Property maintenance prediction system based on big data

Abstract

The invention discloses a property maintenance prediction system based on big data, and the system comprises the steps: monitoring equipment and environment in real time through an intelligent sensor, transmitting the data to a cloud or local storage through a wireless network, and carrying out the historical data processing and cleaning through a distributed architecture; predicting the equipment fault risk through machine learning, establishing a customized prediction model for different equipment, analyzing the life cycle and predicting the service life of the equipment; for old equipment, constructing a life extension model by using historical fault data; the system generates intelligent decision suggestions, pushes equipment health states to property management personnel, automatically formulates maintenance plans and priorities, and dispatches maintenance personnel; during fault prediction, early warning is given out in real time, related personnel are reminded through a short message or an APP, meanwhile, maintenance operation is recorded, historical data are analyzed, and future maintenance plans and budget are optimized; the system dynamically adjusts the life prediction model of the old equipment by accumulating historical data, and ensures the accuracy and efficiency of equipment health management.

Description

Technical Field

[0001] The present invention belongs to the field of property maintenance prediction based on big data, and particularly relates to a property maintenance prediction system based on big data. Background Art

[0002] A property maintenance prediction system based on big data utilizes big data analysis technology to collect, analyze, and process data of various equipment, facilities, and environments in property management to predict and prevent possible failures or problems, thereby optimizing the property management and maintenance process. It obtains real-time data from various equipment within the property (such as elevators, air conditioners, HVAC, etc.) and environmental monitoring systems (such as temperature and humidity, air pressure, etc.); through big data analysis technology, using methods such as machine learning and data mining, it analyzes the operating status of the equipment, identifies information such as the usage patterns of the equipment, historical failure records, and maintenance cycles, and establishes a prediction model; according to the analysis results, the system can predict the failure risk and wear condition of the equipment and issue a warning notice to remind property management personnel to perform necessary maintenance or replacement; according to the prediction results, the system can automatically generate and adjust the maintenance plan, optimize the allocation of maintenance resources, avoid over-maintenance or missed maintenance, reduce costs, and improve work efficiency; it provides data-driven decision support for property management personnel to help formulate reasonable repair, replacement, and budget plans.

[0003] However, although the existing property maintenance prediction system based on big data has a significant effect in improving property management efficiency and preventing failures, there are also some defects and challenges. For some old equipment or facilities with relatively old technologies, the data may be insufficient, and it is difficult for the system to make effective predictions based on the existing data. Summary of the Invention

[0004] Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide a property maintenance prediction system based on big data. For old equipment, by accumulating more historical failure data, a device life prediction model is established, and dynamic adjustment is performed according to the actual operating conditions of the equipment.

[0005] The technical solution adopted by the present invention to solve its technical problems is as follows:

[0006] A property maintenance prediction system based on big data includes:

[0007] A data collection and transmission module, which is used to install various intelligent sensors to monitor the operating status of equipment and the environment within the property in real time, and transmit the data collected by the equipment sensors to the cloud or local storage through a wireless network;

[0008] A data processing and analysis module, which is used to store the historical data of the equipment based on a distributed storage architecture and perform data processing, and process the original data through data cleaning technology;

[0009] A prediction and optimization model module, which is used to predict the failure risk of equipment by using machine learning. For different equipment, customized prediction models are established respectively to conduct life cycle analysis on the equipment, predict the expected life of the equipment, and for old equipment, establish a life extension prediction model through historical failure data;

[0010] A decision support module, which is used to generate intelligent decision-making suggestions based on big data analysis, send maintenance and resource allocation according to the equipment health status to property management personnel, and automatically generate maintenance plans and priorities, and automatically dispatch maintenance personnel;

[0011] An execution module, which is used to issue a real-time warning when a potential failure is predicted, remind relevant personnel to take actions by means of text messages or APPs, record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets.

[0012] Preferably, various intelligent sensors are installed to monitor the operating status of equipment and the environment in the property in real time, and the data collected by the equipment sensors are transmitted to the cloud or stored locally through a wireless network as:

[0013] Collect equipment and environmental data through installed temperature and humidity sensors, vibration sensors, and air pressure sensors, and use Wi-Fi, LoRa, Zigbee, 5G wireless communication technologies to transmit the data from the sensors to the cloud or store it locally in real time;

[0014] Use AWS IoT, Azure IoT, Google Cloud IoT cloud platforms to store and manage equipment and environmental data.

[0015] Preferably, based on the equipment historical data stored in a distributed storage architecture and data processing, the raw data is processed through data cleaning technology as:

[0016] Use a deduplication algorithm to remove duplicates by comparing the timestamps and device ID fields of the data, use the mean, median, and most common values to fill in missing values, and use regression models and KNN machine learning models to predict the missing values;

[0017] Use the Z-score or IQR method to detect outliers, and use the Isolation Forest, K-means clustering, and DBSCAN algorithms to automatically identify abnormal data points;

[0018] Convert the data to data with zero mean and unit variance, and the formula is:

[0019]

[0020] Among them, X is the raw data, μ is the mean of the data, and σ is the standard deviation;

[0021] Compress the data to a specified range, and the formula is:

[0022]

[0023] Align the time series data to ensure that the timestamps of different data sources are consistent. Resample the data according to the timestamps to unify the time interval of the data, and perform interpolation processing at the time points where data is missing. Smooth the data by moving average and weighted average methods.

[0024] Preferably, use machine learning to predict the failure risk of the device. For different devices, establish customized prediction models respectively, perform life cycle analysis on the devices, predict the expected life of the devices, and for old devices, establish a life extension prediction model through historical failure data as:

[0025] By analyzing the historical data of the device, extract the life cycle characteristics of the device, and combine survival analysis, regression models and time series analysis machine learning algorithms to predict the failure risk of the device based on these characteristics;

[0026] Identify the failure modes of the device through isolation forest, K-means clustering, One-Class SVM anomaly detection algorithms in machine learning, and predict the remaining life of the device through linear regression, random forest regression, and deep neural network regression models;

[0027] Use regression models or classification models to predict whether the device will fail. By detecting abnormal fluctuations during the operation of the device, predict the upcoming failures of the device, and based on the historical failure records of the device, predict when to perform maintenance or replace components to extend the service life of the device.

[0028] Preferably, generate intelligent decision-making suggestions based on big data analysis, send to property managers for maintenance and resource allocation according to the device health status, and automatically generate maintenance plans and priorities, and automatically dispatch maintenance personnel as:

[0029] Through the installed sensors and Internet of Things technology, collect the operation data of the device in real time, integrate the historical failure records, maintenance logs, repair history and usage of the device, analyze the health status of the device with machine learning algorithms, and calculate the failure risk and remaining life of the device;

[0030] Based on the device health status, failure prediction results and resource utilization, the intelligent decision-making model generates maintenance suggestions, and the suggestions include whether to perform maintenance or replace components, the failure priority of the device, and the dispatched resources;

[0031] Automatically generate regular or emergency maintenance plans based on failure risks, remaining equipment life, and equipment importance, allocate maintenance tasks according to priorities, and allocate maintenance personnel and spare part resources according to maintenance requirements, personnel capabilities, and resource availability factors using linear programming, genetic algorithm optimization algorithms;

[0032] Automatically dispatch maintenance personnel to perform maintenance tasks according to real-time data, maintenance plans, and resource allocation results.

[0033] Preferably, when potential failures are predicted, issue early warnings in real time, remind relevant personnel to take actions via text messages or apps, record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets as follows:

[0034] Analyze the real-time data and historical data of the equipment based on a machine learning model. When the model identifies potential failure risks in the equipment, the system automatically triggers an early warning, analyzes the health status of the equipment in real time, and evaluates the severity and likelihood of the failure;

[0035] When potential failures are detected and predicted, notify maintenance personnel of tasks and relevant details via text messages, emails, or dedicated apps, including equipment problem descriptions, maintenance priorities, and maintenance times. After maintenance personnel perform maintenance tasks, update the maintenance progress in real time;

[0036] The system details and records the specific operations and processing procedures of each maintenance, ensures that the maintenance history is completely traceable, and integrates the historical maintenance data of all equipment, including maintenance frequencies, failure types, and maintenance costs. Through data analysis, identify potential maintenance patterns and trends;

[0037] Through the analysis of historical maintenance data, the system can predict future equipment maintenance requirements, optimize maintenance plans according to equipment usage, failure probabilities, and budget constraints, and continuously learn and optimize the prediction model based on new maintenance data and failure records.

[0038] Preferably, the operation formula of the property maintenance prediction system based on big data is as follows:

[0039] Failure prediction is achieved through regression analysis or classification algorithms. The formula for linear regression is:

[0040]

[0041] Where:

[0042] is the prediction target;

[0043] βo is the intercept term;

[0044] β1, β2,... is the regression coefficient of the model;

[0045] is the input feature;

[0046] ∈ is the error term;

[0047] For classification problems, the probability of a fault is predicted using logistic regression. The formula for logistic regression is:

[0048]

[0049] Where:

[0050] is the probability of a fault occurring;

[0051] Other symbols are the same as in linear regression;

[0052] Based on the scheduling of optimized maintenance resources, using linear programming or genetic algorithm methods, in terms of scheduling and resource allocation, linear programming is used to maximize efficiency or minimize cost. The formula for linear programming is:

[0053]

[0054] Where:

[0055] are the coefficients of the objective function;

[0056] are the decision variables;

[0057] are the coefficients in the constraint matrix;

[0058] b1, b2,... are the resource or time constraints;

[0059] By solving this linear programming problem, the system determines how to allocate resources to achieve cost minimization or efficiency maximization;

[0060] The formula for the fitness function is:

[0061]

[0062] Where:

[0063] is the solution of the fitness;

[0064] are the weights for each objective;

[0065] are the various parts of the solution;

[0066] For time series data predicting equipment failures, common methods include ARIMA. The formula for the ARIMA model is:

[0067] Y t = μ + φ1Y t＝1 + φ2Y t＝2 +... + φ p Y t＝p + θ1∈ t＝1 + θ2∈ t＝2 +... + θ q ∈ t＝q + ∈ t

[0068] Where:

[0069] Y t is the observed value at time t;

[0070] μ is the constant term;

[0071] φ1, φ2,..., φ p are autoregressive parameters;

[0072] θ1, θ2,..., θ q are moving average parameters;

[0073] ∈ t is white noise;

[0074] By using the ARIMA model, analyze the historical data of equipment failures to predict the time of future upcoming failures.

[0075] Another technical problem to be solved by the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the property maintenance prediction system based on big data as described in any one of the above.

[0076] Another technical problem to be solved by the present invention is to provide a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the property maintenance prediction system based on big data.

[0077] The beneficial effects of the present invention are:

[0078] Through the data collection and transmission module, the operating status of equipment and the environment can be monitored in real time, ensuring that property managers can timely understand the health status of the equipment; through the prediction and optimization model, the failure risk and lifespan of the equipment can be predicted in advance, so as to carry out preventive maintenance, avoid sudden equipment failures, and reduce equipment downtime and maintenance costs; based on big data analysis, the system will automatically generate equipment maintenance plans and priorities, and optimize the allocation of maintenance resources through intelligent scheduling; the execution module records each maintenance operation and analyzes historical data to evaluate the maintenance effect and further optimize future maintenance plans and budgets; for old equipment, an extended lifespan prediction model is established through historical failure data to reasonably extend the service life of the equipment, avoid premature equipment replacement, and reduce equipment replacement costs; all maintenance and repair operations are recorded by the system, and the operation history is made transparent for future tracking, analysis, and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 It is a schematic flowchart of the property maintenance prediction system based on big data of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention. The present invention is described more specifically by way of example in the following paragraphs. The advantages and features of the present invention will be clearer according to the following description and claims.

[0081] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0082] Embodiment

[0083] The property maintenance prediction system based on big data includes:

[0084] A data collection and transmission module, which is used to install various intelligent sensors to monitor the operating status of equipment and the environment in the property in real time, and transmit the data collected by the equipment sensors to the cloud or local storage through a wireless network;

[0085] A data processing and analysis module, which is used to store equipment historical data based on a distributed storage architecture and perform data processing, and process the original data through data cleaning technology;

[0086] A prediction and optimization model module is used to predict the failure risk of equipment using machine learning. For different equipment, customized prediction models are established respectively to conduct lifecycle analysis on the equipment, predict the expected lifespan of the equipment, and for old equipment, an extended lifespan prediction model is established through historical failure data.

[0087] A decision support module is used to generate intelligent decision-making suggestions based on big data analysis, send maintenance and resource allocation based on the equipment health status to property management personnel, automatically generate maintenance plans and priorities, and automatically dispatch maintenance personnel.

[0088] An execution module is used to issue real-time warnings when potential failures are predicted, remind relevant personnel to take actions via text messages or APPs, record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets.

[0089] Through predictive maintenance, potential failures can be identified in advance, sudden equipment downtime can be avoided, and the impact of equipment failures on business operations can be reduced. The system can identify the fatigue and wear of equipment, predict and prompt upcoming failures, allowing maintenance personnel to handle them in advance and avoid the occurrence of emergencies. Through optimizing the maintenance plan, the system can determine which equipment needs emergency maintenance and which can be postponed, thus avoiding unnecessary maintenance expenditures. Through regular maintenance and the extended lifespan prediction model, the lifecycle of equipment can be extended. For old equipment, by analyzing historical failure data, the service life of the equipment can be calculated, and a plan to extend the service life of the equipment can be proposed accordingly, thereby reducing the equipment replacement frequency and cost. Automatically generating maintenance plans and dispatching maintenance personnel greatly improves the efficiency of property management, avoids the cumbersome and delays of manual processing. The intelligent suggestions of the decision support module help management personnel understand the equipment health status in real time and respond quickly, reducing the possibility of slow response and human errors. Recording each maintenance operation in detail, the data records generated by the system are traceable, providing a transparent data source for subsequent analysis and reporting. Maintenance personnel and management can view the operating status and maintenance history of equipment at any time, improving work transparency and facilitating supervision and auditing. Data-driven intelligent decision support, based on the real-time data and historical performance of equipment, the system can intelligently generate maintenance decisions and prioritize them, automatically dispatch maintenance personnel to ensure the optimal use of resources. Through the analysis of historical failures and maintenance data, the system can reasonably predict future maintenance expenditures, help the property management department formulate more accurate budgets, and avoid budget overruns or resource waste.

[0090] Install various intelligent sensors to monitor the operating status of equipment and the environment in the property in real time, and transmit the data collected by the equipment sensors to the cloud or local storage via wireless network as:

[0091] Collect device and environmental data through installed temperature and humidity sensors, vibration sensors, and air pressure sensors, and use Wi-Fi, LoRa, Zigbee, 5G wireless communication technologies to transmit the data from the sensors to the cloud or local storage in real time;

[0092] Adopt cloud platforms such as AWS IoT, Azure IoT, and Google Cloud IoT to store and manage device and environmental data.

[0093] Monitor the device and environmental status in real time through sensors such as temperature and humidity, vibration, and air pressure, and potential device failures or environmental problems can be detected in a timely manner; by collecting and analyzing environmental data in real time, it can ensure that the device operates under the best conditions and avoid device damage caused by abnormal environments (such as too high temperature or too high humidity); through big data analysis, the maintenance plan can be optimized according to the actual operating status of the device and environmental conditions, avoiding unnecessary manual inspections and frequent repairs; using the cloud platform for data storage and analysis can provide real-time data reports and decision-making support for property managers; the cloud platform provides powerful data storage capabilities and supports the processing of high concurrency and large amounts of data; through the encryption and security measures of modern cloud platforms, data can be effectively protected during transmission and storage, avoiding data leakage and damage. In addition, cloud platforms usually have high availability and disaster tolerance capabilities to ensure the reliable storage and real-time access of device data; using platforms such as AWS IoT, Azure IoT, and Google Cloud IoT can facilitate device management, data processing, and analysis. These platforms provide complete APIs and development tools, enabling the system to be easily integrated with other systems, improving the overall management convenience.

[0094] Based on the device historical data stored in a distributed storage architecture and perform data processing. The raw data is processed through data cleaning technology as follows:

[0095] Adopt a deduplication algorithm to remove duplicate items by comparing the timestamps and device ID fields of the data, use the mean, median, and most common values to fill in missing values, and use regression models and KNN machine learning models to predict missing values;

[0096] Adopt the Z-score or IQR method to detect outliers, and use isolation forest, K-means clustering, and DBSCAN algorithms to automatically identify abnormal data points;

[0097] Convert the data to data with zero mean and unit variance. The formula is:

[0098]

[0099] where X is the raw data, μ is the mean of the data, and σ is the standard deviation;

[0100] Compress the data to a specified range, and the formula is:

[0101]

[0102] Align the time series data to ensure that the timestamps of different data sources are consistent. Resample the data according to the timestamps to unify the time interval of the data, and perform interpolation at the time points where data is missing. Smooth the data by moving average and weighted average methods.

[0103] By means of data cleaning such as deduplication, filling missing values, and outlier detection, the accuracy, integrity, and consistency of the data can be ensured; based on the distributed storage architecture, it is able to efficiently store and process a large amount of device historical data, and at the same time support the fast access and parallel processing of massive data; data standardization and normalization can eliminate the scale differences between different features, enabling machine learning algorithms to more effectively learn the potential patterns of the data; by using algorithms such as Isolation Forest, K-means clustering, and DBSCAN, abnormal data points can be automatically detected and identified, providing support for device fault warning, anomaly detection, and predictive analysis; operations such as aligning, resampling, and interpolating the time series data can ensure that the time series data is analyzed under a unified time dimension; the process of data cleaning and preprocessing makes the data more in line with the analysis requirements, eliminates noise, ensures the stability and consistency of the data, further improves the usability of the data, and provides more valuable information for subsequent analysis and modeling; the cleaned data can support various different types of analysis, such as statistical analysis, trend prediction, anomaly detection, etc. Through further analysis of the data by machine learning models, potential laws and trends can be discovered, providing in-depth insights for device management and decision-making.

[0104] Adopt machine learning to predict the failure risk of equipment. For different equipment, customized prediction models are established respectively. Conduct a life cycle analysis of the equipment, predict the expected life of the equipment, and for old equipment, establish a life extension prediction model through historical failure data as:

[0105] By analyzing the historical data of the equipment, extract the life cycle characteristics of the equipment, and combine survival analysis, regression models, and time series analysis machine learning algorithms to predict the failure risk of the equipment based on these characteristics;

[0106] Identify the failure modes of the equipment through anomaly detection algorithms such as Isolation Forest, K-means clustering, and One-Class SVM in machine learning, and predict the remaining life of the equipment through linear regression, random forest regression, and deep neural network regression models;

[0107] Use a regression model or a classification model to predict whether a device will malfunction. By detecting abnormal fluctuations during the operation of the device, predict the upcoming malfunctions of the device. Based on the historical failure records of the device, predict when to perform maintenance or replace components to extend the service life of the device.

[0108] Through accurate fault prediction, the fault risks of the device can be identified in advance, reducing the occurrence of unexpected malfunctions; predicting the remaining life and fault risks of the device allows for targeted maintenance and component replacement before a fault occurs, avoiding high repair costs and production losses caused by device failures; based on the life extension prediction model, the service life of the device can be extended by optimizing the maintenance timing, replacing components in a timely manner, etc.; through accurate fault prediction, the maintenance and downtime arrangements of the device can be optimized, avoiding waste of resources caused by sudden failures. Predictive maintenance can better arrange the maintenance plan, making device maintenance more efficient and precise; the entire process is automated through machine learning algorithms, which can dynamically learn based on the historical data of the device, automatically adjust the prediction model, and improve the accuracy of prediction; through data-driven fault prediction and life extension analysis, managers can make more scientific decisions before device failures occur, prepare spare parts, allocate personnel, and adjust production plans in advance, providing decision support for device operation and maintenance; by monitoring abnormal fluctuations during device operation, potential safety hazards can be detected in a timely manner, avoiding safety accidents caused by device failures, and ensuring the safety of the production environment and employees.

[0109] Generate intelligent decision-making suggestions based on big data analysis, send to property managers for maintenance and resource allocation according to the device health status, and automatically generate maintenance plans and priorities, and automatically dispatch maintenance personnel as follows:

[0110] Through installed sensors and Internet of Things technology, collect the operation data of the device in real time, integrate the historical failure records, maintenance logs, repair history, and usage of the device, analyze the health status of the device using machine learning algorithms, and calculate the fault risks and remaining life of the device;

[0111] Based on the device health status, fault prediction results, and resource utilization, the intelligent decision-making model generates maintenance suggestions, including whether to perform maintenance or replace components, the fault priority of the device, and the dispatched resources;

[0112] Automatically generate regular or emergency maintenance plans according to the fault risks, remaining life of the device, and device importance, and allocate maintenance tasks according to priorities. According to factors such as maintenance requirements, personnel capabilities, and resource availability, use linear programming and genetic algorithm optimization algorithms to allocate maintenance personnel and spare part resources;

[0113] Automatically dispatch maintenance personnel to perform maintenance tasks according to real-time data, maintenance plans, and resource allocation results.

[0114] Through real-time data collection and analysis, it is possible to accurately monitor the health status and failure risks of equipment, perform repairs and maintenance in a timely manner, reduce sudden equipment failures and equipment downtime; by predicting the failure time and remaining life of equipment, it is possible to carry out regular or emergency repairs in advance, avoiding the expensive repair costs and downtime losses caused by equipment failures; by carrying out predictive maintenance in a timely manner, it ensures that the equipment can operate in the best condition, thereby extending the service life of the equipment; by adopting intelligent decision-making and optimization algorithms, it is possible to reasonably allocate resources according to maintenance requirements, personnel capabilities and resource availability, avoiding waste and duplicate allocation of resources; based on big data analysis and machine learning models, the system can provide accurate and reliable maintenance suggestions and automatically generate emergency and regular maintenance plans, enabling managers to make decisions more quickly and effectively; this solution helps property managers achieve intelligent management through automated data collection, analysis and decision-making, thereby improving management levels and work efficiency.

[0115] When potential failures are predicted, issue early warnings in real time, remind relevant personnel to take actions via text messages or apps, and record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets as follows:

[0116] Analyze the real-time and historical data of the equipment based on the machine learning model. When the model identifies potential failure risks in the equipment, the system automatically triggers an early warning, analyzes the health status of the equipment in real time, and evaluates the severity and likelihood of the failure;

[0117] When potential failures are detected and predicted, notify maintenance personnel of the tasks and relevant details via text messages, emails or dedicated apps, including equipment problem descriptions, repair priorities, repair times. After maintenance personnel execute the repair tasks, update the repair progress in real time;

[0118] The system details and records the specific operations and processing procedures of each repair, ensures the integrity and traceability of the repair history, and integrates the historical maintenance data of all equipment, including repair frequencies, failure types, repair costs. Through data analysis, identify potential maintenance patterns and trends;

[0119] Through the analysis of historical maintenance data, the system can predict the future maintenance requirements of the equipment, and optimize the maintenance plan according to the equipment usage, failure probability and budget constraints. Based on the new maintenance data and failure records, the system continuously learns and optimizes the prediction model.

[0120] Through real-time early warning and intelligent fault prediction, maintenance personnel can take actions before serious equipment failures occur, reducing the number of emergency repairs and equipment downtime; through predictive maintenance, unnecessary maintenance and replacements are reduced, over-maintenance is avoided, thus reducing maintenance costs; by promptly detecting potential faults and performing repairs, the service life of equipment can be greatly extended, and equipment scrapping and downtime caused by faults can be reduced; the system helps managers discover common fault patterns and maintenance trends of equipment through in-depth analysis of historical maintenance data, thereby optimizing maintenance strategies; the system continuously optimizes the fault prediction model through machine learning, adjusts the prediction accuracy according to new data, and makes future maintenance plans more accurate and efficient; the system records all maintenance operations to ensure the integrity and traceability of maintenance history, and managers can view the equipment health status, maintenance records, and maintenance costs at any time; based on predictive analysis and historical maintenance data, the system can accurately predict future equipment maintenance requirements, thus formulating reasonable maintenance plans and budgets.

[0121] The operation formula of the property maintenance prediction system based on big data is:

[0122] Fault prediction is achieved through regression analysis or classification algorithms. The formula for linear regression is:

[0123]

[0124] Where:

[0125] is the prediction target;

[0126] βo is the intercept term;

[0127] β1, β2,... are the regression coefficients of the model;

[0128] are the input features;

[0129] ∈ is the error term;

[0130] For classification problems, logistic regression is used to predict the probability of faults. The formula for logistic regression is:

[0131]

[0132] Where:

[0133] is the probability of a fault occurring;

[0134] Other symbols are the same as those in linear regression;

[0135] Based on the scheduling of optimized maintenance resources, using linear programming or genetic algorithm methods, in terms of scheduling and resource allocation, linear programming is used to maximize efficiency or minimize cost. The formula for linear programming is:

[0136]

[0137] Where:

[0138] Are the coefficients of the objective function;

[0139] Are the decision variables;

[0140] Are the coefficients in the constraint matrix;

[0141] b1, b2,... Are the resource or time constraints;

[0142] By solving this linear programming problem, the system determines how to allocate resources to achieve cost minimization or efficiency maximization;

[0143] The formula for the fitness function is:

[0144]

[0145] Where:

[0146] Is the solution Of the fitness;

[0147] Are the weights of each objective;

[0148] Are the various parts of the solution;

[0149] For time series data predicting equipment failures, common methods include ARIMA. The formula for the ARIMA model is:

[0150] Y t = μ + φ1Y t＝1 + φ2Y t＝2 +... + φ p Y t＝p + θ1∈ t＝1 + θ2∈ t＝2 +... + θ q ∈ t＝q + ∈ t

[0151] Where:

[0152] Y t Is the observed value at time t;

[0153] μ is a constant term;

[0154] φ1, φ2,..., φ p are autoregressive parameters;

[0155] θ1, θ2,..., θ q are moving average parameters;

[0156] ∈ t is white noise;

[0157] Through the ARIMA model, analyze the historical data of equipment failures and predict the future failure time that is about to occur.

[0158] Through regression analysis and logistic regression, the probability of equipment failure and the time of failure occurrence can be accurately predicted, improving the prediction accuracy; through linear programming and genetic algorithms, the scheduling and allocation of maintenance resources can be optimized; the data-driven prediction model can identify potential failures in advance, reducing the costs of sudden failures and emergency repairs; by continuously monitoring and predicting the health status of equipment, preventive maintenance measures can be taken in a timely manner, extending the service life of equipment, reducing equipment downtime, and improving the overall operation efficiency of equipment; the system provides accurate and real-time data support for decision-makers by analyzing historical maintenance data, equipment status data, and resource scheduling data, helping to formulate more scientific and efficient maintenance plans; through continuously optimized models and algorithms, the system can flexibly adjust the maintenance plan and budget, making adjustments according to the equipment usage situation and failure prediction results, thereby improving the adaptability and accuracy of the maintenance plan.

[0159] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the big data-based property maintenance prediction system as described above.

[0160] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the big data-based property maintenance prediction system as described above.

[0161] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0162] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the functions described above.

[0163] The above embodiments of the present invention do not limit the protection scope of the present invention. The embodiments of the present invention are not limited to this. All kinds of modifications, substitutions, or changes made to the above structure of the present invention according to the above content of the present invention, in accordance with the common general knowledge and conventional means in the art, without departing from the above basic technical idea of the present invention, shall fall within the protection scope of the present invention.

Claims

1. A property maintenance prediction system based on big data, characterized in that, It includes: The data collection and transmission module is used to install various intelligent sensors, monitor the operating status of equipment and the environment in the property in real time, and transmit the data collected by the equipment sensors to the cloud or local storage through a wireless network; The data processing and analysis module is used to store the historical data of the equipment based on a distributed storage architecture, perform data processing, and process the original data through data cleaning technology; The prediction and optimization model module is used to adopt machine learning to predict the failure risk of equipment. For different equipment, customized prediction models are established respectively to analyze the life cycle of the equipment, predict the expected life of the equipment, and for old equipment, establish a life extension prediction model through historical failure data; The decision support module is used to generate intelligent decision-making suggestions based on big data analysis, send maintenance and resource allocation according to the equipment health status to property management personnel, automatically generate maintenance plans and priorities, and automatically dispatch maintenance personnel; The execution module is used to issue a real-time warning when a potential failure is predicted, remind relevant personnel to take actions by means of text messages or APPs, record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets.

2. The property maintenance prediction system based on big data according to claim 1, wherein Installing various intelligent sensors, monitoring the operating status of equipment and the environment in the property in real time, and transmitting the data collected by the equipment sensors to the cloud or local storage is: Collect equipment and environmental data through installed temperature and humidity sensors, vibration sensors, and air pressure sensors, and use Wi-Fi, LoRa, Zigbee, 5G wireless communication technologies to transmit the data from the sensors to the cloud or local storage in real time; Adopt AWS IoT, Azure IoT, Google Cloud IoT cloud platforms to store and manage equipment and environmental data.

3. The property maintenance prediction system based on big data according to claim 2, characterized in that Based on the historical data of the equipment stored in a distributed storage architecture, perform data processing, and process the original data through data cleaning technology is: Adopt a deduplication algorithm to remove duplicates by comparing the timestamps and device ID fields of the data, use the mean, median, and most common values to fill in missing values, and use regression models and KNN machine learning models to predict the missing values; Adopt the Z-score or IQR method to detect outliers, and use isolation forest, K-means clustering, and DBSCAN algorithms to automatically identify abnormal data points; Convert the data to data with zero mean and unit variance, and the formula is: Where X is the original data, μ is the mean of the data, and σ is the standard deviation; Compress the data to a specified range, and the formula is: Align time series data to ensure that the timestamps of different data sources are consistent, resample the data according to the timestamps, unify the time intervals of the data, perform interpolation processing at the time points where data is missing, and smooth the data through moving average and weighted average methods.

4. The property maintenance prediction system based on big data according to claim 3, characterized in that, Adopt machine learning to predict the failure risk of equipment. For different equipment, customized prediction models are established respectively to analyze the life cycle of the equipment, predict the expected life of the equipment, and for old equipment, establish a life extension prediction model through historical failure data is: By analyzing the historical data of the device, extracting the life cycle characteristics of the device, and combining survival analysis, regression models, and time series analysis machine learning algorithms, predict the failure risk of the device based on these characteristics; Identify the failure modes of the device through isolation forest, K-means clustering, and One-Class SVM anomaly detection algorithms in machine learning, and predict the remaining life of the device through linear regression, random forest regression, and deep neural network regression models; Use regression models or classification models to predict whether the device will fail, predict the upcoming failures of the device by detecting abnormal fluctuations during the operation of the device, and predict when to perform maintenance or replace components based on the historical failure records of the device to extend the service life of the device.

5. The property maintenance prediction system based on big data according to claim 4, characterized in that, Generate intelligent decision-making suggestions based on big data analysis, send to property managers for maintenance and resource allocation according to the device health status, and automatically generate maintenance plans and priorities, and automatically schedule maintenance personnel as: Through installed sensors and Internet of Things technology, collect the operation data of the device in real time, integrate the historical failure records, maintenance logs, repair history, and usage of the device, and analyze the health status of the device with machine learning algorithms to calculate the failure risk and remaining life of the device; Based on the device health status, failure prediction results, and resource utilization, the intelligent decision-making model generates maintenance suggestions, including whether to perform maintenance or replace components, the failure priority of the device, and the scheduled resources; Automatically generate regular or emergency maintenance plans according to the failure risk, remaining life of the device, and device importance, allocate maintenance tasks according to priorities, and allocate maintenance personnel and spare parts resources using linear programming and genetic algorithm optimization algorithms according to maintenance requirements, personnel capabilities, and resource availability factors; Automatically schedule maintenance personnel to perform maintenance tasks according to real-time data, maintenance plans, and resource allocation results.

6. The property maintenance prediction system based on big data according to claim 5, characterized in that, When potential failures are predicted, issue early warnings in real time, remind relevant personnel to take actions via text messages or apps, and record all maintenance and repair operations, analyze historical maintenance data, and optimize future maintenance plans and budgets as: Analyze the real-time data and historical data of the device based on machine learning models. When the model identifies potential failure risks in the device, the system automatically triggers an early warning, analyzes the health status of the device in real time, and evaluates the severity and likelihood of the failure; When potential failures are detected and predicted, notify maintenance personnel of tasks and relevant details via text messages, emails, or dedicated apps, including device problem descriptions, maintenance priorities, and maintenance times. After maintenance personnel perform maintenance tasks, update the maintenance progress in real time; The system details record the specific operations and processing procedures of each maintenance to ensure the integrity and traceability of the maintenance history, and integrate the historical maintenance data of all devices, including maintenance frequencies, failure types, and maintenance costs. Through data analysis, identify potential maintenance patterns and trends; Through the analysis of historical maintenance data, the system can predict the future maintenance requirements of the device, optimize the maintenance plan according to the usage of the device, failure probability, and budget constraints, and the system continuously learns and optimizes the prediction model based on new maintenance data and failure records.

7. The property maintenance prediction system based on big data according to claim 6, wherein, The operation formula is: Fault prediction is achieved through regression analysis or classification algorithms. The formula for linear regression is: Where: is the predicted target; βo is the intercept term; is the regression coefficient of the model; is the input feature; ∈ is the error term; For classification problems, logistic regression is used to predict the probability of faults. The formula for logistic regression is: Where: is the probability of a failure occurring; Other symbols are the same as those in linear regression; Based on the scheduling of optimized maintenance resources, linear programming or genetic algorithm methods are adopted. In terms of scheduling and resource allocation, linear programming is used to maximize efficiency or minimize cost. The formula for linear programming is: Where: is the coefficient of the objective function; is a decision variable; is a coefficient in the constraint matrix; For resource or time constraints; By solving this linear programming problem, the system determines how to allocate resources to achieve cost minimization or efficiency maximization; The formula for the fitness function is: Where: To solve the fitness; The weight for each target; For each part of the solution; For time series data predicting equipment faults, common methods include ARIMA. The formula for the ARIMA model is: Y t = μ + φ1Y t＝1 + φ2Y t＝2 +... + φ p Y t＝p + θ1∈ t＝1 + θ2∈ t＝2 +... +θ q ∈ t＝q +∈ t Where: Y t is the observed value at time t; μ is the constant term; φ1, φ2,..., φ p are autoregressive parameters; θ1, θ2,..., θ q are the moving average parameters; ∈ t is white noise; Through the ARIMA model, historical data of equipment faults is analyzed to predict the time of upcoming faults in the future.

8. An electronic device, characterized in that, Including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a property maintenance prediction system based on big data as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements a property maintenance prediction system based on big data as described in any one of claims 1-7.

Images