A current state monitoring method and monitor for scientific research instruments

Abstract

The present invention discloses a current state monitoring method and monitor for scientific research instruments, and relates to the technical field of measuring electrical variables. The current state monitoring method for scientific research instruments includes: constructing at least one historical instrument use sequence according to the historical working time of each scientific research instrument; screening at least one reference instrument use sequence matching the current instrument use sequence from each historical instrument use sequence according to the current instrument use sequence corresponding to each scientific research instrument; determining the predicted operating condition of the scientific research instrument according to the sequence similarity between the reference instrument use sequence and the current instrument use sequence; inputting the predicted operating condition into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument; comparing the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument, and the actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system.

Description

Technical Field

[0001] The present invention relates to the technical field of measuring electric variables, and in particular to a current state monitoring method and a monitor for scientific research instruments. Background Art

[0002] In scientific research, various scientific research instruments are often used. In order to ensure the repeatability of the experimental process, it is necessary to ensure the stable operation of various scientific research instruments. However, during the operation of scientific research instruments, the current state may be abnormal due to various reasons, which may invalidate the experiment, damage the instrument, or even cause a safety accident. Therefore, it is necessary to monitor the current state of scientific research instruments reasonably.

[0003] In the existing method, a monitoring system is usually used to monitor the scientific research instrument in real time. When the current state of the scientific research instrument exceeds a preset threshold, an alarm will be generated.

[0004] However, the existing method relies on a fixed preset threshold to monitor the current state, which only considers the current change and rated operating power of the scientific research instrument itself. It does not take into account the different current states caused by the different operating conditions of the scientific research instrument in different experimental processes, resulting in low monitoring accuracy. Summary of the invention

[0005] The embodiment of the present invention provides a current status monitoring method and a monitor for a scientific research instrument, which can improve the monitoring accuracy of the current status of the scientific research instrument.

[0006] In one aspect of an embodiment of the present invention, a method for monitoring the current state of a scientific research instrument is provided, comprising:

[0007] According to the historical working time of each scientific research instrument in the laboratory, at least one historical instrument usage sequence of the laboratory is constructed, and the historical instrument usage sequence is used to characterize the working order of each scientific research instrument in the historical time period;

[0008] According to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory, at least one reference instrument usage sequence matching the current instrument usage sequence is obtained from each historical instrument usage sequence;

[0009] Determine the predicted operating conditions of the scientific research instrument based on the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence;

[0010] The predicted operating condition is input into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument, where the current state model is a model trained based on the historical operating condition of the scientific research instrument and the corresponding historical current state;

[0011] The predicted current state is compared with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument. The actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the laboratory's intelligent sensing system.

[0012] In one aspect of an embodiment of the present invention, a current state monitor for a scientific research instrument is provided, comprising:

[0013] A sequence construction module is used to construct at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory, and the historical instrument usage sequence is used to characterize the working order of each scientific research instrument in the historical time period;

[0014] A sequence screening module is used to screen at least one reference instrument usage sequence that matches the current instrument usage sequence from each historical instrument usage sequence according to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory;

[0015] A working condition determination module, used to determine the predicted operating condition of the scientific research instrument based on the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence;

[0016] A current prediction module is used to input the predicted operating conditions into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument. The current state model is a model trained based on the historical operating conditions of the scientific research instrument and the corresponding historical current state;

[0017] The current comparison module is used to compare the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument. The actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the laboratory's intelligent sensing system.

[0018] In the current state monitoring method of scientific research instruments provided by the embodiment of the present invention, a current state model is obtained by training in advance according to the historical operating conditions of the scientific research instruments and the corresponding historical current states. Then, according to the historical instrument usage sequences of the laboratory and the current instrument usage sequences corresponding to the scientific research instruments, the predicted operating conditions corresponding to the scientific research instruments are estimated. The predicted current state of the scientific research instruments can be accurately predicted by simply inputting the predicted operating conditions into the current state model corresponding to the scientific research instruments. At the same time, the actual current state of the scientific research instruments is collected according to the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system of the laboratory. Finally, the actual current state is compared with the predicted current state to accurately obtain the current monitoring results of the scientific research instruments. In this way, the embodiment of the present invention accurately collects the actual current state of the scientific research instruments through the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system, and accurately determines the predicted current state corresponding to the predicted operating conditions of the scientific research instruments according to the current state model. The different current states corresponding to the different operating conditions of the scientific research instruments are fully taken into account, so as to improve the monitoring accuracy of the current state of the scientific research instruments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0020] Figure 1 A schematic flow chart of a first method for monitoring the current state of a scientific research instrument provided by an embodiment of the present invention;

[0021] Figure 2 A schematic flow chart of a second method for monitoring the current state of a scientific research instrument provided by an embodiment of the present invention;

[0022] Figure 3 A schematic flow chart of a third method for monitoring the current state of a scientific research instrument provided by an embodiment of the present invention;

[0023] Figure 4 A schematic structural diagram of a current status monitor for a scientific research instrument provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the current state monitoring method and monitor of a scientific research instrument proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0025] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0026] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of laws and regulations.

[0027] It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present invention, but it does not mean that the applicant has or will necessarily use the solution.

[0028] In the existing methods, a monitoring system is usually used to monitor scientific research instruments in real time. When the current state of the scientific research instrument exceeds the preset threshold, an alarm will be triggered. However, the existing method relies on a fixed preset threshold for current state monitoring, which only considers the current change and rated operating power of the scientific research instrument itself. It does not take into account the different current states caused by the different operating conditions of the scientific research instrument in different experimental processes, resulting in low monitoring accuracy.

[0029] The object of the present invention is to provide a method and monitor for monitoring the current state of a scientific research instrument. In the method for monitoring the current state of a scientific research instrument provided in an embodiment of the present invention, a current state model is obtained by training in advance according to the historical operating conditions of the scientific research instrument and the corresponding historical current state. Then, according to the historical instrument usage sequences of the laboratory and the current instrument usage sequences corresponding to the scientific research instrument, the predicted operating conditions corresponding to the scientific research instrument are estimated. The predicted current state of the scientific research instrument can be accurately predicted by simply inputting the predicted operating conditions into the current state model corresponding to the scientific research instrument. At the same time, the actual current state of the scientific research instrument is collected according to the intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system of the laboratory. Finally, the actual current state is compared with the predicted current state to accurately obtain the current monitoring result of the scientific research instrument. In this way, the embodiment of the present invention accurately collects the actual current state of the scientific research instrument through the intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system, and accurately determines the predicted current state corresponding to the predicted operating condition of the scientific research instrument according to the current state model. The different current states corresponding to different operating conditions of the scientific research instrument are fully taken into account, so that the monitoring accuracy of the current state of the scientific research instrument can be improved.

[0030] The following describes a specific embodiment of a current status monitoring method and monitor for a scientific research instrument provided by an embodiment of the present invention.

[0031] The following first introduces a current status monitoring method for a scientific research instrument provided by an embodiment of the present invention.

[0032] Figure 1 A flow chart of a method for monitoring the current status of a scientific research instrument is provided. The method for monitoring the current status of a scientific research instrument can be applied to a monitor. The method for monitoring the current status of a scientific research instrument can include the following S101 to S105.

[0033] S101, construct at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory, where the historical instrument usage sequence is used to characterize the working order of each scientific research instrument in the historical time period.

[0034] In this embodiment, a plurality of scientific research instruments are provided in the laboratory, such as a spectrometer, a mass spectrometer, and a centrifuge.

[0035] The historical instrument usage sequence is used to characterize the usage order of various scientific research instruments in the historical scientific research process of the target scientific research project, that is, which scientific research instrument was used first and which scientific research instrument was used later. In the same scientific research project, the usage order of scientific research instruments is fixed.

[0036] As an example, the monitor collects the historical working time data of each scientific research instrument, and then constructs the historical instrument usage sequence corresponding to the target scientific research project based on the collected historical working time data. For example, the historical instrument usage sequence can be expressed as [mass spectrometer (start time, end time), spectrometer (start time, end time), centrifuge (start time, end time)].

[0037] S102: According to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory, at least one reference instrument usage sequence matching the current instrument usage sequence is screened from each historical instrument usage sequence.

[0038] In this embodiment, the current instrument usage sequence is used to characterize the usage order between various scientific research instruments in the current scientific research process of the target scientific research project.

[0039] As an example, the monitor records the usage of each scientific research instrument in the current time period in real time and constructs the current instrument usage sequence. Then, the current instrument usage sequence is compared with each historical instrument usage sequence to find a reference instrument usage sequence in the historical instrument usage sequence that matches the current instrument usage sequence.

[0040] Among them, the matching condition can be preset as the usage order of each scientific research instrument in the instrument usage sequence, and each historical instrument usage sequence is screened through the matching condition to screen out at least one reference instrument usage sequence that is the same as the usage order of each scientific research instrument in the current instrument usage sequence.

[0041] S103, determining the predicted operating condition of the scientific research instrument according to the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence.

[0042] In this embodiment, sequence similarity is used to characterize the possibility that the sequence used by the reference instrument can represent the sequence used by the current instrument.

[0043] The predicted operating conditions are used to characterize the estimated operating conditions of scientific research instruments.

[0044] As an example, the monitor uses a sequence similarity algorithm (eg, dynamic time warping DTW, longest common subsequence LCS, etc.) to calculate the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence.

[0045] Then, the reference instrument usage sequence with the highest sequence similarity is selected as the final reference instrument usage sequence corresponding to each scientific research instrument, and the operating conditions of each scientific research instrument in the final reference instrument usage sequence are determined as the predicted operating conditions of each scientific research instrument in the current scientific research process.

[0046] S104, input the predicted operating conditions into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument. The current state model is a model trained based on the historical operating conditions and corresponding historical current states of the scientific research instrument.

[0047] In this embodiment, the monitor uses machine learning or deep learning algorithms (such as support vector machines SVM, neural networks NN, etc.) to train the current state model based on the historical operating conditions of the scientific research instrument and the corresponding historical current states. The input of the current state model is the predicted operating condition, and the output of the current state model is the predicted current state.

[0048] After obtaining the predicted operating conditions of the scientific research instrument, the monitor only needs to input the determined predicted operating conditions into the trained current state model to obtain the predicted current state of the scientific research instrument.

[0049] S105, comparing the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument. The actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system of the laboratory.

[0050] In this embodiment, the monitor collects the actual current state of the scientific research instrument in real time through the intelligent sensor in the laboratory's intelligent sensing system.

[0051] Then, the predicted current state is compared with the actual current state to calculate the current deviation. If the current deviation is greater than the preset deviation threshold, the current monitoring result indicates that the scientific research instrument is abnormal, and an alarm is issued to notify the researcher to handle it.

[0052] In the current state monitoring method of scientific research instruments provided in this embodiment, a current state model is obtained by training in advance based on the historical operating conditions of the scientific research instruments and the corresponding historical current states. Then, the predicted operating conditions corresponding to the scientific research instruments are estimated according to the historical instrument usage sequences of the laboratory and the current instrument usage sequences corresponding to the scientific research instruments. The predicted current state of the scientific research instruments can be accurately predicted by simply inputting the predicted operating conditions into the current state model corresponding to the scientific research instruments. At the same time, the actual current state of the scientific research instruments is collected according to the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system of the laboratory. Finally, the actual current state is compared with the predicted current state to accurately obtain the current monitoring results of the scientific research instruments. In this way, the embodiment of the present invention accurately collects the actual current state of the scientific research instruments through the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system, and accurately determines the predicted current state corresponding to the predicted operating conditions of the scientific research instruments according to the current state model. The different current states corresponding to the different operating conditions of the scientific research instruments are fully taken into account, so as to improve the monitoring accuracy of the current state of the scientific research instruments.

[0053] As an optional embodiment, Figure 2 As shown, S101 may specifically include the following S201-S205:

[0054] S201, calculating the usage time difference between the first scientific research instrument and the second scientific research instrument according to the historical working time of the first scientific research instrument and the second scientific research instrument, where the first scientific research instrument and the second scientific research instrument are any one of the scientific research instruments;

[0055] S202, determining a linkability evaluation value of the first scientific research instrument and the second scientific research instrument according to the usage time difference between the first scientific research instrument and the second scientific research instrument, wherein the linkability evaluation value is used to characterize the possibility of linking the first scientific research instrument and the second scientific research instrument for use;

[0056] S203, when the linkability evaluation value is greater than a preset evaluation threshold, determining that the first scientific research instrument and the second scientific research instrument are in a link relationship;

[0057] S204, looping and executing, according to the historical working time of the first scientific research instrument and the second scientific research instrument, calculating the usage time difference between the first scientific research instrument and the second scientific research instrument, until a preset stop condition is met, and obtaining a plurality of link relationships;

[0058] S205: Construct at least one historical instrument usage sequence of the laboratory according to each link relationship.

[0059] In this embodiment, the historical working time is used to characterize the working time of the scientific research instrument in the historical time period, including the start working time and the end working time. For example, the scientific research instrument A starts to be used at 16:00 on the previous day and ends at 18:00 on the previous day, that is, 16:00-18:00 on the previous day is the historical working time of the scientific research instrument A.

[0060] The linkability evaluation value is used to characterize the possibility that two scientific research instruments can be linked for use. When the linkability evaluation values ​​of scientific research instrument A and scientific research instrument B are greater than the preset evaluation threshold, it means that scientific research instrument A and scientific research instrument B are two devices used in succession. Among them, the one that started working earlier in the historical working time is used first, and the one that started working later in the historical working time is used later.

[0061] As an example, the monitor first collects the historical working time data of all scientific research instruments in the laboratory, and then calculates the usage time difference between any two scientific research instruments (respectively recorded as the first scientific research instrument and the second scientific research instrument) in the same historical time period.

[0062] Among them, scientific research instruments may be used multiple times in the historical period. In this case, the usage time difference can be determined by the following formula 1:

[0063] Δt α，β，i =min(t β，j,0 -t α,i,e ) Formula 1

[0064] In the formula, Δt α,β,i represents the time difference between the i-th consecutive use of scientific research instrument α and scientific research instrument β, t α,i,e represents the time when the scientific research instrument α is used for the i-th time, t β,j,0 It indicates the time when the jth use of the scientific research instrument β ends, and min indicates the minimum value.

[0065] Among them, min(t β，j,0 -t α,i,e ) means that the minimum value is selected from multiple time differences between the scientific research instrument α and the scientific research instrument β as the usage time difference between the scientific research instrument α and the scientific research instrument β.

[0066] The greater the time difference between the use of the first scientific research instrument and the second scientific research instrument, the less likely it is that the first scientific research instrument and the second scientific research instrument are used successively; the smaller the time difference between the use of the first scientific research instrument and the second scientific research instrument, the greater the likelihood that the first scientific research instrument and the second scientific research instrument are used successively.

[0067] Therefore, the monitor sets the linkability evaluation value of the first scientific research instrument and the second scientific research instrument according to the size relationship of the use time difference. For example, the reciprocal of the use time difference can be used as the linkability evaluation value.

[0068] At the same time, a preset evaluation threshold is set to determine whether the two instruments form a link relationship. If the linkability evaluation value of the first scientific research instrument and the second scientific research instrument is greater than the preset evaluation threshold, it is determined that there is a link relationship between them.

[0069] Then, any two other scientific research instruments in the laboratory are combined to obtain a scientific research instrument pair. The above steps are repeated for all scientific research instrument pairs in the laboratory until a preset stop condition is met (for example, a certain number of iterations is reached or all scientific research instrument pairs are traversed), thereby obtaining the link relationship between multiple scientific research instruments in the laboratory.

[0070] Finally, a sequence construction algorithm (such as greedy algorithm, dynamic programming, etc.) is used to arrange the scientific research instruments with link relationships in chronological order to form a series of historical instrument usage sequences.

[0071] Through this embodiment, at least one historical instrument usage sequence can be constructed based on the historical working time of each scientific research instrument in the laboratory. This helps to determine the predicted operating conditions of the current scientific research instrument based on the historical instrument usage sequence. The different current states corresponding to different operating conditions of the scientific research instrument are fully considered, thereby improving the monitoring accuracy of the current state of the scientific research instrument.

[0072] As an optional embodiment, S202 may specifically include:

[0073] According to the usage counts of the first scientific research instrument and the second scientific research instrument, a difference between the usage counts of the first scientific research instrument and the second scientific research instrument is obtained;

[0074] The difference in the number of uses is multiplied by the standard deviation of the difference in the use time between the first scientific research instrument and the second scientific research instrument to obtain an initial evaluation value;

[0075] The initial evaluation value is inversely normalized to obtain linkable evaluation values ​​of the first scientific research instrument and the second scientific research instrument.

[0076] In this embodiment, the linkability evaluation value can be specifically determined by the following formula 2:

[0077] L α,β =exp[-|ΔN α，β ×σ i (Δt α,β,i )|] Formula 2

[0078] Where, L α,β represents the linkable evaluation value of scientific research instrument α and scientific research instrument β, Δt α,β,i represents the time difference between the i-th consecutive use of scientific research instrument α and scientific research instrument β, σ i Indicates the calculation of standard deviation, exp[―||] indicates the inverse normalization calculation, ΔN α,β It represents the difference in the number of times scientific research instrument α and scientific research instrument β are used.

[0079] Among them, ΔN α,β =|N α -N β |, N α That is, the number of times the scientific research instrument α is used, N β This is the number of times scientific research instrument β is used. The difference between the number of times scientific research instrument α and scientific research instrument β are used is ΔN α,β The larger the value, or the time difference Δt between the i-th consecutive use of scientific research instrument α and scientific research instrument β α,β,i The larger the value, the less likely it is that the scientific research instrument α and the scientific research instrument β will be used in succession. That is, the linkable evaluation value L of the scientific research instrument α and the scientific research instrument β is α,β The smaller.

[0080] The linkable evaluation value L of scientific research instrument α and scientific research instrument β α,β The smaller it is, the lower the possibility that scientific research instrument α and scientific research instrument β can be linked and used.

[0081] Through this embodiment, according to the number of times the first scientific research instrument and the second scientific research instrument are used, and the difference in the time of use between the first scientific research instrument and the second scientific research instrument, the linkable evaluation value between the first scientific research instrument and the second scientific research instrument can be accurately evaluated, thereby facilitating the subsequent accurate construction of the historical instrument use sequence according to the linkable evaluation value.

[0082] As an optional embodiment, S102 may specifically include:

[0083] Screening each historical instrument usage sequence to obtain a candidate instrument usage sequence including a target scientific research instrument, where the target scientific research instrument is any one of the scientific research instruments;

[0084] Splitting each candidate instrument usage sequence according to the target scientific research instrument to obtain each candidate instrument usage subsequence containing the target scientific research instrument;

[0085] Calculate the temporal similarity between the current instrument usage sequence corresponding to the target scientific research instrument and each candidate instrument usage subsequence;

[0086] Each candidate instrument usage subsequence whose time sequence similarity is greater than or equal to a preset similarity threshold is determined as a reference instrument usage sequence of the current instrument usage sequence.

[0087] In this embodiment, the target scientific research instrument is used to characterize the scientific research instrument to be monitored, and the candidate instrument usage sequence is used to characterize the historical instrument usage sequence including the target scientific research instrument.

[0088] Time series similarity is used to characterize the degree of similarity between two time series under a preset metric. As an example, time series similarity can be determined by a dynamic time warping (DTW) algorithm.

[0089] As an example, the monitor screens the obtained historical instrument usage sequences, removes the historical instrument usage sequences that do not contain the target scientific research instrument, and obtains the candidate instrument usage sequences. Then, each candidate instrument usage sequence is split after the location of the target scientific research instrument to obtain the candidate instrument usage subsequences containing the target scientific research instrument corresponding to each candidate instrument usage sequence.

[0090] Then, the monitor determines the temporal similarity between the current instrument usage sequence corresponding to the target scientific research instrument and each candidate instrument usage subsequence through the following formula 3:

[0091] T δ，ε =DTW([δ],[ε]) Formula 3

[0092] In the formula, [δ] is used to represent the δth candidate instrument usage subsequence, [ε] is used to represent the εth current instrument usage sequence, DTW([],[]) is used to represent the DTW similarity between two sequences, and T δ,ε It is used to characterize the temporal similarity between the δth candidate instrument usage subsequence and the εth current instrument usage sequence.

[0093] Then, each time sequence similarity is compared with a preset similarity threshold, wherein the preset similarity threshold can be preset to 0.7. Then, each candidate instrument usage subsequence corresponding to a time sequence similarity greater than or equal to the preset similarity threshold is determined as a reference instrument usage sequence of the current instrument usage sequence.

[0094] Through this embodiment, after obtaining the historical instrument usage sequence, a reference instrument usage sequence matching the current instrument usage sequence is further screened. In this way, the predicted operating conditions of the scientific research instrument only need to be analyzed based on the reference instrument usage sequence, without having to analyze each historical instrument usage sequence. This can reduce the computational workload and improve the monitoring efficiency of the current state of the scientific research instrument.

[0095] As an optional embodiment, S103 may specifically include:

[0096] Determining sequence similarity between the reference instrument usage sequence and the current instrument usage sequence according to first instrument parameter configuration information of the reference instrument usage sequence and second instrument parameter configuration information of the current instrument usage sequence;

[0097] According to the sequence similarity, the initial time weight of the weighted moving average algorithm is modified to obtain the modified time weight;

[0098] According to the modified time weight, the predicted operating condition of the scientific research instrument is determined by the weighted moving average algorithm.

[0099] In this embodiment, the first instrument parameter configuration information is used to characterize the parameter configuration of each scientific research instrument in the reference instrument usage sequence, and the second instrument parameter configuration information is used to characterize the parameter configuration of each scientific research instrument in the current instrument usage sequence.

[0100] The weighted moving average algorithm is a method of giving different weights to the observed values, obtaining the moving average according to the different weights, and determining the predicted value based on the final moving average.

[0101] As an example, the monitor first determines the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence by the following formula 4:

[0102]

[0103] In the formula, E ε,γ It is used to characterize the sequence similarity between the γth reference instrument sequence and the εth current instrument sequence, T ε,γ It is used to characterize the timing similarity between the γth reference instrument sequence and the εth current instrument sequence, N a,γ The total number of scientific instruments used to characterize the γth reference instrument sequence γ, ε a,c The parameter value corresponding to the instrument parameter c of the scientific research instrument a in the εth current instrument usage sequence, γ a,c It is used to characterize the parameter value corresponding to the instrument parameter c of the scientific research instrument a in the γth reference instrument sequence. exp(―||) indicates the inverse proportional normalization calculation. μ a [] indicates the mean of the similarity of each instrument parameter corresponding to scientific research instrument a.

[0104] Among them, exp(―|ε a,c ―γ a,c |) represents the calculation of the similarity between the instrument parameter c of the scientific research instrument a in the εth current instrument usage sequence and the instrument parameter c of the scientific research instrument a in the γth reference instrument sequence, μ c [exp(―|ε a,c ―γ a,c |)] means calculating the mean of the similarity between the scientific research instrument a in the εth current instrument usage sequence and the scientific research instrument a in the γth reference instrument sequence. It is used to accumulate the mean values ​​of the similarities between the instrument parameters corresponding to the scientific research instruments in the εth current instrument usage sequence and the γth reference instrument sequence.

[0105] Among them, the temporal similarity T between the γth reference instrument sequence and the εth current instrument sequence is ε,γ The larger the value is, or the more similar the instrument parameters corresponding to each scientific research instrument in the γth reference instrument sequence and the εth current instrument usage sequence are, the sequence similarity E between the γth reference instrument sequence and the εth current instrument usage sequence is. ε,γ The bigger.

[0106] Then, after obtaining the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence, the initial time weight of the weighted moving average algorithm is corrected according to the sequence similarity. The product of the sequence similarity and the preset initial weight is used as the corrected time weight of the weighted moving average algorithm. Finally, according to the corrected time weight, the predicted operating condition of the scientific research instrument is determined by the weighted moving average algorithm.

[0107] Through this embodiment, according to the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence, the initial time weight of the weighted moving average algorithm is corrected to obtain the corrected time weight. Thus, according to the corrected time weight, the predicted operating condition of the scientific research instrument is determined by the weighted moving average algorithm. By accurately predicting the predicted operating condition of the scientific research instrument, the monitoring accuracy of the current state of the scientific research instrument can be improved.

[0108] As an optional embodiment, Figure 3 As shown, before S104, the current state monitoring method of the scientific research instrument may further include the following S301-S305:

[0109] S301, determining the similarity of multiple parameters of the target scientific research instrument and the preset reference instrument according to the first instrument parameter of the target scientific research instrument under the target working condition and the second instrument parameters of multiple batches of preset reference instruments;

[0110] S302, determining the reference degree of the second instrument parameter according to the parameter similarity and the interval time length of the corresponding second instrument parameter, where the interval time length is used to characterize the time difference between the acquisition time of the historical current state of the preset reference instrument under the second instrument parameter and the current time;

[0111] S303, correcting the historical current state of the preset reference instrument according to the reference degree of each second instrument parameter, to obtain the corrected current state of the target scientific research instrument under the target working condition;

[0112] S304, cyclically executing the process of determining the similarity of multiple parameters of the target scientific research instrument and the preset reference instrument according to the first instrument parameter of the target scientific research instrument under the target working condition and the second instrument parameters of multiple batches of the preset reference instrument, until various working conditions of the target scientific research instrument are traversed to obtain the corrected current state of the target scientific research instrument under various working conditions;

[0113] S305, constructing a current state model corresponding to the target scientific research instrument according to the corrected current state of the target scientific research instrument under various working conditions.

[0114] In this embodiment, the scientific research instrument may include a plurality of instrument parameters. For example, when the scientific research instrument is a scanning electron microscope, the instrument parameters may include resolution, maximum field of view, magnification range, electron beam current, and the like.

[0115] The interval time refers to the time difference between the acquisition time of the historical current state of the preset reference instrument under the second instrument parameter of the corresponding batch and the current time. The closer the time, the higher the reference degree of the second instrument parameter is usually.

[0116] As an example, the monitor first retrieves the configuration information of the target scientific research instrument, which may include the instrument name and instrument model. Then, the preset reference instrument corresponding to the target scientific research instrument and multiple batches of second instrument parameters corresponding to the preset reference instrument are screened in the preset reference instrument database. At the same time, the first instrument parameters of the target scientific research instrument under the target working condition in the historical time period are obtained.

[0117] Then, a similarity algorithm (such as cosine similarity, Euclidean distance, etc.) is used to calculate the parameter similarity between the first instrument parameters of the target scientific research instrument and the second instrument parameters of each batch of the preset reference instrument.

[0118] Then, for each batch of second instrument parameters, the reference degree of the second instrument parameters of the batch is determined according to the parameter similarity and the corresponding interval time. Specifically, a weighted average method can be adopted to multiply the parameter similarity by a weight coefficient inversely proportional to the interval time to obtain the reference degree of the second instrument parameters of the batch.

[0119] Then, according to the reference degree of the second instrument parameter of each batch, the historical current state of the preset reference instrument is corrected to obtain the corrected current state of the target scientific research instrument under the target working condition. Specifically, the corrected current state of the target scientific research instrument under the target working condition can be determined by the following formula 5:

[0120] I′ w =μ n (I w,k , C w,k ) Formula 5

[0121] Among them, I′ w Used to characterize the corrected current state of the target scientific research instrument when the working condition is w, I w,k It is used to characterize the historical current state corresponding to the second instrument parameter of the kth batch when the working condition of the preset reference instrument is w, C w,k It is used to characterize the reference degree corresponding to the second instrument parameter of the kth batch when the working condition of the preset reference instrument is w, μ n (I, C) is used to represent the calculated average value of the corrected current states corresponding to n batches of the second instrument parameter, where n is the number of batches of the second instrument parameter.

[0122] At the same time, in order to build a comprehensive current state model, it is necessary to traverse various working conditions of the target scientific research instrument. Therefore, for each working condition of the target scientific research instrument, the above steps are executed cyclically to obtain the corrected current state of the target scientific research instrument under each working condition.

[0123] Finally, according to the corrected current state of the target scientific research instrument under various working conditions, a current state model corresponding to the target scientific research instrument is constructed, wherein the model can be a mathematical expression, a neural network model or a data-based prediction model.

[0124] Through this embodiment, the parameter similarity and time factor between the target scientific research instrument and the preset reference instrument are comprehensively considered to realize the construction of the current state model of the target scientific research instrument. In this way, the predicted current state of the target scientific research instrument can be accurately obtained by simply inputting the working conditions of the target scientific research instrument into the current state model, thereby improving the accuracy and reliability of the current state measurement.

[0125] As an optional embodiment, S301 may specifically include:

[0126] The following steps are executed repeatedly until the second instrument parameters of each batch are traversed:

[0127] Performing difference calculation on a first instrument parameter of a target scientific research instrument under target working conditions and a second instrument parameter of a preset reference instrument to obtain an average parameter difference between the first instrument parameter and the second instrument parameter;

[0128] The average difference of parameters is inversely normalized to obtain the parameter similarity between the target scientific research instrument and the preset reference instrument.

[0129] In this embodiment, the parameter similarity can be specifically determined by the following formula 6:

[0130] S w,k =exp[-|μ τ (τ w,k -τ w,0 )|] Formula 6

[0131] In the formula, S w,k Used to characterize the parameter similarity between the first instrument parameter of the target scientific research instrument and the second instrument parameter of the kth batch of the preset reference instrument when the working condition is w, τ w,k The parameter value used to characterize the second instrument parameter of batch k when the working condition is w, τ w,0 It is used to characterize the parameter value of the first instrument parameter of the target scientific research instrument when the working condition is w. exp(―||) is used to characterize the inverse proportional normalization calculation. τ () is used to characterize the average value calculated for the difference values ​​of the corresponding parameters between the first instrument parameter and the second instrument parameter of the kth batch.

[0132] The greater the difference in parameter value between the first instrument parameter and the second instrument parameter of the kth batch of the preset reference instrument, the lower the parameter similarity between the first instrument parameter and the second instrument parameter of the kth batch of the preset reference instrument.

[0133] The lower the parameter similarity between the first instrument parameter and the second instrument parameter of the kth batch of the preset reference instrument, the lower the reference degree of the historical current state corresponding to the second instrument parameter of the kth batch of the preset reference instrument.

[0134] Through this embodiment, according to the parameter value of the first instrument parameter of the target scientific research instrument under the target working condition and the parameter value of the second instrument parameter of the preset reference instrument, the parameter similarity between the target scientific research instrument and the preset reference instrument can be accurately calculated. This helps to correct the historical current state of the preset reference instrument according to the parameter similarity, thereby improving the accuracy of model training.

[0135] As an optional embodiment, S302 may specifically include:

[0136] The difference between the acquisition time of the historical current state of the preset reference instrument corresponding to the second instrument parameter and the current time is calculated and the absolute value is taken to obtain the interval time of the second instrument parameter;

[0137] The interval duration of the second instrument parameter is divided by the parameter similarity and then the absolute value is calculated to obtain the initial reference degree;

[0138] The initial reference degree is inversely normalized and calculated to obtain the reference degree of the second instrument parameter.

[0139] In this embodiment, the reference degree of the second instrument parameter can be specifically determined by the following formula 7:

[0140]

[0141] In the formula, C w,k It is used to characterize the reference degree corresponding to the second instrument parameter of the kth batch when the working condition of the preset reference instrument is w. w,k It is used to characterize the parameter similarity between the first instrument parameter of the target scientific research instrument and the second instrument parameter of the kth batch of the preset reference instrument when the working condition is w. exp(―||) is used to characterize the inverse proportional normalization calculation. t w,k Used to characterize the interval time length corresponding to the second instrument parameter of the kth batch when the working condition of the preset reference instrument is w, that is, the interval time length between the collection time and the current time of the historical current state corresponding to the second instrument parameter of the kth batch when the working condition of the preset reference instrument is w.

[0142] The smaller the parameter similarity is, or the longer the interval is, the lower the reference degree corresponding to the second instrument parameter is, that is, the lower the reference degree of the historical current state corresponding to the second instrument parameter is.

[0143] Through this embodiment, the reference degree of the second instrument parameter can be accurately calculated according to the parameter similarity and the interval length of the corresponding second instrument parameter, which is helpful for the subsequent reference degree to correct the historical current state of the preset reference instrument, thereby improving the accuracy of model training.

[0144] As an optional embodiment, S105 may specifically include:

[0145] The predicted current state and the actual current state of the scientific research instrument are calculated and the absolute value is taken to obtain the initial risk index of the scientific research instrument;

[0146] Perform inverse proportional normalization calculation on the initial risk index of the scientific research instrument to obtain the final risk index of the scientific research instrument;

[0147] The final risk index is compared with the preset risk index threshold to obtain the current monitoring result of the scientific research instrument.

[0148] In this embodiment, the final risk index of the scientific research instrument can be specifically determined by the following formula 8:

[0149] F = exp(-|I0-I′0|) Formula 8

[0150] Wherein, F is used to characterize the final risk index of the scientific research instrument, I0 is used to characterize the actual current state of the scientific research instrument, I′0 is used to characterize the predicted current state of the scientific research instrument, exp() is an exponential function with natural constants as the base, and exp(―||) is used to characterize the inverse proportional normalization calculation.

[0151] Among them, the greater the difference between the actual current state I0 of the scientific research instrument and the predicted current state I′0 of the scientific research instrument, the greater the possibility of errors in the operation of the scientific research instrument, and the smaller the final risk index will be.

[0152] Therefore, a preset risk index threshold can be set, for example, the preset risk index threshold can be 0.6. When the final risk index is greater than or equal to the preset risk index threshold, it means that an error has occurred in the operation of the scientific research instrument, and an alarm is issued to notify the researcher to handle it; when the final risk index is less than the preset risk index threshold, it means that the scientific research instrument is operating normally.

[0153] Through this embodiment, the final risk index of the scientific research instrument is obtained according to the predicted current state and the actual current state of the scientific research instrument; then the final risk index is compared with the preset risk index threshold to obtain the current monitoring result of the scientific research instrument. In this way, accurate current monitoring of the scientific research instrument can be achieved to avoid safety accidents caused by failure of the scientific research instrument.

[0154] Based on the current state monitoring method of scientific research instruments, the present invention also provides a specific embodiment of a current state monitor for scientific research instruments.

[0155] Figure 4 A structural schematic diagram of a current status monitor for a scientific research instrument provided in an embodiment of the present application is shown. The current status monitor 400 for a scientific research instrument may include a sequence construction module 410, a sequence screening module 420, an operating condition determination module 430, a current prediction module 440, and a current comparison module 450.

[0156] A sequence construction module 410 is used to construct at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory, and the historical instrument usage sequence is used to represent the working order of each scientific research instrument in the historical time period;

[0157] The sequence screening module 420 is used to screen at least one reference instrument usage sequence matching the current instrument usage sequence from each historical instrument usage sequence according to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory;

[0158] The operating condition determination module 430 is used to determine the predicted operating condition of the scientific research instrument based on the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence;

[0159] The current prediction module 440 is used to input the predicted operating conditions into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument, where the current state model is a model trained based on the historical operating conditions of the scientific research instrument and the corresponding historical current states;

[0160] The current comparison module 450 is used to compare the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument. The actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the laboratory's intelligent sensing system.

[0161] In the current state monitor of scientific research instruments provided in this embodiment, a current state model is obtained by training in advance according to the historical operating conditions of the scientific research instruments and the corresponding historical current states. Then, according to the historical instrument usage sequences of the laboratory and the current instrument usage sequences corresponding to the scientific research instruments, the predicted operating conditions corresponding to the scientific research instruments are estimated. The predicted current state of the scientific research instruments can be accurately predicted by simply inputting the predicted operating conditions into the current state model corresponding to the scientific research instruments. At the same time, the actual current state of the scientific research instruments is collected according to the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system of the laboratory. Finally, the actual current state is compared with the predicted current state to accurately obtain the current monitoring results of the scientific research instruments. In this way, the embodiment of the present invention accurately collects the actual current state of the scientific research instruments through the intelligent sensors corresponding to the scientific research instruments in the intelligent sensing system, and accurately determines the predicted current state corresponding to the predicted operating conditions of the scientific research instruments according to the current state model. The different current states corresponding to the different operating conditions of the scientific research instruments are fully taken into account, so as to improve the monitoring accuracy of the current state of the scientific research instruments.

[0162] It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

[0163] The above is only a specific implementation of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.

Claims

1. A method for monitoring the current state of a scientific research instrument, characterized in that: The method comprises: According to the historical working time of each scientific research instrument in the laboratory, at least one historical instrument usage sequence of the laboratory is constructed, and the historical instrument usage sequence is used to characterize the working order of each scientific research instrument in the historical time period; According to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory, at least one reference instrument usage sequence matching the current instrument usage sequence is obtained from each of the historical instrument usage sequences; Determining a predicted operating condition of the scientific research instrument based on a sequence similarity between the reference instrument usage sequence and the current instrument usage sequence; Inputting the predicted operating condition into a current state model corresponding to the scientific research instrument to obtain a predicted current state of the scientific research instrument, wherein the current state model is a model trained according to the historical working time of the scientific research instrument and the corresponding historical current state; Comparing the predicted current state with the actual current state of the scientific research instrument to obtain a current monitoring result of the scientific research instrument, wherein the actual current state is collected by an intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system of the laboratory; The step of constructing at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory comprises: Calculate the usage time difference between the first scientific research instrument and the second scientific research instrument according to the historical working time of the first scientific research instrument and the second scientific research instrument, wherein the first scientific research instrument and the second scientific research instrument are any one of the scientific research instruments; Determine, according to the difference in usage time between the first scientific research instrument and the second scientific research instrument, a linkability evaluation value of the first scientific research instrument and the second scientific research instrument, wherein the linkability evaluation value is used to characterize the possibility of linking the first scientific research instrument and the second scientific research instrument for use; In the case where the linkability evaluation value is greater than a preset evaluation threshold, determining that the first scientific research instrument and the second scientific research instrument are in a link relationship; Calculating the usage time difference between the first scientific research instrument and the second scientific research instrument according to the historical working time of the first scientific research instrument and the second scientific research instrument in a loop until a preset stop condition is met, thereby obtaining a plurality of the link relationships; Constructing at least one historical instrument usage sequence of the laboratory according to each of the link relationships; The determining, according to the usage time difference between the first scientific research instrument and the second scientific research instrument, a linkability evaluation value of the first scientific research instrument and the second scientific research instrument comprises: Obtaining a difference between the number of times the first scientific research instrument is used and the number of times the second scientific research instrument is used, according to the number of times the first scientific research instrument is used and the number of times the second scientific research instrument is used; Multiplying the difference in the number of uses by the standard deviation of the difference in the use time between the first scientific research instrument and the second scientific research instrument to obtain an initial evaluation value; The initial evaluation value is inversely normalized to obtain linkable evaluation values ​​of the first scientific research instrument and the second scientific research instrument.

2. The current state monitoring method of scientific research instrument according to claim 1, characterized in that: The method of selecting, based on the current instrument usage sequence corresponding to each scientific research instrument in the laboratory, at least one reference instrument usage sequence matching the current instrument usage sequence from each of the historical instrument usage sequences comprises: Screening each of the historical instrument usage sequences to obtain a candidate instrument usage sequence including a target scientific research instrument, wherein the target scientific research instrument is any one of the scientific research instruments; Splitting each of the candidate instrument usage sequences according to the target scientific research instrument to obtain each of the candidate instrument usage subsequences containing the target scientific research instrument; Calculating the temporal similarity between the current instrument usage sequence corresponding to the target scientific research instrument and each of the candidate instrument usage subsequences respectively; Each of the candidate instrument usage subsequences whose timing similarity is greater than or equal to a preset similarity threshold is determined as a reference instrument usage sequence of the current instrument usage sequence.

3. The current state monitoring method of scientific research instrument according to claim 1, characterized in that: Determining the predicted operating condition of the scientific research instrument according to the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence includes: Determining sequence similarity between the reference instrument usage sequence and the current instrument usage sequence according to the first instrument parameter configuration information of the reference instrument usage sequence and the second instrument parameter configuration information of the current instrument usage sequence; According to the sequence similarity, the initial time weight of the weighted moving average algorithm is modified to obtain a modified time weight; According to the modified time weight, the predicted operating condition of the scientific research instrument is determined by the weighted moving average algorithm.

4. The current state monitoring method of scientific research instrument according to claim 1, characterized in that: Before inputting the predicted operating condition into the current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument, the method further includes: Determining the similarity of multiple parameters between the target scientific research instrument and the preset reference instrument according to the first instrument parameters of the target scientific research instrument under the target working condition and the second instrument parameters of multiple batches of preset reference instruments; Determine the reference degree of the second instrument parameter according to the parameter similarity and the corresponding interval duration of the second instrument parameter, wherein the interval duration is used to characterize the time difference between the acquisition time of the historical current state of the preset reference instrument under the second instrument parameter and the current time; According to the reference degree of each of the second instrument parameters, the historical current state of the preset reference instrument is corrected to obtain the corrected current state of the target scientific research instrument under the target working condition; Loop execution determines the similarity of multiple parameters between the target scientific research instrument and the preset reference instrument according to the first instrument parameter of the target scientific research instrument under the target working condition and the second instrument parameter of multiple batches of the preset reference instrument, until various working conditions of the target scientific research instrument are traversed to obtain the corrected current state of the target scientific research instrument under various working conditions; According to the corrected current state of the target scientific research instrument under the various working conditions, a current state model corresponding to the target scientific research instrument is constructed.

5. The current state monitoring method of scientific research instrument according to claim 4, characterized in that: The determining, based on the first instrument parameter of the target scientific research instrument under the target working condition and the second instrument parameter of multiple batches of preset reference instruments, the similarity of multiple parameters between the target scientific research instrument and the preset reference instrument comprises: The following steps are executed in a loop until the second instrument parameters of each batch are traversed: Performing difference calculation on a first instrument parameter of the target scientific research instrument under the target working condition and a second instrument parameter of the preset reference instrument to obtain an average parameter difference between the first instrument parameter and the second instrument parameter; The average difference of the parameters is inversely normalized to obtain the parameter similarity between the target scientific research instrument and the preset reference instrument.

6. The current state monitoring method of scientific research instrument according to claim 4, characterized in that: The determining the reference degree of the second instrument parameter according to the parameter similarity and the corresponding interval duration of the second instrument parameter includes: Calculate the difference between the acquisition time of the historical current state of the preset reference instrument corresponding to the second instrument parameter and the current time, and then take the absolute value to obtain the interval length of the second instrument parameter; The interval duration of the second instrument parameter is divided by the parameter similarity and then the absolute value is calculated to obtain an initial reference degree; The initial reference degree is inversely normalized and calculated to obtain the reference degree of the second instrument parameter.

7. The method for monitoring the current state of a scientific research instrument according to any one of claims 1 to 6, characterized in that: The comparing the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument includes: Calculate the difference between the predicted current state and the actual current state of the scientific research instrument and take the absolute value to obtain the initial risk index of the scientific research instrument; Performing inverse proportional normalization calculation on the initial risk index of the scientific research instrument to obtain the final risk index of the scientific research instrument; The final risk index is compared with a preset risk index threshold to obtain the current monitoring result of the scientific research instrument.

8. A current status monitor for scientific research instruments, characterized in that: The monitor comprises: A sequence construction module, used to construct at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory, wherein the historical instrument usage sequence is used to characterize the working order of each scientific research instrument in the historical time period; The step of constructing at least one historical instrument usage sequence of the laboratory according to the historical working time of each scientific research instrument in the laboratory comprises: Calculate the usage time difference between the first scientific research instrument and the second scientific research instrument according to the historical working time of the first scientific research instrument and the second scientific research instrument, wherein the first scientific research instrument and the second scientific research instrument are any one of the scientific research instruments; Determine, according to the difference in usage time between the first scientific research instrument and the second scientific research instrument, a linkability evaluation value of the first scientific research instrument and the second scientific research instrument, wherein the linkability evaluation value is used to characterize the possibility of linking the first scientific research instrument and the second scientific research instrument for use; In the case where the linkability evaluation value is greater than a preset evaluation threshold, determining that the first scientific research instrument and the second scientific research instrument are in a link relationship; Calculating the usage time difference between the first scientific research instrument and the second scientific research instrument according to the historical working time of the first scientific research instrument and the second scientific research instrument in a loop until a preset stop condition is met, thereby obtaining a plurality of the link relationships; Constructing at least one historical instrument usage sequence of the laboratory according to each of the link relationships; The determining, according to the usage time difference between the first scientific research instrument and the second scientific research instrument, a linkability evaluation value of the first scientific research instrument and the second scientific research instrument comprises: Obtaining a difference between the number of times the first scientific research instrument is used and the number of times the second scientific research instrument is used, according to the number of times the first scientific research instrument is used and the number of times the second scientific research instrument is used; Multiplying the difference in the number of uses by the standard deviation of the difference in the use time between the first scientific research instrument and the second scientific research instrument to obtain an initial evaluation value; Performing inverse proportional normalization calculation on the initial evaluation value to obtain linkable evaluation values ​​of the first scientific research instrument and the second scientific research instrument; A sequence screening module, for screening, according to the current instrument usage sequence corresponding to each scientific research instrument in the laboratory, at least one reference instrument usage sequence matching the current instrument usage sequence from each of the historical instrument usage sequences; A working condition determination module, used to determine the predicted operating condition of the scientific research instrument according to the sequence similarity between the reference instrument usage sequence and the current instrument usage sequence; A current prediction module, used for inputting the predicted operating condition into a current state model corresponding to the scientific research instrument to obtain the predicted current state of the scientific research instrument, wherein the current state model is a model trained according to the historical working time of the scientific research instrument and the corresponding historical current state; The current comparison module is used to compare the predicted current state with the actual current state of the scientific research instrument to obtain the current monitoring result of the scientific research instrument. The actual current state is collected by the intelligent sensor corresponding to the scientific research instrument in the intelligent sensing system of the laboratory.

Images