A multi-point temperature measurement method and system based on digital temperature sensor

Abstract

The present invention discloses a multi-point temperature measurement method and system based on digital temperature sensors, relating to the field of temperature sensor technology. The method comprises: selecting temperature sensors and arranging measurement locations; performing temperature measurements within the measurement location layout range; acquiring measurement data and processing and analyzing it; calibrating and temperature-compensating the temperature sensors based on the analysis results; storing the measurement data and analysis results to generate a storage database; selecting data from the storage database to construct an associated data set and establish a temperature knowledge graph; selecting data from the storage database to construct a prediction data set and establish a temperature prediction model; establishing an intelligent control model based on the temperature knowledge graph and the temperature prediction model, and adjusting and optimizing multi-point temperature measurements. The present invention improves the accuracy, real-time performance, and stability of multi-point temperature measurements by selecting temperature sensors, arranging measurements, calibrating and compensating, and synchronously processing and analyzing measurement data.

Description

Technical Field

[0001] The present invention belongs to the technical field of temperature sensors, and in particular relates to a multi-point temperature measurement method and system based on digital temperature sensors. Background Art

[0002] A digital temperature sensor is a device that converts temperature into a digital output signal. Typically manufactured from semiconductor materials, common types include thermistors (such as NTC and PTC), thermistor networks, digital temperature sensor chips (such as the DS18B20), thermocouples, and silicon-based sensors. These sensors typically interface with a microcontroller or digital signal processor to read and process temperature data. Digital temperature sensors offer advantages such as high accuracy, excellent stability, fast response, ease of integration into digital systems, and strong anti-interference capabilities. They are widely used in industrial control, automotive electronics, medical equipment, household appliances, meteorological measurement, and other technical fields.

[0003] Multi-point temperature measurement refers to measuring the temperature at multiple locations of an area or object at the same time. This measurement method can provide more comprehensive and detailed temperature information, which is beneficial for temperature monitoring and control in complex environments or large systems.

[0004] Although digital temperature sensors have the above advantages, multi-point measurement using digital temperature sensors still has problems such as sensor accuracy differences, improper layout, insufficient calibration, measurement data delays and imperfect data processing, which in turn affect the accuracy, real-time nature and stability of multi-point temperature measurement. Summary of the Invention

[0005] The object of the present invention is to provide a multi-point temperature measurement method and system based on a digital temperature sensor, which can be achieved by the following technical solutions:

[0006] In a first aspect, an embodiment of the present application provides a multi-point temperature measurement method based on a digital temperature sensor, comprising the following steps:

[0007] S1, selecting a temperature sensor and arranging the measurement positions of the temperature sensor;

[0008] S2, performing temperature measurement within the layout range of the measurement position;

[0009] S3, acquiring measurement data and performing data processing and data analysis;

[0010] S4, calibrating the temperature sensor according to the analysis result;

[0011] S5, performing temperature compensation on the temperature sensor according to the analysis result;

[0012] S6, storing the measurement data and the analysis results to generate a storage database;

[0013] S7, selecting data from the storage database to construct a related data set;

[0014] S8, constructing a temperature knowledge graph based on the associated data set;

[0015] S9, selecting data from the storage database to construct a prediction data set;

[0016] S10, constructing a temperature prediction model based on the prediction data set;

[0017] S11, establishing an intelligent control model based on the temperature knowledge graph and the temperature prediction model;

[0018] S12, adjusting and optimizing multi-point temperature measurement according to the intelligent control model;

[0019] The establishment of the intelligent control model specifically includes the following steps:

[0020] Obtaining entities, relationships, and attributes in the temperature knowledge graph and using them as graph features;

[0021] Obtaining the spatiotemporal features and temperature features in the temperature prediction model and using them as prediction features;

[0022] Fusing the atlas features with the prediction features to generate regulatory features;

[0023] Feature selection and model training are performed based on the control features to generate the intelligent control model.

[0024] Preferably, selecting the temperature sensor specifically includes the following steps:

[0025] Get parameter information of several digital temperature sensors;

[0026] Acquire environmental data around the measurement location;

[0027] Get the measurement requirements of the temperature sensor;

[0028] The parameter information and the environmental data are analyzed, and a temperature sensor is selected based on the measurement requirements.

[0029] Preferably, a multi-factor decision analysis algorithm is used to select a suitable temperature sensor, specifically comprising the following steps:

[0030] Use measurement needs as decision-making objectives;

[0031] Incorporating parameter information and environmental data into decision-making factors;

[0032] Converting the decision factors into quantifiable decision indicators;

[0033] Collecting data associated with the decision indicators and determining weights of the decision indicators;

[0034] Establishing a decision model according to the decision factors and the weights;

[0035] Outputting a plurality of decision plans through the decision model, evaluating the plurality of decision plans, and outputting evaluation results;

[0036] Performing sensitivity analysis on the decision model and outputting analysis results;

[0037] An optimal decision-making solution is selected according to the evaluation results and the analysis results.

[0038] Preferably, the measurement positions of the temperature sensors are arranged, specifically comprising the following steps:

[0039] Determine a temperature measurement area according to measurement requirements and set a measurement range within the temperature measurement area;

[0040] Selecting several reference measurement points within the measurement range;

[0041] forming an initial measurement interval according to the reference measurement point;

[0042] Performing temperature detection on the initial measurement interval using a first measurement interval and obtaining a reference measurement value;

[0043] Selecting actual measurement points and determining the ultimate measurement interval based on the reference measurement values;

[0044] Using a second measurement interval to perform temperature detection on the final measurement interval and obtain an actual measurement value;

[0045] forming a multi-point temperature measurement area according to the actual measurement values;

[0046] The multi-point temperature measurement area is analyzed and all measurement positions within the area are obtained.

[0047] Preferably, step S3 includes:

[0048] S31, adding a timestamp and a synchronized clock during the multi-point temperature measurement process and obtaining measurement data of the multi-point temperature measurement, specifically including:

[0049] Generate a synchronization signal before temperature measurement, the synchronization signal is used to control the temperature sensor to perform synchronous acquisition;

[0050] During the temperature measurement process, the measurement time is recorded by time stamp and synchronized clock;

[0051] After the temperature measurement is completed, the position information of each measuring point is recorded;

[0052] Correspondingly associating the time information of temperature measurement with the position information and generating a time-position sequence table;

[0053] S32, processing and analyzing the measurement data to generate a first analysis result and a second analysis result, specifically including:

[0054] Preprocessing the measurement data to remove error data and incomplete data;

[0055] performing smoothing and standardization processing on the measurement data;

[0056] Performing outlier detection on the preprocessed measurement data and generating a first analysis result based on the outlier detection;

[0057] Obtain reference measurement values ​​and actual measurement values ​​in the measurement location layout;

[0058] Acquire mutation data appearing therein according to changes between the reference measurement value and the actual measurement value;

[0059] Determining a temperature gradient and a mutation position according to the mutation data, and generating a second analysis result therefrom;

[0060] S33: Generate a first calibration solution according to the first analysis result, and generate a second compensation solution according to the second analysis result.

[0061] Preferably, the outlier detection is specifically expressed as follows: ;

[0062] in, is the expected normal value; represents the standard value; t represents the time point;

[0063] The standard value Also expressed as: ;

[0064] in, Represents the standard deviation of global data; λ is used to control the size of the standard deviation;

[0065] Regarding obtaining the above mutation data, specifically:

[0066] Acquiring continuous temperature data of the reference measurement value and the actual measurement value;

[0067] Before filtering the continuous temperature data, edge position detection is performed on the continuous temperature data to determine whether the edge position is a temperature mutation location. If so, the temperature mutation location is marked; if not, smoothing filtering is performed on the edge position.

[0068] Obtain mutation data based on the marked temperature mutation location.

[0069] Preferably, regarding edge position detection, specifically:

[0070] Perform difference processing on adjacent different temperature points in continuous temperature data;

[0071] When the absolute value of the difference between at least two consecutive adjacent different temperature points is greater than a preset threshold, the area between the at least two consecutive adjacent different temperature points is determined to be a suspected temperature mutation point;

[0072] To determine whether the edge position is a temperature mutation point, specifically:

[0073] Compare the edge detection results with the preset accuracy threshold;

[0074] If the edge detection result is greater than the accuracy threshold, then this edge is where the temperature changes suddenly;

[0075] If the edge detection result is less than or equal to the accuracy threshold, the edge is a noise mutation point.

[0076] Preferably, steps S7 to S8 specifically include the following:

[0077] Selecting related data from the storage database to construct a related data set;

[0078] Performing entity recognition and relationship extraction based on the associated dataset;

[0079] Perform knowledge representation on the extracted entities and relationships and construct a temperature knowledge graph;

[0080] Identifying entities in the associated data, and grouping the entities in pairs to form entity pairs;

[0081] For each entity pair, obtaining a data vector of the associated data;

[0082] Extracting representative features according to the data vector and performing feature screening on the representative features;

[0083] Predicting the entity relationship of the entity pair based on the representation features;

[0084] Constructing the temperature knowledge graph based on the entity relationships corresponding to the entity pairs;

[0085] The associated data refers to data that are associated with each other.

[0086] Preferably, steps S9 to S10 specifically include the following:

[0087] Selecting prediction data from the storage database to construct a prediction data set;

[0088] Acquire several sample data groups and test data groups based on the prediction data set;

[0089] Using the sample data set to perform feature selection and extraction, including spatiotemporal features and temperature features;

[0090] Performing model training according to the spatiotemporal characteristics and temperature characteristics;

[0091] Using the test data set to perform verification and generate the temperature prediction model;

[0092] The predicted data refers to data with a directional / predictive relationship between the data.

[0093] In a second aspect, an embodiment of the present application provides a multi-point temperature measurement system based on a digital temperature sensor, comprising a sensor layout module, a temperature measurement module, a data analysis module, a measurement feedback module, a model generation module, and an intelligent control module;

[0094] The sensor layout module is used to select temperature sensors and arrange measurement positions of the temperature sensors;

[0095] The temperature measurement module is used to measure the temperature within the layout range of the measurement position and obtain measurement data;

[0096] The data analysis module is used to process and analyze the measurement data and obtain analysis results;

[0097] The measurement feedback module is used to calibrate and temperature compensate the temperature sensor according to the analysis results;

[0098] The model generation module is used to construct an associated data set and a predicted data set and generate a temperature knowledge graph and a temperature prediction model respectively;

[0099] The intelligent control module establishes an intelligent control model based on the temperature knowledge graph and the temperature prediction model, and adjusts and optimizes multi-point temperature measurements according to the intelligent control model.

[0100] The beneficial effects of the present invention are:

[0101] (1) The present application first selects a suitable temperature sensor and defines a measurement area based on the selected temperature sensor, then performs a position layout within the measurement area and defines a number of detection points; then performs temperature measurement within the layout range and obtains measurement data; obtains the final analysis result by processing and analyzing the measurement data, and calibrates the temperature sensor based on the analysis result to ensure that the temperature sensor can be adjusted for each measurement, thereby improving the accuracy of the temperature sensor; and by detecting multiple detection points at the same time and through data acquisition and data processing, it can be ensured that the data of each detection point is at the same time (synchronized), thereby ensuring accuracy while reducing delays and errors caused by asynchronous operations and improving the real-time performance of the temperature sensor; further, the present invention also performs temperature compensation on the temperature sensor based on the analysis result, thereby eliminating the problem factor that the temperature at different positions has a large difference due to the influence of temperature gradient when measuring temperature in a large area, further ensuring the accuracy and stability of the temperature measurement; in addition, the present embodiment will measure the data The data and analysis results are stored to generate a storage database, and some data are selected from the storage database as associated data to construct an associated data set, and then some data are selected from the storage database as predicted data to construct a predicted data set; then, a temperature knowledge graph is constructed based on the associated data set. By storing the temperature data of different measurement locations and different measurement times in the temperature knowledge graph and establishing associations between the data, the trend, law and influencing factors of temperature changes can be better reflected; then, a temperature prediction model is constructed based on the predicted data set, and a temperature change model is established according to historical measurement data and temperature influencing factors. Real-time temperature measurement data is input into the prediction model to generate real-time prediction results, thereby predicting and reflecting temperature changes to a certain extent, and providing more comprehensive and accurate temperature prediction and analysis; finally, an intelligent control model is established based on the temperature knowledge graph and the temperature prediction model. By combining the knowledge graph with the prediction model, the analysis and prediction capabilities of temperature data are improved and a more accurate control strategy is provided, thereby making the measurement of multi-point temperature more accurate and stable.

[0102] (2) The present invention selects and arranges digital temperature sensors before performing multi-point temperature measurement to ensure and improve the accuracy of temperature measurement. After screening and determining the temperature sensors to be used for measurement, the measurement range and measurement area are determined according to the measurement requirements, and the measurement positions within the measurement area are arranged. Through reasonable position layout, the problem of distorted or inaccurate measurement results caused by temperature measurement is avoided, thereby improving the accuracy of temperature measurement.

[0103] (3) The present invention adds timestamps and synchronous clocks during data processing and analysis to avoid data delay and asynchrony problems, thereby improving data response speed, measurement synchronization and accuracy.

[0104] (4) The present invention integrates the temperature data of multiple points by constructing a temperature knowledge graph to conduct comprehensive data analysis, which helps to discover the temperature correlation, trend and change pattern between different locations, and provide more comprehensive information for decision-making; it can also timely discover abnormal temperature changes and issue early warnings and handle them by comparing and analyzing the temperature data of different measurement points, which helps to prevent potential problems from occurring; and by understanding the temperature distribution in different areas, corresponding regulatory measures can be taken to improve energy utilization efficiency and reduce energy costs.

[0105] (5) The present invention can predict future temperature change trends based on the laws of temperature change by constructing a temperature prediction model, thereby reducing the occurrence of emergencies; it can also provide strong data support to help formulate reasonable temperature measurement strategies; and it can also improve the accuracy of temperature measurement by predicting temperature.

[0106] (6) The present invention combines the temperature knowledge graph with the temperature prediction model to generate an intelligent control model. This model combines the advantages of the above two models, can more accurately reflect the temperature changes and generate control strategies based on the changes, ensuring measurement accuracy and anti-interference ability, thereby improving the accuracy, real-time performance and stability of multi-point temperature measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] For better understanding and implementation, the technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0108] Figure 1 A flowchart of a multi-point temperature measurement method based on a digital temperature sensor provided in an embodiment of the present application;

[0109] Figure 2 A flow chart of the steps for selecting a temperature sensor provided in an embodiment of the present application;

[0110] Figure 3 A flowchart of the steps for constructing a temperature knowledge graph provided in an embodiment of the present application;

[0111] Figure 4 Flowchart of the steps for constructing a temperature prediction model provided in the embodiment of this application

[0112] Figure 5 This is a structural diagram of a multi-point temperature measurement system based on a digital temperature sensor provided in an embodiment of the present application. DETAILED DESCRIPTION

[0113] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of methods and systems consistent with certain aspects of the present application, as detailed in the appended claims.

[0114] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a," "an," "the," and "the" used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to any or all possible combinations of one or more of the associated listed items.

[0115] The following describes in detail the specific implementation methods, features and effects of the present invention in conjunction with the accompanying drawings and preferred embodiments.

[0116] Example 1

[0117] In existing technical solutions, digital temperature sensors may affect the accuracy of multi-point temperature measurement due to reasons such as accuracy differences, arrangement position, and insufficient calibration. The real-time performance of multi-point temperature measurement may also be affected due to delays in measurement data and defects in imperfect temperature data processing. In addition, the temperature measurement area may be affected by temperature gradients, which may affect both accuracy and real-time performance. Therefore, to address the multi-point temperature measurement problem of digital temperature sensors, this application provides the following solution:

[0118] See also Figures 1 to 4 The present invention provides a multi-point temperature measurement method based on a digital temperature sensor, comprising the following steps:

[0119] S1, selecting a temperature sensor and arranging the measurement positions of the temperature sensor;

[0120] S2, performing temperature measurement within the layout range of the measurement position;

[0121] S3, acquiring measurement data and performing data processing and data analysis;

[0122] S4, calibrating the temperature sensor according to the analysis result;

[0123] S5, performing temperature compensation on the temperature sensor according to the analysis result;

[0124] S6, storing the measurement data and the analysis results to generate a storage database;

[0125] S7, selecting data from the storage database to construct a related data set;

[0126] S8, constructing a temperature knowledge graph based on the associated data set;

[0127] S9, selecting data from the storage database to construct a prediction data set;

[0128] S10, constructing a temperature prediction model based on the prediction data set;

[0129] S11, establishing an intelligent control model based on the temperature knowledge graph and the temperature prediction model;

[0130] S12: Adjust and optimize multi-point temperature measurement according to the intelligent control model.

[0131] Specifically, since multi-point temperature measurement is to detect the temperature of multiple locations in an area or object at the same time, the layout of the measurement locations and the synchronization of the measurement data are important factors in improving the accuracy of temperature measurement. Therefore, the present application first selects a suitable temperature sensor and delineates the measurement area based on the selected temperature sensor, then performs a position layout within the measurement area and delineates a number of detection points; then, temperature measurement is performed within the layout range and measurement data is acquired; the final analysis result is obtained by processing and analyzing the measurement data, and the temperature sensor is calibrated based on the analysis result to ensure that the temperature sensor can be adjusted for each measurement, thereby improving the accuracy of the temperature sensor. Simultaneous detection of multiple detection points and data collection and data processing can ensure that the data of each detection point is at the same time (synchronous), thereby ensuring accuracy while reducing the delay and error caused by asynchronous operation and improving the real-time performance of the temperature sensor; further, this embodiment will also perform temperature compensation on the temperature sensor based on the analysis results, thereby eliminating the problem of large temperature differences at different locations due to the influence of temperature gradient when measuring temperature in a large area, further ensuring the accuracy and stability of temperature measurement; in addition, this embodiment will measure the data The data and analysis results are stored to generate a storage database, and some data are selected from the storage database as associated data to construct an associated data set, and then some data are selected from the storage database as predicted data to construct a predicted data set; then, a temperature knowledge graph is constructed based on the associated data set. By storing the temperature data of different measurement locations and different measurement times in the temperature knowledge graph and establishing associations between the data, the trend, law and influencing factors of temperature changes can be better reflected; then, a temperature prediction model is constructed based on the predicted data set, and a temperature change model is established according to the historical measurement data and temperature influencing factors. The real-time temperature measurement data is input into the prediction model to generate real-time prediction results, thereby predicting and reflecting the temperature changes to a certain extent, and providing more comprehensive and accurate temperature prediction and analysis; finally, an intelligent control model is established based on the temperature knowledge graph and the temperature prediction model. By combining the knowledge graph with the prediction model, the analysis and prediction capabilities of the temperature data are improved and a more accurate control strategy is provided, thereby making the measurement of multi-point temperature more accurate and stable.

[0132] Since digital temperature sensors from different batches or models may have differences in accuracy and stability, this may lead to temperature inconsistencies in multi-point measurements. In addition, different measurement requirements and factors such as temperature range, accuracy, response time, and installation method may also affect the accuracy of temperature measurement. Therefore, before performing multi-point temperature measurement, this embodiment selects and arranges digital temperature sensors to ensure and improve temperature measurement accuracy.

[0133] In one embodiment provided in the present application, selecting a temperature sensor specifically includes the following steps:

[0134] Obtain parameter information of several digital temperature sensors; the parameter information includes but is not limited to accuracy, type, performance, model and installation method;

[0135] Obtain environmental data around the measurement location; the environmental data includes but is not limited to the temperature, humidity, heat source distribution, power supply voltage, and ventilation conditions of the measurement location;

[0136] Obtain the measurement requirements of the temperature sensor; the measurement requirements mainly include the purpose of the temperature sensor, measurement range requirements, measurement area demarcation, measurement object or area, accuracy requirements, response time, communication interface, etc.; it should be noted that the measurement requirements may be modified and adjusted according to different requirements. They can be requirements entered by the user in real time, requirements set in advance, or measurement requirements provided by other means. This embodiment does not make specific limitations here.

[0137] The parameter information and the environmental data are analyzed, and a temperature sensor is selected based on the measurement requirements as a sensor for measuring the multi-point temperature. It should be noted that this embodiment does not specifically limit the specific method or means used for analyzing the parameter information and environmental data. The purpose of the analysis is to use the parameter information and environmental information as analysis factors, analyze them with the measurement requirements as a reference factor, and select the optimal digital temperature sensor based on these three factors.

[0138] Specifically, this embodiment adopts a multi-factor decision analysis algorithm to select an applicable temperature sensor, which specifically includes the following steps:

[0139] Take measurement requirements as decision-making goals; clarify the goals and objectives of decision-making, and ensure that all decision-making plans and final decision results are related to the decision-making goals;

[0140] Consider parameter information and environmental data as decision factors; these data involve multiple variables and conditions, such as accuracy, performance, temperature, humidity, heat source distribution, accuracy requirements, and usage.

[0141] Convert the decision factors into quantifiable decision indicators to facilitate comparison and evaluation;

[0142] Collecting data associated with the decision indicators and determining weights of the decision indicators;

[0143] Establishing a decision model according to the decision factors and the weights;

[0144] Outputting a plurality of decision plans through the decision model, evaluating the plurality of decision plans, and outputting evaluation results;

[0145] Perform a sensitivity analysis on the decision model and output the analysis results; performing a sensitivity analysis on the decision model can examine the impact of different factors or weight changes on the results, thereby improving the robustness of the decision model and further improving the reliability and accuracy of the final output results.

[0146] The optimal decision solution is selected based on the evaluation results and the analysis results. The evaluation results can reflect the multiple decision solutions that best meet the decision objectives and have the highest evaluation scores. The analysis results can eliminate the influence of different factors or weight changes to ensure the accuracy of the solution selection. Therefore, the solution ultimately determined based on these two results is the optimal decision solution selected in this embodiment. Therefore, the optimal decision solution is the optimal digital temperature sensor selected in this embodiment.

[0147] It should be noted that the temperature sensors referred to in this application only refer to digital temperature sensors, including but not limited to digital temperature sensor chips, thermistors, infrared temperature sensors, digital thermometers, and environmental sensors.

[0148] It is understood that the selection of temperature sensors is based on the various measurement requirements described above. Specifically, after understanding the measurement requirements, the temperature sensor that best meets the requirements and requirements is selected by analyzing the parameter information of several temperature sensors and the surrounding environmental information. This sensor is then used as the measurement sensor for subsequent multi-point temperature measurements. The selection of temperature sensors is fundamental to multi-point temperature measurement. This embodiment selects temperature sensors by analyzing various data points, which offers better applicability and compatibility than conventional random selection of sensors. The sensors selected in this manner have higher precision, ensuring consistent temperature across multi-point measurements and improving the accuracy of subsequent multi-point measurements to a certain extent.

[0149] Since temperature sensors are affected by environmental factors such as heat sources and temperature gradients, after screening and determining the temperature sensors for measurement, it is necessary to determine the measurement range and measurement area based on the measurement requirements, and layout the measurement positions within the measurement area. A reasonable layout can avoid distorted or inaccurate measurement results caused by temperature measurement.

[0150] In one embodiment provided in the present application, the measurement positions of the temperature sensors are arranged, specifically comprising the following steps:

[0151] Determine the temperature measurement area based on the measurement requirements and set the measurement range within the temperature measurement area. The measurement range is set based on the measurement requirements as a reference. Although the above measurement requirements may also have range requirements, they are not an accurate range. Therefore, when setting the specific range, it is necessary to standardize it according to the specific situation. The measurement range will also be continuously adjusted according to the measurement situation, so that the results of each measurement are more accurate.

[0152] Several reference measurement points are selected within the measurement range. The initial selection of reference measurement points is random, while subsequent selections are subject to regular changes and adjustments based on the results of the previous or multiple selections. For example, a number of reference measurement points, such as A, B, C, and D, are randomly selected for the first time. After performing multiple measurements in subsequent steps, it is found that the selected reference measurement points A and B provide high precision and meet the measurement requirements. Therefore, the selection of subsequent reference measurement points is adjusted around A and B. After several selections, the selection of subsequent reference measurement points will show regular changes. It should be noted that because each selection of temperature sensors, measurement ranges, and measurement points is designed to eliminate interfering factors and make the measurement results more accurate, the regular changes observed after several selections also reflect, to a certain extent, changes in environmental factors around the measurement location. For example, changes in heat source distribution and temperature will affect the selection of the measurement range. Conversely, regular changes in measurement points also indicate regular changes in heat source distribution and temperature. It is understandable that although the selection of subsequent measurement points will show regular changes, each selection of measurement points is still a random selection.

[0153] An initial measurement interval is formed based on the reference measurement points. Because the reference measurement points are randomly selected, the range of the initial measurement interval formed by the reference measurement points will be smaller than the set measurement range. The purpose of forming the initial measurement interval is to divide the set measurement range more finely based on the selection of the reference measurement points. By narrowing the measurement range, each temperature measurement is more precise, thereby improving the precision and accuracy of each measurement. It should be noted that the initial measurement interval can be one or more, and this embodiment does not specifically limit the number of initial measurement intervals.

[0154] A first measurement interval is used to perform temperature detection on the initial measurement interval and obtain a reference measurement value; the above-mentioned first measurement interval represents the time interval for measuring temperature, and a "rough measurement" is performed on the initial measurement interval by setting a basic measurement interval. The "rough measurement" means that the first measurement interval is longer and the range of the initial measurement interval is larger. It is a rough measurement within a larger range.

[0155] According to the reference measurement value, the actual measurement point is selected and the final measurement interval is determined; after measuring the initial measurement interval, the reference measurement value is obtained, which represents the temperature value obtained by the current measurement to a certain extent. Because each reference measurement point has a reference measurement value, multiple reference measurement points can reflect the temperature changes in the current measurement area to a certain extent; in addition, the actual measurement point refers to the temperature changes presented by the reference measurement value, excluding some measurement points that have an impact on the multi-point temperature measurement, such as the influence of temperature gradient, inaccurate measurement values ​​during the measurement process, the influence of humidity and ventilation conditions, and other environmental factors. Influence; This application selects actual measurement points (points that will not have a significant impact on multi-point temperature measurement) by excluding measurement points with some influencing factors. This approach largely ensures the accuracy of subsequent multi-point measurements and improves the accuracy of its measurements; then the above-mentioned actual measurement points are used as the measurement points required for the "fine measurement" based on the above-mentioned "rough measurement", and "fine measurement" refers to the measurement range that is further narrowed after the actual measurement points are selected. It is also the measurement range that can improve the measurement accuracy and precision after excluding many influencing factors; it should be noted that the above-mentioned ultimate measurement interval is the measurement range determined by the actual measurement points for the above-mentioned "fine measurement".

[0156] A second measurement interval is used to detect the temperature of the ultimate measurement interval and obtain the actual measurement value; the above-mentioned second measurement interval refers to the measurement time interval further adjusted according to the "rough measurement", that is: the time interval is further shortened on the basis of the first measurement interval, and is shortened to the minimum value when necessary. In this way, the temperature value in the shortest time interval can be obtained as much as possible, which is conducive to observing temperature changes.

[0157] It should be noted that the use of the first measurement interval and the second measurement interval is a process of gradually shortening the time interval. Its function is to observe the temperature changes within the interval by shortening the time interval. Different time intervals can reflect different changes, and the temperature values ​​obtained in shorter intervals are more representative and accurate, which can improve the accuracy of multi-point temperature measurement to a certain extent, and can also ensure the real-time performance of multi-point temperature measurement.

[0158] A multi-point temperature measurement region is formed based on the actual measurement values; after obtaining the actual measurement values, they are used as measurement points for the multi-point measurement, and these points form a region for the multi-point temperature measurement. It should be noted that the process from the initial measurement interval to the final measurement interval and then to the multi-point temperature measurement region is a process of continuously narrowing the range while continuously improving the accuracy. It is also a process of continuously eliminating the influence of temperature gradients, heat sources, and other environmental factors. Therefore, the multi-point temperature measurement region ultimately formed is the measurement region with the fewest influencing factors, which ensures that the values ​​obtained from the final multi-point temperature measurement are sufficiently accurate.

[0159] The multi-point temperature measurement area is analyzed and all measurement locations within the area are obtained. The multi-point temperature measurement area formed by the actual measurement values ​​includes all measurement points used for multi-point measurement. Therefore, by analyzing and comparing the measurement values ​​within this area, the multi-point measurement locations are determined (each measurement value corresponds to a measurement location). In this embodiment, after all measurement locations and the layout range are determined, multi-point temperature measurements are performed within the layout range.

[0160] Specifically, in this embodiment, the measurement requirements are first analyzed to determine a temperature measurement area, and a measurement range is set within this area. The measurement area can be a device, a device component, or other product object or area. Next, several reference measurement points are selected within the aforementioned measurement range to form an initial measurement interval. The initial measurement interval is then temperature-detected using a first measurement interval to obtain a reference measurement value. Actual measurement points are then selected based on the reference measurement value and a final measurement interval is determined. The final measurement interval is then temperature-detected using a second measurement interval to obtain an actual measurement value. A multi-point temperature measurement area is then formed based on the actual measurement value. Finally, the multi-point temperature measurement area is analyzed and all measurement locations within the area are obtained. Through these steps, this embodiment eliminates the influence of temperature gradients, heat source distribution, and other environmental factors on multi-point temperature detection, ensuring the real-time and accuracy of temperature measurement and improving the precision of multi-point measurement.

[0161] In an embodiment provided in this application, step S3 specifically includes the following contents:

[0162] S31, adding a timestamp and a synchronized clock during the multi-point temperature measurement process and obtaining measurement data of the multi-point temperature measurement, specifically including:

[0163] Generate a synchronization signal before temperature measurement, which is used to control the temperature sensor to perform synchronous acquisition;

[0164] During the temperature measurement process, the measurement time is recorded by time stamp and synchronized clock;

[0165] After the temperature measurement is completed, the position information of each measuring point is recorded;

[0166] The time information of the temperature measurement is associated with the position information and a time-position sequence table is generated; the time-position sequence table can reflect the information obtained at which position the temperature measurement is performed each time (at each time point).

[0167] It should be noted that since digital temperature sensors need to implement multi-point measurement during data acquisition, and multi-point measurement may cause measurement asynchrony due to the locations of multiple measurement points, when performing multi-point measurement in this embodiment, synchronization of multi-point measurement is ensured by generating a synchronization signal before measurement and adding a timestamp and synchronization clock during measurement.

[0168] S32, processing and analyzing the measurement data, specifically including:

[0169] Preprocessing the measurement data to remove error data and incomplete data;

[0170] performing smoothing and standardization processing on the measurement data;

[0171] Performing outlier detection on the preprocessed measurement data and generating a first analysis result based on the outlier detection;

[0172] Regarding outlier detection, it is specifically expressed as: ;

[0173] in, is the expected normal value; represents the standard value; t represents the time point;

[0174] The standard value It can be expressed as: ;

[0175] in, Represents the standard deviation of the global data; and λ controls the size of the standard deviation;

[0176] Note: If the absolute value of the difference between the measured value at time t and the expected normal value is greater than the standard deviation, it means that the measured value at that moment is an outlier.

[0177] It is understandable that the above-mentioned abnormal points can reflect the abnormalities and interferences that occur in the temperature sensor during the measurement process. After obtaining the above-mentioned abnormal points, the temperature sensor can be calibrated according to the abnormal points for the above-mentioned interference, thereby improving the precision and accuracy of the sensor.

[0178] Obtaining reference measurement values ​​and actual measurement values ​​in the above measurement location layout;

[0179] The mutation data appearing therein is obtained according to the changes between the reference measurement value and the actual measurement value; regarding the mutation data, specifically: the reference measurement value is the value obtained in the first "rough measurement", and the actual measurement value is the value obtained in the second "fine measurement". Changes will occur from the reference measurement value to the actual measurement value, including but not limited to: affected by the temperature gradient, the temperature at some locations is too high, and the temperature at some locations is too low; or affected by the heat source distribution, the temperature change is constantly changing with the heat source distribution; therefore, the above mutation data can reflect to a certain extent the influence and changes of the above-mentioned temperature gradient, heat source distribution and other influencing factors on temperature measurement.

[0180] A temperature gradient and a mutation location are determined based on the mutation data, and a second analysis result is generated based on this. The temperature gradient represents the rate at which temperature changes with location in space. The second analysis result is a solution to the temperature gradient, i.e., a temperature compensation solution, obtained by analyzing the temperature gradient and the mutation location.

[0181] It should be noted that the acquisition of the above mutation data is as follows:

[0182] Acquiring continuous temperature data of the reference measurement value and the actual measurement value;

[0183] Before filtering the continuous temperature data, edge position detection is performed on the continuous temperature data to determine whether the edge position is a temperature mutation location. If so, the temperature mutation location is marked; if not, smoothing filtering is performed on the edge position.

[0184] Obtain mutation data based on the marked temperature mutation location.

[0185] Regarding edge position detection, specifically:

[0186] Perform difference processing on adjacent different temperature points in continuous temperature data;

[0187] When the absolute value of the difference between at least two consecutive adjacent different temperature points is greater than a preset threshold, the area between the at least two consecutive adjacent different temperature points is determined to be a suspected temperature mutation point;

[0188] To determine whether the edge position (suspected temperature mutation location) is a temperature mutation location, specifically:

[0189] Compare the edge detection results with the preset accuracy threshold;

[0190] If the edge detection result is greater than the accuracy threshold, then this edge is where the temperature changes suddenly;

[0191] If the edge detection result is less than or equal to the accuracy threshold, the edge is a noise mutation point.

[0192] It should be noted that this embodiment obtains mutation data by detecting the edge position of continuous temperature data and determining whether the edge position is a temperature mutation point, and performs smoothing filtering on the edge position data based on the judgment result, thereby eliminating noise in the data and improving data accuracy.

[0193] It is understandable that the above mutation data can reflect the temperature change (such as temperature gradient, etc.) to a certain extent, and compensation processing for the above mutation data can improve the precision and accuracy of temperature measurement.

[0194] S33: Generate a first calibration solution according to the first analysis result, and generate a second compensation solution according to the second analysis result.

[0195] Specifically, the first analysis result generated above is a solution calibrated with respect to the measured temperature, and the second analysis result is a solution compensated with respect to the measured temperature.

[0196] In an embodiment provided in the present application, various data generated during the above-mentioned temperature measurement process are stored and a storage database is generated, including real-time data of each measurement, historical data of multiple measurements, various analysis results and processing solutions generated, etc. The storage database can also represent the above-mentioned measurement data in the form of a heat distribution map, thereby better presenting the heat distribution and then reflecting the temperature changes.

[0197] In an embodiment provided in this application, steps S7 to S8 specifically include the following:

[0198] Selecting associated data from the storage database to construct an associated data set; wherein the associated data refers to data with associations between data, including but not limited to temperature values ​​(values ​​obtained by multi-point measurements), timestamps, measurement point locations, measurement sensor information, and environmental factors that affect and are related to temperature changes.

[0199] Performing entity recognition and relationship extraction based on the associated dataset;

[0200] Perform knowledge representation on the extracted entities and relationships and construct a temperature knowledge graph;

[0201] Among them, regarding entity recognition, relationship extraction and knowledge representation, specifically including:

[0202] Entities in the associated data are identified, and the entities are grouped into two to form an entity pair; in this embodiment, the entities include a measurement location, a measurement range, a measurement point, a measurement device, a measurement time, a measurement temperature, and a measurement environment.

[0203] For each entity pair, obtaining a data vector of the associated data;

[0204] Extracting representative features according to the data vector and performing feature screening on the representative features;

[0205] The entity relationship of the entity pair is predicted according to the representation features; for example, the relationship between the measured temperature and the measured time, the relationship between the measured position and the measured temperature, etc.

[0206] The temperature knowledge graph is constructed based on the entity relationships corresponding to the entity pairs.

[0207] It can be understood that this embodiment constructs a temperature knowledge graph, which can integrate the temperature data of multiple points for comprehensive data analysis, helping to discover the temperature correlation, trends and change patterns between different locations, and provide more comprehensive information for decision-making; it can also timely discover abnormal temperature changes and issue early warnings and processing by comparing and analyzing the temperature data of different measurement points, helping to prevent potential problems from occurring; and by understanding the temperature distribution in different areas, corresponding regulatory measures can be taken to improve energy utilization efficiency and reduce energy costs.

[0208] In an embodiment provided in this application, steps S9 to S10 specifically include the following:

[0209] Predicted data is selected from the storage database to construct a predicted data set; wherein, predicted data refers to data with a directional / predictive relationship between data. In multi-point temperature measurement, multiple sets of data will appear, each set of data including multiple temperature values, each corresponding to a measurement point; this data can be used to analyze the temperature distribution at different locations, such as whether the temperature at each point in an area is uniform, or whether there are local temperature differences. Predicted data refers to data that predicts or estimates future events based on existing data or models. In the multi-point temperature measurement of this embodiment, predicted data refers to the temperature conditions at a certain time or location in the future, or the temperature changes at a certain time, predicted based on historical data or environmental data. In this embodiment, predicted data also includes, but is not limited to, temperature values ​​(values ​​obtained from multi-point measurements), timestamps, measurement point locations, information about the measurement sensors, and environmental factors that influence and relate to temperature changes.

[0210] Based on the predicted data set, several sample data groups and test data groups are obtained; wherein, the ratio of the sample data group to the test data group can be 8:2, 9:1, or 7:3. The specific ratio is set according to actual needs and is not specifically limited in this embodiment.

[0211] Using the sample data set to perform feature selection and extraction, including spatiotemporal features and temperature features;

[0212] Performing model training according to the spatiotemporal characteristics and temperature characteristics;

[0213] The test data set is used to perform model validation and generate the temperature prediction model.

[0214] The spatiotemporal features refer to the feature representations of the measurement point positions and measurement time points, while the temperature features refer to the feature representations of the temperature measurement values ​​and temperature changes.

[0215] It can be understood that the temperature prediction model constructed in this embodiment can predict future temperature change trends based on the law of temperature change, thereby reducing the occurrence of emergencies; it can also provide strong data support to help formulate reasonable temperature measurement strategies; it can also improve the accuracy of temperature measurement by predicting temperature.

[0216] In an embodiment provided in the present application, steps S11 to S12 specifically include the following contents: establishing the intelligent control model specifically includes the following steps:

[0217] Obtaining entities, relationships, and attributes in the temperature knowledge graph and using them as graph features;

[0218] Obtaining the spatiotemporal features and temperature features in the temperature prediction model and using them as prediction features;

[0219] Fusing the atlas features with the prediction features to generate regulatory features;

[0220] Perform feature selection and model training based on the control features to generate the intelligent control model;

[0221] Regarding feature fusion: This can be achieved by mapping the information in the temperature knowledge graph to the feature space of the temperature prediction model. For example, the entity attributes in the temperature knowledge graph can be used as input features of the prediction model, or more complex features can be constructed using the relationship information in the temperature knowledge graph.

[0222] In this embodiment, the entity relationship in the graph feature can be integrated with the spatiotemporal feature and temperature feature in the prediction feature. For example, there is a correspondence between the measurement point and the measurement temperature in the entity relationship, and there is also the measurement time and measurement point in the spatiotemporal feature and the measurement temperature in the temperature feature. Therefore, the above-mentioned measurement time, measurement location and measured temperature value can be integrated together to generate a control feature for representing temperature changes; and this control feature is a feature generated by the control strategy based on the temperature change situation.

[0223] It can be understood that this application combines the temperature knowledge graph and the temperature prediction model to generate an intelligent control model. This model combines the advantages of the above two models, can more accurately reflect the temperature changes, and generate control strategies based on the changes; for example: when the temperature knowledge graph and the temperature prediction model find that the temperature values ​​obtained by measuring the temperature at the measurement location at a certain point in time show a large temperature difference, it means that there is a temperature gradient or heat source at this location, which will cause errors in multi-point temperature measurements. Therefore, it is necessary to adjust the measurement strategy in time and change the measurement location, thereby improving the accuracy, real-time and stability of multi-point temperature measurements.

[0224] In summary, the present application reduces the inconsistency and measurement accuracy of temperature measurement by selecting temperature sensors and reasonably arranging measurement positions; reduces data delays and asynchrony during the measurement process through timestamps and synchronized clocks; generates different analysis results through data processing and analysis and calibrates and temperature compensates the sensors separately, reducing the interference caused by temperature gradients, heat source distribution and other environmental factors; reflects and predicts temperature changes by constructing a temperature knowledge graph and a temperature prediction model, and adjusts the measurement strategy in real time through an intelligent control model; in summary, the present embodiment ensures measurement accuracy, response time and anti-interference ability from many aspects, thereby improving the accuracy, real-time nature and stability of multi-point temperature measurement.

[0225] Example 2

[0226] See also Figure 5 , an embodiment of the present application provides a multi-point temperature measurement system based on a digital temperature sensor, including a sensor layout module, a temperature measurement module, a data analysis module, a measurement feedback module, a model generation module and an intelligent control module;

[0227] The sensor layout module is used to select temperature sensors and arrange measurement positions of the temperature sensors;

[0228] The temperature measurement module is used to measure the temperature within the layout range of the measurement position and obtain measurement data;

[0229] The data analysis module is used to process and analyze the measurement data and obtain analysis results;

[0230] The measurement feedback module is used to calibrate and temperature compensate the temperature sensor according to the analysis results;

[0231] The model generation module is used to construct an associated data set and a predicted data set and generate a temperature knowledge graph and a temperature prediction model respectively;

[0232] The intelligent control module establishes an intelligent control model based on the temperature knowledge graph and the temperature prediction model and adjusts and optimizes multi-point temperature measurements according to the intelligent control model.

[0233] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0234] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0235] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0236] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A multi-point temperature measurement method based on a digital temperature sensor, characterized in that: The steps include: Selecting temperature sensors and arranging measurement positions of the temperature sensors; Performing temperature measurement within the layout of the measurement locations; Acquire measurement data and perform data processing and data analysis, including: S31, adding a timestamp and a synchronized clock during the multi-point temperature measurement process and obtaining measurement data of the multi-point temperature measurement; S32, performing data processing and analysis on the measurement data to generate a first analysis result and a second analysis result, specifically including: Performing outlier detection on the preprocessed measurement data and generating a first analysis result based on the outlier detection; determining a temperature gradient and a mutation position according to the mutation data, and generating a second analysis result therefrom; S33, generating a first calibration scheme according to the first analysis result, and generating a second compensation scheme according to the second analysis result; storing the measurement data, the first analysis result, and the second analysis result to generate a storage database; Selecting data from the storage database to construct a related data set; Constructing a temperature knowledge graph based on the associated data set specifically includes: Selecting related data from the storage database to construct a related data set; wherein the related data represents data having a relationship between the data; Selecting prediction data from the storage database to construct a prediction data set; the prediction data represents data with a directional / predictive relationship between the data; Building a temperature prediction model based on the prediction data set specifically includes: Acquire several sample data groups and test data groups based on the prediction data set; Using the sample data set to perform feature selection and extraction, including spatiotemporal features and temperature features; Performing model training according to the spatiotemporal characteristics and temperature characteristics; Using the test data set to perform verification and generate the temperature prediction model; The spatiotemporal features refer to the characteristic representation of the measurement point position and the measurement time point; the temperature features refer to the characteristic representation of the temperature measurement value and the temperature change; Establishing an intelligent control model based on the temperature knowledge graph and the temperature prediction model; Adjusting and optimizing multi-point temperature measurements according to the intelligent control model; The establishment of the intelligent control model specifically includes the following steps: Obtaining entities, relationships, and attributes in the temperature knowledge graph and using them as graph features; Obtaining the spatiotemporal features and temperature features in the temperature prediction model and using them as prediction features; Fusing the atlas features with the prediction features to generate regulatory features; Perform feature selection and model training based on the control features to generate the intelligent control model; Regarding feature fusion: it is achieved by mapping the information in the temperature knowledge graph to the feature space of the temperature prediction model, specifically: using the entity attributes in the temperature knowledge graph as the input features of the temperature prediction model, or using the relationship information in the temperature knowledge graph to construct more complex features.

2. The multi-point temperature measurement method based on a digital temperature sensor according to claim 1, characterized in that: Selecting a temperature sensor includes the following steps: Get parameter information of several digital temperature sensors; Acquire environmental data around the measurement location; Get the measurement requirements of the temperature sensor; The parameter information and the environmental data are analyzed, and a temperature sensor is selected based on the measurement requirements.

3. The multi-point temperature measurement method based on a digital temperature sensor according to claim 2, characterized in that: A multi-factor decision analysis algorithm is used to select the appropriate temperature sensor, which includes the following steps: Use measurement needs as decision-making objectives; Incorporating parameter information and environmental data into decision-making factors; Converting the decision factors into quantifiable decision indicators; Collecting data associated with the decision indicators and determining weights of the decision indicators; Establishing a decision model according to the decision factors and the weights; Outputting a plurality of decision plans through the decision model, evaluating the plurality of decision plans, and outputting evaluation results; Performing sensitivity analysis on the decision model and outputting analysis results; An optimal decision-making solution is selected according to the evaluation results and the analysis results.

4. The multi-point temperature measurement method based on a digital temperature sensor according to claim 1, characterized in that: The measurement positions of the temperature sensors are arranged, specifically comprising the following steps: Determine a temperature measurement area according to measurement requirements and set a measurement range within the temperature measurement area; Selecting several reference measurement points within the measurement range; forming an initial measurement interval according to the reference measurement point; Performing temperature detection on the initial measurement interval using a first measurement interval and obtaining a reference measurement value; Selecting actual measurement points and determining the ultimate measurement interval based on the reference measurement values; Using a second measurement interval to perform temperature detection on the final measurement interval and obtain an actual measurement value; forming a multi-point temperature measurement area according to the actual measurement values; The multi-point temperature measurement area is analyzed and all measurement positions within the area are obtained.

5. The multi-point temperature measurement method based on a digital temperature sensor according to claim 1, characterized in that: said selecting data from said storage database to construct an associated data set; Building a temperature knowledge graph based on the associated data set also includes: Performing entity recognition and relationship extraction based on the associated dataset; Perform knowledge representation on the extracted entities and relationships and construct a temperature knowledge graph; Identifying entities in the associated data, and grouping the entities in pairs to form entity pairs; For each entity pair, obtaining a data vector of the associated data; Extracting representative features according to the data vector and performing feature screening on the representative features; Predicting the entity relationship of the entity pair based on the representation features; The temperature knowledge graph is constructed based on the entity relationships corresponding to the entity pairs.

6. A multi-point temperature measurement system based on a digital temperature sensor, applied to the multi-point temperature measurement method based on a digital temperature sensor according to any one of claims 1 to 5, characterized in that: It includes sensor layout module, temperature measurement module, data analysis module, model generation module and intelligent control module; The sensor layout module is used to select temperature sensors and arrange measurement positions of the temperature sensors; The temperature measurement module is used to measure the temperature within the layout range of the measurement position and obtain measurement data; The data analysis module is used to process and analyze the measurement data and obtain analysis results, including: S31, adding a timestamp and a synchronized clock during the multi-point temperature measurement process and obtaining measurement data of the multi-point temperature measurement; S32, performing data processing and analysis on the measurement data to generate a first analysis result and a second analysis result, specifically including: Performing outlier detection on the preprocessed measurement data and generating a first analysis result based on the outlier detection; determining a temperature gradient and a mutation position according to the mutation data, and generating a second analysis result therefrom; S33, generating a first calibration scheme according to the first analysis result, and generating a second compensation scheme according to the second analysis result; storing the measurement data, the first analysis result, and the second analysis result to generate a storage database; The model generation module is used to construct an associated dataset and a predicted dataset and generate a temperature knowledge graph and a temperature prediction model respectively, specifically including: Selecting related data from the storage database to construct a related data set; wherein the related data represents data having a relationship between the data; Selecting prediction data from the storage database to construct a prediction data set; the prediction data represents data with a directional / predictive relationship between the data; Building a temperature prediction model based on the prediction data set specifically includes: Acquire several sample data groups and test data groups based on the prediction data set; Using the sample data set to perform feature selection and extraction, including spatiotemporal features and temperature features; Performing model training according to the spatiotemporal characteristics and temperature characteristics; Using the test data set to perform verification and generate the temperature prediction model; The spatiotemporal features refer to the characteristic representation of the measurement point position and the measurement time point; the temperature features refer to the characteristic representation of the temperature measurement value and the temperature change; The intelligent control module establishes an intelligent control model based on the temperature knowledge graph and the temperature prediction model, and adjusts and optimizes multi-point temperature measurements according to the intelligent control model; The establishment of the intelligent control model specifically includes the following steps: Obtaining entities, relationships, and attributes in the temperature knowledge graph and using them as graph features; Obtaining the spatiotemporal features and temperature features in the temperature prediction model and using them as prediction features; Fusing the atlas features with the prediction features to generate regulatory features; Perform feature selection and model training based on the control features to generate the intelligent control model; Regarding feature fusion: it is achieved by mapping the information in the temperature knowledge graph to the feature space of the temperature prediction model, specifically: using the entity attributes in the temperature knowledge graph as the input features of the temperature prediction model, or using the relationship information in the temperature knowledge graph to construct more complex features.

Images