Air conditioning equipment fault early warning and intelligent operation and maintenance service system
By combining multi-dimensional data collection and an improved deep learning model with a fault diagnosis knowledge base, the problems of incomplete data collection, inaccurate early warning, and lack of targeted maintenance solutions in the air conditioning equipment operation and maintenance system have been solved. This has enabled efficient and accurate fault early warning and personalized operation and maintenance, reducing costs and improving equipment safety and user satisfaction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DONGGUAN XIANGKE INTELLIGENT CONTROL EQUIP CO LTD
- Filing Date
- 2026-01-26
- Publication Date
- 2026-05-12
AI Technical Summary
Existing air conditioning equipment operation and maintenance systems suffer from problems such as incomplete data collection, low accuracy of fault early warning, and lack of targeted operation and maintenance solutions, resulting in high equipment repair costs and numerous safety hazards.
It employs a multi-dimensional data acquisition module, an intelligent data preprocessing module, a precise fault early warning module, a dynamic operation and maintenance decision-making module, a cloud-based collaborative management module, and a user interaction terminal module. Combined with an improved deep learning model and a fault diagnosis knowledge base, it achieves multi-dimensional data acquisition, precise fault early warning, and personalized operation and maintenance decision-making.
It achieves comprehensive and accurate data collection on the operating status of air conditioning equipment, efficient and accurate fault early warning, and highly targeted operation and maintenance solutions, thereby reducing operation and maintenance costs and improving the safety, stability, and user experience of the equipment.
Smart Images
Figure CN122015229A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent operation and maintenance and fault diagnosis technology, and in particular to an air conditioning equipment fault early warning and intelligent operation and maintenance service system. Background Technology
[0002] As the core equipment for indoor environmental regulation, air conditioning equipment is widely used in various places such as residences, commercial complexes, industrial plants, and medical institutions. Its operational stability directly affects environmental comfort and the normal operation of production activities. With the increase in the service life of air conditioning equipment and the complex changes in the operating environment, problems such as compressor failure, pipeline leakage, decreased cooling efficiency, and electrical system failure occur frequently. These failures not only cause the air conditioning equipment to shut down and affect the user experience, but may also cause permanent damage to the equipment due to the expansion of the failure, increase maintenance costs, and even cause safety accidents.
[0003] Currently, the operation and maintenance of air conditioning equipment mainly relies on traditional manual inspection and post-event repair methods, which have obvious limitations: manual inspection is inefficient, requires a lot of manpower, and is limited by the experience of the inspectors, making it difficult to discover potential faults in the equipment.
[0004] The reactive maintenance mode can only remedy the problem after it occurs, and cannot provide early warnings, resulting in long equipment downtime and affecting normal use.
[0005] To address the aforementioned issues, some companies have introduced simplified air conditioning operation and maintenance systems. However, existing systems still suffer from three major flaws: incomplete data collection, with most systems only collecting a small amount of key electrical parameters or temperature data, ignoring parameters crucial for fault diagnosis such as core component vibration, dust concentration in the operating environment, and pipeline pressure changes. Furthermore, the lack of a data calibration mechanism makes it difficult to guarantee the accuracy of the collected data.
[0006] The accuracy of fault warning is low. Existing systems mostly use traditional threshold judgment or simple machine learning models, which cannot effectively capture the complex spatial features and long-term and short-term time dependencies in the data, and are prone to false warnings or missed warnings.
[0007] The operation and maintenance solutions lack specificity. Most systems use standardized solution templates without fully incorporating personalized information such as equipment model, service life, operating environment, and historical fault records. This results in poor applicability of the operation and maintenance solutions, low maintenance efficiency, and high operation and maintenance costs.
[0008] Therefore, in response to the problems of incomplete data collection, low accuracy of fault early warning, and lack of targeted operation and maintenance solutions in existing air conditioning operation and maintenance systems, it is necessary to develop an intelligent service system that can achieve comprehensive multi-dimensional data collection, accurate fault early warning, and personalized operation and maintenance decision-making. This system has significant practical significance and application value for improving the operation and maintenance efficiency of air conditioning equipment, reducing operation and maintenance costs, and ensuring the safe and stable operation of equipment. Summary of the Invention
[0009] The purpose of this invention is to at least solve one of the technical problems existing in the prior art, and to provide an air conditioning equipment fault early warning and intelligent operation and maintenance service system that can solve the above-mentioned problems.
[0010] To achieve the above objectives, the present invention provides the following technical solution: an air conditioning equipment fault early warning and intelligent operation and maintenance service system, comprising: a multi-dimensional data acquisition module, an intelligent data preprocessing module, a precise fault early warning module, a dynamic operation and maintenance decision module, a cloud-based collaborative management module, a user interaction terminal module, and a fault diagnosis knowledge base module;
[0011] The multi-dimensional data acquisition module is used to collect the operating status parameters, environmental impact parameters and basic information of the entire life cycle of the air conditioning equipment from all aspects, and transmit the collected data to the intelligent data preprocessing module in real time.
[0012] The intelligent data preprocessing module is used to perform layered cleaning, adaptive noise reduction, standardized transformation and multi-source data fusion processing on the collected data, and output a high-quality standardized dataset, which is transmitted to the accurate fault early warning module and the cloud collaborative management module respectively.
[0013] The precise fault early warning module performs real-time analysis of standardized datasets based on an improved deep learning fusion model, identifies the types and levels of equipment fault risks, generates early warning information including fault location and risk trends, and pushes it synchronously to the cloud collaborative management module and the user interaction terminal module.
[0014] The dynamic operation and maintenance decision-making module, relying on the fault diagnosis knowledge base module and combining the historical data, early warning information and basic equipment information stored in the cloud collaborative management module, generates targeted operation and maintenance plans and dynamically optimizes the plans based on operation and maintenance implementation feedback.
[0015] The cloud-based collaborative management module is used to store all data, early warning information, and operation and maintenance plans, realize data interaction and collaborative scheduling between modules, and provide data security protection functions.
[0016] The user interaction terminal module is used to display differentiated information to users with different roles and supports user operation command input and operation and maintenance process traceability;
[0017] The fault diagnosis knowledge base module stores typical fault cases of air conditioning equipment, fault feature database, and operation and maintenance solution templates, providing knowledge support for fault identification and operation and maintenance decisions.
[0018] Preferably, the multi-dimensional data acquisition module includes: a core component sensor group, an environmental sensing unit, an electrical parameter acquisition unit, and an information input and calibration unit;
[0019] The core component sensor group includes vibration sensors, temperature sensors, and pressure sensors installed on the compressor, evaporator, condenser, and throttling device, used to collect real-time operating parameters of the core components;
[0020] The environmental sensing unit is used to collect temperature, humidity, dust concentration and air pressure data of the air conditioning operating environment;
[0021] The electrical parameter acquisition unit is used to acquire the air conditioner's operating current, voltage, power, and power factor;
[0022] The information entry and calibration unit is used to enter the equipment model, factory parameters, installation information, and historical maintenance records, and to periodically calibrate the sensor data.
[0023] Preferably, the processing flow of the intelligent data preprocessing module is as follows:
[0024] The first step is to adopt a hierarchical cleaning strategy based on the 3σ criterion and the isolated forest algorithm to remove abnormal data, and to fill missing data by interpolation or migration based on similar working conditions according to the data type.
[0025] The second step is to use an adaptive wavelet threshold denoising algorithm to dynamically adjust the wavelet basis function and the number of decomposition layers according to the data noise intensity to achieve accurate denoising.
[0026] The third step is to use the Z-score standardization method to transform data of different dimensions to the standard normal distribution interval;
[0027] The fourth step involves using a multi-source data fusion algorithm based on an attention mechanism to assign dynamic weights to operating parameters, environmental parameters, and basic information, and then fusing them to output a standardized dataset.
[0028] Preferably, the improved deep learning fusion model in the accurate fault warning module is a CNN-BiLSTM-Attention fusion model, which includes a CNN feature extraction layer, a BiLSTM temporal analysis layer, an attention enhancement layer, and a risk decision layer connected in sequence.
[0029] The CNN feature extraction layer is used to extract spatial features and local correlation features from the data;
[0030] The BiLSTM time series analysis layer is used to capture the long-term and short-term time dependencies of the data.
[0031] The attention enhancement layer is used to strengthen the weights of key fault features;
[0032] The risk decision-making layer outputs the fault risk level (no risk, low risk, medium risk, high risk) and the corresponding fault type.
[0033] Preferably, the training process of the improved deep learning fusion model includes:
[0034] S1. Construct a labeled dataset containing normal operation data, potential fault data, and typical fault data, and divide it into training set and test set in an 8:2 ratio.
[0035] S2, initialize model parameters, set the learning rate to 0.001~0.003, the number of iterations to 120~150, and the batch size to 32~64;
[0036] S3 uses the AdamW optimizer and Focal Loss loss function for model training and prevents overfitting through an early stopping mechanism.
[0037] S4. Use the test set to verify the model performance. Training is completed when the fault identification accuracy is ≥96% and the recall is ≥95%. Otherwise, adjust the model structure and parameters and retrain.
[0038] Preferably, the process by which the dynamic operation and maintenance decision module generates an operation and maintenance plan is as follows:
[0039] S1 receives warning information from the accurate fault warning module and retrieves matching fault characteristics and cases from the fault diagnosis knowledge base module.
[0040] S2 combines the device's historical operating data, maintenance records, and current operating status stored in the cloud-based collaborative management module to analyze the root cause of the fault;
[0041] S3 generates a targeted solution based on fault type, equipment service life, operating environment and user maintenance cost budget, including fault handling steps, required spare parts, tool list and operating procedures.
[0042] S4 receives feedback from operations and maintenance personnel on the effectiveness of the solution implementation, iterates and optimizes the solution, and updates it to the fault diagnosis knowledge base module.
[0043] Preferably, the cloud-based collaborative management module includes: a distributed data storage unit, a real-time data interaction unit, a data encryption and access control unit, and a data backup unit;
[0044] The distributed data storage unit adopts a hybrid storage architecture of MySQL and MongoDB to store structured data and unstructured data respectively.
[0045] The real-time data interaction unit uses the MQTT communication protocol to achieve low-latency data transmission.
[0046] The data encryption and access control unit uses the AES-256 encryption algorithm to encrypt data during transmission and storage, and assigns user permissions based on role-based access control policies.
[0047] The data backup unit employs a multi-replica backup mechanism in different locations to ensure data security and integrity.
[0048] Preferably, the user interaction terminal module includes a web management terminal for maintenance personnel, a mobile APP for ordinary users, and a device management backend;
[0049] The web management terminal for maintenance personnel supports real-time monitoring of equipment operation data, viewing of fault warning details, editing and distribution of maintenance plans, entry of maintenance records, and updating of the knowledge base.
[0050] The mobile app for regular users supports fault warning alerts, viewing simple maintenance suggestions, submitting repair requests, and checking maintenance progress.
[0051] The device management backend supports system parameter configuration, sensor calibration management, and module status monitoring.
[0052] Preferably, the fault diagnosis knowledge base module adopts an incremental update mechanism, including a fault feature library, a case library, and a solution template library;
[0053] The fault feature database stores typical fault feature parameter thresholds for different models of air conditioning equipment.
[0054] The case library stores historical fault handling cases and effectiveness evaluation data;
[0055] The solution template library stores standardized operation and maintenance solution templates for different fault types and equipment models, which can be dynamically adjusted according to actual working conditions.
[0056] Preferably, the information input and calibration unit has an automatic calibration reminder function, which generates periodic calibration reminders based on the sensor's usage time, ambient humidity, and the stability of the collected data, and records the calibration results;
[0057] When the deviation of the sensor data exceeds a preset threshold, a sensor fault warning is issued, prompting replacement or repair.
[0058] Compared with the prior art, the beneficial effects of the present invention are:
[0059] 1. This air conditioning equipment fault early warning and intelligent operation and maintenance service system has comprehensive and accurate data collection: Through multi-dimensional sensor layout and regular calibration mechanism, it realizes comprehensive collection of basic information on the operating status, environmental parameters, electrical performance and the entire life cycle of air conditioning equipment, and the accuracy of data collection is improved to over 98%, which solves the problems of incomplete data collection and poor accuracy of existing systems.
[0060] 2. The air conditioning equipment fault early warning and intelligent operation and maintenance service system provides accurate and efficient fault early warning: The improved CNN-BiLSTM-Attention fusion model can fully capture the complex features of the data, with a fault identification accuracy of ≥96% and a recall rate of ≥95%, effectively reducing the false alarm and missed alarm rates, realizing early warning and accurate location of faults, and gaining sufficient time for operation and maintenance work.
[0061] 3. This air conditioning equipment fault early warning and intelligent operation and maintenance service system has highly targeted operation and maintenance solutions: it generates operation and maintenance solutions based on the fault diagnosis knowledge base and the personalized characteristics of the equipment, and combines implementation feedback for dynamic optimization, which significantly improves the applicability of the solutions, reduces operation and maintenance costs by more than 30%, and improves operation and maintenance efficiency by more than 50%, solving the problem of the lack of targeted operation and maintenance solutions in existing systems.
[0062] 4. The air conditioning equipment fault early warning and intelligent operation and maintenance service system is data-secure and reliable: the encryption, access control and off-site backup mechanism of the cloud collaborative management module ensure the security of data transmission and storage, and avoid data leakage and loss; the hybrid storage architecture and low-latency communication protocol realize efficient data storage and real-time interaction, and improve the overall operating efficiency of the system.
[0063] 5. The air conditioning equipment fault early warning and intelligent operation and maintenance service system offers an excellent user experience: the differentiated user interaction terminal design meets the usage needs of users with different roles, the operation is convenient and efficient, the operation and maintenance process is traceable, and the user satisfaction is improved. Attached Figure Description
[0064] The present invention will be further described below with reference to the accompanying drawings and embodiments:
[0065] Figure 1 This is a schematic block diagram of an air conditioning equipment fault early warning and intelligent operation and maintenance service system according to the present invention. Detailed Implementation
[0066] This section will describe in detail specific embodiments of the present invention. Preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and overall technical solution of the present invention, but they should not be construed as limiting the scope of protection of the present invention.
[0067] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.
[0068] In the description of this invention, terms such as greater than, less than, and exceeding are understood to exclude the stated number, while terms such as above, below, and within are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.
[0069] In the description of this invention, unless otherwise explicitly defined, terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.
[0070] Please refer to Figure 1 The present invention provides a technical solution: an air conditioning equipment fault early warning and intelligent operation and maintenance service system, including a multi-dimensional data acquisition module, an intelligent data preprocessing module, a precise fault early warning module, a dynamic operation and maintenance decision module, a cloud collaborative management module, a user interaction terminal module, and a fault diagnosis knowledge base module. Each module realizes data interaction through wired or wireless communication.
[0071] Multi-dimensional data acquisition module: Overcoming the shortcomings of existing systems with single data acquisition, it constructs a comprehensive acquisition system, realizing multi-dimensional data acquisition of air conditioning equipment operating status, environmental impact and electrical performance through core components such as sensor group, environmental sensing unit and electrical parameter acquisition unit;
[0072] An information entry and calibration unit has been added to enter basic information about the entire life cycle of the equipment. At the same time, sensor data is calibrated regularly to ensure the accuracy and completeness of the collected data and provide reliable data support for subsequent accurate analysis.
[0073] The intelligent data preprocessing module addresses outliers, missing values, and noise interference in the collected data. It employs a comprehensive processing strategy that integrates layered cleaning, adaptive denoising, standardization transformation, and attention mechanism fusion. The cleaning method combines the 3σ criterion with the isolated forest algorithm to accurately remove outlier data. The adaptive wavelet threshold denoising algorithm dynamically adjusts parameters based on noise intensity to achieve efficient denoising. Standardization processing unifies the data units, and the attention mechanism fusion algorithm strengthens the weight of key data, outputting a high-quality standardized dataset.
[0074] Precise fault warning module: It adopts an improved CNN-BiLSTM-Attention fusion model to solve the problem that traditional models cannot effectively capture complex features. The CNN layer extracts data space features and local correlation features, the BiLSTM layer captures long-term and short-term time dependencies, the attention layer strengthens key fault features, and finally the risk decision layer accurately outputs the fault risk level and fault type, so as to achieve early warning and accurate location of faults.
[0075] Dynamic Operation and Maintenance Decision Module: Relying on the fault diagnosis knowledge base module, combined with equipment historical data, real-time early warning information and basic information, it breaks through the limitations of the existing system's standardized solutions. By analyzing the root causes of faults, combined with the personalized characteristics of equipment and user cost budgets, it generates targeted operation and maintenance solutions.
[0076] At the same time, the operation and maintenance plan is dynamically optimized based on the feedback from the operation and maintenance implementation, so as to realize the iterative upgrade of the operation and maintenance plan and improve the applicability of the plan;
[0077] Cloud-based collaborative management module: As the core hub of the system, it adopts a hybrid storage architecture to achieve efficient storage of all data and ensures real-time data interaction between modules through a low-latency communication protocol;
[0078] Add data encryption, access control and off-site backup functions to ensure data security and reliability, and enable efficient collaborative operation of various modules;
[0079] User interaction terminal module: Different interaction terminals are designed to meet the needs of users with different roles. The web management terminal for operation and maintenance personnel provides full-featured operation and maintenance management support, while the mobile APP for ordinary users enables convenient information viewing and application submission. The device management backend ensures stable system operation and improves the user experience for different users.
[0080] Fault Diagnosis Knowledge Base Module: Adopting an incremental update mechanism, it constructs a knowledge base system that includes a fault feature library, a case library, and a solution template library, providing rich knowledge support for fault identification and operation and maintenance decisions. At the same time, the knowledge base is continuously improved through incremental updates, enhancing the long-term reliability and adaptability of the system.
[0081] Example:
[0082] This embodiment provides an air conditioning equipment fault early warning and intelligent operation and maintenance service system, including a multi-dimensional data acquisition module, an intelligent data preprocessing module, a precise fault early warning module, a dynamic operation and maintenance decision module, a cloud collaborative management module, a user interaction terminal module, and a fault diagnosis knowledge base module. Each module realizes data interaction through a 5G network and a WiFi network.
[0083] Multi-dimensional data acquisition module:
[0084] Core component sensor group: High-precision vibration sensor (model: ADXL355), temperature sensor (model: DS18B20), and pressure sensor (model: MPX5700) are selected and installed in key positions of the air conditioning compressor, evaporator, condenser, and throttling device, respectively.
[0085] The vibration sensor sampling frequency is set to 100Hz, and the acquisition range is ±8g.
[0086] The temperature sensor has a measurement range of -55℃ to 125℃ and an accuracy of ±0.5℃.
[0087] The pressure sensor has a measurement range of 0~700kPa and an accuracy of ±1%FS;
[0088] Environmental sensing unit: An integrated environmental sensor (model: BME280) is selected and installed near the indoor and outdoor units of the air conditioner to collect ambient temperature (-40℃~85℃, accuracy ±0.1℃), humidity (0~100%RH, accuracy ±3%RH), air pressure (300~1100hPa, accuracy ±1hPa) and dust concentration sensor (model: PMS5003), with a collection range of 0.3~10μm;
[0089] Electrical parameter acquisition unit: A three-phase power acquisition module (model: DTSU666) is selected to acquire the air conditioner's operating current (0~100A), voltage (0~450V), power and power factor, with an accuracy of 0.5 class;
[0090] Information entry and calibration unit: adopts an industrial touch screen (model: TPC1561Hi), and maintenance personnel can enter information such as equipment model, manufacturing date, rated parameters, installation location, and historical maintenance records through the touch screen;
[0091] The system generates calibration reminders based on sensor usage time (every 3 months) and ambient humidity (shortened to 1 month when humidity > 80%). It uses standard calibration equipment to calibrate the sensor, records the calibration results, and issues a sensor fault warning when the calibration deviation exceeds ±2%.
[0092] Intelligent data preprocessing module:
[0093] Data processing is achieved using an industrial control computer (model: IPC-610L). The specific process is as follows:
[0094] Layered cleaning: First, the 3σ criterion is used to remove outlier data that deviates from the mean by 3 times the standard deviation. Then, the isolated forest algorithm is used to identify and remove hidden outlier data.
[0095] For missing data, linear interpolation is used to fill numerical data, and migration imputation based on similar working conditions is used for categorical data.
[0096] Adaptive denoising: The db4 wavelet basis function is used to dynamically adjust the number of decomposition layers (2 to 4 layers) according to the data noise intensity. The wavelet coefficients are processed by an improved threshold function (λ=σ√(2lnN), where σ is the noise standard deviation and N is the data length) to reconstruct the denoised data.
[0097] Standardization transformation: The data is transformed to the standard normal distribution interval with a mean of 0 and a standard deviation of 1 by using the Z-score standardization formula x'=(x-μ) / σ (where x is the original data, μ is the mean, and σ is the standard deviation).
[0098] Multi-source data fusion: An attention-based fusion algorithm is adopted. The attention weights of each data source are calculated through a fully connected layer. The weights range from 0 to 1. The standardized data are weighted and summed according to the weights to obtain the fused standardized dataset, which is then transmitted to the accurate fault warning module and the cloud collaborative management module.
[0099] Precise fault early warning module:
[0100] Implemented using an embedded AI chip (model: NVIDIA Jetson Xavier NX), with a built-in improved CNN-BiLSTM-Attention fusion model:
[0101] Model construction: The CNN layer consists of 2 convolutional layers (3×3 kernels, with 32 and 64 neurons respectively) and 2 max pooling layers (2×2 pooling kernels); the BiLSTM layer consists of 2 hidden layers, each with 128 neurons.
[0102] The attention layer employs the Bahdanau attention mechanism;
[0103] The risk decision layer consists of one fully connected layer (64 neurons) and one output layer (softmax activation function, outputting 4 risk levels).
[0104] Model training: Collect operating data (including normal operation, potential faults, and typical faults) from 100 different models of air conditioners, totaling 1 million data points. After labeling, the data is divided into training and test sets in an 8:2 ratio.
[0105] The learning rate was set to 0.002, the number of iterations to 130, and the batch size to 64. The AdamW optimizer and Focal Loss loss function were used for training.
[0106] By using an early stopping mechanism (stopping training if the loss does not decrease after 10 consecutive rounds of testing) to prevent overfitting, the final model achieved a fault identification accuracy of 96.8% and a recall rate of 95.6%.
[0107] Fault warning: After receiving the standardized dataset, input the trained model, output the fault risk level and fault type (such as compressor failure, pipeline leakage, electrical failure, etc.), generate warning information containing fault location, risk trend and warning time, and transmit it to the cloud collaborative management module and user interaction terminal module through the 5G network.
[0108] Dynamic operation and maintenance decision-making module:
[0109] This is implemented using a cloud server (Alibaba Cloud ECS g6.xlarge), and it communicates and connects with the cloud-based collaborative management module and fault diagnosis knowledge base module.
[0110] Fault Analysis: Obtain early warning information, historical equipment operation data and basic information from the cloud, call up matching cases and fault characteristics in the fault diagnosis knowledge base, and determine the root cause of the fault through causal analysis algorithm;
[0111] Solution generation: Based on the fault type, equipment service life, operating environment and user's maintenance cost budget, the system calls a basic template from the solution template library, dynamically adjusts the fault handling steps, spare parts specifications and tool list, and generates a personalized maintenance solution.
[0112] Solution optimization: Operation and maintenance personnel provide feedback on the implementation effect of the solution through the web management terminal (such as whether the fault has been resolved and whether the operation and maintenance cost has been exceeded). The system uses reinforcement learning algorithm to optimize the solution and update it to the solution template library to achieve incremental updates of the knowledge base.
[0113] Cloud-based collaborative management module:
[0114] Implemented using Alibaba Cloud server clusters, including:
[0115] Distributed data storage unit: MySQL database stores structured data such as basic device information and operation and maintenance records, while MongoDB database stores unstructured data such as raw sensor data and early warning information. The storage capacity can be dynamically expanded.
[0116] Real-time data interaction unit: Adopts MQTT communication protocol, with data transmission latency ≤100ms, to realize real-time data interaction between modules;
[0117] Data encryption and access control unit: Employs AES-256 encryption algorithm to encrypt transmitted and stored data;
[0118] Based on the RBAC permission model, different operation permissions are assigned to operation and maintenance personnel, ordinary users, and administrators;
[0119] Data backup unit: It adopts a multi-replica backup mechanism in different locations, automatically backs up data to Alibaba Cloud OSS every day at midnight, and retains the backup data for 3 months to ensure data security;
[0120] User interaction terminal module:
[0121] Web management terminal for maintenance personnel: Adopting a B / S architecture, it supports access via a browser and includes functions such as real-time monitoring of equipment operation data, viewing of fault warning details, editing and distribution of maintenance plans, entry of maintenance records, sensor calibration management, and knowledge base updates.
[0122] Mobile app for general users: Supports Android and iOS systems, and features include fault warning alerts, viewing simple maintenance suggestions, submitting repair requests, checking maintenance progress, and providing feedback.
[0123] Equipment management backend: Supports system parameter configuration (such as sensor sampling frequency, early warning threshold), module status monitoring, log query and system upgrade;
[0124] Fault Diagnosis Knowledge Base Module:
[0125] The database uses MySQL for storage, including:
[0126] Fault Feature Library: Stores the characteristic parameter thresholds for 50 typical air conditioner faults, such as the vibration frequency range corresponding to compressor faults and the pressure change range corresponding to pipe leaks;
[0127] Case Library: Stores 2000+ historical fault handling cases, including fault symptoms, handling process, and implementation results;
[0128] Solution Template Library: Stores 50+ standardized operation and maintenance solution templates for different fault types and equipment models.
[0129] The workflow of this embodiment is as follows: A multi-dimensional data acquisition module collects various types of data in real time, calibrates them, and then transmits them to an intelligent data preprocessing module; the intelligent data preprocessing module performs full-process data processing and outputs a standardized dataset; the accurate fault early warning module analyzes the data using an improved fusion model, generates early warning information, and pushes it out; the dynamic operation and maintenance decision-making module combines knowledge base and cloud data to generate personalized operation and maintenance solutions; users view information, submit applications, and implement operation and maintenance through an interactive terminal; the cloud-based collaborative management module realizes system-wide data storage, interaction, and security; and the fault diagnosis knowledge base module continuously updates incrementally to improve system performance.
[0130] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.
Claims
1. An air conditioning equipment fault early warning and intelligent operation and maintenance service system, characterized in that, include: The system includes a multi-dimensional data acquisition module, an intelligent data preprocessing module, a precise fault early warning module, a dynamic operation and maintenance decision-making module, a cloud-based collaborative management module, a user interaction terminal module, and a fault diagnosis knowledge base module. The multi-dimensional data acquisition module is used to collect the operating status parameters, environmental impact parameters and basic information of the entire life cycle of the air conditioning equipment from all aspects, and transmit the collected data to the intelligent data preprocessing module in real time. The intelligent data preprocessing module is used to perform layered cleaning, adaptive noise reduction, standardized transformation and multi-source data fusion processing on the collected data, and output a high-quality standardized dataset, which is transmitted to the accurate fault early warning module and the cloud collaborative management module respectively. The precise fault early warning module performs real-time analysis of standardized datasets based on an improved deep learning fusion model, identifies the types and levels of equipment fault risks, generates early warning information including fault location and risk trends, and pushes it synchronously to the cloud collaborative management module and the user interaction terminal module. The dynamic operation and maintenance decision-making module, relying on the fault diagnosis knowledge base module and combining the historical data, early warning information and basic equipment information stored in the cloud collaborative management module, generates targeted operation and maintenance plans and dynamically optimizes the plans based on operation and maintenance implementation feedback. The cloud-based collaborative management module is used to store all data, early warning information, and operation and maintenance plans, realize data interaction and collaborative scheduling between modules, and provide data security protection functions. The user interaction terminal module is used to display differentiated information to users with different roles and supports user operation command input and operation and maintenance process traceability; The fault diagnosis knowledge base module stores typical fault cases of air conditioning equipment, fault feature database, and operation and maintenance solution templates, providing knowledge support for fault identification and operation and maintenance decisions.
2. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The multi-dimensional data acquisition module includes: a core component sensor group, an environmental sensing unit, an electrical parameter acquisition unit, and an information input and calibration unit; The core component sensor group includes vibration sensors, temperature sensors, and pressure sensors installed on the compressor, evaporator, condenser, and throttling device, used to collect real-time operating parameters of the core components; The environmental sensing unit is used to collect temperature, humidity, dust concentration and air pressure data of the air conditioning operating environment; The electrical parameter acquisition unit is used to acquire the air conditioner's operating current, voltage, power, and power factor; The information entry and calibration unit is used to enter the equipment model, factory parameters, installation information, and historical maintenance records, and to periodically calibrate the sensor data.
3. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The processing flow of the intelligent data preprocessing module is as follows: The first step is to adopt a hierarchical cleaning strategy based on the 3σ criterion and the isolated forest algorithm to remove abnormal data, and to fill missing data by interpolation or migration based on similar working conditions according to the data type. The second step is to use an adaptive wavelet threshold denoising algorithm to dynamically adjust the wavelet basis function and the number of decomposition layers according to the data noise intensity to achieve accurate denoising. The third step is to use the Z-score standardization method to transform data of different dimensions to the standard normal distribution interval; The fourth step involves using a multi-source data fusion algorithm based on an attention mechanism to assign dynamic weights to operating parameters, environmental parameters, and basic information, and then fusing them to output a standardized dataset.
4. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The improved deep learning fusion model in the accurate fault warning module is a CNN-BiLSTM-Attention fusion model, which includes a CNN feature extraction layer, a BiLSTM temporal analysis layer, an attention enhancement layer, and a risk decision layer connected in sequence. The CNN feature extraction layer is used to extract spatial features and local correlation features from the data; The BiLSTM time series analysis layer is used to capture the long-term and short-term time dependencies of the data. The attention enhancement layer is used to strengthen the weights of key fault features; The risk decision-making layer outputs the fault risk level (no risk, low risk, medium risk, high risk) and the corresponding fault type.
5. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 4, characterized in that: The training process of the improved deep learning fusion model includes: S1. Construct a labeled dataset containing normal operation data, potential fault data, and typical fault data, and divide it into training set and test set in an 8:2 ratio. S2, initialize model parameters, set the learning rate to 0.001~0.003, the number of iterations to 120~150, and the batch size to 32~64; S3 uses the AdamW optimizer and Focal Loss loss function for model training and prevents overfitting through an early stopping mechanism. S4. Use the test set to verify the model performance. Training is completed when the fault identification accuracy is ≥96% and the recall is ≥95%. Otherwise, adjust the model structure and parameters and retrain.
6. An air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The process by which the dynamic operation and maintenance decision module generates an operation and maintenance plan is as follows: S1 receives warning information from the accurate fault warning module and retrieves matching fault characteristics and cases from the fault diagnosis knowledge base module. S2 combines the device's historical operating data, maintenance records, and current operating status stored in the cloud-based collaborative management module to analyze the root cause of the fault; S3 generates a targeted solution based on fault type, equipment service life, operating environment and user maintenance cost budget, including fault handling steps, required spare parts, tool list and operating procedures. S4 receives feedback from operations and maintenance personnel on the effectiveness of the solution implementation, iterates and optimizes the solution, and updates it to the fault diagnosis knowledge base module.
7. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The cloud-based collaborative management module includes: a distributed data storage unit, a real-time data interaction unit, a data encryption and access control unit, and a data backup unit; The distributed data storage unit adopts a hybrid storage architecture of MySQL and MongoDB to store structured data and unstructured data respectively. The real-time data interaction unit uses the MQTT communication protocol to achieve low-latency data transmission. The data encryption and access control unit uses the AES-256 encryption algorithm to encrypt data during transmission and storage, and assigns user permissions based on role-based access control policies. The data backup unit employs a multi-replica backup mechanism in different locations to ensure data security and integrity.
8. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The user interaction terminal module includes a web management terminal for maintenance personnel, a mobile APP for ordinary users, and a device management backend. The web management terminal for maintenance personnel supports real-time monitoring of equipment operation data, viewing of fault warning details, editing and distribution of maintenance plans, entry of maintenance records, and updating of the knowledge base. The mobile app for regular users supports fault warning alerts, viewing simple maintenance suggestions, submitting repair requests, and checking maintenance progress. The device management backend supports system parameter configuration, sensor calibration management, and module status monitoring.
9. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The fault diagnosis knowledge base module adopts an incremental update mechanism, including a fault feature library, a case library, and a solution template library; The fault feature database stores typical fault feature parameter thresholds for different models of air conditioning equipment. The case library stores historical fault handling cases and effectiveness evaluation data; The solution template library stores standardized operation and maintenance solution templates for different fault types and equipment models, which can be dynamically adjusted according to actual working conditions.
10. The air conditioning equipment fault early warning and intelligent operation and maintenance service system according to claim 1, characterized in that: The information input and calibration unit has an automatic calibration reminder function. Based on the sensor's usage time, ambient humidity, and the stability of the collected data, it generates periodic calibration reminders and records the calibration results. When the deviation of the sensor data exceeds a preset threshold, a sensor fault warning is issued, prompting replacement or repair.