Negative pressure ward seven-constant-parameter intelligent coordinated regulation and control system based on medical internet of things

Abstract

The invention relates to a negative pressure ward seven-constant parameter intelligent coordinated regulation and control system based on the medical internet of things, and the system is characterized in that the system comprises an information monitoring module, a performance analysis module, a demand prediction module, a signal management module, a medical data interaction module, and an execution mechanism; the information monitoring module collects environment data and operation data through a multi-parameter sensing unit. And the performance analysis module constructs a regional performance coefficient and a system performance index, evaluates constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise and constant illumination states in real time, and is used for realizing high-precision cooperative control of temperature, humidity, pressure difference, cleanliness, fresh air volume, noise and illumination and dynamic adaptation of a medical process.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for medical environments, and in particular to an intelligent collaborative control system for seven constant parameters in a negative pressure ward based on the Internet of Things in medicine. Background Technology

[0002] As a core setting for treating patients with infectious diseases, the safety, stability, and energy efficiency management of the indoor environment in hospital negative pressure wards are of paramount importance.

[0003] Currently, negative pressure wards face three major problems: First, there is the issue of multi-parameter decoupling and loss of control. Key parameters such as temperature, humidity, pressure difference, and cleanliness are controlled independently, with temperature fluctuations exceeding 5°C and pressure difference deviations exceeding ±3Pa, making it difficult to meet the requirements of pressure difference stability (±1Pa) and humidity range (30%-60% RH) specified in GB 50849-2023. Second, there is high energy consumption, with air conditioning generally relying on high air exchange rates (≥12 times / h), accounting for 40%-50% of the total energy consumption of the ward, and lacking a dynamic optimization mechanism. Third, there is insufficient medical adaptation, with no deep integration with the medical information system, making it unable to respond to events such as the start and stop of computed tomography (CT) equipment, resulting in a disconnect between environmental control and medical processes, and requiring manual intervention as frequently as 2.3 times per day in the ward.

[0004] Driven by policies such as the "Three-Year Action Plan for the Construction of New Internet of Things Infrastructure," smart healthcare scenarios are urgently demanding intelligent environments. There is a pressing need to develop a seven-constant parameter collaborative control system that integrates medical IoT to achieve coordinated control of seven constant parameters: temperature, humidity, pressure difference, cleanliness, fresh air volume, noise, and light. This would significantly improve energy efficiency and adaptability to medical processes while ensuring infection control safety.

[0005] The seven constant parameters refer to constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise, and constant light intensity.

[0006] The Layer 7 Protocol for Health Information Exchange (HIPE) is an international standard protocol for exchanging healthcare information. It establishes an international standard for the electronic exchange of clinical and administrative data between different software applications in the healthcare field. This standardized health information transmission protocol facilitates electronic transmission between different applications in the medical field. HIPE brings together standard formats used by different vendors to design interfaces between application software, allowing various healthcare institutions to exchange data between heterogeneous systems.

[0007] Therefore, overcoming the aforementioned shortcomings has become an important issue that urgently needs to be addressed by those skilled in the art. Summary of the Invention

[0008] This invention overcomes the shortcomings of the above-mentioned technologies and provides a seven-constant parameter intelligent collaborative control system for negative pressure wards based on the Internet of Things in healthcare. It achieves rapid pressure difference recovery time, reduced risk of pathogen spread, rapid response to medical events, and significantly improved aerosol capture efficiency during nebulization therapy. It also reduces energy consumption and maintains seven constant parameters: constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise, and constant light intensity.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: A smart collaborative control system for seven constant parameters in a negative pressure ward based on the Internet of Things in healthcare includes: a smart ward platform, an information monitoring module, a performance analysis module, a demand forecasting module, an information management module, a medical data interaction module, and an actuator. The information monitoring module collects environmental and operational data through multi-parameter sensing units. The performance analysis module constructs regional performance coefficients and system performance indices, and evaluates the status of the seven constant parameters (temperature, humidity, pressure, cleanliness, fresh air volume, noise, and illumination) in real time. The demand forecasting module predicts environmental load based on a long short-term memory network model and generates an electricity demand coefficient by combining internal and external environmental factor correlation functions. The information management module collects data from the demand forecasting module and the medical data interaction module and issues instructions. The actuator receives instructions from the information management module and performs real-time and predictive control. The medical data interaction module connects to the smart ward platform through a health information exchange layer 7 protocol, responds to patient temperature events, and initiates an environmental reset protocol to maintain the seven constant parameters (temperature, humidity, pressure, cleanliness, fresh air volume, noise, and illumination).

[0010] Preferably, as described above, the information monitoring module includes a multi-parameter sensing unit and an edge computing node deployed in the negative pressure ward. It deploys multiple types of sensors to collect real-time environmental parameters and system operation information inside and outside the negative pressure ward, and preprocesses and encrypts the raw data through the edge computing node.

[0011] Preferably, the performance analysis module described above constructs a multi-dimensional performance evaluation model, including regional performance evaluation and system performance evaluation, to evaluate the performance of the seven constant systems of the entire negative pressure ward: constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise, and constant light.

[0012] Preferably, the demand forecasting module described above includes a Long Short-Term Memory (LSTM) network model prediction unit and a correlation influence function library. The LTM network model is a deep learning model specifically designed for processing time series data and capturing its long-term dependencies. It is used for environmental parameter demand forecasting and equipment fault early warning. The LTM network model prediction unit takes meteorological platform data, historical energy consumption data, and treatment plans as input and outputs predicted environmental parameter values ​​for future periods. The correlation influence function library contains correlation functions for external environmental factors and internal environmental factors to generate electricity demand forecasting coefficients.

[0013] Preferably, as described above, the information management module includes a real-time control unit and a predictive control unit. In the real-time control unit, when the calculated value of the internal environmental factor function exceeds the second-level threshold, it indicates that the internal environmental demand of the ward is changing drastically, and the system triggers a level II regional control signal; when the calculated value of the external environmental factor function exceeds the first-level threshold, it indicates that external climate conditions are causing significant interference to the system, and the system triggers a level I system control signal to adjust the ventilation strategy of the entire ward; the predictive control unit predicts the distribution of pollutants based on a convolutional neural network aerosol diffusion model and adjusts the exhaust direction in advance; it predicts the probability of equipment failure based on a long short-term memory network model and generates a maintenance early warning signal.

[0014] Preferably, the medical data interaction module described above is interconnected with the smart ward platform through the health information exchange layer 7 protocol to solve the problem of medical data silos and obtain patients' vital signs data, medical staff's work trajectories and medical equipment operating status in real time.

[0015] Preferably, the negative pressure ward's internal and external environmental parameters, as described above, include external factor parameters and internal factor parameters, and the operational information includes energy consumption parameters and mode parameters.

[0016] Preferably, as described above, the regional performance assessment calculates the regional operating energy consumption index based on energy consumption parameters, constructs the regional operating status index through model parameters, comprehensively obtains the regional operating performance coefficient, and calculates the system performance index by weighting the operating performance coefficients of each region, thereby realizing real-time assessment of the operating status of the entire negative pressure ward system.

[0017] Preferably, the system performance index is calculated by weighting the operating performance coefficients of each region as described above. It is a comprehensive assessment of the performance of all areas, used to reflect the overall effectiveness of the entire ward system. The calculation formula is as follows: ); ; ; ; in, It is the regional operating performance coefficient, which aims to balance the operating status and energy consumption economy of a single region; This refers to the total number of areas within a negative pressure isolation ward. It is the regional weighting coefficient. The standardization factor is RUNa, which is the regional operating status index; ENGa is the regional operating energy consumption index; and Pa is the actual power. It is the rated power. λ1 is the weighting coefficient for the operating status, and λ2 is the weighting coefficient for energy consumption.

[0018] Preferably, the prediction and control unit described above monitors aerosol concentration using a laser particle counter, and combines this with data on temperature, humidity, wind speed, and personnel location to use a convolutional neural network model to predict the distribution of pollutants in the breathing zone and adjust the exhaust confluence direction in advance; it also analyzes the fan bearing temperature and motor current spectrum to use a long short-term memory network model to predict the probability of filter clogging.

[0019] Compared with the prior art, the beneficial effects of the present invention are: 1. Through the above-mentioned intelligent collaborative control system for seven constant parameters in negative pressure wards based on the Internet of Things in healthcare, the environmental control is refined, and the internal and external environmental parameters of the ward are monitored and analyzed in real time. The accuracy of the collaborative control of the seven constant parameters is significantly improved. By constructing a pressure difference-pathogen diffusion coupling model (V=λ·exp(-μ·|ΔP|)), the pressure difference recovery time after a sudden disturbance (such as the opening of the access control) is reduced from more than 120 seconds in the traditional system to 15 seconds, which meets the standard of less than 60 seconds specified in GB 50849-2023, effectively blocking the path of pathogen diffusion. The system adopts a radiant capillary network + dual cold source dehumidification technology to achieve a temperature control accuracy of ±0.5℃ and a humidity fluctuation of less than ±5%RH, avoiding patient discomfort caused by a temperature difference of more than 5℃ in summer in the traditional system.

[0020] 2. This invention provides a seven-constant parameter intelligent collaborative control system for negative pressure wards based on the Internet of Things in healthcare. It interconnects with a smart ward platform through a seventh-layer health information exchange protocol, responding to medical events in real time. When a patient's body temperature exceeds 38.5℃, it automatically triggers an environmental reset protocol (temperature 24±1℃, fresh air volume ≥40m³ / (h·person)). During nebulization therapy, it activates directional airflow organization (bedside exhaust + top-down air supply), achieving a local pressure difference greater than 10Pa and improving aerosol capture efficiency by 18%. Based on the pre-booking time of CT / DR equipment, it compensates for heat load in advance, suppressing temperature and humidity fluctuations in the equipment area within ±0.8℃, achieving deep adaptation to the medical scenario. It introduces energy consumption assessment and power demand prediction models to optimize system operation modes and power resource allocation, reducing system energy consumption, decreasing hospital operating costs, and achieving high-efficiency energy utilization.

[0021] 3. The present invention provides a seven-constant parameter intelligent collaborative control system for negative pressure wards based on the Internet of Things in healthcare. It uses a long short-term memory network prediction model to dynamically adjust the ventilation rate: during idle periods, it reduces the ventilation rate from 12 times / h to 6-8 times / h, with a comprehensive energy saving rate of more than 32%. It uses a convolutional neural network aerosol distribution model to predict and implement local pressurization, reducing the demand for high ventilation throughout the entire area while maintaining a fine particulate matter density of ≤35μg / m³, thereby reducing annual carbon emissions by 28 tons / ward, achieving synergistic optimization of energy consumption and infection control.

[0022] 4. The present invention provides a seven-constant parameter intelligent collaborative control system for negative pressure wards based on the Internet of Things in healthcare. Based on a long short-term memory network fault early warning model based on vibration spectrum analysis, it can identify the risk of filter blockage 7 days in advance (accuracy greater than 90%), reducing maintenance costs by 30%. The edge computing nodes realize 256-bit encrypted transmission of data according to the advanced encryption standard, which meets the requirements of the "Medical and Health Data Security Specification", reducing the risk of leakage by 95% and enabling intelligent upgrade of operation and maintenance. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the workflow of the present invention.

[0024] Figure 2 This is a schematic diagram of the control flow for the medical data interaction module. Detailed Implementation

[0025] The following examples further illustrate the features of the present invention and other related features in detail, so as to facilitate understanding by those skilled in the art.

[0026] In this embodiment, it should be understood that the terms "middle," "upper," "lower," "top," "right side," "left end," "above," "back," "center," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0027] like Figure 1 As shown, the overall architecture follows the closed-loop control principle of intelligent systems, and is divided into five modules according to data flow and function: information monitoring, performance analysis, demand forecasting, signal management, and medical data interaction. Each module is clearly hierarchical and logically coupled, together forming an integrated intelligent control closed loop of "perception-assessment-prediction-decision-execution". The specific implementation is as follows: The system's data foundation originates from the information monitoring module, which acts as the "sensing layer" deployed in the negative pressure ward. This module includes multi-parameter sensing units and edge computing nodes. The multi-parameter sensing units consist of a temperature and humidity sensor (model SHT35, accuracy ±0.2℃), a micro differential pressure transmitter (model MPM480, range ±50Pa, accuracy ±0.5%), a fine particulate matter laser sensor (model PMS5003, range 0-1000μg / m³), a carbon dioxide sensor (SenseAir S8, ±30ppm), an airflow meter (error ±2%FS), a sound level meter (model TES-1358, range 30-130dB, accuracy ±1.5dB), an illuminance sensor (model BH1750, range 0-65535lux), and a current transformer (CT-0.5 class). These units are responsible for real-time acquisition of environmental parameters (such as temperature, humidity, differential pressure, and particulate matter concentration) and equipment operating data (such as current, voltage, and airflow) within and outside the ward. Edge computing nodes perform preliminary preprocessing on the raw data, including smoothing data fluctuations using the Kalman filter algorithm, performing data normalization, transmitting the data to the upper layer after adding timestamps and regional identifiers through a 256-bit encryption channel of the Advanced Encryption Standard, thus providing the system with a real-time and reliable data source.

[0028] The data is then fed into the performance analysis module (evaluation layer). This module calculates the regional operating energy consumption index based on energy consumption parameters (current, voltage, and power), and constructs a regional operating status index based on mode parameters (such as cooling temperature difference and fresh air volume) and environmental parameters. Finally, it calculates the regional operating performance coefficient (PerA) and the system performance index (PerH) using a comprehensive formula. When the index falls below a preset threshold, an early warning message is generated (e.g., "Area 3 energy consumption is abnormal; it is recommended to check the fresh air valve opening") and pushed to maintenance personnel through the platform, achieving transparent performance management and early fault detection.

[0029] The collected data is encrypted and transmitted to the intelligent analysis module, where the performance analysis module performs dynamic evaluation. The performance analysis module calls pre-stored compensation functions to calculate the regional operating energy consumption index based on energy consumption parameters (current, voltage, and power). It also constructs a regional operating status index using mode parameters (cooling temperature difference, heating temperature difference, dehumidification fan speed, fresh air volume, noise weighting, and light compensation), ultimately obtaining the regional operating performance coefficient. System performance evaluation is then performed by weighting the operating performance coefficients of each region to calculate the system performance index, enabling real-time assessment of the entire negative pressure ward system's operating status.

[0030] The system then enters the intelligent decision-making phase, which is completed collaboratively by the demand forecasting and signal management modules. The demand forecasting module ("forecasting layer") receives performance evaluation results, external meteorological data, and hospital treatment plans. It uses a long short-term memory network model to predict environmental load and equipment failures, and generates power demand forecasting coefficients by calculating the internal and external environmental factor function (Fin / Fou). The signal management module ("decision-making layer") makes judgments based on this. Its real-time control unit compares the Fin / Fou function value with the set thresholds (H1, H2) and triggers different levels of control signals. The forecasting and control unit uses a convolutional neural network aerosol diffusion model to predict and adjust ventilation strategies or generate maintenance work orders in advance, realizing the transition from passive response to active intervention.

[0031] Ultimately, the decision-making instructions are implemented through the medical data interaction module and the execution equipment. This module serves as the "application layer" interface. It interconnects with the hospital's smart ward platform via the health information exchange layer 7 protocol to obtain real-time patient vital signs (such as body temperature), medical equipment status, and treatment schedules. When medical events such as "patient body temperature > 38.5℃" or "nebulization therapy started" are detected, they are converted into high-priority control instructions (such as initiating an environmental reset protocol) and sent to the signal management module for execution. The final instructions (such as adjusting the fresh air volume, changing the water temperature, or switching lights on and off) are sent to the end-point execution devices such as variable frequency fresh air units, radiant capillary networks, air valves, and lighting fixtures to achieve precise environmental control.

[0032] The system forms a complete closed-loop feedback loop. The actuator's actions change the environmental state of the ward, and the new data is collected again by the information monitoring module, initiating a new cycle of "perception-assessment-prediction-decision-execution". This achieves continuous self-optimization and high-precision, low-energy-consumption coordinated control of seven constant parameters.

[0033] like Figure 2 As shown, the control method of the medical data interaction module of the present invention adopts a dual-path parallel architecture, and the specific implementation process is as follows: The system continuously monitors data. When the access control sensor (model: AH49E) detects that the ward door is open, the medical event response path is triggered. The edge computing node reads the real-time differential pressure value in parallel, retrieves the historical differential pressure fluctuation data of the ward over the past hour (sampling frequency 1Hz), and simultaneously acquires indoor temperature and humidity, personnel density calculated by the ultra-wideband positioning system, and fine particulate matter concentration monitored by the laser particle counter (accuracy ±10%). Based on the historical and real-time data, the edge computing node performs fusion analysis to generate a dynamic compensation curve for the fan speed: N(t) = N0 + 0.2·e^(-0.5t)·sin(2πt), where N0 is the current speed and t is the time variable. The signal management module sends this to the variable frequency fresh air unit (response time ≤3 seconds) for execution. The system continuously monitors the feedback from the differential pressure sensor to ensure that the differential pressure recovers to the set range (-5Pa ±1Pa) within 15 seconds.

[0034] When there are no high-priority medical events, the system performs routine checks, such as automatically checking all seven constant environmental parameters every 5 minutes. If the parameters meet the standards, the system optimizes the ventilation rate (e.g., reducing it to 6-8 times / hour) based on a long short-term memory network model to achieve energy savings; if the standards are not met, the system initiates a collaborative control strategy using internal environmental factor functions (Fin) and external environmental factor functions (Fou), predicting the load and adjusting the actuators through the long short-term memory network model. All data is encrypted and transmitted back to the smart ward platform, forming a closed loop of "data collection, intelligent analysis, precise control, and feedback optimization".

[0035] Furthermore, this invention achieves scenario-based collaborative control through deep integration with the medical process via the seventh-layer health information exchange protocol. For example, when a patient's body temperature is detected to exceed 38.5℃, the temperature control protocol is automatically triggered, stabilizing the room temperature from 26℃ to 24±0.3℃ within 30 minutes. The fresh air volume is dynamically increased from 30m³ / h to no less than 42m³ / h, and the enhanced operation mode of the high-efficiency air filter is activated in conjunction with this. In actual testing, the pathogen capture efficiency in this mode can reach 92.5% (compared to 78% in the conventional mode), while ensuring that the noise level is controlled below 38 decibels.

[0036] For example, based on the computed tomography (CT) scan equipment reservation information from the smart ward platform, the system pre-cools the equipment room to 22°C one hour in advance to compensate for the heat load generated during equipment operation, and temporarily reduces the fresh air volume of adjacent wards by 18% to maintain the overall energy balance of the system. Regarding light and shadow control, during postoperative recovery, the system automatically activates a silent mode (≤35 dB) and a low-light mode (200 lux illuminance); during endoscopic examinations, the local illuminance of the examination area is increased to 500 lux and the background light source is turned off.

[0037] The medical data interaction module enables patient status response. When the patient's body temperature exceeds 38.5℃, it automatically sets the temperature to 24±1℃, the fresh air volume to no less than 40m³ / (h·person), and activates the high-efficiency air filter enhancement mode. For medical process adaptation, during nebulization therapy, it activates the "bedside directional exhaust + top-down airflow" mode, with a local pressure difference greater than 10Pa. For equipment collaborative control, it compensates for temperature and humidity fluctuations caused by the heat load of the computed tomography / digital X-ray imaging equipment based on the scheduled time. When the patient is in the postoperative recovery period, it automatically activates the silent mode and low-light mode, with a volume no greater than 35 decibels and a light intensity of 200 lux. During endoscopic examinations, it increases the local illuminance to 500 lux and turns off the background light source.

[0038] The above are merely typical embodiments of the present invention. In addition, the present invention may have many other specific implementations. All technical solutions formed by equivalent substitution or equivalent transformation fall within the scope of protection claimed by the present invention.

Claims

1. A medical Internet of Things-based negative pressure ward seven constant parameter intelligent collaborative regulation system, characterized in that, The application relates to a smart ward platform, an information monitoring module, a performance analysis module, a demand prediction module, an information management module, a medical data interaction module and an execution mechanism. The information monitoring module collects environmental data and operation data through a multi-parameter sensing unit; the performance analysis module constructs regional performance coefficients and system performance indexes and evaluates seven constant parameter states of constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise and constant illumination in real time; the demand prediction module predicts environmental load based on a long short-term memory network model, generates a power demand coefficient by combining an internal environmental factor correlation function and an external environmental factor correlation function; the information management module collects data fed back by the demand prediction module and the medical data interaction module and gives instructions; the execution mechanism receives the instructions given by the information management module and performs real-time regulation and prediction regulation; and the medical data interaction module connects the smart ward platform through a health information exchange seventh layer protocol, responds to a patient temperature event and starts an environmental reset protocol to maintain the seven indexes of constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise and constant illumination.

2. The medical Internet of Things-based negative pressure ward seven-constant parameter intelligent collaborative regulation system according to claim 1, characterized in that: The information monitoring module comprises a multi-parameter sensing unit and an edge computing node arranged in a negative pressure ward, a plurality of types of sensors are arranged, external environmental parameters and system operation information in the negative pressure ward are collected in real time, and the original data are preprocessed and encrypted and transmitted through the edge computing node.

3. The medical Internet of Things-based negative pressure ward seven-constant parameter intelligent collaborative regulation system according to claim 1, characterized in that: The performance analysis module constructs a multi-dimensional performance evaluation model, including regional performance evaluation and system performance evaluation, and evaluates the seven constant system performances of constant temperature, constant humidity, constant pressure, constant cleanliness, constant fresh air volume, constant noise and constant illumination in the entire negative pressure ward.

4. The medical Internet of Things-based negative pressure ward seven-constant parameter intelligent collaborative regulation system according to claim 1, characterized in that: The demand prediction module comprises a long short-term memory network model prediction unit and a correlation influence function library, the long short-term memory network model is a deep learning model specially used for processing time series data and capable of capturing long-term dependence relationship of the time series data, is used for environmental parameter demand prediction and device fault early warning, meteorological platform data, historical energy consumption data and diagnosis and treatment plans are input into the long short-term memory network model prediction unit, and environmental parameter prediction values in a future period are output; the correlation influence function library comprises an external environmental factor correlation function and an internal environmental factor correlation function, and generates a power demand prediction coefficient.

5. The medical Internet of Things-based negative pressure ward seven constant parameter intelligent collaborative regulation system according to claim 1, characterized in that: The information management module comprises a real-time regulation unit and a prediction regulation unit, in the real-time regulation unit, when a calculation value of the internal environmental factor function exceeds a second threshold value, it is indicated that the internal environmental demand of the ward changes dramatically, and the system triggers a II-level regional regulation signal; when a calculation value of the external environmental factor function exceeds a first threshold value, it is indicated that external climate conditions significantly interfere with the system, and the system triggers a I-level system regulation signal to adjust a ventilation strategy of the entire ward; The prediction regulation unit predicts pollutant distribution based on a convolutional neural network aerosol diffusion model and adjusts an exhaust direction in advance; The long short-term memory network model is used for predicting a device fault probability and generating a maintenance early warning signal.

6. The medical Internet of Things-based negative pressure ward seven-constant parameter intelligent collaborative regulation system according to claim 1, characterized in that: The medical data interaction module is interconnected with the smart ward platform through a health information exchange seventh layer protocol, solves a medical data island problem, and acquires patient vital sign data, medical staff work tracks and medical device operation states in real time.

7. The medical Internet of Things-based negative pressure ward seven constant parameter intelligent collaborative regulation system according to claim 2, characterized in that: The external environment parameters of the negative pressure ward include external factor parameters and internal factor parameters, and the operation information includes energy consumption parameters and mode parameters.

8. The medical Internet of Things-based negative pressure ward seven constant parameter intelligent collaborative regulation system according to claim 3, characterized in that: The regional performance evaluation is based on the calculation of the regional operation energy consumption index based on the energy consumption parameters, the construction of the regional operation state index through the mode parameters, the comprehensive obtaining of the regional operation performance coefficient, the weighted calculation of the system performance index through the regional operation performance coefficients, and the real-time evaluation of the entire negative pressure ward system operation state.

9. The medical Internet of Things-based negative pressure ward seven constant parameter intelligent collaborative regulation system according to claim 8, characterized in that: The system performance index is weighted calculated by the performance coefficients of the regions , which is a comprehensive evaluation of all the region performances and is used to reflect the overall performance of the entire ward system, The calculation formula is as follows: )； ； ； ； wherein, is the zone operating performance coefficient, aimed at balancing the operating state of the individual zone with the energy consumption economy; is the total number of zones in the negative pressure ward, is the zone weight coefficient, is the standardization factor, RUNa is the zone operating state index, ENGa is the zone operating energy consumption index, Pa is the actual power, is the rated power, is the operating state weight coefficient, and λ2 is the energy consumption weight coefficient.

10. The medical Internet of Things-based negative pressure ward seven-constant parameter intelligent collaborative regulation system according to claim 5, characterized in that: The prediction and regulation unit monitors the aerosol concentration through a laser particle counter, combines with the temperature and humidity, wind speed, and personnel position data, adopts a convolutional neural network model to predict the respiratory area pollutant distribution, and adjusts the exhaust flow direction in advance; analyzes the fan bearing temperature and motor current spectrum, adopts a long short-term memory network model to predict the filter screen blockage probability.

Images