Intelligent temperature monitoring system for engine

Abstract

The invention discloses an intelligent temperature monitoring system for an engine, and relates to the field of temperature monitoring. The system comprises a temperature measurement instruction receiving module used for receiving a temperature measurement instruction of an engine; the temperature feature prediction module performs temperature feature fitting to obtain a temperature prediction distribution feature point cloud; the temperature measurement decision module is used for obtaining a temperature measurement feature decision; the temperature measurement adjustment factor analysis module is used for generating a temperature measurement adjustment factor; the temperature measurement decision compensation module is used for generating an optimized temperature measurement characteristic decision; and the temperature monitoring module is used for monitoring temperature. The technical problems that a traditional engine temperature monitoring system usually depends on a temperature sensor at a fixed position, so that engine temperature monitoring is low in flexibility, poor in adaptability and low in intelligence are solved, the flexibility and expandability of temperature monitoring are improved through a multi-point distributed sensing network and modular design, and the system is suitable for popularization and application. The system can automatically adjust measurement parameters, and the adaptability and stability of the system are improved.

Description

Technical Field

[0001] This application relates to the technical field of temperature monitoring, and specifically to an intelligent temperature monitoring system for engines. Background Art

[0002] With the rapid development of the automotive industry, the engine, as the core component of a vehicle, the stability and reliability of its performance are crucial for the operation of the entire vehicle. Temperature is one of the key factors affecting engine performance. Excessive or too low temperature may damage the engine and even cause failures. Therefore, it is particularly important to monitor and intelligently control the engine temperature in real time.

[0003] Currently, there are already various systems for monitoring engine temperature on the market. These systems generally include components such as temperature sensors, data acquisition modules, processing units, and display interfaces. These systems can collect temperature data of various parts of the engine in real time, analyze and judge through the processing unit, and finally display the temperature information on the interface for operators to refer to. However, these traditional systems still have deficiencies in terms of intelligence, accuracy, and real-time performance, and are difficult to meet the higher requirements for engine temperature monitoring in modern engines.

[0004] In summary, the traditional engine temperature monitoring system usually relies on temperature sensors at fixed positions, resulting in technical problems such as low flexibility, poor adaptability, and low intelligence in engine temperature monitoring. Summary of the Invention

[0005] Based on this, it is necessary to provide an intelligent temperature monitoring system for engines to solve the technical problems that the traditional engine temperature monitoring system usually relies on temperature sensors at fixed positions, resulting in low flexibility, poor adaptability, and low intelligence in engine temperature monitoring. Through a multi-point distributed sensing network and modular design, the flexibility and scalability of temperature monitoring are improved, enabling the system to automatically adjust measurement parameters, and improving the adaptability and stability of the system.

[0006] Based on this, a temperature intelligent monitoring system for an engine is provided. The system includes: a temperature measurement instruction receiving module for receiving a temperature measurement instruction of the engine, where the temperature measurement instruction includes a future temperature measurement time domain corresponding to the engine and a future control feature decision; a temperature feature prediction module for fitting the temperature features of the engine based on the future temperature measurement time domain and the future control feature decision to obtain temperature prediction distribution feature points; a temperature measurement decision module for making a temperature monitoring feature decision on the engine based on the temperature prediction distribution feature points to obtain a temperature measurement feature decision; a temperature measurement adjustment factor analysis module for predicting risks for the engine based on the temperature prediction distribution feature points and generating a temperature measurement adjustment factor in combination with a predetermined risk trigger probability; a temperature measurement decision compensation module for optimizing and compensating the temperature measurement feature decision based on the temperature measurement adjustment factor to generate an optimized temperature measurement feature decision; and a temperature monitoring module for monitoring the temperature of the engine based on the optimized temperature measurement feature decision.

[0007] The above temperature intelligent monitoring system for an engine solves the technical problems that the traditional engine temperature monitoring system usually relies on temperature sensors at fixed positions, resulting in low flexibility, poor adaptability, and low intelligence in engine temperature monitoring. Through a multi-point distributed sensing network and modular design, the flexibility and scalability of temperature monitoring are improved, enabling the system to automatically adjust measurement parameters and enhancing the adaptability and stability of the system.

[0008] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically enumerates the specific embodiments of this application. Brief Description of the Drawings

[0009] Figure 1 It is a structural block diagram of a temperature intelligent monitoring system for an engine in an embodiment; Figure 2 It is a schematic flow diagram of generating optimized temperature measurement data of a temperature intelligent monitoring system for an engine in an embodiment.

[0010] Description of the reference numerals: temperature measurement instruction receiving module 11, temperature feature prediction module 12, temperature measurement decision module 13, temperature measurement adjustment factor analysis module 14, temperature measurement decision compensation module 15, temperature monitoring module 16. Detailed Description of the Embodiments

[0011] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0012] As Figure 1 shown, the present application provides a temperature intelligent monitoring system for an engine, and the system includes: a temperature measurement instruction receiving module 11, which is configured to receive a temperature measurement instruction of the engine, wherein the temperature measurement instruction includes a future temperature measurement time domain and a future control feature decision corresponding to the engine.

[0013] An engine is a machine that can convert other forms of energy into mechanical energy and is a power mechanical device. Its main function is to convert other forms of energy (such as chemical energy, thermal energy, etc.) into mechanical energy, so as to drive various equipment or vehicles to operate. A temperature intelligent monitoring system is a comprehensive system integrating high-precision temperature sensing, data processing, intelligent control and remote monitoring technologies. The system can realize real-time monitoring, analysis, early warning and control of the engine temperature. The present application provides a temperature intelligent monitoring system for an engine, a comprehensive system integrating modern sensing technology, data processing technology and communication technology, aiming to monitor and analyze the temperature data of each key part of the engine in real time to ensure the safe and efficient operation of the engine and prevent failures caused by overheating or overcooling.

[0014] The temperature measurement instruction receiving module 11 in the engine temperature intelligent monitoring system is responsible for receiving and parsing the temperature measurement instructions for the engine. The temperature measurement instruction receiving module 11 is the input interface of the system and is responsible for receiving the temperature measurement instructions sent from the upper computer, the remote monitoring center or the driver interface. Parse the received temperature measurement instructions and extract the key information therein, including the future temperature measurement time domain corresponding to the engine and the future control feature decision. It indicates the specific time period or cycle for temperature monitoring of the engine. This helps the system arrange monitoring tasks according to actual needs and avoid unnecessary resource waste. For example, the system may need to perform high-frequency monitoring in the first few minutes after the engine starts to quickly respond to temperature changes; while the monitoring frequency can be appropriately reduced during the stable operation stage. Based on the analysis of the engine's historical operation data and the evaluation of the current operation state, a temperature control strategy or decision formulated in advance. It includes starting or adjusting the cooling system at a specific temperature threshold, adjusting the engine operation parameters to optimize the temperature distribution, etc. The purpose of the future control feature decision is to ensure that the engine operates within the optimal working temperature range through anticipatory control means and prevent failures caused by overheating or overcooling. The temperature measurement instruction receiving module 11 receives the temperature measurement instructions through a preset communication interface. Parse the received instructions and extract key information such as the future temperature measurement time domain and the future control feature decision. Transmit the parsed information to other modules of the system (such as the data acquisition and processing unit, the intelligent control unit, etc.) for corresponding monitoring and control tasks. When necessary, feedback the instruction reception and execution status to the instruction sender to ensure the reliability of communication and the stability of the system. The temperature measurement instruction receiving module 11 plays a key role in information transfer and decision execution in the engine temperature intelligent monitoring system. Its efficient and accurate operation is of great significance for ensuring the safe and efficient operation of the engine.

[0015] A temperature feature prediction module 12, wherein the temperature feature prediction module 12 performs temperature feature fitting on the engine based on the future temperature measurement time domain and the future control feature decision to obtain a temperature prediction distribution feature point cloud.

[0016] The main function of the temperature feature prediction module 12 is to predict the temperature features of the engine by using the known or predicted future temperature measurement time domain (i.e., the time frame of temperature change within the predicted time period) and future control feature decisions (such as the expected settings of control parameters like engine speed, load, cooling system status, etc.). Collect current and historical engine operation data, including temperature readings, control parameters, etc. These data will be used to establish the basis of the prediction model. Based on the collected data, statistical methods, machine learning algorithms (such as neural networks, support vector machines, random forests, etc.) or deep learning techniques are used to train one or more prediction models. These models will be able to learn the relationship between the engine temperature and control parameters and predict future temperature features. Determine the time range of the temperature change to be predicted, i.e., the future temperature measurement time domain. This time range can be set according to actual needs, such as a few minutes, several hours or longer. Input or predict the control feature decisions of the engine within a future period of time, and these decisions will affect the temperature change of the engine. Use the established prediction model and the input future control feature decisions to fit the temperature features of the engine. This process will generate one or more predicted temperature change curves or point clouds, which describe the possible distribution of the engine temperature within the future temperature measurement time domain. Output a temperature prediction distribution feature point cloud, which contains all the values that the engine temperature may reach within the future temperature measurement time domain and their probability distributions. This is of great significance for formulating effective cooling strategies, optimizing engine performance, preventing overheating, etc. In summary, the temperature feature prediction module 12 plays a crucial role in the engine management system, and it provides strong support for the optimized operation of the engine by using advanced prediction techniques.

[0017] A temperature measurement decision module 13, and the temperature measurement decision module 13 is used to make temperature monitoring feature decisions on the engine based on the temperature prediction distribution feature point cloud to obtain temperature measurement feature decisions.

[0018] The temperature measurement decision-making module 13 is responsible for formulating and executing temperature monitoring feature decisions based on the temperature prediction distribution feature point cloud. The function of this module is to convert the results of temperature prediction into actual control instructions to ensure that the engine operates within a predetermined temperature range, thereby optimizing performance and extending lifespan. The temperature measurement decision-making module 13 first receives the temperature prediction distribution feature point cloud from the temperature feature prediction module 12. This point cloud contains all the values that the engine temperature may reach within the future temperature measurement time domain and their probability distributions. It deeply analyzes the received temperature prediction results to identify potential temperature risk areas, such as high-temperature areas that may exceed the safety range or low-temperature areas that may affect performance. Based on the results of the temperature prediction analysis, the module formulates corresponding temperature monitoring feature decisions. These decisions may include adjusting the working state of the cooling system (such as increasing or decreasing the coolant flow rate, adjusting the fan speed), adjusting the engine load, changing the engine speed, etc., to maintain the engine operating within the ideal temperature range. The decisions are optimized according to real-time data and historical experience. For example, if it is predicted that the engine temperature may rise sharply during a certain period, more proactive cooling measures will be taken in advance. Once the temperature measurement feature decisions are formulated, the module will convert these decisions into specific control instructions and send them to the corresponding actuators (such as the cooling system, fuel injection system, etc.) through the engine management system. During the process of executing the decisions, the module continuously monitors the temperature change of the engine and fine-tunes the decisions according to the actual situation. At the same time, the module also feeds back the monitoring results to the temperature feature prediction module 12 to consider more actual factors in future predictions. In summary, the temperature measurement decision-making module 13 plays a crucial role in the engine management system, ensuring that the engine can operate stably within a predetermined temperature range, thereby improving the overall performance and reliability of the system.

[0019] The temperature measurement adjustment factor analysis module 14, and the temperature measurement adjustment factor analysis module 14 is used to perform risk prediction on the engine based on the temperature prediction distribution feature point cloud and generate a temperature measurement adjustment factor in combination with a predetermined risk trigger probability.

[0020] The temperature measurement adjustment factor analysis module 14 is a key analysis and decision support module in the engine management system. It predicts the risks of the engine based on the temperature prediction distribution characteristic point cloud and generates temperature measurement adjustment factors in combination with a predetermined risk trigger probability. Using the temperature prediction distribution characteristic point cloud, it predicts the risks that the engine may face in the future temperature measurement time domain. These risks may include overheating, overcooling, or excessive temperature fluctuations, etc., which will have an adverse impact on the performance and lifespan of the engine. Combining the predetermined risk trigger probability (a threshold set according to factors such as engine type, operating conditions, historical data, etc.), it evaluates the predicted risks to determine whether the likelihood of their occurrence reaches the level that requires measures to be taken. According to the risk prediction results and the risk trigger probability assessment, corresponding temperature measurement adjustment factors are generated. These adjustment factors will be used as the basis for subsequent control decisions to guide how to adjust the operating state of the engine or the working parameters of the cooling system to reduce risks and optimize engine performance. It receives the temperature prediction distribution characteristic point cloud from the temperature characteristic prediction module 12. Parses the point cloud data to identify potential risk areas (such as high-temperature areas, low-temperature areas, or areas with large temperature fluctuations). Evaluates the potential impact of these risk areas on the performance and lifespan of the engine. Sets or calls the predetermined risk trigger probability threshold. Compares the predicted risks with the threshold to determine whether the likelihood of risk occurrence reaches the level that requires measures to be taken. According to the risk prediction results and the risk trigger probability assessment, formulates specific adjustment strategies. Generates temperature measurement adjustment factors, which may include cooling system adjustment parameters (such as coolant flow rate, fan speed, etc.), engine load adjustment suggestions, engine speed adjustment ranges, etc. Outputs the generated temperature measurement adjustment factors to the engine management system or the corresponding actuator. Monitors the operating state and temperature changes of the adjusted engine and collects feedback data for subsequent optimization and improvement. Through the above work process and key point summary, the temperature measurement adjustment factor analysis module 14 can predict the risks of the engine based on the temperature prediction distribution characteristic point cloud and generate effective temperature measurement adjustment factors, providing strong support for the safe and efficient operation of the engine.

[0021] The temperature measurement decision compensation module 15 is used to optimize and compensate the temperature measurement feature decision based on the temperature measurement adjustment factor to generate an optimized temperature measurement feature decision.

[0022] The temperature measurement decision compensation module 15 is a very important component in the temperature measurement system. It is mainly responsible for optimizing and adjusting the original temperature measurement feature decision according to the temperature measurement adjustment factor to generate a more accurate and reliable optimized temperature measurement feature decision. This mechanism is very crucial in various application scenarios, such as industrial automation, environmental monitoring and other fields, where the accuracy of temperature data directly affects the performance and stability of the system. It receives the original temperature data from the temperature sensor and possibly other relevant parameters (such as ambient humidity, pressure, etc.). These data constitute the preliminary temperature measurement features. Based on these original data, the system makes a preliminary feature decision, that is, to judge whether the current temperature condition meets the preset conditions or standards. This decision may be based on a simple threshold judgment or a more complex algorithm model. The temperature measurement adjustment factor is dynamically determined according to various factors such as the system operating state, external environment changes, and sensor aging. These factors are designed to correct the temperature measurement deviation caused by factors such as sensor errors and environmental interference. Use these adjustment factors to optimize and compensate the preliminary temperature measurement feature decision. This step involves various technical means such as data filtering, error correction, and trend prediction to ensure that the finally output temperature data is closer to the true value. The temperature data after optimization and compensation will be output as the optimized temperature measurement feature decision for subsequent system control, data analysis and other processes. The temperature measurement decision compensation module 15 plays a crucial role in the temperature measurement system. By means of the optimization and compensation mechanism, it ensures the accuracy and reliability of temperature data, providing a strong guarantee for the stable operation and efficient management of the system.

[0023] The temperature monitoring module 16 is configured to monitor the temperature of the engine based on the optimized temperature measurement feature decision.

[0024] The temperature monitoring module 16 is a key component specifically designed for temperature monitoring of the engine. Based on the optimized temperature measurement feature decision-making, this module can monitor the operating temperature of the engine in real-time and accurately, ensuring that the engine operates under the best working conditions and preventing failures caused by overheating or overcooling. The temperature monitoring module 16 is responsible for receiving data from the optimized temperature measurement feature decision-making system and precisely monitoring the engine temperature based on this data. It ensures that the engine temperature remains within the preset safe range, improving the operating efficiency and reliability of the engine. The temperature monitoring module 16 first receives the optimized temperature data from the optimized temperature measurement feature decision-making system. These data have filtered out noise and errors and have high accuracy and reliability. Using high-precision temperature sensors (such as thermocouples, thermistors, etc.), the module conducts real-time temperature monitoring on key parts of the engine (such as cylinders, cooling systems, lubricating oil circuits, etc.). The sensors convert the collected temperature data into electrical signals, which are processed through signal conditioning circuits such as amplification and filtering for subsequent processing and analysis. The received temperature data is sent to the microprocessor in the module. According to the preset temperature threshold and algorithm logic, it determines whether the current engine temperature is within the safe range and generates corresponding control instructions. In summary, the temperature monitoring module 16 is an essential and indispensable part of the engine management system. By monitoring the engine temperature in real-time and accurately and performing intelligent control based on the optimized temperature measurement feature decision-making, it ensures that the engine operates under the best working conditions and improves the performance and reliability of the engine.

[0025] Further, the system is used to perform the following steps: Model the engine according to the temperature feature prediction module to obtain an engine simulation model; based on the future temperature measurement time domain and the temperature feature prediction module, perform K times of simulated control on the engine simulation model according to the future control feature decision-making to generate K engine temperature simulation data sets, where K is a positive integer greater than 1; calculate the spatio-temporal feature median value according to the K engine temperature simulation data sets to generate an engine temperature prediction sequence set; perform feature clustering on the engine temperature prediction sequence set according to the component composition feature information of the engine to generate the temperature prediction distribution feature point cloud.

[0026] Collect the operating data of the engine under different working conditions, including but not limited to engine speed, load, cooling water flow rate, fuel consumption rate, etc. Based on the working principle and thermodynamic principle of the engine, use tools such as MATLAB / Simulink to establish a physical model of the engine. The model should be able to reflect the dynamic characteristics and heat exchange process of the engine. According to the actual parameters of the engine (such as the number of cylinders, displacement, compression ratio, etc.) and experimental data, set and calibrate the parameters of the simulation model to ensure the accuracy and reliability of the model. Based on the operating requirements and performance objectives of the engine, make control feature decisions for a future period of time. These decisions may include adjusting the engine speed, load, cooling water flow rate, etc. Use optimization algorithms (such as genetic algorithms, particle swarm algorithms, etc.) to optimize the control feature decisions to obtain the best control effect. Apply the optimized control feature decisions to the engine simulation model for K times of simulated control. Each simulated control represents the operating state of the engine at different future time points. During each simulated control process, record the temperature change data of the engine to generate K engine temperature simulation data sets. Preprocess the K engine temperature simulation data sets, including data cleaning, denoising, interpolation and other operations. Use statistical methods (such as mean, median, etc.) to calculate the temperature concentration value at each time point to generate an engine temperature prediction sequence set. According to the component composition feature information of the engine (such as cylinder position, cooling system structure, etc.), extract key features from the engine temperature prediction sequence set. Perform clustering analysis on the extracted features to generate a temperature prediction distribution feature point cloud. Each point in the point cloud represents the possible temperature distribution of the engine at a future time point.

[0027] For example, a simulation model is established for a four-cylinder gasoline engine. According to the actual structure of the engine, such as the cylinder layout, intake and exhaust system, cooling system, etc., a corresponding physical model is built in Simulink. Then, by adjusting the parameters of the model (such as combustion efficiency, heat transfer coefficient, etc.), the simulation results are made to match the actual experimental data. Set K = 10, that is, 10 times of simulation control are carried out. Each control is based on different combinations of engine speed and load. Through the simulation run, 10 sets of engine temperature simulation data are obtained, and each data set contains the temperature change of the engine under different working conditions. After data processing and calculation of the central value, a set of engine temperature prediction sequences containing 10 time points is obtained. Then, according to the component composition characteristic information of the engine (such as the cooling effect difference of the cylinders, the flow path of the cooling water, etc.), key features are extracted from the prediction sequence set. Finally, the K-means clustering algorithm is used to cluster and analyze the features, generating a cloud of characteristic points of the temperature prediction distribution. Each point in the cloud represents the possible temperature distribution range of the engine at a certain future time point under different working conditions. Through the above steps, the temperature feature prediction module can be used to model and simulate the control of the engine, generating multiple sets of engine temperature simulation data. Then, through the calculation of the central value of the spatio-temporal feature set and the feature clustering analysis, a cloud of characteristic points of the engine temperature prediction distribution is generated. This process not only helps to comprehensively understand the temperature change of the engine under different future working conditions, but also provides important data support for subsequent temperature measurement decisions.

[0028] Further, the system is used to perform the following steps: Obtain the first component characteristic structure information of the engine; perform feature association on the cloud of characteristic points of the temperature prediction distribution according to the first component characteristic structure information to obtain the first component structure temperature association cloud; activate the temperature measurement decision-making module, where the temperature measurement decision-making module includes multi-dimensional temperature measurement decision-making factors, and the multi-dimensional temperature measurement decision-making factors include temperature monitoring positions and temperature monitoring frequencies; based on the first component structure temperature association cloud, according to the temperature measurement decision-making module, obtain the first component temperature measurement decision, where the first component temperature measurement decision includes the first component temperature measurement spatial domain characteristic information and the first component temperature measurement frequency domain characteristic information; add the first component temperature measurement decision to the temperature measurement characteristic decision, and continue to perform temperature monitoring characteristic decision on the engine according to the cloud of characteristic points of the temperature prediction distribution and the temperature measurement decision-making module to generate the temperature measurement characteristic decision.

[0029] Obtain the detailed characteristic structure information of the first component (such as cylinder head, piston, crankshaft, etc.) in the engine. This information usually includes the position, size, material, heat conduction characteristics of the component, and the connection method with other components, etc. This information is crucial for understanding the temperature distribution and changes of the component during engine operation. According to the obtained characteristic structure information of the first component, perform feature association on the previously generated temperature prediction distribution feature point cloud. The purpose of this step is to correspond the temperature prediction data in the point cloud with the specific component structure, so as to obtain the first component structure temperature association point cloud. This associated point cloud can more intuitively display the temperature distribution of the first component under different future working conditions. Activate the temperature measurement decision-making module. This module contains multi-dimensional temperature measurement decision-making factors, such as temperature monitoring positions and temperature monitoring frequencies. These factors are carefully designed according to the actual situation of the engine and monitoring requirements, aiming to ensure the accuracy and effectiveness of temperature monitoring. Based on the first component structure temperature association point cloud and the multi-dimensional temperature measurement decision-making factors in the temperature measurement decision-making module, formulate the temperature measurement decision for the first component. This decision includes two main parts, the first component temperature measurement airspace characteristic information, which specifies the specific positions of temperature monitoring. These positions are usually the areas where the temperature changes most significantly on the component or have the greatest impact on engine performance. By arranging temperature sensors at these positions, the most representative temperature data can be obtained. The first component temperature measurement frequency domain characteristic information determines the temperature monitoring frequency. This frequency depends on the temperature change speed of the component and monitoring requirements. For components with faster temperature changes, a higher monitoring frequency is required to ensure the real-time and accuracy of the data; while for components with slower temperature changes, the monitoring frequency can be appropriately reduced to save resources. Add the formulated temperature measurement decision for the first component to the original temperature measurement characteristic decision. In this way, the temperature measurement characteristic decision contains detailed temperature measurement schemes for multiple components of the engine. Then, according to the updated temperature measurement characteristic decision and the temperature prediction distribution feature point cloud, continue to make temperature monitoring characteristic decisions for the engine. This process may be an iterative process. With the addition of new monitoring data and the change of engine working conditions, the temperature measurement characteristic decision will also be continuously updated and optimized.

[0030] For example, formulate a temperature measurement decision for the cylinder head of the engine. First, obtain the characteristic structure information of the cylinder head, including its position, size, material, etc. Then, associate this information with the temperature prediction distribution feature point cloud to obtain the cylinder head structure temperature association point cloud. Next, activate the temperature measurement decision-making module and formulate a temperature measurement decision according to the structural characteristics of the cylinder head and the temperature prediction data. For example, temperature sensors may be set respectively at the upper part (close to the combustion chamber) and the lower part (connected to the cooling system) of the cylinder head, and different monitoring frequencies are set to capture the details of temperature changes. Finally, add this decision to the temperature measurement characteristic decision and continue the similar decision-making process for other components of the engine.

[0031] Further, the system is used to perform the following steps: Set the retrieval subject distribution, where the retrieval subject distribution includes the engine and multiple engines of the same model corresponding to the engine; retrieve the temperature monitoring feature decision records according to the retrieval subject distribution to obtain the engine component structure temperature correlation point cloud records and the engine component temperature measurement decision point cloud records; perform feature recognition on the engine component temperature measurement decision point cloud records according to the multi-dimensional temperature measurement decision factors to obtain the engine component temperature measurement decision point cloud feature records; perform supervised training on the temperature measurement decision learning model according to the engine component structure temperature correlation point cloud records and the engine component temperature measurement decision point cloud feature records, and obtain the temperature measurement decision accuracy every time a predetermined number of times of training is completed; when the temperature measurement decision accuracy is greater than the predetermined temperature measurement decision accuracy, generate the temperature measurement decision module.

[0032] Set a distribution including multiple retrieval subjects, where the main retrieval subject is the target engine and multiple engines of the same model corresponding to it. These engines of the same model are structurally similar to the target engine, but may have different temperature distributions and changes due to manufacturing differences, usage wear, etc. Incorporating them into the retrieval subject distribution helps improve the generalization ability and adaptability of the temperature measurement decision. According to the set retrieval subject distribution, retrieve the temperature monitoring feature decision records. This includes retrieving the engine component structure temperature correlation point cloud records and the engine component temperature measurement decision point cloud records related to the target engine and its engines of the same model from the database. These records contain the temperature distributions and temperature measurement decision situations of the engine components under different working conditions and different time periods, and are important data sources for training the temperature measurement decision learning model. For the retrieved engine component temperature measurement decision point cloud records, use multi-dimensional temperature measurement decision factors for feature recognition. The multi-dimensional temperature measurement decision factors include the temperature monitoring position, temperature monitoring frequency, etc., and they jointly define the feature space of the temperature measurement decision. Through feature recognition, key feature information can be extracted from the temperature measurement decision point cloud records to form the engine component temperature measurement decision point cloud feature records. Use the engine component structure temperature correlation point cloud records and the engine component temperature measurement decision point cloud feature records as inputs to perform supervised training on the temperature measurement decision learning model. During the training process, the model will learn how to identify key temperature change patterns from the component structure temperature correlation point cloud and generate corresponding temperature measurement decisions based on these patterns. After each predetermined number of times of training, evaluate the temperature measurement decision accuracy of the model to understand the learning effect and generalization ability of the model. When the accuracy of the temperature measurement decision learning model is greater than the predetermined temperature measurement decision accuracy, it is considered that the model already has sufficient accuracy and stability to generate the final temperature measurement decision module. This module will contain the trained model parameters and algorithm logic and can automatically generate corresponding temperature measurement decisions according to the structure temperature correlation point cloud of the engine components.

[0033] For example, assume that a temperature measurement decision-making module is being developed for a certain model of aeroengine. First, historical operation data of this model of engine and multiple engines of the same model are collected, including temperature monitoring records and temperature measurement decision-making records under different operating conditions. Then, a retrieval subject distribution is set up and these data are incorporated into it. Next, multi-dimensional temperature measurement decision-making factors are used to identify the characteristics of the temperature measurement decision-making records, and key characteristic information is extracted. Then, using these characteristics and the component structure temperature correlation point cloud data, the temperature measurement decision-making learning model is trained. After multiple iterative trainings, the temperature measurement decision-making accuracy of the model reaches the predetermined requirements. Finally, a temperature measurement decision-making module is generated based on the trained model and deployed into the actual application scenario. In this way, when the engine is running, the temperature measurement decision-making module can automatically generate accurate temperature measurement decisions based on the real-time component structure temperature correlation point cloud data.

[0034] Further, the system is used to perform the following steps: Obtain a predetermined risk coefficient; according to the predetermined risk coefficient, establish a risk prediction factor, where the risk prediction factor is the probability greater than or equal to the predetermined risk coefficient; based on the risk prediction factor and the temperature prediction distribution characteristic point cloud, conduct risk prediction to obtain multiple component risk trigger coefficients; determine whether the multiple component risk trigger coefficients are greater than or equal to the predetermined risk trigger probability to obtain the temperature measurement adjustment factor.

[0035] Determine a predetermined risk coefficient. This coefficient is obtained by comprehensively considering various factors such as the engine's design specifications, operating history, safety standards, and industry experience. It represents an acceptable risk level, that is, when the risk level of a component exceeds this coefficient, it is considered necessary to take additional monitoring or preventive measures. Based on the predetermined risk coefficient, establish a risk prediction factor. The risk prediction factor is one or more indicators used to quantify the likelihood of future risks occurring in a component. In this context, the risk prediction factor is defined as the probability of being greater than or equal to the predetermined risk coefficient. This means that if the risk prediction value (i.e., the probability of being greater than or equal to the predetermined risk coefficient) of a certain component at a future time point is high, then the component is considered to have a high risk. To establish the risk prediction factor, various factors need to be considered, such as the component's working environment, material properties, historical failure data, current operating status, etc. These factors can be quantified into specific risk prediction values through statistical analysis, machine learning models, or other prediction techniques. Based on the risk prediction factor and the temperature prediction distribution characteristic point cloud, conduct a risk prediction for engine components. The temperature prediction distribution characteristic point cloud provides the temperature distribution of the component under different operating conditions and times, while the risk prediction factor provides a risk assessment based on these temperature distributions. By combining the two, the risk trigger coefficients of multiple components at different future time points can be calculated. The risk trigger coefficient is a quantitative indicator used to represent the likelihood of a component experiencing a risk at a future time point. It may be obtained through a comprehensive calculation based on the temperature prediction results and the risk prediction factor. Judge whether the risk trigger coefficients of multiple components are greater than or equal to the predetermined risk trigger probability. The predetermined risk trigger probability is a set threshold used to determine whether the component risk has reached the level where measures need to be taken. If the risk trigger coefficient of a certain component is greater than or equal to this threshold, then it is considered that the component has a high risk and corresponding measures need to be taken to reduce the risk. According to the results of the risk prediction, obtain a temperature measurement adjustment factor. The temperature measurement adjustment factor is an indicator used to guide the adjustment of the temperature monitoring strategy. It includes multiple aspects, such as the adjustment of the monitoring frequency, the optimization of the monitoring location, the improvement of the monitoring accuracy, etc. According to the different risk trigger coefficients, different temperature measurement adjustment strategies can be formulated for different components, and these strategies can be implemented by modifying the relevant parameters in the temperature measurement decision-making module.

[0036] For example, assume that a risk prediction is being performed on the cylinder head of an engine. First, a predetermined risk coefficient of 0.05 is determined. Then, risk prediction factors are established based on factors such as the working environment, material properties, and historical failure data of the cylinder head. Next, risk prediction is performed using the temperature data in the temperature prediction distribution characteristic point cloud and the risk prediction factors, and the risk trigger coefficients of the cylinder head at different future time points are calculated. Finally, these coefficients are compared with a predetermined risk trigger probability (such as 0.05), and it is found that the risk trigger coefficient at a certain time point exceeds the threshold. Therefore, a temperature measurement adjustment factor is obtained, and based on this, the temperature monitoring strategy of the cylinder head is adjusted, such as increasing the monitoring frequency or optimizing the monitoring location.

[0037] Further, the system is used to perform the following steps: According to the temperature prediction distribution characteristic point cloud, extract the temperature prediction distribution point cloud of the first engine component; obtain the first engine component attribute characteristic information corresponding to the temperature prediction distribution point cloud of the first engine component; according to the first engine component attribute characteristic information, load the first engine component temperature distribution point cloud sample set and the first engine component risk coefficient sample set, where the first engine component temperature distribution point cloud sample set and the first engine component risk coefficient sample set have a one-to-many relationship; based on the risk prediction factor, according to the first engine component risk coefficient sample set, calculate the first component risk trigger sample set, where the first component risk trigger sample set has a one-to-one correspondence with the first engine component temperature distribution point cloud sample set; according to the first engine component temperature distribution point cloud sample set and the first engine component risk coefficient sample set, build the first engine component risk trigger prediction channel; based on the temperature prediction distribution point cloud of the first engine component, according to the first engine component risk trigger prediction channel, generate the first component risk trigger coefficient, and add the first component risk trigger coefficient to the multiple component risk trigger coefficients.

[0038] From the overall temperature prediction distribution feature point cloud, extract the temperature prediction distribution point cloud related to the first engine component. This point cloud data contains the temperature prediction values of the component under different working conditions and times, and is the basis for subsequent risk prediction. Obtain the attribute feature information of the first engine component. This information may include the material, size, position, working environment, etc. of the component, which are crucial for understanding the temperature distribution and risk characteristics of the component. According to the attribute feature information of the first engine component, two sample sets are loaded: the temperature distribution point cloud sample set of the first engine component and the risk coefficient sample set of the first engine component. These two sample sets have a one-to-many relationship, that is, one temperature distribution point cloud sample may correspond to multiple risk coefficient samples, because the risk coefficient may vary due to various factors. Based on the risk prediction factor and the risk coefficient sample set of the first engine component, calculate the first component risk trigger sample set. This sample set is obtained by screening and calculating the risk coefficient samples, and it reflects the possibility of the component having risks under different temperature distributions. During the calculation process, statistical models, machine learning algorithms or other prediction techniques can be used to evaluate the probability of risk triggering. According to the temperature distribution point cloud sample set of the first engine component and the risk coefficient sample set of the first engine component, build the risk trigger prediction channel of the first engine component. This prediction channel is a model or algorithm that maps the temperature distribution point cloud to the risk trigger coefficient, and it can automatically generate the corresponding risk trigger coefficient according to the input temperature prediction distribution point cloud. Based on the temperature prediction distribution point cloud of the first engine component and the built risk trigger prediction channel, generate the risk trigger coefficient of the first component. This coefficient is a quantitative indicator used to represent the possibility of the component having risks at a certain future time point or working condition. Add this coefficient to the set of risk trigger coefficients of multiple components for overall risk assessment and monitoring. Through the above steps, detailed risk trigger prediction and evaluation are carried out for the first engine component. This process not only considers the temperature distribution characteristics of the component, but also combines the attribute feature information and historical risk data of the component, thereby improving the accuracy and reliability of risk prediction. The finally generated risk trigger coefficient will be used to guide the adjustment and optimization of the temperature monitoring strategy.

[0039] Further, the system is used to perform the following steps: Locate the temperature measurement feature decision according to the temperature measurement adjustment factor, and determine the temperature measurement decisions of multiple components to be adjusted; generate multiple temperature measurement adjustment levels according to the temperature measurement adjustment factor; respectively optimize the temperature measurement airspace features of the temperature measurement decisions of the multiple components to be adjusted according to the multiple temperature measurement adjustment levels, and generate multiple first optimized component temperature measurement decisions; based on the multiple temperature measurement adjustment levels, optimize the temperature measurement frequency domain features according to the multiple first optimized component temperature measurement decisions, and generate multiple second optimized component temperature measurement decisions; update the temperature measurement feature decision according to the multiple second optimized component temperature measurement decisions to obtain the optimized temperature measurement feature decision.

[0040] According to the temperature measurement adjustment factor, locate the components that need to adjust the temperature monitoring strategy, namely multiple components to be adjusted. These components may be components with a relatively high risk trigger coefficient, or components that need to be monitored more frequently or accurately based on historical data and experience judgment. Generate a temperature measurement adjustment level for each component to be adjusted according to the temperature measurement adjustment factor. This level reflects the priority and intensity of component temperature monitoring, and may be obtained by comprehensively considering various factors such as the risk trigger coefficient, component importance, and complexity of the working environment. Different adjustment levels correspond to different monitoring frequencies, precisions, and coverage ranges. According to the temperature measurement adjustment level, optimize the temperature measurement airspace features of the temperature measurement decisions of multiple components to be adjusted. The airspace feature optimization mainly focuses on the layout and coverage range of the monitoring points. Specifically, it may increase the monitoring point density in high-risk areas, or adjust the positions of the monitoring points to better capture the key information of temperature changes. By optimizing the airspace features, the accuracy and comprehensiveness of temperature monitoring can be improved. On the basis of the airspace feature optimization, further optimize the temperature measurement frequency domain features. The frequency domain feature optimization mainly focuses on the adjustment of the monitoring frequency. According to the temperature measurement adjustment level, the monitoring frequency of high-risk components may be increased to capture temperature anomalies more timely; while for components with lower risks, the monitoring frequency can be appropriately reduced to save resources. By optimizing the frequency domain features, the timeliness and economy of temperature monitoring can be ensured. Update the entire temperature measurement feature decision system according to the multiple second optimized component temperature measurement decisions. This process includes integrating the optimized component temperature measurement decisions into the original decision system, and adjusting relevant parameters and rules to ensure the coordination and consistency of the entire system. The updated temperature measurement feature decision will be used as the basis and guidance for future temperature monitoring, which helps to improve the operation safety and reliability of the engine. Through the above steps, the temperature measurement feature decision of the engine is comprehensively optimized according to the temperature measurement adjustment factor. This optimization process covers multiple aspects such as the location of components to be adjusted, the generation of temperature measurement adjustment levels, the optimization of temperature measurement airspace and frequency domain features, and the update of temperature measurement feature decisions. By implementing these optimization measures, the efficiency and accuracy of temperature monitoring can be improved, providing more powerful guarantee for the safe operation of the engine.

[0041] Such as Figure 2As shown, further, the system is used to perform the following steps: Connect the temperature monitoring module to obtain engine temperature measurement data; obtain the induction data of the temperature measurement device corresponding to the engine temperature measurement data; perform anomaly monitoring based on the induction data of the temperature measurement device to obtain the anomaly detection result of the temperature measurement device; optimize and compensate the engine temperature measurement data according to the anomaly detection result of the temperature measurement device to generate optimized temperature measurement data.

[0042] Connect the temperature monitoring module to the temperature measurement device (such as a temperature sensor) on the engine. Once the connection is successful, the temperature monitoring module can receive and process the temperature data from the temperature measurement device in real time, that is, the engine temperature measurement data, which is the basis for subsequent analysis and optimization. In addition to the engine temperature data, it is also necessary to obtain the induction data of the temperature measurement device related to the temperature monitoring process. This includes the operating status of the temperature measurement device, the working environment parameters (such as temperature, humidity, vibration, etc.), and possible device failure or error codes. These induction data are crucial for evaluating the accuracy and reliability of the temperature measurement device. Use the obtained induction data of the temperature measurement device to perform anomaly monitoring. By comparing the induction data with the preset normal range or threshold, possible anomalies of the temperature measurement device can be identified, such as sensor failure, signal interference, or data transmission error, etc. Anomalies may lead to inaccurate or missing temperature data. Once an anomaly is detected, record and generate the anomaly detection result of the temperature measurement device. After obtaining the anomaly detection result of the temperature measurement device, it is necessary to optimize and compensate the engine temperature measurement data according to these results. The purpose of compensation is to eliminate or reduce the temperature data error caused by the anomaly of the temperature measurement device. The compensation methods may include data correction. For the incorrect data caused by sensor failure or signal interference, historical data, model prediction, or data from adjacent sensors can be used for correction. Data interpolation. If the data is missing due to transmission error or device failure, interpolation methods can be used to fill in the missing data points to maintain the continuity and integrity of the data. Weight adjustment. In the case of multiple sensors monitoring simultaneously, the weight of the data of each sensor can be adjusted according to the reliability and accuracy of each sensor to improve the accuracy of the overall temperature data. After the above optimization and compensation steps, the optimized engine temperature measurement data will be obtained. These data are more accurate and reliable and can more truly reflect the actual temperature state of the engine. These optimized temperature measurement data will be used as an important basis for subsequent analysis and decision-making to help improve the operating efficiency and safety of the engine.

[0043] In summary, the beneficial effects of this application include: 1. It improves the flexibility and comprehensiveness of temperature monitoring, can more accurately reflect the overall temperature distribution of the engine, and timely discovers problems such as local overheating.

[0044] 2. It enhances the adaptability and intelligence of the system, and can maintain high measurement accuracy and stability even under complex and changeable working conditions.

[0045] 3. It realizes real-time monitoring and intelligent analysis of the engine temperature, can predict potential temperature problems in advance, and provides timely warnings and decision-making support for operators.

[0046] 4. It improves the remote monitoring ability and fault diagnosis efficiency of the system, reduces the maintenance cost and downtime. At the same time, it also makes it possible for remote maintenance and optimization of the engine.

[0047] For specific embodiments of the intelligent temperature monitoring system for engines, reference can be made to the embodiments described above, which will not be elaborated here. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to each of the above modules.

[0048] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0049] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application.

Claims

1. An intelligent temperature monitoring system for an engine, characterized in that: The system comprises: A temperature measurement instruction receiving module, the temperature measurement instruction receiving module is used to receive a temperature measurement instruction of the engine, wherein the temperature measurement instruction includes a future temperature measurement time domain and a future control feature decision corresponding to the engine; A temperature feature prediction module, wherein the temperature feature prediction module performs temperature feature fitting on the engine based on the future temperature measurement time domain and the future control feature decision to obtain a temperature prediction distribution feature point cloud; A temperature measurement decision module, the temperature measurement decision module is used to perform a temperature monitoring feature decision on the engine based on the temperature prediction distribution feature point cloud to obtain a temperature measurement feature decision; A temperature measurement adjustment factor analysis module, the temperature measurement adjustment factor analysis module is used to perform risk prediction on the engine based on the temperature prediction distribution feature point cloud, and generate a temperature measurement adjustment factor in combination with a predetermined risk trigger probability; A temperature measurement decision compensation module, the temperature measurement decision compensation module is used to optimize and compensate the temperature measurement feature decision based on the temperature measurement adjustment factor to generate an optimized temperature measurement feature decision; A temperature monitoring module is used to monitor the temperature of the engine based on the optimized temperature measurement feature decision.

2. The intelligent temperature monitoring system for an engine according to claim 1, characterized in that: The temperature feature prediction module performs temperature feature fitting on the engine based on the future temperature measurement time domain and the future control feature decision to obtain a temperature prediction distribution feature point cloud, including: Modeling the engine according to the temperature characteristic prediction module to obtain an engine simulation model; Based on the future temperature measurement time domain and the temperature feature prediction module, performing K simulation controls on the engine simulation model according to the future control feature decision to generate K engine temperature simulation data sets, where K is a positive integer greater than 1; Performing simultaneous and spatial feature concentration calculation based on the K engine temperature simulation data sets to generate an engine temperature prediction sequence set; The engine temperature prediction sequence set is subjected to feature clustering according to the component composition feature information of the engine to generate the temperature prediction distribution feature point cloud.

3. The intelligent temperature monitoring system for an engine as claimed in claim 1, characterized in that: The temperature measurement decision module is used to perform temperature monitoring feature decision on the engine based on the temperature prediction distribution feature point cloud to obtain a temperature measurement feature decision, including: Obtaining characteristic structural information of a first component of the engine; Performing feature association on the temperature prediction distribution feature point cloud according to the feature structure information of the first component to obtain a first component structure temperature associated point cloud; Activate the temperature measurement decision module, wherein the temperature measurement decision module includes a multi-dimensional temperature measurement decision factor, and the multi-dimensional temperature measurement decision factor includes a temperature monitoring position and a temperature monitoring frequency; Based on the temperature-related point cloud of the first component structure, and according to the temperature measurement decision module, a first component temperature measurement decision is obtained, wherein the first component temperature measurement decision includes spatial domain feature information of the first component temperature measurement and frequency domain feature information of the first component temperature measurement; The first component temperature measurement decision is added to the temperature measurement feature decision, and the temperature monitoring feature decision of the engine is continued according to the temperature prediction distribution feature point cloud and the temperature measurement decision module to generate the temperature measurement feature decision.

4. The intelligent temperature monitoring system for an engine as claimed in claim 3, characterized in that: The steps of constructing the temperature measurement decision module include: Setting a search subject distribution, wherein the search subject distribution includes the engine and a plurality of engines of the same model corresponding to the engine; Perform temperature monitoring feature decision record retrieval according to the retrieval subject distribution to obtain engine component structure temperature associated point cloud records and engine component temperature measurement decision point cloud records; Performing feature recognition on the engine component temperature measurement decision point cloud record according to the multi-dimensional temperature measurement decision factor to obtain the engine component temperature measurement decision point cloud feature record; Performing supervised training on the temperature measurement decision learning model according to the temperature-related point cloud record of the engine component structure and the temperature measurement decision point cloud feature record of the engine component, and obtaining the temperature measurement decision accuracy after each predetermined number of trainings; When the temperature measurement decision accuracy is greater than a predetermined temperature measurement decision accuracy, the temperature measurement decision module is generated.

5. The intelligent temperature monitoring system for an engine as claimed in claim 1, characterized in that: The temperature measurement adjustment factor analysis module is used to perform risk prediction on the engine based on the temperature prediction distribution feature point cloud, and generate a temperature measurement adjustment factor in combination with a predetermined risk trigger probability, including: Obtaining a predetermined risk factor; According to the predetermined risk coefficient, a risk prediction factor is established, wherein the risk prediction factor is a probability greater than / equal to the predetermined risk coefficient; Perform risk prediction based on the risk prediction factor and the temperature prediction distribution feature point cloud to obtain risk trigger coefficients of multiple components; It is determined whether the risk trigger coefficients of the multiple components are greater than / equal to the predetermined risk trigger probability, and the temperature measurement adjustment factor is obtained.

6. The intelligent temperature monitoring system for an engine as claimed in claim 5, characterized in that: Risk prediction is performed based on the risk prediction factor and the temperature prediction distribution feature point cloud to obtain multiple component risk trigger coefficients, including: Extracting a first engine component temperature prediction distribution point cloud according to the temperature prediction distribution feature point cloud; Obtaining attribute feature information of a first engine component corresponding to the predicted distribution point cloud of the temperature of the first engine component; According to the first engine component attribute characteristic information, a first engine component temperature distribution point cloud sample set and a first engine component risk coefficient sample set are loaded, wherein the first engine component temperature distribution point cloud sample set and the first engine component risk coefficient sample set have a one-to-many relationship; Based on the risk prediction factor, a first component risk triggering sample set is calculated according to the first engine component risk coefficient sample set, wherein the first component risk triggering sample set has a one-to-one correspondence with the first engine component temperature distribution point cloud sample set; Building a first engine component risk trigger prediction channel according to the first engine component temperature distribution point cloud sample set and the first engine component risk coefficient sample set; Based on the first engine component temperature prediction distribution point cloud and according to the first engine component risk trigger prediction channel, a first component risk trigger coefficient is generated, and the first component risk trigger coefficient is added to the multiple component risk trigger coefficients.

7. The intelligent temperature monitoring system for an engine as claimed in claim 1, characterized in that: The temperature measurement decision compensation module is used to optimize and compensate the temperature measurement feature decision based on the temperature measurement adjustment factor to generate an optimized temperature measurement feature decision, including: Locating the temperature measurement feature decision according to the temperature measurement adjustment factor, and determining a plurality of temperature measurement decisions of components to be adjusted; Generating a plurality of temperature measurement adjustment levels according to the temperature measurement adjustment factor; According to the multiple temperature measurement adjustment levels, respectively optimize the temperature measurement spatial characteristics of the multiple temperature measurement decisions of the components to be adjusted to generate multiple first optimized component temperature measurement decisions; Based on the multiple temperature measurement adjustment levels, optimizing the temperature measurement frequency domain characteristics according to the multiple first optimized component temperature measurement decisions, and generating multiple second optimized component temperature measurement decisions; The temperature measurement feature decision is updated according to the multiple second optimized component temperature measurement decisions to obtain the optimized temperature measurement feature decision.

8. The intelligent temperature monitoring system for an engine as claimed in claim 1, characterized in that: The system further comprises a monitoring and compensation module, wherein the monitoring and compensation module is used for: Connecting the temperature monitoring module to obtain engine temperature measurement data; Obtaining temperature measurement device sensing data corresponding to the engine temperature measurement data; Perform abnormal monitoring based on the sensing data of the temperature measuring device to obtain abnormal detection results of the temperature measuring device; The engine temperature measurement data is optimized and compensated according to the abnormality detection result of the temperature measurement equipment to generate optimized temperature measurement data.

Images