Self-made electronic nose device special for detecting citrus huanglongbing and detection method

Abstract

The invention provides a homemade electronic nose device special for detecting the citrus huanglongbing and a detection method, and solves the technical problems that different infection stages of the citrus huanglongbing are difficult to identify accurately and special electronic nose detection equipment for the citrus huanglongbing is lacked. According to the technical scheme, the self-made electronic nose device special for detecting the citrus huanglongbing and the detection method are provided; the self-made electronic nose device comprises a gas collection chamber, a test module and a data acquisition module; the detection method comprises the following steps: step 1, preheating the sensor and cleaning the electronic nasal cavity; 2, collecting leaves of a target plant, cleaning and cutting into pieces; step 3, determining each experiment parameter; 4, setting the flow velocity of each device in the test gas path; 5, performing data analysis and feature extraction on the response curve of each sensor in the electronic nose; and 6, establishing a classification model. The design and construction of the special detection device for detecting the Huanglongbing and different infection stages are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of plant disease detection, and in particular to a self-made electronic nose device and a detection method for citrus Huanglongbing disease detection. Background Art

[0002] Citrus is one of the most important fruits in my country, with an annual output of 41.38 million tons, accounting for 20.4% of the country's fruit production. Citrus Huanglongbing is spread by citrus psyllids and grafting. It is highly contagious and there is no effective treatment. It is often prevented and controlled by detecting and removing diseased trees. It is called citrus cancer. About 100 million citrus plants are infected with Huanglongbing worldwide. Southern Jiangxi, Guangdong and surrounding areas have become the hardest hit areas of Huanglongbing, which has seriously restricted the healthy development of the citrus industry.

[0003] At present, the detection methods of Huanglongbing mainly include naked eye observation, microscopic observation, PCR detection, atlas technology, and detection methods based on volatile organic compounds. Among them, naked eye observation is highly subjective and has large errors; laboratory-based microscopic observation and PCR detection methods require sample pretreatment and require expensive equipment and professional operator support, with cumbersome steps and high costs, and are difficult to apply on a large scale in actual production; the detection method based on atlas technology analyzes the spectrum and image of the obtained citrus leaves or the whole plant, which is limited by the symptoms of Huanglongbing leaves and has poor recognition ability for early Huanglongbing samples; the detection method based on plant volatile organic compounds is a detection technology that has emerged in recent years. It uses the plant's response mechanism to stress and analyzes the volatile substances produced by plant stress to achieve the purpose of diagnosing Huanglongbing. However, due to the lack of dedicated gas-sensitive sensing equipment for Huanglongbing, the current Huanglongbing detection method based on volatile organic compounds produced by plant stress reactions relies on large-scale experimental instruments, such as GC-MS, GC-IMS, etc. and professional operators. The operation is cumbersome and expensive, and it is difficult to promote and popularize on a large scale. Therefore, there is an urgent need for alternative dedicated Huanglongbing detection equipment.

[0004] The prior art discloses a method for detecting citrus Huanglongbing based on an electronic nose, which uses a commercial electronic nose to identify Huanglongbing samples with typical symptoms. Since the commercial electronic nose is not specific to Huanglongbing, the recognition effect is poor for Huanglongbing samples that are not symptomatic in the early stage or confused with nutrient deficiency in the late stage.

[0005] How to solve the above technical problems is the subject faced by the present invention. Summary of the invention

[0006] The purpose of the present invention is to provide a homemade electronic nose device and a detection method for citrus Huanglongbing detection. In view of the technical problems existing in the prior art, such as complicated operation, high detection cost, poor recognition ability of Huanglongbing at different infection stages, and lack of detection devices dedicated to Huanglongbing, a homemade electronic nose device and a detection method for citrus Huanglongbing detection are proposed. The change reaction of volatile organic compounds in plants to disease stress is used as the detection basis, and the design and construction of a dedicated detection device for Huanglongbing detection and the accurate identification of samples at different infection stages, such as the early asymptomatic stage, the mid-term typical symptom stage and the late stage of mixed deficiency symptoms, are realized.

[0007] In order to achieve the above-mentioned purpose of the invention, the technical solution adopted by the present invention is specifically: a self-made electronic nose device specially used for citrus Huanglongbing detection, including: a gas collection chamber, a test module and a data acquisition module.

[0008] The gas collecting chamber includes an incubator, a temperature controller, and a gas collecting bottle, an air filter, a temperature control probe, and a heating plate located in the incubator. The gas collecting bottle is placed in the center of the incubator, the air filter is used to balance the air pressure in the gas collecting bottle, the temperature control probe is placed next to the gas collecting bottle, and is used to provide real-time feedback of the temperature in the incubator, and the heating plate is placed at the bottom of the incubator. The temperature controller is connected to the temperature control probe and the heating plate, and adjusts the heating plate in real time according to the temperature feedback from the temperature control probe to ensure that the temperature in the incubator is constant.

[0009] The test module includes a test chamber and a test gas circuit, wherein the test chamber provides a closed environment for the gas to be tested to fully contact the sensor, so as to ensure the consistency and repeatability of the test conditions. The test gas circuit is used to ensure that the gas in the test chamber is in a stable flow state during the entire detection process, and the gas to be tested is brought into the test chamber for testing through a carrier gas. The test gas circuit is composed of a micro vacuum pump, a precision gas flow stabilizing valve and a glass rotor flowmeter, and is connected through an odorless sealed air pipe to achieve stable control of the airflow to meet the experimental requirements.

[0010] The test chamber is composed of an air inlet chamber, a reaction chamber and an air outlet chamber, and the air inlet chamber, the reaction chamber and the air outlet chamber are all provided with fixed pin holes, and the three chambers are fixedly connected by fixed pins. A food-grade odorless silicone plate is clamped at the connection of the three chambers to ensure the overall sealing of the chamber. The air inlet chamber is provided with an air inlet hole, the air outlet chamber is provided with an air outlet hole, and the reaction chamber is provided with a fixed pin hole, an exhaust hole, a wire groove and a sensor fixing groove. The fixed pin hole is located on the outside of the entire chamber, the size of the sensor fixing groove matches the size of the sensor, and the sensor is sealed and fixed in the sensor fixing groove by a food-grade odorless sealant, ensuring that the gas to be tested can only enter the air outlet chamber through the exhaust hole. The circuit module matched with the sensor is arranged outside the test chamber, and the sensor connection line is reasonably led out through the wire groove to connect with the external sensor module, which saves space and avoids the influence of complex lines on the test process.

[0011] The data acquisition module includes a data acquisition card and data acquisition software. The data acquisition card model is NIUSB-6008. The data acquisition software is compiled using LabVIEW software. Its main functions include various parameter settings, sensor signal acquisition, real-time data display, data storage and export, etc.

[0012] Preferably, the volume of the gas collecting bottle is 200 mL.

[0013] Preferably, the gas collecting chamber is kept at a constant temperature of 40°C, and the temperature control range of the temperature controller is set to 40±1°C.

[0014] Preferably, the heating plate is a PTC heating plate.

[0015] Preferably, the air intake mode of the test air circuit is a dynamic air intake mode of one in and one out.

[0016] Preferably, the air inlet cavity is in a cone barrel shape, and the air outlet cavity is in a hemispherical shape.

[0017] Preferably, the number of sensors is set to 8, which are distributed in a circle facing the air inlet, so that each sensor can contact the gas almost simultaneously and evenly, avoiding inconsistent response times of the sensors and sensor reading deviations caused by local concentration differences.

[0018] Preferably, the sensor is a silicon dioxide gas sensor, and its model and corresponding target gas are TGS822 (organic solvents such as ethanol), TGS822TF (hydrogen, carbon monoxide), TGS826 (ammonia), TGS2600 (hydrogen, ethanol, methane, carbon monoxide), MQ3B (ethanol), MQ136 (hydrogen sulfide), MQ138 (aldehydes, alcohols, ketones, aromatic compounds), and WSP1110 (nitrogen dioxide).

[0019] Preferably, the exhaust hole in the reaction chamber is located directly behind the sensor fixing groove.

[0020] The present invention also provides a detection method of the self-made electronic nose device for citrus Huanglongbing detection, comprising:

[0021] Step 1: Start the device before use, preheat the sensor and clean the electronic nose chamber to achieve a stable working environment;

[0022] Step 2: Collect leaves from the target plant, wash them and cut them into pieces;

[0023] Step 3: Determine the experimental parameters, take a certain amount of leaves to be tested, put them into a gas collecting bottle and seal it, put the gas collecting bottle into the gas collecting chamber, and collect gas under a constant temperature environment;

[0024] Step 4: Set the flow rate of each device in the test gas path, inject the gas to be tested in the gas collecting bottle into the electronic nose test chamber, and obtain the sensor response signal;

[0025] Step 5: Perform data analysis and feature extraction on the response curves of each sensor in the electronic nose;

[0026] Step 6: Establish a classification model and use pattern recognition algorithms to identify samples at different infection stages of Huanglongbing.

[0027] Preferably, the step three is specifically as follows: according to the previous large-scale multi-group comparative experiments, the specific experimental parameters are determined as follows: gas collection temperature (40°C), gas collection time (10 min), sample weight (0.2 g), gas collection volume (200 mL), 0.2 g of the sample to be tested is taken, placed in a 200 mL gas collection bottle and sealed, the gas collection bottle is placed in the gas collection chamber, and the gas is collected for 10 min at a constant temperature of 40°C;

[0028] Preferably, step 4 is specifically as follows: set the air inlet flow rate of the electronic nose system to 300mL / min, the air outlet flow rate to 30mL / min, the sensor cleaning time to 180s, and the measurement time to 300s. After each test, pass the filtered air into the test chamber to initialize the electronic nose sensor array. The sampling interval is set to 1s, and the sensor data is collected and recorded through the designed data acquisition platform. Each sensor collects a total of 300 data points, with a total of 8 sensors, and finally one sample is represented by 2400 data points;

[0029] Preferably, the step five is specifically as follows: based on the results of multiple groups of comparative experiments in the early stage, using extreme value features to extract features of the electronic nose signal, wherein the extreme value features refer to the sensor response values ​​at the peaks or valleys of the signal curve, which reflect the composition and concentration of volatile organic compounds, and the response intensity of the sensor to specific volatile substances;

[0030] Preferably, the step six is ​​specifically as follows: according to the results of multiple groups of comparative experiments in the early stage, a classification model is established using linear discriminant analysis (LDA) to identify Huanglongbing samples.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1) The present invention proposes a manufacturing scheme for an electronic nose system specifically for Huanglongbing detection. The gas collecting chamber, test chamber and air path system of the electronic nose system are designed through airflow simulation, chamber structure design and three-dimensional modeling, thereby ensuring the sealing and stability of the detection environment and realizing the customized manufacturing of the electronic nose device.

[0033] 2) Based on the large-scale professional odor analysis instrument GC-MS / GC-IMS, the sensor selection of the electronic nose device dedicated to Huanglongbing detection was determined, the specificity of the gas sensor for the odor of Huanglongbing was enhanced, and through the combination of optimal experimental parameters, feature extraction and pattern recognition algorithms, effective identification of samples at different infection stages of early, middle and late Huanglongbing was achieved, achieving significant beneficial effects of the device and method involved in the present invention on Huanglongbing detection, and having practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0035] Figure 1 A schematic diagram of the structure of a self-made electronic nose device for citrus Huanglongbing detection according to an embodiment of the present invention;

[0036] Figure 2A schematic diagram of the test chamber structure of a self-made electronic nose device for citrus Huanglongbing detection according to an embodiment of the present invention;

[0037] Figure 3 It is a schematic diagram of the LDA classification results of Huanglongbing positive and Huanglongbing positive samples using a homemade electronic nose according to an embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram of GC-MS analysis results of volatile organic compounds in leaf samples of different citrus varieties according to an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of GC-IMS analysis results of volatile organic compounds in leaf samples of different citrus varieties according to an embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of the LDA classification results of detecting healthy and symptomatic Huanglongbing samples using a commercial PEN3 electronic nose according to an embodiment of the present invention;

[0041] Figure 7 This is a schematic diagram of the LDA classification results of using the commercial PEN3 electronic nose to detect healthy, zinc-deficient, symptomatic Huanglongbing and zinc-deficient Huanglongbing samples according to an embodiment of the present invention;

[0042] Figure 8 This is a schematic diagram of the contribution index weights of various sensors of the commercial PEN3 electronic nose for Huanglongbing detection according to an embodiment of the present invention;

[0043] Fig. 9 The following are top views of the airflow simulation diagram of the reaction chamber of a self-made electronic nose device for citrus Huanglongbing detection according to an embodiment of the present invention: (a) is a schematic diagram showing that the exhaust hole is located at the rear side of the sensor; (b) is a schematic diagram showing that the exhaust hole is located directly behind the sensor;

[0044] Fig.10 1. Airflow simulation comparison diagram of a test cavity of a self-made electronic nose device for citrus Huanglongbing detection according to an embodiment of the present invention: (a) is a schematic diagram of a conical air inlet cavity; (b) is a schematic diagram of a hemispherical air inlet cavity.

[0045] Fig.11 This is a product picture of a homemade electronic nose device specifically for citrus Huanglongbing detection according to an embodiment of the present invention.

[0046] Wherein, the accompanying drawings are marked as follows:

[0047] 1. Insulation box; 2. Temperature controller; 3. Gas collecting bottle; 4. Air filter; 5. Temperature control probe; 6. Heating plate; 7. Miniature vacuum pump; 8. Precision gas steady flow valve; 9. Glass rotor flowmeter; 10. Air inlet chamber; 11. Reaction chamber; 12. Air outlet chamber; 13. Pin hole; 14. Air inlet hole; 15. Air outlet hole; 16. Exhaust hole; 17. Wire groove; 18. Sensor fixing groove; 19. Sensor; 20. Data acquisition card. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0049] Example 1

[0050] See also Figure 1 According to the self-made electronic nose device for citrus Huanglongbing detection provided in this embodiment, it includes a gas collection chamber, a test module and a data acquisition module.

[0051] Specifically, Figure 1 As shown, the gas collecting chamber includes an incubator 1, a temperature controller 2, a gas collecting bottle 3, an air filter 4, a temperature control probe 5 and a heating plate 6 located in the incubator 1. The gas collecting bottle 3 has a volume of 200 mL and is placed in the center of the incubator 1. The air filter 4 is used to balance the air pressure in the gas collecting bottle 3. The temperature control probe 5 is placed next to the gas collecting bottle 3 for real-time feedback of the temperature in the incubator 1. The heating plate 6 is placed at the bottom of the incubator 1 and a PTC heating plate is used for heating. The temperature controller 2 is connected to the temperature control probe 5 and the heating plate 6. The heating plate 6 is adjusted in real time according to the temperature feedback from the temperature control probe 5. The temperature control range is 40±1°C, which can ensure that the temperature in the incubator 1 is constant.

[0052] The test module includes a test chamber and a test gas circuit, wherein the test chamber provides a closed environment for the gas to be tested to fully contact the sensor to ensure the consistency and repeatability of the test conditions. The test gas circuit is used to ensure that the gas in the test chamber is in a stable flow state during the entire detection process, and the gas to be tested is brought into the test chamber for testing through the carrier gas. The test gas circuit is composed of a micro vacuum pump 7, a precision gas flow regulating valve 8 and a glass rotor flowmeter 9, which are connected through an odorless sealed air pipe to achieve stable control of the airflow to meet the experimental requirements.

[0053] like Figure 2As shown, the test chamber consists of an air inlet chamber 10, a reaction chamber 11 and an air outlet chamber 12. In order to increase the stability of the gas flow in the chamber, the test chamber adopts a dynamic air inlet method of one in and one out. The air inlet chamber 10, the reaction chamber 11 and the air outlet chamber 12 are all provided with a fixed pin hole 13, and the three chambers are fixedly connected by the fixed pin 13. A food-grade odorless silicone plate is clamped at the connection of the three chambers to ensure the overall sealing of the chamber. The air inlet chamber 10 is provided with an air inlet hole 14, and the air outlet chamber 12 is provided with an air outlet hole 15. As shown Figure 3 As shown, the reaction chamber 11 is provided with a fixing pin hole 13, an exhaust hole 16, a wire groove 17 and a sensor fixing groove 18. The fixing pin hole 13 is located outside the entire chamber, and the exhaust hole 16 is located directly behind the sensor fixing groove 18. The size of the sensor fixing groove 18 matches the size of the sensor 19. The number of sensors 19 is set to 8, which are distributed in a circle facing the air inlet 14, so that each sensor can contact the gas almost simultaneously and evenly, avoiding inconsistent sensor response time and sensor reading deviation caused by local concentration differences. The sensor 19 is sealed and fixed in the sensor fixing groove 18 by food-grade odorless sealant, so that the reaction surface of the sensor 19 is directly exposed to the air inlet chamber 10, ensuring that the gas to be measured can only enter the air outlet chamber 12 through the exhaust hole 16. Sensor 19 uses a silicon dioxide gas sensor, and the target gases corresponding to the eight sensors are TGS822 (ethanol and other organic solvents), TGS822TF (hydrogen, carbon monoxide), TGS826 (ammonia), TGS2600 (hydrogen, ethanol, methane, carbon monoxide), MQ3B (ethanol), MQ136 (hydrogen sulfide), MQ138 (aldehydes, alcohols, ketones, aromatic compounds), and WSP1110 (nitrogen dioxide). The circuit module matched with sensor 19 is set outside the test cavity, and the connection line of sensor 19 is reasonably led out through the wire groove 17 and connected to the external sensor module, which saves space and avoids the influence of complex lines on the test process.

[0054] Example 2

[0055] On the basis of Example 1, the following describes a detection method for a homemade electronic nose device specifically for citrus Huanglongbing detection provided in this example, and evaluates its ability to distinguish six types of samples: healthy, magnesium-deficient, zinc-deficient, early-stage non-symptomatic Huanglongbing, mid-stage symptomatic Huanglongbing, and late-stage zinc-deficient Huanglongbing.

[0056] The number and types of leaf samples in this embodiment are as follows: the citrus variety to which the leaf samples belong is ponkan, including 71 healthy ones, 66 magnesium-deficient ones, 61 zinc-deficient ones, 52 early-stage non-symptomatic Huanglongbing ones, 77 mid-stage symptomatic Huanglongbing ones, and 61 late-stage zinc-deficient Huanglongbing ones. The samples were collected from an orchard base in Huangyan City, Zhejiang Province under the guidance of experts in November 2022, and the Huanglongbing infection status was determined by qPCR testing.

[0057] Step 1: Start the device before use, preheat the sensor and clean the electronic nose chamber to achieve a stable working environment;

[0058] Step 2: Collect leaves from the target plant, wash them and cut them into pieces;

[0059] Step 3: Determine the experimental parameters, take a certain amount of leaves to be tested, put them into a gas collecting bottle and seal it, put the gas collecting bottle into the gas collecting chamber, and collect gas under a constant temperature environment;

[0060] Specifically, based on the previous large-scale multi-group comparative experiments, the specific experimental parameters were determined as follows: gas collection temperature (40°C), gas collection time (10 min), sample weight (0.2 g), and gas collection volume (200 mL). 0.2 g of the sample to be tested was placed in a 200 mL gas collection bottle and sealed. The gas collection bottle was placed in the gas collection chamber and the gas was collected for 10 min at a constant temperature of 40°C.

[0061] Step 4: Set the flow rate of each device in the test gas path, inject the gas to be tested in the gas collecting bottle into the electronic nose test chamber, and obtain the sensor response signal;

[0062] Specifically, the air inlet flow rate of the electronic nose system is set to 300mL / min, the air outlet flow rate is set to 30mL / min, the sensor cleaning time is set to 180s, and the measurement time is set to 300s. After each test, the filtered air is passed into the test chamber to initialize the electronic nose sensor array. The sampling interval is set to 1s, and the sensor data is collected and recorded through the designed data acquisition platform. Each sensor collects a total of 300 data points, with a total of 8 sensors, and finally one sample is represented by 2400 data points;

[0063] Step 5: Perform data analysis and feature extraction on the response curves of each sensor in the electronic nose;

[0064] Specifically, based on the results of multiple groups of comparative experiments in the early stage, extreme value features are used to extract features of electronic nose signals. The extreme value features refer to the sensor response values ​​at the peaks or troughs of the signal curve, which reflect the composition and concentration of volatile organic compounds, as well as the response intensity of the sensor to specific volatile substances.

[0065] Step 6: Establish a classification model and use pattern recognition algorithms to identify samples at different infection stages of Huanglongbing.

[0066] Specifically, based on the results of multiple groups of comparative experiments in the early stage, a classification model was established using linear discriminant analysis (LDA) to identify Huanglongbing samples.

[0067] In order to test the sensitivity of the homemade electronic nose to the specific odor of Huanglongbing, the samples tested included Huanglongbing leaves at different infection stages and some nutrient-deficient leaves, a total of six categories: healthy, magnesium-deficient, zinc-deficient Huanglongbing, mild Huanglongbing and obvious Huanglongbing. LDA was used for pattern recognition, and its confusion matrix is ​​shown in Table 1. The overall accuracy of the model was 83.33%, among which the recognition rates of zinc-deficient, zinc-deficient Huanglongbing, mild Huanglongbing and obvious Huanglongbing were relatively high, all reaching more than 85%.

[0068] Table 1 Confusion matrix of six types of leaves identified using homemade electronic nose

[0069]

[0070] The main purpose of this example is to identify Huanglongbing infection at different symptomatic stages. Therefore, all types of samples are divided into two categories, and the classification results are statistically analyzed according to Huanglongbing positive samples and Huanglongbing negative samples. The confusion matrix is ​​shown in Table 2. The classification accuracy of the model is 93.59%. Samples at three different infection stages: mild Huanglongbing, obvious Huanglongbing, and zinc deficiency Huanglongbing can all be effectively identified.

[0071] Table 2 Classification results for Huanglongbing infected samples

[0072]

[0073] LDA was used to visualize the sample distribution of diseased and uninfected leaves. The results are as follows: Figure 3 As shown. There is a clear classification boundary between the samples stressed by Huanglongbing (mild Huanglongbing, obvious Huanglongbing and zinc-deficient Huanglongbing) and the samples not stressed by Huanglongbing (healthy, magnesium-deficient and zinc-deficient). Since plants release unique volatile organic compounds when they are stressed by diseases, they show obvious differences in the electronic nose sensor signals. The results of this embodiment prove that the homemade electronic nose device and detection method for citrus Huanglongbing detection proposed in this embodiment can effectively enhance the ability to obtain Huanglongbing characteristics, and have a significant beneficial effect on the identification of Huanglongbing stress at different infection stages, with an accuracy rate of 93.59%.

[0074] experiment:

[0075] The main purpose of this experimental case is to determine the selection of Huanglongbing-specific odor sensor for the homemade electronic nose based on the weight analysis results of specific volatiles caused by Huanglongbing and commercial electronic nose sensors.

[0076] This experimental case achieves the above objectives through the following three steps: Step 1) Use GC-MS and GC-IMS to detect volatile organic compounds in samples infected with Huanglongbing, and screen out specific volatile components caused by Huanglongbing through analysis; Step 2) Use principal component analysis to perform weight analysis on the contribution of each sensor in the commercial electronic nose; Step 3) Determine the sensor selection for the homemade electronic nose.

[0077] Step 1): Use GC-MS and GC-IMS to detect volatile organic compounds in samples infected with Huanglongbing, and screen out specific volatile components caused by Huanglongbing through analysis:

[0078] The number and types of leaf samples in this step are as follows: A total of 24 citrus trees of four citrus varieties (Ponkan, Local Early, Murcott and Newhall) were tested for health status, and a total of 12 healthy trees (3 trees of each variety, as 3 parallels) and 12 symptomatic Huanglongbing trees were identified. Three leaves with typical symptoms collected from the same branch of the same tree were used as a sample unit for subsequent testing. The samples were collected from the orchard base in Huangyan City, Zhejiang Province under the guidance of experts in November 2021 and the infection status of Huanglongbing was determined by qPCR testing.

[0079] The volatile organic compound content of citrus leaves was detected by GC-MS. Each sample was tested three times in parallel, and the average value was taken as the final result for analysis. Before the test, solid phase headspace microextraction technology was used to capture the volatile gas of citrus leaves. A sample unit of 0.7g of mesophyll tissue was placed in a 20mL sample bottle, and trans, trans-2,4-nonadienal was used as the internal standard substance. 200μL of internal standard solution (concentration of 80.69μg / mL of trans, trans-2,4-nonadienal / n-hexane) was added to each sample bottle. Solid phase headspace microextraction was completed by AOC-2000 multifunctional automatic sampler (AOC-2000, Shimadzu, Japan). After the extraction was completed, the extraction head was returned to the sampler and pulled out from the sample bottle to complete the capture of volatile gas components in the headspace of citrus leaves.

[0080] The GC-IMS detection system consists of an automatic headspace sampler (CTC-PAL, CTC Analytics AG, Switzerland), a capillary gas chromatography column (WAX, 15m*0.53mm*1μm), a gas chromatography ion mobility spectrometer, and GC-IMS LibrarySearch (V1.0.3) analysis software. A sample unit of 1g fresh citrus leaves was taken, cut into small pieces, placed in a 20mL headspace sampling glass bottle, incubated at 40℃ for 20min, and 0.2mL of the sample headspace was injected into the gas chromatography column in splitless mode using an injection needle heated to 85℃. Then, nitrogen was used as a carrier to pre-separate the measured gas and then ionized in the ionization chamber. The generated ionized ions were sent to an IMS drift tube with a length of 10cm, and the working temperature of the drift tube was 45℃. The drift gas was nitrogen, and the flow rate was set to 150mL / min. Three parallel tests were performed for each sample, using C4-C9 ketones as external references to facilitate the calculation of the retention indices of the detected volatile organic compounds under the same chromatographic conditions. Volatile compounds were identified by comparing the retention indices with the drift times of standard compounds in the GC-IMS database. The relative content of the identified volatile compounds was expressed as the average value of the peak area.

[0081] The results showed that 143 volatile organic compounds were detected by GC-MS analysis of healthy and symptomatic leaves of four citrus varieties, which were divided into six categories: alcohols, aldehydes, alkenes, aromatics, esters and other compounds; 99 volatile compounds were detected by GC-IMS analysis of healthy and symptomatic leaves of four citrus varieties, which were divided into eight categories: aromatics, heterocyclics, esters, aldehydes, ketones, alcohols and others, among which the substances with the largest content were alkenes, alcohols and aldehydes. Figure 4 and Figure 5 As shown in the figure, based on the analysis results of GC-MS and GC-IMS, the volatiles of citrus leaves mainly include aromatic compounds, aldehydes, alcohols and olefins. After being infected with Huanglongbing, the composition and concentration of their volatile components changed, among which aromatic compounds, aldehydes and olefins increased slightly while alcohols decreased significantly. The changes in these substances can be used as a basis for detecting Huanglongbing infection.

[0082] Step 2): The contribution of each sensor in the commercial electronic nose is analyzed by principal component analysis:

[0083] The number and types of leaf samples in this step are as follows: the citrus variety to which the leaf samples belonged was Ponkan, including 114 healthy ones, 40 zinc-deficient ones, 124 mid-stage symptomatic Huanglongbing ones, and 16 late-stage zinc-deficient Huanglongbing ones. The samples were collected from an orchard base in Huangyan City, Zhejiang Province under the guidance of experts in November 2021, and the Huanglongbing infection status was determined by qPCR testing.

[0084] The PEN3 electronic nose (PEN3, Airsense, German) was used to collect volatiles from citrus leaves. The PEN3 electronic nose consists of a sampling system, a sensor array, and data acquisition software. Before data acquisition, the beaker containing the leaves was placed in a constant temperature oven to enrich the volatile gases. During the collection, the probe and air filter of the electronic nose sampling system were inserted into the beaker at the same time. The gas in the beaker was pumped into the electronic nose chamber by a micro air pump to react with the sensor array, and then the sensor signal was collected and saved by the data acquisition software. The characteristics of the MOS sensor array are shown in Table 3.

[0085] Table 3 PEN3 electronic nose sensor characteristics

[0086]

[0087]

[0088] Based on the previous comparative experiments, the experimental parameters were determined as follows: gas collection temperature (40°C), gas collection time (20 min), sample weight (0.2 g), feature extraction method as extreme value feature, and pattern recognition algorithm as linear discriminant analysis (LDA). LDA analysis was performed on healthy and symptomatic Huanglongbing samples. The results are shown in the figure. Figure 6 As shown in the figure, there are obvious clusters between healthy samples and symptomatic Huanglongbing samples in the LDA two-dimensional graph, and the confidence ellipses of the two types of samples do not overlap, which shows an excellent classification effect. This confirms the feasibility of detecting Huanglongbing based on gas sensitive sensor technology and lays the foundation for subsequent homemade electronic nose equipment.

[0089] In order to test the ability of commercial electronic noses to identify different types of Huanglongbing, two types of samples, zinc-deficient and zinc-deficient Huanglongbing, were added to the test samples. The samples were divided into two categories according to the infection status of Huanglongbing, where negative samples included healthy and zinc-deficient samples, and positive samples included symptomatic Huanglongbing and zinc-deficient Huanglongbing. When zinc-deficient Huanglongbing samples were added, the ability of PEN3 electronic nose to identify positive samples of Huanglongbing decreased. The classification results of LDA are shown in Figure 2. Figure 7 As shown in the figure, there is a large overlap in the data points of the positive and negative samples of Huanglongbing. Since the commercial electronic nose needs to consider a wide range of applicability in order to adapt to different application scenarios, the sensors configured for Huanglongbing detection are poorly targeted to the specific odor of Huanglongbing, which can easily lead to misjudgment. In order to improve the sensitivity of the Huanglongbing odor response, it is necessary to design a dedicated sensor array and a supporting gas path system for the sensitive odor of Huanglongbing, and create a homemade electronic nose device dedicated to citrus Huanglongbing detection.

[0090] Considering the excellent classification ability of PEN3 electronic nose for healthy and symptomatic Huanglongbing samples, the index weights of each sensor were calculated based on principal component analysis to determine the contribution of each sensor to Huanglongbing identification. The weights of the 10 sensors in PEN3 electronic nose for Huanglongbing detection are as follows: Figure 8 As shown, the sensors with weights exceeding 10% are MOS2, MOS4, MOS7 and MOS9. As shown in Table 3, their corresponding target gases are nitrogen oxides, hydrogen, sulfides and aromatic compounds. Therefore, this gas type sensor should be preferred when customizing a homemade electronic nose device for citrus Huanglongbing detection.

[0091] Step 3): Determine the sensor type for the homemade electronic nose device for citrus Huanglongbing detection:

[0092] The homemade electronic nose device designed in this embodiment for the detection of citrus Huanglongbing uses a SnO2 gas sensor. According to the analysis results of step 1) GC-MS and GC-IMS, the volatile substances that change greatly in the leaves of citrus after infection with Huanglongbing mainly include aromatic compounds, aldehydes, alcohols and olefins. The changes in these substances can be used as a basis for the detection of Huanglongbing infection. At the same time, considering that the target gases of the sensors with high contribution in the commercial electronic nose in step 2) include nitrogen oxides, hydrogen, sulfides and aromatic compounds, the 8 sensor models used in the homemade electronic nose device for the detection of citrus Huanglongbing in this embodiment are TGS822, TGS822TF, TGS826, TGS2600, MQ3B, MQ136, MQ138, and WSP1110, respectively, and their corresponding target gases are shown in Table 4.

[0093] Table 4 Sensor models and characteristics of the homemade electronic nose device for citrus Huanglongbing detection

[0094]

[0095] experiment:

[0096] The main purpose of this experimental case is to determine the optimal air chamber structure and hole layout of the homemade electronic nose through airflow simulation, so as to achieve the optimal electronic nose chamber structure design.

[0097] In the reaction chamber, eight sensors are arranged in a circular pattern facing the air inlet, so that each sensor can contact the gas almost simultaneously and evenly, avoiding inconsistent sensor response times and sensor reading deviations caused by local concentration differences. The effects of different exhaust hole positions on the sensors are compared, and the airflow in the chamber is simulated. The results are shown in Figure 2. Fig. 9 When the exhaust hole is located directly behind, the airflow to be measured can be evenly distributed in every corner of the sensor surface, maximizing the probability of the gas to be measured being captured by the sensor.

[0098] The schematic diagram comparing the gas flow effects of the hemispherical chamber structure and the conical chamber structure of the air inlet cavity is shown in Fig.10 As shown in the figure, compared with the hemispherical structure, the conical structure design can effectively reduce the turbulence of the gas in the chamber, so that the gas passes through the sensor array more evenly and stably, improving the efficiency and stability of the contact between the gas and the sensor. In addition, the conical air chamber is also conducive to maximizing the use of the airflow dynamics in a limited space, ensuring that each sensor can be exposed to a sufficient concentration of the gas to be tested, thereby improving the overall detection sensitivity and accuracy of the electronic nose.

[0099] like Fig.10 As shown in the figure, the gas outlet cavity is designed to be hemispherical, which can effectively reduce the cleaning dead corners that may appear during the gas flow process and is easy to maintain. At the same time, the hemispherical design can also reduce the generation of turbulence and improve the stability of gas flow. In short, the improved test cavity has well solved the problems that are prone to occur in the traditional test cavity, such as inconsistent sensor response time, concentration gradient differences between sensors, and difficulty in cleaning. It also has a clever appearance design and is easy to install, meeting the actual use needs.

[0100] 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 principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-made electronic nose device for citrus Huanglongbing detection, characterized in that: The self-made electronic nose device includes a gas collection chamber, a test module and a data acquisition module; The gas collecting chamber comprises an insulated box (1), a temperature controller (2), a gas collecting bottle (3) located in the insulated box (1), an air filter (4), a temperature control probe (5) and a heating plate (6); The gas collecting bottle (3) is placed in the center of the heat preservation box (1), the air filter (4) is used to balance the air pressure in the gas collecting bottle (3), the temperature control probe (5) is placed next to the gas collecting bottle (3) and is used to provide real-time feedback of the temperature in the heat preservation box (1), the heating plate (6) is placed at the bottom of the heat preservation box (1), the temperature controller (2) is connected to the temperature control probe (5) and the heating plate (6) respectively, and the heating plate (6) is adjusted in real time according to the temperature feedback from the temperature control probe (5) to ensure that the temperature in the heat preservation box (1) is constant; The test module comprises a micro vacuum pump (7), a precision gas flow stabilizing valve (8), a glass rotor flowmeter (9), an air inlet chamber (10), a reaction chamber (11) and an air outlet chamber (12); The micro vacuum pump (7) is connected to the gas collecting bottle (3) through a silicone gas tube, the micro vacuum pump (7), the precision gas flow stabilizing valve (8), the glass rotor flow meter (9) and the air inlet chamber (10) are connected through a silicone tube, and the air inlet chamber (10), the reaction chamber (11) and the air outlet chamber (12) are fixed through a pin hole (13); The data acquisition module comprises a data acquisition card (20) and LabVIEW data acquisition software; the data acquisition card (20) is connected to a sensor (19) in a test cavity (11) via a signal line; and the data acquisition card (20) transmits the collected sensor signal to the LabVIEW data acquisition software via a USB line.

2. The self-made electronic nose device for citrus Huanglongbing detection according to claim 1, characterized in that: The test gas circuit is used to ensure that the gas in the test chamber is in a stable flow state during the entire detection process, and the gas to be tested is brought into the test chamber for testing through the carrier gas; The test gas route is composed of a micro vacuum pump (7), a precision gas flow stabilizing valve (8) and a glass rotor flow meter (9) connected through an odorless sealed air pipe, so as to achieve stable control of the airflow; The test chamber is composed of an air inlet chamber (10), a reaction chamber (11) and an air outlet chamber (12); the air inlet chamber (10), the reaction chamber (11) and the air outlet chamber (12) are all provided with a plurality of fixing pin holes (13), and the air inlet chamber (10), the reaction chamber (11) and the air outlet chamber (12) are fixedly connected by the fixing pins (13); A food-grade odorless silica gel plate is sandwiched between the connection points of the air inlet chamber (10), the reaction chamber (11) and the air outlet chamber (12) to ensure the overall sealing of the chamber; The air inlet cavity (10) is provided with an air inlet hole (14) at one end, and an air outlet cavity (12) is provided at the other end, and the air outlet cavity (12) is provided with an air outlet hole (15); The reaction chamber (11) is also provided with a plurality of exhaust holes (16), wire grooves (17) and sensor fixing grooves (18) at equal intervals, and a plurality of fixing pin holes (13) are circumferentially arranged on the outer edge of the entire chamber; The exhaust hole (16) is located on the circumference between the sensor fixing groove (18) and the fixing pin hole (13), and the size of the sensor fixing groove (18) matches the size of the sensor (19); The number of the sensors (19) is set to 8, and they are distributed in a circular pattern facing the air inlet (14), so that each sensor (19) contacts the gas simultaneously and evenly; The sensor (19) is sealed and fixed in the sensor fixing groove (18) by food-grade odorless sealant, so that the reaction surface of the sensor (19) is placed in the air inlet cavity (10), and the gas to be measured enters the air outlet cavity (12) through the exhaust hole (16); The sensor (19) is a silicon dioxide gas sensor; a circuit module matched with the sensor (19) is arranged outside the test cavity, and a connection line of the sensor (19) is led out through a wire groove (17) and connected to an external sensor module.

3. The self-made electronic nose device for citrus Huanglongbing detection according to claim 2, characterized in that: The data acquisition module includes a data acquisition card and data acquisition software. The data acquisition card model is NI USB-6008. The data acquisition software is compiled using LabVIEW software, including various parameter settings, sensor signal acquisition, real-time data display, data storage and export.

4. The self-made electronic nose device for citrus Huanglongbing detection according to claim 3, characterized in that: The volume of the gas collecting bottle (3) is 200 mL; the temperature of the insulated box (1) is maintained at a constant temperature of 40° C., and the temperature control range of the temperature controller (2) is set to 40±1° C.

5. The self-made electronic nose device for citrus Huanglongbing detection according to claim 4, characterized in that: Preferably, the heating plate (6) is a PTC heating plate.

6. The self-made electronic nose device for citrus Huanglongbing detection according to claim 3, characterized in that: The air intake mode of the test air circuit is: a dynamic air intake mode of one in and one out; The air inlet cavity (10) is in a cone barrel shape, and the air outlet cavity (12) is in a hemispherical shape.

7. The self-made electronic nose device for detecting citrus Huanglongbing according to claim 3, characterized in that: The number of the sensors (19) is set to 8, which are distributed in a circumference facing the air inlet (14), so that each sensor (19) contacts the gas simultaneously and evenly.

8. The self-made electronic nose device for citrus Huanglongbing detection according to claim 3, characterized in that: The sensor (19) is a silicon dioxide type gas sensor, and its model and corresponding target gas are TGS822 ethanol organic solvent, TGS822TF hydrogen or carbon monoxide, TGS826 ammonia, TGS2600 hydrogen or ethanol or methane or carbon monoxide, MQ3B ethanol, MQ136 hydrogen sulfide, MQ138 aldehydes or alcohols or ketones or aromatic compounds, and WSP1110 nitrogen dioxide.

9. A detection method based on the self-made electronic nose device for citrus Huanglongbing detection according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Start the device before use, preheat the sensor (19) and clean the electronic nose chamber to achieve a stable working environment; Step 2: Collect leaves from the target plant, wash them and cut them into pieces; Step 3: Determine each experimental parameter, take a certain amount of leaves to be tested, put them into a gas collecting bottle (3) and seal them, put the gas collecting bottle (3) into an insulated box (1), and collect gas in a constant temperature environment; Step 4: setting the flow rate of each device in the test gas path, injecting the gas to be tested in the gas collecting bottle (3) into the electronic nose test chamber, and obtaining the response signal of the sensor (19); The step 4 is: setting the air inlet flow rate of the electronic nose system to 300 mL / min, the air outlet flow rate to 30 mL / min, the sensor (19) cleaning time to 180 s, the measurement time to 300 s, after each test is completed, passing the filtered air into the test chamber to initialize the electronic nose sensor array, the sampling interval is set to 1 s, and the sensor data is collected and recorded through the designed data acquisition platform, each sensor (19) collects a total of 300 data points, a total of 8 sensors (19), and finally one sample is represented by 2400 data points; Step 5: performing data analysis and feature extraction on the response curves of each sensor (19) in the electronic nose; The step 5 specifically comprises: extracting features of the electronic nose signal using extreme value features according to the results of multiple groups of comparative experiments in the previous stage, wherein the extreme value features refer to the sensor response values ​​at the peaks or valleys of the signal curve, reflecting the composition and concentration of volatile organic compounds, as well as the response intensity of the sensor (19) to specific volatile substances; Step 6: Establish a classification model and use pattern recognition algorithms to identify samples at different infection stages of Huanglongbing; The step six is ​​specifically as follows: based on the results of multiple groups of comparative experiments in the early stage, a classification model is established using linear discriminant analysis (LDA) to identify Huanglongbing samples.

10. The detection method of the self-made electronic nose device for citrus Huanglongbing detection according to claim 9, characterized in that: In the step three, based on the previous large-scale multi-group comparative experiments, the specific experimental parameters are determined as follows: gas collection temperature 40°C, gas collection time 10 min, sample weight 0.2 g, gas collection volume 200 mL, 0.2 g of the sample to be tested is taken and placed in a 200 mL gas collection bottle (3) and sealed, and the gas collection bottle (3) is placed in an incubator (1) to collect gas for 10 min at a constant temperature of 40°C.

Images