An analysis and learning control system for an electronic nose

Abstract

The invention discloses an analysis and learning control system for an electronic nose, belonging to the technical field of electronic nose control. The control system comprises a factor analysis module and a detection control module. The factor analysis module is used for analyzing a target electronic nose to determine various environmental influencing factors, and further analyzing each environmental influencing factor to determine various combined segmented intervals corresponding to the electronic nose and corresponding error influence curves. The detection control module is used for performing detection control, determining each target electronic nose, and marking each target electronic nose with a corresponding similar mark. A target electronic nose is selected from each target electronic nose with each similar mark as a standard electronic nose. Detection data of each target electronic nose is acquired in real time. A detection learning model is established. A standard detection result of the standard electronic nose is acquired. Detection data of a target electronic nose having the same similar mark as the standard electronic nose is corrected by the detection learning model to obtain corrected detection data of each target electronic nose.

Description

Technical Field

[0001] The invention belongs to the technical field of electronic nose control, in particular to an analysis and learning control system of an electronic nose. Background Art

[0002] Electronic nose technology, an intelligent device that mimics the mammalian olfactory system, has shown broad application prospects in recent years across multiple fields. Composed primarily of a sensor array, a signal processing system, and a pattern recognition system, it enables rapid and accurate detection of target gases or volatile compounds.

[0003] Since the concept of the electronic nose was first proposed, the technology has garnered widespread attention from both academia and industry. After decades of development, electronic nose technology has made significant progress and has played a vital role in areas such as food safety, environmental monitoring, and medical diagnostics. For example, in the food safety field, electronic nose technology can rapidly assess food freshness, spoilage, and adulteration by detecting volatile compounds in food. In environmental monitoring, electronic nose technology can be used to monitor harmful gases and pollutants in the air in real time, providing scientific support to environmental protection agencies.

[0004] However, despite the success of electronic nose technology, much remains to be done. Existing electronic nose systems mostly rely on traditional pattern recognition algorithms, which often perform poorly when dealing with complex gas mixtures and struggle to accurately identify target gases. Furthermore, existing systems are poorly adaptable to environmental changes, severely impacting their performance once environmental conditions change.

[0005] For example, the Chinese patent application with publication number CN108535426A discloses an electronic nose control system for food monitoring and early warning, which includes an air circuit unit and a circuit unit. The air circuit unit includes an air filter device, an electromagnetic three-way valve, a container to be tested, a sensor chamber, a flow meter and a vacuum pump; the circuit unit includes a main control module, a sensor detection module, a control circuit module and a human-computer interaction module, and the human-computer interaction module includes an RS232 serial port, an LED prompt light, a display screen, a wireless module and a wireless terminal; the RS232 serial port, the LED prompt light and the display screen are all connected to the second output end of the main control module, and the RS232 serial port, the wireless module and the wireless terminal are connected in sequence. By adopting common circuits and electronic components, the structure is simple, the detection speed is fast, the detection accuracy is high, and the cost is reduced; however, the above-mentioned problems still exist.

[0006] Based on this, the present invention provides an analysis and learning control system for an electronic nose. Summary of the Invention

[0007] In order to solve the problems existing in the above solutions, the present invention provides an analysis and learning control system for an electronic nose.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] An analysis and learning control system for an electronic nose, comprising a factor analysis module and a detection control module;

[0010] The factor analysis module is used to analyze the target electronic nose to determine various environmental influencing factors, and to analyze each of the environmental influencing factors to determine each combination segment interval corresponding to the electronic nose and the corresponding error influence curve.

[0011] Furthermore, the methods for analyzing various environmental influencing factors include:

[0012] Obtain impact analysis data corresponding to each environmental influencing factor; analyze the impact analysis data to determine the factor interval corresponding to the environmental influencing factor, split and merge the factor interval corresponding to each environmental influencing factor to obtain each combined segmented interval; and generate a corresponding error impact curve based on each combined segmented interval.

[0013] Furthermore, the method for splitting and merging the factor intervals of each environmental influencing factor includes:

[0014] Step SA1: Set floating points within each factor interval;

[0015] Step SA2: setting a reference point combination according to each floating point, and determining a point combination to be selected according to the reference point combination;

[0016] Step SA3: Evaluate the reference point combination and the candidate point combination using a preset judgment model to obtain corresponding judgment values; merge the candidate point combination with a judgment value of 0 with the reference point combination to obtain an initial combination interval;

[0017] Step SA4: Determine a new candidate point combination based on the initial combination interval; evaluate the reference point combination and the candidate point combination within the initial combination interval using the judgment model to obtain a corresponding judgment value; merge the candidate point combination with a judgment value of 0 with the initial combination interval to obtain a new initial combination interval;

[0018] Step SA5: looping step SA4 until there is no candidate point combination with a judgment value of 0, marking the current initial combination interval as a combination segment interval;

[0019] Step SA6: Loop steps SA2 to SA5 until there is no reference point combination, and obtain the segmented intervals of each combination.

[0020] Furthermore, the expression of the judgment model is ;

[0021] Where: q is the input data, and the output data is the judgment value 1 or 0.

[0022] Furthermore, the method for determining environmental impact factors includes:

[0023] Establishing a reference library, wherein the reference library is used to store each reference electronic nose information and each reference factor corresponding to each reference electronic nose information;

[0024] Obtain target electronic nose information, analyze the target electronic nose information and each reference electronic nose information in the reference library through a preset equivalent analysis model, and obtain equivalent evaluation values between the target electronic nose information and each reference electronic nose information.

[0025] Equivalent electronic noses are determined based on equivalent evaluation values, reference factors corresponding to each equivalent electronic nose are obtained, and the obtained reference factors are integrated to obtain initial factors of the target electronic nose; simulation experiments are conducted on each initial factor to determine the environmental influencing factors of the target electronic nose.

[0026] Furthermore, the expression of the equivalent analysis model is ;

[0027] Where: s is the target electronic nose information and the reference electronic nose information; the output data is the equivalent evaluation value 1 or 0.

[0028] The detection control module is used to perform detection control, determine each target electronic nose, obtain target environment data of each target electronic nose, extract features of the target environment data according to various environmental influencing factors, obtain target environment features, match corresponding combined segment intervals according to each target environment feature, mark each target electronic nose with the same combined segment interval with a corresponding similar mark; and select one target electronic nose from each target electronic nose with each similar mark as a standard electronic nose;

[0029] Acquire detection data of each target electronic nose in real time, the detection data including collected data and detection results;

[0030] Establish a corresponding detection learning model based on the deep learning network;

[0031] Obtain standard detection results of the standard electronic nose; input the standard detection results and corresponding detection data of the standard electronic nose into the detection learning model, and correct the detection data of the target electronic nose with the same type of mark as the standard electronic nose through the detection learning model to obtain the detection correction data of each target electronic nose.

[0032] Furthermore, the method for obtaining the standard test results includes:

[0033] Acquire detection data corresponding to a standard electronic nose and determine each candidate gas; match the error influence curve corresponding to each candidate gas according to the combined segmented interval corresponding to the standard electronic nose; obtain a single detection result for each candidate gas according to each error influence curve; identify the error value corresponding to each single detection result according to the error influence curve; correct each single detection result according to each single detection result and the error value to obtain a standard detection result.

[0034] Furthermore, the method for determining the standard electronic nose includes:

[0035] Acquire the detection data of each target electronic nose with the same or similar label, identify the gas type and gas concentration corresponding to each detection data; after sorting, obtain the various gas types, and set the gas statistics template according to the various gas types;

[0036] The detection data of each target electronic nose is counted using a gas statistics template to obtain gas statistics of each gas electronic nose;

[0037] The gas type is marked as i, i = 1, 2, ..., n, n is a positive integer; the target electronic nose is marked as j, j = 1, 2, ..., m, m is a positive integer; the gas concentration is marked as Cij;

[0038] According to the formula Calculate the gas evaluation value of the corresponding gas type in each target electronic nose;

[0039] Where: PAij is the gas assessment value;

[0040] According to the formula Calculate the priority value of each target electronic nose;

[0041] Where: PUYj is the priority value;

[0042] The target electronic nose with the largest priority value is selected as the standard electronic nose.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] Through the mutual cooperation between the factor analysis module and the detection control module, the electronic noses in the area can be analyzed, learned and controlled. By setting up the detection control module, the detection and control of the electronic noses can be realized, and the detection accuracy of the electronic noses can be improved. At the same time, based on the detection situation of the standard electronic nose, dynamic correction can be achieved for each target electronic nose with the same label. On the premise of improving the detection accuracy, excessive replacement and modification of existing electronic noses can be avoided, thereby improving resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0047] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] like Figure 1 As shown, an analysis and learning control system of an electronic nose includes a factor analysis module and a detection control module;

[0049] The factor analysis module is used to analyze various environmental factors, determine the environmental factors that have an impact on the type of electronic nose, and analyze each environmental factor to determine the impact of different environmental factor data on the detection of different types of gases. The specific process is as follows:

[0050] Obtain a large amount of historical electronic nose detection data, classify the same electronic nose historical detection data into a category, identify the differences between the categories, and based on the differences, calculate the factors that affect the environment of this type of electronic nose and determine the various environmental influencing factors.

[0051] And determine the corresponding impact analysis data based on each environmental impact data, that is, the relevant impact data corresponding to the environmental impact factors;

[0052] Analyze the impact analysis data to determine the factor interval corresponding to the environmental impact factor, that is, the data range in which the environmental impact factor may appear in actual application; split and merge the factor intervals corresponding to each environmental impact factor to obtain each combined segment interval;

[0053] According to the different electronic nose control parameters, the error influence curve of each single gas is generated in each combined segmented interval; that is, the horizontal axis is each electronic nose control parameter, and the vertical axis is the detection error of the electronic nose control parameter in the environment. The statistical setting is based on a large amount of data, and the average value is used to represent the error; the environmental impact in the same combined segmented interval is the same; and the error influence curve is generated for a single gas; therefore, in the above process, when selecting the corresponding historical data for analysis, it can be first classified according to the type of gas; determine the influence of different electronic nose control parameters of different gas types in the combined segmented interval.

[0054] Methods for splitting and merging factor intervals corresponding to various environmental influencing factors include:

[0055] A floating point is set in each factor interval. The floating point is the environmental influencing factor data corresponding to the factor interval. The corresponding environmental influencing factor data is determined according to the position of the floating point in the factor interval. A judgment model is established. The judgment model is used to judge whether the environmental influencing factor data combination corresponding to each floating point has the same influence on gas detection analysis as the previous environmental influencing factor data combination. Training and judgment are performed based on the detection results corresponding to the environmental influencing factor data combination, without considering the influence of different electronic nose control parameters, that is, only considering the influence of the environment. Because the environmental influencing factor data is obtained from the historical detection data of the electronic nose or the simulation experiment results, the corresponding detection results can be determined, and then they can be organized as training data for establishing the judgment model. The expression is: , where: q is the input data, i.e., the environmental impact factor data combination corresponding to each floating point, and the output data is the judgment value 1 or 0; different analysis results refer to the different environmental impacts of the current floating point's environmental impact factor data combination and the environmental impact factor data combination compared with it on the detection and analysis;

[0056] A reference point combination is set based on each floating point, that is, a floating point combination corresponding to any non-combined segmented interval; a candidate point combination is determined based on the reference point combination, and the candidate point combination is used as the reference point combination. The floating points are changed in the order of the factor intervals to form a floating point combination that is different from the reference point combination but not within the combined segmented interval;

[0057] The reference point combination and the candidate point combination are evaluated by the judgment model to obtain the corresponding judgment value; the candidate point combination with a judgment value of 0 is merged with the reference point combination to obtain an initial combination interval; a new candidate point combination is determined based on the initial combination interval, and the reference point combination and the candidate point combination within the initial combination interval are evaluated by the judgment model to obtain the corresponding judgment value; the candidate point combination with a judgment value of 0 is merged with the initial combination interval, and so on, until there is no candidate point combination with a judgment value of 0, and the current initial combination interval is marked as a combination segment interval;

[0058] Re-determine the reference point combination, and so on, until there is no reference point combination, and obtain the segmented intervals of each combination.

[0059] In one embodiment, if the electronic nose is a new product of the enterprise, it will result in insufficient material analysis data, which may easily lead to omissions and missing items. Even if it is set up by professionals based on their experience, the above problems may still occur. If a large number of experimental simulations are carried out without guidance, it will result in a large number of experimental simulations, which will reduce efficiency and be difficult to implement in actual application. Based on this, the following method is used to determine the environmental impact factors for the new electronic nose, including:

[0060] Acquiring various electronic nose information such as the electronic nose's detection method, collection method, data processing method, and device components, and marking the electronic nose as a target electronic nose for differentiation; marking the corresponding electronic nose information as target electronic nose information;

[0061] Acquire currently available electronic noses that can clearly identify their environmental influencing factors, mark the corresponding electronic noses as reference electronic noses, mark the environmental influencing factors as reference factors, obtain electronic nose information of the reference electronic noses, mark it as reference electronic nose information, integrate the reference electronic nose information and the corresponding reference factors, and establish a reference library;

[0062] According to the simulation, a large number of electronic nose information comparison results are set, that is, whether they can be equivalently referenced to the corresponding environmental influencing factors, to form a large number of training sets, that is, according to their actual detection conditions, whether the two will be affected by the corresponding environment, and can also be evaluated in combination with the corresponding similarity. If it is higher than a certain preset value, it is considered to meet the equivalent conditions, otherwise it is not met and is abnormal data; based on the isolation forest algorithm, a corresponding equivalent analysis model is established, which will be considered as normal data if it is equivalent, otherwise it is considered as abnormal data. The input data is two electronic nose information, and the output data is the equivalent evaluation value 1 or 0; the expression is ; Where: s is the input data, i.e. the target electronic nose information and the reference electronic nose information; the output data is the equivalent evaluation value 1 or 0;

[0063] The target electronic nose information and the reference electronic nose information in the reference library are analyzed using an equivalent analysis model. The reference electronic noses with an equivalent evaluation value of 0 are marked as equivalent electronic noses. The reference factors of the equivalent electronic noses are integrated, which is equivalent to a union, to determine the initial factors of the target electronic nose.

[0064] Conduct simulation experiments on each initial factor to determine whether it is affected by the initial factor, determine the environmental influencing factors of the target electronic nose and the corresponding impact analysis data, that is, the relevant simulation experiment data.

[0065] In other embodiments, if the existing technology can realize the setting of the corresponding combined segment intervals and the corresponding error influence curve, the existing technology can also be selected and applied as needed.

[0066] By setting up a factor analysis module, intelligent analysis of the target electronic nose can be achieved, and the impact of various environmental factors on the target electronic nose can be determined, which facilitates subsequent detection adjustments based on actual environmental conditions and improves gas detection accuracy.

[0067] The detection control module is used to perform detection control, determine target electronic noses, i.e., electronic noses of the same type that need to be detected and analyzed, i.e., electronic noses of the same type that are being affected by work; obtain target environment data of each target electronic nose, i.e., environmental data corresponding to the gas to be detected; extract features of the target environment data according to each environmental influencing factor to obtain target environment features, i.e., environmental influencing factor data corresponding to each environmental influencing factor; match corresponding combined segmentation intervals according to each target environment feature, mark corresponding similarity marks on target electronic noses with the same combined segmentation intervals; and select a target electronic nose from each target electronic nose with each similar mark as a standard electronic nose;

[0068] Acquire the detection data of each target electronic nose in real time, that is, perform detection according to the original working mode and obtain the corresponding detection results. The detection data includes the collected data and the detection results;

[0069] A corresponding detection learning model is established based on a deep learning network, and a corresponding training set is manually established for training. The training set includes input data and output data. The input data is the detection data of the standard electronic nose, the standard detection results, and the detection data of other target electronic noses with the same type of labels. The output data is the detection correction data of the detection data of other target electronic noses. That is, the difference between the detection data of the standard electronic nose and the standard detection results is used to correct other detection data with the same influencing conditions, adjust the detection results in the detection data, and obtain detection correction data. Training is carried out through the established training set, and the intelligent model after successful training is marked as a detection learning model.

[0070] The detection data and standard detection results corresponding to the standard electronic nose are obtained; the standard detection results and the corresponding detection data are input into the detection learning model, and the detection data of the target electronic nose with the same type of mark as the standard electronic nose is corrected by the detection learning model to obtain the detection correction data of each target electronic nose.

[0071] Methods for obtaining standard test results include:

[0072] Obtain the corresponding test data, that is, the test data at the corresponding time of the corresponding analysis standard test result;

[0073] Identify the gases that may be present in the target environment and mark them as candidate gases; match the error influence curves corresponding to the candidate gases according to the combined segmented intervals;

[0074] Identify the electronic nose control parameter corresponding to the minimum error in each error influence curve, and mark the corresponding electronic nose control parameter as a single gas parameter;

[0075] Adjust the target electronic nose to a single gas parameter, perform gas detection, and obtain a single detection result for the selected gas; repeat the above steps to detect other selected gases and obtain a single detection result for each selected gas; identify the error value corresponding to each single detection result based on the error influence curve;

[0076] Each single test result is corrected according to the single test result and the error value to obtain a standard test result.

[0077] By setting up a detection control module, the detection control of the electronic nose can be realized and the detection accuracy of the electronic nose can be improved. At the same time, based on the detection situation of the standard electronic nose, dynamic correction can be achieved for each target electronic nose with the same label. On the premise of improving the detection accuracy, excessive replacement and modification of the existing electronic nose can be avoided, thereby improving resource utilization.

[0078] The method of selecting a target electronic nose as a standard electronic nose from each target electronic nose of the same type includes:

[0079] Acquire detection data from target electronic noses with the same or similar tags, and identify the gas type and gas concentration corresponding to each detection data; after sorting, obtain the various gas types present; set a gas statistical template based on the various gas types present, that is, a statistical template that includes each gas type, and calculate the corresponding gas concentration; if the target electronic nose does not include the gas type, its gas concentration is 0;

[0080] The detection data of each target electronic nose is counted using a gas statistics template to obtain gas statistics of each gas electronic nose;

[0081] The gas type is marked as i, i = 1, 2, ..., n, n is a positive integer; the target electronic nose is marked as j, j = 1, 2, ..., m, m is a positive integer; the gas concentration is marked as Cij;

[0082] According to the formula Calculate the gas evaluation value of the corresponding gas type in each target electronic nose;

[0083] Where: PAij is the gas assessment value;

[0084] According to the formula Calculate the priority value of each target electronic nose;

[0085] Where: PUYj is the priority value;

[0086] The target electronic nose with the largest priority value is selected as the standard electronic nose.

[0087] Methods for correcting each single test result according to each single test result and the error value include:

[0088] Obtain the influence of different types of gases on a single detection result when performing a single gas analysis, and then make adjustments based on the corresponding error value; specifically, conduct simulation experiments based on the above steps to determine several groups of training sets, the training sets include input data and output data, the input data includes each single detection result and error value, and the output data includes the detection standard result composed of each corrected single detection result; that is, simulate various detection situations to obtain each single detection result and error value, as well as accurate detection results under the set simulation background; establish a corresponding single analysis model based on the CNN network or DNN network, analyze the single analysis model through the established training set, and analyze the single analysis model after successful training to obtain the corresponding standard detection result. In other embodiments, other methods can also be applied for correction to obtain the corresponding standard detection result.

[0089] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0090] One embodiment of the present application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps and methods of any of the aforementioned electronic nose analysis and learning control systems. Persons skilled in the art will appreciate that all or part of the processes in the aforementioned method embodiments can be implemented by instructing the relevant hardware through the computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the aforementioned method embodiments. Any reference to memory, storage, database, or other medium provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), Direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM).

[0091] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0092] The above description is only a preferred embodiment of the present invention and does not limit the scope of the patent of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied to other related technical fields, is also included in the scope of patent protection of the present invention.

[0093] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

[0094] For ease of description, the above devices are described separately based on their functions and units. Of course, when implementing this application, the functions of each unit can be implemented in the same or multiple software and / or hardware components. Those skilled in the art will appreciate that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0095] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 The functionality specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or a function specified in multiple boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

Claims

1. An analysis and learning control system for an electronic nose, characterized in that: Including factor analysis module and detection and control module; The factor analysis module is used to analyze the target electronic nose, determine various environmental influencing factors, and analyze each of the environmental influencing factors to determine each combined segment interval corresponding to the electronic nose and the corresponding error influence curve; The detection control module is used to perform detection control, determine each target electronic nose, obtain target environment data of each target electronic nose, extract features of the target environment data according to various environmental influencing factors, obtain target environment features, match corresponding combined segment intervals according to each target environment feature, mark each target electronic nose with the same combined segment interval with a corresponding similar mark; and select one target electronic nose from each target electronic nose with each similar mark as a standard electronic nose; Acquire detection data of each target electronic nose in real time, the detection data including collected data and detection results; Establish a corresponding detection learning model based on the deep learning network; Obtaining a standard detection result of the standard electronic nose; inputting the standard detection result and corresponding detection data of the standard electronic nose into a detection learning model; and correcting the detection data of target electronic noses having the same type of markers as the standard electronic nose using the detection learning model to obtain correction data for detection of each target electronic nose; Methods for analyzing various environmental factors include: Obtain impact analysis data corresponding to each environmental influencing factor; The impact analysis data is analyzed to determine the factor intervals corresponding to the environmental impact factors, the factor intervals corresponding to the environmental impact factors are split and merged to obtain each combined segmented interval; and the corresponding error impact curve is generated based on each combined segmented interval.

2. The analysis and learning control system of an electronic nose according to claim 1, characterized in that: Methods for splitting and merging factor intervals of various environmental influencing factors include: Step SA1: Set floating points within each factor interval; Step SA2: setting a reference point combination according to each floating point, and determining a point combination to be selected according to the reference point combination; Step SA3: Evaluate the reference point combination and the candidate point combination using a preset judgment model to obtain corresponding judgment values; merge the candidate point combination with a judgment value of 0 with the reference point combination to obtain an initial combination interval; Step SA4: Determine a new candidate point combination based on the initial combination interval; evaluate the reference point combination and the candidate point combination within the initial combination interval using the judgment model to obtain a corresponding judgment value; merge the candidate point combination with a judgment value of 0 with the initial combination interval to obtain a new initial combination interval; Step SA5: looping step SA4 until there is no candidate point combination with a judgment value of 0, marking the current initial combination interval as a combination segment interval; Step SA6: Loop steps SA2 to SA5 until there is no reference point combination, and obtain the segmented intervals of each combination.

3. The analysis and learning control system of an electronic nose according to claim 2, characterized in that: The expression of the judgment model is ; Where: q is the input data, and the output data is the judgment value 1 or 0.

4. The analysis and learning control system of an electronic nose according to claim 2, characterized in that: Methods for determining environmental impact factors include: Establishing a reference library, wherein the reference library is used to store each reference electronic nose information and each reference factor corresponding to each reference electronic nose information; Obtain target electronic nose information, analyze the target electronic nose information and each reference electronic nose information in the reference library through a preset equivalent analysis model, and obtain equivalent evaluation values between the target electronic nose information and each reference electronic nose information. Equivalent electronic noses are determined based on equivalent evaluation values, reference factors corresponding to each equivalent electronic nose are obtained, and the obtained reference factors are integrated to obtain initial factors of the target electronic nose; simulation experiments are conducted on each initial factor to determine the environmental influencing factors of the target electronic nose.

5. The analysis and learning control system of an electronic nose according to claim 4, characterized in that: The expression of the equivalent analysis model is ; Where: s is the target electronic nose information and the reference electronic nose information; the output data is the equivalent evaluation value 1 or 0.

6. The analysis and learning control system of an electronic nose according to claim 1, characterized in that: Methods for obtaining standard test results include: Acquire detection data corresponding to a standard electronic nose and determine each candidate gas; match the error influence curve corresponding to each candidate gas according to the combined segmented interval corresponding to the standard electronic nose; obtain a single detection result for each candidate gas according to each error influence curve; identify the error value corresponding to each single detection result according to the error influence curve; correct each single detection result according to each single detection result and the error value to obtain a standard detection result.

7. The analysis and learning control system of an electronic nose according to claim 6, characterized in that: Methods for determining a standard electronic nose include: Acquire the detection data of each target electronic nose with the same or similar label, identify the gas type and gas concentration corresponding to each detection data; after sorting, obtain the various gas types, and set the gas statistics template according to the various gas types; The detection data of each target electronic nose is counted using a gas statistics template to obtain gas statistics of each gas electronic nose; The gas type is marked as i, i = 1, 2, ..., n, n is a positive integer; the target electronic nose is marked as j, j = 1, 2, ..., m, m is a positive integer; the gas concentration is marked as Cij; According to the formula Calculate the gas evaluation value of the corresponding gas type in each target electronic nose; Where: PAij is the gas assessment value; According to the formula Calculate the priority value of each target electronic nose; Where: PUYj is the priority value; The target electronic nose with the largest priority value is selected as the standard electronic nose.

Images