System and method for monitoring and analyzing energy consumption of 5G communication base station

Abstract

The invention relates to the field of energy consumption monitoring, and particularly discloses an energy consumption monitoring analysis system and method for a 5G communication base station, and the method comprises the steps: firstly obtaining historical 5G communication base station energy consumption data, 5G communication base station current energy consumption data and 5G communication base station environment data, and then carrying out the feature extraction and correlation analysis of the three data through a deep learning technology, and finally, through a classifier, whether a 5G communication base station energy consumption abnormity early warning is sent out is judged, so that a potential energy consumption abnormity mode is identified more accurately, the accuracy of an early warning system is improved, quick response is made, necessary measures are taken, and the processing efficiency and decision-making capability of the system are improved.

Description

Technical Field

[0001] The present application relates to the field of energy consumption monitoring, and more specifically, to an energy consumption monitoring and analysis system and method for a 5G communication base station. Background Art

[0002] 5G base stations are wireless communication devices that support fifth-generation mobile communication technology (5G). They are responsible for providing high-speed data transmission, low-latency communications, and wide coverage within the network. They utilize high-frequency spectrum bands and advanced antenna technology to achieve higher network performance and greater user capacity than 4G. They are primarily used to support connections for smartphones, IoT devices, and other wireless communication applications. 5G base stations connect to the core network to send and receive data, ensuring that user communication needs are met in diverse locations and environments.

[0003] Considering that mobile communication access utilizes tens of thousands of base stations, whose energy consumption is primarily electricity, the increasing cost of electricity and the expansion of mobile networks have led to a gradual increase in electricity costs for base station equipment rooms. Existing base station energy consumption monitoring and analysis systems can only monitor, count, and store energy consumption at each base station. However, since base station energy consumption is affected by a variety of external factors, it is impossible to directly determine whether base station energy consumption exceeds standards based on monitored energy consumption.

[0004] Therefore, a system and method for monitoring and analyzing energy consumption of 5G communication base stations is desired. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a 5G communication base station energy consumption monitoring and analysis system and method, which first obtains historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data and 5G communication base station environmental data, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to determine whether to issue a 5G communication base station energy consumption abnormality warning, thereby more accurately identifying potential energy consumption abnormality patterns, improving the accuracy of the warning system, and then quickly responding and taking necessary measures to improve system processing efficiency and decision-making capabilities.

[0006] According to one aspect of the present application, a 5G communication base station energy consumption monitoring and analysis system is provided, which includes:

[0007] 5G communication base station energy consumption data acquisition module, used to obtain historical 5G communication base station energy consumption data, 5G communication base station current energy consumption data and 5G communication base station environmental data;

[0008] A 5G communication base station energy consumption data extraction module is used to extract a communication base station energy consumption collaborative correlation feature vector and a 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current energy consumption data of the 5G communication base station and the 5G communication base station environment data;

[0009] The energy consumption monitoring data analysis module is used to determine whether to issue an abnormal energy consumption warning for the 5G communication base station based on the collaborative correlation feature vector of the communication base station energy consumption and the semantic feature vector of the 5G communication base station environment.

[0010] According to another aspect of the present application, a method for monitoring and analyzing energy consumption of a 5G communication base station is provided, which includes:

[0011] Obtain historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data, and 5G communication base station environmental data;

[0012] Extracting a communication base station energy consumption collaborative correlation feature vector and a 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current 5G communication base station energy consumption data, and the 5G communication base station environment data;

[0013] Based on the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector, it is determined whether to issue a 5G communication base station energy consumption abnormality warning.

[0014] Compared with the existing technology, the present application provides a 5G communication base station energy consumption monitoring and analysis system and method, which first obtains historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data and 5G communication base station environmental data, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a classifier is used to determine whether to issue an abnormal energy consumption warning for the 5G communication base station, thereby more accurately identifying potential energy consumption abnormality patterns, improving the accuracy of the early warning system, and then quickly responding and taking necessary measures to improve system processing efficiency and decision-making capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 This is a block diagram of an energy consumption monitoring and analysis system for a 5G communication base station according to an embodiment of the present application.

[0017] Figure 2This is a block diagram of a 5G communication base station energy consumption data extraction module in the energy consumption monitoring and analysis system of a 5G communication base station according to an embodiment of the present application.

[0018] Figure 3 This is a block diagram of a communication base station energy consumption collaborative feature extraction unit in the energy consumption monitoring and analysis system of a 5G communication base station according to an embodiment of the present application.

[0019] Figure 4 This is a block diagram of the energy consumption monitoring data analysis module in the energy consumption monitoring and analysis system of the 5G communication base station according to an embodiment of the present application.

[0020] Figure 5 This is a flowchart of the energy consumption monitoring and analysis method of a 5G communication base station according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0022] Figure 1 This is a block diagram of the energy consumption monitoring and analysis system of the 5G communication base station in the embodiment of the present application. Figure 1 As shown, the energy consumption monitoring and analysis system 100 of the 5G communication base station according to the embodiment of the present application includes: a 5G communication base station energy consumption data acquisition module 110, which is used to obtain historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data and 5G communication base station environmental data; a 5G communication base station energy consumption data extraction module 120, which is used to extract the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the 5G communication base station current energy consumption data and the 5G communication base station environment data; an energy consumption monitoring data analysis module 130, which is used to determine whether to issue a 5G communication base station energy consumption abnormality warning based on the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector.

[0023] In the aforementioned 5G communication base station energy consumption monitoring and analysis system 100, the 5G communication base station energy consumption data acquisition module 110 is used to acquire historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data, and 5G communication base station environmental data. It should be understood that 5G communication base stations are the core equipment supporting fifth-generation mobile communication technology (5G). They provide fast data transmission, low-latency communication services, and wide coverage within the network. By utilizing high-frequency spectrum bands and advanced antenna technology, 5G base stations not only achieve higher network performance than 4G, but also support more user connections, including smartphones, IoT devices, and other wireless communication applications. 5G base stations, through their connection to the core network, are responsible for data transmission and reception, ensuring that user communication needs are met in various geographic locations and environments. Because modern mobile communication networks involve a large number of base stations, the energy consumption of these base stations primarily comes from electricity. With rising electricity costs and the expansion of network scale, electricity expenses for base station equipment rooms are also increasing. Current communication base station energy consumption monitoring systems can only monitor, count, and store energy consumption. However, since the energy consumption of base stations is affected by a variety of external factors, monitoring of energy consumption data alone cannot directly determine whether the base station has exceeded energy consumption standards. Therefore, in the technical solution of this application, by obtaining historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data, and 5G communication base station environmental data, and combining deep learning technology, in order to determine whether it is necessary to issue an abnormal energy consumption warning for a 5G communication base station, potential abnormal energy consumption patterns can be identified more accurately, thereby improving the accuracy of the early warning system. This enables rapid response and the adoption of necessary measures, thereby improving the processing efficiency and decision-making capabilities of the system.

[0024] Specifically, historical energy consumption data provides information about base station energy usage over different time periods and operating conditions, helping to identify energy consumption patterns and trends. By analyzing this data, operators can identify periods of high energy consumption or anomalies and take corrective action. Current energy consumption data reflects base station energy consumption under real-time conditions, helping operators monitor and manage energy consumption in a timely manner to ensure efficient and stable base station operation. Environmental data, such as temperature and humidity, directly impacts base station energy consumption. For example, high temperatures increase cooling requirements, which in turn increases energy consumption. By combining environmental data with energy consumption data, the impact of external factors on energy consumption can be more accurately assessed, allowing optimization of cooling systems and other related equipment to reduce unnecessary energy consumption. Furthermore, this data integration helps predict future energy consumption needs, optimize resource allocation, and avoid energy waste. Overall, comprehensive access to and analysis of this data not only improves network operational efficiency and reduces energy costs, but also supports environmental sustainability goals by implementing energy-saving measures and using renewable energy to reduce carbon footprints. Through precise data-driven management, operators can achieve more efficient energy utilization, improving network cost-effectiveness and reliability.

[0025] In the above-mentioned 5G communication base station energy consumption monitoring and analysis system 100, the 5G communication base station energy consumption data extraction module 120 is used to extract the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current energy consumption data of the 5G communication base station and the 5G communication base station environment data. It should be understood that combining these two feature vectors can form a comprehensive energy consumption management framework. By analyzing the energy consumption collaborative correlation features, energy consumption prediction and optimization can be achieved, while through the environmental semantic features, actual environmental conditions can be adapted and adjusted. This data-driven approach not only improves the energy efficiency of the base station, but also helps operators reduce operating costs, reduce energy waste, and support environmental sustainability goals. Ultimately, through intelligent data analysis and management, operators can operate 5G networks more efficiently and improve network reliability and economy.

[0026] Figure 2 FIG. 5 is a block diagram of a 5G communication base station energy consumption data extraction module in an energy consumption monitoring and analysis system for a 5G communication base station according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the 5G communication base station energy consumption data extraction module 120 includes: a communication base station energy consumption collaborative feature extraction unit 121, used to perform feature extraction on the historical 5G communication base station energy consumption data and the current energy consumption data of the 5G communication base station to obtain the communication base station energy consumption collaborative correlation feature vector; a 5G communication base station environment data feature extraction unit 122, used to perform feature extraction on the 5G communication base station environment data to obtain the 5G communication base station environment semantic feature vector.

[0027] It should be understood that current energy consumption data reflects the energy consumption of the base station under real-time conditions. It provides energy consumption data for the base station's current operating state and can be used for real-time monitoring and management. By combining current data with historical data, it is possible to assess whether the base station's immediate energy consumption conforms to expected trends and promptly identify anomalies. For example, if current energy consumption suddenly increases, feature vector analysis can help quickly identify the cause and take necessary adjustments. By extracting the collaborative correlation feature vector of communication base station energy consumption, historical data and current data can be effectively combined to explore the correlation and influence between them. This feature vector can reveal the interaction between energy consumption and various factors, such as changes in network traffic, fluctuations in equipment load, and the impact of external environmental conditions. In this way, operators can understand how these factors collaboratively affect energy consumption and use these insights to optimize energy management strategies.

[0028] Furthermore, extracting features from 5G base station environmental data to generate 5G base station environmental semantic feature vectors is of great significance in many respects. Environmental data typically includes factors such as temperature, humidity, and wind speed, which directly impact base station performance and energy consumption. For example, higher temperatures increase base station cooling requirements, thereby increasing energy consumption. By extracting features from environmental data, these environmental variables can be converted into meaningful feature vectors, enabling a better understanding and management of environmental impacts on base stations. Environmental semantic feature vectors can help reveal how environmental conditions correlate with base station energy consumption and operating status. After extracting these feature vectors, a relationship model between environmental conditions and energy consumption can be established. For example, by analyzing the impact of temperature and humidity on the cooling system, cooling system configuration can be optimized to reduce unnecessary energy consumption. Furthermore, environmental semantic feature vectors can provide real-time and predictive environmental impact assessments, enabling proactive action when environmental conditions change to ensure stable base station operation.

[0029] Figure 3 FIG. 1 is a block diagram of a communication base station energy consumption collaborative feature extraction unit in an energy consumption monitoring and analysis system for a 5G communication base station according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the communication base station energy consumption collaborative feature extraction unit 121 includes: a historical 5G communication base station energy consumption semantic encoding subunit 1211, which is used to pass the historical 5G communication base station energy consumption data through a historical 5G communication base station energy consumption data semantic encoder based on a Taming model to obtain a historical 5G communication base station energy consumption feature vector; a 5G communication base station current energy consumption semantic encoding subunit 1212, which is used to pass the current energy consumption data of the 5G communication base station through a 5G communication base station current energy consumption data semantic encoder based on a Taming model to obtain a 5G communication base station current energy consumption feature vector; a communication base station energy consumption collaborative feature aggregation subunit 1213, which is used to perform feature aggregation on the historical 5G communication base station energy consumption feature vector and the 5G communication base station current energy consumption feature vector to obtain the communication base station energy consumption collaborative association feature vector.

[0030] As you can see, the Taming model is a powerful data processing and encoding tool that can transform complex historical energy consumption data into feature vectors with high-dimensional semantic information. The Taming model excels at extracting useful information from large-scale, high-dimensional data and converting this information into structured feature vectors. Using the Taming model's semantic encoder for historical 5G base station energy consumption data, the energy consumption data can be effectively reduced in dimensionality and encoded, preserving the most important features. This feature extraction helps identify underlying patterns and trends in the data, such as seasonal variations in energy consumption and fluctuations in energy consumption under different operating conditions. The historical 5G base station energy consumption feature vectors provide a refined data representation that better supports further data analysis and model training. Feature vectors condense the complex information in historical data into a concise format, making data analysis, forecasting, and optimization more efficient. Operators can use these feature vectors to forecast energy consumption, identify anomalies, and develop targeted optimization strategies. The Taming model-based encoder can also address noise and inconsistencies in the data, improving data quality. Converting historical energy consumption data into feature vectors reduces data complexity and redundancy, improving model stability and accuracy. This is crucial for achieving long-term energy efficiency management and optimization. Specifically, the historical 5G communication base station energy consumption data is segmented to obtain a historical 5G communication base station energy consumption word sequence; the embedding layer of the historical 5G communication base station energy consumption data semantic encoder based on the Taming model is used to map each historical 5G communication base station energy consumption word in the historical 5G communication base station energy consumption word sequence to a word embedding vector to obtain a sequence of historical 5G communication base station energy consumption word embedding vectors; the converter-based Bert model of the historical 5G communication base station energy consumption data semantic encoder based on the Taming model is used to perform global context semantic encoding on the sequence of historical 5G communication base station energy consumption word embedding vectors to obtain multiple historical energy consumption feature vectors; and, the multiple feature vectors are cascaded to obtain the historical 5G communication base station energy consumption feature vector.

[0031] Furthermore, 5G base station current energy consumption data is processed through a Taming-based semantic encoder to generate 5G base station current energy consumption feature vectors. This approach aims to provide precise insights and effective control for real-time management and optimization of base station energy consumption. Current energy consumption data provides a base station's energy consumption at a specific moment, providing both real-time and timely information. The primary purpose of processing current energy consumption data is to monitor and dynamically adjust energy consumption strategies in real time, ensuring that base stations maintain efficient operation under changing operating conditions. By inputting current energy consumption data into the Taming-based encoder, this complex and high-dimensional data is converted into structured and easily analyzable feature vectors. The Taming model is an advanced data processing technique that effectively encodes and reduces the dimensionality of complex data. The Taming-based semantic encoder for 5G base station current energy consumption data extracts important features and patterns from real-time energy consumption data. These feature vectors contain key information about energy consumption patterns, device load changes, network traffic fluctuations, and other key information. This approach aims to improve data availability and analysis efficiency, enabling operators to quickly identify and respond to current energy consumption situations. By generating the current energy consumption feature vector of the 5G communication base station, the real-time status of the current energy consumption can be reflected, supporting real-time monitoring and early warning systems. Operators can detect abnormal energy consumption based on these feature vectors, take timely adjustment measures, and prevent potential problems from escalating. Specifically, the embedding layer of the 5G communication base station current energy consumption data semantic encoder based on the Taming model is used to convert the current energy consumption data of the 5G communication base station into an embedding vector to obtain a sequence of current energy consumption embedding vectors; the converter-based Bert model of the 5G communication base station current energy consumption data semantic encoder based on the Taming model is used to perform global context semantic encoding on the sequence of current energy consumption embedding vectors to obtain multiple current energy consumption feature vectors; and, the multiple current energy consumption feature vectors are cascaded to obtain the current energy consumption feature vector of the 5G communication base station.

[0032] Furthermore, historical energy consumption feature vectors contain information on long-term energy consumption trends and cyclical changes. This data helps reveal base station energy consumption patterns over different time periods, seasonal fluctuations, and their relationship to network traffic. This historical perspective provides a stable baseline, enabling operators to understand normal energy consumption patterns and ranges of variation. Current energy consumption feature vectors reflect the energy consumption status of base stations under immediate operating conditions. This real-time data helps monitor current energy consumption and identify and respond to immediate energy consumption anomalies. The immediacy of current data enables operators to adjust strategies promptly to optimize base station operating efficiency. By combining historical and current feature vectors, a comprehensive understanding of base station energy consumption can be achieved. Historical data provides long-term trends and patterns, while current data reflects immediate conditions. Combining these two types of information helps identify the relationship between long-term trends and short-term anomalies, providing more accurate energy consumption analysis.

[0033] In a specific embodiment of the present application, the communication base station energy consumption collaborative feature aggregation subunit 1213 includes: associating the historical 5G communication base station energy consumption feature vector and the current energy consumption feature vector of the 5G communication base station to obtain a communication base station energy consumption collaborative correlation feature matrix; and passing the communication base station energy consumption collaborative correlation feature matrix through an energy consumption data association encoder with mutually transposed convolution kernels to obtain a communication base station energy consumption collaborative correlation feature vector.

[0034] It should be understood that the historical 5G communication base station energy consumption feature vectors contain information on long-term energy consumption trends, patterns, and periodic changes. These features reflect the energy consumption behavior and changes of the base station over different time periods. The current energy consumption feature vector provides real-time energy consumption data, reflecting the actual energy consumption status of the base station under current conditions. These two types of features describe the energy consumption of the base station from two dimensions: long-term and short-term. By correlating the historical and current energy consumption feature vectors to generate a collaborative correlation feature matrix for communication base station energy consumption, it is possible to reveal the relationship between base station energy consumption at different time scales. For example, by combining historical trends with current status, it is possible to identify whether short-term anomalies are consistent with long-term trends, helping to understand whether current energy consumption is within the normal range.

[0035] Furthermore, the base station energy consumption collaborative correlation feature matrix is ​​processed through an energy consumption data association encoder with transposed convolution kernels to obtain a base station energy consumption collaborative correlation feature vector. This approach aims to deeply explore potential relationships in the data and improve the efficiency and accuracy of energy consumption data analysis. The base station energy consumption collaborative correlation feature matrix is ​​generated by correlating historical and current energy consumption feature vectors, providing a comprehensive perspective on base station energy consumption, including information on both long-term trends and short-term dynamics. To further refine this data and extract deeper features, an energy consumption data association encoder with transposed convolution kernels is employed. This encoder with transposed convolution kernels is an advanced neural network technique that efficiently processes and transforms feature matrices. Specifically, the convolution kernels, through their local receptive fields and weight sharing mechanisms, effectively extract important patterns and structural information from the matrix. With transposed convolution kernels, the encoder processes the feature matrix symmetrically, capturing the symmetries and relationships within the feature matrix. Through convolution operations, the encoder maps a high-dimensional feature matrix into a low-dimensional feature space. This process helps to remove redundant information, retain key features, and generate a more refined and abstract communication base station energy consumption collaborative correlation feature vector. Specifically, each layer of the first convolutional neural network model of the energy consumption data association encoder with mutually transposed convolution kernels performs the following on the input data in the forward pass of the layer: performing convolution processing, pooling processing and nonlinear processing based on the first two-dimensional convolution kernel on the input data to obtain a first activation feature map; and performing convolution processing, pooling processing and nonlinear activation processing based on the second two-dimensional convolution kernel on the first activation feature map to obtain a second activation feature map, wherein the first two-dimensional convolution kernel and the second two-dimensional convolution kernel are transposed to each other; wherein the input of the first layer of the first convolutional neural network model of the energy consumption data association encoder with mutually transposed convolution kernels is the communication base station energy consumption collaborative correlation feature matrix, and the output of the last layer of the first convolutional neural network model of the energy consumption data association encoder with mutually transposed convolution kernels is the communication base station energy consumption collaborative correlation feature vector.

[0036] In a specific embodiment of the present application, the historical 5G communication base station energy consumption characteristic vector and the current energy consumption characteristic vector of the 5G communication base station are associated to obtain a communication base station energy consumption collaborative correlation characteristic matrix, including: configuring the operating environment required by the Velocity mechanism; loading the historical 5G communication base station energy consumption characteristic vector historical_energy_vector and the current 5G communication base station energy consumption characteristic vector current_energy_vector; using the matrix multiplication function matrix_multiply provided by the Velocity mechanism to calculate the correlation matrix of historical_energy_vector and current_energy_vector, and storing the calculation result as the communication base station energy consumption collaborative correlation characteristic matrix energy_correlation_matrix; storing the calculated energy_correlation_matrix to the file energy_correlation_matrix.npy, and outputting the result to the console or log file; running the deployment code to verify the calculation result.

[0037] In this way, the historical and current energy consumption data of 5G communication base stations can be efficiently processed and analyzed to obtain the key energy consumption collaborative correlation feature matrix, providing data support for subsequent energy consumption optimization and prediction.

[0038] Part of the deployment code is shown below.

[0039] import numpy as np

[0040] from velocity import matrix_multiply

[0041] #1. Load historical and current energy consumption feature vectors

[0042] historical_energy_vector=np.load('historical_energy_vector.npy')

[0043] current_energy_vector=np.load('current_energy_vector.npy')

[0044] #2. Ensure that the feature vector dimensions are consistent

[0045] assert historical_energy_vector.shape == current_energy_vector.shape, "feature vector dimensions are inconsistent"

[0046] #3. Calculate matrix product

[0047] energy_correlation_matrix=matrix_multiply(historical_energy_vector,current_energy_vector)

[0048] #4.Store the results

[0049] np.save('energy_correlation_matrix.npy',energy_correlation_matrix)

[0050] #5. Output results

[0051] print("The calculation of the communication base station energy consumption collaborative correlation characteristic matrix is ​​completed, and the results have been stored in energy_correlation_matrix.npy")

[0052] As you can understand, the deployment code above first ensures that the required runtime environment for the Velocity mechanism is configured and loads the historical and current 5G base station energy consumption feature vectors. These feature vectors are typically stored as NumPy arrays to facilitate subsequent matrix operations. Then, using the matrix_multiply function provided by the Velocity mechanism, the code calculates the product of the historical and current energy consumption feature vectors. This step is the core of the entire calculation process. Through matrix multiplication, it effectively captures the synergistic relationship between the two feature vectors. After the calculation is complete, the code stores the resulting synergistic correlation feature matrix in the file energy_correlation_matrix.npy for subsequent analysis and use. Furthermore, the results are printed to the console for immediate viewing and verification. Finally, the code runs a deployment verification step to ensure the accuracy and consistency of the calculation results. If errors or exceptions occur, users can use the error messages to debug and correct them, ensuring a smooth deployment process.

[0053] This deployment code quickly and efficiently calculates the energy consumption synergy correlation feature matrix for 5G base stations. This matrix not only contains key information about historical and current energy consumption data, but also reveals potential correlations and trends between them. By leveraging the Velocity mechanism's matrix multiplication function, the code can quickly process and calculate large amounts of data, significantly improving data processing efficiency. By analyzing the synergy correlation feature matrix, users can better understand the energy consumption patterns of 5G base stations and develop more effective energy optimization strategies. Furthermore, this matrix can be used to train energy consumption prediction models, improving prediction accuracy and reliability. The code's clear structure makes it easy to understand and modify, allowing users to adapt and expand it to meet their specific needs. For example, additional feature vector processing steps can be added or integrated into a larger data processing pipeline. This deployment code provides a powerful and flexible tool for analyzing 5G base station energy consumption data, helping users achieve better results in energy management and optimization.

[0054] Specifically, the Velocity mechanism is an efficient and flexible computing framework designed for large-scale data processing and complex computing tasks. By optimizing memory management and parallel computing techniques, it significantly improves the execution speed and resource utilization of tasks such as matrix operations and deep learning model training. The Velocity mechanism not only supports seamless integration of multiple data formats and computing libraries, but also provides a rich set of tools and interfaces to facilitate developers in quickly building and deploying high-performance computing applications.

[0055] In a specific embodiment of the present application, the 5G communication base station environment data feature extraction unit 122 includes: performing data preprocessing on the 5G communication base station environment data to obtain a 5G communication base station environment word embedding vector sequence; passing the 5G communication base station environment word embedding vector sequence through a 5G communication base station environment data semantic encoder to obtain the 5G communication base station environment semantic feature vector.

[0056] It's understandable that 5G base station environmental data typically includes information such as temperature, humidity, air pressure, noise level, and wind speed, which significantly impacts base station operation and performance. However, this environmental data is often raw, heterogeneous, and contains significant noise and irregularities. Therefore, data preprocessing is crucial to transform this raw data into a more structured and analyzable form. Raw environmental data may contain missing values, outliers, or noise. The data cleansing process ensures data quality and consistency by filling in missing values, correcting outliers, and removing noise. This step ensures the accuracy of subsequent analysis. After these preprocessing steps, the resulting 5G base station environmental word embedding vector sequences can be directly used by machine learning models for more accurate analysis and prediction. These vector sequences effectively represent the inherent characteristics and relationships of environmental data, thereby enhancing insight and decision-making capabilities for base station performance and energy management. In short, data preprocessing and the application of word embeddings are key steps to ensure that environmental data can be maximized in machine learning models. Specifically, the 5G communication base station environment data is segmented to obtain a base station environment word sequence; the embedding layer of the sequence encoder of the Clip model is used to map each base station environment word in the base station environment word sequence into a word embedding vector to obtain a 5G communication base station environment word embedding vector sequence.

[0057] Furthermore, the 5G base station environment word embedding vector sequence is processed through the 5G base station environment data semantic encoder to obtain the 5G base station environment semantic feature vector. This is to extract deeper semantic information from the embedding vector, making the analysis of environmental data more meaningful and accurate. The environmental word embedding vector sequence is obtained by converting the raw environmental data (such as temperature, humidity, and air pressure) into a continuous embedding vector. These vector sequences retain the basic characteristics of the environmental data and can reflect the relative relationships between the data. However, the embedding vector sequence may still lack deep semantic understanding of the data and capture contextual information. The introduction of the 5G base station environment data semantic encoder aims to further process these embedding vectors to extract more complex and abstract semantic features. Specifically, the semantic encoder can perform deeper processing on the embedding vector sequence to extract semantic relationships in the data. Through the encoding process, it maps the word embedding vectors to a new feature space, so that similar environmental features have closer relationships in this new space. This processing helps to better understand and describe the meaning and interrelationships of environmental data. The semantic encoder can extract high-level features from the embedding vectors through complex neural network structures (such as LSTM and Transformer). By encoding the vector sequence, the model can identify and learn the underlying patterns and structures in the data, thereby generating more meaningful and useful environmental semantic feature vectors. Specifically, the 5G communication base station environment data semantic encoder uses a transformer-based Bert model to perform global contextual semantic encoding on the 5G communication base station environment word embedding vector sequence to obtain multiple base station environment feature vectors; and the multiple base station environment feature vectors are cascaded to obtain the 5G communication base station environment semantic feature vector.

[0058] In the aforementioned 5G base station energy consumption monitoring and analysis system 100, the energy consumption monitoring data analysis module 130 is configured to determine whether to issue an abnormal 5G base station energy consumption warning based on the base station energy consumption collaborative correlation feature vector and the 5G base station environmental semantic feature vector. It should be understood that the base station energy consumption collaborative correlation feature vector integrates historical and current base station energy consumption data, providing dynamic trends and correlation patterns of base station energy consumption. These feature vectors can reflect the normal fluctuation range and behavioral patterns of base station energy consumption. The 5G base station environmental semantic feature vectors, through in-depth semantic analysis of environmental data, provide information on the impact of environmental factors on base station energy consumption. These feature vectors encompass the potential environmental impacts on base station operation, such as the impact of temperature, humidity, and air pressure on energy consumption. Combining these two features allows for a comprehensive analysis of base station energy consumption, considering not only temporal variations in energy consumption but also the impact of environmental factors, thereby providing a more accurate assessment of energy consumption status. By combining the feature vectors of energy consumption and environmental data, a normal pattern and prediction model for base station energy consumption can be established. When actual energy consumption data deviates significantly from the expected pattern, the system can identify potential anomalies. For example, if a base station's energy consumption suddenly increases under normal environmental conditions, it may indicate an equipment malfunction or abnormal operation.

[0059] Figure 4 FIG. 1 is a block diagram of an energy consumption monitoring data analysis module in an energy consumption monitoring and analysis system of a 5G communication base station according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the energy consumption monitoring data analysis module 130 includes: a communication base station feature fusion unit 131, which is used to fuse the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector to obtain a communication base station energy consumption anomaly judgment feature vector; a communication base station feature optimization unit 132, which is used to perform fine-grained feature internal structure optimization based on feature decomposition on the communication base station energy consumption anomaly judgment feature vector to obtain an optimized communication base station energy consumption anomaly judgment feature vector; a 5G communication base station energy consumption anomaly warning judgment unit 133, which is used to pass the optimized communication base station energy consumption anomaly judgment feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether to issue a 5G communication base station energy consumption anomaly warning.

[0060] It should be understood that the energy consumption collaborative correlation feature vector and the environmental semantic feature vector provide different levels of information. The energy consumption feature vector reflects the energy consumption behavior of the base station, while the environmental feature vector reveals the impact of external factors on energy consumption. Combining the two can obtain a more comprehensive perspective on anomaly detection. Through fusion, the complex relationship between energy consumption data and environmental factors can be taken into account. For example, if an environmental factor (such as increased temperature) causes an increase in energy consumption, and this increase exceeds the expected range of historical data patterns, the fused feature vector can more accurately identify whether there is an anomaly.

[0061] In particular, in the technical solution of the present application, the energy consumption collaborative correlation feature vector and the environmental semantic feature vector may form dimensional redundancy after fusion, making it difficult for the classifier to distinguish between key abnormal patterns and noise. For example, seasonal fluctuations in historical energy consumption may produce similar mappings in the feature space with periodic changes in ambient temperature, resulting in blurred classification boundaries. In addition, the original features may contain noise components that are irrelevant to the abnormal pattern, such as sensor errors or sudden interference. These noises will reduce the sensitivity of the model to real energy consumption anomalies. Therefore, in the technical solution of the present application, the communication base station energy consumption anomaly judgment feature vector is optimized by performing fine-grained feature internal structure optimization based on feature decomposition on the communication base station energy consumption anomaly judgment feature vector to obtain an optimized communication base station energy consumption anomaly judgment feature vector.

[0062] Among them, the fine-grained feature internal structure optimization based on feature decomposition of the communication base station energy consumption abnormality judgment feature vector is performed to obtain an optimized communication base station energy consumption abnormality judgment feature vector, including: first, calculating the full-dimensional fine-grained autocorrelation topological structure matrix of the communication base station energy consumption abnormality judgment feature vector, which is expressed as follows:

[0063]

[0064]

[0065]

[0066] v i ,v j ∈V

[0067] Among them, V represents the abnormal energy consumption judgment feature vector of the communication base station, v i and v j They represent the i-th and j-th eigenvalues ​​of the communication base station energy consumption anomaly judgment feature vector, w1, w2, w3 and w4 represent different weight hyperparameters, ⊙ represents matrix dot product, D1 represents the forward weighted matrix of the communication base station energy consumption anomaly judgment feature, D2 represents the reverse weighted matrix of the communication base station energy consumption anomaly judgment feature, Represents the value of the (i, j)th position of the forward weighted matrix of the abnormal energy consumption judgment feature of the communication base station, Indicates the value of the (i, j)th position of the inverse weighted matrix of the abnormal energy consumption judgment feature of the communication base station, M i,j Represents a full-dimensional fine-grained autocorrelation topology matrix.

[0068] That is, by quantifying the statistical dependence and structural correlation between the various dimensions of the communication base station energy consumption anomaly judgment feature vector, the multi-dimensional interactive relationship implicit in the energy consumption collaborative correlation feature vector and the environmental semantic feature vector is converted into a topological matrix form that can be directly analyzed. The generated full-dimensional fine-grained autocorrelation topological structure matrix comprehensively covers the fine-grained correlation information of all dimensions of the communication base station energy consumption anomaly judgment feature vector, laying a mathematical foundation for the subsequent analysis of the internal structure of the feature and identification of key abnormal patterns, improving the communication base station energy consumption anomaly judgment feature vector's ability to represent energy consumption anomaly patterns, and ensuring that the classifier can more accurately identify abnormal states.

[0069] Secondly, the full-dimensional fine-grained autocorrelation topological structure matrix is ​​subjected to eigendecomposition to obtain a set of fine-grained eigendecomposition coding vectors for judging abnormal energy consumption of communication base stations, which is expressed as follows:

[0070]

[0071] Where Λ represents a diagonal matrix, λ1 and λ m denote the first and mth eigenvalues ​​of the diagonal matrix, respectively, (·) T represents the transpose of the vector, U represents the set of fine-grained intrinsic decomposition encoding vectors of the abnormal energy consumption judgment feature of the communication base station, x1, x2 and x m They respectively represent the first, second and mth communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vectors in the set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vectors.

[0072] Specifically, by performing eigendecomposition on the fully dimensional, fine-grained autocorrelation topological structure matrix, the complex correlation structure in the feature vectors for determining abnormal energy consumption in communication base stations, formed by the fusion of historical and current energy consumption data and environmental factors, is transformed into an orthogonal basis set guided by the intrinsic structure of the data. This decouples the internal dependencies of the features and provides operational mathematical units for the subsequent targeted optimization of each independent structural pattern. This results in a set of fine-grained eigendecomposition encoding vectors for determining abnormal energy consumption in communication base stations, ranked by importance. These vectors form independent fundamental change patterns in the original feature space, effectively separating the coupling relationship between energy consumption data and environmental semantics, as well as redundant associations between feature dimensions.

[0073] Then, information compression is performed on each communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vector in the set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vectors to obtain a set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vectors, which is expressed as follows:

[0074]

[0075] Among them, x i represents the fine-grained intrinsic decomposition coding vector of the energy consumption anomaly judgment feature of the communication base station in the set of fine-grained intrinsic decomposition coding vectors, ||·|| represents the Euclidean norm, and y i Represents the fine-grained intrinsic decomposition compression coding vector of the abnormal energy consumption judgment feature of the i-th communication base station.

[0076] Specifically, a nonlinear transformation function is used to adjust the norm and compress the dynamic range of the fine-grained intrinsic decomposition encoding vectors of the communication base station energy consumption anomaly judgment features. This captures higher-order dependencies between energy consumption data and environmental semantic features that transcend linear correlations, breaking the linear framework limitations of feature decomposition. This allows the discriminative information related to energy anomalies in the fine-grained intrinsic decomposition encoding vectors to be retained, suppressing noise interference and redundant signals. After information compression, the fine-grained intrinsic decomposition encoding vectors of the communication base station energy consumption anomaly judgment features are transformed into compressed fine-grained intrinsic decomposition encoding vectors containing high-order statistical characteristics, laying the foundation for subsequent feature weight optimization through intrinsic saliency factors.

[0077] Next, the intrinsic significance factor of each communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vector in the set of the communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vector is calculated to obtain a set of intrinsic significance factors, which is expressed as follows:

[0078]

[0079] Among them, α and β represent different weight parameters, ||·||1 represents the first norm, ||·||2 represents the second norm, L represents the length of the fine-grained intrinsic decomposition compression coding vector of the abnormal judgment feature of the communication base station energy consumption, a i represents y i The corresponding intrinsic significance factor.

[0080] Specifically, by calculating scalar values ​​for the fine-grained intrinsic decomposition and compression code vectors of the communication base station energy anomaly judgment features, the value of the structural information carried by each component is quantified, providing a basis for dynamic importance assessment based on the modulated state for subsequent fusion processing, and achieving adaptive value reconstruction of the fine-grained intrinsic decomposition and compression code vectors of the communication base station energy anomaly judgment features. The resulting set of intrinsic significance factors can dynamically characterize the actual contribution of each fine-grained intrinsic decomposition and compression code vector in energy anomaly judgment, providing a precise basis for weight allocation for fine-grained weighted fusion of feature vectors.

[0081] Then, the set of intrinsic saliency factors is normalized to obtain a set of intrinsic saliency weight factors, which is expressed as follows:

[0082] w i =Softmax(a i )

[0083] Among them, Softmax(·) represents the normalization function, w i Indicates a i The corresponding intrinsic saliency weight factor.

[0084] Specifically, through Softmax normalization, the intrinsic significance factors are converted into a set of intrinsic significance weight factors whose sum is 1. This standardizes the numerical scale and controls the sharpness of the weight distribution, avoiding numerical instability or extreme bias toward individual fine-grained intrinsic decomposition compressed code vectors of the energy anomaly judgment features during subsequent fusion. This provides a stable and controllable weight basis for the adaptive weighted fusion of the energy anomaly judgment features of the communication base station. The resulting set of intrinsic significance weight factors quantifies the relative importance of each fine-grained intrinsic decomposition compressed code vector of the energy anomaly judgment features of the communication base station using a probability distribution, ensuring that the contribution of each fine-grained intrinsic decomposition compressed code vector of the energy anomaly judgment features of the communication base station is reasonably distributed.

[0085] Finally, based on the set of intrinsic significance weight factors, the set of fine-grained intrinsic decomposition compression coding vectors of the communication base station energy consumption anomaly judgment feature is fine-grainedly fused to obtain an optimized communication base station energy consumption anomaly judgment feature vector, which is expressed as follows:

[0086]

[0087] Among them, V ' Represents the optimized communication base station energy consumption abnormality judgment feature vector.

[0088] Specifically, a set of intrinsic saliency weight factors is used to perform a weighted combination of the fine-grained intrinsic decomposition and compression coding vectors of the communication base station energy consumption anomaly judgment features. Through fine-grained operations at the intrinsic component level, key information in the features is enhanced and noise is suppressed. This results in an optimized communication base station energy consumption anomaly judgment feature vector that integrates historical energy consumption, current energy consumption, and environmental semantic features, providing more discriminative input for downstream energy consumption anomaly classification tasks. Through this weighted fusion, the optimized communication base station energy consumption anomaly judgment feature vector condenses the key structural information in the communication base station energy consumption anomaly judgment feature vector. Through nonlinear modulation and weighted fusion, the ability to characterize energy consumption anomaly patterns is enhanced, making it more robust than the communication base station energy consumption anomaly judgment feature vector.

[0089] Furthermore, the optimized feature vector for determining abnormal energy consumption in communication base stations incorporates energy consumption synergy correlation features and environmental semantic features, integrating the base station's energy consumption pattern and environmental influencing factors to form a comprehensive basis for determining abnormalities. However, directly interpreting the optimized feature vector for determining abnormal energy consumption in communication base stations and determining whether an abnormality exists is difficult and time-consuming. Therefore, the introduction of a classifier automates this process, converting the optimized feature vector for determining abnormal energy consumption in communication base stations into easily understandable classification results. The classifier operates based on a machine learning algorithm. It uses training data to learn how to classify different feature vectors into predefined categories, such as "normal" or "abnormal." During training, the classifier learns the relationship between the feature vector for determining abnormal energy consumption in communication base stations and abnormal conditions, building a model that issues early warnings and notifies relevant personnel to take necessary measures, thereby reducing losses and maintenance costs.

[0090] In summary, the embodiment of the present application first obtains historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data and 5G communication base station environmental data, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a classifier is used to determine whether to issue an abnormal energy consumption warning for the 5G communication base station, thereby more accurately identifying potential energy consumption abnormality patterns, improving the accuracy of the early warning system, and then quickly responding and taking necessary measures to improve system processing efficiency and decision-making capabilities.

[0091] As described above, the energy consumption monitoring and analysis system 100 for a 5G communication base station according to an embodiment of the present application can be implemented in various terminal devices. In one example, the energy consumption monitoring and analysis system 100 for a 5G communication base station can be integrated into the terminal device as a software module and / or a hardware module. For example, the energy consumption monitoring and analysis system 100 for a 5G communication base station can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the energy consumption monitoring and analysis system 100 for a 5G communication base station can also be one of the many hardware modules of the terminal device.

[0092] Alternatively, in another example, the energy consumption monitoring and analysis system 100 of the 5G communication base station and the terminal device may also be separate devices, and the energy consumption monitoring and analysis system 100 of the 5G communication base station may be connected to the terminal device through a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0093] Figure 5 Flowchart of the energy consumption monitoring and analysis method of a 5G communication base station according to an embodiment of the present application. Figure 5 As shown, according to the energy consumption monitoring and analysis method of the 5G communication base station in the embodiment of the present application, it includes: S110, obtaining historical 5G communication base station energy consumption data, current energy consumption data of the 5G communication base station and 5G communication base station environmental data; S120, extracting the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current energy consumption data of the 5G communication base station and the 5G communication base station environment data; S130, based on the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector, judging whether to issue an abnormal 5G communication base station energy consumption warning.

[0094] Here, those skilled in the art will understand that the specific operations of each step in the above-mentioned 5G communication base station energy consumption monitoring and analysis method have been referred to above. Figures 1 to 4 The energy consumption monitoring and analysis system of the 5G communication base station has been introduced in detail, and therefore, its repeated description will be omitted.

[0095] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any other combination. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0096] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0097] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0098] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0099] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0100] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A 5G communication base station energy consumption monitoring and analysis system, characterized in that: include: 5G communication base station energy consumption data acquisition module, used to obtain historical 5G communication base station energy consumption data, 5G communication base station current energy consumption data and 5G communication base station environmental data; A 5G communication base station energy consumption data extraction module is used to extract a communication base station energy consumption collaborative correlation feature vector and a 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current energy consumption data of the 5G communication base station and the 5G communication base station environment data; The energy consumption monitoring data analysis module is used to determine whether to issue an abnormal energy consumption warning for the 5G communication base station based on the collaborative correlation feature vector of the communication base station energy consumption and the semantic feature vector of the 5G communication base station environment.

2. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 1, characterized in that: The 5G communication base station energy consumption data extraction module includes: A communication base station energy consumption collaborative feature extraction unit, configured to extract features from the historical 5G communication base station energy consumption data and the current energy consumption data of the 5G communication base station to obtain a communication base station energy consumption collaborative correlation feature vector; The 5G communication base station environment data feature extraction unit is used to extract features from the 5G communication base station environment data to obtain the 5G communication base station environment semantic feature vector.

3. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 2, characterized in that: The communication base station energy consumption collaborative feature extraction unit includes: A historical 5G communication base station energy consumption semantic encoding subunit, configured to obtain a historical 5G communication base station energy consumption feature vector by passing the historical 5G communication base station energy consumption data through a historical 5G communication base station energy consumption data semantic encoder based on a Taming model; A 5G communication base station current energy consumption semantic encoding subunit, configured to pass the 5G communication base station current energy consumption data through a 5G communication base station current energy consumption data semantic encoder based on a Taming model to obtain a 5G communication base station current energy consumption feature vector; The communication base station energy consumption collaborative feature aggregation subunit is used to perform feature aggregation on the historical 5G communication base station energy consumption feature vector and the current energy consumption feature vector of the 5G communication base station to obtain the communication base station energy consumption collaborative association feature vector.

4. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 3, characterized in that: The communication base station energy consumption collaborative feature aggregation subunit includes: Correlating the historical 5G communication base station energy consumption feature vector with the current energy consumption feature vector of the 5G communication base station to obtain a communication base station energy consumption collaborative correlation feature matrix; The communication base station energy consumption cooperative correlation feature matrix is ​​passed through an energy consumption data association encoder with mutually transposed convolution kernels to obtain a communication base station energy consumption cooperative correlation feature vector.

5. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 4, characterized in that: Associating the historical 5G communication base station energy consumption feature vector with the current energy consumption feature vector of the 5G communication base station to obtain a communication base station energy consumption collaborative correlation feature matrix, including: Configure the operating environment required by the Velocity mechanism; Load the historical 5G communication base station energy consumption feature vector historical_energy_vector and the current 5G communication base station energy consumption feature vector current_energy_vector; Use the matrix multiplication function matrix_multiply provided by the Velocity mechanism to calculate the correlation matrix of historical_energy_vector and current_energy_vector, and store the calculation result as the communication base station energy consumption collaborative correlation feature matrix energy_correlation_matrix; Store the calculated energy_correlation_matrix into the file energy_correlation_matrix.npy and output the result to the console or log file; Run the deployed code to verify the calculation results.

6. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 5, characterized in that: The 5G communication base station environment data feature extraction unit includes: Performing data preprocessing on the 5G communication base station environment data to obtain a 5G communication base station environment word embedding vector sequence; The 5G communication base station environment word embedding vector sequence is passed through a 5G communication base station environment data semantic encoder to obtain the 5G communication base station environment semantic feature vector.

7. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 6, characterized in that: The energy consumption monitoring data analysis module includes: A communication base station feature fusion unit, configured to fuse the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector to obtain a communication base station energy consumption anomaly judgment feature vector; A communication base station feature optimization unit, configured to perform fine-grained feature internal structure optimization based on feature decomposition on the communication base station energy consumption abnormality judgment feature vector to obtain an optimized communication base station energy consumption abnormality judgment feature vector; The 5G communication base station energy consumption abnormality warning judgment unit is used to pass the optimized communication base station energy consumption abnormality judgment feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether to issue a 5G communication base station energy consumption abnormality warning.

8. The energy consumption monitoring and analysis system for a 5G communication base station according to claim 7, characterized in that: The communication base station feature optimization unit includes: Calculating a full-dimensional fine-grained autocorrelation topological structure matrix of the communication base station energy consumption anomaly judgment feature vector; Performing eigendecomposition on the full-dimensional fine-grained autocorrelation topological structure matrix to obtain a set of fine-grained eigendecomposition coding vectors for judging abnormal energy consumption of communication base stations; Performing information compression on each communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vector in the set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition coding vectors to obtain a set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vectors; Calculating the intrinsic significance factor of each communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vector in the set of communication base station energy consumption anomaly judgment feature fine-grained intrinsic decomposition compressed coding vectors to obtain a set of intrinsic significance factors; Normalizing the set of intrinsic saliency factors to obtain a set of intrinsic saliency weight factors; Based on the set of the intrinsic significance weight factors, the set of fine-grained intrinsic decomposition compression coding vectors of the communication base station energy consumption anomaly judgment feature is fine-grainedly fused to obtain an optimized communication base station energy consumption anomaly judgment feature vector.

9. A method for monitoring and analyzing energy consumption of a 5G communication base station, characterized in that: include: Obtain historical 5G communication base station energy consumption data, current 5G communication base station energy consumption data, and 5G communication base station environmental data; Extracting a communication base station energy consumption collaborative correlation feature vector and a 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current 5G communication base station energy consumption data, and the 5G communication base station environment data; Based on the communication base station energy consumption collaborative correlation feature vector and the 5G communication base station environment semantic feature vector, it is determined whether to issue a 5G communication base station energy consumption abnormality warning.

10. The energy consumption monitoring and analysis method of a 5G communication base station according to claim 9, characterized in that: Extracting a communication base station energy consumption collaborative correlation feature vector and a 5G communication base station environment semantic feature vector from the historical 5G communication base station energy consumption data, the current 5G communication base station energy consumption data, and the 5G communication base station environment data, including: Performing feature extraction on the historical 5G communication base station energy consumption data and the current energy consumption data of the 5G communication base station to obtain a collaborative correlation feature vector of the communication base station energy consumption; Feature extraction is performed on the 5G communication base station environment data to obtain the 5G communication base station environment semantic feature vector.

Images