Gas chromatograph-mass spectrometer fault detection device and detection method

Abstract

The invention relates to the technical field of gas chromatograph-mass spectrometer, and discloses a gas chromatograph-mass spectrometer fault detection device, which comprises a computer processor and a control power supply, and is characterized in that the computer processor is in data connection with a temperature detection assembly and an air tightness detection assembly; the computer processor is in data connection with a pressure detection assembly, and the control power supply is electrically connected with a voltage detection assembly. According to the gas chromatograph-mass spectrometer fault detection device and detection method, comprehensive data detection is carried out by arranging a temperature detection assembly, an air tightness detection assembly, a pressure detection assembly and a voltage detection assembly, detected data are processed through a computer processor, and then historical fault data are called from the computer processor; and constructing a fault model based on a neural network according to the fault data, and analyzing the abnormal data of the gas chromatograph-mass spectrometer body through the fault model, thereby rapidly checking the fault problem of the gas chromatograph-mass spectrometer body.

Description

Technical Field

[0001] The invention relates to the technical field of gas chromatography-mass spectrometry, and in particular to a gas chromatography-mass spectrometry fault detection device and a detection method. Background Art

[0002] GC-MS is an instrument that combines gas chromatograph and mass spectrometer. Mass spectrometry can perform qualitative analysis, while chromatography can separate and analyze organic compounds. However, due to the complex structure of gas chromatograph and mass spectrometer, and the large number of internal components, GC-MS may fail during use. Therefore, it is necessary to use GC-MS fault detection device to troubleshoot GC-MS.

[0003] The conventional method for troubleshooting GC-MS is to use a variety of instruments and equipment to check and confirm the possible causes of the failure one by one. However, due to the complex structure of the GC-MS and the large number of internal components, the causes of the failure are also large, which leads to a long troubleshooting time and cumbersome troubleshooting steps. The operation volume in fault detection is large and extremely inconvenient.

[0004] Meanwhile, after searching, a gas chromatography-mass spectrometer fault detection device is disclosed in an invention patent with Chinese patent publication number CN115951002A. The gas chromatography-mass spectrometer fault detection device in the invention patent detects vacuum degree and voltage data of the gas chromatography-mass spectrometer through equipment such as a vacuum gauge, a voltage sensor and a communication device, and sorts the detected data and the acquired mass chromatogram data by fault priority through a constructed fault model. However, since there are many causes of gas chromatography-mass spectrometer failures, the gas chromatography-mass spectrometer fault detection device is difficult to obtain more comprehensive fault data only through a vacuum gauge and a voltage sensor. Therefore, the fault detection accuracy of the fault detection device needs to be further improved. Therefore, a gas chromatography-mass spectrometer fault detection device and a detection method are proposed. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the shortcomings of the prior art, the present invention provides a gas chromatography-mass spectrometer fault detection device and a detection method, which have the advantages of collecting fault data from multiple aspects and quickly troubleshooting faults by building a training model, thereby solving the problem that the gas chromatography-mass spectrometer fault detection device in the above-mentioned background technology has a small data collection range and cannot realize comprehensive fault troubleshooting.

[0007] (II) Technical solution

[0008] In order to achieve the purpose of collecting fault data from multiple aspects and quickly troubleshooting faults by building a training model, the present invention provides the following technical solution: a gas chromatography-mass spectrometer fault detection device, including a computer processor and a control power supply, the computer processor data is connected to a temperature detection component, the computer processor data is connected to an air tightness detection component, the computer processor data is connected to a pressure detection component, and the control power supply is electrically connected to a voltage detection component.

[0009] Preferably, the control power supply is electrically connected to the gas spectrometer body through a cable, the computer processor includes a data acquisition module, a data processing module and a data storage module, the gas spectrometer body includes an ion source and an analysis detector, the analysis detector receives ions and generates a mass spectrometer signal, and transmits the mass spectrometer signal to the data acquisition module through the analysis detector, and the data processing module is connected to an image display for generating mass spectrometer image data according to the acquired mass spectrometer signal.

[0010] Preferably, the gas chromatography-mass spectrometer body includes an injection port and a chromatographic interface, the temperature detection component includes a temperature controller arranged on the gas chromatography-mass spectrometer body, which is used to control the temperature values ​​of the ion source, the injection port and the chromatographic interface, the temperature detection component includes a temperature detector, which is used to detect the actual temperature data of the ion source, the injection port and the chromatographic interface, and the temperature detector is data-connected to the data acquisition module.

[0011] Preferably, the airtightness detection component includes a gas leak detector, which is a thermal conductivity gas sensor structure. The pipeline connection components of the gas spectrometer body are detected according to the thermistor of the thermal conductivity gas sensor. The gas spectrometer body includes a mechanical pump and a vacuum pump. The airtightness detection component includes a vacuum gauge connected to the mechanical pump and the vacuum pump through an electrode connector. The vacuum gauge and the gas leak detector are both data-connected to the data acquisition module.

[0012] Preferably, the gas chromatography-mass spectrometer body includes an injection chamber, a separation chamber and a detection chamber, the pressure detection component includes an air pressure sensor, the air pressure sensor is provided with a pressure detection terminal, the pressure detection terminal includes a pressure-sensitive film and a resistor element electrically connected to the pressure-sensitive film, and is used to detect the air pressure data of the injection chamber, the separation chamber and the detection chamber, the pressure detection component includes a gas flow sensor, the gas flow sensor is provided with a vortex street velocity measurement probe, and is used to detect the gas flow data in the gas chromatography-mass spectrometer body pipeline, and the air pressure sensor and the gas flow sensor are both data-connected to the data acquisition module.

[0013] Preferably, the voltage detection component includes a multimeter and a communication data line connected to the multimeter interface, which is used to detect the line data of the control power supply and the gas chromatography-mass spectrometer body. The multimeter is connected to the data acquisition module through the transmission interface, and the data storage module stores normal instrument data and historical fault data.

[0014] A method for detecting a gas chromatography-mass spectrometer fault comprises the following steps:

[0015] S1. Collect the working data of the control power supply and the gas chromatography-mass spectrometer body through the temperature detection component, the air tightness detection component, the pressure detection component and the voltage detection component respectively;

[0016] S2. The collected working data is transmitted to the data processing module through the data acquisition module for processing to obtain an abnormal data set;

[0017] S3. Collect historical fault data from the data storage module and establish a fault detection model;

[0018] S4. Input the collected abnormal data set into the fault detection model, analyze and output the problem data type of the gas chromatography-mass spectrometer body.

[0019] Preferably, the historical fault data is obtained by accessing the data storage module through the communication interface to form a historical fault data set S n (S n =X1,X2,X3,...X n ), for the historical fault data set S n Perform data standardization processing, the standardization processing formula is:

[0020] Where Z i represents the eigenvalue of the i-th sample after standardization, X i represents the eigenvalue of the i-th sample, Represents the historical fault data set S n The average value of s represents the historical fault data set S n The standard deviation of .

[0021] Preferably, for the historical fault data set S n After standardization, we get the fault data set Z n , divide the fault data set Z n A training sample set and a test sample set are obtained, and an input layer, a convolution layer, a pooling layer, an activation function ReLU and an output layer are established, as well as a loss function, an optimizer and training parameters are constructed to construct a convolutional neural network model, wherein the convolution layer includes a first convolution layer C1, a second convolution layer C2 and a third convolution layer C3, the pooling layer includes a first pooling layer S1 and a second pooling layer S2, and the training parameters include the number of training times, the number of tests and the number of training rounds.

[0022] Preferably, the sample data of the training sample set is input into the input layer, and converted into a one-dimensional vector after feature extraction through the convolution layer and the pooling layer. The obtained one-dimensional vector is input into the output layer, and the fault problem model is output. The error is back-propagated and the model parameters are updated according to the optimizer. The model is repeatedly trained and iterated until the model converges or reaches the set number of training rounds. The model is evaluated according to the test sample set, the abnormal data set is input into the model, and the problem data type of the gas chromatography-mass spectrometer body is output.

[0023] (III) Beneficial effects

[0024] Compared with the prior art, the present invention provides a gas chromatography-mass spectrometry instrument fault detection device and detection method, which have the following beneficial effects:

[0025] The gas chromatography-mass spectrometer fault detection device and detection method perform comprehensive data detection on the gas chromatography-mass spectrometer body by setting a temperature detection component, an air tightness detection component, a pressure detection component and a voltage detection component, transmit the detected data to a computer processor for processing, then retrieve historical fault data from the computer processor, build and train a fault model based on a neural network according to the fault data, analyze the abnormal data of the gas chromatography-mass spectrometer body through the fault model, and thus quickly find out the fault problem of the gas chromatography-mass spectrometer body. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is an overall schematic diagram of the gas chromatography-mass spectrometer fault detection device of the present invention;

[0027] Figure 2 It is a schematic diagram of the structure of the gas chromatography-mass spectrometer and the control power supply of the present invention;

[0028] Figure 3 This is a schematic diagram of the structure of the temperature detection component of the present invention;

[0029] Figure 4 It is a schematic diagram of the structure of the airtightness detection component of the present invention;

[0030] Figure 5 This is a schematic diagram of the structure of the pressure detection component of the present invention;

[0031] Figure 6 This is a schematic diagram of the structure of the voltage detection component of the present invention;

[0032] Figure 7 The present invention is a flow chart of the gas chromatography-mass spectrometer fault detection method.

[0033] In the figure: 1. Computer processor; 2. Control power supply; 3. Temperature detection component; 301. Temperature controller; 302. Temperature detector; 4. Air tightness detection component; 401. Gas leak detector; 402. Vacuum gauge; 5. Pressure detection component; 501. Air pressure sensor; 502. Gas flow sensor; 6. Voltage detection component; 601. Multimeter; 602. Communication data cable; 7. Gas chromatography-mass spectrometry instrument body. DETAILED DESCRIPTION

[0034] 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.

[0035] See also Figure 1-7 A gas chromatography-mass spectrometry fault detection device includes a computer processor 1 and a control power supply 2. The computer processor 1 is data-connected with a temperature detection component 3, the computer processor 1 is data-connected with a gas tightness detection component 4, the computer processor 1 is data-connected with a pressure detection component 5, and the control power supply 2 is electrically connected with a voltage detection component 6.

[0036] Furthermore, the control power supply 2 is electrically connected to the gas spectrometer body 7 through a cable, the computer processor 1 includes a data acquisition module, a data processing module and a data storage module, the gas spectrometer body 7 includes an ion source and an analysis detector, the analysis detector receives ions and generates a mass spectrometer signal, and transmits the mass spectrometer signal to the data acquisition module through the analysis detector. The data processing module is connected to an image display for generating mass spectrometer image data according to the acquired mass spectrometer signal.

[0037] Specifically, the gas chromatography-mass spectrometer body 7 mainly includes a gas path system and a circuit system, which can be divided into a carrier gas system, a sampling system, a chromatographic column and column temperature, a detection system and a data recording system according to functional units. The mixture sample is separated by the chromatographic column and enters the ion source of the mass spectrometer, and ions are ionized in the ion source. After passing through the mass analyzer and the detector, the ions become mass spectrometry signals and are input into the computer processor 1. The curve of ion intensity changing with time is displayed on the image display, thereby obtaining a mass spectrum.

[0038] Furthermore, the gas chromatography-mass spectrometer body 7 includes an injection port and a chromatographic interface, the temperature detection component 3 includes a temperature controller 301 arranged on the gas chromatography-mass spectrometer body 7, which is used to control the temperature values ​​of the ion source, the injection port and the chromatographic interface, and the temperature detection component 3 includes a temperature detector 302, which is used to detect the actual temperature data of the ion source, the injection port and the chromatographic interface, and the temperature detector 302 is data-connected to the data acquisition module.

[0039] Specifically, the temperature controller 301 controls the heating devices on the ion source, the injection port and the chromatographic interface, and then controls the temperature value. If the ion source temperature is too low, the sample is adsorbed; if the temperature of the gas chromatographic interface is too low and the final temperature of the gas chromatographic heating program is too low, the high boiling point compound sensitivity is low and the peak shape is poor; if the temperature of the injector and the gas chromatographic interface is too low, the peak tailing problem will occur; if the temperature of the injector is too low and the gas chromatographic heating is too slow, the chromatographic peak will be too wide.

[0040] Therefore, the temperature detector 302 is used to collect the actual temperature data of the ion source, injection port and chromatographic interface respectively, which is convenient for troubleshooting and temperature control operation, judges the abnormality of temperature data, and transmits the detected temperature data to the data acquisition module.

[0041] Furthermore, the airtightness detection component 4 includes a gas leak detector 401, which is a thermal conductivity gas sensor structure. The pipeline connection components of the gas spectrometer body 7 are detected according to the thermistor of the thermal conductivity gas sensor. The gas spectrometer body 7 includes a mechanical pump and a vacuum pump. The airtightness detection component 4 includes a vacuum gauge 402 connected to the mechanical pump and the vacuum pump through an electrode connector. The vacuum gauge 402 and the gas leak detector 401 are both data-connected to the data acquisition module.

[0042] Specifically, the thermal conductivity gas sensor heats the thermistor and uses the thermal conductivity of the gas to carry away the thermistor. At this time, the thermistor changes its resistance data as the temperature drops. The changed resistance data is converted into a voltage data signal by a signal conditioning and conversion circuit, and the corresponding gas concentration detection is converted by the Wheatstone bridge principle, thereby quickly detecting the air tightness of the pipeline connection part of the gas chromatograph body 7;

[0043] The cold cathode of the vacuum gauge 402 ionizes the gas molecules in the mechanical pump and the vacuum pump, and the air pressure is measured by collecting the ion current to determine the leakage situation, and the detected gas concentration data and vacuum degree data are transmitted to the data acquisition module.

[0044] Furthermore, the gas chromatography-mass spectrometer body 7 includes an injection chamber, a separation chamber and a detection chamber, the pressure detection component 5 includes an air pressure sensor 501, the air pressure sensor 501 is provided with a pressure detection terminal, the pressure detection terminal includes a pressure-sensitive film and a resistor element electrically connected to the pressure-sensitive film, and is used to detect the air pressure data of the injection chamber, the separation chamber and the detection chamber, the pressure detection component 5 includes a gas flow sensor 502, the gas flow sensor 502 is provided with a vortex street velocity measurement probe, and is used to detect the gas flow data in the pipeline of the gas chromatography-mass spectrometer body 7, and the air pressure sensor 501 and the gas flow sensor 502 are both connected to the data acquisition module.

[0045] Specifically, the pressure-sensitive film is deformed by the gas pressure, so that the resistor element generates a resistance change signal, and the signal processing circuit inside the air pressure sensor 501 converts the resistance change signal into air pressure data, thereby detecting the air pressure data of the injection chamber, the separation chamber, and the detection chamber;

[0046] The vortex velocity measuring probe of the gas flow sensor 502 adopts an embedded piezoelectric crystal structure to convert the Karman vortex frequency into a voltage pulse signal, thereby detecting the gas flow data in the pipeline through the change of the voltage pulse signal, judging whether there is a foreign object blocking the pipeline, and transmitting the air pressure data and gas flow data to the data acquisition module.

[0047] Furthermore, the voltage detection component 6 includes a multimeter 601 and a communication data line 602 connected to the interface of the multimeter 601, which is used to detect the line data of the control power supply 2 and the gas chromatography-mass spectrometer body 7. The multimeter 601 is connected to the data acquisition module through the transmission interface, and the data storage module stores normal instrument data and historical fault data.

[0048] Specifically, the voltage and current data of each connection line of the control power supply 2 and the gas chromatography-mass spectrometer body 7 are detected by the multimeter 601, and the abnormal lines of the control power supply 2 and the gas chromatography-mass spectrometer body 7 are determined, and the abnormal line data is transmitted to the data acquisition module, and stored in the data storage module after being processed by the data processing module.

[0049] A method for detecting a gas chromatography-mass spectrometer fault comprises the following steps:

[0050] S1. Collect the working data of the control power supply 2 and the gas chromatography-mass spectrometer body 7 through the temperature detection component 3, the air tightness detection component 4, the pressure detection component 5 and the voltage detection component 6 respectively;

[0051] S2. The collected working data is transmitted to the data processing module through the data acquisition module for processing to obtain an abnormal data set;

[0052] S3. Collect historical fault data from the data storage module and establish a fault detection model;

[0053] S4. Input the collected abnormal data set into the fault detection model, analyze and output the problem data type of the gas chromatography-mass spectrometer body 7.

[0054] Furthermore, the historical fault data is obtained by accessing the data storage module through the communication interface to form a historical fault data set S n (S n =X1,X2,X3,...X n ), for the historical fault data set S n Perform data standardization, the standardization formula is:

[0055] Where Z i represents the eigenvalue of the i-th sample after standardization, X i represents the eigenvalue of the i-th sample, Represents the historical fault data set S n The average value of s represents the historical fault data set S n The standard deviation of .

[0056] Specifically, standardized fault data is obtained by standardizing historical fault data, and training sample sets and test sample sets are obtained by dividing the fault data to perform model training and testing.

[0057] Furthermore, for the historical fault dataset S n After standardization, we get the fault data set Z n , divide the fault data set Z n The training sample set and the test sample set are obtained, and the input layer, convolution layer, pooling layer, activation function ReLU and output layer are established, as well as the loss function, optimizer and training parameters are constructed to construct a convolutional neural network model. The convolution layer includes the first convolution layer C1, the second convolution layer C2 and the third convolution layer C3, the pooling layer includes the first pooling layer S1 and the second pooling layer S2, and the training parameters include the number of training times, the number of tests and the number of training rounds.

[0058] Specifically, the first convolution layer C1 receives the output of the input layer, receives the output of the first convolution layer C1 through the first pooling layer S1, receives the output of the first pooling layer S1 through the second convolution layer C2, receives the output of the second convolution layer C2 through the second pooling layer S2, receives the output of the second pooling layer S2 through the third convolution layer C3, and inputs the output of the third convolution layer C3 to the output layer.

[0059] Furthermore, the sample data of the training sample set is input into the input layer, and the feature is extracted through the convolution layer and the pooling layer and then converted into a one-dimensional vector. The obtained one-dimensional vector is input into the output layer, and the fault problem model is output. The error is back-propagated according to the optimizer and the model parameters are updated. The model is repeatedly trained and iterated until the model converges or reaches the set number of training rounds. The model is evaluated according to the test sample set, the abnormal data set is input into the model, and the problem data type of the gas chromatography-mass spectrometer body 7 is output.

[0060] Specifically, the model is optimized through iterative training of sample data of the training sample set, and the accuracy of the model is evaluated using the test sample set. The abnormal data set is input to the model, and then the problem data type is output.

[0061] In summary, the gas chromatography-mass spectrometer fault detection device and detection method, by setting the temperature detection component 3, the airtightness detection component 4, the pressure detection component 5 and the voltage detection component 6, perform comprehensive data detection on the gas chromatography-mass spectrometer body 7, transmit the detected data to the computer processor 1 for processing, then retrieve the historical fault data from the computer processor 1, build and train a fault model based on a neural network according to the fault data, analyze the abnormal data of the gas chromatography-mass spectrometer body 7 through the fault model, so as to quickly find out the fault problem of the gas chromatography-mass spectrometer body 7.

[0062] 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. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0063] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A gas chromatography-mass spectrometry instrument fault detection device, characterized in that: The invention comprises a computer processor (1) and a control power supply (2), wherein the computer processor (1) is data-connected to a temperature detection component (3), the computer processor (1) is data-connected to an airtightness detection component (4), the computer processor (1) is data-connected to a pressure detection component (5), and the control power supply (2) is electrically connected to a voltage detection component (6).

2. The gas chromatography-mass spectrometry fault detection device according to claim 1, characterized in that: The control power supply (2) is electrically connected to a gas spectrometer body (7) via a cable. The computer processor (1) comprises a data acquisition module, a data processing module and a data storage module. The gas spectrometer body (7) comprises an ion source and an analysis detector. The analysis detector receives ions and generates a mass spectrum signal, and transmits the mass spectrum signal to the data acquisition module via the analysis detector. The data processing module is connected to an image display for generating mass spectrum image data according to the acquired mass spectrum signal.

3. A gas chromatography-mass spectrometry fault detection device according to claim 2, characterized in that: The gas chromatography-mass spectrometer body (7) comprises an injection port and a chromatographic interface. The temperature detection component (3) comprises a temperature controller (301) arranged on the gas chromatography-mass spectrometer body (7) and used to control the temperature values ​​of the ion source, the injection port and the chromatographic interface. The temperature detection component (3) comprises a temperature detector (302) and used to detect actual temperature data of the ion source, the injection port and the chromatographic interface. The temperature detector (302) is data-connected to a data acquisition module.

4. A gas chromatography-mass spectrometry fault detection device according to claim 3, characterized in that: The airtightness detection component (4) comprises a gas leak detector (401), which is a thermal conductivity gas sensor structure. The pipeline connection component of the gas chromatograph body (7) is detected according to the thermistor of the thermal conductivity gas sensor. The gas chromatograph body (7) comprises a mechanical pump and a vacuum pump. The airtightness detection component (4) comprises a vacuum gauge (402) connected to the mechanical pump and the vacuum pump via an electrode connector. The vacuum gauge (402) and the gas leak detector (401) are both data-connected to a data acquisition module.

5. A gas chromatography-mass spectrometry fault detection device according to claim 4, characterized in that: The gas chromatography-mass spectrometer body (7) comprises an injection chamber, a separation chamber and a detection chamber; the pressure detection component (5) comprises an air pressure sensor (501); the air pressure sensor (501) is provided with a pressure detection terminal; the pressure detection terminal comprises a pressure-sensitive film and a resistor element electrically connected to the pressure-sensitive film, and is used to detect air pressure data of the injection chamber, the separation chamber and the detection chamber; the pressure detection component (5) comprises a gas flow sensor (502); the gas flow sensor (502) is provided with a vortex velocity measurement probe, and is used to detect gas flow data in a pipeline of the gas chromatography-mass spectrometer body (7); the air pressure sensor (501) and the gas flow sensor (502) are both data-connected to a data acquisition module.

6. A gas chromatography-mass spectrometry fault detection device according to claim 2, characterized in that: The voltage detection component (6) comprises a multimeter (601) and a communication data line (602) connected to an interface of the multimeter (601), and is used to detect line data of a control power supply (2) and a gas chromatography-mass spectrometer body (7). The multimeter (601) is connected to a data acquisition module via a transmission interface, and the data storage module stores normal instrument data and historical fault data.

7. A method for detecting faults of a gas chromatography-mass spectrometer, characterized in that: The method comprises a gas chromatography-mass spectrometer fault detection device as described in claims 1 to 6, and further comprises the following operating steps: S1. Collecting the working data of the control power supply (2) and the gas chromatography-mass spectrometer body (7) through the temperature detection component (3), the air tightness detection component (4), the pressure detection component (5) and the voltage detection component (6); S2. The collected working data is transmitted to the data processing module through the data acquisition module for processing to obtain an abnormal data set; S3. Collect historical fault data from the data storage module and establish a fault detection model; S4. Input the collected abnormal data set into the fault detection model, analyze and output the problem data type of the gas chromatography-mass spectrometer body (7).

8. A method for detecting faults in a gas chromatograph-mass spectrometer according to claim 7, characterized in that: Access the data storage module through the communication interface to obtain historical fault data and form a historical fault data set S n (S n =X1,X2,X3,...X n ), for the historical fault data set S n Perform data standardization processing, the standardization processing formula is: Where Z i represents the eigenvalue of the i-th sample after standardization, X i represents the eigenvalue of the i-th sample, Represents the historical fault data set S n The average value of s represents the historical fault data set S n The standard deviation of .

9. A method for detecting faults in a gas chromatograph-gas spectrometer according to claim 8, characterized in that: For the historical fault dataset S n After standardization, we get the fault data set Z n , divide the fault data set Z n A training sample set and a test sample set are obtained, and an input layer, a convolution layer, a pooling layer, an activation function ReLU and an output layer are established, as well as a loss function, an optimizer and training parameters are constructed to construct a convolutional neural network model, wherein the convolution layer includes a first convolution layer C1, a second convolution layer C2 and a third convolution layer C3, the pooling layer includes a first pooling layer S1 and a second pooling layer S2, and the training parameters include the number of training times, the number of tests and the number of training rounds.

10. A method for detecting faults in a gas chromatograph-mass spectrometer according to claim 9, characterized in that: The sample data of the training sample set is input into the input layer, and the feature is extracted through the convolution layer and the pooling layer and then converted into a one-dimensional vector. The obtained one-dimensional vector is input into the output layer, and the fault problem model is output. The error is back-propagated according to the optimizer and the model parameters are updated. The model is repeatedly trained and iterated until the model converges or reaches the set number of training rounds. The model is evaluated according to the test sample set, the abnormal data set is input into the model, and the problem data type of the gas chromatography-mass spectrometer body (7) is output.

Images