Control method of indoor air purifier based on Internet of Things

Abstract

The invention relates to the technical field of air purifiers of the Internet of Things, and discloses a control method of an indoor air purifier based on the Internet of Things, which comprises the following steps: step 1, acquiring indoor air quality parameters and equipment running states through a sensor on the air purifier to obtain state vectors containing multiple variables, the state vector comprises PM2.5 concentration, carbon dioxide concentration, total volatile organic compound concentration, humidity, filter element accumulated use time and equipment operation noise, and the state vector is used for describing information of air quality and equipment operation state; and 2, according to the state vector, control variables of the air purifier are defined, and the control variables comprise the air speed and the operation mode. Through combined monitoring of multiple sensors, indoor air quality parameters and equipment operation states are collected, construction of multivariable state vectors is achieved, comprehensive and accurate environment and equipment operation data are obtained, support is provided for follow-up optimization control, and the environment sensing capacity of the air purifier is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things air purifiers, and in particular to a control method for an indoor air purifier based on the Internet of Things. Background Art

[0002] The global environmental pollution problem is becoming increasingly serious, especially with the acceleration of industrialization and urbanization, outdoor and indoor air pollution has become one of the important factors threatening human health. The problem of indoor air pollution is particularly prominent. Due to insufficient ventilation, the use of chemical materials, and the infiltration of outdoor pollutants in modern homes and offices, the air quality is poor and even higher than the outdoor pollution level for a long time. Studies have shown that pollutants such as PM2.5, carbon dioxide, total volatile organic compounds, and formaldehyde in indoor air can pose a serious threat to the human respiratory system, immune system, and even cardiovascular system, affecting the comfort and health of the human body.

[0003] With the intensification of air pollution and the improvement of people's health awareness, the application of indoor air purifiers has gradually become a focus of attention. However, existing air purifiers have many problems in technical application and are difficult to meet users' increasingly complex needs for indoor air quality management. The main manifestations are as follows:

[0004] Traditional air purifiers are usually equipped with a single type and a small number of sensors, while indoor air pollution is complex, including multiple pollutants such as PM2.5, carbon dioxide, and total volatile organic compounds, and humidity, noise, and the operating status of the equipment itself have a significant impact on the purification effect. Existing technologies cannot achieve multi-dimensional monitoring of indoor air quality and equipment operating status, resulting in the purifier being unable to accurately identify air pollution conditions, reducing purification efficiency.

[0005] Air purifiers are mostly optimized for a single goal, but in actual application scenarios, users can balance and optimize purification effect, energy consumption, filter life and operating noise according to different needs. Existing technologies lack the ability to dynamically balance multiple goals, especially in the trade-off between efficient purification and energy saving and consumption reduction, and cannot meet users' optimization needs for different scenarios.

[0006] The operation control of existing air purifiers usually adjusts the operating parameters based on real-time monitoring data. It is impossible to predict the changing trend of air quality in the future, and the equipment cannot take countermeasures in advance, resulting in delayed response and poor purification effect.

[0007] Traditional air purifiers are not smart enough and lack user interaction experience. The device operation mode is usually manually adjusted by the user, and the interactive interface cannot intuitively display air quality information and device operation status. Users cannot understand the current air conditions and the effect of the device on air quality improvement in real time. At the same time, existing technologies cannot dynamically adjust the device's operation goals according to changes in user needs and usage scenarios.

[0008] Therefore, those skilled in the art provide a control method for an indoor air purifier based on the Internet of Things to solve the above-mentioned problems. Summary of the invention

[0009] In view of the deficiencies in the prior art, the present invention provides a control method for an indoor air purifier based on the Internet of Things to solve the problems raised in the above background technology.

[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions: A control method for an indoor air purifier based on the Internet of Things, comprising:

[0011] Step 1: Collect indoor air quality parameters and equipment operation status through sensors on the air purifier to obtain a state vector containing multiple variables. The state vector includes PM2.5 concentration, carbon dioxide concentration, total volatile organic compound concentration, humidity, cumulative filter element usage time and equipment operation noise. The state vector is used to describe information about air quality and equipment operation status;

[0012] Step 2: Define the control variables of the air purifier according to the state vector. The control variables include wind speed and operation mode. The wind speed is used to control the air handling capacity of the air purifier. The operation modes include high-efficiency mode, silent mode and energy-saving mode. The control variables are used to adjust the operation parameters of the air purifier to adapt to different air states and operation requirements.

[0013] Step 3: Based on the state vector and control variables, a target optimization function is constructed to describe the optimization requirements of the air purifier operation. The target optimization function includes energy consumption cost, air quality deviation cost, filter life loss cost and noise cost. The construction of the target optimization function is used to balance among multiple optimization objectives to determine the operation strategy.

[0014] Step 4: By analyzing the historical changes of the state vector and combining the adjustment of the control variables, a state transition model is established to describe the impact of the air purifier operation on the air quality and operating status. The state transition model is used to describe the dynamic change relationship between the air quality and the equipment status over time after the air purifier is operated. The variables in the state transition model include the change in air pollutant concentration, the change in humidity, the cumulative value of the filter element usage time, and the relationship between noise and operating parameters. The state transition model provides a basic basis for environmental changes for optimal control;

[0015] Step 5: Determine the control variable value of the air purifier based on the target optimization function and the state transition model. The adjustment of the control variable is used to implement the operation strategy of the air purifier under different air conditions. The result of the optimized control is transmitted to the fan and actuator through the execution module of the equipment to complete the dynamic adjustment of the operating parameters of the air purifier.

[0016] Preferably, the target optimization function constructed in step 3 specifically comprises the following formula:

[0017] J t =α1E t +α2Q t +α3L t +α4N t ,

[0018] Among them, J t The target optimization function is used to comprehensively describe the optimization target of the air purifier operation;

[0019] E t is the energy cost, which represents the relationship between wind speed and power consumption, and is defined as E t =P t Δt,

[0020] Where P t =β1w t +β2,w t represents wind speed, β1 and β2 are wind speed power consumption coefficients, and Δt represents the time step;

[0021] Q t is the air quality deviation cost, defined as:

[0022] Q t =γ1(p t -p thr ) 2 +γ2(c t -c thr ) 2 +γ3(v t -v thr ) 2 ,

[0023] Among them, p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration;

[0024] L t is the filter life loss cost, defined as Among them, m t is the operation mode;

[0025] N t is the noise cost, defined as Among them, n t For operating noise;

[0026] α1, α2, α3, α4, γ1, γ2, γ3, δ, ∈ are the coefficients of optimization weights.

[0027] Preferably, the air quality deviation cost Q in the objective optimization function t The health thresholds in meet the following settings:

[0028] PM2.5 concentration threshold p thr 35 micrograms per cubic meter;

[0029] CO2 concentration threshold c thr 800ppm;

[0030] Total volatile organic compound concentration threshold v thr It is 0.6ppm.

[0031] Preferably, the state transition model in step 4 includes the dynamic impact of the operation of the air purifier on the air quality and the device state, including:

[0032] PM2.5 concentration change: Δp t =-η p w t +∈ p,t ,

[0033] Among them, Δp t is the change in PM2.5 concentration, η p is the purification efficiency coefficient, w t is the wind speed, ∈ p,t is random environmental noise;

[0034] Change in carbon dioxide concentration: Δc t =-η c w t +∈ c,t ,

[0035] Where Δc t is the change in carbon dioxide concentration, η c is the purification efficiency coefficient, w t is the wind speed, ∈ c,t is the carbon dioxide environmental noise;

[0036] Total volatile organic compound concentration change: Δv t =-η v w t +∈ v,t ,

[0037] Where Δv t is the change in total volatile organic compound concentration, η v is the purification efficiency coefficient, w t is the wind speed, ∈ v,t is the total volatile organic compound ambient noise;

[0038] Humidity change: Δh t =η h -ξ h w t +∈ h,t ,

[0039] Where Δh t is the humidity change, η h and h is the coefficient of humidity increase and humidity decrease, w t is the wind speed, ∈ h,t is the humidity environmental noise;

[0040] Cumulative change in filter element usage time: Δl t =Δt,

[0041] Among them, Δl t is the change in the cumulative use time of the filter element, and Δt is the time step;

[0042] Running noise: n t =f(w t ,m t ),

[0043] Among them, n t is the operating noise, f(w t ,m t ) represents the functional relationship between wind speed and operation mode and noise, w t is the wind speed, m t is the running mode.

[0044] Preferably, the optimization control of the air purifier is accelerated by a reinforcement learning model, the input of the reinforcement learning model is the current state vector, the output is the adjustment value of the control variable, the reinforcement learning model is trained using a Q learning algorithm, and the loss function is defined as:

[0045]

[0046] Among them, L is the loss function, r t represents the immediate reward value, γ represents the discount factor, s t+1 is the state vector at the next moment, u t+1 is the control variable at the next moment, Q(s t ,u t) means in state s t Take the control variable u t The comprehensive value of

[0047] The instant reward value r t Defined as the negative value of the objective optimization function, satisfying:

[0048] r t =-(α1E t +α2Q t +α3L t +α4N t ),

[0049] Among them, E t is the energy cost, Q t is the air quality deviation cost, L t is the filter life loss cost, N t is the noise cost, and α1, α2, α3, and α4 are the coefficients of the optimization weights.

[0050] Preferably, the air purifier realizes optimized control through the Internet of Things platform, including the following modules:

[0051] Data acquisition module, used to collect variables of the state vector;

[0052] An optimization calculation module is used to calculate the adjustment value of the control variable;

[0053] An execution module, used to adjust the wind speed and operation mode according to the control variables;

[0054] The user interaction module is used to set optimization goals and operation modes through mobile terminals and other terminal devices.

[0055] Preferably, the optimization calculation module calculates the adjustment value of the control variable based on a multi-objective optimization method, and the value ranges of the weight coefficients α1, α2, α3, and α4 in the objective function of the optimization calculation are set by the user and adaptively adjusted through historical data, wherein:

[0056] α1 represents the weight coefficient of energy consumption optimization, and its value range is [0.1, 0.9];

[0057] α2 represents the weight coefficient of air quality optimization, and its value range is [0.1, 0.9];

[0058] α3 represents the weight coefficient of filter element life optimization, and its value range is [0.1, 0.9];

[0059] α4 represents the weight coefficient of noise optimization, and its value range is [0.1, 0.9];

[0060] If the sum of weight coefficients meets the following constraints, the optimization calculation module outputs the balance optimization result:

[0061] α1+α2+α3+α4=1.

[0062] Preferably, the weight coefficient of the historical data adaptively adjusted in the optimization calculation module is based on the following formula:

[0063] in, represents the weight of the i-th optimization target at time t;

[0064] η represents the step size of weight update, satisfying η>0;

[0065] Denotes the target optimization function J t About weight α i The partial derivative of is used to measure the influence of the current optimization objective on the overall optimization result;

[0066] i∈{1,2,3,4} represents the index of different optimization objectives, including energy consumption optimization, air quality optimization, filter life optimization and noise optimization.

[0067] Preferably, the user interaction module is used to display the weight adjustment result of the optimization target in real time and dynamically set the optimization target according to the user's operation mode, wherein:

[0068] In high-efficiency mode, the weight of the optimization target satisfies α2>α1, α3, α4, giving priority to optimizing air quality;

[0069] In silent mode, the weight of the optimization target satisfies α4>α1, α2, α3, giving priority to reducing the equipment operation noise;

[0070] In energy-saving mode, the weights of the optimization objectives satisfy α1>α2, α3, α4, giving priority to reducing equipment energy consumption;

[0071] In the equilibrium mode, the weights of the optimization objectives satisfy α1=α2=α3=α4=0.25, achieving a balance among the optimization objectives.

[0072] Preferably, the air quality parameters in the operating state vector collected by the data collection module are fused and calculated according to the following weighted average formula to generate a comprehensive air quality index AQI:

[0073]

[0074] Among them, AQI t Indicates the current air quality index.

[0075] w p 、wc 、w v is the weight coefficient of PM2.5, carbon dioxide and total volatile organic compounds, satisfying w p +w c +w v =1,

[0076] p t 、c t 、v t Indicates the current concentration of PM2.5, carbon dioxide and total volatile organic compounds;

[0077] p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration;

[0078] The comprehensive air quality index AQI t It is used as the input basis for dynamically adjusting the control variables in the optimization calculation module.

[0079] The present invention provides a control method for an indoor air purifier based on the Internet of Things, which has the following beneficial effects:

[0080] 1. The present invention collects indoor air quality parameters and equipment operating status through joint monitoring of multiple sensors, realizes the construction of multi-variable state vectors, obtains comprehensive and accurate environmental and equipment operating data, provides support for subsequent optimization control, and significantly improves the environmental perception ability of the air purifier.

[0081] 2. The present invention realizes the dynamic trade-off between different operating objectives of the air purifier by constructing a target optimization function including energy consumption cost, air quality deviation cost, filter element life loss cost and noise cost, and obtains a balanced optimization effect of the equipment among efficient purification, energy saving and consumption reduction, extended filter element life and reduced noise, effectively solving the technical problem of single target and inability to take both into account in the operation of traditional equipment.

[0082] 3. The present invention establishes a state transition model to describe the dynamic relationship between the operation of the air purifier and the air quality parameters and the equipment status, and combines the reinforcement learning algorithm to optimize the control variables in real time, thereby achieving accurate prediction of the dynamic changes of air quality, achieving the effect of rapid response and precise control of the equipment, and significantly improving the adaptability and response speed of the equipment to complex air environments.

[0083] 4. The present invention realizes the dynamic setting of the operating mode and optimization target of the air purifier by the user through the combination of the Internet of Things platform and the user interaction module, obtains the precise adaptation effect of the equipment in different scenarios, meets the user's personalized needs for multi-scenario applications, and significantly improves the intelligence level of the air purifier and the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0085] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0086] The present invention is described in detail below in conjunction with the accompanying drawings:

[0087] Example:

[0088] Please see attached Figure 1 The embodiment of the present invention provides a control method of an indoor air purifier based on the Internet of Things, comprising:

[0089] Step 1: Collect indoor air quality parameters and equipment operation status through sensors on the air purifier to obtain a state vector containing multiple variables. The state vector includes PM2.5 concentration, carbon dioxide concentration, total volatile organic compound concentration, humidity, cumulative filter element usage time and equipment operation noise. The state vector is used to describe information about air quality and equipment operation status;

[0090] Step 2: Define the control variables of the air purifier according to the state vector. The control variables include wind speed and operation mode. The wind speed is used to control the air handling capacity of the air purifier. The operation modes include high-efficiency mode, silent mode and energy-saving mode. The control variables are used to adjust the operation parameters of the air purifier to adapt to different air states and operation requirements.

[0091] Step 3: Based on the state vector and control variables, a target optimization function is constructed to describe the optimization requirements of the air purifier operation. The target optimization function includes energy consumption cost, air quality deviation cost, filter life loss cost and noise cost. The construction of the target optimization function is used to balance among multiple optimization objectives to determine the operation strategy.

[0092] Step 4: By analyzing the historical changes of the state vector and combining the adjustment of the control variables, a state transition model is established to describe the impact of the air purifier operation on the air quality and operating status. The state transition model is used to describe the dynamic change relationship between the air quality and the equipment status over time after the air purifier is operated. The variables in the state transition model include the change in air pollutant concentration, the change in humidity, the cumulative value of the filter element usage time, and the relationship between noise and operating parameters. The state transition model provides a basic basis for environmental changes for optimal control;

[0093] Step 5: Determine the control variable value of the air purifier based on the target optimization function and the state transition model. The adjustment of the control variable is used to implement the operation strategy of the air purifier under different air conditions. The result of the optimized control is transmitted to the fan and actuator through the execution module of the equipment to complete the dynamic adjustment of the operating parameters of the air purifier.

[0094] Benefits of step 1: Use multiple sensors to collect the core parameters of indoor air quality and the operating status of equipment to generate a comprehensive state vector. The multi-dimensional monitoring method can accurately reflect the indoor air quality and equipment operating status, make up for the limitation of traditional equipment relying on single parameter monitoring, and provide comprehensive data support for subsequent optimization control;

[0095] Benefits of step 2: By defining wind speed and operation mode as control variables, the operating parameters of the air purifier can be flexibly adjusted. Wind speed is used to dynamically control air handling capacity, and the operation mode can be automatically switched according to user needs or air quality changes, realizing the air purifier's ability to adapt to various environmental conditions, effectively solving the problem of a single operation mode of traditional equipment;

[0096] Benefits of step 3: By constructing the target optimization function, the core optimization requirements in the operation of the air purifier are quantified into energy consumption cost, air quality deviation cost, filter life loss cost and noise cost. This optimization method can balance multiple technical indicators and achieve a dynamic trade-off between high-efficiency purification and low energy consumption and low noise. Compared with the single-target optimization method of traditional equipment, it can adapt to the needs of different usage scenarios;

[0097] Benefits of step 4: By establishing a state transition model, the dynamic relationship between air quality and equipment status after the air purifier is running can be accurately described. The model includes the dynamic evolution of key variables such as air pollutant concentration, humidity, filter element usage time and operating noise. Combined with historical data, it can predict future air quality trends, make up for the deficiency that traditional equipment can only rely on current data to adjust operating parameters, and provide a basis for dynamic response and optimization control of equipment;

[0098] Benefits of step 5: By comprehensively analyzing the target optimization function and the state transition model, the control variable value can be generated to dynamically adjust the operating parameters of the air purifier. The results of the optimization control are directly transmitted to the fan and actuator through the execution module of the equipment, ensuring that the equipment responds quickly to changes in air quality, improves purification efficiency, reduces energy consumption, and at the same time, extends the service life of the filter element.

[0099] The target optimization function constructed in step 3 specifically includes the following formula:

[0100] J t =α1E t +α2Q t +α3L t +α4N t ,

[0101] Among them, J t The target optimization function is used to comprehensively describe the optimization target of the air purifier operation;

[0102] E t is the energy cost, which represents the relationship between wind speed and power consumption, and is defined as E t =P t Δt,

[0103] Where P t =β1w t +β2,w t represents wind speed, β1 and β2 are wind speed power consumption coefficients, and Δt represents the time step;

[0104] Q t is the air quality deviation cost, defined as:

[0105] Q t =γ1(p t -p thr ) 2 +γ2(c t -c thr ) 2 +γ3(v t -v thr ) 2 ,

[0106] Among them, p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration;

[0107] L t is the filter life loss cost, defined as Among them, m t is the operation mode;

[0108] Nt is the noise cost, defined as Among them, n t For operating noise;

[0109] α1, α2, α3, α4, γ1, γ2, γ3, δ, ∈ are the coefficients of optimization weights.

[0110] The air quality deviation cost Q in the objective optimization function t The health thresholds in meet the following settings:

[0111] PM2.5 concentration threshold p thr 35 micrograms per cubic meter;

[0112] CO2 concentration threshold c thr 800ppm;

[0113] Total volatile organic compound concentration threshold v thr It is 0.6ppm.

[0114] The construction of the objective optimization function is one of the important cores of this invention. By quantifying the energy consumption, air quality, filter life and noise of the air purifier, it provides a mathematical basis for multi-objective optimization. By adjusting the optimization weights, dynamically balancing the priorities of different objectives, achieving the efficient adaptability of the equipment in different usage scenarios, improving the operating efficiency and intelligence level of the equipment, significantly improving the user experience, and at the same time reducing the cost of equipment use and environmental impact, it has a wide range of application value.

[0115] The state transition model in step 4 includes the dynamic impact of air purifier operation on air quality and device status, including:

[0116] PM2.5 concentration change: Δp t =-η p w t +∈ p,t ,

[0117] Among them, Δp t is the change in PM2.5 concentration, η p is the purification efficiency coefficient, w t is the wind speed, ∈ p,t is random environmental noise;

[0118] Change in carbon dioxide concentration: Δc t =-η c w t +∈ c,t ,

[0119] Among them, Δc t is the change in carbon dioxide concentration, η c is the purification efficiency coefficient, wt is the wind speed, ∈ c,t is the carbon dioxide environmental noise;

[0120] Total volatile organic compound concentration change: Δv t =-η v w t +∈ v,t ,

[0121] Where Δv t is the change in total volatile organic compound concentration, η v is the purification efficiency coefficient, w t is the wind speed, ∈ v,t is the total volatile organic compound ambient noise;

[0122] Humidity change: Δh t =η h -ξ h w t +∈ h,t ,

[0123] Where Δh t is the humidity change, η h and h is the coefficient of humidity increase and humidity decrease, w t is the wind speed, ∈ h,t is the humidity environmental noise;

[0124] Cumulative change in filter element usage time: Δl t =Δt,

[0125] Among them, Δl t is the change in the cumulative use time of the filter element, and Δt is the time step;

[0126] Running noise: n t =f(w t ,m t ),

[0127] Among them, n t is the operating noise, f(w t ,m t ) represents the functional relationship between wind speed and operation mode and noise, w t is the wind speed, m t is the running mode.

[0128] The state transition model is an important foundation of the present invention. By dynamically modeling air quality parameters, humidity, filter status, and operating noise, the system comprehensively describes the dynamic impact of air purifier operation on indoor air and equipment status. The introduction of the model enhances the device's ability to predict changes in air quality and significantly improves the accuracy and intelligence of operating parameter adjustments. The dynamic characteristics of the state transition model provide a scientific basis for optimized control, enabling the device to operate efficiently and reliably to meet the diverse needs of users in complex air environments.

[0129] The optimization control of the air purifier is accelerated by the reinforcement learning model. The input of the reinforcement learning model is the current state vector, and the output is the adjustment value of the control variable. The reinforcement learning model is trained using the Q learning algorithm, and the loss function is defined as:

[0130]

[0131] Among them, L is the loss function, r t represents the immediate reward value, γ represents the discount factor, s t+1 is the state vector at the next moment, u t+1 is the control variable at the next moment, Q(s t ,u t ) means in state s t Take the control variable u t The comprehensive value of

[0132] Instant reward value r t Defined as the negative value of the objective optimization function, satisfying:

[0133] r t =-(α1E t +α2Q t +α3L t +α4N t ),

[0134] Among them, E t is the energy cost, Q t is the air quality deviation cost, L t is the filter life loss cost, N t is the noise cost, and α1, α2, α3, and α4 are the coefficients of the optimization weights.

[0135] The present invention realizes the intelligent and efficient optimization control by introducing the reinforcement learning model. The reinforcement learning model takes the current state vector as input and quickly generates the control variable adjustment value through the Q learning algorithm, which significantly improves the real-time response capability and optimization efficiency of the equipment. At the same time, through the design of instant reward value, the reinforcement learning model can dynamically balance the multi-objective optimization requirements, making the operation strategy of the equipment in complex environments more flexible and accurate. Combined with the iterative learning ability of historical data, the model shows strong robustness and adaptability, providing strong support for the intelligent upgrade of air purifiers.

[0136] The air purifier is optimized and controlled through the IoT platform, including the following modules:

[0137] Data acquisition module, used to collect variables of the state vector;

[0138] An optimization calculation module is used to calculate the adjustment value of the control variable;

[0139] An execution module, used to adjust the wind speed and operation mode according to the control variables;

[0140] The user interaction module is used to set optimization goals and operation modes through mobile terminals and other terminal devices.

[0141] The optimization calculation module calculates the adjustment value of the control variable based on the multi-objective optimization method. The value range of the weight coefficients α1, α2, α3, and α4 in the objective function of the optimization calculation is set by the user and adaptively adjusted through historical data, where:

[0142] α1 represents the weight coefficient of energy consumption optimization, and its value range is [0.1, 0.9];

[0143] α2 represents the weight coefficient of air quality optimization, and its value range is [0.1, 0.9];

[0144] α3 represents the weight coefficient of filter element life optimization, and its value range is [0.1, 0.9];

[0145] α4 represents the weight coefficient of noise optimization, and its value range is [0.1, 0.9];

[0146] If the sum of weight coefficients meets the following constraints, the optimization calculation module outputs the balance optimization result:

[0147] α1+α2+α3+α4=1.

[0148] The present invention realizes optimization control through the Internet of Things platform, relying on the coordinated operation of the data acquisition module, the optimization calculation module, the execution module and the user interaction module to create an intelligent and automated air purifier control system. Through the close cooperation of the data acquisition and optimization calculation modules, the device can accurately sense the air quality and operating status and dynamically adjust the operating parameters. Through the user interaction module, the user can flexibly set the optimization target and operating mode. Through multi-objective optimization and weight coefficient adjustment, the device can meet the diverse needs of different scenarios. At the same time, it takes into account energy consumption, purification effect, filter element life and noise optimization, significantly improves the performance, intelligence level and user experience of the equipment, and has a wide range of application value.

[0149] The adaptive adjustment weight coefficient of historical data in the optimization calculation module is based on the following formula:

[0150]

[0151] in, represents the weight of the i-th optimization target at time t;

[0152] η represents the step size of weight update, satisfying η>0;

[0153] Denotes the target optimization function J t About weight α i The partial derivative of is used to measure the influence of the current optimization objective on the overall optimization result;

[0154] i∈{1,2,3,4} represents the index of different optimization objectives, including energy consumption optimization, air quality optimization, filter life optimization and noise optimization.

[0155] The present invention realizes the dynamic update of weight coefficients and the precise balance of optimization targets by adaptively adjusting the formula of historical data in the optimization calculation module. The data-driven adaptive adjustment mechanism can automatically adjust the weights according to the contribution of the current optimization target to the total result, avoid target conflicts and improve optimization efficiency. At the same time, it ensures stability and convergence through step size control, adapts to the needs of various scenarios and improves the personalized optimization ability of the equipment, solves the limitations of traditional fixed weight optimization, significantly improves the intelligence level and operational adaptability of the equipment, and further improves the operation effect of the air purifier in complex environments.

[0156] The user interaction module is used to display the weight adjustment results of the optimization target in real time and dynamically set the optimization target according to the user's operation mode, where:

[0157] In high-efficiency mode, the weight of the optimization target satisfies α2>α1, α3, α4, giving priority to optimizing air quality;

[0158] In silent mode, the weight of the optimization target satisfies α4>α1, α2, α3, giving priority to reducing the equipment operation noise;

[0159] In energy-saving mode, the weights of the optimization objectives satisfy α1>α2, α3, α4, giving priority to reducing equipment energy consumption;

[0160] In the equilibrium mode, the weights of the optimization objectives satisfy α1=α2=α3=α4=0.25, achieving a balance among the optimization objectives.

[0161] The user interaction module displays the optimization target weight adjustment results in real time and allows users to dynamically set optimization targets according to the operating mode, so as to achieve precise control and multi-target optimization of the device in different scenarios. Through the flexible switching of high-efficiency mode, silent mode, energy-saving mode and balanced mode, the device can dynamically meet the personalized needs of users, while improving air quality, reducing energy consumption, extending filter life and reducing noise, significantly enhancing the interactive experience between the device and the user, and combining with intelligent algorithms to further improve the convenience and operating efficiency of the device, providing comprehensive technical support for modern smart air purifiers.

[0162] The air quality parameters in the operating state vector collected by the data acquisition module are fused and calculated according to the following weighted average formula to generate the comprehensive air quality index AQI:

[0163]

[0164] Among them, AQI t Indicates the current air quality index.

[0165] w p 、w c 、w v is the weight coefficient of PM2.5, carbon dioxide and total volatile organic compounds, satisfying w p +w c +w v =1,

[0166] p t 、c t 、v t Indicates the current concentration of PM2.5, carbon dioxide and total volatile organic compounds;

[0167] p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration;

[0168] Comprehensive air quality index AQI t It is used as the input basis for dynamically adjusting the control variables in the optimization calculation module.

[0169] The present invention calculates the comprehensive air quality index through the weighted average formula, provides a unified and accurate input basis for the optimization calculation module to dynamically adjust the control variables, and supports the flexible adjustment of the weight coefficient to meet the needs of multiple scenarios. Fusion calculation enhances the stability and adaptability of data processing, making the control decision of the equipment more intelligent and reliable. Ultimately, it significantly improves the optimization control capability and user experience of the air purifier, and has a wide range of application value.

[0170] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A control method for an indoor air purifier based on the Internet of Things, characterized in that: include: Step 1: Collect indoor air quality parameters and equipment operation status through sensors on the air purifier to obtain a state vector containing multiple variables. The state vector includes PM2.5 concentration, carbon dioxide concentration, total volatile organic compound concentration, humidity, cumulative filter element usage time and equipment operation noise. The state vector is used to describe information about air quality and equipment operation status; Step 2: Define the control variables of the air purifier according to the state vector. The control variables include wind speed and operation mode. The wind speed is used to control the air handling capacity of the air purifier. The operation modes include high-efficiency mode, silent mode and energy-saving mode. The control variables are used to adjust the operation parameters of the air purifier to adapt to different air states and operation requirements. Step 3: Based on the state vector and control variables, a target optimization function is constructed to describe the optimization requirements of the air purifier operation. The target optimization function includes energy consumption cost, air quality deviation cost, filter life loss cost and noise cost. The construction of the target optimization function is used to balance among multiple optimization objectives to determine the operation strategy. Step 4: By analyzing the historical changes of the state vector and combining the adjustment of the control variables, a state transition model is established to describe the impact of the air purifier operation on the air quality and operating status. The state transition model is used to describe the dynamic change relationship between the air quality and the equipment status over time after the air purifier is operated. The variables in the state transition model include the change in air pollutant concentration, the change in humidity, the cumulative value of the filter element usage time, and the relationship between noise and operating parameters. The state transition model provides a basic basis for environmental changes for optimal control; Step 5: Determine the control variable value of the air purifier based on the target optimization function and the state transition model. The adjustment of the control variable is used to implement the operation strategy of the air purifier under different air conditions. The result of the optimized control is transmitted to the fan and actuator through the execution module of the equipment to complete the dynamic adjustment of the operating parameters of the air purifier.

2. The control method of an indoor air purifier based on the Internet of Things according to claim 1, characterized in that: The target optimization function constructed in step 3 specifically includes the following formula: J t =α1E t +α2Q t +α3L t +α4N t , Among them, J t The target optimization function is used to comprehensively describe the optimization target of the air purifier operation; E t is the energy cost, which represents the relationship between wind speed and power consumption and is defined as E t =P t Δt, Where P t =β1w t +β2,w t represents wind speed, β1 and β2 are wind speed power consumption coefficients, and Δt represents the time step; Q t is the air quality deviation cost, defined as: Q t =γ1(p t -p thr ) 2 +γ2(c t -c thr ) 2 +γ3(v t -v thr ) 2 , Among them, p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration; L t is the filter life loss cost, defined as Among them, m t is the operation mode; N t is the noise cost, defined as Among them, n t For operating noise; α1, α2, α3, α4, γ1, γ2, γ3, δ, ∈ are coefficients of optimization weights.

3. The control method of an indoor air purifier based on the Internet of Things according to claim 2, characterized in that: The air quality deviation cost Q in the objective optimization function t The health thresholds in meet the following settings: PM2.5 concentration threshold p thr 35 micrograms per cubic meter; CO2 concentration threshold c thr 800ppm; Total volatile organic compound concentration threshold v thr It is 0.6ppm.

4. The control method of an indoor air purifier based on the Internet of Things according to claim 1, characterized in that: The state transition model in step 4 includes the dynamic impact of the air purifier operation on the air quality and the device state, including: PM2.5 concentration change: Δp t =-η p w t +∈ p,t , Among them, Δp t is the change in PM2.5 concentration, η p is the purification efficiency coefficient, w t is the wind speed, ∈ p,t is random environmental noise; Change in carbon dioxide concentration: Δc t =-η c w t +∈ c,t , Where Δc t is the change in carbon dioxide concentration, η c is the purification efficiency coefficient, w t is the wind speed, ∈ c,t is the carbon dioxide environmental noise; Total volatile organic compound concentration change: Δv t =-η v w t +∈ v,t , Where Δv t is the change in total volatile organic compound concentration, η v is the purification efficiency coefficient, w t is the wind speed, ∈ v,t is the total volatile organic compound ambient noise; Humidity change: Δh t =η h -ξ h w t +∈ h,t , Where Δh t is the humidity change, η h and h is the coefficient of humidity increase and humidity decrease, w t is the wind speed, ∈ h,t is the humidity environmental noise; Cumulative change in filter element usage time: Δl t =Δt, Among them, Δl t is the change in the cumulative use time of the filter element, and Δt is the time step; Running noise: n t =f(w t ,m t ), Among them, n t is the operating noise, f(w t ,m t ) represents the functional relationship between wind speed and operation mode and noise, w t is the wind speed, m t is the running mode.

5. The control method of an indoor air purifier based on the Internet of Things according to claim 1, characterized in that: The optimization control of the air purifier is accelerated by a reinforcement learning model, the input of which is the current state vector, and the output is the adjusted value of the control variable. The reinforcement learning model is trained using a Q learning algorithm, and the loss function is defined as: Among them, L is the loss function, r t represents the immediate reward value, γ represents the discount factor, s t+1 is the state vector at the next moment, u t+1 is the control variable at the next moment, Q(s t ,u t ) means in state s t Take the control variable u t The comprehensive value of The instant reward value r t Defined as the negative value of the objective optimization function, satisfying: r t =-(α1E t +α2Q t +α3L t +α4N t ), Among them, E t is the energy cost, Q t is the air quality deviation cost, L t is the filter life loss cost, N t is the noise cost, and α1, α2, α3, and α4 are the coefficients of the optimization weights.

6. The control method of an indoor air purifier based on the Internet of Things according to claim 1, characterized in that: The air purifier is optimized and controlled through the Internet of Things platform, including the following modules: Data acquisition module, used to collect variables of the state vector; An optimization calculation module is used to calculate the adjustment value of the control variable; An execution module, used to adjust the wind speed and operation mode according to the control variables; The user interaction module is used to set optimization goals and operation modes through mobile terminals and other terminal devices.

7. The control method of an indoor air purifier based on the Internet of Things according to claim 6, characterized in that: The optimization calculation module calculates the adjustment value of the control variable based on the multi-objective optimization method. The value range of the weight coefficients α1, α2, α3, and α4 in the objective function of the optimization calculation is set by the user and adaptively adjusted through historical data, wherein: α1 represents the weight coefficient of energy consumption optimization, and its value range is [0.1, 0.9]; α2 represents the weight coefficient of air quality optimization, and its value range is [0.1, 0.9]; α3 represents the weight coefficient of filter element life optimization, and its value range is [0.1, 0.9]; α4 represents the weight coefficient of noise optimization, and its value range is [0.1, 0.9]; If the sum of weight coefficients meets the following constraints, the optimization calculation module outputs the balance optimization result: α1+α2+α3+α4＝1。 8. The control method of an indoor air purifier based on the Internet of Things according to claim 7, characterized in that: The historical data adaptive adjustment weight coefficient in the optimization calculation module is based on the following formula: in, represents the weight of the i-th optimization target at time t; η represents the step size of weight update, satisfying η>0; Denotes the target optimization function J t About weight α i The partial derivative of is used to measure the influence of the current optimization objective on the overall optimization result; i∈{1,2,3,4} represents the index of different optimization objectives, including energy consumption optimization, air quality optimization, filter life optimization and noise optimization.

9. The control method of an indoor air purifier based on the Internet of Things according to claim 8, characterized in that: The user interaction module is used to display the weight adjustment result of the optimization target in real time and dynamically set the optimization target according to the user's operation mode, wherein: In high-efficiency mode, the weight of the optimization target satisfies α2>α1, α3, α4, giving priority to optimizing air quality; In silent mode, the weight of the optimization target satisfies α4>α1, α2, α3, giving priority to reducing the equipment operation noise; In energy-saving mode, the weights of the optimization objectives satisfy α1>α2, α3, α4, giving priority to reducing equipment energy consumption; In the equilibrium mode, the weights of the optimization objectives satisfy α1=α2=α3=α4=0.25, achieving a balance among the optimization objectives.

10. The control method of an indoor air purifier based on the Internet of Things according to claim 6, characterized in that: The air quality parameters in the running state vector collected by the data collection module are fused and calculated according to the following weighted average formula to generate a comprehensive air quality index AQI: Among them, AQI t Indicates the current air quality index. w p 、w c 、w v is the weight coefficient of PM2.5, carbon dioxide and total volatile organic compounds, satisfying w p +w c +w v =1, p t 、c t 、v t Indicates the current concentration of PM2.5, carbon dioxide and total volatile organic compounds; p thr 、c thr 、v thr The health thresholds for PM2.5 concentration, carbon dioxide concentration and total volatile organic compound concentration; The comprehensive air quality index AQI t It is used as the input basis for dynamically adjusting the control variables in the optimization calculation module.

Images