Communication infrastructure intelligent operation and maintenance management system based on cloud computing

Abstract

The invention discloses a communication infrastructure intelligent operation and maintenance management system based on cloud computing. The system collects multi-source operation and maintenance data through the data acquisition module, and the multi-source operation and maintenance data is transmitted to the cloud computing center module after being processed by the data transmission and preprocessing module. The cloud computing center module performs deep analysis on data by using a plurality of intelligent algorithm-based sub-modules such as equipment state evaluation, fault prediction, resource optimization and allocation, operation and maintenance decision suggestion and the like to generate an operation and maintenance strategy, the operation and maintenance strategy is implemented by the operation and maintenance operation execution module, and the operation and maintenance condition is visually presented by means of the visual display and monitoring module. According to the method, the powerful computing power and storage capacity of cloud computing are utilized, and multiple algorithm advantages such as machine learning and time sequence analysis are combined, so that intelligence and automation of operation and maintenance management of the communication infrastructure are realized, the operation and maintenance efficiency and quality are effectively improved, and stable and reliable operation of communication service is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication infrastructure intelligent operation and maintenance management system based on cloud computing. Background Art

[0002] With the rapid development of communication technology, the scale of communication infrastructure (such as base stations, communication cables, data centers, etc.) is becoming increasingly large and widely distributed, and its operation and maintenance management faces many challenges. Traditional operation and maintenance management methods often rely on manual inspections, empirical judgments, and post-fault handling, which have problems such as low efficiency, untimely fault detection, and unreasonable resource allocation. For example, manual inspections are difficult to fully monitor the operating status of the infrastructure in real time, and potential fault hazards are easily missed; when faced with complex network environments and diverse equipment failures, there is a lack of effective intelligent analysis methods to quickly locate problems and provide accurate solutions. Therefore, there is an urgent need for an intelligent operation and maintenance management system based on cloud computing, which can achieve efficient and accurate operation and maintenance management of communication infrastructure with the help of the powerful computing and storage capabilities of cloud computing and intelligent algorithms. Summary of the invention

[0003] The main purpose of the present invention is to provide a cloud computing-based intelligent operation and maintenance management system for communication infrastructure to achieve efficient and accurate operation and maintenance management of communication infrastructure.

[0004] To achieve the above objectives, this cloud computing-based communication infrastructure intelligent operation and maintenance management system is mainly composed of the following core modules:

[0005] Data acquisition module:

[0006] By deploying various sensors (temperature sensors, humidity sensors, voltage sensors, flow sensors, performance monitoring probes, etc.) on key equipment of the communication infrastructure (base station equipment, servers, switches, transmission equipment, etc.), the operating parameters of the equipment (such as equipment temperature, humidity environment, power supply voltage, network traffic, processor usage, etc.) and status information (equipment online and offline status, alarm information, etc.) are collected in real time. At the same time, the network interface is used to collect external data related to the communication infrastructure, such as geographic environment data (weather conditions and topography at the location of the base station), network topology information (connection relationships between various devices, link bandwidth, etc.) and user usage data (communication traffic requirements of users in various regions, service usage frequency, etc.), to provide a comprehensive and multi-dimensional data foundation for subsequent operation and maintenance analysis.

[0007] Data transmission and preprocessing module:

[0008] The collected data is transmitted to the cloud computing platform in real time using high-speed and stable communication protocols (such as 5G, fiber optic networks, etc.). During the transmission process, the data is pre-processed, including data format unification (converting data from different sources and formats into a standard format that the system can recognize and process), data cleaning (removing duplicate, erroneous, and invalid data records), and data compression (reasonably compressing massive data without affecting the accuracy of subsequent analysis to improve transmission and storage efficiency) to ensure that the data quality entering the cloud computing platform is reliable and convenient for subsequent processing.

[0009] Cloud computing center module:

[0010] As the core computing and storage platform of the entire system, the cloud computing center has a powerful computing resource pool (including CPU, GPU and other computing units, which can be flexibly allocated according to task requirements) and massive data storage capabilities (using a distributed storage system to ensure high reliability and scalability of data). After receiving the pre-processed data, the module implements intelligent operation and maintenance management of the communication infrastructure based on multiple functional sub-modules, mainly including:

[0011] Equipment status assessment submodule: Use the status assessment algorithm based on machine learning, take the historical normal operation data and real-time collected data as training samples, and build the equipment status assessment model (for example, use support vector machine, random forest and other algorithms to build a classification model to determine whether the equipment is in normal, potential failure, failure and other different status categories). By extracting and analyzing the features of the current operating parameters and status information of the equipment, inputting them into the model, and outputting the current status assessment results of the equipment, the real-time monitoring of the health status of the equipment can be achieved.

[0012] Fault prediction submodule: Based on time series analysis algorithms (such as ARIMA model, Prophet model, etc.) and deep learning algorithms (such as long short-term memory network LSTM, etc.), the historical operation data of the equipment is mined and analyzed to capture the time series characteristics and rules in the data, and predict the time point, fault type, and fault probability of possible future equipment failures, so as to provide early warning of potential failure risks so that operation and maintenance personnel can prepare in advance.

[0013] Resource optimization and allocation submodule: Combining linear programming algorithms and genetic algorithms, with the goal of meeting communication service quality requirements (such as ensuring network bandwidth requirements of users in various regions, reducing communication latency, etc.) and minimizing operation and maintenance costs (such as equipment energy consumption, manpower allocation costs, etc.), according to the real-time load of each communication infrastructure equipment (reflected by collected network traffic, equipment utilization rate and other data), geographical location distribution and resource reserves, formulate the optimal resource allocation plan (such as adjusting base station power, allocating server resources, dispatching operation and maintenance personnel, etc.) to achieve efficient use and reasonable allocation of resources.

[0014] Operation and maintenance decision-making recommendation submodule: Based on the results of equipment status assessment, fault prediction and resource optimization and allocation, the algorithm based on rule engine and expert system is used to comprehensively consider multiple factors such as the overall operation status of the communication infrastructure, operation and maintenance strategy, historical fault handling experience, etc., to generate targeted operation and maintenance decision-making recommendations (such as immediately repairing a faulty equipment, focusing on monitoring potential faulty equipment, adjusting resource allocation according to business peaks, etc.), provide clear and actionable guidance for operation and maintenance personnel, and assist them in making quick and accurate operation and maintenance decisions.

[0015] Operation and maintenance execution module:

[0016] Receive operation and maintenance decision suggestions from the cloud computing center module, and convert them into specific operation and maintenance operation instructions (such as equipment restart, parameter adjustment, maintenance task dispatch, etc.), and issue instructions to the corresponding equipment or operation and maintenance personnel through automated operation and maintenance tools (such as Ansible, SaltStack, etc.) or interacting with the operation and maintenance personnel's mobile APP, so as to realize the automated execution of communication infrastructure operation and maintenance operations and improve operation and maintenance efficiency and accuracy.

[0017] Visual display and monitoring module:

[0018] The various operation and maintenance information (equipment status data, fault prediction results, resource allocation plans, operation and maintenance decision suggestions, etc.) obtained by the cloud computing center module analysis and processing are visualized, through intuitive charts (such as bar charts showing equipment utilization rate, line charts showing fault trends, maps marking base station locations and status, etc.), graphics (equipment topology diagrams showing the connection and operating status of each device) and real-time monitoring interfaces (real-time updates of equipment alarm information, operation and maintenance operation execution progress, etc.), it is convenient for operation and maintenance managers to fully and intuitively understand the overall operation and maintenance status of communication infrastructure, facilitate timely discovery of problems, tracking of operation and maintenance progress, and evaluation of operation and maintenance results.

[0019] The cloud computing-based communication infrastructure intelligent operation and maintenance management system provided by the present invention:

[0020] 1. Efficient processing and storage based on cloud computing: With the powerful computing and storage capabilities of the cloud computing platform, it can process massive amounts of communication infrastructure operation data in real time, breaking through the limitations of traditional single-machine processing capabilities and ensuring the efficient and stable operation of the system when facing large-scale, distributed communication infrastructure.

[0021] 2. Accurate operation and maintenance decisions driven by intelligent algorithms: Through the comprehensive application of multiple intelligent algorithms, the entire process from equipment status assessment, fault prediction to resource optimization and allocation and operation and maintenance decision recommendations is intelligentized, overcoming the shortcomings of traditional operation and maintenance relying on manual experience and judgment, improving the timeliness of fault discovery, the accuracy of operation and maintenance decisions, and the rationality of resource allocation, effectively reducing operation and maintenance costs and improving the quality of communication services.

[0022] 3. Collaboration between visual monitoring and automated execution: Visual display allows operation and maintenance personnel to grasp the overall situation in real time, while automated operation and maintenance operation execution quickly transforms decisions into actual actions. The two work together to greatly improve the efficiency and convenience of operation and maintenance management, ensuring the smooth progress of the entire operation and maintenance process. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0024] Figure 1 It is a principle block diagram of a communication infrastructure intelligent operation and maintenance management system based on cloud computing in one embodiment of the present application;

[0025] Figure 2 This is a block diagram of the cloud computing center module in the embodiment of the present application;

[0026] Figure 3 It is a flowchart based on the time series analysis algorithm and the deep learning algorithm in the embodiment of the present application;

[0027] Figure 4 It is a flowchart of the basic construction of the linear programming algorithm in the embodiment of the present application;

[0028] Figure 5 is a flow chart of an algorithm based on a rule engine and an expert system in an embodiment of the present application;

[0029] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0032] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0033] In one embodiment, if Figure 1 As shown, the present application discloses a cloud computing-based communication infrastructure intelligent operation and maintenance management system. The high-definition adaptive video surveillance system is mainly composed of the following core modules:

[0034] Data acquisition module:

[0035] By deploying various sensors (temperature sensors, humidity sensors, voltage sensors, flow sensors, performance monitoring probes, etc.) on key equipment in the communication infrastructure (base station equipment, servers, switches, transmission equipment, etc.), we can comprehensively collect equipment operating parameters (such as equipment temperature, humidity environment, power supply voltage, network traffic, processor usage, etc.) and status information (equipment online and offline status, alarm information, etc.). At the same time, we use network interfaces to collect external data related to the communication infrastructure, such as geographic environment data (weather conditions and topography at the base station location), network topology information (connection relationships between various devices, link bandwidth, etc.) and user usage data (communication traffic requirements of users in various regions, service usage frequency, etc.), to lay a solid foundation for subsequent data processing and operation and maintenance analysis.

[0036] Data transmission and preprocessing module:

[0037] The collected data is transmitted to the cloud computing platform in real time using high-speed and stable communication protocols (such as 5G, fiber optic networks, etc.). During the transmission process, the module performs the following preprocessing operations:

[0038] Data format unification: Data from different sensors, different devices, and different formats (for example, some sensors output data in binary format, some in XML format, etc.) are converted into a standardized format that the system can process. JSON format is usually used for unification to facilitate subsequent data parsing and processing operations.

[0039] Data cleaning: Remove duplicate, erroneous and invalid data records through set rules and algorithms. For example, for the data collected by the temperature sensor, if the temperature values ​​collected at multiple consecutive time points exceed the reasonable fluctuation range of the normal operating temperature range of the equipment (the upper and lower thresholds can be set according to the equipment technical manual and historical data), and do not conform to the normal temperature change law (such as a large jump change in a short period of time), then these data may be judged as erroneous data and are removed; for repeatedly collected data, only the valid records collected for the first time are retained to improve the quality and availability of the data.

[0040] Data compression: Under the premise of ensuring that the accuracy of subsequent analysis is not affected, use appropriate data compression algorithms (such as GZIP compression algorithm) to reasonably compress massive data. The GZIP compression algorithm is based on the LZ77 algorithm and Huffman coding. By finding repeated strings and representing them with shorter codes, it effectively reduces the storage space occupied by data, improves data transmission efficiency, and enables data to be transmitted to the cloud computing platform more quickly.

[0041] Cloud computing center module:

[0042] Reference Figure 2 As the core computing and storage platform of the system, the cloud computing center relies on a powerful computing resource pool (covering multiple computing units such as CPU and GPU, which can be flexibly allocated on demand) and massive data storage capabilities (using a distributed storage system to ensure high reliability and scalability of data), and uses multiple functional sub-modules to achieve intelligent operation and maintenance management. The data processing algorithms of each sub-module are as follows:

[0043] Equipment status assessment submodule:

[0044] Using the machine learning-based state assessment algorithm, specifically using the support vector machine (SVM) and random forest algorithms to build the equipment state assessment model. The following is the detailed algorithm flow:

[0045] Data preparation and feature extraction: Collect a large amount of historical normal operation data and a certain amount of labeled fault data (for example, clearly mark the device status as normal, potential fault, fault, etc.), and extract key features from these data as model input. For communication equipment, the extracted features may include parameters such as device temperature, voltage, network traffic, and performance indicators of key components (such as CPU usage and memory usage of the server). These features are combined into feature vectors. For example, for a server device, the feature vector can be expressed as temperature value, voltage value, network traffic, usage rate, and memory usage rate.

[0046] Model training:

[0047] Support Vector Machine (SVM): The extracted feature vectors and the corresponding equipment status labels are input into the SVM model as training samples. The SVM algorithm aims to find an optimal hyperplane so that data samples of different status categories (such as normal and faulty) can be separated to the greatest extent in the feature space. The original feature space is mapped to a higher-dimensional space through the kernel function (common kernel functions include linear kernel, polynomial kernel, Gaussian kernel, etc., and the appropriate kernel function is selected according to the distribution characteristics of the data. For example, the Gaussian kernel function is usually used for nonlinearly separable data) to better find a linearly separable hyperplane, and then train an SVM model that can accurately classify equipment status.

[0048] Random Forest: A random forest model is also constructed using feature vectors and device status labels as training samples. Random Forest is an ensemble learning algorithm based on decision trees. It generates multiple sub-datasets from the original training data set through random sampling (sampling with replacement, i.e. Bootstrap sampling), and then builds a decision tree for each sub-dataset. In the process of building a decision tree, a portion of features are randomly selected from all features each time (for example, the number of features is set to the square root of the total number of features) to determine the best split node, thereby reducing the variance of the model and improving the generalization ability. Finally, multiple decision trees are combined to determine the device status category through a voting mechanism (for classification problems, each tree predicts a category, and the category with the most votes is the final prediction result).

[0049] Status assessment: The same feature extraction operation is performed on the equipment operating parameter data collected in real time to generate a real-time feature vector, which is input into the trained SVM and random forest models. The two models respectively output the prediction results of the equipment status (such as normal, potential fault, fault, etc.). Then, the prediction results of the two models are comprehensively considered (for example, a weighted average method can be used to assign corresponding weights according to the accuracy of the model in historical verification. If the SVM model has a higher accuracy, a relatively higher weight is assigned). Finally, the current status assessment result of the equipment is determined to achieve real-time monitoring of the equipment health status.

[0050] Fault prediction submodule:

[0051] Reference Figure 3 Based on the combination of time series analysis algorithms (such as ARIMA model, Prophet model, etc.) and deep learning algorithms (such as long short-term memory network LSTM, etc.), the historical operation data of the equipment is deeply mined and analyzed. The specific algorithm process is as follows:

[0052] Time series analysis (taking the ARIMA model as an example):

[0053] Data stationarity test: For the historical time series data of a key operating parameter of the equipment (such as the CPU usage of the server), a stationarity test is first performed. The commonly used method is the ADF test (Augmented Dickey-Fuller Test). If the data is non-stationary, a differential operation (such as the first-order difference, calculating the difference between the data of two adjacent time points to make the data stable) is used to make it meet the stationarity requirements.

[0054] Model order determination: Determine the parameters of the ARIMA model, p (autoregression order, indicating how many lagged self-data are used to predict the current data), q (moving average order, that is, how many lagged prediction errors are used to predict the current data), and d (difference order, the number of differences in the previous stabilization process). Determine the optimal order of the model by observing the autocorrelation function (ACF) graph and the partial autocorrelation function (PACF) graph, and combining the information criterion (such as the AIC criterion, select the p, q, d combination that minimizes the AIC value).

[0055] Model training and prediction: The ARIMA model is constructed and trained according to the determined order of the time series data after stabilization. After the trained model is obtained, the latest data at several time points (determined by the number of lagged data required by the model) are input to predict the changing trend of the equipment's operating parameters in the future and the time points when abnormalities may occur, which can be used as a reference for fault prediction.

[0056] Deep learning analysis (taking LSTM network as an example):

[0057] Data preprocessing and feature engineering: Normalize the historical time series data of multiple relevant operating parameters of the device (such as CPU usage, memory usage, network traffic, etc.) (the commonly used normalization method is Min-Max normalization, which maps the data to the interval of 0,1), and then divide it according to the time window (for example, taking the data every 10 minutes as a time window) to form an input sample sequence. The output corresponding to each input sample sequence is the device operating parameter value of the next time window, so as to construct the training data set and test data set.

[0058] Network construction and training: Construct an LSTM network structure, including an input layer, several hidden layers (the number of LSTM units in the hidden layer is reasonably set according to data complexity and computing resources, such as 64, 128, etc.) and an output layer. During the training process, use a suitable optimization algorithm (such as the Adam optimizer) and a loss function (such as the mean square error loss function, which is used to measure the difference between the predicted value and the true value) to train the network. Through multiple iterations, adjust the network's weight parameters so that the network can accurately capture the complex dependencies and change patterns of the equipment operating parameters in the time series.

[0059] Fault prediction: The latest time series data collected in real time is processed in the same preprocessing method and then input into the trained LSTM network. The network outputs the predicted values ​​of various operating parameters of the equipment at multiple time points in the future. Combined with the set fault threshold (set according to the normal operating range of the equipment and historical fault data, such as CPU usage exceeding 90% is considered a possible fault), the type and probability of possible future faults of the equipment are judged to further enrich the results of fault prediction.

[0060] Comprehensive prediction results: The prediction results of the ARIMA model and the LSTM network are integrated, for example, through weighted averaging or rule-based integration (the integration rules are set according to the prediction accuracy of different algorithms in different scenarios) to obtain the final fault prediction results, including key information such as the time point when the equipment may fail in the future, the fault type, and the failure probability, so as to provide early warnings to operation and maintenance personnel so that they can be prepared for the response.

[0061] Resource optimization and allocation submodule:

[0062] Reference Figure 4 , combining linear programming algorithm and genetic algorithm to formulate the optimal resource allocation plan. The specific algorithm is implemented as follows:

[0063] Linear programming algorithm basic construction:

[0064] Objective function setting: To meet the communication service quality requirements (such as ensuring the network bandwidth requirements of users in each area, the average download rate of users in the area covered by each base station can be set to not be lower than a certain threshold; controlling communication delay, requiring the end-to-end communication delay not to exceed the specified value, etc.) and minimizing operation and maintenance costs (covering equipment energy consumption, manpower deployment costs, etc., for example, quantifying the power consumption cost of the base station and the working hours of the operation and maintenance personnel, etc.), construct the objective function. For example, the objective function can be expressed as Among them C i represents the operation and maintenance cost coefficient of the i-th equipment, x i is the deployment quantity or state variable of the corresponding equipment, H j represents the labor cost coefficient of the j-th type of operation and maintenance personnel, y j To allocate the corresponding number of personnel, the optimal solution for resource allocation is found by optimizing this objective function.

[0065] Determination of constraints: Constraints are set based on the actual situation of the communication infrastructure, including limitations on equipment resources (such as the maximum available amount of resources such as the server's CPU and memory; the limited power adjustment range of base stations, etc.), differences in demand in geographical regions (users in different regions have different demands for network bandwidth and service quality), and the association between devices (such as the configuration adjustment of certain devices requires simultaneous consideration of the carrying capacity of other devices connected to it, etc.). For example, for server resource allocation, the constraints can be expressed as Where R i represents the amount of resources (such as CPU resources) that the i-th server can provide, x i is the number of servers deployed, D k It represents the total demand of users in the kth region for the resource, so as to ensure that resource allocation meets the needs of users in each region.

[0066] Genetic algorithm optimization solution:

[0067] Coding and population initialization: Encode the resource allocation plan (such as the power setting of each base station, the resource allocation of each server, the deployment arrangement of operation and maintenance personnel, etc.), usually using binary coding or real number coding (select according to the characteristics of the problem, for the allocation plan of continuous variables, real number coding is more suitable). Randomly generate the initial population, and each individual in the population represents a possible resource allocation plan.

[0068] Definition of fitness function: The objective function value constructed by linear programming is used as the fitness function. That is, the higher the fitness value (the smaller the objective function value, the lower the operation and maintenance cost and the better the service quality requirements are met), the better the resource allocation plan of the individual.

[0069] Genetic operations: Through selection operations (such as roulette selection, tournament selection, etc., excellent individuals are selected according to the fitness ratio of individuals to enter the next generation of the population), crossover operations (for encoded individuals, some coding fragments are exchanged according to a certain crossover probability to simulate gene recombination and generate new individuals) and mutation operations (randomly change certain coding bits of individuals with a smaller mutation probability to increase the diversity of the population and avoid falling into the local optimal solution), the population is continuously iterated and evolved, so that the individuals in the population gradually approach the optimal resource allocation plan.

[0070] Solution determination: After multiple rounds of iterative optimization of the genetic algorithm, when the fitness value of the population converges or reaches the preset number of iterations, the individual with the highest fitness is selected from the final population and decoded to obtain the optimal resource allocation plan, so as to achieve efficient utilization and reasonable allocation of resources to cope with the changes in resource demand of communication infrastructure in different operating scenarios.

[0071] Operation and maintenance decision suggestion submodule:

[0072] Reference Figure 5 , using algorithms based on rule engines and expert systems, comprehensively considering multiple factors such as the overall operation status of the communication infrastructure, operation and maintenance strategies, and historical fault handling experience, to generate targeted operation and maintenance decision suggestions. The specific algorithm process is as follows:

[0073] Rule engine construction: Based on the expertise in the field of operation and maintenance and previous operation and maintenance practice experience, a series of rules are formulated, which are presented in the form of "if...then..." For example, "If the status of a base station equipment is evaluated as a potential fault and the probability of fault prediction exceeds 30%, it is recommended to immediately arrange for operation and maintenance personnel to carry relevant detection tools to go to the site for detailed inspection"; "If the resource allocation plan involves cross-regional server resource adjustment and the adjustment range exceeds 50%, it is necessary to notify the operation and maintenance person in charge of the relevant area in advance for coordination and communication", etc. These rules are stored in the rule base, and the rule engine is built so that it can match and infer rules based on the real-time incoming equipment status evaluation results, fault prediction information, resource allocation plan and other data.

[0074] Expert system integration: The expert system consists of two parts: the knowledge base and the inference engine. The knowledge base stores a large number of historical fault handling cases, operation and maintenance specifications of different equipment, and the overall architecture and operation principles of the communication infrastructure, etc., which are organized and managed in a structured form (such as knowledge graphs, database tables, etc.). The inference engine uses logical reasoning methods (such as forward reasoning, reverse reasoning, etc., and selects the appropriate reasoning method according to the actual situation) to conduct a comprehensive analysis based on the information passed in by the rule engine and the knowledge in the knowledge base. For example, when encountering a new type of fault situation, by searching the knowledge base for similar fault handling experience and corresponding operation and maintenance strategies, combined with the current specific equipment status and resource conditions, a more practical operation and maintenance decision recommendation is generated.

[0075] Suggestion generation and output: The rule engine works in collaboration with the expert system. After receiving relevant data from other sub-modules, it quickly performs rule matching, knowledge query and logical reasoning, and ultimately generates specific operation and maintenance decision suggestions, such as repair suggestions for a certain faulty equipment, key monitoring tips for potential faulty equipment, and specific operational guidance for adjusting resource allocation according to business peaks. These suggestions are then output to the operation and maintenance execution module to provide clear and actionable guidance for operation and maintenance personnel, helping them to quickly make accurate operation and maintenance decisions.

[0076] Operation and maintenance execution module:

[0077] Receive operation and maintenance decision suggestions from the cloud computing center module and convert them into specific operation and maintenance operation instructions (such as equipment restart, parameter adjustment, maintenance task dispatch, etc.). With the help of automated operation and maintenance tools (such as Ansible, SaltStack, etc.) or interacting with the mobile APP of the operation and maintenance personnel, the instructions are issued to the corresponding equipment or operation and maintenance personnel to realize the automated execution of communication infrastructure operation and maintenance operations, and improve the efficiency and accuracy of operation and maintenance.

[0078] Visual display and monitoring module:

[0079] The various operation and maintenance information (equipment status data, fault prediction results, resource allocation plans, operation and maintenance decision suggestions, etc.) obtained by the cloud computing center module analysis and processing are visualized, through intuitive charts (such as bar charts showing equipment utilization rate, line charts showing fault trends, maps marking base station locations and status, etc.), graphics (equipment topology diagrams showing the connection and operating status of each device) and real-time monitoring interfaces (real-time updates of equipment alarm information, operation and maintenance operation execution progress, etc.), it is convenient for operation and maintenance managers to fully and intuitively understand the overall operation and maintenance status of communication infrastructure, facilitate timely discovery of problems, tracking of operation and maintenance progress, and evaluation of operation and maintenance results.

[0080] The present embodiment is a cloud computing-based intelligent operation and maintenance management system for communication infrastructure. The system collects multi-source operation and maintenance data through the data acquisition module, and transmits it to the cloud computing center module after being processed by the data transmission and preprocessing module. The cloud computing center module uses multiple sub-modules based on intelligent algorithms such as equipment status assessment, fault prediction, resource optimization and allocation, and operation and maintenance decision suggestions to conduct in-depth analysis of the data, generate operation and maintenance strategies, and then implement them by the operation and maintenance operation execution module, while intuitively presenting the operation and maintenance situation with the help of the visualization display and monitoring module. The present invention utilizes the powerful computing power and storage capacity of cloud computing, combined with the advantages of various algorithms such as machine learning and time series analysis, to realize the intelligent and automated operation and maintenance management of communication infrastructure, effectively improve the efficiency and quality of operation and maintenance, and ensure the stable and reliable operation of communication services.

[0081] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A cloud computing-based communication infrastructure intelligent operation and maintenance management system, characterized in that: include: The data acquisition module is used to collect equipment operating parameters, status information and external related data by deploying various sensors on key equipment of the communication infrastructure and using network interfaces; The data transmission and preprocessing module is used to transmit the collected data to the cloud computing platform using a high-speed and stable communication protocol, and perform preprocessing operations such as format unification, cleaning, and compression on the data; The cloud computing center module includes an equipment status assessment submodule, a fault prediction submodule, a resource optimization and allocation submodule, and an operation and maintenance decision suggestion submodule. It uses a machine learning-based status assessment algorithm, a time series analysis algorithm and a deep learning algorithm, a linear programming algorithm and a genetic algorithm, and an algorithm based on a rule engine and an expert system to achieve status assessment, fault prediction, resource optimization and allocation, and operation and maintenance decision suggestions for communication infrastructure equipment; An operation and maintenance execution module is used to receive the operation and maintenance decision suggestions from the cloud computing center module and convert them into specific operation and maintenance operation instructions, and realize the automatic execution of the operation and maintenance operations through the automated operation and maintenance tools or by interacting with the mobile terminal APP of the operation and maintenance personnel; The visualization and monitoring module is used to visualize various operation and maintenance information obtained by the cloud computing center module analysis and processing, so that the operation and maintenance management personnel can intuitively understand the overall operation and maintenance status of the communication infrastructure.

2. According to claim 1, a cloud computing-based communication infrastructure intelligent operation and maintenance management system is characterized in that: The equipment status assessment submodule adopts machine learning-based status assessment algorithms including support vector machine, random forest and other algorithms. The equipment status assessment model is constructed with historical normal operation data and real-time collected data as training samples. The current status assessment result of the equipment is output by extracting and analyzing features of the current operating parameters and status information of the equipment.

3. According to the cloud computing-based communication infrastructure intelligent operation and maintenance management system of claim 1, it is characterized in that: The fault prediction submodule uses time series analysis algorithms including ARIMA model, Prophet model, etc., and deep learning algorithms including long short-term memory network LSTM, etc., which predicts the time point, fault type, and fault probability of possible future equipment failures by mining and analyzing the historical operation data of the equipment.

4. According to claim 1, a cloud computing-based communication infrastructure intelligent operation and maintenance management system is characterized in that: The resource optimization and allocation submodule combines linear programming algorithm and genetic algorithm to meet the communication service quality requirements and minimize operation and maintenance costs, and formulates the optimal resource allocation plan according to the real-time load, geographical location distribution and resource reserves of each communication infrastructure equipment.

5. According to claim 1, a cloud computing-based communication infrastructure intelligent operation and maintenance management system is characterized in that: The operation and maintenance decision suggestion submodule uses an algorithm based on a rule engine and an expert system, comprehensively considers multiple factors such as the overall operation status of the communication infrastructure, operation and maintenance strategies, historical fault handling experience, etc., and generates targeted operation and maintenance decision suggestions.

6. The cloud computing-based communication infrastructure intelligent operation and maintenance management system according to claim 1, characterized in that: The data transmission and preprocessing module adopts high-speed and stable communication protocols including 5G, optical fiber network, etc. When unifying the data format, data from different sources and in different formats are converted into JSON format, and data compression adopts a suitable compression algorithm such as GZIP compression.

Images