Intelligent management and control platform and method based on base station management

Abstract

The invention relates to an intelligent management and control platform and method based on base station management. The intelligent management and control platform and method are applied to operation state monitoring and regulation and control optimization of a plurality of communication base stations. The method comprises the following steps: S10, acquiring operation data of a plurality of communication base stations in a target area, and performing structured processing on the data to generate operation data input in a standard format; s20, performing fusion analysis on the structured operation data, and calculating an operation state score value and a corresponding state label of each base station; s30, based on the score value and the state label, constructing a risk prediction model, identifying a potential fault trend of each base station, and outputting a risk level and early warning information; s40, matching a regulation and control strategy template from a strategy library according to the risk level and the early warning information, and automatically executing corresponding scheduling operation; and S50, collecting a feedback result of the regulation and control operation, and calculating an operation performance difference before and after regulation and control to optimize a parameter weight in the regulation and control strategy library.

Description

Technical Field

[0001] The present invention relates to the field of communication base station management, and particularly relates to an intelligent control platform and method based on base station management. Background Art

[0002] With the continuous expansion and complexity of communication networks, especially in the 5G and future network architectures, the number and distribution density of communication base stations have increased significantly. The operating status of base stations is affected by various factors, including communication load, power status, signal quality, and external environment. In the prior art, the monitoring and regulation of the operating status of base stations mostly rely on manual analysis and static rule configuration, lacking adaptive and intelligent management means, and it is difficult to meet the requirements of multi-dimensional data fusion and real-time regulation in complex environments. In addition, the response speed and accuracy of traditional methods in fault detection and regulation optimization are relatively low, and the potential information of large-scale data cannot be fully utilized. Based on this, an intelligent control platform is proposed, which can analyze the operating status of base stations in real time, predict potential risks, and adaptively adjust regulation strategies by introducing deep learning algorithms and automated strategy matching mechanisms, thereby improving the stability and management efficiency of communication base stations. Summary of the Invention

[0003] The present invention provides an intelligent control method based on base station management, which includes: S10. Obtain the operation data of multiple communication base stations in the target area, and perform structured processing on the data to generate an operation data input in a standard format; S20. Perform fusion analysis on the structured operation data, and calculate the operation status score value and corresponding status label of each base station; S30. Based on the score value and status label, construct a risk prediction model, identify the potential fault trend of each base station, and output the risk level and warning information; S40. According to the risk level and warning information, match the regulation strategy template from the strategy library and automatically execute the corresponding scheduling operation; S50. Collect the feedback results of the regulation operation, and calculate the difference in operation performance before and after regulation, which is used to optimize the parameter weights in the regulation strategy library.

[0004] An intelligent control method based on base station management as described above, wherein obtaining the operation data of multiple communication base stations in the target area and performing structured processing on the data to generate an operation data input in a standard format includes: Automatically obtain the real-time operation data of multiple communication base stations in the target area, and the operation data includes load conditions, power status, signal quality, and environmental parameters; Perform formatting processing and timestamp sorting on the collected data to generate a standardized input data set; Denoise the input dataset to eliminate invalid or abnormal data to ensure data quality.

[0005] An intelligent control method based on base station management as described above, wherein, perform fusion analysis on structured operation data, and calculate the operation status score value and corresponding status label for each base station, including: Extract multi-dimensional features such as load, voltage, temperature, and signal quality from the input dataset, and construct a feature vector; Input the feature vector into the fusion analysis model for comprehensive evaluation, and generate the operation status score value for each base station; Generate corresponding status labels according to the calculation results of the score values, and store the score values and status labels in the database.

[0006] An intelligent control method based on base station management as described above, wherein, based on the score values and status labels, construct a risk prediction model, identify the potential fault trend of each base station, and output the risk level and warning information, including: Use the score values and status labels combined with historical data to train the risk prediction model, and generate prediction parameters for identifying fault trends; Analyze the operation status of the base station based on the trained risk prediction model, and generate the corresponding risk level; Generate warning information according to the risk level, and transmit the warning information to the policy library for subsequent policy matching.

[0007] An intelligent control method based on base station management as described above, wherein, according to the risk level and warning information, match the regulation policy template from the policy library and automatically execute the corresponding scheduling operations, including: Retrieve the regulation policy template that matches the current status from the policy library according to the risk level and warning information; Execute the selected regulation policy template through the automated scheduling module, including power adjustment, load transfer, or standby activation operations; Record the execution results of the regulation operations, and store the execution results in the database for subsequent feedback analysis.

[0008] An intelligent control method based on base station management as described above, wherein, collect the feedback results of the regulation operations, and calculate the difference in operation performance before and after regulation, for optimizing the parameter weights in the regulation policy library, including: Collect the operation status data after the regulation operation is completed, and generate a feedback dataset for analysis; Calculate and analyze the feedback dataset to determine the change range of the performance indicators before and after regulation; Update the parameter weights or policy templates in the policy library according to the calculation results to improve the self-adaptability and accuracy of the policy.

[0009] An intelligent control method based on base station management as described above, wherein the policy library includes a rule template module and a deep learning recommendation module. The rule template module is used to retrieve regulation policies through label indexing, and the deep learning recommendation module is used to generate adaptive regulation policies based on real-time data.

[0010] The present invention also provides an intelligent control platform based on base station management, which includes: A data collection module, used to automatically obtain the operation data of multiple communication base stations and perform formatting processing; A status analysis module, used to perform fusion analysis and status score calculation on structured data; A risk prediction module, used to generate warning information based on the score value and historical data; A policy matching module, used to retrieve and execute regulation policies from the policy library; A feedback optimization module, used to collect regulation feedback data and update the parameters in the policy library.

[0011] The beneficial effects achieved by the present invention are as follows: Through the organic combination of intelligent data collection, status analysis, risk prediction, policy matching, and feedback optimization modules, the present invention realizes real-time monitoring, risk prediction, and automatic regulation of communication base stations. The system can adaptively adjust the parameters in the policy library, improve the accuracy and self-learning ability of regulation policies, and thus enhance the stability and management efficiency of the operation of communication base stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0013] Figure 1 It is a flowchart of an intelligent control method based on base station management provided in Embodiment 1 of the present application; Figure 2 It is a schematic diagram of an intelligent control platform based on base station management provided in Embodiment 2 of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0015] Embodiment 1 As Figure 1 shown, Embodiment 1 of the present application provides an intelligent control method based on base station management, including the following steps: S10. Obtain the operation data of multiple communication base stations in the target area, and perform structured processing on the data to generate a standard format operation data input; The system obtains and processes the real-time operation data of multiple communication base stations in the target area through an automated acquisition and processing mechanism. The aim is to ensure the integrity and standardization of the data, and provide high-quality input data for subsequent data analysis and risk prediction. It includes the following sub-steps: S11. Automatically obtain the real-time operation data of multiple communication base stations in the target area, and the operation data includes load conditions, power supply status, signal quality, and environmental parameters; The system first automatically obtains the real-time operation data of multiple communication base stations in the target area, including load conditions, power supply status, signal quality, and environmental parameters. To achieve comprehensive coverage and accurate acquisition, the system collects the above data in real time through the monitoring module or remote acquisition module configured on each communication base station, and uses a distributed data acquisition architecture for transmission and storage to ensure the continuity and integrity of the data.

[0016] S12. Perform formatting processing and timestamp sorting on the collected data to generate a standardized input data set; After the data collection is completed, the system performs formatting processing and timestamp sorting on the collected raw data to generate a standardized input data set. The data of different base stations is uniformly converted into a data structure that conforms to a predetermined standard, so as to support the efficient reading and processing of subsequent analysis modules. At the same time, the system performs timestamp sorting on all data records to ensure the temporal consistency and accuracy of the data.

[0017] S13. Perform denoising processing on the input data set to remove invalid or abnormal data to ensure data quality.

[0018] The system performs denoising processing on the generated input data set to remove invalid or abnormal data to ensure data quality. In this process, the system detects and filters abnormal data through a built-in rule engine and statistical methods, and automatically removes incorrect data according to the set tolerance standard and data characteristics. After processing, the generated standard format operation data input will be stored in the database to support the fusion analysis and status scoring of subsequent steps.

[0019] S20. Perform fusion analysis on the structured operation data, and calculate the operation status score value and corresponding status label of each base station; In step S20, the system deeply analyzes and calculates the structured operation data, aiming to accurately evaluate the operation status of each base station and provide scientific evaluation results in a quantitative manner. Specifically, it includes the following sub-steps: S21. Extract multi-dimensional features such as load, voltage, temperature, and signal quality from the input dataset and construct a feature vector; The system extracts features in multiple dimensions from the input dataset, including but not limited to: load features, including the number of user connections, network throughput, concurrent request numbers, etc., reflecting the communication load level of the base station; power supply features, including voltage, current, power consumption, etc., reflecting the power usage and health status of the base station equipment; temperature features, including the internal temperature of the equipment, ambient temperature, etc., reflecting the stability and safety of the equipment working environment; signal quality features, including signal-to-noise ratio, bit error rate, received signal strength indication, etc., reflecting the communication quality and link stability; environmental features, including external environmental parameters such as humidity, air pressure, wind speed, etc., which have an indirect impact on the base station performance.

[0020] Based on the above multiple feature dimensions, the system constructs a high-dimensional feature vector for each base station: , where n is the total number of features; represents the feature values under different categories; the feature vector X represents the global state of the current base station.

[0021] S22. Input the feature vector into the fusion analysis model for comprehensive evaluation and generate the operation status score value of each base station; The system inputs the feature vector X into the fusion analysis model, evaluates it by integrating multiple feature dimensions, and generates the operation status score value S of each base station. The following calculation formula is used in this process: , where n represents the feature vector dimension; i is the feature index used to distinguish different feature values; m is the total number of intermediate state variables; j is the index of the intermediate state variable used to distinguish different state variables; p is the total number of external environment variables, representing the number of independent environment parameters related to the system operation; k is the index of the external environment; L is the number of perturbation functions included in each environment variable; l is the index of the environmental perturbation function used to describe different influencing factors of the environment variable; represents the weight of the feature, indicating the importance of each feature in the score calculation, which is determined based on historical data training and feature importance analysis; is the non-linear mapping function of the feature, used to process the non-linear relationship of the feature; T represents the time interval length, indicating the total time range of data collection and analysis; represents the evolution function of the feature over time, reflecting the state of the feature at different time points; represents the time change rate of the feature, characterizing the change rate of the feature in the time dimension; is an intermediate state variable, representing the state parameter generated by the combined action of multiple features; is a non-linear mapping of the intermediate state variable, used to extract high-level features; is a function of the state variable changing with time, reflecting the dynamic change of the system state; is an external environmental variable, representing external factors that affect the operating state of the base station, such as meteorological conditions, geographical location, etc.; is the influence factor of the environmental variable, used to measure the degree of influence of the external environment on the score; The disturbance function of the environmental variable changing with time, characterizing the dynamics of the environmental impact.

[0022] S23. Generate corresponding state labels according to the calculation result of the score value, and store the score value and the state label in the database.

[0023] After the score calculation is completed, the system generates corresponding state labels according to the score result to identify the operating states of different base stations. The generation process of the state label includes three links: classification rule setting, label generation, and storage update.

[0024] First, the system divides the operating states of each base station into three levels according to the level of the score value S: normal, warning, and fault. The classification rule is based on the comparison between the score value and the preset threshold to determine the label, where L represents the state label generated by the system; S is the operating state score value; and are thresholds obtained through training and optimization by machine learning algorithms based on historical data and operation records.

[0025] After the system generates the state label, it stores the label and the score result in the database together for subsequent query and analysis. After each score calculation, the system automatically updates the records in the database to maintain the real-time and accuracy of the data.

[0026] S30. Based on the score value and the state label, construct a risk prediction model, identify the potential fault trend of each base station, and output the risk level and warning information; In step S30, the system constructs a risk prediction model based on the dataset of the score value and the state label, combined with historical operation records and label information, to identify the potential fault trend of each base station. It includes the following sub-steps: S31. Use the score value and the state label combined with historical data to train the risk prediction model and generate prediction parameters for identifying the fault trend; The system preprocesses the collected historical data, including the score value, status label, and original feature data of the base station operation. To ensure the quality and consistency of the input data, the system performs matching and alignment operations on these data, and uses a joint denoising algorithm based on the moving average method and Gaussian filtering to remove outliers and noise in the data. During the data standardization process, the system selects the Z-Score standardization method to make the mean of each feature dimension 0 and the standard deviation 1, so as to ensure that data of different scales can effectively participate in the training.

[0027] After completing the data preprocessing, the system extracts important features closely related to risk prediction from the standardized dataset and constructs a feature vector set. To reduce data redundancy and improve model performance, the system extracts the most informative principal components through principal component analysis (PCA), and at the same time applies the recursive feature elimination (RFE) algorithm to sort and screen the features, and retains the features with the highest contribution to the prediction effect. The finally obtained feature vector set is represented as a high-dimensional vector X and is used for subsequent deep learning model training.

[0028] In the process of model construction, the system selects the long short-term memory network (LSTM) as the core risk prediction model. The LSTM model consists of an input layer, an LSTM layer, a fully connected layer, and an output layer. The input layer is used to receive the feature vector, and the LSTM layer models the time series features of the input data through a network structure composed of multiple neurons. To ensure that the model can capture long-term dependence relationships, the system sets the number of LSTM layers to three, each layer contains 128 neurons, and the tanh activation function is used.

[0029] During the training process, the output layer of the LSTM model generates a single output value, which is used to represent the risk score value of the base station. Through the backpropagation algorithm, the system continuously updates the weight matrix and bias vector, and evaluates and adjusts them on the validation set. After the training is completed, the optimized model parameters are saved to the database to support subsequent inference and analysis.

[0030] S32. Analyze the operating status of the base station based on the trained risk prediction model and generate the corresponding risk level; The system uses the trained LSTM model to perform real-time analysis on the operating status of the base station. The current score value and feature vector of the base station are passed as inputs to the LSTM model, and the model calculates and outputs a risk score value through forward propagation. The system comprehensively evaluates the operating status of the base station by analyzing the change rules of each feature in the time series and combining the key features extracted during the model training process.

[0031] The risk assessment formula is as follows: , where R is the final risk score value, which is used to evaluate the operating status of the base station and the potential fault trend; Represents the feature weight; Represents the i-th input feature; T represents the length of the time interval, indicating the time range for data collection and analysis; t is the independent variable, representing the continuous change from the starting time to the analysis time T; Is the feature correlation coefficient; Is the non-linear mapping function of the feature, reflecting the complex relationship of the feature at different time points; Is used to reflect sudden events; Represents the interaction kernel function between features, indicating the coupling relationship between different features; Represents the time decay factor, used to indicate that the influence of the feature on the score gradually weakens with the event; Is the non-linear transformation function of the intermediate state variable, used to extract the potential relationship in the high-dimensional feature space; Is the time evolution function of the state variable, indicating the dynamic change of the feature over time; Is the influence function of the environmental variable, indicating the interference effect of the external environment on the base station score; p represents the total number of environmental variables; m is the number of intermediate state variables, indicating the number of features in the high-dimensional space generated during the data processing; n represents the total number of input features, indicating the dimension of the original features collected from the base station. After generating the risk score value, the system compares the score result with the preset threshold and classifies it according to different risk levels. When the risk score value is lower than the high-risk threshold, the system determines it as a high-risk state, indicating that there are serious faults or abnormal conditions in the base station; when the score value is between the high-risk threshold and the medium-risk threshold, the system determines it as a medium-risk state, indicating that there are potential fault trends or performance degradations in the base station; when the score value is higher than the medium-risk threshold, the system determines it as a low-risk state, indicating that the operating state of the base station is relatively stable.

[0032] S33. Generate warning information according to the risk level and transmit the warning information to the policy library for subsequent policy matching.

[0033] The system generates warning information according to the risk score value and the corresponding risk level and transmits it to the policy library for subsequent policy matching.

[0034] The generation of the warning information is based on the results of the risk score value calculation and the risk level analysis, and includes the following key contents: Base station identifier, used to uniquely identify the currently monitored communication base station, including the base station number or geographical location code; Risk score value, the comprehensive risk score value calculated by the system, reflecting the operating status and potential risk level of the current base station; Risk level, divided into three levels according to the risk score value and preset thresholds: high risk, medium risk, and low risk; Timestamp, recording the specific time when the warning information is generated to ensure the timeliness and traceability of the data; Feature contribution analysis, the system conducts a detailed analysis of the contribution of each feature, lists the features that have the greatest impact on the risk score value, and provides a contribution ranking; External environment impact factors, including the interference impact of environmental variables on the operating status of the base station, such as temperature, humidity, wind speed, etc.; Data source identifier, indicating the original data source used to calculate the risk score, facilitating data traceability and verification.

[0035] The generated warning information is stored in the system database in a structured format and transmitted through the network to the policy library for centralized management. The system updates and records the warning information in real time to ensure the accuracy and timeliness of the policy matching process. In addition, through the comprehensive analysis of the warning information, the system can provide visual reports and trend charts to show the risk changes of different base stations within a specific time period.

[0036] In the policy library, the system parses the warning information through the rule engine and automatically triggers different policy response processes according to the risk level. In the high-risk state, the system immediately executes the emergency policy response program; in the medium-risk state, the system selects the best regulation plan through the policy library and implements optimization operations; in the low-risk state, the system maintains the monitoring state and conducts regular re-inspections and evaluations.

[0037] S40. According to the risk level and warning information, match the regulation policy template from the policy library and automatically execute the corresponding scheduling operation; In step S40, based on the warning information generated by the aforementioned risk score value and risk level, the system automatically retrieves the regulation policy template that matches the current state from the policy library and executes the corresponding scheduling operation through the automated scheduling module. This process includes the following sub-steps: S41. Retrieve the regulation policy template that matches the current state from the policy library according to the risk level and warning information; The system first retrieves the regulation policy template that matches the current state from the policy library according to the risk score value, risk level, feature contribution analysis, and environmental impact factors included in the warning information. The templates in the policy library are divided into three types: high-risk policies, medium-risk policies, and low-risk policies.

[0038] High-risk strategies are mainly used to handle severe faults or abnormal states, usually including emergency power adjustment, load transfer, or activation of standby base stations. Medium-risk strategies are used to optimize the performance and stability of base stations, reducing the impact of potential risks through fine power adjustment and resource scheduling. Low-risk strategies are used for routine maintenance and monitoring operations to ensure the long-term stability of the system.

[0039] S42. Execute the selected regulation strategy template through the automated scheduling module, including power adjustment, load transfer, or standby activation operations; After retrieving the strategy template, the system transfers the selected regulation strategy template to the automated scheduling module for execution. The core tasks of the automated scheduling module include: power adjustment, real-time adjustment of the transmission power of the base station to optimize the signal coverage and transmission quality; load transfer, transferring part of the current base station's load to other neighboring base stations to relieve the overload pressure or balance resource allocation; standby activation, automatically activating the standby base station to take over the work of the faulty base station in a high-risk state, thus ensuring the continuity and stability of communication.

[0040] During the execution of the regulation strategy, the system monitors the effect of the operation in real time through a feedback mechanism and makes dynamic adjustments according to the feedback information.

[0041] S43. Record the execution result of the regulation operation and store the execution result in the database for subsequent feedback analysis.

[0042] The system records the execution result of the regulation operation in detail and stores it in the database for subsequent feedback analysis and strategy optimization.

[0043] The recorded content of the regulation operation result includes: execution time, recording the start time and end time of the regulation strategy execution; execution result, including the specific operation results of power adjustment, load transfer, and standby activation; performance difference, recording the difference in the base station operation performance indicators before and after regulation to measure the effect of the regulation strategy; status feedback, including the change in the risk score value, risk level, and environmental impact factor of the current base station; strategy template identifier, recording the number of the executed regulation strategy template for subsequent optimization and improvement.

[0044] The system stores all the recorded information in the database and optimizes and updates the regulation templates in the strategy library through periodic analysis and regression algorithms.

[0045] S50. Collect the feedback results of the regulation operation and calculate the difference in the operation performance before and after regulation for optimizing the parameter weights in the regulation strategy library.

[0046] In step S50, the system collects feedback on the operating state after the regulation operation and calculates the performance difference. By deeply analyzing and comparing the feedback data, the system can continuously optimize the parameter weights and policy templates in the regulation strategy library to improve the adaptability and accuracy of the strategy. Specifically, it includes the following sub-steps: S51. Collect the operating state data after the regulation operation and generate a feedback data set for analysis; After the regulation operation is completed, the system automatically collects the operating state data of the base station, including the current risk score value, risk level, feature contribution degree, and external environment impact factors. The system formats and structurally stores the collected data to generate a feedback data set for subsequent analysis and optimization.

[0047] The content of the feedback data set includes: base station identifier, which is used to uniquely identify the currently monitored base station; regulation strategy template identifier, which records the number of the used regulation strategy template; risk score value and risk level before execution, which record the operating state of the base station before the regulation operation; risk score value and risk level after execution, which record the operating state of the base station after the regulation operation; external environment impact factors, which record the external environment parameters that affect the operation of the base station during the regulation; execution time and interval, which record the time interval from the start of the regulation operation to the completion of the feedback collection.

[0048] S52. Calculate and analyze the feedback data set to determine the change range of the performance indicators before and after the regulation; The system calculates the performance indicator difference before and after the regulation according to the data in the feedback data set. The calculation process includes: calculating the risk score value difference, quantifying the effect of the regulation strategy by comparing the risk score values before and after execution; analyzing the change of the risk level, judging the effectiveness and accuracy of the regulation strategy according to the comparison of the risk levels before and after the regulation; adjusting the external environment impact factors: determining the influence degree of different environmental factors on the operating state of the base station through regression analysis and Bayesian optimization algorithm.

[0049] The system constructs multiple performance indicator matrices for difference analysis and adaptive adjustment of the optimization strategy. The results of the difference analysis will be directly used to optimize the parameter weights and policy templates in the strategy library.

[0050] S53. Update the parameter weights or policy templates in the strategy library according to the calculation results to improve the adaptability and accuracy of the strategy.

[0051] The system updates the parameter weights or policy templates in the policy library according to the calculation results to improve the adaptability and accuracy of the policies. This includes: parameter weight optimization, where the system dynamically adjusts the weight coefficients of each feature based on the results of difference analysis to ensure the accuracy and stability of the policy in different environments and states; policy template optimization, where the system improves the configuration and execution mechanism of the existing policy template through comprehensive analysis of historical data and feedback data; and policy library update, where the optimized parameter weights and policy templates are synchronously updated to the policy library for policy matching and execution.

[0052] Embodiment 2 As Figure 2 shown, Embodiment 2 of the present application provides an intelligent management and control platform based on base station management, including: A data acquisition module 21 for automatically acquiring the operation data of multiple communication base stations and performing formatting processing; The data acquisition module 21 collects data from multiple communication base stations in the target area in real time or periodically through a standardized data acquisition protocol and interface. The collected data includes, but is not limited to, the following types: load data, such as the number of user connections, network throughput, concurrent request quantity, etc., which reflects the communication load and service capacity of the base station; power status data, such as voltage, current, power consumption, etc., which reflects the power supply situation and equipment health status of the base station; signal quality data, such as signal-to-noise ratio, bit error rate, received signal strength indication, etc., which reflects the communication link quality and stability of the base station; and environmental parameter data, such as temperature, humidity, air pressure, wind speed, etc., which reflects the potential impact of the external environment on the operation status of the base station.

[0053] The data acquisition module 21 performs preliminary processing on the collected data, including formatting, denoising, and standardization operations. Formatting processing refers to uniformly converting data from different sources into a standard structured format for subsequent analysis. The denoising operation uses a method combining moving average and Gaussian filtering to eliminate outliers and noise interference during the acquisition process. The standardization operation adjusts the data distribution of each feature to a standard form with a mean of 0 and a standard deviation of 1 through the Z-Score standardization method, thereby eliminating the influence of different data scales on the analysis results.

[0054] The dataset that has undergone formatting processing and standardization is stored in the system database to support the calculations of the subsequent status analysis module 22 and the modeling of the risk prediction module 23.

[0055] A status analysis module 22 for performing fusion analysis and status score calculation on the structured data; The status analysis module 22 comprehensively analyzes the operation status of multiple communication base stations and calculates the status scores by calling the standardized data set generated by the data collection module 21. Its working principle includes three main links: feature extraction, fusion analysis, and score calculation.

[0056] First, the system extracts features closely related to the operation status of the base station from the standardized data set to construct a feature vector set. The extracted features include load conditions, power status, signal quality, and environmental parameters, forming a complete feature set X. The feature extraction process uses principal component analysis (PCA) and recursive feature elimination (RFE) algorithms to ensure the accuracy and effectiveness of feature selection.

[0057] In the fusion analysis stage, the status analysis module 22 comprehensively analyzes the feature vector set using deep learning algorithms. The system uses a long short-term memory network (LSTM) model to model and process time series data, thereby capturing the mutual relationships and evolution laws between different features. Through multi-level processing and non-linear mapping of the feature vectors, the system generates a status score value based on the current input data.

[0058] The score calculation process is based on the weighted sum and time series analysis of multiple features. The contribution degree of each feature is determined by the feature weight. The system automatically optimizes the weight values of each feature through the model training process to ensure the accuracy and stability of score calculation.

[0059] The generated status score value is used to identify the operation status of the current base station and provides crucial input for subsequent risk prediction. The system generates corresponding status labels according to the size of the score value and the analysis results of feature contribution degrees, and stores the score results in the database for query and analysis.

[0060] The risk prediction module 23 is used to generate warning information based on the score value and historical data; The risk prediction module 23 predicts and analyzes the potential risks of the base station by calling the status score value and corresponding status labels generated by the status analysis module 22, and combining historical operation data and external environmental impact factors. Its working principle includes three main links: score value analysis, trend prediction, and warning information generation.

[0061] In the score value analysis stage, the system first standardizes the score value provided by the status analysis module 22 and conducts a comparative analysis with historical data. By constructing a feature matrix and a time series analysis model, the system can identify the change trend of the score value and the potential relationship between features. Combining the feature contribution degree analysis and environmental impact factors, the system can accurately judge the risk level of the base station.

[0062] In the trend prediction stage, the risk prediction module 23 uses a risk prediction model based on deep learning (Long Short-Term Memory Network, LSTM) to model and analyze the input data. Through the comprehensive processing of the score value and historical data, the system can identify the time-dependent relationships and mutation trends of different features, and generate corresponding risk score values through prediction algorithms.

[0063] By analyzing and predicting the change trend of the score value, the system generates multiple risk indicators, including: the risk score value, which represents the comprehensive risk level of the current base station; the risk level, which is divided into three levels: high risk, medium risk, and low risk by comparing the risk score value with a preset threshold; the trend coefficient, which is used to reflect the change rate and direction of the score value, indicating the trend of risk increase or decrease.

[0064] In the warning information generation stage, the system generates corresponding warning information according to the analysis results of the risk score value and the risk level. The warning information includes the base station identifier, risk score value, risk level, timestamp, and feature contribution degree analysis, etc. The system stores the generated warning information in the database and transmits it through the network to the policy matching module 24 for subsequent policy retrieval and execution.

[0065] The policy matching module 24 is used to retrieve and execute regulation policies from the policy library; The core functions of the policy matching module 24 include three links: policy template retrieval, policy execution, and feedback recording.

[0066] In the policy template retrieval stage, the system automatically retrieves the regulation policy template that matches the current state from the policy library according to the warning information generated by the risk prediction module 23, including the risk score value, risk level, trend coefficient, and feature contribution degree analysis. The policy templates stored in the policy library are classified according to the risk level, including high-risk policies, medium-risk policies, and low-risk policies.

[0067] High-risk policies are mainly used to handle serious faults or abnormal states, usually including emergency power adjustment, load transfer, or activation operations of standby base stations; medium-risk policies are used to optimize the performance and stability of base stations, reducing the impact of potential risks through fine power adjustment and resource scheduling; low-risk policies are used for routine maintenance and monitoring operations to ensure the long-term stability of the system.

[0068] In the policy execution stage, the policy matching module 24 passes the selected regulatory policy template to the automated scheduling module and executes it. The system processes the selected policy template through the automated scheduling module, including: power adjustment, real-time adjustment of the transmission power of the base station according to the current risk score value and feature contribution degree to optimize the signal coverage and transmission quality; load transfer, transferring part of the load of the current base station to other neighboring base stations to relieve the overload pressure or balance resource allocation; standby activation, automatically activating the standby base station to take over the work of the faulty base station in a high-risk state, thus ensuring the continuity and stability of communication.

[0069] During the policy execution process, the system monitors the effect of the regulatory operation through a real-time feedback mechanism and makes dynamic adjustments according to the feedback information. The data generated during the execution process includes the start time and end time of the policy execution, operation results, and performance differences, etc.

[0070] In the feedback recording stage, the policy matching module 24 stores the execution results and relevant data in the database to support subsequent analysis and policy optimization by the feedback optimization module 25.

[0071] The feedback optimization module 25 is used to collect regulatory feedback data and update the parameters in the policy library.

[0072] The core functions of the feedback optimization module 25 include three links: feedback data collection, performance difference analysis, and policy library update.

[0073] In the feedback data collection stage, the system collects data on the operating state after the regulatory operation and generates a complete feedback data set. The feedback data set includes the following key contents: base station identifier, used to uniquely identify the currently monitored communication base station; regulatory policy template identifier, recording the number of the regulatory policy template used to ensure data traceability and optimization; risk score value and risk level before regulation, recording the operating state and related indicators of the base station before the regulatory operation; risk score value and risk level after regulation: recording the operating state and improvement effect of the base station after the regulatory operation; external environment impact factor, recording the external environment parameters that affect the operation of the base station during regulation, such as temperature, humidity, wind speed, etc.; execution time and interval, recording the start time and end time of the regulatory operation, as well as the collection time interval of the feedback data.

[0074] In the performance difference analysis stage, the system calculates the differences in various indicators before and after the regulatory operation through the analysis of the feedback data set, including the change in the risk score value, the improvement of the risk level, and the influence degree of the environmental factor. The system uses multi-dimensional matrix analysis and regression algorithms to quantify and compare the performance of each feature under different policy templates.

[0075] During the policy library update phase, the system optimizes and updates the parameter weights and policy templates in the policy library according to the results of performance difference analysis. Specifically, it includes: parameter weight optimization, adaptively adjusting the weight parameters according to the performance of different features during the regulation process to improve the accuracy of risk prediction and policy matching; policy template optimization, updating and reconstructing the templates in the policy library to optimize the selection mechanism and execution efficiency of regulation policies; policy library update and synchronization, versioning the optimized policy library and synchronizing it to the core database of the system to ensure the efficiency and accuracy of subsequent policy matching processes.

[0076] Corresponding to the above embodiment, an embodiment of the present invention provides a computer storage medium, including: at least one memory and at least one processor; The memory is used to store one or more program instructions; The processor is used to run one or more program instructions to execute an intelligent control method based on base station management.

[0077] Corresponding to the above embodiment, an embodiment of the present invention provides a computer-readable storage medium. The computer storage medium contains one or more program instructions, and the one or more program instructions are used to be executed by a processor to execute an intelligent control method based on base station management.

[0078] The embodiment disclosed by the present invention provides a computer-readable storage medium. Computer program instructions are stored in the computer-readable storage medium. When the computer program instructions run on a computer, the computer is enabled to execute the above-mentioned intelligent control method based on base station management.

[0079] In the embodiment of the present invention, the processor may be an integrated circuit chip with signal processing capabilities. The processor may be a general-purpose processor, a digital signal processor (DSP for short), an application-specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0080] The various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware decoding processor, or can be executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers, etc. The processor reads the information in the storage medium and combines its hardware to complete the steps of the above method.

[0081] The storage medium can be a memory, for example, it can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0082] Among them, the non-volatile memory can be read-only memory (ROM for short), programmable read-only memory (PROM for short), erasable programmable read-only memory (EPROM for short), electrically erasable programmable read-only memory (EEPROM for short), or flash memory.

[0083] The volatile memory can be random access memory (RAM for short), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM for short), dynamic random access memory (DRAM for short), synchronous dynamic random access memory (SDRAM for short), double data rate synchronous dynamic random access memory (DDR SDRAM for short), enhanced synchronous dynamic random access memory (ESDRAM for short), synchronous link dynamic random access memory (SLDRAM for short), and direct rambus random access memory (DRRAM for short).

[0084] The storage medium described in the embodiments of the present invention is intended to include but not be limited to these and any other suitable types of memories.

[0085] Those skilled in the art should be able to realize that in one or more of the above examples, the functions described in the present invention can be implemented by a combination of hardware and software. When applying software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. The computer-readable medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium accessible by a general-purpose or special-purpose computer.

[0086] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent control method based on base station management, characterized in that, It includes the following steps: S10. Obtain the operation data of multiple communication base stations in the target area, and perform structured processing on the data to generate operation data input in a standard format; S20. Conduct fusion analysis on the structured operation data, and calculate the operation status score value and corresponding status label for each base station; S30. Based on the score value and status label, construct a risk prediction model, identify the potential fault trend of each base station, and output the risk level and warning information; S40. According to the risk level and warning information, match the regulation strategy template from the strategy library and automatically execute the corresponding scheduling operation; S50. Collect the feedback results of the regulation operation, and calculate the difference in operation performance before and after regulation, which is used to optimize the parameter weights in the regulation strategy library.

2. The intelligent control method based on base station management according to claim 1, wherein Obtaining the operation data of multiple communication base stations in the target area, and performing structured processing on the data to generate operation data input in a standard format includes the following sub-steps: Automatically obtain the real-time operation data of multiple communication base stations in the target area, and the operation data includes load conditions, power supply status, signal quality, and environmental parameters; Perform formatting processing and timestamp sorting on the collected data to generate a standardized input data set; Perform denoising processing on the input data set to eliminate invalid or abnormal data to ensure data quality.

3. An intelligent control method based on base station management according to claim 1, characterized in that, Conducting fusion analysis on the structured operation data, and calculating the operation status score value and corresponding status label for each base station includes the following sub-steps: Extract multi-dimensional features of load, voltage, temperature, and signal quality from the input data set, and construct a feature vector; Input the feature vector into the fusion analysis model for comprehensive evaluation, and generate the operation status score value for each base station; Generate corresponding status labels based on the calculation results of the score value, and store the score value and status label in the database.

4. The intelligent control method based on base station management according to claim 1, characterized in that Based on the score value and status label, construct a risk prediction model, identify the potential fault trend of each base station, and output the risk level and warning information includes the following sub-steps: Use the score value and status label combined with historical data to train the risk prediction model, and generate prediction parameters for identifying fault trends; Analyze the operation status of the base station based on the trained risk prediction model, and generate the corresponding risk level; Generate warning information according to the risk level, and transmit the warning information to the strategy library for subsequent strategy matching.

5. The intelligent control method based on base station management according to claim 1, characterized in that According to the risk level and warning information, match the regulation strategy template from the strategy library and automatically execute the corresponding scheduling operation includes the following sub-steps: Retrieve the regulation strategy template that matches the current status from the strategy library according to the risk level and warning information; Execute the selected regulation strategy template through the automated scheduling module, including power adjustment, load transfer, or standby activation operations; Record the execution results of the regulation operation, and store the execution results in the database for subsequent feedback analysis.

6. The intelligent control method based on base station management according to claim 1, characterized in that Collect the feedback results of the regulation operation, and calculate the difference in operation performance before and after regulation, which is used to optimize the parameter weights in the regulation strategy library includes the following sub-steps: Collect the operation status data after the regulation operation is completed, and generate a feedback data set for analysis; Perform calculation and analysis on the feedback data set to determine the change range of performance indicators before and after regulation; Update the parameter weights or policy templates in the policy library according to the calculation results to improve the adaptability and accuracy of the policy.

7. The intelligent control method based on base station management according to claim 1, characterized in that The policy library includes a rule template module and a deep learning recommendation module. The rule template module is used to retrieve regulatory policies through label indexing, and the deep learning recommendation module is used to generate adaptive regulatory policies based on real-time data.

8. An intelligent control platform based on base station management, characterized in that, It includes: A data collection module for automatically obtaining the operation data of multiple communication base stations and performing formatting processing; A status analysis module for performing fusion analysis and status score calculation on structured data; A risk prediction module for generating warning information based on the score value and historical data; A policy matching module for retrieving and executing regulatory policies from the policy library; A feedback optimization module for collecting regulatory feedback data and updating the parameters in the policy library.

Images