Building operation and maintenance management system based on Internet of Things

Abstract

The invention discloses a building operation and maintenance management system based on the Internet of Things, and relates to the technical field of the Internet of Things, and the system comprises the following components: a fault prediction module which selects a feature most related to an equipment fault from extracted features, selects a proper machine learning algorithm according to the features of data to construct a fault prediction model, and transmits the fault prediction model to a server; according to the invention, model construction is carried out on data of the data acquisition module through a logistic regression model formula and a loss function of the fault prediction module, and real-time data of the equipment is compared with the logistic regression model formula, so that a mapping relation between normal operation and a fault state of the equipment is accurately captured; in combination with the optimization process of the loss function, the model can continuously optimize own parameters and improve the prediction accuracy, so that the system can obtain equipment state data in real time and immediately perform fault prediction, thereby taking corresponding maintenance measures in time.

Description

Technical Field

[0001] The present invention relates to the field of building operation and maintenance management, and in particular to a building operation and maintenance management system based on the Internet of Things. Background Art

[0002] With the rapid development of information technology, Internet of Things technology has been widely used in various fields, especially in building operation and maintenance management, playing an increasingly important role. The building operation and maintenance management system is a comprehensive management tool that aims to use and manage buildings normally from the completion of the building acceptance to the demolition process through the use of operating technology, facilities and equipment and building users, so as to increase investment returns, reduce operating costs, and extend the use time of the building.

[0003] However, traditional building operation and maintenance management methods often rely on manual inspections and post-maintenance, and lack the ability to predict equipment failure trends and perform preventive maintenance. This method is not only inefficient, but also often makes it difficult to detect equipment failures in a timely manner, resulting in long equipment downtime and high maintenance costs. At the same time, traditional operation and maintenance management methods cannot fully utilize the data advantages brought by IoT technology, and cannot realize the intelligent and automated building operation and maintenance management.

[0004] Therefore, how to combine IoT technology and machine learning algorithms to achieve accurate prediction of equipment failure trends and automation of preventive maintenance is still an urgent problem to be solved in the field of building operation and maintenance management. However, the building operation and maintenance management systems on the market cannot yet achieve the function of fault prediction. Therefore, building an operation and maintenance management system that can perform preventive maintenance and fault prediction is of great significance to the construction industry in reducing maintenance costs, extending equipment service life, and improving the efficiency and intelligence level of building operation and maintenance management. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a building operation and maintenance management system based on the Internet of Things technology. By real-time collection, processing and analysis of equipment operation data, and using machine learning algorithms to predict equipment failure trends, preventive maintenance can be achieved, downtime caused by equipment failure can be reduced, equipment service life can be extended, and the efficiency and intelligence level of building operation and maintenance management can be improved.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a preventive maintenance management system, the system comprising the following components:

[0007] Data acquisition module: Link the data collector with the IoT sensor node, obtain data from the sensor node according to the set sampling frequency, so as to reflect the real-time operation status of the equipment. The collected data will be temporarily stored in the internal storage of the data collector. To ensure the integrity and accuracy of the data, this module usually performs preliminary verification and filtering on the data;

[0008] Data analysis module: remove duplicate data from the collected raw data, pre-process missing values, extract meaningful features from the pre-processed data to calculate statistics and indicators, and conduct more in-depth data analysis;

[0009] Fault prediction module: Select the features most relevant to equipment failure from the extracted features, select the appropriate machine learning algorithm based on the data features to build a fault prediction model so that it can learn the relationship between equipment failure and features, evaluate the performance of the model using cross-validation, and then optimize the model based on the evaluation results;

[0010] Preventive maintenance module: Specify the preventive maintenance plan for the equipment based on the prediction results. When it is predicted that the equipment is about to fail, the system will automatically trigger the maintenance plan. At the same time, the system can also optimize the maintenance plan and improve maintenance efficiency based on the actual operation status and maintenance history of the equipment.

[0011] Furthermore, the specific steps of the data acquisition module are:

[0012] (1) Data collection: connecting the data collector to the IoT sensor to detect the operating status and parameters of the equipment in real time;

[0013] (2) Data storage: The collected data is initially verified and filtered, and then temporarily stored in the internal memory of the data collector.

[0014] Furthermore, the data analysis module specifically comprises the following steps:

[0015] (1) Data processing: The collected raw data is preprocessed to remove duplicate data, handle missing values, and extract meaningful features from the preprocessed data for calculating statistics and indicators;

[0016] (2) Data analysis: Conduct more in-depth data analysis based on the extracted features.

[0017] Furthermore, the fault prediction module specifically comprises the following steps:

[0018] (1) Model selection and training: Select the logistic regression model formula and loss function based on data features to build a fault prediction model. Use historical data to train the model so that it can learn the relationship between equipment failures and features.

[0019] (2) Model optimization and prediction: Evaluate the performance of the model through cross-validation, optimize the model based on the evaluation results, compare real-time data with the trained model, and identify the trend of possible equipment failures.

[0020] Furthermore, the preventive maintenance module includes the following specific steps:

[0021] (1) Receiving prediction results: receiving the output from the fault prediction module, i.e., the probability and expected occurrence time of equipment failure;

[0022] (2) Formulate maintenance plan: Based on the fault prediction results, determine the interval between failures according to the failure rate analysis algorithm, and specify a preventive maintenance plan in combination with the equipment's maintenance history and operating status;

[0023] (3) Plan optimization and execution: Considering the actual operation status and maintenance cost of the equipment, the maintenance plan is optimized to ensure the effectiveness and economy of the maintenance plan, and the corresponding maintenance tasks are performed according to the optimized maintenance plan;

[0024] (4) Recording and feedback: Record the execution status and results of maintenance tasks, and feed this information back to the fault prediction module, and continuously improve the prediction module through linear regression algorithm.

[0025] Furthermore, the fault prediction module constructs a fault prediction model by adopting a logistic regression model formula and a loss function, and trains the model to learn the mapping relationship between the normal operation of the equipment and the fault state. The logistic regression model formula is: Load the training data set and divide the data set into a feature matrix X∈R m×n Where m is the number of samples, n is the number of features and the target vector y∈R m , initialize the parameter vector θ∈R n , usually set to a zero vector or a random decimal vector, set the learning rate α and the number of iterations num_iterations, build the initialization model, and calculate the prediction logic h for each sample θ (x (i) ), the logistic regression model formula is According to the target vector y and the predicted probability h θ (x (i) ), for calculating the loss function J(θ), the loss function calculation formula is , where m is the number of samples, for each iteration (from 1 to num_iterations): calculate the loss function J(θ) with respect to each parameter θ j The partial derivative (gradient) of is: Repeat the above iterative process until the preset number of iterations num_iterations is reached, or the loss function J(θ) converges to a sufficiently small value, use the validation set (if any) to evaluate the performance of the trained logistic regression model, calculate the accuracy, recall, and F1 score indicators, and for new data points, input their features into the trained logistic regression model and use the model formula to calculate the predicted probability h θ (x (i) ), according to the preset probability threshold (usually 0.5), the predicted probability is converted into a binary classification label, that is, if h θ (x (i) ) is greater than or equal to the threshold, it is predicted as a positive class (labeled as 1), otherwise it is predicted as a negative class (labeled as 0), and the prediction result is applied to the actual scene.

[0026] Furthermore, according to the Internet of Things-based building operation and maintenance management system of claim 5, it is characterized in that the preventive maintenance module adopts a failure rate analysis algorithm to predict the possible future failure trend of the equipment, and the failure path analysis algorithm adopts an exponential failure rate model: When the shape parameter m = 1, the failure rate model is simplified to: Among them, t0 is the characteristic life, that is, the mean time between failures (MTBF) of the product. To use this model, you can calculate the mean time between failures (MTBF) of the equipment and then approximate it with its reciprocal. Once you have The model can be used to predict the failure rate of the equipment at any time in the future.

[0027] Furthermore, the preventive maintenance module uses machine learning technology to continuously record the historical maintenance data of the equipment and the current operating status of the equipment, conduct model evaluation, and optimize the model structure according to the evaluation results. The machine learning technology needs to extract meaningful features from the collected data. These features may include operating temperature, temperature and pressure. Assuming that we extract a feature, then for each operation sample, we can get a feature vector. Use these feature vectors and corresponding labels to train a machine learning model. A linear regression model can be used, and its mathematical form is: y=w1x1+w2x2+…+w n x n +b, where y is the predicted failure outcome, w1,w2,…,w n is the weight coefficient, b is the bias term, these parameters need to be learned through training data. During the training process, a loss function is usually used to measure the prediction error of the model, and the loss function is minimized through an optimization algorithm. For linear regression problems, the commonly used loss function is the mean square error (MSE): Where m is the number of samples, y i is the true skill level score of the i-th sample, is the model's prediction score for the i-th sample.

[0028] Furthermore, the preventive maintenance module has intelligent and adaptive capabilities, and can continuously optimize the fault prediction model and maintenance plan according to the operating conditions and historical data of the equipment, thereby improving the intelligent level of operation and maintenance management.

[0029] Furthermore, the preventive maintenance module supports remote monitoring and management. Users can access the system interface through the network, view the equipment status in real time, receive fault warning information, and perform remote operation and maintenance.

[0030] Compared with the existing technology, the experimental instrument operation behavior recording system has the following beneficial effects:

[0031] 1. The present invention constructs a model for the data of the data acquisition module through the logistic regression model formula and loss function of the fault prediction module and compares the real-time data of the equipment with it, accurately capturing the mapping relationship between the normal operation of the equipment and the fault state. Combined with the optimization process of the loss function, the model can continuously optimize its own parameters and improve the accuracy of the prediction, which enables the system to obtain equipment status data in real time and immediately predict faults, so as to take corresponding maintenance measures in time.

[0032] 2. Through the failure rate analysis algorithm adopted by the preventive maintenance module, the system can obtain the operating status data of the equipment in real time, and perform dynamic prediction in combination with the failure rate analysis algorithm, so that the system can continuously update the prediction results as the equipment is used over time, ensuring that the prediction results are always consistent with the actual status of the equipment. At the same time, the system can also continuously optimize the maintenance plan according to the actual operation status and maintenance history of the equipment, further improving maintenance efficiency.

[0033] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 This is the process operation diagram of the building operation and maintenance management system based on the Internet of Things. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Example 1

[0038] Building operation and maintenance management system based on the Internet of Things, the system includes the following components:

[0039] Data acquisition module: Link the data collector with the IoT sensor node, obtain data from the sensor node according to the set sampling frequency, so as to reflect the real-time operation status of the equipment. The collected data will be temporarily stored in the internal storage of the data collector. To ensure the integrity and accuracy of the data, this module usually performs preliminary verification and filtering on the data;

[0040] Data analysis module: pre-processes the collected raw data to ensure data consistency and comparability, extracts meaningful features from the processed data and conducts a deeper analysis, and uses learning algorithms to build models to predict the operating status or failure trend of equipment;

[0041] Fault prediction module: Select the features most relevant to equipment failure from the extracted features, select the appropriate machine learning algorithm based on the data features to build a fault prediction model so that it can learn the relationship between equipment failure and features, evaluate the performance of the model using cross-validation, and then optimize the model based on the evaluation results;

[0042] Preventive maintenance module: Specify the preventive maintenance plan for the equipment based on the prediction results. When it is predicted that the equipment is about to fail, the system will automatically trigger the maintenance plan. At the same time, the system can also optimize the maintenance plan and improve maintenance efficiency based on the actual operation status and maintenance history of the equipment.

[0043] The fault prediction module constructs a fault prediction model by using the logistic regression model formula and loss function, and trains the model to learn the mapping relationship between the normal operation of the equipment and the fault state. The logistic regression model formula is: Load the training data set and divide the data set into the feature matrix X∈R m×n (where m is the number of samples and n is the number of features) and the target vector y∈R m , initialize the parameter vector θ∈R n, usually set to a zero vector or a random decimal vector, set the learning rate α and the number of iterations num_iterations, build the initialization model, and calculate the prediction logic h for each sample θ (x (i) ), the logistic regression model formula is According to the target vector y and the predicted probability h θ (x (i) ), calculate the loss function J(θ), the loss function calculation formula is J Where m is the number of samples, for each iteration (from 1 to num_iterations): calculate the loss function J(θ) with respect to each parameter θ j The partial derivative (gradient) of is: Repeat the above iterative process until the preset number of iterations num_iterations is reached, or the loss function J(θ) converges to a sufficiently small value, use the validation set (if any) to evaluate the performance of the trained logistic regression model, calculate the accuracy, recall, and F1 score indicators, and for new data points, input their features into the trained logistic regression model and use the model formula to calculate the predicted probability h θ (x (i) ), according to the preset probability threshold (usually 0.5), the predicted probability is converted into a binary classification label, that is, if h θ (x (i) ) is greater than or equal to the threshold, it is predicted as a positive class (labeled as 1), otherwise it is predicted as a negative class (labeled as 0), and the prediction result is applied to the actual scene.

[0044] The data from the data acquisition module is modeled through the logistic regression model formula and loss function of the fault prediction module and the real-time data of the equipment is compared with it to accurately capture the mapping relationship between the normal operation of the equipment and the fault status. Combined with the optimization process of the loss function, the model can continuously optimize its own parameters and improve the accuracy of the prediction. This enables the system to obtain equipment status data in real time and make fault predictions immediately, so that corresponding maintenance measures can be taken in time.

[0045] Example 2

[0046] Building operation and maintenance management system based on the Internet of Things, the system includes the following components:

[0047] Data acquisition module: Link the data collector with the IoT sensor node, obtain data from the sensor node according to the set sampling frequency, so as to reflect the real-time operation status of the equipment. The collected data will be temporarily stored in the internal storage of the data collector. To ensure the integrity and accuracy of the data, this module usually performs preliminary verification and filtering on the data;

[0048] Data analysis module: pre-processes the collected raw data to ensure data consistency and comparability, extracts meaningful features from the processed data and conducts a deeper analysis, and uses learning algorithms to build models to predict the operating status or failure trend of equipment;

[0049] Fault prediction module: Select the features most relevant to equipment failure from the extracted features, select a suitable machine learning algorithm based on the characteristics of the data to build a fault prediction model so that it can learn the relationship between equipment failure and features, evaluate the performance of the model using cross-validation, and then optimize the model based on the evaluation results.

[0050] Preventive maintenance module: Specify the preventive maintenance plan for the equipment based on the prediction results. When it is predicted that the equipment is about to fail, the system will automatically trigger the maintenance plan. At the same time, the system can also optimize the maintenance plan and improve maintenance efficiency based on the actual operation status and maintenance history of the equipment.

[0051] The preventive maintenance module uses a failure rate analysis algorithm to predict the possible future failure trend of the equipment, and the failure path analysis algorithm uses an exponential failure rate model: When the shape parameter m = 1, the failure rate model is simplified to: Among them, t0 is the characteristic life, that is, the mean time between failures (MTBF) of the product, which means that under normal conditions, the probability of a component failure at the next moment is independent of the component life. To use this model, it is necessary to determine the value of λ, which is usually done by analyzing the historical failure data of the equipment. The mean time between failures (MTBF) of the equipment can be calculated, and then its inverse can be used to approximate λ. Once the value of λ is obtained, this model can be used to predict the failure rate of the equipment at any time in the future.

[0052] Through the failure rate analysis algorithm adopted by the preventive maintenance module, the system can obtain the operating status data of the equipment in real time, and perform dynamic prediction in combination with the failure rate analysis algorithm, so that the system can continuously update the prediction results as the equipment is used over time, ensuring that the prediction results are always consistent with the actual status of the equipment. This real-time and dynamic nature enables the system to detect potential faults in a timely manner and take corresponding preventive maintenance measures, which can not only avoid downtime losses caused by equipment failures, but also reduce unnecessary maintenance operations, thereby optimizing resource utilization. At the same time, the system can also continuously optimize the maintenance plan based on the actual operation of the equipment and maintenance history, further improving maintenance efficiency.

[0053] Example 3

[0054] The steps of the system based on Example 1 and Example 2 are as follows:

[0055] (1) System deployment and configuration

[0056] Select appropriate IoT sensors and device interfaces based on the equipment layout and operation and maintenance requirements within the building, and install them on the corresponding equipment;

[0057] Configure the system's network connection to ensure that the data acquisition module can transmit data to the data processing center in real time;

[0058] Customize the parameters and algorithms of the data analysis module, fault prediction module and preventive maintenance module according to actual needs.

[0059] (2) Data collection and processing

[0060] The acquisition module collects the operation data of the equipment in real time through the IoT sensors and device interfaces. The collected data is transmitted to the data processing center through the wireless network to ensure the real-time and accuracy of the data.

[0061] The analysis module receives data from the data acquisition module and performs data cleaning, denoising and feature extraction to improve the quality and availability of the data. By comparing historical data with real-time data, it analyzes the operating status of the equipment, performance change trends and possible fault hazards.

[0062] (3) Failure prediction and maintenance plan specification

[0063] The fault prediction module uses machine learning algorithms to predict the equipment's fault probability, fault type, and fault occurrence time based on the equipment's operating data and historical fault records;

[0064] Based on the prediction results, the preventive maintenance module formulates a preventive maintenance plan for the equipment, including maintenance time, maintenance content, and required maintenance resources.

[0065] (4) Maintenance execution and monitoring

[0066] When the system triggers a maintenance plan, the operation and maintenance manager arranges maintenance personnel, prepares maintenance tools, and performs corresponding maintenance work according to the system prompts;

[0067] During the maintenance process, the system monitors the operating status and maintenance progress of the equipment in real time to ensure the smooth progress of the maintenance work.

[0068] (5) System optimization and upgrade

[0069] When the system triggers a maintenance plan, the operation and maintenance manager arranges maintenance personnel, prepares maintenance tools, and performs corresponding maintenance work according to the system prompts;

[0070] During the maintenance process, the system monitors the operating status and maintenance progress of the equipment in real time to ensure the smooth progress of the maintenance work.

[0071] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A building operation and maintenance management system based on Internet of Things technology, characterized in that: The system consists of the following components: Data acquisition module: Link the data collector with the IoT sensor node, obtain data from the sensor node according to the set sampling frequency, so as to reflect the real-time operation status of the equipment. The collected data will be temporarily stored in the internal storage of the data collector. To ensure the integrity and accuracy of the data, this module usually performs preliminary verification and filtering on the data; Data analysis module: remove duplicate data from the collected raw data, pre-process missing values, extract meaningful features from the pre-processed data to calculate statistics and indicators, and conduct more in-depth data analysis; Fault prediction module: Select the features most relevant to equipment failure from the extracted features, select the appropriate machine learning algorithm based on the data features to build a fault prediction model so that it can learn the relationship between equipment failure and features, evaluate the performance of the model using cross-validation, and then optimize the model based on the evaluation results; Preventive maintenance module: Specify the preventive maintenance plan for the equipment based on the prediction results. When it is predicted that the equipment is about to fail, the system will automatically trigger the maintenance plan. At the same time, the system can also optimize the maintenance plan and improve maintenance efficiency based on the actual operation status and maintenance history of the equipment.

2. The building operation and maintenance management system based on the Internet of Things according to claim 1 is characterized in that: The specific steps of the data acquisition module are as follows: (1) Data collection: connecting the data collector to the IoT sensor to detect the operating status and parameters of the equipment in real time; (2) Data storage: The collected data is initially verified and filtered, and then temporarily stored in the internal memory of the data collector.

3. The building operation and maintenance management system based on the Internet of Things according to claim 1 is characterized in that: The data analysis module has the following specific steps: (1) Data processing: The collected raw data is preprocessed to remove duplicate data, handle missing values, and extract meaningful features from the preprocessed data for calculating statistics and indicators; (2) Data analysis: Conduct more in-depth data analysis based on the extracted features.

4. The building operation and maintenance management system based on the Internet of Things according to claim 1 is characterized in that: The specific steps of the fault prediction module are as follows: (1) Model selection and training: Select the logistic regression model formula and loss function based on data features to build a fault prediction model. Use historical data to train the model so that it can learn the relationship between equipment failures and features. (2) Model optimization and prediction: Evaluate the performance of the model through cross-validation, optimize the model based on the evaluation results, compare real-time data with the trained model, and identify the trend of possible equipment failures.

5. The building operation and maintenance management system based on the Internet of Things according to claim 1 is characterized in that: The specific steps of the preventive maintenance module are: (1) Receiving prediction results: receiving the output from the fault prediction module, i.e., the probability and expected occurrence time of equipment failure; (2) Formulate maintenance plan: Based on the fault prediction results, determine the interval between failures according to the failure rate analysis algorithm, and specify a preventive maintenance plan in combination with the equipment's maintenance history and operating status; (3) Plan optimization and execution: Considering the actual operation status and maintenance cost of the equipment, the maintenance plan is optimized to ensure the effectiveness and economy of the maintenance plan, and the corresponding maintenance tasks are performed according to the optimized maintenance plan; (4) Recording and feedback: Record the execution status and results of maintenance tasks, and feed this information back to the fault prediction module, and continuously improve the prediction module through linear regression algorithm.

6. The building operation and maintenance management system based on the Internet of Things according to claim 4 is characterized in that The fault prediction module constructs a fault prediction model through a machine learning algorithm, trains the model, and learns the mapping relationship between the normal operation of the equipment and the fault state. The machine learning algorithm includes a logistic regression model formula and a loss function. The logistic regression model formula is: Load the training data set and divide the data set into a feature matrix X∈R m×n (where m is the number of samples and n is the number of features) and the target vector y∈R m , initialize the parameter vector θ∈R n When , it is usually set to a zero vector or a random decimal vector, set the learning rate α and the number of iterations num_iterations, then build the initialization model and calculate the prediction logic h for each sample θ (x (i) ), the logistic regression model formula is According to the target vector y and the predicted probability h θ (x (i) ), for calculating the loss function J(θ), the loss function calculation formula is Where m is the number of samples, for each iteration (from 1 to num_iterations): calculate the loss function J(θ) with respect to each parameter θ j The partial derivative (gradient) of is: Repeat the above iterative process until the preset number of iterations num_iterations is reached, or the loss function J(θ) converges to a sufficiently small value. Evaluate the performance of the trained logistic regression model by using the validation set (if any), calculate the accuracy, recall, and F1 score indicators, and for new data points, input their features into the trained logistic regression model and use the model formula to calculate the predicted probability h θ (x (i) ), according to the preset probability threshold (usually 0.5), the predicted probability is converted into a binary classification label, that is, if h θ (x (i) ) is greater than or equal to the threshold, it is predicted as a positive class (labeled as 1), otherwise it is predicted as a negative class (labeled as 0), and the prediction result is applied to the actual scene.

7. The building operation and maintenance management system based on the Internet of Things according to claim 5 is characterized in that ,The preventive maintenance module adopts a failure rate analysis algorithm to predict the future ,failure rate analysis algorithm adopts an exponential failure rate model: When the shape parameter m = 1, the failure rate model is simplified to: Among them, t0 is the characteristic life, that is, the mean time between failures (MTBF) of the product. To use this model, you can calculate the mean time between failures (MTBF) of the equipment and then approximate it with its reciprocal. Once you have The model can be used to predict the failure rate of the equipment at any time in the future.

8. The building operation and maintenance management system based on the Internet of Things according to claim 5 is characterized in that ,The preventive maintenance module uses machine learning technology to continuously ,record the historical maintenance data and the current operation status ,of the equipment, and the equipment, perform model evaluation, ,and optimize the model structure according to the evaluation results.,The machine learning technology needs to extract meaningful features from the collected data. ,These features may include the temperature, the temperature and the pressure of the operation. ,Assuming that we extract features, then for each operation sample, we can get a feature vector. ,Use these feature vectors and the corresponding labels to train a machine learning ,model, a linear regression model can be used, and its mathematical form is : y=w1x1+w2x2+…+w n x n +b, where y is the predicted failure outcome, w1,w2,…,w n is the weight coefficient, b is the bias term, these parameters need to be learned through training data. During the training process, a loss function is usually used to measure the prediction error of the model, and the loss function is minimized through an optimization algorithm. For linear regression problems, the commonly used loss function is the mean square error (MSE): Where m is the number of samples, y i is the true skill level score of the i-th sample, is the model's prediction score for the i-th sample.

9. According to the IoT-based building operation and maintenance management system of claim 1, the preventive maintenance module has intelligent and adaptive capabilities, and can continuously optimize the fault prediction model and maintenance plan according to the operating conditions and historical data of the equipment, thereby improving the intelligence level of operation and maintenance management.

10. The experimental instrument operation behavior recording system according to claim 1, characterized in that ,The preventive maintenance module supports remote monitoring and management. Users can access the system interface through the ,network, to view the equipment status in real time, receive ,fault warning information, and perform remote operation

Images