Wireless network hotspot-based aroma diffuser intelligent control method and aroma diffuser

Abstract

The invention belongs to the technical field of Internet of Things, and discloses an aroma diffuser intelligent control method based on wireless network hotspots and an aroma diffuser. Comprising the steps of collecting pressure distribution data and judging whether an opening instruction is generated or not; if the starting instruction is generated, driving characteristic data are collected in real time; processing the driving characteristic data to obtain characteristic processing data; analyzing the feature processing data, and determining a working mode; regulating and controlling aromatherapy parameters according to the feature processing data and the working mode; the air quality in the vehicle can be improved, and a comfortable and fresh in-vehicle environment is created for a driver, so that the driving experience of a user is improved; personalized fragrance selection is achieved, potential safety hazards caused by traditional manual operation are reduced, the influence on driving safety is small, and convenience and comfort in the driving process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and more specifically, to an intelligent control method of an aromatherapy machine based on a wireless network hotspot and an aromatherapy machine. Background Art

[0002] As people pay more attention to the quality of life in cars, car aromatherapy machines, as an important device to enhance driving experience, have gradually attracted attention; traditional car aromatherapy machines are usually manually operated, and users need to adjust the fragrance through physical buttons or knobs, which is not convenient to use and it is difficult to achieve personalized fragrance selection and intelligent management; this manual control method limits the user experience to a certain extent, especially during driving, distracting operations will bring safety hazards.

[0003] With the advancement of intelligent control technology, intelligent aromatherapy machines based on wireless networks are widely favored by users; through wireless network connection, users can remotely control the aromatherapy machine to achieve various personalized operations such as timer switch and aromatherapy concentration adjustment; in addition, the intelligent aromatherapy machine can also make adaptive adjustments according to environmental information to improve the user experience and aromatherapy effect; for example, the patent with announcement number CN111110902A discloses a control method, device, storage medium and electronic device for an aromatherapy machine; including: obtaining environmental information of the current environment and human state information in the current environment; inputting the current human state information and the current environmental information into the trained first deep learning model to obtain the control mode of the aromatherapy machine; according to the control mode of the aromatherapy machine, controlling the aromatherapy smell, aromatherapy concentration, light brightness, light color and music playback of the aromatherapy machine; this invention can realize intelligent control of the aromatherapy machine according to the current human state and the current environmental state;

[0004] However, although the above technology realizes the intelligent control of the aromatherapy machine, the control basis relies on human physiological information and brain wave information, which usually requires users to wear specific equipment for collection. When the user is driving, wearing the equipment will interfere with the user's vision and touch, causing the user's attention to be distracted and the comfort to be reduced, affecting the user's reaction ability, and further affecting driving safety and increasing the risk of accidents.

[0005] In view of this, the present invention proposes an intelligent control method of an aromatherapy machine based on a wireless network hotspot and an aromatherapy machine to solve the above problems. Summary of the invention

[0006] In order to overcome the above defects of the prior art and achieve the above objectives, the present invention provides the following technical solutions: an intelligent control method for an aromatherapy machine based on a wireless network hotspot, comprising:

[0007] S1: Collect pressure distribution data and determine whether to generate an opening instruction;

[0008] S2: If an opening command is generated, the driving characteristic data is collected in real time;

[0009] S3: Process the driving feature data to obtain feature processing data;

[0010] S4: Analyze the feature processing data and determine the working mode;

[0011] S5: Regulate aromatherapy parameters according to feature processing data and working mode.

[0012] Further, the pressure distribution data is pressure distribution information on the car seat, and the pressure distribution data includes n pressure data, where n is an integer greater than 1;

[0013] The method for determining whether to generate an opening instruction includes:

[0014] Add n pressure data in sequence and divide by n to obtain the pressure average value; preset a pressure threshold and compare the pressure average value with the pressure threshold; if the pressure average value is greater than or equal to the pressure threshold, generate a preliminary opening instruction; if the pressure average value is less than the pressure threshold, do not generate a preliminary opening instruction;

[0015] If a preliminary opening instruction is generated, the distribution distance is obtained; the distribution distance is the distance between every two adjacent pressure sensors; the pressure data corresponding to every two adjacent pressure sensors are taken as a group of pressure sets, and the pressure difference corresponding to each group of pressure sets is calculated in turn; the expression of the pressure difference is: ΔP = |P2-P1|; where ΔP is the pressure difference, P2 is a pressure data in the pressure set, and P1 is another pressure data in the pressure set; each pressure difference is divided by the corresponding distribution distance to obtain the pressure gradient; all pressure gradients are added in turn, and then divided by the number of distances to obtain the pressure gradient mean; a gradient threshold is preset, and the pressure gradient mean is compared with the gradient threshold; if the pressure gradient mean is greater than or equal to the gradient threshold, an opening instruction is generated; if the pressure gradient mean is less than the gradient threshold, no opening instruction is generated.

[0016] Furthermore, the driving characteristic data includes environmental data and time data; the environmental data includes temperature, humidity, gas concentration and wind speed; the time data includes driving time and driving duration;

[0017] The method for obtaining feature processing data comprises:

[0018] A group of historical feature data is obtained, where the historical feature data is driving feature data collected at historical moments, and a is an integer greater than 1; the driving feature data collected in real time is marked as real-time feature data, anomaly detection is performed on each data in the real-time feature data, the data with anomalies is marked as abnormal data, the abnormal data is re-collected, and the abnormal data in the real-time feature data is replaced with the corresponding re-collected data to obtain feature processing data; wherein the anomaly detection method for each data in the real-time feature data is consistent; the method for detecting anomaly of temperature includes:

[0019] Preset the number of analyses b; use the real-time characteristic data and the historical characteristic data as the analysis characteristic data, subtract the corresponding neighbor temperature from each temperature in the analysis characteristic data, obtain the temperature difference corresponding to each temperature, and the neighbor temperature corresponding to the temperature is the remaining a temperatures in the analysis characteristic data; use all the temperature differences corresponding to each temperature as a temperature set, and the temperature set corresponds to the temperature one by one; sort the temperature differences in each temperature set from small to large, and generate a temperature sorting table corresponding to each temperature set; use the temperature difference ranked in the bth place in each temperature sorting table as the reference distance of the temperature corresponding to the corresponding temperature set, and use the neighbor temperature corresponding to the first b temperature differences in each temperature sorting table as the standard temperature of the temperature corresponding to the corresponding temperature set;

[0020] According to the reference distance of each temperature and the corresponding temperature difference, the adjacent distance between each temperature and the corresponding standard temperature is determined; according to the adjacent distance between each temperature and the corresponding standard temperature, the local density of each temperature is calculated; the temperature in the real-time characteristic data is marked as the real-time temperature, and the discrete coefficient ls corresponding to the real-time temperature is calculated according to the adjacent distance between each temperature and the corresponding standard temperature; the judgment coefficient d is preset, 0<d<1; if |ls-1|≥d, the real-time temperature is marked as abnormal data; if |ls-1|<d, the real-time temperature is not marked.

[0021] Furthermore, the expression of the adjacent distance is: xl(p,q)=max(jl(p),wc(p,q)); wherein xl(p,q) is the adjacent distance between the pth temperature in the analysis characteristic data and the corresponding qth standard temperature, max is the maximum value function, jl(p) is the reference distance of the pth temperature in the analysis characteristic data, wc(p,q) is the temperature difference between the pth temperature in the analysis characteristic data and the corresponding qth standard temperature, p∈[1,a+1], q∈[1,b];

[0022] The expression of local density is: Where md(p) is the local density of the pth temperature in the analyzed characteristic data;

[0023] The expression of the coefficient of dispersion is: Where md′ is the local density corresponding to the real-time temperature, and md(q) is the local density of the qth standard temperature corresponding to the real-time temperature.

[0024] Furthermore, the step of determining the working mode includes:

[0025] Step S401: Analyze the gas concentration in the feature processing data to evaluate the air quality; replace the gas concentration in the feature processing data with the air quality to obtain feature replacement data;

[0026] Step S402: construct multiple fuzzy sets for each data in the feature replacement data;

[0027] Step S403: converting each data in the feature replacement data into the membership degree of each corresponding fuzzy set by using fuzzification technology;

[0028] Step S404: define fuzzy rules;

[0029] Step S405: matching the fuzzified feature replacement data with the fuzzy rules, performing fuzzy reasoning, and obtaining fuzzy reasoning results, where the fuzzy reasoning results are the membership degrees corresponding to each working mode;

[0030] Step S406: Compare each degree of membership in the fuzzy inference result, and take the working mode corresponding to the degree of membership with the largest value as the working mode corresponding to the feature processing data;

[0031] In step S401, the method for evaluating air quality includes:

[0032] The gas concentration in the feature processing data is used as the analysis data, and the analysis data is input into the trained quality assessment model to predict the corresponding air quality; the training process of the quality assessment model includes:

[0033] G groups of analysis data are collected in advance, and corresponding air qualities are set for the g groups of analysis data, where g is an integer greater than 1, and the analysis data and the corresponding air qualities are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a quality assessment model, and the quality assessment model uses a set of predicted air qualities corresponding to each set of analysis data as output, and uses the actual air quality corresponding to each set of analysis data as a prediction target, where the actual air quality is the pre-set air quality corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as a training target; the quality assessment model is trained until the sum of prediction errors converges and the training is stopped; the quality assessment model is a deep neural network model.

[0034] Furthermore, the aromatherapy parameters include aromatherapy concentration, working time and release interval; the aromatherapy concentration is the amount of aroma released, the working time is the duration of continuous working after the aromatherapy machine is turned on each time, and the release interval is the time for the aroma to be released at intervals;

[0035] The step of regulating the aromatherapy parameters comprises:

[0036] Step S51: constructing a parameter set, the parameter set including M groups of analysis parameters;

[0037] Step S52: According to the working mode, m groups of control parameters are selected from the parameter set, where 1<m<M;

[0038] Step S53: Calculate the comfort level corresponding to each set of control parameters;

[0039] Step S54: based on comfort, select S groups of selection parameters from m groups of control parameters, 1≤S<m;

[0040] Step S55: if S=1, the aromatherapy parameters are regulated according to the selected parameters; if S>1, the energy consumption value corresponding to each set of selected parameters is calculated;

[0041] Step S56: sort the S group of selection parameters from large to small according to the energy consumption value, mark the selection parameter at the front as the best parameter, and adjust the aromatherapy parameter according to the best parameter.

[0042] Furthermore, in step S51, the method for constructing the parameter set is: obtaining a parameter range, the parameter range including the range corresponding to each parameter in the aromatherapy parameter; randomly selecting a value from each range within the parameter range to construct a set of analysis parameters, constructing a total of M groups of analysis parameters, where M is an integer greater than 1, and the M groups of analysis parameters are all different;

[0043] In step S53, the method for calculating the comfort corresponding to each set of control parameters includes:

[0044] Each set of control parameters and feature replacement data is taken as a set of test parameters, and the test parameters correspond to the control parameters one by one; each set of test parameters is input into the trained comfort assessment model to predict the corresponding comfort; the training process of the comfort assessment model is consistent with the training process of the quality assessment model, and both are deep neural network models.

[0045] Furthermore, in step S52, the step of selecting m groups of control parameters from the parameter set includes:

[0046] Step S521: preset the neighborhood distance and the number of neighbors;

[0047] Step S522: Take each set of analysis parameters in the parameter set as a node and calculate the node distance between every two nodes; the expression of the node distance is: Where, L ij is the node distance between the i-th node and the j-th node, x ir is the rth parameter in the analysis parameters corresponding to the i-th node, x jr is the rth parameter in the analysis parameters corresponding to the jth node, r∈[1,3];

[0048] Step S523: construct the neighborhood of each node according to the neighborhood distance; count the number of nodes in the neighborhood corresponding to each node according to the node distance, and mark them as the number of neighboring points; that is, count the number of node distances corresponding to each node whose value is less than or equal to the neighborhood distance;

[0049] Step S524: Mark the node whose number of neighbors is greater than or equal to the number of neighbors as a center point;

[0050] Step S525: randomly select a center point that is not marked as a selected point, mark it as a selected point, and update the current point to a selected point; construct a new cluster and mark it as the current cluster; add the current point and all nodes in the neighborhood corresponding to the current point to the current cluster;

[0051] Step S526: Mark all the center points in the current cluster except the current point as extension points, and add all the nodes in the neighborhood corresponding to the extension point to the current cluster;

[0052] Step S527: looping step S526 until all nodes in the neighborhood corresponding to the extension point in the current cluster are added to the current cluster, the loop ends, and the current cluster is marked as a completed cluster;

[0053] Step S528: looping steps S525 to S527 until all center points are marked as selected points, the loop ends, and all completed clusters are obtained, and the completed clusters correspond to the working modes one by one;

[0054] Step S529: According to the determined working mode, m groups of analysis parameters in the corresponding completed cluster are obtained and marked as control parameters.

[0055] Furthermore, in step S54, the method of selecting S groups of selection parameters from m groups of control parameters includes:

[0056] According to comfort, m groups of control parameters are sorted from large to small to generate a parameter sorting table; a deviation threshold is preset; the control parameter at the front of the parameter sorting table is marked as a selection parameter, and the control parameter not marked as a selection parameter is marked as a screening parameter; the comfort of each screening parameter is subtracted from the comfort of the selection parameter in turn to obtain a comfort deviation; the comfort deviation is compared with the deviation threshold, and the screening parameter corresponding to the comfort deviation less than or equal to the deviation threshold is marked as a selection parameter, and the screening parameter corresponding to the comfort deviation greater than the deviation threshold is not marked;

[0057] In step S55, the method for calculating the energy consumption value corresponding to each set of selection parameters includes:

[0058] Each set of test parameters is input into the trained energy consumption assessment model to predict the corresponding energy consumption value; the training process of the energy consumption assessment model is consistent with the training process of the quality assessment model, and both are deep neural network models.

[0059] An aromatherapy machine comprises a storage unit, a central processing unit and a computer program stored in the storage unit and executable on the central processing unit. When the central processing unit executes the computer program, the intelligent control method of the aromatherapy machine based on a wireless network hotspot is implemented.

[0060] The technical effects and advantages of the aromatherapy machine intelligent control method based on wireless network hotspot and the aromatherapy machine of the present invention are as follows:

[0061] By collecting pressure distribution data, it automatically determines whether to turn on the aromatherapy machine; and by real-time collection of various environmental data and driving characteristics in the car, fuzzy logic analysis is used to determine the appropriate working mode; then after parameter screening, comfort and energy consumption evaluation and other steps, the aromatherapy parameters of the aromatherapy machine are dynamically adjusted to improve the air quality in the car, create a comfortable and fresh car environment for the driver, and thus enhance the user's driving experience; it not only realizes personalized fragrance selection, but also reduces the safety hazards brought by traditional manual operation, has little impact on driving safety, and improves convenience and comfort during driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of an intelligent control method for an aromatherapy machine based on a wireless network hotspot according to Embodiment 1 of the present invention;

[0063] Figure 2 This is a flow chart of the aromatherapy parameter control method according to Example 1 of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Example 1

[0066] See also Figure 1 As shown, this embodiment provides an intelligent control method for an aromatherapy machine based on a wireless network hotspot, and the method includes:

[0067] S1: Collect pressure distribution data and determine whether to generate an opening instruction.

[0068] The pressure distribution data is the pressure distribution information on the car seat. The pressure distribution data includes n pressure data, where n is an integer greater than 1. The pressure data is collected by pressure sensors distributedly installed under the car seat.

[0069] The method for determining whether to generate an opening instruction includes:

[0070] Add n pieces of pressure data in sequence, and then divide by n to obtain an average pressure value; preset a pressure threshold value, which is preset by a person skilled in the art according to actual conditions; compare the average pressure value with the pressure threshold value; if the average pressure value is greater than or equal to the pressure threshold value, generate a preliminary opening instruction; if the average pressure value is less than the pressure threshold value, do not generate a preliminary opening instruction; it should be understood that the purpose of generating a preliminary opening instruction is that when the average pressure value on the car seat changes, it does not necessarily mean that the driver has entered the car, but may be because the driver has placed an object on the car seat, so further analysis is required;

[0071] If a preliminary opening instruction is generated, the distribution distance is obtained; the distribution distance is the distance between every two adjacent pressure sensors; the pressure data corresponding to every two adjacent pressure sensors are taken as a group of pressure sets, and the pressure difference corresponding to each group of pressure sets is calculated in turn; the expression of the pressure difference is: ΔP = |P2-P1|; where ΔP is the pressure difference, P2 is a pressure data in the pressure set, and P1 is another pressure data in the pressure set; each pressure difference is divided by the corresponding distribution distance to obtain the pressure gradient; all pressure gradients are added in turn, and then divided by the number of distances to obtain the pressure gradient mean; a preset gradient threshold is preset by a technician in this field according to actual conditions; the pressure gradient mean is compared with the gradient threshold; if the pressure gradient mean is greater than or equal to the gradient threshold, an opening instruction is generated; if the pressure gradient mean is less than the gradient threshold, no opening instruction is generated; it should be understood that when the driver sits on a car seat, the pressure is often concentrated in a specific area of ​​the seat, such as the center of the seat, so the pressure distribution is usually unevenly distributed; while the center of gravity of the heavy object placed on the seat is relatively evenly distributed, and the weight can be more evenly distributed on the seat, so the pressure distribution is usually evenly distributed.

[0072] S2: If an opening instruction is generated, driving characteristic data is collected in real time.

[0073] Driving characteristic data include environmental data and time data;

[0074] Environmental data includes temperature, humidity, gas concentration, and wind speed;

[0075] Temperature is the temperature inside the car, and the temperature is obtained by a temperature sensor installed in the car (such as a thermocouple sensor, a digital temperature sensor, etc.); temperature has a significant effect on the volatilization speed and concentration of the aromatherapy. The higher the temperature, the faster the molecular movement in the aromatherapy, the faster the volatilization speed of the aromatherapy, and the faster the growth rate of the aromatherapy concentration in the car, and vice versa. Humidity is the humidity inside the car, and the humidity is obtained by a humidity sensor installed in the car (such as a capacitive humidity sensor, a resistive humidity sensor, etc.); humidity has a significant effect on the volatilization speed and concentration of the aromatherapy. The higher the humidity, the slower the volatilization speed of the aromatherapy, and the slower the growth rate of the aromatherapy concentration in the car, and vice versa. Gas concentration is the concentration of different gases in the car, and the gas concentration is obtained by a multi-gas sensor module installed in the car (such as a Figaro sensor). TGS, MiCS-5524, etc.), and multiple gas sensors are integrated in the multi-gas sensor module; gas concentration is used to reflect the air quality in the car. The worse the air quality, the more it is necessary to increase the working time and aromatherapy concentration of the aromatherapy machine to improve the air quality in the car, and vice versa; wind speed is the speed of air flow in the car, and the wind speed is obtained through the wind speed sensor installed in the car (such as thermal wind speed sensor, turbine wind speed sensor, etc.); wind speed will affect the diffusion effect of aromatherapy. The higher the wind speed, the faster the aroma will spread inside the car, and vice versa; therefore, it is necessary to intelligently adjust the aromatherapy concentration and working time of the aromatherapy machine according to environmental data to ensure the balance of comfort, air quality and aromatherapy effect in the car environment, avoid the aroma being too strong or too weak, and achieve a comfortable and efficient in-vehicle aromatherapy environment.

[0076] Time data include driving time and driving duration;

[0077] The driving time is the current moment, and the driving time is obtained through the on-board clock system in the car; the driving time is the time from the car start time to the driving time, and the driving time is obtained as follows: when the vehicle ignition sensor installed in the car engine (such as crankshaft position sensor, ignition coil sensor, etc.) detects that the car is started, the timer built into the vehicle control unit starts timing, and the recorded time is the driving time; when driving at night or driving for a long time, due to dim light, the driver is prone to fatigue and sleepiness, so the aromatherapy machine can appropriately increase the aromatherapy concentration and working time, which helps the driver stay awake and alert.

[0078] It should be noted that the data transmission between the multiple sensors and the aromatherapy machine mentioned above is all achieved through wireless network hotspots.

[0079] S3: Process the driving feature data to obtain feature processing data.

[0080] Methods for obtaining feature processing data include:

[0081] Acquire a group of historical feature data, where the historical feature data is driving feature data collected at historical moments, and a is an integer greater than 1; mark the driving feature data collected in real time as real-time feature data, perform anomaly detection on each data in the real-time feature data, mark the data with anomalies as abnormal data, re-collect the abnormal data, replace the abnormal data in the real-time feature data with the corresponding re-collected data, and obtain feature processing data; wherein the anomaly detection method for each data in the real-time feature data is consistent.

[0082] Methods for detecting temperature anomalies include:

[0083] The analysis quantity b is preset, and the analysis quantity b is preset by technical personnel in this field according to actual conditions; the real-time characteristic data and the historical characteristic data are used as the analysis characteristic data, and the corresponding neighbor temperature is subtracted from each temperature in the analysis characteristic data to obtain the temperature difference corresponding to each temperature, and the neighbor temperature corresponding to the temperature is the remaining a temperatures in the analysis characteristic data; all the temperature differences corresponding to each temperature are used as a temperature set, and the temperature set corresponds to the temperature one by one; the temperature differences in each temperature set are sorted from small to large, and a temperature sorting table corresponding to each temperature set is generated; the temperature difference ranked in the bth place in each temperature sorting table is used as the reference distance of the temperature corresponding to the corresponding temperature set, and the neighbor temperature corresponding to the first b temperature differences in each temperature sorting table is used as the standard temperature of the temperature corresponding to the corresponding temperature set;

[0084] According to the reference distance of each temperature and the corresponding temperature difference, the adjacent distance between each temperature and the corresponding standard temperature is determined; the expression of the adjacent distance is: xl(p,q)=max(jl(p),wc(p,q)); where xl(p,q) is the adjacent distance between the pth temperature in the analysis feature data and the corresponding qth standard temperature, max is the maximum value function, jl(p) is the reference distance of the pth temperature in the analysis feature data, wc(p,q) is the temperature difference between the pth temperature in the analysis feature data and the corresponding qth standard temperature, p∈[1,a+1], q∈[1,b];

[0085] According to the adjacent distance between each temperature and the corresponding standard temperature, the local density of each temperature is calculated. The expression of local density is: Where md(p) is the local density of the pth temperature in the analyzed characteristic data;

[0086] The temperature in the real-time characteristic data is marked as the real-time temperature. According to the adjacent distance between each temperature and each corresponding standard temperature, the dispersion coefficient corresponding to the real-time temperature is calculated; the expression of the dispersion coefficient is: Where ls is the dispersion coefficient corresponding to the real-time temperature, md′ is the local density corresponding to the real-time temperature, and md(q) is the local density of the qth standard temperature corresponding to the real-time temperature;

[0087] The judgment coefficient r is preset, 0<d<1, and the judgment coefficient d is preset by technical personnel in this field according to actual conditions; if |ls-1|≥d, the real-time temperature is marked as abnormal data; if |ls-1|<d, the real-time temperature is not marked.

[0088] S4: Analyze the feature processing data and determine the working mode.

[0089] Working modes include refreshing mode (increasing the concentration of aromatherapy to help drivers stay alert and awake), relaxing mode (reducing the concentration of aromatherapy to provide comfortable fragrance, suitable for the rest period after long driving), fresh mode (maintaining medium aromatherapy concentration to ensure fresh air but not produce too strong fragrance, suitable for poor air quality in the car), etc.

[0090] The steps to determine the working mode include:

[0091] Step S401: Analyze the gas concentration in the feature processing data to evaluate the air quality; replace the gas concentration in the feature processing data with the air quality to obtain feature replacement data;

[0092] Step S402: construct multiple fuzzy sets for each data in the feature replacement data; for example, the fuzzy set corresponding to temperature is low temperature, medium temperature, high temperature, etc., the fuzzy set corresponding to air quality is good quality, medium quality, poor quality, etc., and the fuzzy set corresponding to driving time is short time, medium time, long time, etc.;

[0093] Step S403: convert each data in the feature replacement data into the membership of each corresponding fuzzy set through fuzzification technology; fuzzification is the process of converting an exact value into the membership corresponding to the fuzzy set, and fuzzification technology includes, for example, a triangular membership function, a trapezoidal membership function, etc.; for example, if the value of the wind speed is low, it is inferred that the membership of the low wind speed is 0.9, the membership of the medium wind speed is 0.1, and the membership of the high wind speed is 0;

[0094] Step S404: define fuzzy rules, which are defined based on expert knowledge or relevant literature; for example, if the temperature is low, the humidity is high, and the duration is long, then the probability of inferring that the working mode is the refreshing mode is high; if the quality is poor, the wind speed is medium, and the duration is short, then the probability of inferring that the working mode is the refreshing mode is high;

[0095] Step S405: Match the fuzzified feature replacement data with the fuzzy rules, perform fuzzy reasoning, and obtain the fuzzy reasoning result, which is the membership degree corresponding to each working mode; the fuzzy reasoning method is, for example, the Mamdan i or Sugeno fuzzy reasoning method; the fuzzy reasoning result is, for example, the membership degree of the refreshing mode is 0.1, the membership degree of the relaxing mode is 0.3, and the membership degree of the refreshing mode is 0.6;

[0096] Step S406: Compare each membership degree in the fuzzy inference result, and take the working mode corresponding to the membership degree with the largest value as the working mode corresponding to the feature processing data.

[0097] In the above step S401, the method for evaluating air quality includes:

[0098] The gas concentration in the feature processing data is used as the analysis data, and the analysis data is input into the trained quality assessment model to predict the corresponding air quality; the training process of the quality assessment model includes:

[0099] G groups of analysis data are collected in advance, and corresponding air qualities are set for the g groups of analysis data, where g is an integer greater than 1, and the analysis data and the corresponding air qualities are converted into a corresponding set of feature vectors; the air qualities corresponding to the analysis data are collected by a technician in the historical aromatherapy machine control process, and each group of analysis data is analyzed in turn based on actual experience, the air quality corresponding to each group of analysis data is evaluated, and the corresponding air quality is set for the g groups of analysis data in turn;

[0100] Each set of feature vectors is used as the input of the quality assessment model. The quality assessment model takes a set of predicted air quality corresponding to each set of analysis data as output, and takes the actual air quality corresponding to each set of analysis data as the prediction target. The actual air quality is the pre-set air quality corresponding to the analysis data. The training goal is to minimize the sum of the prediction errors of all analysis data. The calculation formula of the prediction error is η w =(β w -ε w ) 2 , where η w is the prediction error, w is the group number of the eigenvector corresponding to the analyzed data, β w is the predicted air quality corresponding to the wth group of analysis data, ε w is the actual air quality corresponding to the wth group of analysis data; the quality assessment model is trained until the sum of the prediction errors reaches convergence and the training is stopped.

[0101] The above-mentioned quality assessment model is specifically a deep neural network model; it includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function introduces nonlinearity, allowing the network to learn more complex patterns and features.

[0102] S5: Regulate aromatherapy parameters according to feature processing data and working mode.

[0103] Aromatherapy parameters include aromatherapy concentration, working time and release interval; aromatherapy concentration is the amount of aroma released, working time is the length of time the aromatherapy machine continues to work each time it is turned on, and release interval is the time between aroma releases.

[0104] like Figure 2 As shown, the steps of regulating the aromatherapy parameters include:

[0105] Step S51: constructing a parameter set, the parameter set including M groups of analysis parameters;

[0106] Step S52: According to the working mode, m groups of control parameters are selected from the parameter set, where 1<m<M;

[0107] Step S53: Calculate the comfort level corresponding to each set of control parameters;

[0108] Step S54: based on comfort, select S groups of selection parameters from m groups of control parameters, 1≤S<m;

[0109] Step S55: if S=1, the aromatherapy parameters are regulated according to the selected parameters; if S>1, the energy consumption value corresponding to each set of selected parameters is calculated;

[0110] Step S56: sort the S group of selection parameters from large to small according to the energy consumption value, mark the selection parameter at the front as the best parameter, and adjust the aromatherapy parameter according to the best parameter.

[0111] In the above step S51, the method for constructing the parameter set is: obtaining a parameter range, the parameter range includes the range corresponding to each parameter in the aromatherapy parameter, and the parameter range is obtained by a technician in this field according to the relevant technical parameters of the aromatherapy machine; randomly selecting a value from each range within the parameter range to construct a set of analysis parameters, and constructing a total of M groups of analysis parameters, where M is an integer greater than 1, and the M groups of analysis parameters are all different.

[0112] In the above step S52, the step of selecting m groups of control parameters from the parameter set includes:

[0113] Step S521: preset the neighborhood distance and the number of neighbors; the neighborhood distance and the number of neighbors are preset by those skilled in the art according to actual conditions;

[0114] Step S522: Take each set of analysis parameters in the parameter set as a node and calculate the node distance between every two nodes; the expression of the node distance is: Where, L ij is the node distance between the i-th node and the j-th node, x ir is the rth parameter in the analysis parameters corresponding to the i-th node, x jr is the rth parameter in the analysis parameters corresponding to the jth node, r∈[1,3];

[0115] Step S523: construct the neighborhood of each node according to the neighborhood distance; count the number of nodes in the neighborhood corresponding to each node according to the node distance, and mark them as the number of neighboring points; that is, count the number of node distances corresponding to each node whose value is less than or equal to the neighborhood distance;

[0116] Step S524: Mark the node whose number of neighbors is greater than or equal to the number of neighbors as a center point;

[0117] Step S525: randomly select a center point that is not marked as a selected point, mark it as a selected point, and update the current point to a selected point; construct a new cluster and mark it as the current cluster; add the current point and all nodes in the neighborhood corresponding to the current point to the current cluster;

[0118] Step S526: Mark all the center points in the current cluster except the current point as extension points, and add all the nodes in the neighborhood corresponding to the extension point to the current cluster;

[0119] Step S527: looping step S526 until all nodes in the neighborhood corresponding to the extension point in the current cluster are added to the current cluster, the loop ends, and the current cluster is marked as a completed cluster;

[0120] Step S528: looping steps S525 to S527 until all center points are marked as selected points, the loop ends, all completed clusters are obtained, and technicians in this field match the completed clusters with the working modes one by one according to actual experience;

[0121] Step S529: According to the determined working mode, m groups of analysis parameters in the corresponding completed cluster are obtained and marked as control parameters.

[0122] In the above step S53, the method for calculating the comfort corresponding to each set of control parameters includes:

[0123] Each set of control parameters and feature replacement data is taken as a set of test parameters, and the test parameters correspond to the control parameters one by one; each set of test parameters is input into the trained comfort assessment model to predict the corresponding comfort; the training process of the comfort assessment model is consistent with the training process of the quality assessment model, and both are deep neural network models.

[0124] In the above step S54, the method of selecting S groups of selection parameters from m groups of control parameters includes:

[0125] According to comfort, m groups of control parameters are sorted from large to small to generate a parameter sorting table; a deviation threshold is preset, and the deviation threshold is pre-set by technical personnel in the field according to actual conditions; the control parameter ranked first in the parameter sorting table is marked as a selection parameter, and the control parameter not marked as a selection parameter is marked as a screening parameter; the comfort of each screening parameter is subtracted from the comfort of the selection parameter in turn to obtain a comfort deviation; the comfort deviation is compared with the deviation threshold, and the screening parameter corresponding to the comfort deviation less than or equal to the deviation threshold is marked as a selection parameter, and the screening parameter corresponding to the comfort deviation greater than the deviation threshold is not marked.

[0126] In the above step S55, the method for calculating the energy consumption value corresponding to each set of selection parameters includes:

[0127] Each set of test parameters is input into the trained energy consumption assessment model to predict the corresponding energy consumption value; the training process of the energy consumption assessment model is consistent with the training process of the quality assessment model, and both are deep neural network models.

[0128] This embodiment automatically determines whether to turn on the aromatherapy machine by collecting pressure distribution data; and uses fuzzy logic analysis to determine the appropriate working mode by collecting various environmental data and driving characteristics in the car in real time; and then through the steps of parameter screening, comfort and energy consumption evaluation, it realizes dynamic regulation of the aromatherapy parameters of the aromatherapy machine to improve the air quality in the car, create a comfortable and fresh car environment for the driver, and thus enhance the user's driving experience; it not only realizes personalized fragrance selection, but also reduces the safety hazards caused by traditional manual operation, has little impact on driving safety, and improves the convenience and comfort during driving.

[0129] Example 2

[0130] This embodiment discloses an aromatherapy machine, including a storage unit, a central processing unit, and a computer program stored in the storage unit and executable on the central processing unit. When the central processing unit executes the computer program, the operation mode of the above-mentioned intelligent control method of an aromatherapy machine based on a wireless network hotspot is implemented.

[0131] Example 3

[0132] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the central processing unit, an intelligent control method for an aromatherapy machine based on a wireless network hotspot according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, a volatile storage unit and / or a non-volatile storage unit. The volatile storage unit may include, for example, a random access memory unit (RAM) and a cache memory unit (cache). The non-volatile storage unit may include, for example, a read-only memory unit (ROM), a hard disk, a flash memory, etc.

[0133] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be run by a central processing unit to execute instructions corresponding to the method steps provided by the present application, for example: a method for intelligent control of an aromatherapy machine based on a wireless network hotspot. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0134] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0135] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent control method for an aromatherapy machine based on a wireless network hotspot, characterized in that: include: S1: Collect pressure distribution data and determine whether to generate an opening instruction; S2: If an opening command is generated, the driving characteristic data is collected in real time; S3: Process the driving feature data to obtain feature processing data; S4: Analyze the feature processing data and determine the working mode; S5: Regulate aromatherapy parameters according to feature processing data and working mode.

2. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 1 is characterized in that: The pressure distribution data is pressure distribution information on the car seat, and the pressure distribution data includes n pressure data, where n is an integer greater than 1; The method for determining whether to generate an opening instruction includes: Add n pressure data in sequence and divide by n to obtain the pressure average value; preset a pressure threshold and compare the pressure average value with the pressure threshold; if the pressure average value is greater than or equal to the pressure threshold, generate a preliminary opening instruction; if the pressure average value is less than the pressure threshold, do not generate a preliminary opening instruction; If a preliminary opening instruction is generated, the distribution distance is obtained; the distribution distance is the distance between every two adjacent pressure sensors; the pressure data corresponding to every two adjacent pressure sensors are taken as a group of pressure sets, and the pressure difference corresponding to each group of pressure sets is calculated in turn; the expression of the pressure difference is: ΔP = |P2-P1|; where ΔP is the pressure difference, P2 is a pressure data in the pressure set, and P1 is another pressure data in the pressure set; each pressure difference is divided by the corresponding distribution distance to obtain the pressure gradient; all pressure gradients are added in turn, and then divided by the number of distances to obtain the pressure gradient mean; a gradient threshold is preset, and the pressure gradient mean is compared with the gradient threshold; if the pressure gradient mean is greater than or equal to the gradient threshold, an opening instruction is generated; if the pressure gradient mean is less than the gradient threshold, no opening instruction is generated.

3. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 2 is characterized in that: The driving characteristic data includes environmental data and time data; the environmental data includes temperature, humidity, gas concentration and wind speed; the time data includes driving time and driving duration; The method for obtaining feature processing data comprises: A group of historical feature data is obtained, where the historical feature data is driving feature data collected at historical moments, and a is an integer greater than 1; the driving feature data collected in real time is marked as real-time feature data, anomaly detection is performed on each data in the real-time feature data, the data with anomalies is marked as abnormal data, the abnormal data is re-collected, and the abnormal data in the real-time feature data is replaced with the corresponding re-collected data to obtain feature processing data; wherein the anomaly detection method for each data in the real-time feature data is consistent; the method for detecting anomaly of temperature includes: Preset the number of analyses b; use the real-time characteristic data and the historical characteristic data as the analysis characteristic data, subtract the corresponding neighbor temperature from each temperature in the analysis characteristic data, obtain the temperature difference corresponding to each temperature, and the neighbor temperature corresponding to the temperature is the remaining a temperatures in the analysis characteristic data; use all the temperature differences corresponding to each temperature as a temperature set, and the temperature set corresponds to the temperature one by one; sort the temperature differences in each temperature set from small to large, and generate a temperature sorting table corresponding to each temperature set; use the temperature difference ranked in the bth place in each temperature sorting table as the reference distance of the temperature corresponding to the corresponding temperature set, and use the neighbor temperature corresponding to the first b temperature differences in each temperature sorting table as the standard temperature of the temperature corresponding to the corresponding temperature set; According to the reference distance of each temperature and the corresponding temperature difference, the adjacent distance between each temperature and the corresponding standard temperature is determined; according to the adjacent distance between each temperature and the corresponding standard temperature, the local density of each temperature is calculated; the temperature in the real-time characteristic data is marked as the real-time temperature, and the discrete coefficient ls corresponding to the real-time temperature is calculated according to the adjacent distance between each temperature and the corresponding standard temperature; the judgment coefficient d is preset, 0<d<1; if |ls-1|≥d, the real-time temperature is marked as abnormal data; if |ls-1|<d, the real-time temperature is not marked.

4. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 3 is characterized in that: The expression of adjacent distance is: xl(p,q)=max(jl(p),wc(p,q)); where xl(p,q) is the adjacent distance between the pth temperature in the analysis feature data and the corresponding qth standard temperature, max is the maximum value function, jl(p) is the reference distance of the pth temperature in the analysis feature data, wc(p,q) is the temperature difference between the pth temperature in the analysis feature data and the corresponding qth standard temperature, p∈[1,a+1], q∈[1,b]; The expression of local density is: Where md(p) is the local density of the pth temperature in the analyzed characteristic data; The expression of the coefficient of dispersion is: Where md′ is the local density corresponding to the real-time temperature, and md(q) is the local density of the qth standard temperature corresponding to the real-time temperature.

5. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 4 is characterized in that: The step of determining the working mode comprises: Step S401: Analyze the gas concentration in the feature processing data to evaluate the air quality; replace the gas concentration in the feature processing data with the air quality to obtain feature replacement data; Step S402: construct multiple fuzzy sets for each data in the feature replacement data; Step S403: converting each data in the feature replacement data into the membership degree of each corresponding fuzzy set by using fuzzification technology; Step S404: define fuzzy rules; Step S405: matching the fuzzified feature replacement data with the fuzzy rules, performing fuzzy reasoning, and obtaining fuzzy reasoning results, where the fuzzy reasoning results are the membership degrees corresponding to each working mode; Step S406: Compare each degree of membership in the fuzzy inference result, and take the working mode corresponding to the degree of membership with the largest value as the working mode corresponding to the feature processing data; In step S401, the method for evaluating air quality includes: The gas concentration in the feature processing data is used as the analysis data, and the analysis data is input into the trained quality assessment model to predict the corresponding air quality; the training process of the quality assessment model includes: G groups of analysis data are collected in advance, and corresponding air qualities are set for the g groups of analysis data, where g is an integer greater than 1, and the analysis data and the corresponding air qualities are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a quality assessment model, and the quality assessment model uses a set of predicted air qualities corresponding to each set of analysis data as output, and uses the actual air quality corresponding to each set of analysis data as a prediction target, where the actual air quality is the pre-set air quality corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as a training target; the quality assessment model is trained until the sum of prediction errors converges and the training is stopped; the quality assessment model is a deep neural network model.

6. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 5 is characterized in that: The aromatherapy parameters include aromatherapy concentration, working time and release interval; the aromatherapy concentration is the amount of aroma released, the working time is the duration of continuous working after the aromatherapy machine is turned on each time, and the release interval is the time interval of aroma release; The step of regulating the aromatherapy parameters comprises: Step S51: constructing a parameter set, the parameter set including M groups of analysis parameters; Step S52: According to the working mode, m groups of control parameters are selected from the parameter set, where 1<m<M; Step S53: Calculate the comfort level corresponding to each set of control parameters; Step S54: based on comfort, select S groups of selection parameters from m groups of control parameters, 1≤S<m; Step S55: if S=1, the aromatherapy parameters are regulated according to the selected parameters; if S>1, the energy consumption value corresponding to each set of selected parameters is calculated; Step S56: sort the S group of selection parameters from large to small according to the energy consumption value, mark the selection parameter at the front as the best parameter, and adjust the aromatherapy parameter according to the best parameter.

7. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 6 is characterized in that: In step S51, the method for constructing the parameter set is: obtaining a parameter range, the parameter range including the range corresponding to each parameter in the aromatherapy parameter; randomly selecting a value from each range within the parameter range to construct a set of analysis parameters, constructing a total of M groups of analysis parameters, where M is an integer greater than 1, and the M groups of analysis parameters are all different; In step S53, the method for calculating the comfort corresponding to each set of control parameters includes: Each set of control parameters and feature replacement data is used as a set of test parameters, and the test parameters correspond to the control parameters one by one; Each set of test parameters is input into the trained comfort assessment model to predict the corresponding comfort; the training process of the comfort assessment model is consistent with that of the quality assessment model, and both are deep neural network models.

8. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 7 is characterized in that: In step S52, the step of selecting m groups of control parameters from the parameter set includes: Step S521: preset the neighborhood distance and the number of neighbors; Step S522: Take each set of analysis parameters in the parameter set as a node and calculate the node distance between every two nodes; the expression of the node distance is: Where, L ij is the node distance between the i-th node and the j-th node, x ir is the rth parameter in the analysis parameters corresponding to the i-th node, x jr is the rth parameter in the analysis parameters corresponding to the jth node, r∈[1,3]; Step S523: construct the neighborhood of each node according to the neighborhood distance; count the number of nodes in the neighborhood corresponding to each node according to the node distance, and mark them as the number of neighboring points; that is, count the number of node distances corresponding to each node whose value is less than or equal to the neighborhood distance; Step S524: Mark the node whose number of neighbors is greater than or equal to the number of neighbors as a center point; Step S525: randomly select a center point that is not marked as a selected point, mark it as a selected point, and update the current point to a selected point; construct a new cluster and mark it as the current cluster; add the current point and all nodes in the neighborhood corresponding to the current point to the current cluster; Step S526: Mark all the center points in the current cluster except the current point as extension points, and add all the nodes in the neighborhood corresponding to the extension point to the current cluster; Step S527: looping step S526 until all nodes in the neighborhood corresponding to the extension point in the current cluster are added to the current cluster, the loop ends, and the current cluster is marked as a completed cluster; Step S528: looping steps S525 to S527 until all center points are marked as selected points, the loop ends, and all completed clusters are obtained, and the completed clusters correspond to the working modes one by one; Step S529: According to the determined working mode, m groups of analysis parameters in the corresponding completed cluster are obtained and marked as control parameters.

9. The intelligent control method of aromatherapy machine based on wireless network hotspot according to claim 8 is characterized in that: In step S54, the method of selecting S groups of selection parameters from m groups of control parameters includes: According to comfort, m groups of control parameters are sorted from large to small to generate a parameter sorting table; a deviation threshold is preset; the control parameter at the front of the parameter sorting table is marked as a selection parameter, and the control parameter not marked as a selection parameter is marked as a screening parameter; the comfort of each screening parameter is subtracted from the comfort of the selection parameter in turn to obtain a comfort deviation; the comfort deviation is compared with the deviation threshold, and the screening parameter corresponding to the comfort deviation less than or equal to the deviation threshold is marked as a selection parameter, and the screening parameter corresponding to the comfort deviation greater than the deviation threshold is not marked; In step S55, the method for calculating the energy consumption value corresponding to each set of selection parameters includes: Each set of test parameters is input into the trained energy consumption assessment model to predict the corresponding energy consumption value; the training process of the energy consumption assessment model is consistent with the training process of the quality assessment model, and both are deep neural network models.

10. An aromatherapy machine, comprising a storage unit, a central processing unit, and a computer program stored in the storage unit and executable on the central processing unit, characterized in that: When the central processing unit executes the computer program, the intelligent control method of the aromatherapy machine based on the wireless network hotspot described in any one of claims 1 to 9 is implemented.

Images