Host remote control terminal

Abstract

The invention provides a host remote control terminal, and relates to the technical field of remote control, and the terminal comprises a monitoring sequence obtaining module which is used for obtaining a plurality of combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by a combustible gas monitoring host; the integrated recovery generation module is used for obtaining a plurality of generated combustible gas concentration sequence sets and generated humidity sequence sets; the accuracy offset analysis module is used for obtaining accuracy offset parameters; and the host calibration module is used for remotely controlling the combustible gas monitoring host to calibrate according to the calibration period. The technical problems of high difficulty and long period of remote control and accuracy calibration of the combustible gas monitoring host due to the influence of environmental factors and a transmission process can be solved, the monitoring precision and efficiency are improved through remote accuracy offset analysis and calibration of the combustible gas monitoring host, and the reliability of the combustible gas monitoring host is improved. And the complexity and time consumption of field operation are reduced.

Description

Technical Field

[0001] The present application relates to the field of remote control technology, and in particular to a host remote control terminal. Background Art

[0002] With the continuous improvement of industrial automation and intelligence, various monitoring hosts have been widely used in many fields, especially in safety fields such as fire protection, electrical fire and combustible gas monitoring. These hosts are usually used to monitor and control important equipment in real time to ensure the safety of industrial production, construction and other places. However, traditional monitoring and control equipment has some limitations, especially when it is necessary to remotely operate and accurately calibrate monitoring hosts of various brands and models, there are many technical obstacles.

[0003] The accuracy of the monitoring host is easily affected by environmental factors (such as temperature, humidity, etc.) and usage time. Existing calibration methods mostly rely on manual intervention and need to be performed on-site, which is long and inefficient. This not only increases the operation and maintenance costs, but also increases the risks caused by manual operation. Summary of the invention

[0004] The purpose of this application is to provide a host remote control terminal to solve the technical problems of difficulty and long cycle in remote control and accuracy calibration of combustible gas monitoring host due to environmental factors and the influence of transmission process.

[0005] In view of the above problems, the present application provides a host remote control terminal.

[0006] The present application provides a host remote control terminal, which includes: a monitoring sequence acquisition module, which is used to acquire multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by a combustible gas monitoring host; an integrated recovery generation module, which is used to perform integrated recovery generation on the multiple combustible gas concentration sequences and humidity sequences, and obtain multiple generated combustible gas concentration sequence sets and generated humidity sequence sets; an accuracy offset analysis module, which is used to perform accuracy offset analysis of the combustible gas monitoring host based on the multiple generated combustible gas concentration sequence sets and generated humidity sequence sets, and obtain accuracy offset parameters, wherein the accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis; a host calibration module, which is used to configure a calibration period according to the accuracy offset parameters, and remotely control the combustible gas monitoring host to perform calibration according to the calibration period.

[0007] The technical solution provided in this application has at least the following technical effects or advantages: The above-mentioned host remote control terminal first obtains a variety of combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host, and then integrates and restores the multiple combustible gas concentration sequences and humidity sequences to obtain a plurality of generated combustible gas concentration sequence sets and generated humidity sequence sets; then, based on the multiple generated combustible gas concentration sequence sets and generated humidity sequence sets, the combustible gas monitoring host is subjected to accuracy offset analysis to obtain accuracy offset parameters, wherein the accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis; finally, according to the accuracy offset parameters, a calibration period is configured, and the combustible gas monitoring host is remotely controlled to be calibrated according to the calibration period; through the above steps, the technical effect of improving monitoring accuracy and efficiency and reducing the complexity and time consumption of on-site operations through remote accuracy offset analysis and calibration of the combustible gas monitoring host can be achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a schematic diagram of the structure of the host remote control terminal of this application.

[0011] Figure 2 A schematic diagram of the process of obtaining multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host by the host remote control terminal of this application.

[0012] Description of the reference numerals: monitoring sequence acquisition module 11 , integrated restoration generation module 12 , accuracy offset analysis module 13 , host calibration module 14 . DETAILED DESCRIPTION

[0013] This application solves the technical problem of difficulty and long cycle in remote control and accuracy calibration of the combustible gas monitoring host due to the influence of environmental factors and the transmission process by providing a host remote control terminal. It achieves the technical effect of improving monitoring accuracy and efficiency and reducing the complexity and time consumption of on-site operations through remote accuracy offset analysis and calibration of the combustible gas monitoring host.

[0014] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] For examples, please see the attached Figure 1 , the present application provides a host remote control terminal, the terminal specifically includes the following modules: The monitoring sequence acquisition module 11 is used to acquire multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host.

[0016] Specifically, in the monitoring sequence acquisition module 11, a combustible gas monitoring host is used to monitor a variety of combustible gases (such as methane) and humidity in the surrounding environment in real time, and the measured data is transmitted to the control terminal remotely. These data will form a variety of combustible gas concentration sequences and humidity sequences. These sequences include combustible gas concentrations and ambient humidity values ​​at multiple time points. The monitoring host periodically obtains these data according to a preset monitoring frequency and sends them to the remote control terminal in the form of a digital signal to ensure that the control terminal can receive information about environmental changes in real time.

[0017] Further, if Figure 2 As shown, the present application provides a method for obtaining multiple combustible gas concentration sequences and humidity sequences monitored by a combustible gas monitoring host and transmitted remotely, including: According to the preset monitoring and transmission frequency, the concentrations and ambient humidity of multiple combustible gases monitored and remotely transmitted by the combustible gas monitoring host within the most recent multiple timestamps are obtained, wherein the multiple combustible gases include methane, propane and hydrogen; according to the multiple timestamps, the concentrations and ambient humidity of the multiple combustible gases are arranged to obtain multiple combustible gas concentration sequences and humidity sequences.

[0018] In a preferred embodiment, according to actual business needs, the control terminal will pre-set the monitoring transmission frequency, such as every second, every minute or every hour, and the combustible gas monitoring host will collect the concentrations of multiple combustible gases (such as methane, propane, hydrogen) in the environment and the corresponding ambient humidity according to the set monitoring transmission frequency; at each timestamp, the host collects the concentrations of multiple combustible gases and ambient humidity at the current moment to form an original data set, and transmits the monitoring data to the control terminal for further processing through a remote transmission interface (such as RS232, RS485, CAN interface, etc.); after the control terminal receives the data in multiple timestamps, it organizes and arranges the data in timestamp order, and classifies the concentration and humidity of each combustible gas into independent sequences, thereby forming multiple combustible gas concentration sequences and humidity sequences, which can ensure alignment on timestamps for subsequent integrated recovery generation and accuracy offset analysis.

[0019] The integrated restoration generation module 12 is used to perform integrated restoration generation on the multiple combustible gas concentration sequences and humidity sequences to obtain multiple generated combustible gas concentration sequence sets and generated humidity sequence sets.

[0020] Specifically, in the integrated restoration generation module 12, after obtaining a plurality of combustible gas concentration sequences and humidity sequences, these data will have certain deviations due to environmental factors, errors of the combustible gas monitoring host itself, etc. In order to ensure the accuracy of the data, the control terminal will perform integrated restoration on the plurality of combustible gas concentration sequences and humidity sequences. Specifically, the obtained plurality of combustible gas concentration sequences and humidity sequences are input into the pre-built restoration generation channel. Each channel performs restoration for different combustible gas or humidity data. These channels use integrated machine learning methods to process data through multiple restoration paths to simulate and correct the environment. The impact of factors, inherent errors, etc. on the monitoring data. For example, some paths handle slight measurement errors (such as small deviations caused by sensor drift), while other paths handle larger deviations (such as large fluctuations caused by signal noise or environmental changes). These restoration paths repair inaccurate data to make the final generated concentration sequence and humidity sequence more accurate and reliable. After processing, these restoration generation channels output multiple generated combustible gas concentration sequence sets and generated humidity sequence sets. These data can be further used for accuracy offset analysis to ensure the accuracy and consistency of monitoring data, and also provide a basis for subsequent remote control and calibration.

[0021] Furthermore, the present application provides an integrated recovery generation of the plurality of combustible gas concentration sequences and humidity sequences to obtain a plurality of generated combustible gas concentration sequence sets and generated humidity sequence sets, including: Based on integrated machine learning, multiple combustible gas recovery generation channels and humidity recovery generation channels are constructed, wherein each recovery generation channel includes multiple recovery generation paths; the multiple combustible gas concentration sequences and humidity sequences are respectively input into the multiple combustible gas recovery generation channels and humidity recovery generation channels, and the recovery generation obtains multiple generated combustible gas concentration sequence sets and generated humidity sequence sets.

[0022] In a preferred embodiment, by integrating machine learning methods, multiple restoration generation channels are constructed for different combustible gas and humidity data, each channel focuses on processing a certain type of data, such as the concentration or humidity of combustible gas, and each channel contains multiple restoration generation paths, which are responsible for correcting the data according to different data deviations. For example, some paths correct slight errors, while other paths correct larger deviations; then, the obtained multiple combustible gas concentration sequences and humidity sequences are input into these constructed restoration generation channels, and each data sequence is transmitted to the corresponding combustible gas restoration generation channel or humidity restoration generation channel respectively. The multiple restoration generation paths in the channel will process the data in parallel, and make corrections based on the training model to eliminate or reduce errors and fluctuations in the data; in this way, multiple combustible gas concentration sequences and humidity sequences are restored and generated, and finally multiple generated combustible gas concentration sequence sets and generated humidity sequence sets are obtained. These generated sequence sets contain more accurate and reliable monitoring data, providing a more precise basis for subsequent analysis, calibration and control.

[0023] Furthermore, the present application provides a method for constructing multiple combustible gas recovery generation channels and humidity recovery generation channels based on integrated machine learning, including: By adopting integrated learning and generative adversarial network, a first combustible gas recovery generation channel of a first combustible gas is constructed, wherein the first combustible gas recovery generation channel includes a plurality of first combustible gas recovery generation paths; and a plurality of combustible gas recovery generation channels and humidity recovery generation channels corresponding to other multiple combustible gases are further constructed.

[0024] In an optional embodiment, in order to construct a recovery generation channel for the first combustible gas (such as methane), it is necessary to design a framework that can simultaneously capture the influence of environmental factors and measurement error characteristics based on the method of ensemble learning and generative adversarial network. The framework includes multiple recovery generation paths, and each path is corrected according to different degrees of deviation of the data. Specifically, firstly, the monitoring records of the first combustible gas are collected from the historical records to construct the first combustible gas concentration recovery training samples, including the concentration sequence after transmission and the original concentration sequence on the host side; then, based on the training mechanism of the generative adversarial network, the generator and the discriminator are connected to form a recovery generation channel. Adversarial learning corrects the deviation in the transmitted data. The generator is responsible for generating restored data close to the true value, while the discriminator is used to evaluate the quality of the generated data. After independent training, each first combustible gas restoration generation path can focus on correcting data errors of different degrees of deviation; then, multiple independently trained first combustible gas restoration generation paths are combined in parallel to form a complete first combustible gas restoration generation channel. After completing the construction of the first combustible gas restoration generation channel, a similar method is used to continue to build corresponding restoration generation channels for other combustible gases (such as propane, hydrogen) and humidity data. Each channel also contains multiple restoration generation paths, and the paths can be optimized based on specific data characteristics and influencing factors. Finally, all channels are integrated into the control terminal, which can process the corresponding combustible gas concentration and humidity data respectively, providing high-precision restoration generation data support for subsequent accuracy offset analysis and calibration.

[0025] Furthermore, the present application provides a method for constructing a first combustible gas recovery generation channel of a first combustible gas by using integrated learning and generative adversarial networks, including: According to the monitoring remote transmission record of the first combustible gas in the historical time, the sample first combustible gas concentration sequence set after the remote transmission is collected, and the sample first original combustible gas concentration sequence set recorded in the combustible gas monitoring host is collected; the sample first combustible gas concentration sequence set and the sample first original combustible gas concentration sequence set are divided to obtain multiple first combustible gas concentration restoration training samples; the multiple first combustible gas concentration restoration training samples are respectively used to train multiple first combustible gas restoration generation paths based on the generative adversarial network until the training verification converges; the multiple first combustible gas restoration generation paths are combined to obtain a first combustible gas restoration generation channel.

[0026] In an optional embodiment, when constructing the first combustible gas recovery generation channel, firstly, the remote transmission records of the first combustible gas (such as methane) in the historical time are obtained through the combustible gas monitoring host. These records include concentration data transmitted from the host to the control terminal. By sorting these data, a sample first combustible gas concentration sequence set can be constructed. This sample sequence set reflects the concentration changes of the combustible gas at different time points. In addition, the original combustible gas concentration is collected from the local record of the host to construct a sample first original combustible gas concentration sequence set. This sample sequence set is an unprocessed original measurement result. Then, based on the sample first combustible gas concentration sequence and the sample first The deviation of the original combustible gas concentration sequence is determined by dividing the collected sample first combustible gas concentration sequence set and the sample first original combustible gas concentration sequence set to form multiple training sample sets. For example, the average deviation between 1% and 5% is divided into a training sample set, the average deviation between 5% and 15% is divided into a training sample set, and the average deviation above 15% is divided into a training sample set. Each training sample is composed of a pair of remote transmission data and original data, which is designed to help learn how to recover more accurate measurement results from the transmission data. Then, these divided training samples are used to train the Generative Adversarial Network (GAN), which is composed of a generative adversarial network. The generator is responsible for generating the restored combustible gas concentration data based on the input training samples, while the discriminator is used to judge whether the generated concentration data is close to the real data. During the training process, the generator continuously generates the restored combustible gas concentration data based on the input training samples, and gradually optimizes the quality of the generated data through adversarial learning with the discriminator. The task of the discriminator is to evaluate the gap between the generated data and the real data and judge whether it is real enough. The generator and the discriminator are jointly optimized through back propagation. The generator adjusts its parameters to make the generated data closer to the real data, while the discriminator continuously improves its judgment ability to identify the difference between the generated data and the real data. Multiple rounds of adversarial training are performed until the generator can generate restored data that is very similar to the real data and the discriminator's judgment ability reaches the best, and the training process is completed; through this training process, the generator learns how to effectively recover more accurate monitoring data from the original concentration data, and the correction method learned by the generator is reflected in multiple first combustible gas restoration generation paths, each path corresponding to a different deviation range; finally, multiple fully trained restoration generation paths will be combined into a complete restoration generation channel, which can be used to process the concentration data of the first combustible gas. This channel can generate a more accurate concentration sequence based on the input data for subsequent accuracy analysis and calibration.

[0027] The accuracy offset analysis module is used to perform accuracy offset analysis of the combustible gas monitoring host according to the multiple generated combustible gas concentration sequence sets and generated humidity sequence sets to obtain accuracy offset parameters, wherein the accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis.

[0028] Specifically, in the accuracy offset analysis module 13, the fluctuation of gas concentration and the change of ambient humidity will affect the sensor data in the combustible gas monitoring host, thereby affecting the accuracy of the monitoring data. In order to ensure that the monitoring host can still provide accurate data under various environmental conditions, it is necessary to perform accuracy offset analysis to evaluate the specific impact of these fluctuations and humidity changes on the monitoring results. In this process, the control terminal analyzes multiple generated combustible gas concentration sequences and humidity sequences, and calculates the changes of each data sequence within a unit time (such as 30 minutes). The analysis process will evaluate the impact of gas concentration fluctuations and humidity changes on the accuracy of the monitoring host during this period. The accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis. The fluctuation analysis focuses on the impact of gas concentration fluctuations on sensor detection results, which is carried out through mean calculation, deviation calculation, accuracy offset analysis channel calculation and other steps. The humidity analysis focuses on the impact of humidity changes on monitoring accuracy, which is carried out through mean calculation, accuracy offset analysis channel calculation and other steps. Finally, through these analyses, the accuracy offset parameter can be obtained. This parameter represents the change range of monitoring accuracy per unit time, which helps the control terminal understand the error range of the monitoring host under specific environmental changes, and provides a basis for subsequent calibration and adjustment.

[0029] Further, the present application provides a method of performing an accuracy offset analysis of the combustible gas monitoring host according to the multiple generated combustible gas concentration sequence sets and humidity sequence sets to obtain an accuracy offset parameter, including: Randomly select an initial first generated combustible gas concentration from a plurality of first generated combustible gas concentration sequences in the first combustible gas; randomly select a number of first generated combustible gas concentrations from the plurality of first generated combustible gas concentration sequences, calculate the concentration deviation percentage between the mean of the plurality of first generated combustible gas concentrations and the initial first generated combustible gas concentration, and obtain a first initial combustible gas concentration fluctuation parameter; continue to calculate a plurality of first initial combustible gas concentration fluctuation parameters, and calculate the mean to obtain a first combustible gas concentration fluctuation parameter; continue to calculate a plurality of combustible gas concentration fluctuation parameters based on other plurality of generated combustible gas concentration sequence sets; calculate the humidity mean in the humidity sequence set; perform accuracy offset analysis of the combustible gas monitoring host according to the plurality of combustible gas concentration fluctuation parameters and the humidity mean, and obtain a plurality of fluctuation accuracy offset parameters and humidity accuracy offset parameters; calculate the mean of the plurality of fluctuation accuracy offset parameters and the humidity accuracy offset parameters to obtain an accuracy offset parameter.

[0030] In an optional implementation, an initial first generated combustible gas concentration value is randomly selected from multiple generated first combustible gas concentration sequences of the generated combustible gas concentration sequence set as a reference, and this initial first generated combustible gas concentration value will be used as a reference point for deviation analysis; then, a number of first generated combustible gas concentrations are randomly selected from multiple generated first combustible gas concentration sequences, and the mean of these concentration values ​​is calculated. The specific data volume is determined according to actual business needs, and then the deviation between the mean and the initial first generated combustible gas concentration value is calculated, and the calculated deviation is ratioed with the initial first generated combustible gas concentration value to obtain a concentration deviation percentage. The concentration deviation percentage reflects the fluctuation of the concentration value in the sequence. Through this method, a first initial combustible gas concentration fluctuation parameter (i.e., concentration deviation percentage) can be obtained, which represents the degree of fluctuation of the combustible gas concentration in the time period; then, the same calculation is performed on other first generated combustible gas concentration sequences until each combustible gas concentration sequence is calculated, thereby obtaining multiple first initial combustible gas concentration fluctuation parameters, and then the mean of all first initial combustible gas concentration fluctuation parameters is calculated to obtain the first combustible gas concentration fluctuation parameter, which represents the average level of concentration fluctuation of all concentration sequences in this generated combustible gas concentration sequence set; then, The same calculation is performed on other generated combustible gas concentration sequence sets to obtain multiple combustible gas concentration fluctuation parameters, including the second combustible gas concentration fluctuation parameter, the third combustible gas concentration fluctuation parameter, etc. After calculating the combustible gas concentration fluctuation parameter of each generated combustible gas concentration sequence set, the control terminal will also perform mean calculation on the humidity sequence set to obtain the humidity mean in the humidity sequence set, reflecting the overall level of ambient humidity; further, based on the multiple combustible gas concentration fluctuation parameters and humidity mean, the accuracy offset analysis channel constructed in advance is used to perform accuracy offset analysis on the combustible gas monitoring host. The accuracy offset analysis channel will use the internal multiple The fluctuation accuracy offset analysis branch performs fluctuation accuracy offset analysis on multiple combustible gas concentration fluctuation parameters to generate multiple fluctuation accuracy offset parameters, which represent the impact of concentration fluctuation on monitoring accuracy. The internal humidity accuracy offset analysis branch performs humidity accuracy offset analysis on the humidity mean to generate humidity accuracy offset parameters, which reflect the impact of humidity changes on monitoring accuracy. Finally, the mean of all fluctuation accuracy offset parameters and humidity accuracy offset parameters is calculated to obtain the final accuracy offset parameter, which represents the overall deviation of monitoring accuracy caused by concentration fluctuation and humidity change within a certain period of time, and serves as the basis for subsequent calibration and adjustment. Through the above process, the monitoring accuracy deviation caused by concentration fluctuation and humidity change can be comprehensively analyzed and quantified, providing a basis for further optimizing the stability and accuracy of the combustible gas monitoring host.

[0031] Further, the present application provides for respectively performing accuracy offset analysis of the combustible gas monitoring host according to the multiple combustible gas concentration fluctuation parameters and humidity mean values, and obtaining multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters, including: According to the monitoring accuracy test data of the combustible gas monitoring host, multiple sample combustible gas concentration fluctuation parameter sets and sample humidity mean sets are collected, and the monitoring accuracy change amplitude of the combustible gas monitoring host per unit time under different sample combustible gas concentration fluctuation parameters and sample humidity mean values ​​is obtained, and marked as a sample accuracy offset parameter set; the multiple sample combustible gas concentration fluctuation parameter sets and sample humidity mean sets are respectively used as input features, and the sample accuracy offset parameter set is used as output features to train multiple fluctuation accuracy offset analysis branches and humidity accuracy offset analysis branches, and the accuracy offset analysis channels are obtained by combining them; the multiple combustible gas concentration fluctuation parameters and humidity mean values ​​are respectively input into the accuracy offset analysis channel to obtain multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters.

[0032] In an optional implementation, based on the monitoring accuracy test data of the combustible gas monitoring host, multiple sample combustible gas concentration fluctuation parameter sets and sample humidity mean value sets are collected and obtained, wherein each sample combustible gas concentration fluctuation parameter set corresponds to a combustible gas, and these sample data reflect the influence of the combustible gas concentration fluctuation and humidity on the monitoring accuracy under different environmental conditions; then, the monitoring accuracy variation range of the host corresponding to the combustible gas concentration fluctuation parameter and humidity mean value of each sample in unit time is calculated, and these variation ranges are sample accuracy offset parameters, which represent the monitoring accuracy variation range of the combustible gas monitoring host in unit time under specific concentration fluctuation and humidity conditions; Afterwards, these sample combustible gas concentration fluctuation parameter sets and sample humidity mean value sets are used as input features, and the previously obtained sample accuracy offset parameter set is used as output features to train and generate multiple analysis branches. Each branch is trained for the influence of combustible gas concentration fluctuation and humidity, that is, the fluctuation accuracy offset analysis branch and the humidity accuracy offset analysis branch are trained respectively. These branches can effectively analyze and predict the changes in monitoring accuracy under different concentration fluctuation and humidity conditions by learning the relationship between input features and output features. Multiple fluctuation accuracy offset analysis branches and humidity accuracy offset analysis branches are constructed in the same way, that is, they are all constructed through fully connected neural networks, and can also be based on decision The method is constructed by using a tree, random forest, deep neural network, etc. Taking a fully connected neural network as an example, the network is used to construct a fluctuation accuracy offset analysis branch structure, including an input layer, a hidden layer, an output layer, etc., and then the weight of the network is initialized using a standard initialization method (such as Xavier or He initialization), and a set of sample combustible gas concentration fluctuation parameters of a combustible gas is passed to the neural network as input features. In the forward propagation process, the input features are passed layer by layer through the input layer and the hidden layer (using the ReLU activation function), and the predicted fluctuation accuracy offset parameters are calculated through the output layer. The mean square error (MSE) loss function is used to calculate the loss between the predicted fluctuation accuracy offset parameters and the actual value, and The gradient of the loss to each layer weight is calculated by the back propagation algorithm; then, the parameters of the neural network are optimized using the Adam optimizer, and the weights and biases of each layer are adjusted to minimize the value of the loss function. This process is repeated until the set maximum number of training times is reached or the loss stops decreasing; after the training, the performance of the neural network is evaluated using data not used for training to ensure that the neural network can accurately predict the impact of combustible gas concentration fluctuations and humidity changes on monitoring accuracy deviations. If the accuracy reaches the preset accuracy, the current neural network is used as a fluctuation accuracy deviation analysis branch. If the accuracy does not meet the requirements, the learning rate, batch size and other hyperparameters are adjusted to continue optimization to improve the prediction ability;In the same way as above, the remaining sample combustible gas concentration fluctuation parameter sets and sample accuracy offset parameter sets are used to train multiple fluctuation accuracy offset analysis branches, and the sample humidity mean value set and sample accuracy offset parameter set are used to train the humidity accuracy offset analysis branch. These branches are then combined in parallel to construct an accuracy offset analysis channel. Finally, multiple combustible gas concentration fluctuation parameters and humidity mean values ​​are input into the accuracy offset analysis channel. Through the analysis channel, multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters are obtained. These parameters reflect the accuracy offset of the monitoring host under different fluctuation and humidity conditions, providing a basis for subsequent calibration and adjustment. Through this process, the offset of monitoring accuracy under different environmental conditions can be comprehensively evaluated and analyzed, thereby providing support for further improving monitoring accuracy. ;

[0033] The host calibration module is used to configure the calibration cycle according to the accuracy offset parameter, and remotely control the combustible gas monitoring host to calibrate according to the calibration cycle.

[0034] Specifically, in the host calibration module 14, the control terminal will determine an appropriate calibration cycle based on the obtained accuracy offset parameter. The accuracy offset parameter reflects how much the accuracy of the monitoring host has changed within a certain period of time. Therefore, it can be used to determine when to calibrate to maintain the accuracy of monitoring. If the accuracy offset parameter is large, a shorter calibration cycle will be set. Conversely, if the accuracy offset is small, a longer calibration cycle will be set. Finally, the control terminal will remotely control the combustible gas monitoring host and automatically perform calibration operations according to the configured calibration cycle to ensure that its monitoring results continue to be accurate and reliable.

[0035] Further, the present application provides configuring a calibration cycle according to the accuracy offset parameter, and remotely controlling the combustible gas monitoring host to perform calibration according to the calibration cycle, including: According to the accuracy offset parameter, the time when the accuracy offset amplitude reaches a preset accuracy offset threshold is calculated as the offset calibration time; the offset calibration time is used as the calibration period, and the combustible gas monitoring host is remotely controlled to perform calibration according to the calibration period.

[0036] In a preferred embodiment, the calculated accuracy offset parameter, i.e., the variation of monitoring accuracy per unit time (e.g., the variation every 30 minutes), is used as a basic indicator for evaluating the accuracy variation of the monitoring host. According to the preset accuracy offset threshold (e.g., the maximum allowable accuracy offset range), the time required to reach the threshold is calculated by the ratio of the preset accuracy offset threshold to the accuracy offset parameter. This time represents the maximum time interval that can be allowed before the accuracy offset of the monitoring host may exceed the acceptable range. Subsequently, the calculated offset calibration time is used as the calibration cycle to ensure that the monitoring host completes the calibration operation before the deviation exceeds the acceptable range. This can ensure the accuracy of monitoring and avoid the waste of resources caused by too frequent calibration. Afterwards, according to the set calibration cycle, a calibration instruction is sent to the combustible gas monitoring host through remote control, so that it automatically performs the calibration operation at the end of each calibration cycle. Remote control ensures the automation and efficiency of the operation and reduces the need for manual intervention. Through the above process, the calibration cycle can be flexibly adjusted according to the actual accuracy variation range to ensure that the monitoring host is always maintained in a high-precision working state, while optimizing the utilization of calibration resources.

[0037] In summary, the host remote control terminal provided by this application has the following technical effects: This application obtains multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host through a monitoring sequence acquisition module, performs integrated restoration and generation of multiple combustible gas concentration sequences and humidity sequences through an integrated restoration generation module, obtains multiple generated combustible gas concentration sequence sets and generated humidity sequence sets, performs accuracy offset analysis of the combustible gas monitoring host based on multiple generated combustible gas concentration sequence sets and humidity sequence sets through an accuracy offset analysis module, and obtains accuracy offset parameters, wherein the accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis, configures a calibration cycle based on the accuracy offset parameters through a host calibration module, and remotely controls the combustible gas monitoring host to calibrate according to the calibration cycle. These technical effects jointly solve the technical problems of difficulty and long cycle in remote control and accuracy calibration of the combustible gas monitoring host due to environmental factors and the influence of the transmission process, and achieves the technical effect of improving monitoring accuracy and efficiency and reducing the complexity and time consumption of on-site operations through remote accuracy offset analysis and calibration of the combustible gas monitoring host.

[0038] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0039] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. Host remote control terminal, characterized in that: The terminal comprises: A monitoring sequence acquisition module is used to acquire multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host; An integrated restoration generation module, used for performing integrated restoration generation on the multiple combustible gas concentration sequences and humidity sequences to obtain multiple generated combustible gas concentration sequence sets and generated humidity sequence sets; An accuracy offset analysis module, used to perform accuracy offset analysis of the combustible gas monitoring host according to the multiple generated combustible gas concentration sequence sets and generated humidity sequence sets to obtain accuracy offset parameters, wherein the accuracy offset analysis includes fluctuation accuracy offset analysis and humidity accuracy offset analysis; The host calibration module is used to configure the calibration cycle according to the accuracy offset parameter, and remotely control the combustible gas monitoring host to calibrate according to the calibration cycle.

2. The host remote control terminal according to claim 1, characterized in that: Obtain multiple combustible gas concentration sequences and humidity sequences monitored and remotely transmitted by the combustible gas monitoring host, including: According to a preset monitoring transmission frequency, the concentration and ambient humidity of multiple combustible gases monitored and remotely transmitted by the combustible gas monitoring host within the most recent multiple timestamps are obtained, wherein the multiple combustible gases include methane, propane and hydrogen; The concentrations of the various combustible gases and the ambient humidity are arranged according to the multiple time stamps to obtain multiple combustible gas concentration sequences and humidity sequences.

3. The host remote control terminal according to claim 1, characterized in that: The multiple combustible gas concentration sequences and humidity sequences are integrated and restored to obtain multiple generated combustible gas concentration sequence sets and generated humidity sequence sets, including: Based on integrated machine learning, multiple combustible gas recovery generation channels and humidity recovery generation channels are constructed, wherein each recovery generation channel includes multiple recovery generation paths; The plurality of combustible gas concentration sequences and humidity sequences are respectively input into the plurality of combustible gas recovery generation channels and humidity recovery generation channels, and the recovery generation obtains a plurality of generated combustible gas concentration sequence sets and generated humidity sequence sets.

4. The host remote control terminal according to claim 3, characterized in that: Based on integrated machine learning, multiple combustible gas recovery generation channels and humidity recovery generation channels are constructed, including: Using integrated learning and generative adversarial networks, constructing a first combustible gas recovery generation channel of a first combustible gas, wherein the first combustible gas recovery generation channel includes a plurality of first combustible gas recovery generation paths; Continue to construct multiple combustible gas recovery generation channels and humidity recovery generation channels corresponding to other multiple combustible gases.

5. The host remote control terminal according to claim 4, characterized in that: Using integrated learning and generative adversarial networks, a first combustible gas recovery generation channel for the first combustible gas is constructed, including: According to the monitoring remote transmission record of the first combustible gas in the historical time, the first combustible gas concentration sequence set of the sample after the remote transmission is collected, and the first original combustible gas concentration sequence set of the sample recorded in the combustible gas monitoring host is collected; Dividing the sample first combustible gas concentration sequence set and the sample first original combustible gas concentration sequence set to obtain a plurality of first combustible gas concentration restoration training samples; Using the plurality of first combustible gas concentration restoration training samples respectively, and training a plurality of first combustible gas restoration generation paths based on a generative adversarial network until the training verification converges; The plurality of first combustible gas recovery generation paths are combined to obtain a first combustible gas recovery generation channel.

6. The host remote control terminal according to claim 1, characterized in that: According to the multiple generated combustible gas concentration sequence sets and humidity sequence sets, the combustible gas monitoring host is analyzed for accuracy deviation to obtain accuracy deviation parameters, including: Randomly selecting an initial first generated combustible gas concentration from among a plurality of first generated combustible gas concentration sequences in the first combustible gas; Randomly selecting a number of first generated combustible gas concentrations in the plurality of first generated combustible gas concentration sequences, calculating a concentration deviation percentage between an average of the number of first generated combustible gas concentrations and the initial first generated combustible gas concentration, and obtaining a first initial combustible gas concentration fluctuation parameter; Continue to calculate to obtain multiple first initial combustible gas concentration fluctuation parameters, and calculate the average to obtain the first combustible gas concentration fluctuation parameter; Continuously calculating and obtaining multiple combustible gas concentration fluctuation parameters according to multiple other generated combustible gas concentration sequence sets; Calculating the mean humidity in the humidity sequence set; According to the multiple combustible gas concentration fluctuation parameters and humidity mean values, respectively, an accuracy offset analysis of the combustible gas monitoring host is performed to obtain multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters; The average of the plurality of fluctuation accuracy offset parameters and the humidity accuracy offset parameters is calculated to obtain an accuracy offset parameter.

7. The host remote control terminal according to claim 6, characterized in that: According to the multiple combustible gas concentration fluctuation parameters and humidity mean values, accuracy offset analysis of the combustible gas monitoring host is performed respectively to obtain multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters, including: According to the monitoring accuracy test data of the combustible gas monitoring host, multiple sample combustible gas concentration fluctuation parameter sets and sample humidity mean sets are collected, and the monitoring accuracy variation range of the combustible gas monitoring host per unit time under different sample combustible gas concentration fluctuation parameters and sample humidity mean values ​​is obtained, and marked as a sample accuracy offset parameter set; The plurality of sample combustible gas concentration fluctuation parameter sets and sample humidity mean value sets are respectively used as input features, and the sample accuracy offset parameter set is used as output features, and a plurality of fluctuation accuracy offset analysis branches and humidity accuracy offset analysis branches are trained, and an accuracy offset analysis channel is obtained by combining them; The multiple combustible gas concentration fluctuation parameters and humidity mean values ​​are respectively input into the accuracy offset analysis channel to obtain multiple fluctuation accuracy offset parameters and humidity accuracy offset parameters.

8. The host remote control terminal according to claim 1, characterized in that: According to the accuracy offset parameter, a calibration period is configured, and the combustible gas monitoring host is remotely controlled to perform calibration according to the calibration period, including: According to the accuracy offset parameter, the time when the accuracy offset amplitude reaches a preset accuracy offset threshold is calculated as the offset calibration time; The offset calibration time is used as a calibration period, and the combustible gas monitoring host is remotely controlled to perform calibration according to the calibration period.

Images