Building intelligent operation and maintenance management system and method based on big data analysis

Abstract

The invention discloses a building intelligent operation and maintenance management system and method based on big data analysis, and relates to the field of building intelligent operation and maintenance, and the system comprises a data collection layer including high-precision multi-type sensors and adaptive collection and transmission, a data storage layer including distributed storage and high-redundancy cache, and a big data analysis layer including data cleaning and data storage. The operation and maintenance decision-making layer is used for generating a strategy according to analysis and performing dynamic adjustment; the user interaction layer is used for performing visual interface and convenient feedback; the intelligent early warning module is used for performing dynamic threshold and accurate early warning; the analog simulation optimization module is used for performing model simulation and VR / AR auxiliary decision making; and the operation and maintenance management level and the energy utilization efficiency are improved. The system integrates building data, monitors equipment in real time, accurately predicts faults and energy consumption, optimizes operation and maintenance decisions, improves the management efficiency and the energy utilization rate, reduces the cost, and improves the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of building intelligent operation and maintenance management, and in particular to a building intelligent operation and maintenance management system and method based on big data analysis. Background Art

[0002] With the acceleration of urbanization, the scale of buildings is expanding, and their functions are becoming increasingly complex. Building operation and maintenance management is facing severe challenges. In modern buildings, various equipment systems (such as electrical systems, HVAC systems, water supply and drainage systems, etc.) are interrelated and work together. Their operating status and energy consumption directly affect the comfort, safety and economy of the building.

[0003] Traditional building operation and maintenance management methods mainly rely on manual inspections and experience judgment, which has many limitations. On the one hand, manual inspections are difficult to achieve real-time and comprehensive monitoring of building equipment and systems, and problems can often only be discovered after they occur, resulting in untimely maintenance, affecting the normal use of the building, and may even cause safety accidents. For example, if electrical equipment failures are not discovered in time, they may cause serious consequences such as fires. On the other hand, manual evaluation and decision-making lack accuracy and scientificity, making it difficult to carry out refined energy management and equipment maintenance scheduling based on the actual operation of the building.

[0004] At the same time, with the rapid development of information technology, a huge amount of data has been generated in buildings, but this data has not been effectively integrated and utilized. The data of various equipment systems are often stored in a scattered manner with inconsistent formats, forming "data islands" that cannot provide comprehensive and accurate support for operation and maintenance management. Moreover, traditional data analysis methods are difficult to handle large-scale, multi-source heterogeneous data and cannot tap the potential value behind the data.

[0005] In terms of energy management, high energy consumption in buildings is a global problem. According to statistics, building energy consumption accounts for a considerable proportion of the world's total energy consumption, and existing operation and maintenance management methods are difficult to achieve efficient use of energy. Lack of in-depth analysis and accurate prediction of energy consumption data makes it impossible to detect energy consumption anomalies in a timely manner and take effective optimization measures.

[0006] In addition, building operation and maintenance management involves multiple professional fields, and there is a lack of effective coordination mechanisms between the fields, resulting in poor information transmission and low work efficiency. For example, equipment maintenance personnel and energy management personnel may not be able to work together due to information asymmetry, affecting the overall operation and maintenance effect.

[0007] In summary, there is an urgent need for an innovative building intelligent operation and maintenance management system and method that can integrate various types of data in the building, use big data analysis technology to achieve real-time monitoring of building equipment and systems, fault prediction, energy consumption optimization and intelligent decision-making, improve the efficiency and quality of operation and maintenance management, reduce building energy consumption, and ensure the safety, comfort and sustainable operation of the building. The present invention is proposed based on this background and aims to provide a new solution for building operation and maintenance management. Summary of the invention

[0008] The present invention proposes a building intelligent operation and maintenance management system and method based on big data analysis to solve the problems mentioned in the above-mentioned prior art.

[0009] In order to achieve the above objectives, the present invention adopts the following technical solutions: A building intelligent operation and maintenance management system based on big data analysis, comprising: The data collection layer is equipped with various types of sensors, which automatically identify and adapt to the equipment interfaces in the building, realize plug-and-play data collection, and have self-calibration function. The data is encrypted and transmitted via wired or wireless methods; The data storage layer consists of a distributed file system and a relational database. The distributed file system stores raw data and uses distributed cache technology. The relational database stores structured data and has transaction processing, backup, and recovery functions. The big data analysis layer includes a data cleaning module, which removes noisy and bad data based on a combination of rules and machine learning. The algorithm library contains a variety of analysis algorithms and can be dynamically updated and expanded. The model training module uses machine learning technology to train the equipment failure prediction model based on historical data. The formula is: ,in is the failure probability, X is the operating parameter vector, a and b are the fitting coefficients; the energy consumption optimization model formula is: , where E opt To optimize energy consumption, k1-k7 are fitting coefficients, T represents the equipment operating temperature, N represents the number of equipment, V represents wind speed related parameters, I represents the equipment operating current, Tenv represents the ambient temperature, H represents the ambient humidity, AQ represents the air quality, and cross-validation technology is used during training to improve the generalization ability; The operation and maintenance decision-making layer generates operation and maintenance strategies based on the results of big data analysis, including equipment maintenance, energy management, and personnel scheduling strategies. They are dynamically adjusted based on real-time data. When a sudden equipment failure occurs, they are adjusted and notified within 3 minutes. Notifications and instructions are pushed to the operation and maintenance personnel's terminals. Strategies have priority and traceability, and are optimized in combination with the expert experience library when generated. The user interaction layer provides a visual interface through which operation and maintenance personnel can view system operation data, receive notifications and instructions, and provide feedback on task execution. Users can access the interface through computers and mobile phones. Intelligent early warning module: Real-time monitoring of key system data, dynamic setting of early warning thresholds based on system characteristics and safety standards, and the use of multi-dimensional data analysis technology, taking into account historical data fluctuations, real-time data change rates, and equipment operating status correlation factors. When the data exceeds the early warning threshold, early warning information is sent to operation and maintenance personnel through sound and light alarms, SMS push, and system pop-up windows. The early warning information includes abnormal data details, possible cause analysis, and preliminary solution suggestions, and automatically generates early warning event records; Simulation and optimization module: Based on the digital model of the building system and combined with the results of big data analysis, different operation and maintenance strategies and equipment parameter adjustments are simulated. Virtual reality and augmented reality technologies are used to provide operation and maintenance personnel with an immersive simulation scene experience. Operation and maintenance decisions are optimized based on simulation results, and the best solution is selected for execution.

[0010] Furthermore, the sensors of the data acquisition layer dynamically adjust the acquisition frequency according to the importance of different areas and equipment in the building, and the acquisition frequency of key areas and key equipment is increased. The specific adjustment method is: the system divides the areas according to the building layout and the importance level of the equipment in advance, and sets a basic acquisition frequency for each area and equipment. When it is detected that the operating data of a certain area or equipment fluctuates abnormally or approaches the warning threshold, the acquisition frequency of the area or equipment is automatically increased to 2-5 times the basic frequency. After 10-30 minutes of continuous acquisition, the acquisition frequency is restored or further adjusted according to the stability of the data; the sensor adopts ceramic packaging, electromagnetic shielding cover and other designs.

[0011] Furthermore, the relational database of the data storage layer supports distributed transaction processing and can automatically optimize the data storage structure according to the frequency of data access. In terms of distributed transaction processing, the two-phase commit protocol 2PC or the three-phase commit protocol 3PC is adopted to ensure that data operations on multiple nodes are all successful or all rolled back.

[0012] Furthermore, the algorithm library of the big data analysis layer includes an anomaly detection algorithm based on deep learning, which identifies abnormal data patterns in building systems and warns of potential risks in advance. The algorithms in the algorithm library can be customized according to user needs and actual application scenarios. The deep learning anomaly detection algorithm adopts a convolutional neural network CNN or a recurrent neural network RNN ​​architecture. The training data includes normal operating data and simulated abnormal data. By learning the data, the algorithm can automatically extract data features and identify deviations from normal patterns.

[0013] Furthermore, when generating the operation and maintenance strategy, the operation and maintenance decision-making layer automatically adjusts the energy management strategy according to the energy consumption demand and equipment operation characteristics in different seasons and time periods. For the work performance evaluation of the operation and maintenance personnel, an evaluation model is established based on indicators such as the timeliness and accuracy of task completion, and the effectiveness of feedback issues. The performance score is calculated using the weighted average method, and personnel scheduling is optimized based on the score. Tasks are assigned preferentially to operation and maintenance personnel with higher performance, and personnel with lower performance are trained or adjusted to other positions.

[0014] Furthermore, a method for applying a building intelligent operation and maintenance management system based on big data analysis includes the following steps: S1. Data collection steps: The data collection layer collects data on equipment operation, environment and personnel activities in the building at the set frequency and accuracy. The sensor is self-calibrated before collection, and automatically re-collects or adjusts parameters in case of abnormality. The sensor automatically adapts to the device interface to achieve plug-and-play collection, and dynamically allocates collection resources according to different areas of the building and the importance of equipment; S2, data transmission and storage step: the collected data is encrypted and adaptively transmitted and then stored in the data storage layer. Multi-link redundancy is supported during the transmission process, and automatic switching is performed when the main link fails. Classification and management of data is performed during storage; S3, data analysis step: The big data analysis layer reads data from the storage layer, cleans it, and uses the algorithm library to analyze it, mining data relationships, using the formula: Calculate associations, where X is the operating parameter vector and Y is the energy consumption. Automatically update the algorithm library when new relationships are discovered, and use deep learning anomaly detection algorithms to identify anomalies. S4, model training and prediction steps: The model training module trains the model based on historical and analysis results, cross-validates to improve generalization ability, uses the trained model to predict equipment operation and energy consumption, and the prediction results are used to guide operation and maintenance decisions; S5, Operation and maintenance decision generation step: The operation and maintenance decision layer generates strategies based on the analysis and prediction results, optimizes the expert experience database, and pushes them to the operation and maintenance personnel terminal. If there is no timely response, the automatic upgrade warning will notify the management personnel. The strategy includes priority and traceability, and automatically adjusts the strategy according to different situations; S6, Operation and maintenance execution and feedback step: Operation and maintenance personnel execute tasks according to instructions and feedback the results through the user interaction layer, including task status, time, and problems. The system optimizes subsequent decisions and task allocation based on the feedback; S7, Intelligent early warning step: The intelligent early warning unit monitors key system data in real time and compares it with the preset threshold value, which is determined according to system characteristics and safety standards; S8, Performance evaluation and optimization steps: The performance evaluation unit regularly evaluates system performance, uses multi-indicator comprehensive evaluation, generates reports by comparing historical and industry standards, optimizes the system based on the reports, identifies bottlenecks and provides targeted suggestions, including hardware upgrades and algorithm adjustments; S9, data adaptive optimization step: according to the data analysis results after the system is running, automatically identify the weak links and key areas of data collection, dynamically adjust the distribution and collection parameters of sensors, and at the same time, optimize the data storage strategy according to the importance and change frequency of different types of data; S10. Collaborative operation and maintenance decision-making steps: When making operation and maintenance decisions, conduct data interaction and collaborative analysis with the operation and maintenance systems of surrounding buildings or buildings of the same type, and jointly formulate better operation and maintenance strategies by sharing data and experience.

[0015] Furthermore, in step S1, the data acquisition layer automatically adjusts the acquisition parameters of the sensor according to the real-time data analysis results, and monitors the working status of the sensor in real time, promptly discovers faults and issues alarms. Specifically, the data acquisition layer analyzes the stability, change trends and other characteristics of the collected data in real time. When it is found that the data fluctuation exceeds the normal range or shows an abnormal change trend, the sensor's acquisition frequency and accuracy parameters are automatically adjusted. For sensor working status monitoring, the sensor's communication status, power, and temperature indicators are monitored by periodically sending detection signals and receiving sensor responses. If an abnormality is found, an alarm message is immediately issued.

[0016] Furthermore, in step S3, when analyzing data, the big data analysis layer uses parallel computing technology to accelerate the calculation process, and at the same time visualizes the analysis results to facilitate operation and maintenance personnel to intuitively understand the data relationship. The parallel computing technology uses the MapReduce or Spark framework to divide large-scale data into multiple subtasks, which are assigned to multiple computing nodes for parallel processing. In terms of visualization, bar charts, line charts, scatter plots, and heat maps are used to display the correlation, distribution, and trend change relationship between data. Operation and maintenance personnel select different data dimensions and analysis algorithms through the visualization interface of the user interaction layer, generate corresponding visualization results in real time, and can zoom, translate, and filter charts.

[0017] Furthermore, in step S5, when the operation and maintenance decision-making layer generates strategies, it uses simulation technology to pre-evaluate different strategies and select the optimal strategy to execute. At the same time, it can monitor and evaluate the strategy execution effect in real time and adjust the strategy in time. The simulation technology simulates the operation of different operation and maintenance strategies based on the mathematical model and real-time data of the building system to predict their impact on system performance, energy consumption and cost. The evaluation indicators include equipment failure rate reduction rate, energy consumption saving rate and operation and maintenance cost reduction rate.

[0018] Furthermore, in step S8, the performance evaluation unit evaluates the system again after optimizing it, forming a closed-loop optimization process, continuously improving the system performance to the optimal state, and sharing the optimization experience with other similar building systems. In terms of performance evaluation, a multi-index evaluation method such as the analytic hierarchy process (AHP) or the fuzzy comprehensive evaluation method is used to comprehensively consider the accuracy, stability, response speed, resource utilization, and user satisfaction indicators of the system, set weights for each indicator, and calculate the comprehensive performance score of the system.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This system uses big data analysis to achieve real-time monitoring of the operating status of building equipment and fault prediction, and the prediction model has a high accuracy rate. The operation and maintenance decision-making layer quickly generates and dynamically adjusts strategies based on the analysis results. For example, when equipment suddenly fails, it can respond within 3 minutes with an accuracy rate of over 98%, effectively avoiding downtime losses caused by equipment failure, increasing equipment service life, and reducing maintenance costs.

[0020] The energy consumption optimization model takes into account a variety of factors and realizes efficient energy management through in-depth analysis and accurate prediction of building energy consumption data. Automatically adjusting energy strategies according to different seasons, time periods and equipment operation characteristics can effectively reduce building energy consumption, improve energy utilization efficiency, save a lot of costs for building operations, and comply with the concept of sustainable development.

[0021] The data collection layer ensures the accuracy and comprehensiveness of the data. The collection frequency and accuracy can be flexibly adjusted. The sensor can adapt to a variety of environments and has self-calibration function. The data storage layer realizes the safe storage and efficient management of massive data. The combination and optimization technology of distributed file system and relational database ensures the integrity and query efficiency of data, providing a solid foundation for subsequent analysis.

[0022] The user interaction layer provides a convenient visual interface, which is convenient for operation and maintenance personnel to operate and obtain information, and timely feedback on execution status to optimize system operation. At the same time, the system and method of the present invention can share optimization experience, promote technological progress and management level improvement of the entire building operation and maintenance industry, and promote the development of building intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic block diagram of the building intelligent operation and maintenance management system based on big data analysis proposed by the present invention; Figure 2 This is a schematic block diagram of the building intelligent operation and maintenance management method based on big data analysis proposed in the present invention. DETAILED DESCRIPTION

[0024] 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.

[0025] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0026] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0027] Reference Figure 1-2 :A building intelligent operation and maintenance management system based on big data analysis, including: The data collection layer is equipped with various types of sensors with an accuracy of ±0.01%. The collection frequency can be adjusted between 0.1Hz-10Hz as needed. It can automatically identify and adapt to various equipment interfaces in the building to achieve plug-and-play data collection. It also has a self-calibration function, and the calibration cycle can be set to weekly or monthly. The data is encrypted and transmitted via wired or wireless methods. Wireless transmission supports Wi-Fi, Bluetooth, ZigBee and other protocols, and the transmission rate is not less than 10Mbps. The data storage layer consists of a distributed file system and a relational database. The distributed file system stores raw data, with a capacity that can be expanded to PB level and a redundancy of no less than 3. It uses distributed cache technology, with a cache hit rate of no less than 85%. The relational database stores structured data and has transaction processing, backup and recovery functions. The backup cycle can be set, and the recovery time does not exceed 30 minutes. The big data analysis layer includes a data cleaning module, which removes bad data such as noise based on a combination of rules and machine learning. The algorithm library contains a variety of analysis algorithms and can be dynamically updated and expanded. The model training module uses machine learning technology to train equipment failure prediction models based on historical data (formula: , is the failure probability, X is the operating parameter vector, a and b are fitting coefficients) and energy consumption optimization model (formula: , Eopt To optimize energy consumption, k1-k7 are fitting coefficients, T represents the operating temperature of the equipment, N represents the number of equipment, V represents wind speed related parameters, I represents the operating current of the equipment, Tenv represents the ambient temperature, H represents the ambient humidity, and AQ represents the air quality). The model accuracy rate exceeds 92%, and cross-validation technology is used during training to improve generalization ability; The operation and maintenance decision-making layer generates operation and maintenance strategies based on the results of big data analysis, including equipment maintenance, energy management, and personnel scheduling strategies. It can be adjusted dynamically based on real-time data. When a sudden equipment failure occurs, it will be adjusted and notified within 3 minutes, and notifications and instructions will be pushed to the operation and maintenance personnel's terminal. The accuracy rate exceeds 98%. The strategy has priority and traceability, and is optimized in combination with the expert experience library when it is generated; The user interaction layer provides a visual interface through which operation and maintenance personnel can view system operation data, receive notifications and instructions, and provide feedback on task execution. Users can access the interface through a variety of terminals (such as computers and mobile phones). The interface is easy to operate and the response time does not exceed 1 second. Intelligent early warning module: Real-time monitoring of key system data, dynamic setting of early warning thresholds based on system characteristics and safety standards, and the use of multi-dimensional data analysis technology, taking into account factors such as historical data fluctuations, real-time data change rates, and equipment operating status associations. When data exceeds the early warning threshold, accurate early warning information is sent to operation and maintenance personnel through sound and light alarms, SMS push notifications, system pop-ups, and other methods. The early warning information includes details of abnormal data, possible cause analysis, and preliminary solution recommendations. At the same time, early warning event records are automatically generated to facilitate subsequent tracing and analysis; Simulation optimization module: Based on the digital model of the building system and combined with the results of big data analysis, different operation and maintenance strategies and equipment parameter adjustments are simulated. Virtual reality (VR) and augmented reality (AR) technologies are used to provide operation and maintenance personnel with an immersive simulation scene experience so that they can more intuitively evaluate the effects of strategies and adjustments. Operation and maintenance decisions are optimized based on simulation results, the optimal solution is selected for execution, and continuous learning and accumulation of simulation experience are achieved, the optimization model is automatically updated, and the system's ability to respond to complex operation and maintenance situations is improved.

[0028] In the present invention, the sensor of the data acquisition layer works stably in harsh environments (such as high temperature, high humidity, and strong electromagnetic interference), the measurement error is less than ±0.05%, and the battery life is at least 1.5 years. The sensor adopts a low-power design, and the sensor can dynamically adjust the acquisition frequency according to the importance of different areas and equipment in the building, and the acquisition frequency of key areas and key equipment is increased. The specific adjustment method is: the system divides the area according to the building layout and the importance level of the equipment in advance, and sets a basic acquisition frequency for each area and equipment. When it is detected that the operating data of a certain area or equipment fluctuates abnormally or approaches the warning threshold, the acquisition frequency of the area or equipment is automatically increased to 2-5 times the basic frequency. After continuous acquisition for a period of time (such as 10-30 minutes), the acquisition frequency is restored or further adjusted according to the stability of the data. The sensor adopts advanced heat insulation, moisture-proof, shielding materials and circuit design to ensure stable operation in harsh environments, such as ceramic packaging, electromagnetic shielding cover and other measures. At the same time, the power management module of the sensor chip is optimized to achieve low power design and reduce battery energy consumption.

[0029] In the present invention, the relational database of the data storage layer supports distributed transaction processing, ensures the consistency and integrity of cross-node data operations, improves the system data management capabilities, and can automatically optimize the data storage structure according to the data access frequency to improve data query efficiency. In terms of distributed transaction processing, a two-phase commit protocol (2PC) or a three-phase commit protocol (3PC) is adopted to ensure that data operations on multiple nodes are either all successful or all rolled back. For the monitoring of data access frequency, the system counts the number of accesses to each data block at regular time intervals (such as 5-10 minutes). When the access frequency of a data block exceeds a set threshold (such as 100 times per minute), it is migrated to a high-speed storage area (such as a solid-state hard disk). If the access frequency continues to decrease to a certain level (such as less than 10 times per minute), it is migrated back to the ordinary storage area to achieve optimal utilization of storage resources.

[0030] In the present invention, the algorithm library of the big data analysis layer contains an anomaly detection algorithm based on deep learning, which can more accurately identify abnormal data patterns in the building system and warn of potential risks in advance. The algorithms in the algorithm library can be customized according to user needs and actual application scenarios. The deep learning anomaly detection algorithm adopts a convolutional neural network (CNN) or a recurrent neural network (RNN) architecture. The training data includes normal operation data and simulated abnormal data. By learning a large amount of data, the algorithm can automatically extract data features and identify deviations from normal patterns. For customized development, users can specify the specific data type to be analyzed, the abnormal pattern features of concern, etc. through a graphical interface or a configuration file. The system automatically adjusts the algorithm parameters or combines existing algorithm modules according to user needs to generate a customized analysis algorithm.

[0031] In the present invention, when generating the operation and maintenance strategy, the operation and maintenance decision-making layer can automatically adjust the energy management strategy according to the energy consumption demand and equipment operation characteristics in different seasons and time periods to achieve efficient use of energy. At the same time, the work performance of the operation and maintenance personnel can be evaluated to optimize the personnel scheduling. In terms of energy management strategy adjustment, the system pre-stores typical energy consumption patterns and equipment operation parameter standards in different seasons and time periods, and automatically adjusts the equipment operation mode (such as temperature setting of the air-conditioning system, brightness adjustment of the lighting system, etc.) through comparative analysis in combination with real-time collected data to achieve energy saving. For the evaluation of the work performance of the operation and maintenance personnel, an evaluation model is established based on indicators such as the timeliness, accuracy, and effectiveness of feedback issues of task completion, and the weighted average method is used to calculate the performance score. According to the score, the personnel scheduling is optimized, and tasks are assigned to the operation and maintenance personnel with higher performance first, and the personnel with lower performance are trained or adjusted.

[0032] The present invention also discloses a building intelligent operation and maintenance management method based on big data analysis, comprising the following steps: S1. Data collection steps: The data collection layer collects data on equipment operation, environment and personnel activities in the building at the set frequency and accuracy. The sensor is self-calibrated before collection, and automatically re-collects or adjusts parameters in case of abnormality. The sensor automatically adapts to the device interface to achieve plug-and-play collection, and dynamically allocates collection resources according to different areas of the building and the importance of equipment; S2, data transmission and storage step: the collected data is encrypted and adaptively transmitted and then stored in the data storage layer. Multi-link redundancy is supported during the transmission process, and automatic switching is performed when the main link fails to ensure safe and stable data storage. The data is classified and managed during storage to improve the efficiency of data query and call; S3, data analysis step: The big data analysis layer reads data from the storage layer, cleans it, and uses the algorithm library to analyze it and mine data relationships (such as Calculate associations, where X is the operating parameter vector and Y is energy consumption). The algorithm can be customized and updated. When new relationships are discovered, the algorithm library is automatically updated, and deep learning anomaly detection algorithms are used to identify anomalies. S4, model training and prediction steps: The model training module trains the model based on historical and analysis results, cross-validates to improve generalization ability, adjusts the coefficient to make the accuracy rate exceed 92%, uses the trained model to predict equipment operation and energy consumption, and the prediction results are used to guide operation and maintenance decisions; S5, Operation and maintenance decision generation step: The operation and maintenance decision layer generates strategies based on the analysis and prediction results, optimizes the expert experience database, and pushes them to the operation and maintenance personnel terminal. If there is no timely response, the automatic upgrade warning will notify the management personnel. The strategy includes priority and traceability, and automatically adjusts the strategy according to different situations; S6, Operation and maintenance execution and feedback step: Operation and maintenance personnel execute tasks according to instructions and feedback the results through the user interaction layer, including task status, time, problems, etc. The system optimizes subsequent decisions and task allocation based on the feedback; S7, Intelligent early warning step: The intelligent early warning unit monitors key system data in real time and compares it with the preset threshold (the threshold is determined according to system characteristics and safety standards). When the threshold is exceeded, a multi-form early warning is issued to ensure that abnormalities are discovered in a timely manner and the early warning information is detailed and accurate, so that operation and maintenance personnel can quickly locate the problem; S8. Performance evaluation and optimization steps: The performance evaluation unit evaluates system performance regularly (e.g. monthly or quarterly), uses multiple indicators (including user satisfaction) for comprehensive evaluation, generates reports by comparing historical and industry standards, optimizes the system based on the reports, identifies bottlenecks and provides targeted suggestions, such as hardware upgrades and algorithm adjustments, to continuously improve system performance; Data adaptive optimization steps: According to the data analysis results after the system has been running for a period of time, the weak links and key areas of data collection are automatically identified, and the distribution and collection parameters of sensors are dynamically adjusted to improve the comprehensiveness and accuracy of data collection. At the same time, according to the importance and change frequency of different types of data, the data storage strategy is optimized to improve the efficiency of data storage and retrieval; Collaborative operation and maintenance decision-making steps: When making operation and maintenance decisions, not only the data analysis results within the building system are considered, but also data interaction and collaborative analysis can be performed with the operation and maintenance systems of surrounding buildings or buildings of the same type. By sharing data and experience, better operation and maintenance strategies can be jointly formulated. For example, in terms of energy management, joint optimization scheduling of buildings in the region can be achieved to improve energy efficiency and reduce operation and maintenance costs.

[0033] In the present invention, in step S1, the data acquisition layer can automatically adjust the acquisition parameters of the sensor according to the real-time data analysis results to obtain more valuable data, and can also monitor the working status of the sensor in real time, detect faults in time and issue an alarm. Specifically, the data acquisition layer analyzes the stability, change trend and other characteristics of the collected data in real time. When it is found that the data fluctuation exceeds the normal range or shows an abnormal change trend, it automatically adjusts the acquisition frequency, accuracy and other parameters of the sensor, such as doubling the acquisition frequency or switching to a higher accuracy mode. For the monitoring of the working status of the sensor, by periodically sending a detection signal and receiving the response of the sensor, the key indicators such as the communication status, power, and temperature of the sensor are monitored. Once an abnormality is found (such as communication interruption, low power, high temperature, etc.), an alarm message is immediately issued, and the alarm message is notified to the operation and maintenance personnel through SMS, email, system pop-up window and other methods.

[0034] In the present invention, in step S3, when analyzing data, the big data analysis layer can use parallel computing technology to accelerate the calculation process and improve the analysis efficiency. At the same time, the analysis results can be visualized to facilitate the operation and maintenance personnel to intuitively understand the data relationship. Parallel computing technology uses MapReduce or Spark framework to divide large-scale data into multiple subtasks, distribute them to multiple computing nodes for parallel processing, and finally summarize the results. In terms of visualization, a variety of chart forms such as bar charts, line charts, scatter charts, and heat maps are used to display the correlation, distribution, trend changes and other relationships between data. The operation and maintenance personnel can select different data dimensions and analysis algorithms through the visualization interface of the user interaction layer, generate corresponding visualization results in real time, and can zoom, translate, filter and other operations on the chart to deeply analyze the data.

[0035] In the present invention, in step S5, when the operation and maintenance decision-making layer generates strategies, it can use simulation technology to pre-evaluate different strategies and select the optimal strategy for execution. At the same time, it can monitor and evaluate the strategy execution effect in real time and adjust the strategy in time. Simulation technology simulates the operation of different operation and maintenance strategies (such as equipment maintenance plans, energy management plans, etc.) based on the mathematical model and real-time data of the building system, and predicts their impact on system performance, energy consumption, cost, etc. The evaluation indicators include equipment failure rate reduction rate, energy consumption saving rate, operation and maintenance cost reduction rate, etc. By comparing the simulation results of different strategies, the strategy with the best comprehensive performance is selected for execution. During the strategy execution process, the actual operation data of the system is monitored in real time and compared with the simulation prediction results. When it is found that the actual effect deviates from the expectation, the strategy parameters are adjusted or the strategy is replaced in time to ensure the effectiveness of the operation and maintenance strategy.

[0036] In the present invention, in step S8, the performance evaluation unit evaluates again after optimizing the system to form a closed-loop optimization process, continuously improve the system performance to the optimal state, and at the same time, the optimization experience can be shared with other similar building systems to promote the development of the entire industry. In terms of performance evaluation, a multi-index evaluation method such as the analytic hierarchy process (AHP) or the fuzzy comprehensive evaluation method is adopted to comprehensively consider multiple indicators such as the accuracy, stability, response speed, resource utilization, and user satisfaction of the system, set a weight for each indicator, and calculate the comprehensive performance score of the system. According to the evaluation results, targeted optimization suggestions are proposed for system performance bottlenecks (such as data collection delays, low efficiency of analysis algorithms, inaccurate decision execution, etc.), such as upgrading hardware equipment, improving algorithms, adjusting operation and maintenance strategies, etc. After the optimization is implemented, the performance evaluation is performed again, and the optimization is continuously iterated until the system performance reaches the optimal state. At the same time, the successful experience and effective measures in the system optimization process are organized into a case library, and shared with the operation and maintenance managers of other building systems through channels such as industry forums, technical seminars, and online platforms, so as to promote the technological progress and management level of the entire industry.

[0037] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A building intelligent operation and maintenance management system based on big data analysis, characterized in that: include: The data collection layer is equipped with various types of sensors to automatically identify and adapt to the equipment interfaces in the building. The data is encrypted and transmitted via wired or wireless methods; The data storage layer consists of a distributed file system and a relational database. The distributed file system stores raw data and uses distributed cache technology, while the relational database stores structured data. The big data analysis layer includes a data cleaning module, which removes noisy data based on a combination of rules and machine learning. The algorithm library is dynamically updated and expanded. The model training module uses machine learning technology to train equipment failure prediction models based on historical data, and cross-validation technology is used during training to improve generalization capabilities. The operation and maintenance decision-making layer generates operation and maintenance strategies based on big data analysis results, and dynamically adjusts them based on real-time data. When a sudden equipment failure occurs, adjustments and notifications are made within 3 minutes, and notifications and instructions are pushed to the operation and maintenance personnel's terminals. Strategies have priority and traceability, and are optimized in combination with the expert experience library when generated; The user interaction layer provides a visual interface through which operation and maintenance personnel can view system operation data, receive notifications and instructions, and provide feedback on task execution. Users can access the interface through computers and mobile phones. Intelligent early warning module: real-time monitoring of key system data, dynamic setting of early warning thresholds based on system characteristics and safety standards, and the use of multi-dimensional data analysis technology to comprehensively consider historical data fluctuations, real-time data change rates, and equipment operating status correlation factors. When the data exceeds the early warning threshold, an early warning message is sent to the operation and maintenance personnel; Simulation and optimization module: Based on the digital model of the building system and combined with the results of big data analysis, different operation and maintenance strategies and equipment parameter adjustments are simulated. Virtual reality and augmented reality technologies are used to provide operation and maintenance personnel with an immersive simulation scene experience. Operation and maintenance decisions are optimized based on simulation results, and the best solution is selected for execution.

2. The building intelligent operation and maintenance management system based on big data analysis according to claim 1 is characterized in that: The sensors of the data acquisition layer dynamically adjust the acquisition frequency according to the importance of different areas and equipment in the building, and the acquisition frequency of key areas and key equipment is increased. The specific adjustment method is: the system divides the areas according to the building layout and the importance level of the equipment in advance, and sets a basic acquisition frequency for each area and equipment. When it is detected that the operating data of a certain area or equipment fluctuates abnormally or approaches the warning threshold, the acquisition frequency of the area or equipment is automatically increased to 2-5 times the basic frequency. After 10-30 minutes of continuous acquisition, the acquisition frequency is restored or further adjusted according to the stability of the data; the sensor adopts ceramic packaging, electromagnetic shielding cover and other designs.

3. The building intelligent operation and maintenance management system based on big data analysis according to claim 1 is characterized in that: The relational database of the data storage layer supports distributed transaction processing and can automatically optimize the data storage structure according to the frequency of data access. In terms of distributed transaction processing, the two-phase commit protocol 2PC or the three-phase commit protocol 3PC is adopted to ensure that data operations on multiple nodes are all successful or all rolled back.

4. The building intelligent operation and maintenance management system based on big data analysis according to claim 1 is characterized in that: The algorithm library of the big data analysis layer contains an anomaly detection algorithm based on deep learning, which can identify abnormal data patterns in the building system and warn of potential risks in advance. The algorithms in the algorithm library can be customized according to user needs and actual application scenarios. The deep learning anomaly detection algorithm adopts a convolutional neural network CNN or a recurrent neural network RNN ​​architecture. The training data includes normal operation data and simulated abnormal data. Through learning the data, the algorithm can automatically extract data features and identify deviations from normal patterns. The specific training formula of the equipment failure prediction model is: ,in is the failure probability, X is the operating parameter vector, a and b are the fitting coefficients; the energy consumption optimization model formula is: , where E opt To optimize energy consumption, k1-k7 are fitting coefficients, T represents the equipment operating temperature, N represents the number of equipment, V represents wind speed related parameters, I represents the equipment operating current, Tenv represents the ambient temperature, H represents the ambient humidity, and AQ represents the air quality.

5. The building intelligent operation and maintenance management system based on big data analysis according to claim 1 is characterized in that: When generating the operation and maintenance strategy, the operation and maintenance decision-making layer automatically adjusts the energy management strategy according to the energy consumption demand and equipment operation characteristics in different seasons and time periods. For the work performance evaluation of the operation and maintenance personnel, an evaluation model is established based on indicators such as the timeliness and accuracy of task completion, and the effectiveness of feedback issues. The performance score is calculated using the weighted average method. According to the score, personnel scheduling is optimized, tasks are assigned to operation and maintenance personnel with higher performance first, and personnel with lower performance are trained or adjusted to other positions.

6. A method for applying the building intelligent operation and maintenance management system based on big data analysis according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. Data collection steps: The data collection layer collects data on equipment operation, environment and personnel activities in the building at a set frequency and accuracy. The sensor is self-calibrated before collection, and automatically re-collects or adjusts parameters in case of abnormality, and the sensor automatically adapts to the device interface; S2, data transmission and storage step: the collected data is encrypted and adaptively transmitted and then stored in the data storage layer. Multi-link redundancy is supported during the transmission process, and automatic switching is performed when the main link fails. Classification and management of data is performed during storage; S3, data analysis step: The big data analysis layer reads data from the storage layer, cleans it, and uses the algorithm library to analyze it and mine data relationships; S4, model training and prediction steps: The model training module trains the model based on historical and analysis results, cross-validates to improve generalization ability, uses the trained model to predict equipment operation and energy consumption, and the prediction results are used to guide operation and maintenance decisions; S5, Operation and maintenance decision generation step: The operation and maintenance decision layer generates strategies based on the analysis and prediction results, optimizes the expert experience database, and pushes them to the operation and maintenance personnel terminal. If there is no timely response, the automatic upgrade warning will notify the management personnel. The strategy includes priority and traceability, and automatically adjusts the strategy according to different situations; S6, Operation and maintenance execution and feedback step: Operation and maintenance personnel execute tasks according to instructions and feedback the results through the user interaction layer, including task status, time, and problems. The system optimizes subsequent decisions and task allocation based on the feedback; S7, Intelligent early warning step: The intelligent early warning unit monitors key system data in real time and compares it with the preset threshold value, which is determined according to system characteristics and safety standards; S8. Performance evaluation and optimization steps: The performance evaluation unit regularly evaluates system performance, uses multi-indicator comprehensive evaluation, generates reports by comparing historical and industry standards, optimizes the system based on the reports, identifies bottlenecks, and provides targeted suggestions; S9, data adaptive optimization step: according to the data analysis results after the system is running, automatically identify the weak links and key areas of data collection, dynamically adjust the distribution and collection parameters of sensors, and at the same time, optimize the data storage strategy according to the importance and change frequency of different types of data; S10. Collaborative operation and maintenance decision-making steps: When making operation and maintenance decisions, conduct data interaction and collaborative analysis with the operation and maintenance systems of surrounding buildings or buildings of the same type, and jointly formulate better operation and maintenance strategies by sharing data and experience.

7. The method of building intelligent operation and maintenance management system based on big data analysis according to claim 6 is characterized in that: In step S1, the data acquisition layer automatically adjusts the acquisition parameters of the sensor according to the real-time data analysis results, and monitors the working status of the sensor in real time, discovers faults and issues alarms in time. Specifically, the data acquisition layer analyzes the stability, change trends and other characteristics of the collected data in real time. When it is found that the data fluctuation exceeds the normal range or shows an abnormal change trend, it automatically adjusts the acquisition frequency and accuracy parameters of the sensor. For sensor working status monitoring, it monitors the communication status, power and temperature indicators of the sensor by periodically sending detection signals and receiving the sensor's response. If an abnormality is found, an alarm message is immediately issued.

8. The method of building intelligent operation and maintenance management system based on big data analysis according to claim 6 is characterized in that: In step S3, when analyzing data, the big data analysis layer uses parallel computing technology to accelerate the calculation process and visualize the analysis results, so that the operation and maintenance personnel can intuitively understand the data relationship. The parallel computing technology uses the MapReduce or Spark framework to divide large-scale data into multiple subtasks and distribute them to multiple computing nodes for parallel processing. In terms of visualization, bar charts, line charts, scatter charts, and heat maps are used to display the correlation, distribution, and trend change relationship between data. The operation and maintenance personnel select different data dimensions and analysis algorithms through the visualization interface of the user interaction layer, generate corresponding visualization results in real time, and can zoom, translate, and filter the charts; Using the formula: Calculate the association, X is the operating parameter vector, Y is the energy consumption, automatically update the algorithm library when new relationships are found, and use the deep learning anomaly detection algorithm to identify anomalies.

9. The method of building intelligent operation and maintenance management system based on big data analysis according to claim 6 is characterized in that: In step S5, when generating strategies, the operation and maintenance decision-making layer uses simulation technology to pre-evaluate different strategies and select the optimal strategy for execution. At the same time, it can monitor and evaluate the strategy execution effect in real time and adjust the strategy in time. The simulation technology simulates the operation of different operation and maintenance strategies based on the mathematical model and real-time data of the building system to predict their impact on system performance, energy consumption, and cost. The evaluation indicators include equipment failure rate reduction rate, energy consumption saving rate, and operation and maintenance cost reduction rate.

10. The method of building intelligent operation and maintenance management system based on big data analysis according to claim 6 is characterized in that: In step S8, the performance evaluation unit evaluates the system again after optimizing it, forming a closed-loop optimization process, continuously improving the system performance to the optimal state, and sharing the optimization experience with other similar building systems. In terms of performance evaluation, a multi-index evaluation method such as the analytic hierarchy process (AHP) or the fuzzy comprehensive evaluation method is used to comprehensively consider the accuracy, stability, response speed, resource utilization, and user satisfaction indicators of the system, set weights for each indicator, and calculate the comprehensive performance score of the system.

Images