision trees, and using the random decision trees to isolate outliers; S400, respectively setting abnormal value standards for the temperature data, humidity data, voltage data and current data, and when the collected environmental data is not within the set abnormal value standards, determining that the data is an abnormal value; when the collected environmental data is within the set abnormal value standards, determining that the data is a normal value; S500, based on the three analysis results of step S200, step S300 and step S400, if the three analysis results of the environmental data are all abnormal values, output abnormal value information; if two analysis results of the environmental data are abnormal values, output data warning information; if one analysis result of the environmental data is an abnormal value, output data deviation information; S600: Send the output abnormal value information, data warning information and data deviation information to the computer room management personnel.

[0006] As an optional embodiment of the present invention, optionally, in step S300, constructing a random decision tree includes: S301, dividing the environmental data into a training set and a test set; S302, using information gain as a criterion for feature selection in the training set to obtain a feature subset; S303, using nonlinear transformation to fuse the feature subsets to generate fused features, using the fused features to construct several random decision trees, and using randomly selected features to perform node splitting to generate candidate splitting points: S304, calculating the Gini impurity of each candidate split point, selecting the split point with the smallest Gini impurity as the split node, and recursively performing node splitting until the set stop condition is met, thereby completing the construction of the random decision tree; S305, evaluating the random decision tree using the test set; S306: Based on the evaluation results, iteratively optimize the random decision tree.

[0007] As an optional embodiment of the present invention, optionally, in step S400, setting the abnormal value standard includes: S401, real-time acquisition of historical temperature data, historical humidity data, historical voltage data and historical current data; S402, respectively calculating the average value and standard deviation of the historical temperature data, the historical humidity data, the historical voltage data, and the historical current data; S403, setting an upper limit and a lower limit of abnormal values ​​for each data based on the mean value and the standard deviation; S404, setting an update period for the mean value and the standard deviation, and updating the upper limit and the lower limit of the abnormal value in real time according to the updated mean value and the standard deviation; S405: Calculate a seasonal adjustment factor, and perform seasonal adjustment on the abnormal value upper bound and the abnormal value lower bound according to the real-time seasonal adjustment factor.

[0008] As an optional embodiment of the present invention, optionally, calculating the seasonal adjustment factor in step S405 includes: S4051, performing seasonal trend analysis on the historical temperature data, historical humidity data, historical voltage data, and historical current data to obtain temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics; S4052. Based on the temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics, respectively calculate the temperature seasonal adjustment factor, humidity seasonal adjustment factor, voltage seasonal adjustment factor, and current seasonal adjustment factor.

[0009] As an optional embodiment of the present invention, optionally, the formula for calculating the upper bound of the outlier value and the lower bound of the outlier value is: Outlier upper bound: Outlier lower bound: in, represents the upper limit of outliers, represents the average value, represents the setting coefficient, represents the standard deviation, represents the seasonal adjustment factor.

[0010] In another aspect, the present invention further provides an automated monitoring system for environmental data of an energy-saving computer room, the system comprising an analysis method for environmental data of an energy-saving computer room; The system further comprises: Data collection module, used to collect environmental data of energy-saving computer rooms in real time; A data transmission module, connected to the data acquisition module, for sending the environmental data to a cloud platform; A data analysis module, which is arranged in the cloud platform and is used to execute the analysis method, analyze the environmental data, and generate abnormal value information, data warning information, and data deviation information; An alarm module, connected to the data analysis module, for generating corresponding sound, light and electricity information according to abnormal value information, data alarm information and data deviation information; A data storage module, connected to the data analysis module and the data transmission module, for storing historical environmental data, analysis results and alarm records; The display module is connected with the data storage module, the alarm module and the data analysis module.

[0011] As another optional embodiment of the present invention, optionally, the system further includes: The execution module is connected to the data analysis module and the energy-saving computer room equipment, and is used to adjust the state of the energy-saving computer room equipment according to the analysis result of the data analysis module.

[0012] As another optional embodiment of the present invention, optionally, the system further includes: The automated testing module is used to perform automated testing on the system, obtain test results, and send the test results to the display module.

[0013] As another optional embodiment of the present invention, optionally, the system further includes: Integration module for integrating different energy-saving computer room environment monitoring methods.

[0014] The beneficial effects of the present invention are: 1. The present invention collects and comprehensively analyzes various environmental data such as temperature, humidity, voltage and current in the energy-saving room in real time, uses three analysis methods to conduct in-depth analysis of the environmental data, and conducts comprehensive evaluation based on the three analysis results. It can quickly and accurately identify abnormal values ​​and output corresponding information (abnormal value information, data alarm information and data deviation information) based on different abnormal degrees. This instant feedback mechanism ensures that the room management personnel can respond quickly and deal with potential problems, thereby effectively avoiding equipment failures or energy waste caused by abnormal environmental parameters. In addition, the present invention reduces labor costs and improves the level of intelligence by establishing a model to automatically analyze environmental data.

[0015] 2. In the process of constructing a random decision tree, the present invention uses information gain as the criterion for feature selection, and combines nonlinear transformation to fuse feature subsets to generate more representative fusion features. This strategy not only improves the classification accuracy of the model, but also enhances the model's ability to handle complex data structures. At the same time, by recursively splitting nodes and selecting splitting points based on Gini impurity, the stability and effectiveness of the decision tree model during the construction process are ensured. In addition, by using a test set to evaluate and optimize the random decision tree, the generalization ability and adaptability of the model are further improved.

[0016] 3. The present invention sets the outlier standard by acquiring historical environmental data in real time and calculating its average value and standard deviation. At the same time, the update cycle of the average value and standard deviation, as well as the calculation of the seasonal adjustment factor, are set to achieve dynamic adjustment of the outlier standard. This mechanism ensures that the outlier standard can be automatically adjusted with changes in the environment and seasonal characteristics, thereby improving the accuracy and reliability of outlier detection.

[0017] 4. The system of the present invention integrates multiple functional modules such as data acquisition, transmission, analysis, alarm, storage and display, and realizes comprehensive monitoring and intelligent analysis of energy-saving room environmental data. Through the connection between the execution module and the energy-saving room equipment, the system can automatically adjust the equipment status according to the analysis results to achieve the purpose of energy saving and consumption reduction. In addition, the introduction of the automated test module and the integrated module further improves the automation and intelligence level of the system and reduces the need and cost of manual intervention.

[0018] 5. The implementation of the present invention not only helps to timely discover and solve computer room environmental problems, but also provides computer room managers with rich historical data and analysis results. These data and results can serve as an important basis for computer room management and maintenance, helping managers to understand the operating conditions of the computer room more carefully and formulate more scientific and reasonable maintenance plans and management strategies. At the same time, through system integration and automated processing, the efficiency and accuracy of computer room management are improved, and the possibility of human errors and negligence is reduced.

[0019] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 This is a flow chart of Example 1 of the present invention.

[0021] Figure 2 It is a schematic diagram of the structure of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0022] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0023] Example 1 like Figure 1As shown, a method for analyzing environmental data of an energy-saving computer room includes: S100, collecting environmental data of the energy-saving computer room, classifying the environmental data, and sending the environmental data to a cloud platform, wherein the environmental data includes temperature data, humidity data, voltage data, and current data; It should be noted that in step S100, the temperature data, humidity data, voltage data and current data are collected by corresponding high-precision sensors, and the collected data are transmitted to the cloud platform by wireless transmission; the environmental data is classified to facilitate subsequent data processing and analysis. The classification is based on the data type (such as temperature, humidity, voltage, current) and data source (such as different energy-saving computer rooms). After the classification is completed, various types of environmental data will be packaged and securely sent to the cloud platform to ensure the integrity and real-time nature of the data.

[0024] S200, the cloud platform sets a threshold, obtains an average value based on the environmental data, calculates the distance between the feature point and the average value by a Z-Score algorithm, and when the distance is greater than the threshold, determines that the feature point is an abnormal value; when the distance is less than or equal to the threshold, determines that the feature point is a normal value; It should be noted that in step S200, the cloud platform uses a pre-set threshold to perform anomaly detection. This step is a key step to ensure that the system can accurately identify and respond to abnormal environmental parameters in the energy-saving room. The application of the Z-Score algorithm calculates the distance between each feature point and the average value (i.e., the Z value) and compares it with the threshold to achieve rapid and effective identification of abnormal values. When the Z value of a feature point exceeds the set threshold, the system can immediately mark it as an abnormal value and trigger the corresponding processing mechanism.

[0025] S300, constructing at least two random decision trees, and using the random decision trees to isolate outliers; It should be noted that in order to further improve the accuracy and efficiency of anomaly detection in step S300, the present invention adopts a strategy of constructing at least two random decision trees. These random decision trees are trained by different feature subsets, aiming to capture abnormal patterns in environmental data from different angles and levels. Each random decision tree is an independent classifier, which can process the input environmental data in parallel and gradually narrow the search range of abnormal values ​​on their respective decision paths. Specifically, each random decision tree randomly selects a part of the features as the basis for splitting during the construction process, and divides the data set by recursively selecting the optimal splitting point. This randomness not only increases the diversity of the model, but also helps to avoid overfitting and improve the generalization ability of the model. At the same time, by integrating the prediction results of multiple random decision trees, the present embodiment can use voting or averaging to comprehensively judge the abnormal values ​​in the environmental data, thereby further improving the accuracy and stability of anomaly detection.

[0026] S400, respectively setting abnormal value standards for the temperature data, humidity data, voltage data and current data, and when the collected environmental data is not within the set abnormal value standards, determining that the data is an abnormal value; when the collected environmental data is within the set abnormal value standards, determining that the data is a normal value; It should be noted that this example adopts a flexible setting of abnormal value standards in step S400. Specifically, this step not only takes into account the type of environmental data (such as temperature, humidity, voltage, current), but also fully considers the real-time nature of the data, seasonal changes, and statistical characteristics of historical data. First, for temperature data and humidity data, this embodiment sets a reasonable temperature range and humidity range as the basic abnormal value standard according to the actual operation requirements of the computer room and the working conditions of the equipment. At the same time, the system will also monitor the changing trend of ambient temperature and humidity in real time, and dynamically adjust these standard values ​​in combination with the average value and standard deviation of historical data to adapt to fluctuations caused by seasonal changes or sudden environmental factors. For example, during the high temperature period in summer, the system will automatically increase the upper limit of the temperature abnormal value to reduce false alarms triggered by normal high temperature. For voltage data and current data, this embodiment focuses on ensuring the stability and safety of equipment operation. The system will set the corresponding safety threshold according to the rated voltage and rated current of each device in the computer room.

[0027] S500, based on the three analysis results of step S200, step S300 and step S400, if the three analysis results of the environmental data are all abnormal values, output abnormal value information; if two analysis results of the environmental data are abnormal values, output data warning information; if one analysis result of the environmental data is an abnormal value, output data deviation information; It should be noted that in step S500, the same environmental data needs to be analyzed by three analysis methods of step S200, step S300 and step S400 to obtain three analysis results, and then reanalyzed according to the three analysis results to determine the final abnormality handling strategy. This multi-level verification mechanism is intended to improve the accuracy of the system's judgment of abnormal conditions of energy-saving room environmental data. Only when the environmental data is an abnormal value in the above three analysis methods, it means that the environmental data is abnormal environmental data, and the abnormal value information is output to remind the administrator to pay attention immediately and take corresponding measures. If the environmental data is judged as an abnormal value in two analysis methods, it means that there may be potential problems in the data, but it has not reached the level of emergency treatment. At this time, data warning information is output so that the administrator can monitor in time and prepare for further investigation or adjustment. When only one analysis method considers that the environmental data is abnormal, it may be due to deviations caused by specific conditions or algorithm characteristics. At this time, data deviation information is output for the administrator to refer to and decide whether to conduct in-depth inspections or adjust the abnormal value standards. If the environmental data represents normal values ​​in all three analysis methods, the environmental data is normal.

[0028] S600: Send the output abnormal value information, data warning information and data deviation information to the computer room management personnel.

[0029] It should be noted that in step S600, the timely transmission and communication of information is the key to ensuring the stable operation of the energy-saving computer room. Once the system determines the abnormal state of environmental data through a multi-level analysis mechanism (three analysis methods), whether it is abnormal value information, data alarm information or data deviation information, it needs to be sent to the computer room manager quickly and accurately. To this end, the present invention integrates a variety of communication methods, including but not limited to e-mail, SMS push, instant messaging platform, etc., to ensure the immediate delivery of information. At the same time, the system also provides detailed reporting and logging functions to help managers quickly locate the source of the problem and trace historical data to analyze potential trends and patterns.

[0030] As an optional embodiment of the present invention, optionally, in step S300, constructing a random decision tree includes: S301, dividing the environmental data into a training set and a test set; It should be noted that in step S301, the environmental data is historical environmental data. When the historical environmental data is split into a training set and a test set, it is split randomly to ensure that the historical environmental data in the training set and the test set are different. The data ratio of the training set and the test set is 8:2.

[0031] S302, using information gain as a criterion for feature selection in the training set to obtain a feature subset; It should be noted that in step S302, information gain measures the degree of purity improvement that a feature can bring when dividing a data set. By calculating the information gain value of each feature, this embodiment can select the feature that contributes most to the current data set division, thereby building a more efficient and accurate decision tree. In step S302, the system traverses all features in the training set and calculates the information gain of each feature. Subsequently, the features are sorted according to the size of the information gain value, and the top N features with higher information gain values ​​are selected as feature subsets for subsequent decision tree construction. Such a feature selection strategy helps to reduce the complexity of the decision tree, improve the operating efficiency of the model, and retain the feature information that is most critical to anomaly detection.

[0032] S303, using nonlinear transformation to fuse the feature subsets to generate fused features, using the fused features to construct several random decision trees, and using randomly selected features to perform node splitting to generate candidate splitting points: It should be noted that in step S303, in order to further improve the adaptability and detection ability of the random decision tree model to complex environmental data, this embodiment introduces nonlinear transformation technology to fuse feature subsets. This nonlinear transformation includes but is not limited to polynomial expansion, kernel methods (such as RBF kernels), neural network layers, etc., which can effectively capture the nonlinear relationship between features, thereby generating richer and more expressive fusion features. The generated fusion features not only retain the important information of the original features, but also incorporate new nonlinear combination features, so that the model can learn more complex patterns and structures in the data. Subsequently, several random decision trees are constructed using these fusion features, and each decision tree randomly selects part of the fusion features for node splitting during the construction process to generate candidate splitting points. This randomness not only helps to reduce the risk of overfitting of the model, but also improves the generalization ability of the model.

[0033] S304, calculating the Gini impurity of each candidate split point, selecting the split point with the smallest Gini impurity as the split node, and recursively performing node splitting until the set stop condition is met, thereby completing the construction of the random decision tree; It should be noted that in step S304, during the node splitting process, each candidate split point will be evaluated according to a certain standard (Gini impurity), and the optimal split point will be selected for splitting until the stopping condition (such as the maximum depth of the tree, the minimum number of samples in the node, etc.) is met. In this way, each random decision tree can independently learn part of the rules in the data, and integrate the prediction results of each tree through ensemble learning methods (such as voting, averaging, etc.), so as to obtain the final anomaly detection result.

[0034] S305, evaluating the random decision tree using the test set; It should be noted that in step S305, this embodiment uses a test set to perform a performance evaluation on the constructed random decision tree model to ensure its generalization ability and accuracy on unknown data. The evaluation method is classification evaluation indicators such as recall rate, precision rate, F1 score, etc. These indicators can fully reflect the performance of the model in identifying abnormal data. By comparing the real labels on the test set with the model prediction results, the specific values ​​of the above evaluation indicators can be calculated to evaluate the effectiveness of the random decision tree model.

[0035] S306: Based on the evaluation results, iteratively optimize the random decision tree.

[0036] It should be noted that if the evaluation result of the random decision tree does not meet the expected standard in step S306, the random decision tree needs to be optimized. The optimization method may include adjusting the parameters of the random decision tree (such as the maximum depth of the tree, the number of features considered when splitting, etc.), reselecting feature subsets, or adjusting the strategy of nonlinear transformation, etc. Repeat steps S301 to S306 until the preset number of iterations is reached or the model performance is no longer significantly improved. Through iterative optimization, the optimal random decision tree model can be gradually approached. Each iteration may bring a slight improvement in model performance until a relatively stable state is reached. The selection of the number of iterations requires a balance between model performance and computing resources to avoid overfitting or excessive computing costs.

[0037] As an optional embodiment of the present invention, optionally, in step S400, setting the abnormal value standard includes: S401, real-time acquisition of historical temperature data, historical humidity data, historical voltage data and historical current data; It should be noted that in step S401, after obtaining the historical temperature data, historical humidity data, historical voltage data and historical current data, the data needs to be preprocessed. The preprocessing includes steps such as data cleaning, missing value processing, outlier detection and data normalization to ensure the quality and consistency of the data.

[0038] S402, respectively calculating the average value and standard deviation of the historical temperature data, the historical humidity data, the historical voltage data, and the historical current data; It should be noted that in step S402, the mean and standard deviation are calculated to evaluate the typical fluctuation range and central trend of these historical environmental parameters. The mean provides the center position of the data set, while the standard deviation reflects the degree of dispersion of the data points relative to the mean. These two statistics are crucial for setting reasonable outlier standards.

[0039] S403, setting an upper limit and a lower limit of abnormal values ​​for each data based on the mean value and the standard deviation; It should be noted that, in step S403, the upper and lower bounds of the abnormal value are set in order to clearly define which data points can be considered as abnormal data.

[0040] S404, setting an update period for the mean value and the standard deviation, and updating the upper limit and the lower limit of the abnormal value in real time according to the updated mean value and the standard deviation; It should be noted that the update cycle is set in step S404 to ensure that the outlier standard can be dynamically adjusted as the environmental parameters change, so as to maintain the timeliness and accuracy of the model. Since environmental parameters (such as temperature, humidity, voltage, current, etc.) may change due to various factors such as seasons, weather, and equipment aging, it is very necessary to regularly update these statistics and their corresponding outlier standards. The update cycle can be set according to actual conditions, such as daily, weekly, or monthly, etc., depending on the frequency of changes in environmental parameters and the sensitivity requirements of the model.

[0041] S405: Calculate a seasonal adjustment factor, and perform seasonal adjustment on the abnormal value upper bound and the abnormal value lower bound according to the real-time seasonal adjustment factor.

[0042] It should be noted that the introduction of the seasonal adjustment factor in step S405 is to further refine the setting of the outlier standard, taking into account that environmental parameters are often affected by seasonal factors, such as seasonal fluctuations in temperature, seasonal changes in humidity, etc. These seasonal changes may cause the distribution characteristics of environmental parameters to be different in different time periods, thereby affecting the applicability of the outlier standard. Therefore, by calculating the seasonal adjustment factor and seasonally adjusting the upper and lower bounds of the outlier accordingly, the outlier standard can be made more in line with the actual situation, and the accuracy and stability of the model in different seasons can be improved.

[0043] As an optional embodiment of the present invention, optionally, calculating the seasonal adjustment factor in step S405 includes: S4051, performing seasonal trend analysis on the historical temperature data, historical humidity data, historical voltage data, and historical current data to obtain temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics; It should be noted that in step S4051, by performing time series analysis on historical temperature data, historical humidity data, historical voltage data, and historical current data, their respective seasonal patterns can be identified. Seasonal components are extracted through statistical methods (such as seasonal decomposition, Fourier transform, wavelet analysis, etc.), thereby obtaining temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics. These characteristics describe the periodic fluctuation law of environmental parameters within a year or a specific time period, providing a basis for subsequent seasonal adjustments.

[0044] S4052. Based on the temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics, respectively calculate the temperature seasonal adjustment factor, humidity seasonal adjustment factor, voltage seasonal adjustment factor, and current seasonal adjustment factor.

[0045] It should be noted that in step S4052, based on the extracted temperature seasonality characteristics, humidity seasonality characteristics, voltage seasonality characteristics, and current seasonality characteristics, this embodiment can use these characteristics to quantify the degree of influence of seasonal changes on corresponding environmental parameters, thereby calculating respective seasonal adjustment factors. These adjustment factors will serve as important parameters for dynamically adjusting the abnormal value standard to ensure that the model can maintain high accuracy and stability in different seasons.

[0046] As an optional embodiment of the present invention, optionally, the formula for calculating the upper bound of the outlier value and the lower bound of the outlier value is: Outlier upper bound: Outlier lower bound: in, Indicates the upper limit of outliers; represents the average value; Indicates the setting coefficient; represents standard deviation; represents the seasonal adjustment factor.

[0047] It should be noted that the above formula can be used to calculate the upper and lower bounds of the abnormal values ​​of temperature data, humidity data, voltage data, and current data, so as to obtain the upper and lower bounds of the abnormal values ​​of the corresponding data.

[0048] Example 2 An automatic monitoring system for environmental data of an energy-saving computer room, the system comprising an analysis method for environmental data of an energy-saving computer room; The system further comprises: Data collection module, used to collect environmental data of energy-saving computer rooms in real time; The data acquisition module includes a temperature sensor, a humidity sensor, a voltage sensor and a current sensor.

[0049] A data transmission module, connected to the data acquisition module, for sending the environmental data to a cloud platform; The data transmission module is a wireless communication module that supports communication protocols such as Wi-Fi, 4G / 5G or LoRa, ensuring real-time and stable transmission of collected environmental data (including temperature, humidity, voltage, current, etc.) to the cloud platform. The cloud platform has powerful data processing and storage capabilities, and can receive and store data from multiple data acquisition modules, providing a basis for subsequent data analysis and processing.

[0050] A data analysis module, which is arranged in the cloud platform and is used to execute the analysis method, analyze the environmental data, and generate abnormal value information, data warning information, and data deviation information; The data analysis module is used to implement an analysis method for energy-saving computer room environmental data described in Example 1. When the data analysis module receives the environmental data collected by the acquisition module, it analyzes the collected environmental data through the three analysis methods in the embodiment to obtain three analysis results, and then analyzes the three analysis results. Only when the environmental data is an abnormal value in the above three analysis methods, it means that the environmental data is abnormal environmental data, outputs abnormal value information, and sends the abnormal value information to the execution module. The execution module adjusts the corresponding device status according to the abnormal value information. If the environmental data is determined to be an abnormal value in two analysis methods, it indicates that there may be potential problems in the data, but it has not yet reached the level of emergency treatment. At this time, data warning information is output so that the administrator can monitor in time and prepare for further investigation or adjustment. When only one analysis method considers that the environmental data is abnormal, it may be due to deviations caused by specific conditions or algorithm characteristics. At this time, data deviation information is output for the administrator to refer to and decide whether to conduct in-depth inspections or adjust the abnormal value standards.

[0051] An alarm module, connected to the data analysis module, for generating corresponding sound, light and electricity information according to abnormal value information, data alarm information and data deviation information; The alarm module is an acoustic and optical alarm device, which is integrated into the monitoring interface of the cloud platform or the on-site control room of the energy-saving computer room. Once the data analysis module generates abnormal value information, data warning information or data deviation information, the alarm module will immediately trigger the corresponding acoustic and optical signal. Specifically, for abnormal value information, the alarm module will start a high-brightness red flash and a strong beep to attract the administrator's immediate attention and indicate the urgency of the abnormality. At the same time, on the monitoring interface, the abnormal data point will be displayed with a striking red mark, accompanied by a scrolling prompt information, clearly informing the type, location and possible impact range of the abnormal data. For data alarm information, the alarm module uses a combination of a yellow flash and a relatively gentle beep, which not only reminds the administrator to pay attention, but also avoids causing excessive tension. On the monitoring interface, the alarm data will be highlighted in yellow, with a detailed alarm description and possible processing suggestions, to help the administrator quickly judge and prepare countermeasures. For data deviation information, the alarm module may only remind the administrator through text prompts or slight color changes on the monitoring interface, without triggering an acoustic and optical alarm, so as not to interfere with normal monitoring work. This deviation information will be used as part of daily monitoring to help administrators continuously optimize outlier criteria and monitor system performance.

[0052] A data storage module, connected to the data analysis module and the data transmission module, for storing historical environmental data, analysis results and alarm records; The data storage module is a high-performance database deployed in the core server cluster of the cloud platform to ensure the security, integrity and accessibility of the data. The module adopts a distributed storage architecture, which can efficiently handle the storage needs of massive environmental data and support fast data retrieval and query operations. During the data storage process, the system will classify and store the data in different data tables according to the data type and analysis results, so as to facilitate subsequent rapid call and analysis. Specifically, the data storage module not only stores the real-time collected environmental data (such as temperature, humidity, voltage, current, etc.), but also saves various analysis results processed by the data analysis module, including abnormal value information, data alarm information and data deviation information. In addition, the module also records the detailed information of all alarm events, including the time of occurrence, abnormal type, impact range, processing results, etc., providing valuable data support for subsequent troubleshooting and performance optimization. In order to further improve the security and reliability of data, the data storage module also implements a data backup and recovery mechanism. The system regularly backs up key data to a remote data center to prevent local data loss or damage. At the same time, the system also provides a data recovery function, which can quickly restore data to the most recent backup state when data is lost or the system crashes, ensuring the continuity and integrity of the data.

[0053] The display module is connected with the data storage module, the alarm module and the data analysis module.

[0054] The display module is a visual display platform integrated on the cloud platform monitoring interface, which can present the environmental data, analysis results and alarm status of the energy-saving computer room in real time. It uses advanced graphics and chart technology to transform complex data information into intuitive and easy-to-understand visual elements, allowing administrators to quickly grasp the overall status of the computer room environment and discover and respond to potential problems in a timely manner.

[0055] As another optional embodiment of the present invention, optionally, the system further includes: The execution module is connected to the data analysis module and the energy-saving computer room equipment, and is used to adjust the state of the energy-saving computer room equipment according to the analysis result of the data analysis module.

[0056] The execution module is an intelligent controller that directly communicates and interacts with various devices in the energy-saving computer room (such as air conditioning systems, ventilation equipment, power management systems, etc.). When the data analysis module detects abnormal environmental data and feeds back abnormal value information, data alarm information, or data deviation information to the execution module, the execution module will immediately start the corresponding preset strategy to accurately control the equipment in the computer room to maintain the stability and efficiency of the computer room environment. Specifically, the execution module has a complete set of control logic and algorithms built in, which can automatically calculate the optimal adjustment plan based on different types of abnormal information. For example, in the case of abnormal temperature rise, the execution module will first try to lower the temperature of the computer room by adjusting the cooling mode of the air conditioning system; if the effect is not good, it will further adjust the wind speed and direction of the ventilation equipment to enhance the air circulation in the computer room.

[0057] As another optional embodiment of the present invention, optionally, the system further includes: The automated testing module is used to perform automated testing on the system, obtain test results, and send the test results to the display module.

[0058] The automated test module regularly conducts comprehensive and efficient tests on each module of the system through preset test cases and automated scripts. This module not only covers multiple dimensions such as functional testing, performance testing, and security testing, but also has the ability to simulate faults. The automated test module is closely integrated with other modules of the system (such as data analysis module, alarm module, data storage module, display module, and execution module). Through the mature automated test framework developed by unittest, an automated test system for monitoring and analyzing data in the computer room environment is constructed to simulate data input and operation instructions in a real environment, automatically trigger and monitor the response behavior of the system. During the test, the automated test module will record the execution of each test case in detail, including key information such as test time, test steps, expected results and actual results, so as to facilitate subsequent problem location and repair. Once the test is completed, the automated test module will generate a detailed test report and send the test results to the display module for visual display. On the display module, the administrator can intuitively see an overview of the test results, the passing status of specific test cases, the reasons for failure, and the recommended repair solutions. This design not only improves the transparency and traceability of the test work, but also effectively reduces the cost and error rate of manual intervention.

[0059] As another optional embodiment of the present invention, optionally, the system further includes: Integration module for integrating different energy-saving computer room environment monitoring methods.

[0060] The integration module is a flexible and extensible framework that quickly integrates new monitoring algorithms and analysis models into the system through the CI (continuous integration) process. CT (continuous testing) is performed during the integration process to ensure the stability and reliability of new functions. The updated system is pushed to users in real time through the CD (continuous deployment) process, improving the system update speed and user experience.

[0061] Example 3 A computer-readable storage medium, comprising: a memory having a computer program stored thereon; The processor is used to execute the program in the memory to implement the analysis method for energy-saving computer room environmental data in Example 1.

[0062] It should be noted that the electronic device according to the embodiment of the present disclosure includes a processor and a memory for storing processor executable instructions, wherein the processor is configured to implement any of the above-mentioned cloud resource task scheduling methods with efficient optimization when executing the executable instructions.

[0063] Here, it should be noted that the number of processors can be one or more. At the same time, the electronic device in the embodiment of the present disclosure may also include an input device and an output device. Among them, the processor, memory, input device and output device may be connected through a bus or in other ways, which are not specifically limited here.

[0064] As a computer-readable storage medium, the memory can be used to store software programs, computer executable programs and various modules, such as: the program or module corresponding to the cloud resource task scheduling method with efficient optimization in the embodiment of the present disclosure. The processor executes various functional applications and data processing of the electronic device by running the software programs or modules stored in the memory.

[0065] The input device can be used to receive input numbers or signals. The signal can be a key signal related to user settings and function control of the device / terminal / server. The output device can include a display device such as a display screen.

[0066] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for analyzing environmental data of energy-saving computer rooms, characterized in that: The method comprises: S100, collecting environmental data of the energy-saving computer room, classifying the environmental data, and sending the environmental data to a cloud platform, wherein the environmental data includes temperature data, humidity data, voltage data, and current data; S200, the cloud platform sets a threshold, obtains an average value based on the environmental data, calculates the distance between the feature point and the average value by a Z-Score algorithm, and when the distance is greater than the threshold, determines that the feature point is an abnormal value; when the distance is less than or equal to the threshold, determines that the feature point is a normal value; S300, constructing at least two random decision trees, and using the random decision trees to isolate outliers; S400, respectively setting abnormal value standards for the temperature data, humidity data, voltage data and current data, and when the collected environmental data is not within the set abnormal value standards, determining that the data is an abnormal value; when the collected environmental data is within the set abnormal value standards, determining that the data is a normal value; S500, based on the three analysis results of step S200, step S300 and step S400, if the three analysis results of the environmental data are all abnormal values, output abnormal value information; if two analysis results of the environmental data are abnormal values, output data warning information; if one analysis result of the environmental data is an abnormal value, output data deviation information; S600: Send the output abnormal value information, data warning information and data deviation information to the computer room management personnel.

2. The method for analyzing environmental data of an energy-saving computer room according to claim 1, characterized in that: In step S300, constructing a random decision tree includes: S301, dividing the environmental data into a training set and a test set; S302, using information gain as a criterion for feature selection in the training set to obtain a feature subset; S303, using nonlinear transformation to fuse the feature subsets to generate fused features, using the fused features to construct several random decision trees, and using randomly selected features to perform node splitting to generate candidate splitting points: S304, calculating the Gini impurity of each candidate split point, selecting the split point with the smallest Gini impurity as the split node, and recursively performing node splitting until the set stop condition is met, thereby completing the construction of the random decision tree; S305, evaluating the random decision tree using the test set; S306: Based on the evaluation results, iteratively optimize the random decision tree.

3. The method for analyzing environmental data of an energy-saving computer room according to claim 1, characterized in that: In step S400, setting the abnormal value standard includes: S401, real-time acquisition of historical temperature data, historical humidity data, historical voltage data and historical current data; S402, respectively calculating the average value and standard deviation of the historical temperature data, the historical humidity data, the historical voltage data, and the historical current data; S403, setting an upper limit and a lower limit of abnormal values ​​for each data based on the mean value and the standard deviation; S404, setting an update period for the mean value and the standard deviation, and updating the upper limit and the lower limit of the abnormal value in real time according to the updated mean value and the standard deviation; S405: Calculate a seasonal adjustment factor, and perform seasonal adjustment on the abnormal value upper bound and the abnormal value lower bound according to the real-time seasonal adjustment factor.

4. The method for analyzing environmental data of an energy-saving computer room according to claim 3, characterized in that: Calculating the seasonal adjustment factor in step S405 includes: S4051, performing seasonal trend analysis on the historical temperature data, historical humidity data, historical voltage data, and historical current data to obtain temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics; S4052. Based on the temperature seasonal characteristics, humidity seasonal characteristics, voltage seasonal characteristics, and current seasonal characteristics, respectively calculate the temperature seasonal adjustment factor, humidity seasonal adjustment factor, voltage seasonal adjustment factor, and current seasonal adjustment factor.

5. The method for analyzing environmental data of an energy-saving computer room according to claim 4, characterized in that: The formulas for calculating the upper and lower outlier bounds are: Outlier upper bound: Outlier lower bound: in, represents the upper limit of outliers, represents the average value, represents the setting coefficient, represents the standard deviation, represents the seasonal adjustment factor.

6. An automatic monitoring system for environmental data of an energy-saving computer room, the system comprising an analysis method for environmental data of an energy-saving computer room according to any one of claims 1 to 5; The system further comprises: Data collection module, used to collect environmental data of energy-saving computer rooms in real time; A data transmission module, connected to the data acquisition module, for sending the environmental data to a cloud platform; A data analysis module, arranged in the cloud platform, for executing the analysis method according to claim 1, for analyzing the environmental data, and generating abnormal value information, data warning information and data deviation information; An alarm module, connected to the data analysis module, for generating corresponding sound, light and electricity information according to abnormal value information, data alarm information and data deviation information; A data storage module, connected to the data analysis module and the data transmission module, for storing historical environmental data, analysis results and alarm records; The display module is connected with the data storage module, the alarm module and the data analysis module.

7. An automatic monitoring system for energy-saving computer room environment data according to claim 6, characterized in that: The system further comprises: The execution module is connected to the data analysis module and the energy-saving computer room equipment, and is used to adjust the state of the energy-saving computer room equipment according to the analysis result of the data analysis module.

8. The automatic monitoring system for energy-saving computer room environment data according to claim 6, characterized in that: The system further comprises: The automated testing module is used to perform automated testing on the system, obtain test results, and send the test results to the display module.

9. The automatic monitoring system for energy-saving computer room environment data according to claim 6, characterized in that: The system further comprises: Integration module for integrating different energy-saving computer room environment monitoring methods.

10. A computer-readable storage medium, characterized in that: include: a memory having a computer program stored thereon; A processor is used to execute the program in the memory to implement the method for analyzing environmental data of an energy-saving computer room according to any one of claims 1 to 5.

Images