A signal enhancement method and platform for off-line concentration monitor

Abstract

The present invention discloses a signal enhancement method and platform for an offline concentration monitor, and relates to the field of signal processing technology. The method comprises: arranging a VHP sensor array of an offline concentration monitor in a target area; monitoring VHP concentration through the sensor array, and transmitting the signal array to a processing unit; in the processing unit, generating a distance array according to sensor position information, and performing first-order restoration of the signal array to generate a first restoration signal; indexing humidity information according to the first restoration signal; and performing second-order restoration of the first restoration signal according to the humidity information to generate a second restoration signal as an enhancement processing result. The present invention solves the technical problem of insufficient monitoring accuracy caused by signal attenuation and interference in VHP concentration signal monitoring, and achieves the technical effect of improving the monitoring accuracy of VHP concentration signals through a signal processing method combining first-order restoration and second-order restoration, and improving the processing efficiency of offline signals through humidity information compensation.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to a signal enhancement method and platform for an off-line concentration monitor. Background Art

[0002] With the widespread application of hydrogen peroxide (VHP) in the fields of medicine, pharmaceuticals, and food processing, accurate monitoring of its concentration has become an important technical requirement to ensure the disinfection effect. In the VHP sterilization process, the accuracy of concentration monitoring is directly related to the effectiveness and safety of sterilization. However, there are still many challenges in the existing technology in offline VHP concentration monitoring. On the one hand, due to the long distance between the sensor and the processing unit, the monitoring signal is easily affected by environmental factors such as attenuation, noise interference, and humidity during transmission, resulting in a decrease in signal quality, making it difficult to ensure the accuracy of the monitoring results. On the other hand, the current monitoring equipment lacks systematicness and pertinence in sensor layout, and does not fully consider the spatial characteristics of the target area and the influence of sensor location information on signal distribution, resulting in uneven signal coverage, which further affects the monitoring effect. In addition, as an important variable affecting VHP concentration, the existing monitoring technology has not effectively used humidity information for compensation and correction, which limits the reliability and accuracy of the monitoring results. Summary of the invention

[0003] The present application provides a signal enhancement method and platform for an offline concentration monitor, which is used to solve the technical problem of insufficient monitoring accuracy caused by signal attenuation and interference in VHP concentration signal monitoring.

[0004] In view of the above problems, the present application provides a signal enhancement method and platform for an offline concentration monitor.

[0005] In a first aspect of the present application, a signal enhancement method for an offline concentration monitor is provided, the method comprising: deploying a VHP sensor array of the offline concentration monitor in a target area where VHP concentration monitoring is to be performed, wherein the VHP sensor array includes a sensor position information array; performing VHP concentration monitoring through the VHP sensor array, and transmitting the VHP concentration signal array obtained by monitoring to a processing unit of the offline concentration monitor; within the processing unit, generating a sensor distance array according to the sensor position information array, performing a first-order restoration on the VHP concentration signal array according to the sensor distance array to obtain a first restored VHP concentration signal array; obtaining a humidity information array by indexing according to the first restored VHP concentration signal array; and performing a second-order restoration on the first restored VHP concentration signal array according to the humidity information array to obtain a second restored VHP concentration signal array as a signal enhancement processing result of the offline concentration monitor.

[0006] In a second aspect of the present application, a signal enhancement platform for an offline concentration monitor is provided, the platform comprising: a sensor array deployment module, the sensor array deployment module is used to deploy a VHP sensor array of the offline concentration monitor in a target area to be subjected to VHP concentration monitoring, wherein the VHP sensor array comprises a sensor position information array; a concentration monitoring module, the concentration monitoring module is used to perform VHP concentration monitoring through the VHP sensor array, and transmit the VHP concentration signal array obtained by monitoring to a processing unit of the offline concentration monitor; a first-order restoration generation module, the first-order restoration generation module is used to generate a VHP concentration signal array obtained by monitoring the VHP concentration in the target area to be subjected to VHP concentration monitoring; In the processing unit, a sensor distance array is generated according to the sensor position information array, and the VHP concentration signal array is first-order restored and generated according to the sensor distance array to obtain a first restored VHP concentration signal array; a humidity information acquisition module is used to index and obtain a humidity information array according to the first restored VHP concentration signal array; a second-order restoration generation module is used to perform second-order restoration and generate the first restored VHP concentration signal array according to the humidity information array to obtain a second restored VHP concentration signal array as a signal enhancement processing result of an offline concentration monitor.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] In the target area to be monitored for VHP concentration, a VHP sensor array of an offline concentration monitor is arranged, wherein the VHP sensor array includes a sensor position information array; VHP concentration monitoring is performed through the VHP sensor array, and the VHP concentration signal array obtained by monitoring is transmitted to the processing unit of the offline concentration monitor; in the processing unit, a sensor distance array is generated according to the sensor position information array, and the VHP concentration signal array is first-order restored according to the sensor distance array to obtain a first restored VHP concentration signal array; according to the first restored VHP concentration signal array, a humidity information array is obtained by indexing; according to the humidity information array, the first restored VHP concentration signal array is second-order restored to obtain a second restored VHP concentration signal array as the signal enhancement processing result of the offline concentration monitor. The technical effect of improving the monitoring accuracy of the VHP concentration signal by a signal processing method combining first-order restoration and second-order restoration and improving the processing efficiency of the offline signal by humidity information compensation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic flow chart of a signal enhancement method for an offline concentration monitor provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of the structure of a signal enhancement platform for an offline concentration monitor provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: sensor array layout module 1, concentration monitoring module 2, first-order restoration generation module 3, humidity information acquisition module 4, second-order restoration generation module 5. DETAILED DESCRIPTION

[0013] The present application provides a signal enhancement method and platform for an offline concentration monitor, which is used to solve the technical problem of insufficient monitoring accuracy caused by signal attenuation and interference in VHP concentration signal monitoring.

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0015] Embodiment 1, as Figure 1 As shown, the present application provides a signal enhancement method for an offline concentration monitor, the method comprising:

[0016] In a target area to be monitored for VHP concentration, a VHP sensor array of an offline concentration monitor is arranged, wherein the VHP sensor array includes a sensor position information array.

[0017] The offline concentration monitor is a device used for concentration monitoring in a specific environment. It has efficient signal acquisition and processing capabilities and is usually composed of a host, a sensor array, and auxiliary equipment (such as a Bluetooth printer). During use, the monitor monitors and records the concentration of chemical gases (such as VHP) in the target area through the coordination with the sensor. The sensor is connected to the host through an interface. Different sensor interfaces are usually marked as PORT1, PORT2, etc. to ensure correct connection and correspond to different monitoring functions. After the device is turned on, the user can select the corresponding language, enter the password and log in to the operating system through the host interface, select the sensor type according to the specific monitoring needs, set the verification time and sampling interval (usually 60 seconds or 120 seconds). After the monitoring starts, the sensor will collect the target gas concentration signal in the environment in real time and transmit the data to the host for processing. The host can not only display the current sensor information and concentration values, but also support the instant output of monitoring data through a Bluetooth printer for easy storage and analysis. The device has a flexible operation process and can meet a variety of monitoring needs. By deploying a sensor array in the target area, the offline concentration monitor can shut down all devices after the collection is completed and organize the relevant equipment for reuse. The overall design focuses on the accuracy of data collection and the convenience of operation, and is suitable for concentration monitoring tasks in laboratories, industrial production or disinfection environments.

[0018] In one embodiment, the concentration of VHP (Vaporized Hydrogen Peroxide) refers to the concentration of hydrogen peroxide in a vaporized state in an environment or system, usually expressed in ppm (parts per million) or mg / m³. Hydrogen peroxide is a highly efficient strong oxidant. The vapor-like VHP formed after vaporization is widely used in biomedicine, food processing and other scenarios with strict requirements for microbial control due to its powerful sterilization, disinfection and sterilization capabilities. In order to accurately monitor the VHP concentration in the target area, firstly, the sensor array of the offline concentration monitor is designed and arranged according to the spatial distribution characteristics of the target area. Specifically, multiple high-precision VHP sensors are installed at different locations in the target area according to the pre-set monitoring requirements. The interface of each sensor has a clear identification (such as PORT1, PORT2, etc.), and is connected to the host of the monitor through a dedicated connection line to ensure the accuracy of the array layout and the convenience of operation. In addition, each sensor provides data support for the concentration distribution in the target area by collecting the real-time concentration signal of VHP in the surrounding environment. In addition, to ensure the accuracy of the monitoring data, the location coordinate information of the sensors is recorded and organized into a sensor location information array, which is not only used for spatial correction of the monitoring data, but also provides key parameters for subsequent signal processing steps. Through this layout, comprehensive coverage and accurate monitoring of VHP concentrations in the target area can be achieved, providing a reliable data basis for offline monitoring.

[0019] In a possible implementation, a VHP sensor array of an offline concentration monitor is arranged in a target area to be monitored for VHP concentration, including:

[0020] At multiple monitoring locations within a target area to be monitored for VHP concentration, multiple VHP sensors of an offline concentration monitor are deployed to form a VHP sensor array; a coordinate system of the target area is constructed, the position coordinates of the multiple VHP sensors are obtained, and a sensor position information array of the VHP sensor array is constructed.

[0021] Optionally, the system terminal determines the number and distribution of monitoring locations according to the spatial distribution characteristics of the target area and the requirements of VHP concentration monitoring. Reasonable monitoring locations are selected by analyzing the area, shape and effective monitoring area of ​​the target area and the VHP sensor. Each monitoring location will be used as a node for laying out the VHP sensor. Subsequently, multiple VHP sensors of offline concentration monitors are laid out in the target area to form a VHP sensor array. Each sensor in this VHP sensor array needs to be connected to the monitor host through a dedicated connection line during deployment, and at the same time, the interface identifier (such as PORT1, PORT2, etc.) must be consistent with the port of the host to ensure that the sensor is connected correctly and operates efficiently. After the deployment is completed, a coordinate system of the target area is constructed to accurately record the position of each sensor. The coordinate system of the target area usually takes the host or a fixed point as the origin, and uses two-dimensional coordinates (X, Y) or three-dimensional coordinates (X, Y, Z) to record the position information of the sensor. For each sensor, the corresponding coordinate value is measured and recorded according to its physical installation position. Through a unified coordinate system, the spatial distribution of the sensor in the target area can be accurately described. After that, the position information of all sensors is summarized and organized into a sensor position information array. The array is a matrix structure that contains the unique identification of each sensor and the corresponding coordinate value. By constructing the sensor location information array, not only can the distribution of sensors in the target area be clearly described, but it also provides key support for subsequent data processing steps to ensure the accuracy and consistency of signal acquisition and monitoring results.

[0022] The VHP concentration is monitored by the VHP sensor array, and the VHP concentration signal array obtained by monitoring is transmitted to the processing unit of the offline concentration monitor.

[0023] In one embodiment, through the deployment of the VHP sensor array, the system terminal begins to monitor the VHP concentration in the target area in real time. Each sensor collects the hydrogen peroxide (VHP) concentration value in the surrounding environment according to its location, and these concentration values ​​reflect the gas distribution at different locations in the target area. The sensor converts the collected concentration signal into a digital signal and sends it to the host processing unit of the offline concentration monitor through a connecting line or wireless transmission method (such as Bluetooth, wireless network, etc.). The concentration signal received by the processing unit is stored in an array form to form a VHP concentration signal array. The signal array is organized according to the sensor position information so that each concentration value can be matched one by one with its corresponding spatial position, thereby providing basic data for subsequent signal processing and analysis. Through this monitoring and signal transmission method, the VHP concentration distribution in the target area can be effectively recorded, and the integrity and accuracy of the data can be ensured, which lays the foundation for subsequent concentration analysis, restoration processing and optimization of monitoring results.

[0024] In the processing unit, a sensor distance array is generated according to the sensor position information array, and the VHP concentration signal array is first-order restored according to the sensor distance array to obtain a first restored VHP concentration signal array.

[0025] In one embodiment, in the processing unit, in order to correct the signal attenuation problem caused by the transmission distance between the sensor and the processing unit, it is necessary to generate a sensor distance array based on the sensor position information array, and perform first-order restoration on the VHP concentration signal array based on this. Specifically, the system terminal first uses the sensor coordinates recorded in the sensor position information array and the position coordinates of the processing unit to calculate the transmission distance between each sensor and the processing unit using the Euclidean distance formula. By calculating the distance value of each sensor, a sensor distance array is generated to represent the distance distribution between the sensor and the processing unit. Subsequently, the VHP concentration signal is first-order restored and generated based on the sensor distance array in combination with the ensemble learning method. The greater the distance between the sensor and the processing unit, the higher the attenuation of the signal, so more restoration generation branches are needed to correct the signal deviation. The system terminal allocates the number of restoration generation branches in proportion according to the distance value in the sensor distance array. The greater the distance, the more restoration branches there are. Afterwards, the VHP concentration signals collected by each sensor are respectively input into the corresponding number of restoration generation branches, and the signals are restored through ensemble learning. Each branch corrects the attenuation and noise of the signal by counteracting the generation path. For the signal restoration results of each sensor, the corresponding restoration branch outputs are summarized as the final restored concentration signal value to reduce the impact of single branch errors on the final signal and improve the stability and accuracy of the restoration results. Finally, the first restored VHP concentration signal array generated by the processing unit not only corrects the signal attenuation problem caused by the transmission distance, but also significantly improves the accuracy and reliability of the monitoring signal through the multi-branch restoration mechanism of integrated learning, laying the foundation for subsequent in-depth processing.

[0026] In a possible implementation, in the processing unit, a sensor distance array is generated according to the sensor position information array, and a first-order restoration is performed on the VHP concentration signal array according to the sensor distance array, including:

[0027] According to the position coordinates of the processing unit and in combination with the sensor position information array, the multiple sensor distances between the multiple VHP sensors in the VHP sensor array and the processing unit are calculated to obtain a sensor distance array; according to the multiple sensor distances in the sensor distance array, multiple restoration generation numbers for first-order restoration generation of the multiple VHP concentration signals in the VHP concentration signal array are respectively set; the multiple VHP concentration signals are respectively input into the multiple restoration generation numbers of VHP concentration restoration branches in the first-order restorer for integrated restoration generation to obtain multiple first restored VHP concentration signal sets, wherein the first-order restorer includes multiple trained VHP concentration restoration branches; the multiple first restored VHP concentration signal sets are averaged to obtain multiple first restored VHP concentration signals, and a first restored VHP concentration signal array is constructed.

[0028] Optionally, taking three-dimensional coordinates as an example, the system terminal uses the Euclidean distance formula to calculate the distance between each sensor and the processing unit one by one according to the coordinates of each sensor in the sensor position information array and the position coordinates of the processing unit, and stores the distance values ​​of all sensors in the array in index order to generate a sensor distance array. Subsequently, according to the distance value in the sensor distance array, the corresponding number of restoration generation is set for each VHP concentration signal. Sensors with longer distances have more severe signal attenuation, so more restoration generation branches are allocated, while sensors with shorter distances have fewer branches allocated. This allocation method can ensure the reasonable distribution of restoration resources. After that, each VHP concentration signal in the VHP concentration signal array is sequentially input into multiple restoration generation branches in the first-order restorer according to the branch number requirement. Each branch has been trained with historical data, and can effectively correct the error caused by distance attenuation and noise of the signal, and output a more accurate restoration concentration signal. After completing the integrated restoration generation, the system terminal obtains multiple first restoration VHP concentration signal sets, and each first restoration VHP concentration signal set corresponds to the generation result of multiple restoration generation branches of a VHP concentration signal. Then, the mean value of the first restored VHP concentration signal set corresponding to each VHP concentration signal is calculated, and the calculation result is used as the final restored signal of the sensor. The mean processing can reduce the influence of single branch error and ensure the stability and accuracy of the signal. Finally, the final restored signals of all sensors are sorted into the first restored VHP concentration signal array in sequence. This signal array integrates all sensor data that have undergone first-order restoration processing, more accurately reflects the VHP concentration distribution in the target area, and lays a high-quality data foundation for subsequent processing steps.

[0029] In a possible implementation, according to the multiple sensor distances in the sensor distance array, multiple numbers of restoration generation for performing first-order restoration generation on multiple VHP concentration signals in the VHP concentration signal array are respectively set, including:

[0030] The total number of branches of the multiple VHP concentration restoration branches in the first-order restorer is obtained; the ratio of each sensor distance to the sum of the multiple sensor distances is calculated, multiplied by the total number of branches and rounded to obtain multiple restoration generation numbers.

[0031] Optionally, the system terminal first determines the total number of restoration generation branches available in the first-order restorer. This number is predetermined or determined based on the processing capacity of the first-order restorer, and represents the maximum number of restoration branches that the first-order restorer can process simultaneously. For each sensor, the ratio of the distance between the sensor and the processing unit is calculated to the sum of the distances of all sensors. This ratio is used to represent the relative importance of the sensor signal in the overall attenuation. Subsequently, the ratio of each sensor is multiplied by the total number of branches of the first-order restorer to obtain the number of restoration generation branches required for the sensor. In order to ensure that the number of branches is an integer value, the result needs to be rounded up. For example, if the ratio calculation result of a sensor is 3.2, its restoration generation number will be rounded to 4. For each sensor, the calculated restoration generation number is recorded to form a restoration generation number array. Each value in the array is matched one by one with the corresponding sensor, and is used to guide the subsequent restoration generation branch allocation. Through the above process, the number of restoration generation branches is reasonably allocated according to the relative size of the sensor distance, ensuring that sensors with longer distances and more severe signal attenuation can be allocated more restoration resources, thereby better correcting signal deviations and improving restoration effects.

[0032] In a possible implementation, the training steps of the multiple VHP concentration restoration branches in the first-order restorer include:

[0033] In the VHP concentration monitoring transmission data within the historical time, the transmitted sample VHP concentration signal set and the sample sensor distance set are collected, and the original VHP concentration signal is collected as the sample restored VHP concentration signal set; the sample VHP concentration signal set, the sample sensor distance set and the sample restored VHP concentration signal set are divided into multiple groups of restoration training data; the multiple groups of restoration training data are respectively used to construct and train multiple VHP concentration restoration branches based on the generative adversarial network, and each VHP concentration restoration branch includes a restoration generation path and a restoration adversarial path; after the training of the multiple VHP concentration restoration branches converges, a first-order restorer is obtained by combining them.

[0034] Optionally, key information is extracted from the VHP concentration monitoring transmission data in the historical time, including the sample VHP concentration signal set after transmission (i.e., the signal data attenuated due to distance and interference) and the corresponding sample sensor distance set (i.e., the distance data from the sensor to the processing unit). At the same time, the corresponding original VHP concentration signals of these transmission signals are collected as real signal data to construct a sample restored VHP concentration signal set. These data will serve as the basis for model training. Subsequently, the collected sample VHP concentration signal set, sample sensor distance set, and sample restored VHP concentration signal set are grouped and divided. Each set of data contains the corresponding transmission signal, sensor distance, and target restored signal to form a complete set of restored training data. These data are used to train different restoration branches to enhance the diversity and restoration ability of the model. Afterwards, multiple VHP concentration restoration branches are constructed and trained based on a generative adversarial network (GAN). Each restoration branch contains a restoration generation path and a restoration adversarial path to form a complete generative adversarial network structure. The restoration generation path is used to receive the transmitted sample signal and sensor distance as input, and generate an output that approximates the target restoration signal through a deep neural network; the restoration adversarial path is used to receive the generated signal and the target restoration signal, and judge the closeness of the generated signal to the real signal through feature extraction and discrimination mechanism. The transmission signal, sensor distance and target restoration signal are input into each restoration branch in turn. The generation path attempts to generate a high-quality restoration signal, while the adversarial path continuously improves the ability to distinguish between the generated signal and the real signal. During the training process, the optimization goal of the restoration generation path is to minimize the error between the generated signal and the target signal. The mean square error (MSE) loss function is usually used to measure the accuracy of the generated signal, and the generator weights are optimized through back propagation to make the generated signal gradually approach the target signal. The goal of the restoration adversarial path is to improve the ability to distinguish between the generated signal and the real signal. The cross entropy loss function is used during the training process. By optimizing the classification performance of the discriminator, its ability to distinguish between true and false signals is enhanced. During the entire adversarial training process, the generation path and the adversarial path are optimized alternately. The generation path attempts to deceive the adversarial path to generate a restored signal that is closer to the target signal, while the adversarial path is constantly adjusted to more accurately identify the defects of the generated signal. In order to coordinate the optimization process of the two, an adversarial loss function is also introduced to ensure the balance between the generator and the discriminator. When the error between the generated signal and the target restored signal drops to an acceptable range and the adversarial path cannot significantly distinguish the two, the training of the restoration branch is determined to be completed. At this point, the restoration branch has a strong signal repair capability and can efficiently handle signal distortion problems caused by distance attenuation and noise. By repeating the above training process for multiple sets of restoration training data, multiple restoration branches are gradually constructed, each branch is trained independently, and finally a diverse restoration capability is formed. The combination of these branches not only improves the generalization performance of the model, but also provides strong support for adapting to diverse signal attenuation situations in practical applications.Finally, multiple trained restoration branches are combined into a complete first-order restorer. The first-order restorer can receive transmission signals and distance data in subsequent practical applications to generate high-precision restored VHP concentration signals. Through this process, the constructed first-order restorer can make full use of historical monitoring data, combined with the powerful modeling capabilities of the generative adversarial network, effectively fix the signal attenuation problem and improve the accuracy and reliability of VHP concentration monitoring.

[0035] According to the first restored VHP concentration signal array, the humidity information array is indexed and obtained.

[0036] In one embodiment, according to the data information in the first restored VHP concentration signal array, the relationship between concentration and humidity in the historical monitoring data is used to obtain the corresponding humidity information array through matching or classification method indexing. Specifically, each first restored VHP concentration signal represents a restored concentration value at a sensor position, and there is a certain correlation between these restored concentration values ​​and the ambient humidity. The system knows that the restored signals will be input into the pre-built VHP humidity classifier one by one. The VHP humidity classifier is trained based on historical monitoring data, which includes known VHP concentrations and corresponding humidity information. By learning historical data, the classifier can identify the characteristic relationship between the concentration signal and the humidity information. When the first restored signal is input into the classifier, the classifier will classify the signal according to the mapping relationship that has been mastered, and output the corresponding humidity value. In addition, the humidity classifier can be connected to the monitoring device via Bluetooth to achieve real-time data synchronization and remote transmission of classification results, thereby improving monitoring efficiency. Finally, all humidity values ​​are organized into a humidity information array according to the corresponding relationship with the first restored VHP concentration signal array. This humidity information array provides the necessary environmental parameters for the subsequent second-order restoration process, which is used to further correct the concentration signal and improve the accuracy and reliability of the monitoring results.

[0037] In a possible implementation, indexing and obtaining a humidity information array according to the first restored VHP concentration signal array includes:

[0038] According to the VHP disinfection data in the historical time, the sample VHP concentration set and the corresponding sample humidity information set are collected; the mapping relationship between the sample VHP concentration set and the sample humidity information set is constructed to obtain a VHP humidity classifier; the multiple first restored VHP concentration signals in the first restored VHP concentration signal array are input into the VHP humidity classifier, and the multiple humidity information is obtained by mapping and classification as a humidity information array.

[0039] Optionally, the system terminal extracts a sample VHP concentration set and a corresponding sample humidity information set from the VHP disinfection monitoring data in the historical time. The sample VHP concentration set contains multiple known VHP concentration values, which are recorded under different humidity environments; the sample humidity information set includes environmental humidity data corresponding to these concentration values. These data are usually collected through a sensor array and stored in a database as the basis for training the model. Subsequently, the collected sample VHP concentration set and sample humidity information set are used to analyze the characteristic relationship between the concentration signal and humidity. Through data modeling methods (such as regression analysis, classification trees or machine learning algorithms), a mapping relationship model between the concentration signal and humidity is constructed to form a VHP humidity classifier. The classifier can accurately predict or classify the corresponding humidity information based on the input VHP concentration value. Specifically, taking random forest as an example, the system terminal performs feature analysis on the relationship between concentration and humidity based on the collected sample data, and divides the training set and the validation set to ensure the learning ability and generalization performance of the model. During the training process, the random forest generates multiple subsamples through random sampling with replacement to provide independent training data for each decision tree. In each decision tree, at the beginning of training, the sample data is recursively split and the concentration feature is selected as the split point to minimize the classification confusion of the sample (such as the Gini coefficient). The split data is divided into new nodes to form a branching structure of the tree until the leaf node is reached or the preset tree depth limit is met. Each decision tree is gradually optimized during the training process so that the humidity classification results stored in the leaf node reflect the correlation of the input concentration features as accurately as possible. Through multiple iterations and adjustments, all decision trees learn the mapping rules between concentration and humidity on their respective sub-sample sets. After all decision trees are trained, the random forest combines the classification results of each tree through a majority voting mechanism to generate the final humidity prediction. The trained classifier is evaluated by the validation set data, and the classification effect of the classifier in different humidity intervals is detected by calculating the classification accuracy and generating a confusion matrix. If it is found that the model does not perform well in some intervals, it can be further improved by increasing the number of decision trees, adjusting the depth of the tree, or optimizing the data sampling strategy. Finally, the trained random forest model, as a VHP humidity classifier, can receive the restored VHP concentration signal in real time and quickly match the corresponding humidity information. The output humidity information of the classifier is organized into a humidity information array, which corresponds one-to-one with the first restored VHP concentration signal array, providing reliable environmental parameter support for subsequent signal restoration processing and improving the accuracy and stability of concentration monitoring results. Afterwards, the system terminal inputs each restored concentration signal in the first restored VHP concentration signal array into the VHP humidity classifier one by one. The classifier will classify or predict each input signal according to the constructed mapping relationship, and output the corresponding humidity value, and organize it according to the corresponding relationship with the first restored VHP concentration signal array to form a humidity information array.The humidity information array contains humidity data corresponding to each sensor position or spatial position, and matches the first restored concentration signal one by one. Through the above process, the VHP humidity classifier built based on historical disinfection data can quickly and effectively map the restored concentration signal to the corresponding humidity information and generate a humidity information array, which provides key support for subsequent second-order restoration and concentration signal processing.

[0040] According to the humidity information array, the first restored VHP concentration signal array is subjected to second-order restoration to generate a second restored VHP concentration signal array as a signal enhancement processing result of an offline concentration monitor.

[0041] In one embodiment, humidity is a key factor affecting the accuracy of VHP concentration monitoring. Therefore, after completing the first restored VHP concentration signal array, it is necessary to perform a second-order restoration process in combination with humidity information to further improve the accuracy of the signal. First, according to each signal value in the first restored VHP concentration signal array, the corresponding humidity value is indexed in the humidity information array to ensure that each restored concentration signal can be matched one by one with the correct humidity data. Through this matching method, the necessary environmental parameters are provided for the second-order restoration. In the second-order restoration stage, each first restored concentration signal and its corresponding humidity value are input into the trained second-order restorer for processing. The restorer is constructed based on historical data, and can consider the impact of humidity on the signal. By learning the complex relationship between humidity and concentration, the restored signal is further corrected. In actual operation, each first restored concentration signal and humidity information pair is processed by the second-order restorer to output a second restored concentration signal. All second restored concentration signals are organized into a second restored VHP concentration signal array. The signal array represents high-precision concentration data after humidity correction, which can more accurately reflect the VHP concentration distribution in the target area. After the processing is completed, the system terminal can output a detailed report of the restoration results through the built-in printing function. The report content includes the restored concentration signal, the corresponding humidity information and the final second-order restored concentration signal, which is convenient for operators to archive and analyze the monitoring results. These data, as the signal enhancement processing results of the offline concentration monitor, not only improve the reliability and accuracy of the monitoring results, but also provide convenience for future data analysis and quality control.

[0042] In a possible implementation, performing second-order restoration on the first restored VHP concentration signal array according to the humidity information array to obtain a second restored VHP concentration signal array includes:

[0043] In the VHP concentration monitoring data within the historical time, a set of sample monitoring VHP concentration signals and a set of sample humidity information obtained by monitoring are collected, and an actual set of sample actual VHP concentrations is collected; the sample monitoring VHP concentration signal set and the sample humidity information set are used to generate input data, the sample actual VHP concentration set is used as discriminant data, and a generative adversarial network is used to train a second-order restorer; multiple first restored VHP concentration signals in the first restored VHP concentration signal array are input into the second-order restorer that has converged in training, multiple second restored VHP concentration signals are generated and output, and a second restored VHP concentration signal array is obtained.

[0044] Optionally, in the second-order restoration process, the system terminal first extracts relevant data from the historical VHP concentration monitoring data, including a sample monitoring VHP concentration signal set (recorded transmission signal), a sample humidity information set (corresponding humidity environment parameters), and a sample actual VHP concentration set (real concentration data). The sample monitoring VHP concentration signal and humidity information are used as input data of the generator, and the actual concentration signal is used as the target data of the discriminator to train the second-order restorer through a generative adversarial network (GAN). The generator is responsible for generating the corrected concentration signal, while the discriminator is used to evaluate the closeness between the generated signal and the actual signal. Since the training process is similar to the GAN training process of the first-order restoration, it will not be repeated here. When the second-order restorer training converges, the first restored VHP concentration signal array is input one by one into the trained second-order restorer to generate a corrected concentration signal. In this process, the second-order restorer combines the humidity information of each input signal to further correct the influence of transmission and environmental interference on the concentration signal, and finally generates multiple more accurate second restored VHP concentration signals. All generated second restored signals are organized into a second restored VHP concentration signal array as the final signal enhancement processing result of the offline concentration monitor.

[0045] Embodiment 2, based on the same inventive concept as the signal enhancement method for an offline concentration monitor in the above embodiment, Figure 2As shown, the present application provides a signal enhancement platform for an offline concentration monitor, and the platform and method embodiments in the embodiments of the present application are based on the same inventive concept. The platform includes: a sensor array deployment module 1, the sensor array deployment module 1 is used to deploy the VHP sensor array of the offline concentration monitor in the target area to be monitored for VHP concentration, wherein the VHP sensor array includes a sensor position information array; a concentration monitoring module 2, the concentration monitoring module 2 is used to monitor the VHP concentration through the VHP sensor array, and transmit the VHP concentration signal array obtained by monitoring to the processing unit of the offline concentration monitor; a first-order restoration generation module 3, the first-order restoration generation module 3 is used in the processing unit to generate the VHP concentration signal according to the sensor array. The position information array generates a sensor distance array, and based on the sensor distance array, the VHP concentration signal array is first-order restored to obtain a first restored VHP concentration signal array; a humidity information acquisition module 4, the humidity information acquisition module 4 is used to index and obtain a humidity information array based on the first restored VHP concentration signal array; a second-order restoration generation module 5, the second-order restoration generation module 5 is used to perform a second-order restoration on the first restored VHP concentration signal array based on the humidity information array to obtain a second restored VHP concentration signal array as a signal enhancement processing result of an offline concentration monitor.

[0046] Furthermore, the sensor array layout module 1 also includes:

[0047] At multiple monitoring locations within a target area to be monitored for VHP concentration, multiple VHP sensors of an offline concentration monitor are deployed to form a VHP sensor array; a coordinate system of the target area is constructed, the position coordinates of the multiple VHP sensors are obtained, and a sensor position information array of the VHP sensor array is constructed.

[0048] Furthermore, the first-order restoration generation module 3 also includes:

[0049] According to the position coordinates of the processing unit and in combination with the sensor position information array, the multiple sensor distances between the multiple VHP sensors in the VHP sensor array and the processing unit are calculated to obtain a sensor distance array; according to the multiple sensor distances in the sensor distance array, multiple restoration generation numbers for first-order restoration generation of the multiple VHP concentration signals in the VHP concentration signal array are respectively set; 4 the multiple VHP concentration signals are respectively input into the multiple restoration generation numbers of the VHP concentration restoration branches in the first-order restorer for integrated restoration generation to obtain multiple first restored VHP concentration signal sets, wherein the first-order restorer includes multiple trained VHP concentration restoration branches; the multiple first restored VHP concentration signal sets are averaged to obtain multiple first restored VHP concentration signals, and a first restored VHP concentration signal array is constructed.

[0050] Furthermore, the first-order restoration generation module 3 also includes:

[0051] The total number of branches of the multiple VHP concentration restoration branches in the first-order restorer is obtained; the ratio of each sensor distance to the sum of the multiple sensor distances is calculated, multiplied by the total number of branches and rounded to obtain multiple restoration generation numbers.

[0052] Furthermore, the first-order restoration generation module 3 also includes:

[0053] In the VHP concentration monitoring transmission data within the historical time, the transmitted sample VHP concentration signal set and the sample sensor distance set are collected, and the original VHP concentration signal is collected as the sample restored VHP concentration signal set; the sample VHP concentration signal set, the sample sensor distance set and the sample restored VHP concentration signal set are divided into multiple groups of restoration training data; the multiple groups of restoration training data are respectively used to construct and train multiple VHP concentration restoration branches based on the generative adversarial network, and each VHP concentration restoration branch includes a restoration generation path and a restoration adversarial path; after the training of the multiple VHP concentration restoration branches converges, a first-order restorer is obtained by combining them.

[0054] Furthermore, the humidity information obtaining module 4 also includes:

[0055] According to the VHP disinfection data in the historical time, the sample VHP concentration set and the corresponding sample humidity information set are collected; the mapping relationship between the sample VHP concentration set and the sample humidity information set is constructed to obtain a VHP humidity classifier; the multiple first restored VHP concentration signals in the first restored VHP concentration signal array are input into the VHP humidity classifier, and the multiple humidity information is obtained by mapping and classification as a humidity information array.

[0056] Furthermore, the second-order restoration generation module 5 also includes:

[0057] In the VHP concentration monitoring data within the historical time, a set of sample monitoring VHP concentration signals and a set of sample humidity information obtained by monitoring are collected, and an actual set of sample actual VHP concentrations is collected; the sample monitoring VHP concentration signal set and the sample humidity information set are used to generate input data, the sample actual VHP concentration set is used as discriminant data, and a generative adversarial network is used to train a second-order restorer; multiple first restored VHP concentration signals in the first restored VHP concentration signal array are input into the second-order restorer that has converged in training, multiple second restored VHP concentration signals are generated and output, and a second restored VHP concentration signal array is obtained.

[0058] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0060] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A signal enhancement method for an off-line concentration monitor, characterized in that: The method comprises: Arranging a VHP sensor array of an offline concentration monitor in a target area to be monitored for VHP concentration, wherein the VHP sensor array includes a sensor position information array; Monitoring VHP concentration by using the VHP sensor array, and transmitting the VHP concentration signal array obtained by monitoring to a processing unit of an offline concentration monitor; In the processing unit, a sensor distance array is generated according to the sensor position information array, and the VHP concentration signal array is first-order restored according to the sensor distance array to obtain a first restored VHP concentration signal array; According to the first restored VHP concentration signal array, index to obtain a humidity information array; According to the humidity information array, performing second-order restoration on the first restored VHP concentration signal array to obtain a second restored VHP concentration signal array as a signal enhancement processing result of an offline concentration monitor; Wherein, in the processing unit, a sensor distance array is generated according to the sensor position information array, and a first-order restoration is performed on the VHP concentration signal array according to the sensor distance array, including: According to the position coordinates of the processing unit, in combination with the sensor position information array, multiple sensor distances between multiple VHP sensors in the VHP sensor array and the processing unit are calculated to obtain a sensor distance array; According to the plurality of sensor distances in the sensor distance array, respectively setting a plurality of restoration generation numbers for performing first-order restoration generation on the plurality of VHP concentration signals in the VHP concentration signal array; The multiple VHP concentration signals are respectively input into the multiple restored and generated VHP concentration restoration branches in the first-order restorer for integrated restoration and generation to obtain multiple first restored VHP concentration signal sets, wherein the first-order restorer includes multiple trained VHP concentration restoration branches, and multiple groups of restoration training data are respectively used to construct and train multiple VHP concentration restoration branches based on a generative adversarial network, each VHP concentration restoration branch includes a restoration generation path and a restoration adversarial path; after the training of the multiple VHP concentration restoration branches converges, a first-order restorer is obtained by combining them; the multiple first restored VHP concentration signal sets are averaged to obtain multiple first restored VHP concentration signals, and a first restored VHP concentration signal array is constructed.

2. The signal enhancement method for an offline concentration monitor according to claim 1, characterized in that: In the target area to be monitored for VHP concentration, a VHP sensor array of an offline concentration monitor is deployed, including: At multiple monitoring locations within a target area to be monitored for VHP concentration, multiple VHP sensors of an offline concentration monitor are deployed to form a VHP sensor array; a coordinate system of the target area is constructed, the position coordinates of the multiple VHP sensors are obtained, and a sensor position information array of the VHP sensor array is constructed.

3. The signal enhancement method for an off-line concentration monitor according to claim 1, characterized in that: According to the plurality of sensor distances in the sensor distance array, a plurality of restoration generation numbers for performing first-order restoration generation on the plurality of VHP concentration signals in the VHP concentration signal array are respectively set, including: The total number of branches of the multiple VHP concentration restoration branches in the first-order restorer is obtained; the ratio of each sensor distance to the sum of the multiple sensor distances is calculated, multiplied by the total number of branches and rounded to obtain multiple restoration generation numbers.

4. The signal enhancement method for an off-line concentration monitor according to claim 1, characterized in that: The training steps of the multiple VHP concentration restoration branches in the first-order restorer include: In the VHP concentration monitoring transmission data within the historical time, the sample VHP concentration signal set and the sample sensor distance set after transmission are collected, and the original VHP concentration signal is collected as the sample restored VHP concentration signal set; the sample VHP concentration signal set, the sample sensor distance set and the sample restored VHP concentration signal set are divided into multiple groups of restoration training data.

5. The signal enhancement method for an off-line concentration monitor according to claim 1, characterized in that: According to the first restored VHP concentration signal array, the humidity information array is obtained by indexing, including: According to the VHP disinfection data in the historical time, the sample VHP concentration set and the corresponding sample humidity information set are collected; Constructing a mapping relationship between the sample VHP concentration set and the sample humidity information set to obtain a VHP humidity classifier; The plurality of first restored VHP concentration signals in the first restored VHP concentration signal array are input into the VHP humidity classifier, and a plurality of humidity information are obtained by mapping and classifying as a humidity information array.

6. The signal enhancement method for an off-line concentration monitor according to claim 1, characterized in that: According to the humidity information array, performing second-order restoration on the first restored VHP concentration signal array to obtain a second restored VHP concentration signal array includes: In the VHP concentration monitoring data in the historical time, a set of sample monitoring VHP concentration signals and a set of sample humidity information obtained by monitoring are collected, and an actual set of sample actual VHP concentration is collected; The sample monitoring VHP concentration signal set and the sample humidity information set are used to generate input data, the sample actual VHP concentration set is used as discrimination data, and a generative adversarial network is used to train a second-order restorer; multiple first restored VHP concentration signals in the first restored VHP concentration signal array are input into the second-order restorer that has converged in training, multiple second restored VHP concentration signals are generated and output, and a second restored VHP concentration signal array is obtained.

7. A signal enhancement platform for an offline concentration monitor, characterized in that: The platform is used to execute a signal enhancement method for an offline concentration monitor according to any one of claims 1 to 6, comprising: A sensor array deployment module: deploying a VHP sensor array of an offline concentration monitor in a target area to be monitored for VHP concentration, wherein the VHP sensor array includes a sensor position information array; Concentration monitoring module: performs VHP concentration monitoring through the VHP sensor array, and transmits the VHP concentration signal array obtained by monitoring to the processing unit of the offline concentration monitor; A first-order restoration generation module: in the processing unit, generating a sensor distance array according to the sensor position information array, and performing first-order restoration generation on the VHP concentration signal array according to the sensor distance array to obtain a first restored VHP concentration signal array; Humidity information acquisition module: according to the first restored VHP concentration signal array, index to obtain the humidity information array; second-order restoration generation module: according to the humidity information array, perform second-order restoration generation on the first restored VHP concentration signal array to obtain the second restored VHP concentration signal array as the signal enhancement processing result of the offline concentration monitor.

Images