Control method, device and equipment of single-molecule conductivity characterization instrument and storage medium

Abstract

The invention provides a control method, device and equipment for a single molecule conductivity characterization instrument and a storage medium, and the method comprises the steps: obtaining time sequence data in a single molecule conductivity test process, converting the time sequence data into a multi-dimensional feature vector sequence, and enabling the time sequence data to be measurement data of conductivity changing with time; then, a pre-trained Angle Transform model is called to carry out modeling on the multi-dimensional feature vector sequence so as to capture the incidence relation between different positions in the time sequence data; then, calculating an association difference score of the time sequence data according to the association relationship, and judging whether the sequence is a normal sequence or not according to the association difference score; finally, when it is judged that the time sequence data is a normal sequence, the conductance curves of the time sequence data are stored, and when the number of the stored conductance curves reaches a preset value, a data result is output. The problem that the superiority of the curve cannot be automatically identified in the conductivity measurement process is solved.

Description

Technical Field

[0001] The present invention relates to the field of precision instruments, and in particular to a control method, device, equipment, and storage medium of a single-molecule conductivity characterization instrument. Background Art

[0002] Molecular electronics is an important frontier of interdisciplinary studies in recent years. It explores the potential of single molecules as functional elements, provides a "bottom-up" solution for the miniaturization of electronic devices, and also provides a unique research platform for studying physical and chemical processes such as assembly and reaction at the molecular level. At present, the main method for studying the electrical transport properties of single molecules is scanning tunneling microscopy junction (STM-BJ). However, the existing precision scientific instruments based on STM-BJ technology have a low degree of intelligence. The automated conductivity measuring instrument can only mechanically perform tests according to the preset test process, and relies on the number of saved files to achieve an automatic gear program closed loop. The intelligent identification, self-monitoring and regulation of excellent curves during instrument testing have not yet been realized, resulting in a large amount of invalid data being automatically saved during the test process, and the test efficiency is low.

[0003] In view of this, this application is filed. Summary of the invention

[0004] The invention discloses a control method, device, equipment and storage medium of a single molecule conductivity characterization instrument, aiming to solve the problem that the excellence of a curve cannot be automatically identified during the conductivity measurement process.

[0005] The first embodiment of the present invention provides a control method for a single molecule conductivity characterization instrument, comprising:

[0006] Acquire time series data in a single-molecule conductivity test process, and convert the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of conductivity changing with time;

[0007] Calling a pre-trained Anomaly Transformer model to model the multidimensional feature vector sequence to capture the correlation relationship between different positions in the multidimensional feature vector sequence;

[0008] Calculating the correlation difference score of the time series data according to the correlation relationship, and judging whether it is a normal sequence according to the correlation difference score;

[0009] When it is determined that the time series data is a normal sequence, the conductivity curve of the time series data is saved, and when the number of the saved conductivity curves reaches a preset value, the data result is output.

[0010] Preferably, converting the time series data into a multi-dimensional feature vector sequence is specifically as follows:

[0011] According to the preset active window length and step length, the statistical features in each window are calculated, wherein the statistical features include mean, standard deviation, maximum value, and minimum value;

[0012] Performing Fourier transform on the data in each window to extract spectrum energy and main frequency;

[0013] The statistical features, the spectrum energy and the main frequency are combined to generate a multi-dimensional feature vector sequence.

[0014] Preferably, the calling of the pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the time series data is specifically as follows:

[0015] Embedding the input multi-dimensional feature vector sequence into a high-dimensional space to form an embedding matrix, and combining the position encoding with the embedding matrix to generate modeling information;

[0016] The self-attention mechanism is used to calculate the similarity between the modeling information at each position and all other positions in the sequence to generate an association relationship.

[0017] Preferably, the training process of the Anomaly Transformer model is:

[0018] The normal and abnormal time series data are input into the Transformer encoder respectively to obtain the feature representation of the normal sequence and the feature representation of the abnormal sequence;

[0019] The prior association distribution and sequence association distribution of normal and abnormal sequences are calculated, and the association difference loss function is called to measure the difference between the prior association and sequence association of normal and abnormal sequences based on KL divergence. The Anomaly Transformer model is trained by minimizing the prior association and maximizing the sequence association to amplify the difference between normal and abnormal sequences.

[0020] A second embodiment of the present invention provides a control device for a single molecule conductivity characterization instrument, comprising:

[0021] A feature conversion unit, used to obtain time series data in the single-molecule conductivity test process, and convert the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of the conductivity changing with time;

[0022] A position association unit, used to call a pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the multi-dimensional feature vector sequence;

[0023] a difference judgment unit, configured to calculate a correlation difference score of the time series data according to the correlation relationship, and judge whether it is a normal sequence according to the correlation difference score;

[0024] The result output unit is used to save the conductivity curve of the time series data when it is determined that the time series data is a normal sequence, and output the data result when the number of saved conductivity curves reaches a preset value.

[0025] Preferably, converting the time series data into a multi-dimensional feature vector sequence is specifically as follows:

[0026] According to the preset active window length and step length, the statistical features in each window are calculated, wherein the statistical features include mean, standard deviation, maximum value, and minimum value;

[0027] Performing Fourier transform on the data in each window to extract spectrum energy and main frequency;

[0028] The statistical features, the spectrum energy and the main frequency are combined to generate a multi-dimensional feature vector sequence.

[0029] Preferably, the position association unit is specifically used for:

[0030] Embedding the input multi-dimensional feature vector sequence into a high-dimensional space to form an embedding matrix, and combining the position encoding with the embedding matrix to generate modeling information;

[0031] The self-attention mechanism is used to calculate the similarity between the modeling information at each position and all other positions in the sequence to generate an association relationship.

[0032] Preferably, the training process of the Anomaly Transformer model is:

[0033] The normal and abnormal time series data are input into the Transformer encoder respectively to obtain the feature representation of the normal sequence and the feature representation of the abnormal sequence;

[0034] The prior association distribution and sequence association distribution of normal and abnormal sequences are calculated, and the association difference loss function is called to measure the difference between the prior association and sequence association of normal and abnormal sequences based on KL divergence. The Anomaly Transformer model is trained by minimizing the prior association and maximizing the sequence association to amplify the difference between normal and abnormal sequences.

[0035] A third embodiment of the present invention provides a control device for a single-molecule conductivity characterization instrument, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a control method for a single-molecule conductivity characterization instrument as described in any one of the above.

[0036] A fourth embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a control method for a single-molecule conductivity characterization instrument as described in any one of the above items.

[0037] Based on the control method, device, equipment, and storage medium of a single-molecule conductivity characterization instrument provided by the present invention, firstly, the time series data in the single-molecule conductivity test process is obtained, and the time series data is converted into a multi-dimensional feature vector sequence, wherein the time series data is the measurement data of the conductivity changing with time; then, the pre-trained Anomaly Transformer model is called to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the multi-dimensional feature vector sequence; then, the association difference score of the time series data is calculated according to the association relationship, and whether it is a normal sequence is determined according to the association difference score; finally, when the time series data is judged to be a normal sequence, the conductivity curve of the time series data is saved, and when the number of saved conductivity curves reaches a preset value, the data result is output. The problem that the excellentness of the curve cannot be automatically identified during the conductivity measurement process is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of a control method of a single molecule conductivity characterization instrument provided by the first embodiment of the present invention;

[0039] Figure 2 It is a schematic diagram of the principle of the instrument hardware system provided by the present invention;

[0040] Figure 3 It is a schematic diagram of the structure of the Anomaly Transformer model provided by the present invention;

[0041] Figure 4 It is a schematic diagram of the test process of the conductivity experiment provided by the present invention;

[0042] Figure 5 It is a schematic diagram of the communication interaction process between the control program and the algorithm program provided by the present invention;

[0043] Figure 6 It is a module schematic diagram of a control device of a single-molecule conductivity characterization instrument provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0044] 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.

[0045] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0046] It should be clear that the described embodiments 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 creative work are within the scope of protection of the present invention.

[0047] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0048] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0049] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0050] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0051] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0052] The invention discloses a control method, device, equipment and storage medium of a single molecule conductivity characterization instrument, aiming to solve the problem that the excellence of a curve cannot be automatically identified during the conductivity measurement process.

[0053] See also Figure 1 The first embodiment of the present invention provides a control method for a single-molecule conductance characterization instrument, which can be executed by a control device of the single-molecule conductance characterization instrument (hereinafter referred to as the control device), and in particular, is executed by one or more processors in the control device to at least implement the following steps:

[0054] S101, acquiring time series data in a single-molecule conductivity test process, and converting the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of conductivity changing with time;

[0055] In this embodiment, the control device can be a terminal with data processing capabilities such as a desktop computer, a laptop computer, a server, a workstation, etc., which can establish communication with the single-molecule conductivity characterization instrument to receive the data collected by it. The control device can be installed with a corresponding operating system and application software, and the functions required by this embodiment are realized through the combination of the operating system and the application software.

[0056] Specifically, in this embodiment, the overall control principle of the instrument is as follows: Figure 2 The control system is divided into the PC control layer, the self-developed controller control layer and the Myr io FPGA control layer. The PC end exchanges data with Myr io and the self-developed controller through the USB interface, corresponding to Figure 2 Marks ① and ② in the figure. Marks ③, ④ and ⑤ respectively show the control process of the self-developed controller on the BJ motor driver input control signal, output bias value and output piezoelectric ceramic scanning voltage value. Mark ⑥ represents the acquisition of the needle tip input signal, ⑦ represents the Myri io output control signal to the linear module driver, and marks ⑧ and ⑨ show the Myri io FPGA realizing the signal acquisition of the photoelectric limiter and the beam sensor respectively. Marks ⑩ and The motor driver and linear module driver are shown to drive the BJ motor and linear module respectively.

[0057] It should be noted that time series data mainly refers to the measurement data collected during the single-molecule conductivity test on the change of conductivity over time. Among them, the measurement data records the change of single-molecule conductivity value over time, forming a time series. These data can be used to analyze the conductivity characteristics and identify the advantages and disadvantages of the conductivity curve. Specifically, time series data includes: (1) Conductance value: represents the change of the value of single-molecule conductivity over time. (2) Timestamp: The specific time point corresponding to each conductivity value, ensuring that the data is orderly and can accurately reflect the dynamic process of conductivity change.

[0058] In this embodiment, the time series data is converted into a multi-dimensional feature vector sequence, which may be:

[0059] Get a time series data to represent the change of the conductivity value of a single molecule over time, which can be expressed as a conductivity value sequence = {G1, G2, G3, G4, G5}. Based on the setting of the sliding window length to 5 and the step length to 1, the sliding window is divided into: {G1, G2, G3, G4, G5}, {G2, G3, G4, G5, G6}.....

[0060] Next, statistical features are calculated for the data in each window, which may include: mean, standard deviation, maximum value, and minimum value;

[0061] Mean: Standard Deviation: Maximum value: max{G1,G2,G3,G4,G5}, minimum value: min{G1,G2,G3,G4,G5}.

[0062] Next, the data in the window is Fourier transformed to extract the spectrum energy and main frequency components; each window corresponds to a feature vector

[0063] F1=[μ,σ,max,min,frequency energy,main frequency...]

[0064] Repeat the above process to obtain the feature vector sequence of the entire time series data. These feature vector sequences not only contain the time series information of the original conductance value, but also introduce statistical, time domain, frequency domain and other information, so that the model can more comprehensively understand and capture the characteristics of time series data, facilitating more accurate anomaly detection and pattern recognition.

[0065] S102, calling a pre-trained Anomaly Transformer model to model the multidimensional feature vector sequence to capture the correlation relationship between different positions in the multidimensional feature vector sequence;

[0066] Please combine Figure 3 , the training process of the Anomaly Transformer model is:

[0067] First, a set of time series data considered normal is input into the Transformer encoder to obtain its encoder output (feature representation), and a set of time series data considered abnormal is input into the Transformer encoder to obtain its encoder output (feature representation). For each time step, its attention weight matrix and output representation are calculated. Among them, the attention weight matrix captures the relationship between the current time step and other time steps; then, the mean and variance of the attention weight and output representation of the normal sequence are calculated, and the mean and variance of the attention weight and output representation of the abnormal sequence are calculated. Next, a loss function is defined to measure the correlation difference between normal and abnormal sequences. The Euclidean distance or cosine similarity can be used to compare their means and variances; the expression of the loss function can be:

[0068]

[0069] Among them, μ is the mean, σ is the variance, and i is each feature dimension of the time series data.

[0070] Finally, the model can be trained by minimizing the loss of normal sequences and maximizing the loss of abnormal sequences. In actual anomaly detection, if the score is high, it means that the association pattern of the sequence to be detected is significantly different from the normal pattern, indicating that the sequence may be abnormal. If the score is low, it means that the association pattern of the sequence to be detected is similar to the normal pattern, indicating that the sequence is normal.

[0071] Furthermore, in this embodiment, the feature vector sequence can be first embedded into a high-dimensional space to form an input embedding matrix. Since the Transformer model itself does not contain sequence information, position encoding needs to be added to retain the time order. The position encoding vector is added to the input embedding so that the model can perceive the order of the sequence. The self-attention mechanism is the core of the Transformer. For the input embedding of each position, the association between different positions is captured by calculating its similarity (attention weight) with all other positions in the sequence. More specifically, the input embedding is linearly transformed into a query, key, and value matrix. The similarity between the query and the key is calculated by dot product, and then the value is weighted and summed to obtain the output of each position. Multiple self-attention heads are calculated in parallel to capture the associated information in different subspaces, and finally the results are spliced ​​and then linearly transformed. The attention output is nonlinearly transformed through a feedforward neural network. This process consists of multiple layers of stacked Transformer encoders, each of which includes a self-attention mechanism and a feedforward neural network.

[0072] Through the self-attention mechanism, the Transformer model can capture the associations between different positions in the sequence, including: Global association: The self-attention mechanism allows the model to pay attention to every position in the entire sequence, so it can capture long-distance dependencies. For example, the model can find the association between the change in conductance at a certain point in time and the change at a previous or later time point. Local association: The multi-head attention mechanism enables the model to capture local features in different subspaces and identify associations within a short distance.

[0073] S103, calculating the correlation difference score of the time series data according to the correlation relationship, and judging whether it is a normal sequence according to the correlation difference score;

[0074] In this embodiment, during the anomaly detection process, the correlation difference loss function is used to compare the similarities and differences between normal and abnormal sequences on the encoder output. The correlation difference loss function measures the correlation difference between normal and abnormal time series, and the model learns to distinguish them by minimizing the loss of normal sequences and maximizing the loss of abnormal sequences.

[0075] The correlation difference score can be defined as the distance measure between the normal pattern and the sequence to be detected:

[0076]

[0077] If the score is high, it means that the correlation pattern of the sequence to be detected is significantly different from the normal pattern, indicating that the sequence may be abnormal. If the score is low, it means that the correlation pattern of the sequence to be detected is similar to the normal pattern, indicating that the sequence is normal.

[0078] S104, when it is determined that the time series data is a normal sequence, the conductance curve of the time series data is saved, and when the number of the saved conductance curves reaches a preset value, the data result is output.

[0079] Please combine Figure 4 When the number of good conductivity curves reaches the target setting value, the instrument will automatically change the sample and output the data results, otherwise the instrument will continue to test.

[0080] In a possible embodiment of the present invention, please combine Figure 5 ,This embodiment uses LabVIEW programming language in instrument control, and Python programming language in intelligent recognition algorithm, so it is necessary to establish an interactive bridge between the two languages. To this end, this embodiment proposes two data interaction solutions.

[0081] Solution A: Indirect interaction based on tdms files

[0082] First, the instrument control software generates data through LabVIEW and stores it in tdms format. This format has efficient read and write performance and good compatibility. Subsequently, the Python script reads these tdms files in real time for data processing and analysis. The intelligent recognition algorithm processes the conductivity curve in real time and counts qualified conductivity curves. Then, the Python script is called through LabVIEW to transmit the processed count values ​​back to the instrument control algorithm process in real time. This feedback mechanism ensures that the instrument can perform automatic closed-loop control based on real-time data, improving the response speed and accuracy of the system.

[0083] Solution b: Call the Python script directly

[0084] LabVIEW directly calls Python scripts for data processing. This method simplifies the data transmission process and improves real-time performance. LabVIEW uses its built-in Python script calling function to achieve real-time recognition of the conductivity curve. After the recognition process is completed, the Python script will directly feed back the curve count value to LabVIEW, thereby realizing automated closed-loop control of the instrument. This method reduces the intermediate steps of data storage and reading, making the system more efficient and faster.

[0085] In order to realize the above two solutions, this embodiment focuses on solving the data interaction problem between LabVIEW and Python. In the first solution, tdms files are used as the medium for data transmission to ensure efficient data transmission and processing. In the second solution, the external script call function of LabVIEW is used to directly run the Python script to realize real-time data processing and feedback.

[0086] Through these two solutions, this embodiment realizes the efficient collaboration between LabVIEW and Python, fully utilizing the advantages of LabVIEW in instrument control and the powerful capabilities of Python in data processing and intelligent recognition algorithms, thereby realizing the intelligent and automatic control of the instrument and improving the overall performance and reliability of the system.

[0087] Please combine Figure 6 The second embodiment of the present invention provides a control device for a single molecule conductance characterization instrument, comprising:

[0088] A feature conversion unit 201 is used to obtain time series data in the single molecule conductivity test process and convert the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of the conductivity changing with time;

[0089] A position association unit 202, configured to call a pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the time series data;

[0090] A difference judgment unit 203 is used to calculate the correlation difference score of the time series data according to the correlation relationship, and judge whether it is a normal sequence according to the correlation difference score;

[0091] The result output unit 204 is used to save the conductivity curve of the time series data when it is determined that the time series data is a normal sequence, and output the data result when the number of saved conductivity curves reaches a preset value.

[0092] Preferably, converting the time series data into a multi-dimensional feature vector sequence is specifically as follows:

[0093] According to the preset active window length and step length, the statistical features in each window are calculated, wherein the statistical features include mean, standard deviation, maximum value, and minimum value;

[0094] Performing Fourier transform on the data in each window to extract spectrum energy and main frequency;

[0095] The statistical features, the spectrum energy and the main frequency are combined to generate a multi-dimensional feature vector sequence.

[0096] Preferably, the position association unit is specifically used for:

[0097] Embedding the input multi-dimensional feature vector sequence into a high-dimensional space to form an embedding matrix, and combining the position encoding with the embedding matrix to generate modeling information;

[0098] The self-attention mechanism is used to calculate the similarity between the modeling information at each position and all other positions in the sequence to generate an association relationship.

[0099] Preferably, the training process of the Anomaly Transformer model is:

[0100] The normal and abnormal time series data are input into the Transformer encoder respectively to obtain the feature representation of the normal sequence and the feature representation of the abnormal sequence;

[0101] The prior association distribution and sequence association distribution of normal and abnormal sequences are calculated, and the association difference loss function is called to measure the difference between the prior association and sequence association of normal and abnormal sequences based on KL divergence. The Anomaly Transformer model is trained by minimizing the prior association and maximizing the sequence association to amplify the difference between normal and abnormal sequences.

[0102] A third embodiment of the present invention provides a control device for a single-molecule conductivity characterization instrument, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a control method for a single-molecule conductivity characterization instrument as described in any one of the above.

[0103] A fourth embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a control method for a single-molecule conductivity characterization instrument as described in any one of the above items.

[0104] Based on the control method, device, equipment, and storage medium of a single-molecule conductivity characterization instrument provided by the present invention, firstly, the time series data in the single-molecule conductivity test process is obtained, and the time series data is converted into a multi-dimensional feature vector sequence, wherein the time series data is the measurement data of the conductivity changing with time; then, the pre-trained Anomaly Transformer model is called to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the time series data; then, the association difference score of the time series data is calculated according to the association relationship, and whether it is a normal sequence is determined according to the association difference score; finally, when the time series data is judged to be a normal sequence, the conductivity curve of the time series data is saved, and when the number of saved conductivity curves reaches a preset value, the data result is output. The problem that the excellentness of the curve cannot be automatically identified during the conductivity measurement process is solved.

[0105] Exemplarily, the computer program described in the third and fourth embodiments of the present invention can be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the control device of the single-molecule conductivity characterization instrument. For example, the device described in the second embodiment of the present invention.

[0106] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the control method of the single-molecule conductivity characterization instrument, and uses various interfaces and lines to connect the various parts of the control method for implementing the single-molecule conductivity characterization instrument.

[0107] The memory can be used to store the computer program and / or module, and the processor realizes various functions of a control method of a single molecule conductivity characterization instrument by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0108] Wherein, if the implemented module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0109] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0110] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A control method for a single molecule conductivity characterization instrument, characterized in that: include: Acquire time series data in a single-molecule conductivity test process, and convert the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of conductivity changing with time; Calling a pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the correlation relationship between different positions in the time series data; Calculating the correlation difference score of the time series data according to the correlation relationship, and judging whether it is a normal sequence according to the correlation difference score; When it is determined that the time series data is a normal sequence, the conductivity curve of the time series data is saved, and when the number of the saved conductivity curves reaches a preset value, the data result is output.

2. The control method of a single molecule conductivity characterization instrument according to claim 1, characterized in that: The specific steps of converting the time series data into a multi-dimensional feature vector sequence are as follows: According to the preset active window length and step length, the statistical features in each window are calculated, wherein the statistical features include mean, standard deviation, maximum value, and minimum value; Performing Fourier transform on the data in each window to extract spectrum energy and main frequency; The statistical features, the spectrum energy and the main frequency are combined to generate a multi-dimensional feature vector sequence.

3. The control method of a single molecule conductivity characterization instrument according to claim 1, characterized in that: The calling of the pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the time series data is specifically: Embedding the input multi-dimensional feature vector sequence into a high-dimensional space to form an embedding matrix, and combining the position encoding with the embedding matrix to generate modeling information; The self-attention mechanism is used to calculate the similarity between the modeling information at each position and all other positions in the sequence to generate an association relationship.

4. The control method of a single molecule conductivity characterization instrument according to claim 1, characterized in that: The training process of the Anomaly Transformer model is as follows: Input normal and abnormal time series data into the Transformer encoder respectively to obtain the feature representation of normal sequence and the feature representation of abnormal sequence; The prior association distribution and sequence association distribution of normal and abnormal sequences are calculated, and the association difference loss function is called to measure the difference between the prior association and sequence association of normal and abnormal sequences based on KL divergence. The Anomaly Transformer model is trained by minimizing the prior association and maximizing the sequence association to amplify the difference between normal and abnormal sequences.

5. A control device for a single-molecule conductivity characterization instrument, characterized in that: include: A feature conversion unit, used to obtain time series data in the single-molecule conductivity test process, and convert the time series data into a multi-dimensional feature vector sequence, wherein the time series data is measurement data of the conductivity changing with time; A position association unit, used to call a pre-trained Anomaly Transformer model to model the multi-dimensional feature vector sequence to capture the association relationship between different positions in the time series data; a difference judgment unit, configured to calculate a correlation difference score of the time series data according to the correlation relationship, and judge whether it is a normal sequence according to the correlation difference score; The result output unit is used to save the conductivity curve of the time series data when it is determined that the time series data is a normal sequence, and output the data result when the number of saved conductivity curves reaches a preset value.

6. The control device of a single molecule conductivity characterization instrument according to claim 5, characterized in that: The specific steps of converting the time series data into a multi-dimensional feature vector sequence are as follows: According to the preset active window length and step length, the statistical features in each window are calculated, wherein the statistical features include mean, standard deviation, maximum value, and minimum value; Performing Fourier transform on the data in each window to extract spectrum energy and main frequency; The statistical features, the spectrum energy and the main frequency are combined to generate a multi-dimensional feature vector sequence.

7. The control device of a single molecule conductivity characterization instrument according to claim 5, characterized in that: The position association unit is specifically used for: Embedding the input multi-dimensional feature vector sequence into a high-dimensional space to form an embedding matrix, and combining the position encoding with the embedding matrix to generate modeling information; The self-attention mechanism is used to calculate the similarity between the modeling information at each position and all other positions in the sequence to generate an association relationship.

8. The control device of a single molecule conductivity characterization instrument according to claim 5, characterized in that: The training process of the Anomaly Transformer model is as follows: Input normal and abnormal time series data into the Transformer encoder respectively to obtain the feature representation of normal sequence and the feature representation of abnormal sequence; The prior association distribution and sequence association distribution of normal and abnormal sequences are calculated, and the association difference loss function is called to measure the difference between the prior association and sequence association of normal and abnormal sequences based on KL divergence. The Anomaly Transformer model is trained by minimizing the prior association and maximizing the sequence association to amplify the difference between normal and abnormal sequences.

9. A control device for a single-molecule conductivity characterization instrument, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a control method for a single-molecule conductivity characterization instrument as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a control method for a single-molecule conductivity characterization instrument as described in any one of claims 1 to 4.

Images