阅读说明：本技术 一种茶油品质快速检测装置 (Tea-seed oil quality short-term test device ) 是由 孙钟雷 李宇 谢仁香 赵久月 张德 于 2019-12-02 设计创作，主要内容包括：本发明公开了一种茶油品质快速检测装置,包括加热器、反应罐和检测箱；其特征在于,反应罐内设有温度传感器,检测箱内安装有烟雾传感器,检测箱的内侧壁上安装有多个气体传感器,检测箱上还安装有数据处理器,数据处理器包括信号放大器、单片机和AD转换电路。通过加热器对待测茶油进行加热,利用温度传感器、烟雾传感器和气体传感器对加热后待测茶油挥发的气体进行检测,从而得出待测茶油的酸价值、烟点和气味品质。本发明整个装置轻小便携、功能齐全,在不损坏茶油原有理化性质的情况下,短时间内采集分析多个检测指标,从而对茶油的综合品质进行评价。(The invention discloses a device for quickly detecting the quality of tea oil, which comprises a heater, a reaction tank and a detection box, wherein the heater is arranged on the reaction tank; the reaction tank is internally provided with a temperature sensor, the detection box is internally provided with a smoke sensor, the inner side wall of the detection box is provided with a plurality of gas sensors, the detection box is also provided with a data processor, and the data processor comprises a signal amplifier, a single chip microcomputer and an AD conversion circuit. The tea oil to be detected is heated by the heater, and the volatile gas of the heated tea oil to be detected is detected by the temperature sensor, the smoke sensor and the gas sensor, so that the acid value, the smoke point and the smell quality of the tea oil to be detected are obtained. The whole device is light, small and portable, has complete functions, and collects and analyzes a plurality of detection indexes in a short time under the condition of not damaging the original physicochemical properties of the tea oil, thereby evaluating the comprehensive quality of the tea oil.)

1. A rapid detection device for tea oil quality comprises a heater (1), a reaction tank (2) for containing tea oil to be detected and a detection box (3), wherein the heater (1) is installed at the bottom of the reaction tank (2) and can heat the reaction tank, the bottom of the detection box (3) is provided with an opening, an opening is formed in the reaction tank (2) corresponding to the position of the opening, and the detection box (3) is communicated with the reaction tank through the opening and the opening;

the device is characterized in that a temperature sensor (4) is also arranged in the reaction tank (2), and the temperature sensor (4) is used for measuring the temperature of the tea oil to be measured in the reaction tank (2);

a smoke sensor (5) is installed at the position of the opening of the detection box (3), and the smoke sensor (5) is used for detecting the smoke concentration generated after the tea oil to be detected in the detection box (3) is heated and detecting the tea oil smoke point; a plurality of gas sensors (6) are mounted on the inner side wall of the detection box (3) and are used for detecting free fatty acids and chemical volatile substances in the tea oil to be detected and inducing the free fatty acids and the chemical volatile substances to generate corresponding electric signals, wherein the gas sensors are lipid gas sensors and are used for detecting the acid value of the tea oil to be detected, and organic volatile gas sensors are used for detecting the odor grade of the tea oil to be detected;

the detection box (3) is also provided with a data processor, the data processor comprises a signal amplifier, a single chip microcomputer and an AD conversion circuit, the signal input end of the signal amplifier is connected with the signal output end of each sensor, the signal amplifier transmits the collected electric signals to the AD conversion circuit, and then the electric signals are transmitted to the single chip microcomputer by the AD conversion circuit for data processing;

the single chip microcomputer processes the electric signals after receiving the electric signals converted after the acquisition of each gas sensor, and the specific processing process is as follows: carrying out normalization processing and signal graphic analysis on the electric signal, preprocessing the electric signal by adopting a difference method and a difference quotient method, extracting the peak value, the peak time, the average value and the peak-out time of a signal curve as characteristic values, and resolving the characteristic values through a resolving model established by a RBF neural network to obtain an acid value and a smell grade;

heating the tea oil to be detected until the tea oil is fuming, feeding back the tea oil to the temperature sensor by the smoke sensor when the smoke sensor detects that the smoke concentration reaches a specified value, and recording the oil temperature at the moment by the temperature sensor to obtain the smoke point of the tea oil to be detected.

2. The rapid detection device for tea oil quality according to claim 1, wherein the RBF neural network is composed of an input layer, a hidden layer and an output layer, wherein,

the gaussian function of the hidden layer is as follows:

in the formula: x is an n-dimensional input vector, namely a characteristic value of a gas sensor signal;

c is the central vector of the ith hidden layer;

σ is the corresponding center width;

the input layer implements a non-linear mapping from x to R (x), and the output layer implements a linear mapping from R (x) to y

In the formula: p is the number of output nodes;

W_ikis the output weight;

y_kthe value is output by network, namely the value is the output value of acid value or the grade value of tea oil.

3. The device for rapidly detecting the quality of the tea oil as claimed in claim 1, wherein the signal data of the gas sensor is subjected to feature extraction, and the maximum peak value of the extracted signal, which is the most basic feature of the sensor signal and reflects the odor intensity of the sample, is recorded as max; the average value can reflect the average information of the odor of the sample and is marked as average; the peak time is the time at which the peak occurs, which reflects the degree of sensitivity to the sample, denoted as time; the three values are used as main characteristic values;

obtaining a feature vector R for each sample can be expressed as:

R＝[max average time]。

4. the device for rapidly detecting the quality of the tea oil according to claim 1, wherein an oil inlet pipe (7) communicated with the reaction tank (2) is installed at the upper part of the reaction tank (2), an oil inlet valve (8) for controlling the opening and closing of the oil inlet pipe (7) is arranged on the oil inlet pipe, an oil outlet pipe (9) communicated with the reaction tank (2) is installed at the bottom of the reaction tank (2), an oil outlet valve (10) opened and closed by a controller is arranged on the oil outlet pipe (9), the inner side wall of the bottom of the reaction tank (2) is an inclined surface, and the inlet end of the oil outlet pipe (9) is installed at the lower side close to the inclined surface.

5. The device for rapidly detecting the quality of the tea oil according to claim 1, wherein the diameter of one side of the reaction tank (2) close to the opening thereof is larger than the diameter of the inner side wall of the bottom of the reaction tank (2).

6. The device for rapidly detecting the quality of the tea oil according to claim 1, wherein an exhaust fan (11) is further installed at an opening of the detection box (3), the exhaust fan (11) is fixedly connected with the inner side wall of the bottom of the detection box (3), an air inlet of the exhaust fan (11) is communicated with an opening of the reaction tank through the opening of the detection box, and the exhaust fan can suck the gas in the reaction tank (2) into the detection box (3); and a switch (12) for controlling the opening and closing of the exhaust fan (11) is also arranged on the outer box body of the detection box (3).

7. The device for rapidly detecting the quality of the tea oil according to claim 1, wherein an exhaust pipe (13) communicated with the inside of the detection box (3) is further installed at the top of the detection box (3), and an exhaust valve for controlling the opening and the closing of the exhaust pipe (13) is installed on the exhaust pipe.

8. The device for rapidly detecting the quality of the camellia oil according to claim 1, wherein an insulation box (14) is further arranged outside the reaction tank (2), a through hole is formed in the top of the insulation box (14), the reaction tank (2) is arranged in the insulation box (14), the opening of the reaction tank (2) penetrates through the through hole to be fixedly connected with the reaction tank (2), and insulation materials are filled between the inner side wall of the insulation box (14) and the outer side wall of the reaction tank (2).

9. The device for rapidly detecting the quality of the tea oil according to claim 1, wherein the heater (1) is a constant temperature heater, a display screen is fixedly connected to the outer side wall of the detection box (3), and the display screen can display the detection result of the processed tea oil.

Technical Field

The invention relates to the field of edible oil processing, in particular to a device for quickly detecting the quality of tea oil.

Background

Tea oil, olive, coconut and palm are called as four major woody oil plants in the world, have the advantages of clear color, fragrant and strong taste, rich nutrition, storage tolerance and the like, are known as 'east olive oil', and are edible oil deeply loved by consumers.

The tea oil is very easily influenced by external factors in the processes of production, processing, storage and transportation, such as oxygen, temperature, moisture and processing methods, which can cause the phenomena of increased moisture content, more tea oil impurities, changed color, weakened fragrance and the like, so that the tea oil product with excellent quality on the market is particularly precious; in addition, some illegal vendors use the tea oil to be well loved by consumers at present, but the tea oil is generally sold and adulterated to gain the violence due to the characteristic of higher price, so that the problems of increase of harmful substances and reduction of physiologically active substances such as vitamins, linolenic acid and the like in the tea oil are solved. At present, the conventional detection method and the instrumental detection method are mainly used for detecting the quality of the tea oil. Most of the traditional detection methods and instrument detection methods are used for detecting and comparing single indexes of the tea oil, and the tea oil detection only judges the single indexes and does not have integrity and representativeness. For example, the method for detecting the acid value, i.e., the content of fatty acid, mainly comprises an acid-base titration method, an electrochemical method, a spectroscopic method, a chromatographic method, an iodometry method adopted in national standards and the like, and although the iodometry method has high accuracy, the method also has the problems of high detection cost, complicated operation, easy pollution to the environment and the like, has certain requirements on workers and environmental conditions, and is not suitable for on-site monitoring and use. The detection of the smell, namely the volatile component, of the tea oil usually adopts a solid phase microextraction-gas chromatography technology, the method depends on a fiber head to adsorb an object to be detected from a sample and desorbs the object to be detected in an analytical device, a solvent-free sample pretreatment technology is adopted, the adulteration amount is calculated through a linear equation, and the technology is long in time consumption, complicated in operation and inaccurate in result. The detection of the tea oil smoke point is based on an image processing technology, the smoke point is judged through image processing, but the detection speed is low, manpower and material resources are consumed, and objective and accurate effects cannot be achieved.

From the above, the existing detection methods have very complicated detection steps, and the problems of relatively complicated pretreatment process, slow detection speed, large amount of required chemical reagents and the like are only applicable to laboratory detection, so that the detection requirements of modern society on the simple, convenient, rapid, accurate and on-site detection of the quality of the edible tea oil are difficult to meet.

Disclosure of Invention

In view of the above-mentioned shortcomings of the prior art, the present invention aims to: the provided device for rapidly detecting the quality of the tea oil solves the problems of complex detection steps, complex pretreatment process, low detection speed and large amount of required chemical reagents in the existing detection method.

In order to solve the technical problems, the invention adopts the following technical scheme:

a rapid detection device for tea oil quality comprises a heater, a reaction tank for containing tea oil to be detected and a detection box, wherein the heater is arranged at the bottom of the reaction tank and can heat the reaction tank;

the reaction tank is also internally provided with a temperature sensor which is used for measuring the temperature of the tea oil to be measured in the reaction tank; a smoke sensor is arranged at the opening of the detection box and used for detecting the smoke concentration generated after the tea oil to be detected in the detection box is heated; the detection box comprises a detection box body, a detection box body and a detection box body, wherein the detection box body is provided with a detection box body, the detection box body is provided with a detection box;

the detection box is also provided with a data processor, the data processor comprises a signal amplifier, a singlechip and an AD conversion circuit, the signal input end of the signal amplifier is connected with the signal output end of each sensor, the signal amplifier transmits the collected electric signals to the AD conversion circuit, and the electric signals are transmitted to the singlechip by the AD conversion circuit for data processing;

the single chip microcomputer processes the electric signals after receiving the electric signals converted after the collection of the odor sensors, and the specific processing process is as follows: carrying out normalization processing and signal graphic analysis on the electric signal, preprocessing the electric signal by adopting a difference method and a difference quotient method, extracting the peak value, the peak time, the average value and the peak-out time of a signal curve as characteristic values, and resolving the characteristic values through a resolving model established by a RBF neural network to obtain an acid value and a smell grade;

heating the tea oil to be detected until the tea oil is fuming, feeding back the tea oil to the temperature sensor by the smoke sensor when the smoke sensor detects that the smoke concentration reaches a specified value, and recording the oil temperature at the moment by the temperature sensor to obtain the smoke point of the tea oil to be detected.

The acid value in the tea oil is detected through the gas sensor, the gas sensor is firstly screened, the gas sensor with high sensitivity to free fatty acid and decomposition products thereof is selected, volatile matters generated by the heated tea oil are detected through the gas sensor, electric signals are acquired, the electric signals are converted into specific data through a standard equation and a prediction model which are input into a single chip microcomputer in advance, and the acid value of the tea oil is judged through the data.

Similarly, the invention also utilizes the gas sensor to detect the odor grade of the tea oil, and has the advantages of rapidness, intelligence and accurate result. Through chemical analysis of the chemical composition of the tea oil smell, volatile substances in the tea oil mainly comprise one or more of aldehydes, ketones, alcohols, esters, alkenes, benzene rings, glucoside degradation products and heterocycles, the gas sensors are screened according to the volatile substances, and a plurality of gas sensors with high sensitivity to one or more substances are respectively selected.

Then, a gas sensor for detecting acid value and a gas sensor for detecting organic volatile substances are combined for use, the volatile substances generated after the tea oil sample to be detected is heated are detected through the gas sensors, electric signals obtained by the gas sensors are subjected to normalization processing and signal graph analysis, preprocessing is carried out by adopting a difference method and a difference quotient method, then stable values of parameters of a peak value, peak time, an average value and peak time of a signal curve are extracted as characteristic values, and an RBF neural network is adopted as a model equation establishing method. Simultaneously, can be according to gaseous flow law and experimental experience, install gas sensor in suitable position, for example, the equal position all around of detection case inside wall top, detection case inside wall makes gaseous abundant and gas sensor contact, has guaranteed the degree of accuracy that detects.

The smoke point of the tea oil sample to be detected can be rapidly detected through the temperature sensor and the smoke sensor, namely the smoke sensor detects the smoke concentration to reach a specified smoke concentration value, the temperature sensor works and detects the temperature value at the moment, and the tea oil smoke point can be accurately and timely detected.

Meanwhile, the data collected by the gas sensor, the temperature sensor and the smoke sensor are processed by the data processor. The signal amplifier is mainly used for amplifying, filtering and denoising sensor signals and preparing for converting and acquiring the signals and mainly comprises a filter circuit and an amplifying circuit. The AD conversion circuit is analog-to-digital conversion, converts analog signals collected by the gas sensor, the temperature sensor, the smoke sensor and the smoke sensor into digital signals, fuses detection results of all systems according to a certain weight, and finally feeds back real-time data. The single chip microcomputer is equivalent to a microcomputer and is the core of operation, and data acquisition, data processing, classification decision, fusion analysis algorithm and the like of multi-sensor fusion are written in the single chip microcomputer.

According to the method, the tea oil is heated, and the gas sensor is used for detecting the content of free fatty acid in the tea oil and comparing the content with the content of standard tea oil, so that the acid value of the tea oil to be detected is reflected; detecting the content of volatile gas in the tea oil by using a gas sensor so as to judge the odor grade of the tea oil; the smoke point of the tea oil is detected by using the temperature sensor, the data of the temperature sensor is compared with the smoke point of the pure tea oil, and the comprehensive quality of the tea oil is detected. The whole device is light, small and portable, has complete functions, and collects and analyzes a plurality of detection indexes in a short time under the condition of not damaging the original physicochemical properties of the tea oil, thereby evaluating the comprehensive quality of the tea oil. The content of free fatty acid in the tea oil is predicted by heating the tea oil and collecting volatile gas, so that the problems of long time consumption and complex operation of the traditional detection are solved, and the method has the characteristics of quickness and intelligence; temperature sensor and smoke transducer's cooperation is used, can gather the oil temperature when the tea-seed oil heating goes out the cigarette, and then judge the quality of tea-seed oil through the oil temperature, has avoided the contingency and the inaccuracy of image processing technique, can carry out quick, harmless, intelligent, comprehensive detection to the quality of tea-seed oil simultaneously. The data processor converts and amplifies the analog signals generated by the gas sensor, the smoke sensor and the temperature sensor into digital signals, so that the tea oil detection and adulteration technology is fused with the informatization technology, and the intelligence and the advancement of the technology are reflected. Whole detection device is light little portable, is convenient for move, puts, and the box is encapsulated situation, isolated external environment, prevents that gas from volatilizing, has guaranteed the accuracy nature that the tea-seed oil smell detected. The invention can adjust the weight of each index according to the requirements of different customers and make a personalized detection scheme for the customers.

Further, the RBF neural network is composed of an input layer, a hidden layer and an output layer, wherein the Gaussian function of the hidden layer is as follows:

in the formula: x is an n-dimensional input vector, namely a characteristic value of a gas sensor signal;

c is the central vector of the ith hidden layer;

σ is the corresponding center width.

The input layer implements a non-linear mapping from x to R (x), and the output layer implements a linear mapping from R (x) to y

In the formula: p is the number of output nodes;

W_ikis the output weight;

y_kthe value is output by network, namely the value is the output value of acid value or the grade value of tea oil.

An RBF network (radial basis function network) is a special forward neural network that has the property of mapping a non-linearly separable mode space to a linearly separable state space. The RBF network structure comprises an input layer, a hidden layer and an output layer, the structure of the RBF network structure is shown in figure 3, the input layer and the hidden layer are directly connected, hidden layer nodes are formed by radial action functions such as Gaussian kernel functions, and output layer nodes are simple linear functions. The function (basis function) in the hidden node will respond to the input signal locally, and when the input signal is close to the central range of the basis function, the hidden node will produce a larger output, and the network has local approximation capability.

Further, the signal data of the gas sensor is subjected to feature extraction, and the maximum peak value of the signal, which is the most basic feature of the sensor signal and reflects the odor intensity of the sample, is recorded as max; the average value can reflect the average information of the odor of the sample and is marked as average; the peak time is the time at which the peak occurs, which reflects the degree of sensitivity to the sample, denoted as time; the three values are used as main characteristic values;

obtaining a feature vector R for each sample can be expressed as:

R＝[max average time]。

further, retort upper portion install with the advance oil pipe that the retort is linked together is equipped with the inlet valve of its switching of control on advancing the oil pipe, the oil pipe that goes out that the retort is linked together is installed to the retort bottom, is equipped with the outlet valve of controller switching on going out the oil pipe, the inside wall of retort bottom is the inclined plane, the entrance point of going out oil pipe is installed in the low side department that is close to the inclined plane. The bottom of retort is the inclined plane, more is favorable to tea-seed oil to flow from the oil pipe, can not appear partial tea-seed oil and remain the condition in the retort bottom.

Further, the diameter of one side of the reaction tank close to the opening of the reaction tank is smaller than the diameter of the inner side wall of the bottom of the reaction tank. The opening of the reaction tank is smaller than the diameter of the inner side wall of the bottom of the reaction tank, the section of the whole reaction tank is in a round table shape, and the reaction tank with the structure is processed by simulating a nasal flow channel of a human sense organ system, so that volatile gas enters the detection device according to the neurophysiological structure of a biological olfactory system and the biochemical mechanism of olfactory sensation and flows out completely or with reduced loss to achieve the purpose of accurately detecting the volatile gas.

Furthermore, an exhaust fan is further installed at the position of the opening of the detection box, the exhaust fan is fixedly connected with the inner side wall of the bottom of the detection box, an air inlet of the exhaust fan is communicated with the opening of the reaction tank through the opening of the detection box, and the exhaust fan can suck gas in the reaction tank into the detection box; and a switch for controlling the opening and closing of the exhaust fan is further mounted on the outer box body of the detection box. The exhaust fan can be fast with the volatile substance suction detection case that the tea-seed oil that awaits measuring produced after the heating in the retort, make the gas sensor in the detection case can be more abundant and timely with the volatile substance contact of the tea-seed oil that awaits measuring.

Furthermore, the top of the detection box is also provided with an exhaust pipe communicated with the inside of the detection box, and the exhaust pipe is provided with an exhaust valve for controlling the opening and closing of the exhaust pipe. The blast pipe can be convenient for detect the gas outgoing in the back detection box that finishes to detect next time. Simultaneously, the blast pipe can also be used for keeping the pressure balance inside and outside the detection case.

Furthermore, the outside of retort still is equipped with the insulation can, and the through-hole has been seted up at the top of insulation can, the retort is arranged in the insulation can and the opening of retort passes through the through-hole with retort fixed connection, and it has heat preservation thermal insulation material to fill between the inside wall of insulation can and the lateral wall of retort. The heat preservation and insulation material is wrapped on the periphery of the reaction tank, so that the heat loss can be reduced, the temperature of the tea oil sample to be detected can be always kept at the most suitable detection temperature, and the accuracy of the detection result is improved.

Further, the heater is a constant temperature heater, the outer side wall of the detection box is fixedly connected with a display screen, and the display screen can display the detection result of the processed tea oil. The constant temperature heater ensures the relative stability of the temperature of the whole detection system, so that the detection process is not influenced by the outside. The constant temperature heater is adopted for heating to enable the temperature of the system to reach the temperature required by tea oil detection and keep for a period of time, the detection temperature is controlled through a single-factor test and an orthogonal test, and the most suitable detection temperature is found, so that the temperature of the constant temperature heater can be directly set to be the most suitable detection temperature, and the detection time is further saved.

The traditional method for detecting the acid value mainly comprises an acid-base titration method, an electrochemical method, a spectrum method, a chromatography method, an iodometry method adopted in national standards and the like, and although the iodometry method is high in accuracy, the traditional method also has the problems of high detection cost, complex operation, easiness in causing certain pollution to the environment and the like, has certain requirements on working personnel and environmental conditions, is not suitable for on-site monitoring and the like. In the traditional method, the volatile components of the tea oil are usually detected by adopting a solid-phase microextraction-gas chromatography technology, the method depends on a solvent-free sample pretreatment technology that a fiber head adsorbs a substance to be detected from a sample and desorbs the substance in analytical equipment, and the adulteration amount is calculated by a linear equation, so that the technology is long in time consumption, complicated in operation and inaccurate in result. The traditional detection of the tea oil smoke point is based on an image processing technology and is judged by image processing, but the detection speed is low, manpower and material resources are consumed, and objective and accurate effects cannot be achieved. The invention overcomes various defects in the traditional detection method of the tea oil, and can quickly and accurately detect the quality of the tea oil by adopting the rapid detection device of the tea oil quality.

Compared with the prior art, the invention has the following beneficial effects:

1. according to the invention, the gas sensor with high sensitivity to free fatty acid is screened, the volatile substances generated by the heated tea oil sample to be detected are detected by the gas sensor, and the acid value of the tea oil sample to be detected is obtained according to further calculation, so that the defects that the traditional method for detecting the acid value is high in cost, complicated to operate, easy to cause certain pollution to the environment, has certain requirements on workers and environmental conditions, and is not suitable for on-site monitoring and use are overcome;

2. according to the invention, through screening the gas sensors with high sensitivity to chemical volatile substances in the tea oil, especially the gas sensors with high sensitivity to one or more of aldehydes, ketones, alcohols, esters, alkenes, benzene rings, glucoside degradation products and/or heterocyclic chemical volatile substances, the gas sensors are used for detecting the chemical volatile substances in the gas generated by heating the tea oil sample to be detected, so that the evaluation on the tea oil smell is more accurate, and the method has important significance for improving the value of the tea oil and ensuring the health of consumers;

3. according to the invention, the temperature sensor and the smoke sensor can be used for timely detecting the oil temperature when the heated tea oil is discharged for the first time, so that the smoke point of the oil product can be accurately detected, and the problems that the traditional method is low in detection speed, consumes manpower and material resources and cannot achieve appreciable and accurate effects are solved;

4. the constant temperature heater is matched with the temperature sensor, so that the temperature of the system reaches and keeps the temperature required by tea oil detection, the detection temperature is controlled through a single-factor test and an orthogonal test, the most suitable detection temperature is found, and meanwhile, the heat-insulating material is filled between the reaction tank and the reaction box, so that the loss of the temperature is reduced, and the influence of temperature change on the detection result is reduced;

5. the device can be used for rapidly, nondestructively, intelligently and comprehensively detecting the quality of the tea oil, is portable, has complete functions, is convenient to move and place, collects and analyzes a plurality of detection indexes in a short time under the condition of not damaging the original physicochemical property of the tea oil, and evaluates the comprehensive quality of the tea oil.

Drawings

Fig. 1 is a schematic structural diagram of a device for rapidly detecting the quality of tea oil according to the present invention.

Fig. 2 is a signal response curve of the gas sensor to tea oil.

Fig. 3 is an RBF neural network.

In the figure: the device comprises a heater 1, a reaction tank 2, a detection box 3, a temperature sensor 4, a smoke sensor 5, a gas sensor 6, an oil inlet pipe 7, an oil inlet valve 8, an oil outlet pipe 9, an oil outlet valve 10, an exhaust fan 11, a switch 12, an exhaust pipe 13 and a heat preservation box 14.

Detailed Description

The invention will be further explained with reference to the drawings and the embodiments.