Intelligent control system and method for preheating time of sensor

Abstract

The invention belongs to the technical field of sensors, and particularly relates to a sensor preheating time intelligent control system and method.The method comprises the steps that firstly, real-time preheating data and historical preheating data of a tetrahydrothiophene sensor are obtained, and the real-time preheating data are filled and continuously updated through a time window; extracting historical preheating associated features and effective preheating associated features, and marking according to a timeline; comparing, outputting historical preheating data meeting a comparison threshold value in a time window section, extracting a predicted sensor power-off time value after the time window section in the historical preheating data, and finally calculating the sensor preheating time value according to the predicted sensor power-off time value. And then starting the tetrahydrothiophene sensor to execute preheating operation. The problem that the reliability and the measurement accuracy of the whole sensor system are reduced due to the fact that the tetrahydrothiophene sensor is powered on again after power failure in the prior art can be solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensors, and in particular relates to an intelligent control system and method for sensor preheating time. Background Art

[0002] In today's industrial and environmental monitoring fields, accurate detection of various gases is crucial. Among them, the WGD(A) wireless methane detector (hereinafter referred to as the sensor) plays a key role as an intrinsically safe detection instrument specially applied to methane gas environments.

[0003] The sensor can not only accurately detect the concentration of methane gas, but also has a wealth of additional functions. It uses NB-LOT wireless communication to upload the detection data to the platform according to the adjustable cycle, so as to facilitate the centralized management and analysis of data. At the same time, it also integrates the tetrahydrothiophene concentration detection function, which can monitor the tetrahydrothiophene content in a specific environment; the environmental temperature and humidity detection function can provide environmental status information; the water immersion alarm and tilt alarm functions can effectively prevent the equipment from being damaged due to external abnormal conditions; the GPS positioning function helps to determine the location of the equipment, which is convenient for tracking and management; the Bluetooth communication function further expands the convenience of its data interaction. In addition, the product also has significant advantages such as high protection level, stable operation, strong environmental adaptability, low power consumption and long service life. These characteristics enable it to be widely used in many complex environments.

[0004] However, under existing technical conditions, the sensor faces some technical challenges. Since it is powered by lithium thionyl chloride batteries, in order to extend the service life of the product, each sensor unit is usually powered off after completing the detection task. However, the tetrahydrothiophene sensor is an electrochemical sensor. When the power is turned off for a certain period of time and the gas concentration is detected again, the electrodes on the tetrahydrothiophene sensor will accumulate charge. This charge accumulation phenomenon will have a serious negative impact on the measurement accuracy of the sensor, causing deviations in the measurement data and failing to accurately reflect the actual tetrahydrothiophene gas concentration. In actual application scenarios, there are often strict requirements for the precise measurement of tetrahydrothiophene gas concentration. For example, in some situations involving chemical production safety monitoring or specific environmental quality assessments, inaccurate measurement data may cause misjudgments, leading to a series of safety hazards or decision-making errors.

[0005] In summary, the current sensor technology is in conflict with the power supply cycle and the measurement accuracy of the tetrahydrothiophene sensor. An innovative technical solution is urgently needed to effectively solve the problem of inaccurate measurement of the tetrahydrothiophene sensor after power failure and power-on without affecting the overall power consumption control and usage cycle, thereby improving the reliability and practicality of the entire sensor system to meet the increasingly complex and stringent gas detection needs. Summary of the invention

[0006] The technical problem solved by the present invention is to provide an intelligent control system and method for sensor preheating time, so as to solve the problem in the prior art that the reliability and measurement accuracy of the entire sensor system are reduced due to power-on after power failure in the tetrahydrothiophene sensor in the wireless methane detector.

[0007] The basic solution provided by the present invention is a sensor preheating time intelligent control system, including a data acquisition module, a time window module, a preheating time prediction module, and a preheating processing module, wherein:

[0008] The data acquisition module is used to acquire real-time preheating data and historical preheating data of the tetrahydrothiophene sensor;

[0009] A time window is constructed in the time window module, and the time window module is used to fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real time;

[0010] The associated feature extraction module is used to extract historical preheating associated features in the historical preheating data according to the timeline, and mark the historical preheating associated features according to the timeline followed;

[0011] The correlation feature extraction module is also used to extract effective preheating correlation features from the real-time preheating data within the time window, and mark the effective preheating correlation features according to the time window timeline;

[0012] The preheating time prediction module is used to compare the marked effective preheating associated features and the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data;

[0013] A sensor preheating model is preset in the preheating processing module, and the preheating processing module is used to input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation according to the predicted sensor power-off time value and the sensor preheating time value.

[0014] Further, the data acquisition module includes an acquisition interface unit, a storage unit and a preprocessing unit, wherein the acquisition interface unit is used to establish a data acquisition channel with tetrahydrothiophene to acquire real-time preheating data of the tetrahydrothiophene sensor;

[0015] The storage unit is used to store the data of the preheating process of the tetrahydrothiophene sensor according to a preset data format each time the preheating operation is completed, and to establish a data index mechanism during the storage process to generate historical preheating data;

[0016] The preprocessing unit is used to perform filtering and standardization processing on the real-time preheating data.

[0017] Furthermore, the time window module includes a time window construction unit, a filling unit and an update unit. The time window construction unit is used to form a time window based on a preset time length as an upper boundary and a lower boundary. The filling unit is used to determine whether the data in the time window is full. If not, the preprocessed real-time preheating data is added to the time window in chronological order. If so, the front-end data of the time window is removed according to a first-in-first-out mechanism, and the new preprocessed real-time preheating data is added to the end of the time window; the update unit is used to enable the filling unit to fill and remove the data in the time window at a preset time interval.

[0018] Furthermore, the associated feature extraction module includes a feature screening and identification unit, a feature vector construction unit, and a feature marking unit, wherein:

[0019] The feature screening and identification unit is used to screen out other sensor data and corresponding timestamp information associated with the preheating of the tetrahydrothiophene sensor from the stored historical preheating data according to the association algorithm, and sort the screened data in the order of timestamps to generate historical preheating associated features;

[0020] The feature screening and identification unit is also used to continuously monitor the real-time preheating data within the time window, and call the association algorithm to screen out other sensor data associated with the preheating of the tetrahydrothiophene sensor from the real-time preheating data, and sort and organize the screened data according to the time sequence in the time window to generate effective preheating related features;

[0021] The feature vector construction unit is used to form a historical preheating associated feature vector and an effective preheating associated feature vector from the historical preheating associated features and the effective preheating associated features generated by the feature screening and identification unit;

[0022] The feature marking unit is used to mark each feature value in the historical preheating associated feature vector with historical time information and feature type information; the feature marking unit is also used to mark each feature value in the effective preheating associated feature vector with real-time time information and feature type information.

[0023] Further, the preheating time prediction module includes a feature comparison unit, a data extraction unit and a time prediction unit, wherein:

[0024] The feature comparison unit is used to compare the marked valid preheating-related feature vectors corresponding to the time window with the marked historical preheating-related feature vectors one by one according to the feature type and time tag information, and determine whether the difference value of the comparison result within the time window exceeds a preset difference threshold, and retain the historical preheating-related feature vectors that exceed the preset difference threshold;

[0025] The data extraction unit is used to extract corresponding historical preheating data from the storage unit according to the retained historical preheating associated feature vector to generate a historical preheating data subset;

[0026] The time prediction unit is used to obtain the preheating time data after the time window period corresponding to each historical preheating data subset according to the historical preheating data subset, as the predicted sensor power-off time value.

[0027] Furthermore, the preheating processing module inputs the predicted sensor power-off time value into the sensor preheating model, and the output sensor preheating time value is specifically:

[0028] The sensor preheating model is constructed according to the linear fitting algorithm. The sensor preheating model is called to receive the predicted sensor power-off time value of the preheating time prediction module, and the sensor preheating time value is generated by linear fitting. The expression is:

[0029] T standard =0.0018T predicted 4 -0.1128T predicted 3 +2.0732T predicted 2 -1.4531T predicted +59.746

[0030] Among them, T standard Indicates the sensor warm-up time value, T predicted Indicates the predicted sensor power-off time value.

[0031] A sensor preheating time intelligent control method, applied to the above-mentioned sensor preheating time intelligent control system, comprises:

[0032] S1: Obtain real-time preheating data and historical preheating data of the tetrahydrothiophene sensor;

[0033] S2: Build a time window, fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real-time time;

[0034] S3: extract historical preheating related features from historical preheating data, and mark the historical preheating related features according to the timeline followed; at the same time, extract effective preheating related features from the real-time preheating data within the time window, and mark the effective preheating related features according to the timeline of the time window;

[0035] S4: Compare the marked effective preheating associated features and the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data;

[0036] S5: Preset a sensor preheating model, input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation according to the predicted sensor power-off time value and the sensor preheating time value.

[0037] The principle and advantage of the present invention are as follows: In the technical solution of the present application, based on the characteristic that the tetrahydrothiophene sensor will affect the measurement accuracy due to the accumulation of charge on the electrode when it is powered on again after power failure, firstly, the real-time preheating data of the tetrahydrothiophene sensor and the historical preheating data stored in the system are obtained through a special data acquisition and processing mechanism. For the real-time preheating data, dynamic management is performed using a time window, and the data is continuously updated within the window to reflect the current state changes of the sensor.

[0038] Next, key preheating-related features are extracted from the historical preheating data and the real-time preheating data within the time window. These features mainly come from various signals sent by other sensors before the tetrahydrothiophene sensor is started, such as low voltage signals, over-temperature signals, water immersion alarm signals, etc., because these signals are potentially associated with the startup and preheating timing of the tetrahydrothiophene sensor.

[0039] Then, with the help of the preheating time prediction module, the marked effective preheating-related features in the time window are compared with the historical preheating-related features. The historical preheating data with a higher matching degree are screened out by setting a threshold, and the corresponding predicted sensor power-off time value is extracted from it.

[0040] Finally, the preheating processing module calculates the preheating time value of the sensor based on the predicted power-off time value of the sensor, and determines the preheating start timing and related operation adjustments of the tetrahydrothiophene sensor based on this value, thereby realizing intelligent control of the preheating time of the tetrahydrothiophene sensor, improving the measurement accuracy after the sensor is powered on again and optimizing the overall energy consumption management.

[0041] Therefore, the advantages of this solution are:

[0042] 1. Improve measurement accuracy: By accurately predicting the preheating time of the tetrahydrothiophene sensor, it is ensured that the charge accumulated in the electrode can be fully cleared, so that the sensor can more accurately reflect the actual gas concentration when re-detecting the gas concentration, effectively avoiding the measurement error caused by the influence of the electrode charge, and improving the reliability and data quality of the sensor in complex application scenarios, providing a more solid foundation for subsequent decision-making based on sensor data;

[0043] 2. Optimize energy management: Since the preheating time of the tetrahydrothiophene sensor can be intelligently controlled according to actual conditions, unnecessary long preheating or insufficient preheating is avoided. Under the premise of ensuring measurement accuracy, the power consumption due to excessive preheating is reduced, and the service life of the lithium thionyl chloride battery is extended. For equipment that relies on battery power and has requirements for the service life, it greatly improves energy utilization efficiency and reduces equipment maintenance costs and energy replenishment frequency;

[0044] 3. Adaptability and flexibility: This solution is based on a comprehensive analysis of historical data and real-time data, and can adapt to different working environments and sensor status changes. Whether facing different temperature and humidity conditions, abnormal signal combinations of other sensors, or changes in sensor performance as the equipment is used for a longer time, the warm-up time prediction and control strategy can be dynamically adjusted through continuously updated data and feature extraction and comparison. It has strong adaptability and flexibility and can be widely used in a variety of complex and changeable industrial and environmental monitoring scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a functional block diagram of an embodiment of the present invention;

[0046] Figure 2 A fitting curve diagram of the power-off time and preheating time of the sensor according to an embodiment of the present invention;

[0047] Figure 3 The figure is a flowchart of an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following is further described in detail through specific implementation methods:

[0049] In the WGD (A) wireless methane detector, an NB-LOT wireless communication module, an environmental temperature and humidity detection module, a water immersion detection alarm module, a tilt alarm module, a GPS positioning module and a tetrahydrothiophene concentration detection module are integrated, wherein the tetrahydrothiophene concentration detection module adopts a tetrahydrothiophene sensor, which is an electrochemical sensor. For an electrochemical sensor, if it is not used for a period of time, its electrodes may accumulate charges, thereby affecting the accuracy of the sensor; on the other hand, the sensor electrodes may absorb other gases in the environment. Therefore, the preheating process can ensure that the sensor is in a state where the detection meets the specifications and is ready to work at any time; based on this, in order to enable the tetrahydrothiophene sensor to be preheated in time in actual use scenarios and to achieve a working state that does not affect its use at any time, the present application needs to design and plan its power-off time. Therefore, a sensor preheating time intelligent control system is proposed, and its embodiment is basically as shown in the attached Figure 1 As shown: a sensor preheating time intelligent control system, including a data acquisition module, a time window module, a preheating time prediction module and a preheating processing module, wherein the data acquisition module is used to obtain real-time preheating data and historical preheating data of the tetrahydrothiophene sensor. In this embodiment, the data acquisition module includes an acquisition interface unit, a storage unit and a preprocessing unit, wherein the acquisition interface unit is used to establish a data acquisition channel with tetrahydrothiophene to obtain real-time preheating data of the tetrahydrothiophene sensor, for example, various electrical signal parameters of the tetrahydrothiophene sensor related to preheating, including the initial charge state of the sensor electrode, the starting current change information of the internal circuit, etc. These electrical signal parameters constitute the basis of the real-time preheating data; at the same time, the data acquisition frequency is set, for example, data is collected every 10 milliseconds to ensure that more detailed changes in the preheating process can be captured.

[0050] The storage unit is used to store the data of the preheating process according to a preset data format each time the tetrahydrothiophene sensor completes a preheating operation, and establish a data index mechanism during the storage process to generate historical preheating data; wherein the storage unit adopts a solid-state hard disk or a flash memory chip, and after each tetrahydrothiophene completes a preheating operation, the relevant data of this preheating process, including the above-mentioned various types of electrical signal parameters, preheating time, environmental conditions and other information, is stored according to the data format of the storage unit, and during the storage process, a data index mechanism is established so that the required historical preheating data can be quickly retrieved according to keywords such as time, sensor number, data type, etc. For example, an index tree is established in date and time order, and when it is necessary to query the historical preheating data within a specific time period, the corresponding data storage location can be quickly located through the established index.

[0051] The preprocessing unit is used to filter and standardize the real-time preheating data. The filtering process uses a median filtering algorithm to sort the collected data sequence according to size and take the middle value as the valid data, thereby smoothing the data curve and improving the stability and reliability of the data.

[0052] Standardization processing is to perform standard processing on the data after filtering, and convert the data of different physical dimensions into a standard format that is convenient for subsequent analysis and processing.

[0053] A time window is constructed in the time window module, and the time window module is used to fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real-time time; in this embodiment, the time window module includes a time window construction unit, a filling unit and an updating unit, wherein the time window construction unit is used to form a time window according to a preset time length as an upper boundary and a lower boundary; in this embodiment, the preset time length can be 10 minutes, that is, the time window is set according to 10 minutes as a time period, and the time window can trace the real-time preheating data collected within 10 minutes.

[0054] The filling unit is used to determine whether the data in the time window is full. If not, the preprocessed real-time preheating data is added to the time window in chronological order. If so, the front-end data of the time window is removed according to the first-in-first-out mechanism, and the new preprocessed real-time preheating data is added to the end of the time window. Specifically, when the new real-time preheating data is collected and preprocessed, first determine whether the current time window is full. If not, directly add the new real-time preheating data to the end of the data sequence in the time window. If the time window is full, the front-end data of the time window is removed according to the first-in-first-out principle, and then the new real-time preheating data is added to the end. In the whole process, the continuity and integrity of the data are ensured.

[0055] The update unit presets an update time interval, for example, updating the data in the time window once every minute to ensure that the data in the time window is always real-time preheated data within the last 10 minutes.

[0056] The associated feature extraction module is used to extract historical preheating associated features from historical preheating data according to the timeline, and mark the historical preheating associated features according to the timeline followed; at the same time, the associated feature extraction module is also used to extract effective preheating associated features from real-time preheating data within the time window, and mark the effective preheating associated features according to the time window timeline; in this embodiment, the associated feature extraction module includes a feature screening and identification unit, a feature vector construction unit, and a feature marking unit, wherein:

[0057] The feature screening and identification unit is used to screen out other sensor data and corresponding timestamp information associated with the preheating of the tetrahydrothiophene sensor from the stored historical preheating data according to the association algorithm, and sort the screened data in the order of timestamps to generate historical preheating related features; specifically, the historical preheating related features include low voltage signal records, over-temperature signal records, water immersion alarm signal records, etc., among which:

[0058] For low voltage signals, we count the number of times they appear before each preheating, their duration, and the interval between them and the start time of preheating. For example, if we find in historical data that low voltage signals appear three times before a preheating event, each time lasting between 5 and 10 seconds, and the average interval between them and the start time of preheating is 2 minutes, these data constitute some historical preheating-related features of low voltage signals.

[0059] For over-temperature signals, analyze the range of over-temperature, frequency of occurrence, and distribution during the preheating cycle. For example, count the proportion of preheating times with different over-temperature ranges (such as 5-10 degrees Celsius over-temperature, 10-15 degrees Celsius over-temperature, etc.) in historical data, and whether the over-temperature signal appears more frequently in the early, middle, or late stages of preheating.

[0060] Similarly, for the water immersion alarm signal, record the relative relationship between the time point of its occurrence and the preheating time, such as the average delay in starting the preheating operation after the water immersion alarm signal appears, and whether there is a specific pattern between multiple water immersion alarms and preheating.

[0061] The feature screening and identification unit is also used to continuously monitor the real-time preheating data within the time window, and call the association algorithm to screen out other sensor data associated with the preheating of the tetrahydrothiophene sensor from the real-time preheating data, and sort and organize the screened data according to the time sequence in the time window to generate effective preheating related features; in this scheme, the effective preheating related features are the same as the historical preheating related features.

[0062] The feature vector construction unit is used to form a historical preheating associated feature vector and an effective preheating associated feature vector for the historical preheating associated features and the effective preheating associated features generated by the feature screening and identification unit respectively; in this scheme, the features such as low voltage signal, over-temperature signal, and water immersion alarm signal in the historical preheating associated features and the effective preheating associated features are combined into a feature vector. For example, the feature vector can be expressed as [number of low voltage signals, average duration of low voltage signals, interval time between low voltage signals and preheating, number of occurrences of over-temperature range 1, number of occurrences of over-temperature range 2, proportion of over-temperature signals in the early stage of preheating, water immersion alarm and preheating delay time, ...], and the historical preheating associated features and the effective preheating associated features are comprehensively described in the form of vectors.

[0063] The feature marking unit is used to mark each feature value in the historical preheating-related feature vector with historical time information and feature type information; the feature marking unit is also used to mark each feature value in the effective preheating-related feature vector with real-time time information and feature type information. In this solution, each feature value in the historical preheating-related feature vector is marked with its corresponding historical time information. For example, for the feature value of the low voltage signal before a certain preheating in the historical data, the specific time point of the preheating is marked, such as "2023-05-10 10:30:00-Number of low voltage signals: 3", so that the corresponding relationship between the feature and time can be clearly defined in the subsequent comparison analysis.

[0064] For valid preheating-related feature vectors, the start and end time of the time window are marked. For example, "time window [2024-12-20 09:00:00-2024-12-20 09:10:00]-number of low voltage signals: 2", so as to determine the time range in which these features are extracted, and to facilitate matching and comparison with historical preheating data on the timeline.

[0065] The feature type information is marked next to each feature element of the feature vector, and the feature type information to which it belongs is also marked. For example, for the "low voltage signal and preheating interval time" element in the historical preheating associated feature vector, it is marked as "low voltage feature-interval time", so that the type and meaning of the feature can be quickly identified during data processing and analysis, improving the accuracy and efficiency of data processing.

[0066] The preheating time prediction module is used to compare the marked effective preheating associated features corresponding to the time window with the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data; specifically, the preheating time prediction module includes a feature comparison unit, a data extraction unit and a time prediction unit, wherein:

[0067] The feature comparison unit is used to compare the corresponding marked effective preheating-related feature vectors in the time window with the marked historical preheating-related feature vectors one by one according to the feature type and time tag information, and determine whether the difference value of the comparison result within the time window exceeds the preset difference threshold, and retain the historical preheating-related feature vectors that exceed the preset difference threshold; specifically, the feature comparison unit will first compare the elements in the effective preheating-related feature vector and the historical preheating-related feature vector one by one according to the feature type and time tag information. For example, compare the number of occurrences of the low voltage signal in the time window with the number of occurrences in similar situations in the historical preheating data, and the matching degree of the amplitude range of the over-temperature signal and other features. By calculating the similarity or difference value between the features, it is determined which historical preheating data is closest to the effective preheating associated features in the current time window, and the degree of closeness is judged by the set difference threshold. For example, for the feature comparison of the number of occurrences of low voltage signals, an allowable error range is set as the threshold. If the difference between the number of occurrences of low voltage signals in the time window and a record in the historical data is within this threshold range, it is considered that the historical data meets the requirements in this feature. Finally, the effective preheating associated feature vector and the historical preheating associated feature vector in the time window are slid and traversed by the weighted summation method until all historical preheating associated feature sets separated by the time period of the time window are obtained, and all features in the historical preheating associated feature set that match the time window are retained.

[0068] The data extraction unit is used to extract corresponding historical preheating data from the storage unit according to the retained historical preheating associated feature vector to generate a historical preheating data subset;

[0069] The time prediction unit is used to obtain the preheating time data after the time window period corresponding to each historical preheating data subset according to the historical preheating data subset as the predicted sensor power-off time value. For example, in the historical preheating data subset with the highest matching degree, the tetrahydrothiophene sensor starts to work 10 minutes after the time period to which it belongs, and the 10 minutes are used as the predicted sensor power-off time value of the current tetrahydrothiophene sensor.

[0070] A sensor preheating model is preset in the preheating processing module, and the preheating processing module is used to input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation according to the predicted sensor power-off time value and the sensor preheating time value, wherein the predicted sensor power-off time value is input into the sensor preheating model, and the output sensor preheating time value is specifically:

[0071] The sensor preheating model is constructed according to the linear fitting algorithm, and the sensor preheating model is called to receive the predicted sensor power-off time value of the preheating time prediction module, and linear fitting is performed to generate the sensor preheating time value, such as Figure 2 As shown, it is the linear fitting result, where Figure 2 The solid line is the measured line, the dotted line is the fitting curve, and the expression is:

[0072] T standard =0.0018T predicted 4 -0.1128T predicted 3 +2.0732T predicted 2 -1.4531T predicted +59.746

[0073] Among them, T standard Indicates the sensor warm-up time value, T predicted Represents the predicted sensor power-off time value,

[0074] In this embodiment, the predicted sensor power-off time value refers to the interval value between restarting the tetrahydrothiophene sensor, and the sensor preheating time value refers to the preheating time required for the tetrahydrothiophene sensor to start and fully start.

[0075] In order to better express the technical solution of the present invention, the collected real sensor data is verified as shown in Table 1 below:

[0076] Table 1

[0077] Preheating sensor power-off time (min) Sensor warm-up time (s) 0 60 5 92 10 161 15 213 20 241 30 265 35 353

[0078] like Figure 3 As shown, in another embodiment of the present embodiment, a sensor preheating time intelligent control method is also included, which is applied to the above-mentioned sensor preheating time intelligent control system, including:

[0079] S1: Obtain real-time preheating data and historical preheating data of the tetrahydrothiophene sensor;

[0080] S2: Build a time window, fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real-time time;

[0081] S3: extract historical preheating related features from historical preheating data, and mark the historical preheating related features according to the timeline followed; at the same time, extract effective preheating related features from the real-time preheating data within the time window, and mark the effective preheating related features according to the timeline of the time window;

[0082] S4: Compare the marked effective preheating associated features and the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data;

[0083] S5: Preset a sensor preheating model, input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and calculate the sensor sleep value according to the predicted sensor power-off time value and the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation after the sensor sleep value.

[0084] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A sensor preheating time intelligent control system, characterized in that: It includes data acquisition module, time window module, preheating time prediction module and preheating processing module, among which: The data acquisition module is used to acquire real-time preheating data and historical preheating data of the tetrahydrothiophene sensor; A time window is constructed in the time window module, and the time window module is used to fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real time; The associated feature extraction module is used to extract historical preheating associated features in the historical preheating data according to the timeline, and mark the historical preheating associated features according to the timeline followed; The correlation feature extraction module is also used to extract effective preheating correlation features from the real-time preheating data within the time window, and mark the effective preheating correlation features according to the time window timeline; The preheating time prediction module is used to compare the marked effective preheating associated features and the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data; A sensor preheating model is preset in the preheating processing module, and the preheating processing module is used to input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation according to the predicted sensor power-off time value and the sensor preheating time value.

2. According to claim 1, the intelligent control system for sensor preheating time is characterized in that: The data acquisition module includes an acquisition interface unit, a storage unit and a preprocessing unit. The acquisition interface unit is used to establish a data acquisition channel with tetrahydrothiophene to obtain real-time preheating data of the tetrahydrothiophene sensor; The storage unit is used to store the data of the preheating process of the tetrahydrothiophene sensor according to a preset data format each time the preheating operation is completed, and to establish a data index mechanism during the storage process to generate historical preheating data; The preprocessing unit is used to perform filtering and standardization processing on the real-time preheating data.

3. According to claim 2, a sensor preheating time intelligent control system is characterized in that: The time window module includes a time window construction unit, a filling unit and an updating unit. The time window construction unit is used to form a time window according to a preset time length as an upper boundary and a lower boundary. The filling unit is used to determine whether the data in the time window is filled. If not, the preprocessed real-time preheating data is added to the time window in chronological order. If so, the front-end data of the time window is removed according to a first-in-first-out mechanism, and the new preprocessed real-time preheating data is added to the end of the time window; the updating unit is used to enable the filling unit to fill and remove the data in the time window according to a preset time interval.

4. The sensor preheating time intelligent control system according to claim 3 is characterized in that: The associated feature extraction module includes a feature screening and identification unit, a feature vector construction unit, and a feature marking unit, wherein: The feature screening and identification unit is used to screen out other sensor data and corresponding timestamp information associated with the preheating of the tetrahydrothiophene sensor from the stored historical preheating data according to the association algorithm, and sort the screened data in the order of timestamps to generate historical preheating associated features; The feature screening and identification unit is also used to continuously monitor the real-time preheating data within the time window, and call the association algorithm to screen out other sensor data associated with the preheating of the tetrahydrothiophene sensor from the real-time preheating data, and sort and organize the screened data according to the time sequence in the time window to generate effective preheating related features; The feature vector construction unit is used to form a historical preheating associated feature vector and an effective preheating associated feature vector from the historical preheating associated features and the effective preheating associated features generated by the feature screening and identification unit; The feature marking unit is used to mark each feature value in the historical preheating associated feature vector with historical time information and feature type information; the feature marking unit is also used to mark each feature value in the effective preheating associated feature vector with real-time time information and feature type information.

5. The sensor preheating time intelligent control system according to claim 4 is characterized in that: The preheating time prediction module includes a feature comparison unit, a data extraction unit and a time prediction unit, wherein: The feature comparison unit is used to compare the marked valid preheating-related feature vectors corresponding to the time window with the marked historical preheating-related feature vectors one by one according to the feature type and time tag information, and determine whether the difference value of the comparison result within the time window exceeds a preset difference threshold, and retain the historical preheating-related feature vectors that exceed the preset difference threshold; The data extraction unit is used to extract corresponding historical preheating data from the storage unit according to the retained historical preheating associated feature vector to generate a historical preheating data subset; The time prediction unit is used to obtain the preheating time data after the time window period corresponding to each historical preheating data subset according to the historical preheating data subset, as the predicted sensor power-off time value.

6. The sensor preheating time intelligent control system according to claim 5 is characterized in that: The preheating processing module inputs the predicted sensor power-off time value into the sensor preheating model, and the output sensor preheating time value is specifically: The sensor preheating model is constructed according to the linear fitting algorithm. The sensor preheating model is called to receive the predicted sensor power-off time value of the preheating time prediction module, and the sensor preheating time value is generated by linear fitting. The expression is: T standard =0.0018T predicted 4 -0.1128T predicted 3 +2.0732T predicted 2 -1.4531T predicted +59.746 Among them, T standard Indicates the sensor warm-up time value, T predicted Indicates the predicted sensor power-off time value.

7. A sensor preheating time intelligent control method, applied to a sensor preheating time intelligent control system as described in any one of claims 1 to 6 above, characterized in that: include: S1: Obtain real-time preheating data and historical preheating data of the tetrahydrothiophene sensor; S2: Build a time window, fill the real-time preheating data into the time window, and continuously update the preheating data in the time window with the real-time time; S3: extract historical preheating related features from historical preheating data, and mark the historical preheating related features according to the timeline followed; at the same time, extract effective preheating related features from the real-time preheating data within the time window, and mark the effective preheating related features according to the timeline of the time window; S4: Compare the marked effective preheating associated features and the marked historical preheating associated features in the time window, output the historical preheating data that meets the comparison threshold in the time window, and extract the predicted sensor power-off time value after the time window in the historical preheating data; S5: Preset a sensor preheating model, input the predicted sensor power-off time value into the sensor preheating model, output the sensor preheating time value, and start the tetrahydrothiophene sensor to perform a preheating operation according to the predicted sensor power-off time value and the sensor preheating time value.

Images