Inflow gas humidity control system based on smart devices

Abstract

The present invention discloses an inflow gas humidity regulation system based on intelligent devices, which relates to the technical field of gas humidity control. The system includes: a user management component for receiving basic user information and user-related feature data sets of target users; a device management component for receiving basic device feature data of target ventilators; an initial regulation and analysis component for performing initial ventilator regulation and analysis based on basic user information, user-related feature data sets and basic device feature data, and generating initial device regulation and control decisions; an inflow gas humidity enhancement component for enhancing the inflow gas humidity of the initial device regulation and control decisions, and generating device regulation and control plans; a humidity tracing and control component, which performs tracing and optimization regulation and control on the device regulation and control plans based on the embedded humidity tracing and control algorithm. The technical problems of low control accuracy and slow regulation response are solved, and the technical effects of fast regulation response and refined control are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas humidity control, and in particular to an inflow gas humidity regulating system based on intelligent devices. Background Art

[0002] Among smart devices, a smart ventilator is a device used to provide breathing support for users and provide air with a certain flow rate and humidity. Specifically, it mechanically delivers a predetermined tidal volume into the user's lungs to assist or replace the user's spontaneous breathing.

[0003] Smart ventilators usually have a gas heating and humidification function, using a humidifier tank or heated humidifier to increase the temperature and humidity of the inhaled gas to simulate the moisture exchange of the respiratory tract under normal physiological conditions. Existing gas humidity technology has technical problems such as low control accuracy and slow adjustment response. Summary of the invention

[0004] The present invention provides an inflow gas humidity regulation system based on intelligent devices to solve the technical problems of low control accuracy and slow regulation response in the prior art, and to achieve the technical effects of fast regulation response and refined control.

[0005] The present invention provides an inflow gas humidity control system based on an intelligent device, wherein the system comprises:

[0006] A user management component, the user management component is used to receive basic user information and a user-related feature data set of a target user, wherein the user-related feature data set includes user vital sign feature data, user environment feature data and user activity feature information;

[0007] A device management component, the device management component is used to receive basic device characteristic data of a target ventilator, the target ventilator having a corresponding relationship with the target user;

[0008] An initial control analysis component, which performs initial ventilator control analysis based on the user basic information, the user-related feature data set, and the device basic feature data to generate an initial device control decision;

[0009] An inflow gas humidity enhancement component, the inflow gas humidity enhancement component is used to enhance the inflow gas humidity of the initial equipment control decision to generate an equipment control plan;

[0010] A humidity tracing and control component is used to trace and optimize the equipment control scheme based on an embedded humidity tracing and control algorithm.

[0011] The present invention discloses an inflow gas humidity regulation system based on smart devices, which is configured to include: a user management component for processing basic user information and associated characteristic data, including physical signs, environment and activity characteristics. A device management component, responsible for processing the basic characteristic data of the ventilator and associating it with the user data. An initial control analysis component, which generates an initial control decision based on user and device data. An inflow gas humidity enhancement component enhances the humidity of the initial control decision. A humidity tracing control component traces and optimizes the control scheme based on an embedded algorithm. The inflow gas humidity regulation system based on smart devices disclosed by the present invention solves the technical problems of low control accuracy and slow regulation response, and achieves the technical effects of fast regulation response and refined control. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a structural schematic diagram of the inflow gas humidity regulation system based on intelligent devices of the present invention.

[0013] Explanation of the reference numerals: user management component 11 , device management component 12 , initial regulation and analysis component 13 , inflow gas humidity enhancement component 14 , humidity tracing and regulation component 15 . DETAILED DESCRIPTION

[0014] The technical solution provided in the embodiments of the present invention is to solve the technical problems of low control accuracy and slow adjustment response in the prior art. The overall idea adopted is as follows:

[0015] First, the user basic information and user-related feature data set of the target user are received through the user management component, wherein the user-related feature data set includes user vital sign feature data, user environment feature data and user activity feature information; then, the device management component receives the basic device feature data of the target ventilator, which has a corresponding relationship with the target user; then, the initial control analysis component performs initial ventilator control analysis based on the user basic information, the user-related feature data set and the basic device feature data, and generates an initial device control decision; then, the inflow gas humidity enhancement component enhances the inflow gas humidity of the initial device control decision and generates a device control plan; finally, the humidity tracing control component performs traceability and optimization control on the device control plan based on the embedded humidity tracing control algorithm.

[0016] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0017] Embodiment 1

[0018] Figure 1 The structure diagram of the inflow gas humidity control system based on the smart device of the present invention is shown in FIG. 1 , wherein the inflow gas humidity control system based on the smart device includes:

[0019] A user management component 11, the user management component is used to receive basic user information and a user-related feature data set of a target user, wherein the user-related feature data set includes user vital sign feature data, user environment feature data and user activity feature information;

[0020] The target user refers to the object that needs the smart device to provide gas supply; the user management component is a module in the system, responsible for managing and processing information related to the user. Specifically, the user management component receives various information from the user through the input interface, including the user's basic information and the user-related feature data set.

[0021] Optionally, the input interface of the user management component includes a direct input interface and an external input interface, wherein the direct input interface is used to connect to a human-computer interaction device and is an interface through which a user can directly input information into the system; the external input interface refers to an interface through which the user management component exchanges data with other external systems. Through external system interfaces, such as APIs and data interfaces, data is exchanged with other external systems to obtain existing structured user basic information and user-related feature data sets.

[0022] Among them, user basic information refers to the user's personal information, such as name, age, gender, contact information, etc. User-related feature data sets refer to various feature data related to users, such as the user's height, weight, health status, etc. Specifically, user-related feature data sets include user physical feature data, user environment feature data, and user activity feature information.

[0023] Furthermore, user vital sign characteristic data refers to characteristic data related to the user's physiological condition and physiological needs; user environment characteristic data refers to characteristic data related to the user's environment, such as temperature, humidity, light, etc. These characteristic data help the system better understand the user's living environment. User activity characteristic information refers to characteristic data related to user activities, such as exercise volume, exercise intensity, etc. The air supply humidity required by the user is different under different exercise intensities. For example, if the user is at a low activity level, such as sleeping or lying in bed for a long time, the respiratory mucosa may become dry, and a corresponding higher humidity air supply is required to keep the user's respiratory tract moist. Correspondingly, if the user's activity level is high, such as walking or living normally, more respiratory moisture will be generated.

[0024] Through this process, the user management component can collect and store various information of the target users, including the users' basic information, user-related feature data sets, etc., to provide support for subsequent user analysis and personalized services.

[0025] A device management component 12, the device management component is used to receive basic device characteristic data of a target ventilator, the target ventilator having a corresponding relationship with the target user;

[0026] The device management component is responsible for managing and processing device-related information. Specifically, the device management component receives the basic characteristic data of the target ventilator, which helps the system better understand and control the ventilator. The basic characteristic data refers to the basic information related to the ventilator, such as working parameters, applicable environment, etc.

[0027] Optionally, there is a corresponding relationship between the target ventilator and the target user, that is, the target ventilator serves the target user. By confirming the corresponding relationship between the target ventilator and the target user, it is ensured that the management and control of the ventilator equipment can be carried out according to the needs of the target user.

[0028] In some embodiments, the device management component is further configured to:

[0029] Receiving equipment monitoring data of the target ventilator;

[0030] Transmitting the device monitoring data to a device status evaluation unit in the device management component to generate a device status evaluation coefficient;

[0031] Inputting the device state evaluation coefficient into a device state verification unit in the device management component, wherein the device state verification unit includes a first device state evaluation threshold and a second device state evaluation threshold, and the first device state evaluation threshold is greater than the second device state evaluation threshold;

[0032] If the device state evaluation coefficient is greater than / equal to the first device state evaluation threshold, generating a control analysis signal, and activating the initial control analysis component according to the control analysis signal;

[0033] If the device state evaluation coefficient is less than the first device state evaluation threshold, and the device state evaluation coefficient is greater than the second device state evaluation threshold, generating a device operation and maintenance signal;

[0034] If the device state evaluation coefficient is less than / equal to the second device state evaluation threshold, a redundant device switching signal is generated.

[0035] Optionally, the device management component is also used for automated maintenance and monitoring of the target ventilator. Specifically, it collects and acquires the equipment monitoring data of the ventilator, evaluates the status of the ventilator, and activates corresponding components or generates corresponding signals based on the status judgment results to ensure the normal operation of the ventilator equipment and the user's treatment safety.

[0036] Among them, equipment monitoring data refers to various real-time data generated by the ventilator during operation, including pressure, flow, temperature, etc. Equipment monitoring data reflects the current working status of the ventilator.

[0037] Exemplarily, the device management component receives device monitoring data through a data interface and a corresponding communication protocol. The data interface includes a hardware interface (such as a serial port, a USB interface), a network interface (such as Wi-Fi, Ethernet), etc. Then, the received raw data is parsed to understand the meaning and structure of the data, and the parsed device monitoring data is stored in a storage medium, such as a database, a file system, or a data structure in memory. For subsequent data processing and analysis.

[0038] Furthermore, the device management component transmits the received device monitoring data to the device status evaluation unit for status evaluation. The device status evaluation unit is a functional unit within the device management component, which is used to evaluate the status of the ventilator according to the level of the device monitoring data and the preset evaluation criteria, and calculate the device status evaluation coefficient, which reflects the health and operating status of the ventilator. Among them, the evaluation criteria include the normal range or standard value of each monitoring indicator.

[0039] For example, the device status evaluation coefficient is generated by calculating the deviation of each monitoring indicator in the evaluation standard from the corresponding standard value or normal range, and extracting characteristic values, including the evaluation deviation, the maximum deviation or the weighted result of multiple statistical values. The specific calculation method is determined according to the design requirements and application requirements of the target ventilator.

[0040] Optionally, the device status verification unit is used to verify the status of the ventilator, and the device status verification unit includes a first device status evaluation threshold and a second device status evaluation threshold set based on a dual threshold method, and the first device status evaluation threshold is greater than the second device status evaluation threshold. The multiple evaluation intervals formed by the first device status evaluation threshold and the second device status evaluation threshold reflect the different degrees of device status, and exemplarily include a first interval representing a normal device, a second interval representing a need for maintenance, and a third interval reflecting an abnormal device that needs to be replaced and switched.

[0041] Specifically, the state of the ventilator is judged according to the comparison result between the device state evaluation coefficient and the threshold. If the device state evaluation coefficient is greater than or equal to the first device state evaluation threshold, the device state verification unit generates a control analysis signal and activates the initial control analysis component to facilitate the subsequent corresponding control operations; if the device state evaluation coefficient is less than the first device state evaluation threshold, but greater than the second device state evaluation threshold, the device state verification unit generates a device operation and maintenance signal to notify and instruct professional technicians to perform equipment operation and maintenance operations, such as maintenance and repair. If the device state evaluation coefficient is less than or equal to the second device state evaluation threshold, it means that there is a serious problem with the target ventilator and the gas supply capacity is insufficient. The device state verification unit generates a redundant device switching signal, which is used to control the system to switch to the backup ventilator device to ensure the continuity and reliability of gas supply.

[0042] An initial control analysis component 13, wherein the initial control analysis component is embedded with a decision function unit constructed based on a neural network, performs initial ventilator control analysis based on the user basic information, the user-related feature data set and the device basic feature data, and generates an initial device control decision;

[0043] The initial control analysis component is the adjustment decision function component in the system, which is used to perform initial control analysis of the ventilator based on the information of the user and the device, and then obtain equipment control decisions with a high degree of personalization and high adaptability to the target user.

[0044] Optionally, the initial control analysis component is embedded with a decision function unit built based on a neural network. The decision function unit takes user basic information, user-related feature data sets and device basic feature data as input data, and generates an initial device control decision through an internal network layer according to the learned decision mapping relationship.

[0045] In some embodiments, the initial control analysis component is embedded with a decision function unit constructed based on a neural network, and performs initial ventilator control analysis based on the user basic information, the user-related feature data set, and the device basic feature data to generate an initial device control decision, including:

[0046] Loading a sample control scenario data set, wherein the sample control scenario data set includes a plurality of sample control scenario data groups that meet a predetermined capacity constraint, and each sample control scenario data group includes sample user basic information, a sample user associated feature data set, and sample device basic feature data;

[0047] Performing confidence control decision sample mining according to the sample control scenario data set to generate a confidence control decision sample set;

[0048] Taking the sample control scene data set as input information and the confident control decision sample set as output information, supervised training is performed on a predetermined base learner, when the control decision error coefficient is less than the control decision error threshold for a preset number of consecutive times;

[0049] The user basic information, the user-related feature data set and the device basic feature data are input into the decision function unit to obtain the initial device regulation decision.

[0050] Specifically, the sample control scenario data set is the original sample data set, which contains multiple sample control scenario data groups of multiple scenarios and multiple control effects, and the number of sample control scenario data groups in the sample control scenario data set meets the predetermined capacity constraint. In other words, the number of samples in the sample control scenario data set meets the preset sample number constraint, ensuring that the richness and data volume of the sample data meet the subsequent training requirements. Among them, each sample control scenario data group includes associated stored sample user basic information, sample user associated feature data set and sample device basic feature data.

[0051] Optionally, confidence control decision sample mining is performed based on the sample control scenario data set to generate a confidence control decision sample set, wherein confidence control decision sample mining refers to mining valuable control decision samples from the sample control scenario data set. Exemplarily, it includes mining control decision samples with similar scenarios and good control results.

[0052] Further, the confidence control decision sample set is used as training data to supervise the predetermined base learner. Exemplarily, the predetermined base learner is a network model constructed by a BP neural network or a fully connected neural network. When the control decision error coefficient is less than the control decision error threshold for a preset number of consecutive times, it is considered that the predetermined base learner has completed the learning of the decision logic and decision mapping relationship contained in the confidence control decision sample set, and the output performance meets the needs of the target scenario, and the predetermined base learner is stored as a decision function unit. Among them, the control decision error coefficient can be accuracy, precision, recall rate or F1 score, etc.

[0053] In some implementations, performing confidence control decision sample mining according to the sample control scenario data set to generate a confidence control decision sample set includes:

[0054] Extracting a first sample control scene data group according to the sample control scene data set, wherein the first sample control scene data group is any one of the sample control scene data sets;

[0055] Taking the first sample regulation scenario data set as a feature retrieval constraint and taking the sample ventilator control scheme as a feature retrieval target;

[0056] Perform ventilator control record feature retrieval according to the feature retrieval constraint and the feature retrieval target to generate a first sample ventilator control scheme set;

[0057] Perform support calculation on each sample solution in the first sample ventilator control solution set to generate a sample solution support calculation result;

[0058] Perform confidence calculation according to the sample solution support calculation result to generate a sample solution confidence calculation result;

[0059] Screening the winning sample solution confidence according to the sample solution confidence calculation result, and matching the first confidence sample ventilator control solution with the first sample ventilator control solution set;

[0060] The first confidence sample ventilator control scheme is added to the confidence regulation decision sample set.

[0061] Optionally, first, a sample control scenario data group is randomly extracted from the sample control scenario data set as the first sample control scenario data group, and then, the feature data in the first sample control scenario data group is used as a feature retrieval constraint to screen out qualified sample ventilator control schemes and generate a first sample ventilator control scheme set. Exemplarily, based on the fuzzy search principle, a ventilator control record feature retrieval is performed to obtain a first sample ventilator control scheme set with higher richness. Among them, the first sample ventilator control scheme set contains multiple ventilator control scheme usage records in a scenario, and the ventilator control schemes in the multiple ventilator control scheme usage records are repeatable.

[0062] Optionally, support and confidence are calculated for each sample solution in the first sample ventilator control solution set to evaluate its applicability in the first sample control scenario data group. Specifically, the support is calculated by evaluating the number of times the sample solution appears in the first sample control scenario data group, and the confidence is the ratio of the support to the total number of times. In other words, the confidence is the proportion of the support.

[0063] Furthermore, based on the sample scheme confidence calculation results, the sample scheme with the highest confidence is screened out as the confidence of the winning sample scheme; then, the first sample ventilator control scheme set is traversed, and the ventilator control scheme corresponding to the confidence of the winning sample scheme is extracted as the first confidence sample ventilator control scheme and stored in the confidence control decision sample set.

[0064] Optionally, the confidence control decision sample set includes multiple confidence sample ventilator control schemes, and the multiple confidence sample ventilator control schemes correspond to multiple ventilator application scenarios, wherein the ventilator application scenario refers to a combination of a set of user basic information, user-associated feature data sets and device basic feature data, and multiple ventilator application scenarios correspond to multiple combinations of user basic information, user-associated feature data sets and device basic feature data.

[0065] By constructing a confidence control decision sample set, it is helpful to quickly call the ventilator control scheme when adjusting the target ventilator, thereby improving decision-making efficiency.

[0066] An inflow gas humidity enhancement component 14, the inflow gas humidity enhancement component is used to enhance the inflow gas humidity of the initial equipment control decision to generate an equipment control plan;

[0067] The inflow gas humidity enhancement component is a humidity control functional component in the system, which is used to optimize and enhance the inflow gas humidity for the initial equipment control decision, thereby ensuring the quality and comfort of air supply.

[0068] In some embodiments, the inflow gas humidity enhancement component is used to enhance the inflow gas humidity of the initial equipment control decision to generate an equipment control plan, including:

[0069] The inflow gas humidity enhancement component includes a gas humidity feature record library and a feature twin recognition unit, wherein the gas humidity feature record library includes multiple groups of humidity feature records, each group of humidity feature records includes historical user basic information, historical user associated feature data sets, historical equipment control decisions and historical inflow gas humidity information;

[0070] According to the feature twin recognition unit, the associated inflow gas humidity recognition is performed on the gas humidity feature record library in combination with the user basic information, the user associated feature data set and the initial equipment control decision to generate an associated inflow gas humidity set;

[0071] Calculating the central value of the associated inflow gas humidity set to generate a confident inflow gas humidity;

[0072] The initial equipment control decision is optimized according to the confident inflow gas humidity to obtain the equipment control solution.

[0073] Optionally, the inflow gas humidity enhancement component includes a gas humidity feature record library and a feature twin recognition unit, wherein the gas humidity feature record library is a database storing multiple groups of humidity feature records, each group of records including historical user basic information, historical user-related feature data sets, historical equipment control decisions, and historical inflow gas humidity information. The feature twin recognition unit is a module in the inflow gas humidity enhancement component, which is used to associate the gas humidity feature record library with the inflow gas humidity recognition in combination with the user basic information, the user-related feature data set, and the initial equipment control decision.

[0074] Further, the concentration value of the associated inflow gas humidity set is calculated, and the confident inflow gas humidity is generated according to the concentration value. The concentration value refers to the value with a high frequency in the associated inflow gas humidity set, and the concentration value reflects the optimal inflow gas humidity under the current user and device characteristics.

[0075] Furthermore, the initial device control decision is adjusted according to the confidence in the inflow gas humidity, including adjusting the gas humidification amount and other related parameters to obtain a better device control solution. Through this process, the inflow gas humidity enhancement component optimizes the initial device control decision according to the characteristics of the user and the device, combined with historical data, and generates a better device control solution, which helps to improve the accuracy and effectiveness of ventilator control.

[0076] In some implementations, according to the feature twin recognition unit, the associated inflow gas humidity recognition is performed on the gas humidity feature record library in combination with the user basic information, the user associated feature data set and the initial equipment control decision to generate an associated inflow gas humidity set, including:

[0077] Extracting a first group of humidity feature records according to the gas humidity feature record library, wherein the first group of humidity feature records is any group of humidity feature records in the gas humidity feature record library;

[0078] Adding the first historical user basic information, the first historical user associated feature data set, and the first historical equipment control decision in the first set of humidity feature records to the first sample feature data set;

[0079] Adding the user basic information, the user-associated feature data set, and the initial device control decision to a baseline feature data set;

[0080] Inputting the first sample feature data set and the reference feature data set into the feature twin recognition unit to obtain a first feature twin recognition index;

[0081] If the first feature twin identification index satisfies the preset feature twin identification constraint, the first historical inflow gas humidity information in the first group of humidity feature records is added to the associated inflow gas humidity set.

[0082] Optionally, to generate an associated inflow gas humidity set, first, randomly select a group of historical data from the gas humidity feature record library as the first group of humidity feature records. Then, add the first historical user basic information, the first historical user associated feature data set, and the first historical equipment control decision in the first group of humidity feature records to the first sample feature data set. At the same time, add the user basic information, the user associated feature data set, and the initial equipment control decision to the baseline feature data set. Next, input the first sample feature data set and the baseline feature data set into the feature twin recognition unit for feature matching and recognition. Among them, the feature twin recognition unit is a functional unit for identifying and matching features in historical data.

[0083] Exemplarily, the feature twin recognition unit is constructed based on a neural network, including: a feature extraction layer based on statistics, convolutional neural network (CNN), recurrent neural network (RNN), etc.; a vectorization layer constructed based on vectorization methods such as Word2Vec and TF-IDF; a recognition layer based on a twin network model, which contains two sub-networks of the same structure, each of which is responsible for processing an input feature vector. The two sub-networks share the same weights to ensure that they learn similar representations. The recognition layer matches the input feature vector. And outputs a matching score indicating the degree of similarity between the feature vector and the feature vector in the historical data. According to the matching score, the similarity of the samples in the first sample feature data set and the benchmark feature data set can be determined, and the output is the first feature twin recognition index.

[0084] Furthermore, if the first feature twin recognition index satisfies the preset feature twin recognition constraint, it means that the first group of humidity feature records corresponding to the first feature twin recognition index has a high similarity with the target user and the target ventilator, and the first historical inflow gas humidity information in the first group of humidity feature records is added to the associated inflow gas humidity set.

[0085] The humidity tracing and regulating component 15 performs tracing and optimizing regulation on the equipment regulation scheme based on the built-in humidity tracing and regulating algorithm.

[0086] In some embodiments, the humidity tracing and controlling component performs tracing and optimizing control on the equipment control scheme based on an embedded humidity tracing and controlling algorithm, including:

[0087] Interacting with the user management component to obtain continuously monitored user feature data, wherein the continuously monitored user feature data includes continuously monitored user vital sign feature data, continuously monitored user environment feature data, and continuously monitored user activity feature information;

[0088] Performing user feature change analysis based on the continuously monitored user feature data and the user-associated feature data set to generate user feature change data;

[0089] Identify the impact of humidity correlation on inflow gas according to the user characteristic change data, and generate a user change correlation characteristic impact coefficient;

[0090] When the user change associated characteristic influence coefficient is greater than / equal to the user change associated characteristic influence threshold, humidity compensation of the inflowing gas is performed according to the user characteristic change data, and a compensation gas humidity adjustment scheme is generated.

[0091] Optionally, the user management component continuously monitors the user and the user environment through monitoring devices or monitoring sensors, obtains continuously monitored user vital sign characteristic data (such as breathing rate, heart rate, etc.), continuously monitored user environment characteristic data (such as temperature, humidity, etc.), and continuously monitored user activity characteristic information (such as exercise volume, activity duration, etc.), so as to capture the target user's state change information.

[0092] Furthermore, based on the monitored user feature data and user-related feature data set, the user feature changes are analyzed to determine the change trend of the user features, including the rate and magnitude of the changes, and generate user feature change data.

[0093] Optionally, analyze the user feature change data, identify the impact of the inflow gas humidity association, analyze and obtain the impact of the inflow gas humidity on the user, and generate the user change association feature impact coefficient. Then, compare the user change association feature impact coefficient with the user change association feature impact threshold. If it is greater than or equal to the threshold, adjust the humidity of the inflow gas according to the user feature change data, and generate a compensating gas humidity adjustment plan to reduce or eliminate the adverse impact on the user. Among them, the user change association feature impact threshold is a preset threshold used to determine whether the user change association feature impact coefficient needs to be compensated for the inflow gas humidity.

[0094] Through this process, the humidity tracing and control component, based on the embedded humidity tracing and control algorithm, traces and optimizes the equipment control plan according to user feature change data, thereby improving the accuracy and effectiveness of the equipment control plan.

[0095] In some embodiments, the humidity tracing and controlling component performs tracing and optimizing control on the equipment control scheme based on an embedded humidity tracing and controlling algorithm, including:

[0096] Interact with the equipment management component to obtain equipment traceability monitoring data;

[0097] Identify equipment deviation characteristics according to the equipment control plan and the equipment traceability monitoring data to generate equipment deviation identification data;

[0098] Identify the impact of humidity correlation on the inflow gas according to the device offset identification data, and generate a device offset humidity correlation impact coefficient;

[0099] When the device offset humidity correlation influence coefficient is greater than / equal to the device offset humidity correlation influence threshold, the inflow gas humidity correction is performed according to the device offset identification data.

[0100] Optionally, as the use time increases, the performance of the target ventilator may change or shift, and the device shift feature identification is performed according to the device control plan and the device traceability monitoring data to obtain the device shift identification data. Among them, the device traceability monitoring data refers to the continuous monitoring data of the device status during the implementation of the device control plan.

[0101] Specifically, the humidity traceability and control component interacts with the equipment management component to obtain continuous monitoring data on the equipment status during the implementation of the equipment control plan.

[0102] Optionally, analyze the equipment control scheme and equipment traceability monitoring data, identify the deviation characteristics of the equipment state, and generate equipment deviation identification data, which is characterized by the change in equipment operating parameters or working performance.

[0103] Optionally, by analyzing the device offset identification data, the influence of the inflow gas humidity on the device state is identified, and the device offset humidity correlation influence coefficient is generated. The influence characteristics of the inflow gas humidity on the device state are obtained based on a statistical analysis method, and illustratively, the influence characteristics of the inflow gas humidity on the device state are represented as a mathematical or machine learning model.

[0104] Optionally, when the device offset humidity correlation influence coefficient is greater than / equal to the device offset humidity correlation influence threshold, the humidity of the inflowing gas is adjusted according to the device offset identification data, and a compensating gas humidity adjustment scheme is generated to reduce or eliminate the adverse effects on the device. The device offset humidity correlation influence threshold is a preset threshold used to determine whether the device offset humidity correlation influence coefficient needs to be corrected for the inflowing gas humidity.

[0105] Through this process, the humidity traceability and control component, based on the embedded humidity traceability and control algorithm, traces and optimizes the equipment status according to the equipment control plan and equipment traceability monitoring data, which helps to improve the accuracy and effectiveness of ventilator control.

[0106] In some embodiments, the humidity traceability and control component is also used to predict abnormal risks based on the equipment traceability monitoring data, generate an equipment abnormal risk coefficient, and generate an abnormal risk alarm signal when the equipment abnormal risk coefficient is greater than / equal to an abnormal risk threshold.

[0107] Optionally, the humidity traceability control component predicts abnormal risk based on the preset deviation warning constraint and the device deviation humidity correlation influence coefficient. The deviation warning constraint includes a mutation constraint and a cumulative constraint. The mutation constraint is characterized by a higher first deviation threshold; the cumulative constraint is characterized by a lower second deviation threshold and the associated deviation duration.

[0108] Exemplarily, when the equipment offset humidity correlation influence coefficient is greater than / equal to the second deviation threshold and the duration exceeds the deviation duration or the equipment offset humidity correlation influence coefficient is greater than / equal to the first deviation threshold, a corresponding equipment abnormal risk coefficient is generated, and a graded abnormal risk alarm is issued based on the abnormal risk threshold.

[0109] In summary, the inflow gas humidity control system based on intelligent devices provided by the present invention has the following technical effects:

[0110] By setting up the user management component, the user's basic information and associated feature data, including physical signs, environment and activity features, are received. The device management component receives the basic feature data of the ventilator and associates it with the user data. The initial control analysis component generates an initial control decision based on the user and device data. The inflow gas humidity enhancement component enhances the humidity of the initial control decision. The humidity traceability control component traces and optimizes the control scheme according to the embedded algorithm. This achieves the technical effect of rapid adjustment response and refined control.

[0111] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An inflow gas humidity control system based on intelligent devices, characterized in that: The system comprises: A user management component, the user management component is used to receive basic user information and a user-related feature data set of a target user, wherein the user-related feature data set includes user vital sign feature data, user environment feature data and user activity feature information; A device management component, the device management component is used to receive basic device characteristic data of a target ventilator, the target ventilator having a corresponding relationship with the target user; An initial control analysis component, which performs initial ventilator control analysis based on the user basic information, the user-related feature data set, and the device basic feature data to generate an initial device control decision; An inflow gas humidity enhancement component, the inflow gas humidity enhancement component is used to enhance the inflow gas humidity of the initial equipment control decision to generate an equipment control plan; A humidity tracing and control component, which performs tracing and optimization control on the equipment control scheme based on an embedded humidity tracing and control algorithm; The humidity tracing and controlling component performs tracing and optimizing control on the equipment control scheme based on the built-in humidity tracing and controlling algorithm, including: Interacting with the user management component to obtain continuously monitored user feature data, wherein the continuously monitored user feature data includes continuously monitored user vital sign feature data, continuously monitored user environment feature data, and continuously monitored user activity feature information; Performing user feature change analysis based on the continuously monitored user feature data and the user-associated feature data set to generate user feature change data; Identify the impact of humidity correlation on inflow gas according to the user characteristic change data, and generate a user change correlation characteristic impact coefficient; When the user change associated characteristic influence coefficient is greater than / equal to the user change associated characteristic influence threshold, humidity compensation of the inflow gas is performed according to the user characteristic change data, and a compensation gas humidity adjustment scheme is generated; Interact with the equipment management component to obtain equipment traceability monitoring data; Identify equipment deviation characteristics according to the equipment control plan and the equipment traceability monitoring data to generate equipment deviation identification data; Identify the impact of humidity correlation on the inflow gas according to the device offset identification data, and generate a device offset humidity correlation impact coefficient; When the device offset humidity correlation influence coefficient is greater than / equal to the device offset humidity correlation influence threshold, the inflow gas humidity correction is performed according to the device offset identification data.

2. The system according to claim 1, characterized in that The device management component is also used to: Receiving equipment monitoring data of the target ventilator; Transmitting the device monitoring data to a device status evaluation unit in the device management component to generate a device status evaluation coefficient; Inputting the device state evaluation coefficient into a device state verification unit in the device management component, wherein the device state verification unit includes a first device state evaluation threshold and a second device state evaluation threshold, and the first device state evaluation threshold is greater than the second device state evaluation threshold; If the device state evaluation coefficient is greater than / equal to the first device state evaluation threshold, generating a control analysis signal, and activating the initial control analysis component according to the control analysis signal; If the device state evaluation coefficient is less than the first device state evaluation threshold, and the device state evaluation coefficient is greater than the second device state evaluation threshold, generating a device operation and maintenance signal; If the device state evaluation coefficient is less than / equal to the second device state evaluation threshold, a redundant device switching signal is generated.

3. The system according to claim 1, characterized in that The initial control analysis component is embedded with a decision function unit built based on a neural network, and performs initial ventilator control analysis based on the user basic information, the user-related feature data set, and the device basic feature data to generate an initial device control decision, including: Loading a sample control scenario data set, wherein the sample control scenario data set includes a plurality of sample control scenario data groups that meet a predetermined capacity constraint, and each sample control scenario data group includes sample user basic information, a sample user associated feature data set, and sample device basic feature data; Performing confidence control decision sample mining according to the sample control scenario data set to generate a confidence control decision sample set; Taking the sample control scene data set as input information and the confident control decision sample set as output information, supervised training is performed on a predetermined base learner, when the control decision error coefficient is less than the control decision error threshold for a preset number of consecutive times; The user basic information, the user-related feature data set and the device basic feature data are input into the decision function unit to obtain the initial device regulation decision.

4. The system according to claim 3, characterized in that Confidence control decision sample mining is performed according to the sample control scenario data set to generate a confidence control decision sample set, including: Extracting a first sample control scene data group according to the sample control scene data set, wherein the first sample control scene data group is any one of the sample control scene data sets; Taking the first sample control scenario data set as a feature retrieval constraint and taking the sample ventilator control scheme as a feature retrieval target; Perform ventilator control record feature retrieval according to the feature retrieval constraint and the feature retrieval target to generate a first sample ventilator control scheme set; Perform support calculation on each sample solution in the first sample ventilator control solution set to generate a sample solution support calculation result; Perform confidence calculation according to the sample solution support calculation result to generate a sample solution confidence calculation result; Screening the winning sample solution confidence according to the sample solution confidence calculation result, and matching the first confidence sample ventilator control solution with the first sample ventilator control solution set; The first confident sample ventilator control scheme is added to the confident regulation decision sample set.

5. The system according to claim 1, characterized in that The inflow gas humidity enhancement component is used to enhance the inflow gas humidity of the initial equipment control decision to generate an equipment control plan, including: The inflow gas humidity enhancement component includes a gas humidity feature record library and a feature twin recognition unit, wherein the gas humidity feature record library includes multiple groups of humidity feature records, each group of humidity feature records includes historical user basic information, historical user associated feature data sets, historical equipment control decisions and historical inflow gas humidity information; According to the feature twin recognition unit, the associated inflow gas humidity recognition is performed on the gas humidity feature record library in combination with the user basic information, the user associated feature data set and the initial equipment control decision to generate an associated inflow gas humidity set; Calculating the central value of the associated inflow gas humidity set to generate a confident inflow gas humidity; The initial equipment control decision is optimized according to the confident inflow gas humidity to obtain the equipment control solution.

6. The system according to claim 5, characterized in that According to the feature twin recognition unit, the associated inflow gas humidity recognition is performed on the gas humidity feature record library in combination with the user basic information, the user associated feature data set and the initial equipment control decision to generate an associated inflow gas humidity set, including: Extracting a first group of humidity feature records according to the gas humidity feature record library, wherein the first group of humidity feature records is any group of humidity feature records in the gas humidity feature record library; Adding the first historical user basic information, the first historical user associated feature data set, and the first historical equipment control decision in the first set of humidity feature records to the first sample feature data set; Adding the user basic information, the user-associated feature data set, and the initial device control decision to a baseline feature data set; Inputting the first sample feature data set and the reference feature data set into the feature twin recognition unit to obtain a first feature twin recognition index; If the first feature twin identification index satisfies the preset feature twin identification constraint, the first historical inflow gas humidity information in the first group of humidity feature records is added to the associated inflow gas humidity set.

7. The system according to claim 1, characterized in that The humidity traceability control component is also used to predict abnormal risks based on the equipment traceability monitoring data, generate an equipment abnormal risk coefficient, and generate an abnormal risk alarm signal when the equipment abnormal risk coefficient is greater than / equal to an abnormal risk threshold.

Images