Citrus aurantium flower tea drying monitoring system based on artificial intelligence technology

Abstract

The invention discloses a citrus aurantium flower tea drying monitoring system based on an artificial intelligence technology. The citrus aurantium flower tea drying monitoring system comprises a sensing layer, a transmission layer, a control layer and an application layer. Wherein the control layer comprises a drying prediction unit and an intelligent control unit; the moisture content prediction unit is used for training a plurality of citrus aurantium flower tea moisture content prediction models by using a neural network and predicting the moisture content of the citrus aurantium flower tea; the intelligent control unit adopts a dynamic Bayesian algorithm to determine a model weight corresponding to prediction of each citrus aurantium flower tea water content prediction model, reconstructs the predicted citrus aurantium flower tea water content according to the model weight, and predicts the aroma locking degree of the citrus aurantium flower tea by utilizing a logistic regression model according to the reconstructed and predicted citrus aurantium flower tea water content; according to the predicted aroma locking degree and moisture content of the citrus aurantium flower tea, the temperature and humidity in the drying process are controlled and adjusted through a fuzzy control method. The operation strategy of the drying equipment is dynamically adjusted, and the drying efficiency and the tea quality consistency are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of tea processing, and in particular to a tea drying monitoring system based on artificial intelligence technology. Background Art

[0002] Daidaihua is the dried flower bud of the Rutaceae plant Daidaihua. Daidaihua has thick petals (1.5-2mm), a thick wax layer, slow water migration, and contains more volatile oils and flavonoids, which are easily decomposed at high temperatures. This makes it very demanding to control the temperature during the drying process of Daidaihua tea. Too high a temperature may destroy the structure and effective ingredients of the petals, such as volatile oils and flavonoids, which may affect the aroma and efficacy. If the temperature is too low, the drying may not be thorough and it is easy to mold. Therefore, the temperature needs to be adjusted in stages, such as a higher temperature in the early stage to remove moisture, and a lower temperature in the later stage to retain the ingredients. In addition, during the drying process, there are also high requirements for the appearance and color of Daidaihua tea.

[0003] The traditional flower tea drying process relies on manual experience and drying equipment with fixed parameters. It is difficult to achieve real-time monitoring and dynamic adjustment of the drying process, resulting in unstable drying effects and uneven quality of scented tea. Traditional drying equipment usually relies on temperature and humidity sensors to control the temperature and humidity during the drying process of scented tea. However, the lack of multi-dimensional data perception such as tea morphology and aroma components can easily lead to problems such as low moisture uniformity, aroma loss, browning of petals, easy sandwich moisture in thick petals, and high risk of mildew. Summary of the invention

[0004] The purpose of the present invention is to provide a drying monitoring system for chrysanthemum tea based on artificial intelligence technology. By predicting the moisture content and fragrance locking degree of chrysanthemum tea, the drying process of chrysanthemum tea is fuzzy controlled in stages in advance, thereby realizing precise dynamic regulation of the drying process of chrysanthemum tea.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A drying monitoring system for daidai flower tea based on artificial intelligence technology, comprising:

[0007] The perception layer uses sensors to collect the environmental parameters of the industrial tea dryer for drying the tea in real time; the environmental parameters include drying temperature, drying humidity, tea moisture and aroma level;

[0008] The transport layer uses wired and / or wireless methods to transmit data within the system;

[0009] A control layer, comprising a drying prediction unit and an intelligent control unit;

[0010] The moisture content prediction unit uses a neural network to train multiple Dai Dai flower tea moisture content prediction models, and uses each Dai Dai flower tea moisture content prediction model to predict the moisture content of the Dai Dai flower tea; the neural network includes a recurrent neural network, a deep neural network, a long short-term memory network (LSTM), a Transformer model, etc.;

[0011] The intelligent control unit uses a dynamic Bayesian algorithm to determine the model weight corresponding to the prediction of each water content prediction model of the daidai flower tea, and reconstructs the predicted water content of the daidai flower tea according to the model weight to obtain the reconstructed water content of the daidai flower tea;

[0012] The fragrance-locking degree of Daidaihua tea is predicted by logistic regression model according to the reconstructed predicted moisture content of Daidaihua tea.

[0013] According to the predicted fragrance-locking degree and moisture content of the tea, the temperature and humidity in the drying process are controlled and adjusted using the fuzzy control method.

[0014] The application layer provides real-time data display, historical data query, drying process monitoring and alarm prompt functions. The application layer can compare the monitored data and predicted data with the standard data to achieve early warning.

[0015] According to the above technical solution, the moisture content prediction model of the daidai flower tea takes the drying temperature, drying humidity, drying wind speed, the initial moisture content of the daidai flower tea, and the prediction time as input, and the moisture content as output.

[0016] According to the above technical solution, the water content of the reconstructed daidai flower tea is:

[0017] θ t =a1f1(t)+a2f2(t)+…+a k f k (t);

[0018] In the formula, θ t represents the reconstructed Dai Dai flower tea moisture content at time t, a1 represents the weight corresponding to the first Dai Dai flower tea moisture content prediction model, a2 represents the weight corresponding to the second Dai Dai flower tea moisture content prediction model, a k represents the weights corresponding to the k prediction models of the moisture content of daidai tea, f1(t) represents the moisture content of daidai tea at time t predicted by the first prediction model of the moisture content of daidai tea, f2(t) represents the moisture content of daidai tea at time t predicted by the second prediction model of the moisture content of daidai tea, f k (t) represents the moisture content of the tea at time t predicted by the kth tea moisture content prediction model.

[0019] Reconstructing the predicted moisture content of the tea according to the model weight can improve the accuracy of the predicted moisture content of the tea.

[0020] According to the above technical solution, a logistic regression model is used to predict the fragrance-locking degree of the daidai flower tea; the logistic regression model:

[0021]

[0022] In the formula, β0 represents the intercept term, β1 represents the weight corresponding to the drying temperature feature, T represents the drying temperature, β2 represents the weight corresponding to the drying time feature, t represents the drying time, β3 represents the weight corresponding to the ambient humidity feature, H represents the ambient humidity, β4 represents the weight corresponding to the moisture content feature of the reconstructed Dai Dai flower tea at time t, and M t Represents the reconstructed moisture content of Daidai flower tea at time t, β5 represents the weight corresponding to the moisture content decrease feature, ΔM represents the moisture content decrease, and M0 represents the initial moisture content of Daidai flower tea.

[0023] Most scented teas pay attention to controlling the temperature and humidity during the drying process by monitoring the moisture content of the scented tea. In fact, the quality of scented tea is not only related to the drying conditions (moisture content) of the tea leaves, but also the degree of aroma locking of the tea leaves. Therefore, in the process of drying tea leaves, the moisture content and the degree of aroma locking can be considered to control the drying temperature and humidity.

[0024] Daidaihua belongs to the Rutaceae family. Its petals are relatively thick and contain volatile oils and aromatic substances. Therefore, special attention should be paid to temperature and humidity when drying to avoid loss of aroma. In the drying process of Daidaihua tea, the first thing is temperature control. In the early stage, low temperature is required to lock the aroma, and then the temperature is gradually increased for dehydration. In terms of humidity, different humidity controls are required at different stages, such as high humidity in the fragrance fixing stage and low humidity in the dehydration stage. The thickness of the petals must also be considered, which may require a longer drying time, and the retention of aroma components is the key, which can be detected using an electronic nose and near-infrared spectroscopy. Therefore, the drying stage of Daidaihua tea is divided into three stages: preheating stage, dehydration stage, and fragrance fixing stage, and they are controlled separately.

[0025] According to the above technical solution, the steps of controlling the temperature and humidity in the drying process by the fuzzy control method include:

[0026] Determine the drying stage of the Dai Dai flower tea, the drying stage of the Dai Dai flower tea is divided into three stages: preheating stage, dehydration stage and fragrance fixing stage;

[0027] The fuzzy control method for the preheating stage controls the temperature and humidity during the drying process of the tea according to the preheating fuzzy rule base; the trapezoidal membership function for fuzzifying the degree of locking fragrance in the preheating stage is (0%, 0%, 20%, 40%); this is because the preheating stage is mainly to allow the tea to absorb moisture and make preliminary preparations for locking fragrance. The membership function of moisture content is (40%, 50%, 70%, 80%).

[0028] The dehydration stage fuzzy control method controls the temperature and humidity during the drying process of bitter orange flower tea according to the dehydration fuzzy rule base; the trapezoidal membership function of the fragrance-locking degree in the dehydration stage is (20%, 40%, 60%, 80%). The dehydration stage is the key stage for fragrance locking. As the moisture is removed, the fragrance-locking degree gradually increases. The membership function of the moisture content is (0%, 10%, 20%, 30%); in the dehydration stage, the moisture content needs to be gradually reduced.

[0029] The flavor-fixing stage fuzzy control method controls the temperature and humidity during the drying process of bitter orange flower tea according to the flavor-fixing fuzzy rule base; the trapezoidal membership function of the fragrance-locking degree in the flavor-fixing stage is (60%, 80%, 100%, 100%). The flavor-fixing stage aims to make the tea reach a relatively high fragrance-locking degree. When the fragrance-locking degree reaches 80% or above, it is considered to achieve an ideal fragrance-locking effect. The membership function of the moisture content is (0%, 7%, 10%, 20%). The flavor-fixing stage requires a relatively low moisture content to achieve the best fragrance-locking effect.

[0030] According to the above technical solution, the preheating fuzzy rule base:

[0031]

[0032] In the table, SL1 represents the fragrance-locking degree of bitter orange flower tea in the preheating stage, HS1 represents the moisture content of bitter orange flower tea in the preheating stage, H h1 represents that the fragrance-locking degree of bitter orange flower tea in the preheating stage is divided into a high fragrance-locking degree after fuzzyfication, M h1 represents that the fragrance-locking degree of bitter orange flower tea in the preheating stage is divided into a medium fragrance-locking degree after fuzzyfication, L h1 represents that the fragrance-locking degree of bitter orange flower tea in the preheating stage is divided into a low fragrance-locking degree after fuzzyfication, H s1 represents that the moisture content of bitter orange flower tea in the preheating stage is divided into a high moisture content after fuzzyfication, M s1 represents that the moisture content of bitter orange flower tea in the preheating stage is divided into a medium moisture content after fuzzyfication, L s1 represents that the moisture content of bitter orange flower tea in the preheating stage is divided into a low moisture content after fuzzyfication, PB T1 represents that the control amount of the temperature in the preheating stage is positive large, PB H1 It shows that the control amount of the humidity in the preheating stage is positive large, ZO represents the control amount of the temperature or humidity in the preheating stage, NB T1 represents that the control amount of the temperature in the preheating stage is negative large, NB H1 represents that the control amount of the humidity in the preheating stage is negative large.

[0033] According to the above technical solution,

[0034]

[0035] In the table, SL2 represents the fragrance-locking degree of the bitter orange flower tea during the dehydration stage, HS2 represents the moisture content of the bitter orange flower tea during the dehydration stage, and H h2 represents that the fragrance-locking degree of the bitter orange flower tea during the dehydration stage is classified as a high fragrance-locking degree after fuzzification, and M h2 represents that the fragrance-locking degree of the bitter orange flower tea during the dehydration stage is classified as a medium fragrance-locking degree after fuzzification, and L h2 represents that the fragrance-locking degree of the bitter orange flower tea during the dehydration stage is classified as a low fragrance-locking degree after fuzzification, and H s2 represents that the moisture content of the bitter orange flower tea during the dehydration stage is classified as a high moisture content after fuzzification, and M s2 represents that the moisture content of the bitter orange flower tea during the dehydration stage is classified as a medium moisture content after fuzzification, and L s2 represents that the moisture content of the bitter orange flower tea during the dehydration stage is classified as a low moisture content after fuzzification, and PB T2 represents that the control amount of the temperature during the dehydration stage is positive large, and PB H2 indicates that the control amount of the humidity during the dehydration stage is positive large, ZO represents the control amount of the temperature or humidity during the dehydration stage, and NB T2 represents that the control amount of the temperature during the dehydration stage is negative large, and NB H2 represents that the control amount of the humidity during the dehydration stage is negative large.

[0036] According to the above technical solution, the fragrance-fixing fuzzy rule base:

[0037]

[0038] In the table, SL3 represents the fragrance-locking degree of the bitter orange flower tea during the fragrance-fixing stage, HS3 represents the moisture content of the bitter orange flower tea during the fragrance-fixing stage, and H h3 represents that the fragrance-locking degree of the bitter orange flower tea during the fragrance-fixing stage is classified as a high fragrance-locking degree after fuzzification, and M h3 represents that the fragrance-locking degree of the bitter orange flower tea during the fragrance-fixing stage is classified as a medium fragrance-locking degree after fuzzification, and L h3 represents that the fragrance-locking degree of the bitter orange flower tea during the fragrance-fixing stage is classified as a low fragrance-locking degree after fuzzification, and H s3 represents that the moisture content of the bitter orange flower tea during the fragrance-fixing stage is classified as a high moisture content after fuzzification, and M s3 represents that the moisture content of the bitter orange flower tea during the fragrance-fixing stage is classified as a medium moisture content after fuzzification, and L s3 represents that the moisture content of the bitter orange flower tea during the fragrance-fixing stage is classified as a low moisture content after fuzzification, and PB T3 represents that the control amount of the temperature during the fragrance-fixing stage is positive large, and PB H3 indicates that the control amount of the humidity during the fragrance-fixing stage is positive large, ZO represents the control amount of the temperature or humidity during the fragrance-fixing stage, and NB T3 represents that the control amount of the temperature during the fragrance-fixing stage is negative large, and NB H3 represents that the control amount of the humidity during the fragrance-fixing stage is negative large.

[0039] According to the above technical solution, the temperature adjustment amount in the preheating stage is [-5°C, +5°C], and the humidity adjustment amount ΔRH is

[0040] [-10%,+10%];

[0041] The temperature adjustment amount ΔT in the dehydration stage is [-10°C, +10°C], and the humidity adjustment amount ΔRH is [-20%, 0%];

[0042] The temperature adjustment amount ΔT in the fragrance fixing stage is [-5°C, +5°C], and the humidity adjustment amount ΔRH is [-10%, 0%].

[0043] Also included is another embodiment, a storage medium for storing computer executable instructions, wherein the computer executable instructions, when executed, implement a drying monitoring system for daidai flower tea based on artificial intelligence technology as described in any one of the above technical solutions.

[0044] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention describes the drying conditions of the tea in the future by predicting the moisture content and fragrance-locking degree of the tea during the drying process, and dynamically regulates the temperature and humidity in the preheating stage, dehydration stage, and fragrance-fixing stage of the tea according to the predicted moisture content and fragrance-locking degree of the tea using a fuzzy control method. The present invention dynamically adjusts the operation strategy of the drying equipment, significantly improving the drying efficiency and tea quality consistency. The present invention is suitable for industrialized tea production, and has the characteristics of intelligence, self-adaptation, and low energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0046] Figure 1 It is a structural schematic diagram of a drying monitoring system for daidai flower tea based on artificial intelligence technology of the present invention. DETAILED DESCRIPTION

[0047] 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.

[0048] The present invention provides a technical solution:

[0049] A drying monitoring system for daidai flower tea based on artificial intelligence technology, characterized by comprising:

[0050] The perception layer uses sensors to collect in real time the environmental parameters of the industrial tea dryer for drying the bitter orange flower tea; the environmental parameters include the drying temperature, drying humidity, tea moisture content, and aroma level.

[0051] The transmission layer uses wired and / or wireless means to transmit data within the system.

[0052] The control layer includes a drying prediction unit and an intelligent control unit;

[0053] The moisture content prediction unit uses neural networks to train multiple moisture content prediction models for bitter orange flower tea, and uses each moisture content prediction model for bitter orange flower tea to predict the moisture content of the bitter orange flower tea; the moisture content prediction model for bitter orange flower tea takes the drying temperature, drying humidity, drying wind speed, initial moisture content of bitter orange flower tea, and prediction time as inputs, and the moisture content as the output.

[0054] The intelligent control unit uses the dynamic Bayesian algorithm to determine the model weight corresponding to each moisture content prediction model for bitter orange flower tea, and reconstructs the predicted moisture content of the bitter orange flower tea according to the model weight to obtain the reconstructed moisture content of the bitter orange flower tea;

[0055] Reconstructed moisture content of bitter orange flower tea:

[0056] θ t = a1f1(t) + a2f2(t) + … + a k f k (t);

[0057] In the formula, θ t represents the reconstructed moisture content of the bitter orange flower tea at time t, a1 represents the weight corresponding to the first moisture content prediction model for bitter orange flower tea, a2 represents the weight corresponding to the second moisture content prediction model for bitter orange flower tea, a k represents the weight corresponding to the kth moisture content prediction model for bitter orange flower tea, f1(t) represents the moisture content of the bitter orange flower tea predicted by the first moisture content prediction model for bitter orange flower tea at time t, f2(t) represents the moisture content of the bitter orange flower tea predicted by the second moisture content prediction model for bitter orange flower tea at time t, f k (t) represents the moisture content of the bitter orange flower tea predicted by the kth moisture content prediction model for bitter orange flower tea at time t.

[0058] According to the reconstructed predicted moisture content of the bitter orange flower tea, use the logistic regression model to predict the aroma locking degree of the bitter orange flower tea; Logistic regression model:

[0059]

[0060] In the formula, β0 represents the intercept term, β1 represents the weight corresponding to the drying temperature feature, T represents the drying temperature, β2 represents the weight corresponding to the drying time feature, t represents the drying time, β3 represents the weight corresponding to the ambient humidity feature, H represents the ambient humidity, β4 represents the weight corresponding to the moisture content feature of the reconstructed Dai Dai flower tea at time t, and M t Represents the reconstructed moisture content of Daidai flower tea at time t, β5 represents the weight corresponding to the moisture content decrease feature, ΔM represents the moisture content decrease, and M0 represents the initial moisture content of Daidai flower tea.

[0061] For example: the initial moisture content is 78%, the drying temperature is 80℃, the time is 60min, the ambient humidity is 35%, the reconstructed predicted moisture content of the daidai flower tea is 23%, ΔM=55%; after training the logistic regression model, β0=-3.2, β1=1.58, β2=0.85, β3=-1.22, β4=-2.13, β5=1.0 are obtained; model output: P≈32%.

[0062] According to the predicted fragrance-locking degree and moisture content of the chrysanthemum tea, the fuzzy control method is used to control the temperature and humidity in the drying process.

[0063] The application layer provides real-time data display, historical data query, drying process monitoring and alarm prompt functions.

[0064] The steps of controlling the temperature and humidity in the drying process by the fuzzy control method include:

[0065] Determine the drying stage of the Dai Dai flower tea, the drying stage of the Dai Dai flower tea is divided into three stages: preheating stage, dehydration stage and fragrance fixing stage;

[0066] The fuzzy control method in the preheating stage controls the temperature and humidity in the drying process of Dai Dai flower tea according to the preheating fuzzy rule base; the trapezoidal membership function for fuzzifying the degree of fragrance locking in the preheating stage is (0%, 0%, 20%, 40%), and the membership function of moisture content is (40%, 50%, 70%, 80%); the temperature adjustment amount in the preheating stage is [-5℃, +5℃], and the humidity adjustment amount ΔRH is

[0067] [-10%,+10%].

[0068] Among them, preheat the fuzzy rule base:

[0069]

[0070] In the table, SL1 represents the degree of aroma locking of Dai Dai flower tea in the preheating stage, HS1 represents the moisture content of Dai Dai flower tea in the preheating stage, and H h1 The degree of aroma locking of Dai Dai flower tea in the preheating stage is fuzzy and divided into high aroma locking degree, M h1It indicates that the fragrance-locking degree of the bitter orange flower tea in the preheating stage is divided into medium fragrance-locking degree after fuzzification, L h1 It indicates that the fragrance-locking degree of the bitter orange flower tea in the preheating stage is divided into low fragrance-locking degree after fuzzification, H s1 It indicates that the moisture content of the bitter orange flower tea in the preheating stage is divided into high water content after fuzzification, M s1 It indicates that the moisture content of the bitter orange flower tea in the preheating stage is divided into medium water content after fuzzification, L s1 It indicates that the moisture content of the bitter orange flower tea in the preheating stage is divided into low water content after fuzzification, PB T1 It indicates that the control amount of the temperature in the preheating stage is positive large, PB H1 It indicates that the control amount of the humidity in the preheating stage is positive large. ZO indicates the control amount of the temperature or humidity in the preheating stage, NB T1 It indicates that the control amount of the temperature in the preheating stage is negative large, NB H1 It indicates that the control amount of the humidity in the preheating stage is negative large.

[0071] The fuzzy control method in the dehydration stage controls the temperature and humidity during the drying process of the bitter orange flower tea according to the dehydration fuzzy rule base; in the dehydration stage, the trapezoidal membership function of the fragrance-locking degree is (20%, 40%, 60%, 80%), and the membership function of the moisture content is (0%, 10%, 20%, 30%); the temperature adjustment amount ΔT in the dehydration stage is [-10°C, +10°C], and the humidity adjustment amount ΔRH is [-20%, 0%].

[0072] Among them, the dehydration fuzzy rule base:

[0073]

[0074] In the table, SL2 represents the fragrance-locking degree of the bitter orange flower tea in the dehydration stage, and HS2 represents the moisture content of the bitter orange flower tea in the dehydration stage, H h2 It indicates that the fragrance-locking degree of the bitter orange flower tea in the dehydration stage is divided into high fragrance-locking degree after fuzzification, M h2 It indicates that the fragrance-locking degree of the bitter orange flower tea in the dehydration stage is divided into medium fragrance-locking degree after fuzzification, L h2 It indicates that the fragrance-locking degree of the bitter orange flower tea in the dehydration stage is divided into low fragrance-locking degree after fuzzification, H s2 It indicates that the moisture content of the bitter orange flower tea in the dehydration stage is divided into high water content after fuzzification, M s2 It indicates that the moisture content of the bitter orange flower tea in the dehydration stage is divided into medium water content after fuzzification, L s2 It indicates that the moisture content of the bitter orange flower tea in the dehydration stage is divided into low water content after fuzzification, PB T2 It indicates that the control amount of the temperature in the dehydration stage is positive large, PB H2 It indicates that the control amount of the humidity in the dehydration stage is positive large. ZO indicates the control amount of the temperature or humidity in the dehydration stage, NB T2 It indicates that the control amount of the temperature in the dehydration stage is negative large, NBH2 It means that the control amount of humidity in the dehydration stage is negative.

[0075] The fuzzy control method in the fragrance fixing stage controls the temperature and humidity in the drying process of daidai flower tea according to the fragrance fixing fuzzy rule base; the trapezoidal membership function of the fragrance locking degree in the fragrance fixing stage is (60%, 80%, 100%, 100%), and the membership function of the moisture content is (0%, 7%, 10%, 20%); the temperature adjustment amount ΔT in the fragrance fixing stage is [-5℃, +5℃], and the humidity adjustment amount ΔRH is [-10%, 0%].

[0076] Among them, the fixed fragrance fuzzy rule base:

[0077]

[0078] In the table, SL3 represents the degree of scent locking of Daidai flower tea in the scent fixing stage, HS3 represents the moisture content of Daidai flower tea in the scent fixing stage, and H h3 The degree of aroma locking of Dai Dai flower tea in the aroma fixing stage is fuzzy and divided into high aroma locking degree, M h3 The degree of aroma locking of Dai Dai flower tea in the aroma fixing stage is fuzzy and divided into medium aroma locking degree, L h3 The degree of aroma locking of Dai Dai flower tea in the aroma fixing stage is fuzzy and divided into low aroma locking degree, H s3 The moisture content of Dai Dai flower tea in the fixed fragrance stage is fuzzy and divided into high moisture content, M s3 The moisture content of the tea in the fixed fragrance stage is fuzzy and divided into medium moisture content, L s3 The moisture content of Dai Dai flower tea in the fixed fragrance stage is fuzzy and divided into low moisture content, PB T3 Indicates that the temperature control amount in the aroma setting stage is positive, PB H3 The control amount of humidity in the fragrance setting stage is positive, ZO represents the control amount of temperature or humidity in the fragrance setting stage, NB T3 Indicates that the temperature control amount in the aroma setting stage is large, NB H3 It means that the control amount of humidity in the fragrance setting stage is negative and large.

[0079] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0080] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A drying monitoring system for Daidai flower tea based on artificial intelligence technology, characterized in that, include: The perception layer uses sensors to collect the environmental parameters of the industrial tea dryer for drying the tea in real time; the environmental parameters include drying temperature, drying humidity, tea moisture and aroma level; The transport layer uses wired and / or wireless methods to transmit data within the system; A control layer, comprising a drying prediction unit and an intelligent control unit; The moisture content prediction unit uses a neural network to train multiple daidai flower tea moisture content prediction models, and uses each daidai flower tea moisture content prediction model to predict the moisture content of the daidai flower tea; The intelligent control unit uses a dynamic Bayesian algorithm to determine the model weight corresponding to the prediction of each water content prediction model of the daidai flower tea, and reconstructs the predicted water content of the daidai flower tea according to the model weight to obtain the reconstructed water content of the daidai flower tea; The fragrance-locking degree of Daidaihua tea is predicted by logistic regression model according to the reconstructed predicted moisture content of Daidaihua tea. According to the predicted fragrance-locking degree and moisture content of the tea, the temperature and humidity in the drying process are controlled and adjusted using the fuzzy control method. The application layer provides real-time data display, historical data query, drying process monitoring and alarm prompt functions.

2. The drying monitoring system for Daidai flower tea based on artificial intelligence technology according to claim 1, wherein The moisture content prediction model of the daidai flower tea takes the drying temperature, drying humidity, drying wind speed, the initial moisture content of the daidai flower tea, and the prediction time as input, and the moisture content as output.

3. The drying monitoring system of bitter orange flower tea based on artificial intelligence technology according to claim 1, characterized in that, The water content of the reconstructed daidai flower tea is: θ t = a1f1(t) + a2f2(t) + … + a k f k (t); In the formula, θ t represents the reconstructed Dai Dai flower tea moisture content at time t, a1 represents the weight corresponding to the first Dai Dai flower tea moisture content prediction model, a2 represents the weight corresponding to the second Dai Dai flower tea moisture content prediction model, a k represents the weights corresponding to the k prediction models of the moisture content of daidai tea, f1(t) represents the moisture content of daidai tea at time t predicted by the first prediction model of the moisture content of daidai tea, f2(t) represents the moisture content of daidai tea at time t predicted by the second prediction model of the moisture content of daidai tea, f k (t) represents the moisture content of the tea at time t predicted by the kth tea moisture content prediction model.

4. The drying monitoring system for Daidai flower tea based on artificial intelligence technology according to claim 1, characterized in that, The degree of fragrance locking of the daidai flower tea is predicted by a logistic regression model; the logistic regression model: In the formula, β0 represents the intercept term, β1 represents the weight corresponding to the drying temperature feature, T represents the drying temperature, β2 represents the weight corresponding to the drying time feature, t represents the drying time, β3 represents the weight corresponding to the ambient humidity feature, H represents the ambient humidity, β4 represents the weight corresponding to the moisture content feature of the reconstructed Dai Dai flower tea at time t, and M t Represents the reconstructed moisture content of Daidai flower tea at time t, β5 represents the weight corresponding to the moisture content decrease feature, ΔM represents the moisture content decrease, and M0 represents the initial moisture content of Daidai flower tea.

5. The drying monitoring system of Citrus aurantium flower tea based on artificial intelligence technology according to claim 1, characterized in that, The steps of controlling the temperature and humidity in the drying process by the fuzzy control method include: Determine the drying stage of the Dai Dai flower tea, the drying stage of the Dai Dai flower tea is divided into three stages: preheating stage, dehydration stage and fragrance fixing stage; The fuzzy control method in the preheating stage controls the temperature and humidity in the drying process of the tea according to the preheating fuzzy rule base; the trapezoidal membership function for fuzzifying the degree of locking fragrance in the preheating stage is (0%, 0%, 20%, 40%), and the membership function of the moisture content is (40%, 50%, 70%, 80%); The fuzzy control method for the dehydration stage controls the temperature and humidity in the drying process of the daidai flower tea according to the dehydration fuzzy rule base; the trapezoidal membership function of the fragrance locking degree in the dehydration stage is (20%, 40%, 60%, 80%), and the membership function of the moisture content is (0%, 10%, 20%, 30%); The fuzzy control method in the fragrance fixing stage controls the temperature and humidity in the drying process of the daidai flower tea according to the fragrance fixing fuzzy rule base; the trapezoidal membership function of the fragrance locking degree in the fragrance fixing stage is (60%, 80%, 100%, 100%), and the membership function of the moisture content is (0%, 7%, 10%, 20%).

6. The drying monitoring system of Citrus aurantium var. amara flower tea based on artificial intelligence technology according to claim 1, characterized in that, The preheated fuzzy rule base: In the table, SL1 represents the aroma-locking degree of the bitter orange flower tea in the preheating stage, HS1 represents the moisture content of the bitter orange flower tea in the preheating stage, H h1 represents that the aroma-locking degree of the bitter orange flower tea in the preheating stage is divided into a high aroma-locking degree after fuzzification, M h1 represents that the aroma-locking degree of the bitter orange flower tea in the preheating stage is divided into a medium aroma-locking degree after fuzzification, L h1 represents that the aroma-locking degree of the bitter orange flower tea in the preheating stage is divided into a low aroma-locking degree after fuzzification, H s1 represents that the moisture content of the bitter orange flower tea in the preheating stage is divided into a high moisture content after fuzzification, M s1 represents that the moisture content of the bitter orange flower tea in the preheating stage is divided into a medium moisture content after fuzzification, L s1 represents that the moisture content of the bitter orange flower tea in the preheating stage is divided into a low moisture content after fuzzification, PB T1 represents that the control amount of the temperature in the preheating stage is positive large, PB H1 It indicates that the control amount of the humidity in the preheating stage is positive large, ZO represents the control amount of the temperature or humidity in the preheating stage, NB T1 represents that the control amount of the temperature in the preheating stage is negative large, NB H1 represents that the control amount of the humidity in the preheating stage is negative large.

7. The drying monitoring system for Daidai flower tea based on artificial intelligence technology according to claim 1, wherein Dehydration fuzzy rule base: In the table, SL2 represents the fragrance-locking degree of bitter orange flower tea in the dehydration stage, HS2 represents the moisture content of bitter orange flower tea in the dehydration stage, H h2 represents that the fragrance-locking degree of bitter orange flower tea in the dehydration stage is classified as high fragrance-locking degree after fuzzification, M h2 represents that the fragrance-locking degree of bitter orange flower tea in the dehydration stage is classified as medium fragrance-locking degree after fuzzification, L h2 represents that the fragrance-locking degree of bitter orange flower tea in the dehydration stage is classified as low fragrance-locking degree after fuzzification, H s2 represents that the moisture content of bitter orange flower tea in the dehydration stage is classified as high water content after fuzzification, M s2 represents that the moisture content of bitter orange flower tea in the dehydration stage is classified as medium water content after fuzzification, L s2 represents that the moisture content of bitter orange flower tea in the dehydration stage is classified as low water content after fuzzification, PB T2 represents that the control amount of the temperature in the dehydration stage is positive large, PB H2 It indicates that the control amount of the humidity in the dehydration stage is positive large, ZO represents the control amount of the temperature or humidity in the dehydration stage, NB T2 represents that the control amount of the temperature in the dehydration stage is negative large, NB H2 represents that the control amount of the humidity in the dehydration stage is negative large.

8. The drying monitoring system for Daidai flower tea based on artificial intelligence technology according to claim 1, characterized in that, Defining aroma fuzzy rule base: In the table, SL3 represents the fragrance-locking degree of the bitter orange flower tea in the fragrance-fixing stage, HS3 represents the moisture content of the bitter orange flower tea in the fragrance-fixing stage, H h3 represents that the fragrance-locking degree of the bitter orange flower tea in the fragrance-fixing stage is divided into a high fragrance-locking degree after fuzzification, M h3 represents that the fragrance-locking degree of the bitter orange flower tea in the fragrance-fixing stage is divided into a medium fragrance-locking degree after fuzzification, L h3 represents that the fragrance-locking degree of the bitter orange flower tea in the fragrance-fixing stage is divided into a low fragrance-locking degree after fuzzification, H s3 represents that the moisture content of the bitter orange flower tea in the fragrance-fixing stage is divided into a high moisture content after fuzzification, M s3 represents that the moisture content of the bitter orange flower tea in the fragrance-fixing stage is divided into a medium moisture content after fuzzification, L s3 represents that the moisture content of the bitter orange flower tea in the fragrance-fixing stage is divided into a low moisture content after fuzzification, PB T3 represents that the control amount of the temperature in the fragrance-fixing stage is positive large, PB H3 represents that the control amount of the humidity in the fragrance-fixing stage is positive large, ZO represents the control amount of the temperature or humidity in the fragrance-fixing stage, NB T3 represents that the control amount of the temperature in the fragrance-fixing stage is negative large, NB H3 represents that the control amount of the humidity in the fragrance-fixing stage is negative large.

9. The drying monitoring system of Citrus aurantium var. amara flower tea based on artificial intelligence technology according to claim 1, wherein The temperature adjustment amount in the preheating stage is [-5°C, +5°C], and the humidity adjustment amount ΔRH is [-10%, +10%]; The temperature adjustment amount ΔT in the dehydration stage is [-10°C, +10°C], and the humidity adjustment amount ΔRH is [-20%, 0%]; The temperature adjustment amount ΔT in the fragrance fixing stage is [-5°C, +5°C], and the humidity adjustment amount ΔRH is [-10%, 0%].

10. A storage medium for storing computer-executable instructions, characterized in that: The computer-executable instructions, when executed, implement an artificial-intelligence technology-based drying monitoring system for Daidai flower tea as described in any one of claims 1-9 above.

Images