Laboratory multi-source data monitoring management method and system

Abstract

The invention provides a multi-source data monitoring management method and system for a laboratory, and belongs to the technical field of laboratory data processing and state control. The method is applied to process monitoring of laboratory reaction equipment, and comprises the following steps: step S100: acquiring a plurality of first monitoring data and a plurality of second monitoring data of a first type of monitoring sensors and a second type of monitoring sensors in a first time period; s200, determining a to-be-converted time domain data set and a non-converted time domain data set; s300, converting the time domain data set to be converted into a frequency domain data set; s400, constructing an N-order data analysis matrix; and S500, performing stability analysis on the N-order data analysis matrix to obtain a process analysis result of the laboratory reaction equipment. According to the technical scheme, whether the reaction process of the laboratory reaction equipment is stable or not can be accurately researched and judged on the basis of multi-source and multi-dimensional different-domain real-time sensing data, so that a stable reaction environment is provided for part of precision experiments, chemical test experiments and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of laboratory data processing and state control, and in particular relates to a laboratory multi-source data monitoring and management method and system, a computer-readable storage medium for implementing the method, a computer program product, and an electronic device. Background Art

[0002] With the popularization of IoT devices and the development of artificial intelligence technology, the current level of laboratory environment management and equipment status monitoring is constantly improving. For example, through artificial intelligence (AI) and machine learning technology, an intelligent monitoring and maintenance system for the laboratory environment is realized. The system can monitor the temperature, humidity, air quality and other environmental parameters in the laboratory in real time, and analyze historical data through machine learning algorithms to predict the potential change trend of the laboratory environment, and automatically adjust air conditioners, humidifiers and other equipment to ensure the stability of the experimental environment. The Chinese invention patent publication CN118094461A provides a laboratory monitoring system and method for multi-information fusion. By integrating multiple information, applying advanced data processing algorithms and integrating adaptive learning mechanisms, it improves the ability to identify, predict and respond to potential security threats in the laboratory, and provides comprehensive and effective security protection for various laboratory environments.

[0003] In actual experiments or testing processes, it is often necessary to further determine whether the currently running reaction equipment is in a comprehensive stable state. For example, some precision reaction tests require the test equipment (reaction equipment) to be turned on for a period of time until the equipment is in a stable state before the test can begin; some chemical reaction processes require ensuring that the reaction equipment itself is continuously in a stable state before the reaction results can be confirmed; in industrial applications, whether the reaction equipment can continue to be in a stable state is also the key to determining whether the output of the current reaction process can form large-scale industrial products.

[0004] In the prior art, multiple sensors are usually used to collect corresponding sensor data, such as pressure, temperature, humidity, vibration and other data, and whether the reaction equipment is in a stable state is judged by whether the data tends to be stable. However, due to the errors of the sensor equipment itself and the accidental and randomness of the test collection time point, the stability judgment result obtained by this single time domain and single trend data analysis method is relatively rough, and cannot accurately provide a stable reaction environment for some precision experiments, chemical testing experiments, etc. Summary of the invention

[0005] The laboratory multi-source data monitoring and management method and system proposed in the present invention, the computer-readable storage medium, the computer program product and the electronic device for implementing the method are proposed to solve the above-mentioned technical problems.

[0006] In a first aspect of the present invention, a laboratory multi-source data monitoring and management method is proposed, the method is applied to process monitoring of laboratory reaction equipment, the laboratory reaction equipment corresponds to multiple types of monitoring sensors; the method is characterized in that the method comprises the following steps:

[0007] S100: Acquire a plurality of first monitoring data of a first type of monitoring sensor in a first time period, and simultaneously acquire a plurality of second monitoring data of a second type of monitoring sensor in the first time period;

[0008] S200: Determine a time domain data set to be converted and a time domain data set not to be converted according to a comparison result of statistical indicators of the plurality of first monitoring data and the plurality of second monitoring data;

[0009] S300: Convert the to-be-converted time domain data set into a frequency domain data set;

[0010] S400: constructing an N-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set; N is an integer greater than 2;

[0011] S500: Perform stability analysis on the N-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0012] The multiple types of monitoring sensors include at least two of image monitoring sensors, noise monitoring sensors, vibration monitoring sensors, temperature monitoring sensors, and humidity monitoring sensors.

[0013] The step S200 specifically includes:

[0014] S201: Calculate a first statistical indicator of the plurality of first monitoring data and a second statistical indicator of the plurality of second monitoring data; the statistical indicator is one of variance, standard deviation, and maximum rate of change;

[0015] S202: When the first statistical indicator is greater than the second statistical indicator, determine that a plurality of first monitoring data are time domain data sets to be converted, and determine that a plurality of second monitoring data are time domain data sets not to be converted.

[0016] The step S500 specifically includes:

[0017] Calculating eigenvalues, eigenvectors, and / or matrix traces based on the N-order data analysis matrix;

[0018] Based on the calculation results of the characteristic roots, the characteristic vectors, and / or the matrix traces, it is determined whether the reaction process of the laboratory reaction device is stable.

[0019] In a second aspect of the present invention, the technical solution belonging to the same inventive concept also includes a multi-source data monitoring and management method for a laboratory, the method being used to determine whether a reaction process of a laboratory reaction device is stable, the laboratory reaction device corresponding to K different types of monitoring sensors;

[0020] The method comprises the following steps:

[0021] S110: Get each monitoring sensor Sen i P sensor data in a first time period;

[0022] S210: Determine a time domain data set to be converted and a time domain data set not to be converted;

[0023] S310: Convert the to-be-converted time domain data set into a frequency domain data set;

[0024] S410: constructing an M-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set, where M is an integer greater than 2;

[0025] S510: Perform stability analysis on the M-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0026] In a third aspect of the present invention, the technical solution belonging to the same inventive concept also includes a multi-source data monitoring and management method for a laboratory, the method being used to determine whether a reaction process of a laboratory reaction device is stable, the laboratory reaction device corresponding to K different types of monitoring sensors;

[0027] The method comprises the following steps:

[0028] S120: Get each monitoring sensor Sen i K sensor data in the first time period;

[0029] S220: Based on each monitoring sensor Sen i Based on the K sensor data in the first time period, a K-order data analysis matrix is ​​constructed;

[0030] S320: Perform stability analysis on the K-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0031] In a fourth aspect of the present invention, in order to implement the methods described in the above three aspects, a laboratory multi-source data monitoring and management system is provided, the system comprising K monitoring sensors of different categories corresponding to laboratory reaction equipment;

[0032] The K monitoring sensors of different categories include at least two of an image monitoring sensor, a noise monitoring sensor, a vibration monitoring sensor, a temperature monitoring sensor, and a humidity monitoring sensor; and a plurality of sensor data are collected through the K monitoring sensors of different categories;

[0033] The system also includes a matrix construction unit, a feature calculation unit, and a stability analysis unit;

[0034] The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data;

[0035] The feature calculation unit calculates at least one matrix feature based on the data analysis matrix;

[0036] The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics;

[0037] Wherein, the matrix characteristics include characteristic roots of the data analysis matrix;

[0038] The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics, specifically including:

[0039] When the absolute value of the characteristic root of the data analysis matrix is ​​less than 1, it is determined that the reaction process of the laboratory reaction equipment is stable.

[0040] The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data, specifically comprising:

[0041] Get each monitoring sensor Sen i K sensor data collected in the same time period;

[0042] Determine the time domain data set to be converted and the time domain data set not to be converted;

[0043] Converting the to-be-converted time-domain data set into a frequency-domain data set;

[0044] Based on the unconverted time domain data set and frequency domain data set, an M-order data analysis matrix is ​​constructed; M is an integer greater than 2.

[0045] The determining of the time domain data set to be converted and the time domain data set not to be converted specifically includes:

[0046] Calculate each monitoring sensor Sen i The maximum rate of change corresponding to the K sensor data collected in the same time period;

[0047] Based on each monitoring sensor Sen i The corresponding maximum rate of change determines the monitoring sensor Sen iWhether the corresponding K sensor data are time domain data sets to be converted.

[0048] The multi-source data monitoring and management methods of laboratories in various aspects mentioned above can be automatically implemented through various forms of electronic devices and computer program instructions; the computer program instructions can be stored in storage media of different forms and loaded into computer electronic devices for execution.

[0049] Therefore, in the fifth aspect of the present invention, a computer-readable storage medium is also provided for storing computer instructions, which, when executed on an electronic device, enables the electronic device to execute various aspects of the multi-source data monitoring and management method for laboratories.

[0050] In the sixth aspect of the present invention, a computer device is also proposed, which includes a processor and a memory, the memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the computer device executes the multi-source data monitoring and management method of the laboratory in the aforementioned aspects.

[0051] In a seventh aspect of the present invention, a computer program product is also proposed, the product comprising a computer program, and when the computer program is executed, the multi-source data monitoring and management method of the laboratory in the aforementioned aspects is implemented.

[0052] The technical solution of the present invention applies the Hurwitz stability criterion to the stability judgment of laboratory equipment for the first time. It can accurately judge whether the reaction process of laboratory reaction equipment is stable based on multi-source and multi-dimensional real-time sensor data in different domains, thereby providing a stable reaction environment for some precision experiments, chemical testing experiments, etc.

[0053] Further advantages of the present invention will be further reflected in detail in the specific embodiments section in conjunction with the drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be 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.

[0055] Figure 1-Figure 3 This is a schematic diagram of the main flow of a laboratory multi-source data monitoring and management method according to different embodiments of the present invention.

[0056] Figure 4 This is a schematic diagram of the functional modules of a laboratory multi-source data monitoring and management system according to an embodiment of the present invention.

[0057] Figure 5 Is implemented Figure 1 or Figure 2 or Figure 3 Schematic diagram of computer-readable storage medium and electronic device of the method steps / process DETAILED DESCRIPTION

[0058] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments that are the same as the present application. Instead, they are only examples of devices and methods that are the same as some aspects of the present application as detailed in the attached claims.

[0059] Figure 1-Figure 3 Three specific implementation schemes of the laboratory multi-source data monitoring and management method under three different embodiments are given. These three schemes are different implementation forms based on the same inventive concept. They will be introduced one by one below.

[0060] See also Figure 1 , a laboratory multi-source data monitoring management method is shown, the method is applied to process monitoring of laboratory reaction equipment, and the laboratory reaction equipment corresponds to multiple types of monitoring sensors. The multiple types of monitoring sensors include at least two of image monitoring sensors, noise monitoring sensors, vibration monitoring sensors, temperature monitoring sensors, and humidity monitoring sensors.

[0061] Of course, under different laboratory equipment and reaction conditions, there may be different monitoring objects, and thus there may be corresponding other different monitoring sensors, such as current, voltage, specific gas concentration, discharge / leakage, specific liquid concentration sensors, etc.

[0062] In practical applications, each target (to be observed) laboratory reaction device corresponds to at least two types of monitoring sensors, that is, each target (to be observed) laboratory reaction device is configured with at least two different monitoring sensors to collect different types of observation data.

[0063] Figure 1 The embodiment is introduced by taking the example of configuring two types of monitoring sensors for each target (to be observed) laboratory reaction equipment.

[0064] Preferably, the first type of sensor is an image monitoring sensor, which is used to collect reaction screen image frames of laboratory reaction equipment; the second type of sensor is a vibration monitoring sensor, which is used to capture vibration amplitude data of laboratory reaction equipment.

[0065] Figure 1The method includes steps S100-S500, and each step is specifically implemented as follows:

[0066] S100: Acquire a plurality of first monitoring data of a first type of monitoring sensor in a first time period, and simultaneously acquire a plurality of second monitoring data of a second type of monitoring sensor in the first time period;

[0067] S200: Determine a time domain data set to be converted and a time domain data set not to be converted according to a comparison result of statistical indicators of the plurality of first monitoring data and the plurality of second monitoring data;

[0068] S300: Convert the to-be-converted time domain data set into a frequency domain data set;

[0069] S400: constructing an N-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set; N is an integer greater than 2;

[0070] S500: Perform stability analysis on the N-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0071] Specifically, in step S100, monitoring data of a plurality of different sampling points of an image monitoring sensor and a vibration monitoring sensor within the same time period are acquired;

[0072] It should be noted that in order to ensure the representativeness of the data and obtain as many sampling time nodes as possible, although the sampling periods (starting point and ending point) of the image monitoring sensor and the vibration monitoring sensor are the same, the sampling time nodes are different.

[0073] For example, assuming that the sampling time period is [t1, t2], the sampling time nodes of the image monitoring sensor are [t1, t1+Δ1, t1+2Δ1, …, t1+kΔ1, t2];

[0074] The sampling time nodes of the vibration monitoring sensor are [t1, t1+Δ2, t1+2Δ2, …, t1+kΔ2, t2];

[0075] Wherein, Δ1≠Δ2; k≥2; preferably, k is an even number.

[0076] In this way, the number of sampling time nodes of the image monitoring sensor is k+2, and the number of sampling time nodes of the vibration monitoring sensor is also k+2, but the total number of sampling time nodes of the two is 2k+2.

[0077] For ease of description, k=2 is taken as an example for description below, but those skilled in the art are aware that the subsequent embodiments are applicable to all situations where k≥2.

[0078] At this time, in step S100, the monitoring data of the image monitoring sensor and the vibration monitoring sensor at four different sampling points in the same time period are obtained as follows:

[0079] A plurality of first monitoring data of the image monitoring sensor: [IM1, IM2, IM3, IM4]; wherein IM i (i=1,2,3,4) is the image data frame acquired by the image monitoring sensor for the i-th time;

[0080] A plurality of second monitoring data of the vibration monitoring sensor: [sM1, sM2, sM3, sM4], where sM j (j=1, 2, 3, 4) is the amplitude data frame obtained by the vibration monitoring sensor for the jth time;

[0081] Next, the process proceeds to step S200 : determining a time domain data set to be converted and a time domain data set not to be converted according to a comparison result of statistical indicators of the plurality of first monitoring data and the plurality of second monitoring data.

[0082] The multiple monitoring data obtained in step S100 are all time domain data, that is, each monitoring data is related to a time node. Most of the existing technologies only perform a horizontal single-dimensional follow-up analysis in the time domain. Although the corresponding stability judgment results can be obtained, the stability judgment results obtained by this single-dimensional single-domain data analysis method are relatively rough and cannot accurately provide a stable reaction environment for some precision experiments, chemical testing experiments, etc.

[0083] To this end, the first improvement of the present invention is to perform time domain-frequency domain conversion. In order to ensure that the representative information of multiple domains can be fully utilized, it is necessary to first determine which time domain data should be converted into frequency domain data, namely, the time domain data set to be converted, and the time domain data that does not need to be converted will be retained, which is called the unconverted time domain data set.

[0084] Specifically, the step S200 specifically includes:

[0085] S201: Calculate a first statistical indicator of the plurality of first monitoring data and a second statistical indicator of the plurality of second monitoring data; the statistical indicator is one of variance, standard deviation, and maximum rate of change;

[0086] S202: When the first statistical indicator is greater than the second statistical indicator, determine that a plurality of first monitoring data are time domain data sets to be converted, and determine that a plurality of second monitoring data are time domain data sets not to be converted.

[0087] The calculation of variance and standard deviation belongs to the prior art and will not be elaborated in detail in this embodiment.

[0088] The maximum rate of change is also one of the common statistical indicators. A common definition is as follows:

[0089] For a set of data sets {d1, d2, …, d N}, N ≥ 2, its maximum change rate maxC is defined as follows:

[0090]

[0091] Among them, max{……} means taking the maximum value among multiple values, min{……} means taking the minimum value among multiple values; avg{……} means taking the average value of multiple values.

[0092] For ease of description, it is assumed that after calculation in step S200, multiple first monitoring data [IM1, IM2, IM3, IM4] of the image monitoring sensor are determined to be time domain data sets to be converted, while multiple second monitoring data [sM1, sM2, sM3, sM4] of the vibration monitoring sensor are time domain data sets that are not converted.

[0093] Next, the process proceeds to step S300: converting the time domain data set to be converted into a frequency domain data set.

[0094] This step is to convert the first monitoring data [IM1, IM2, IM3, IM4] into a frequency domain data set.

[0095] The time-frequency conversion of data can be achieved through various frequency domain transformation models.

[0096] Preferably, in this embodiment, the time domain data can be converted into frequency domain data by Fourier transform method, such as finite Fourier transform. Based on the frequency domain data, the amplitude of each frequency component can be extracted to form an amplitude spectrum, and the main periodic patterns and trends of the frequency domain data can be determined based on the amplitude spectrum.

[0097] The time domain data is converted into frequency domain data, including: constructing a finite Fourier transform model; based on the time domain data, obtaining a time domain signal sequence, the number of sampling points of the time domain data and a time domain data index; based on the time domain signal sequence, the number of sampling points of the time domain data and the time domain data index, converting the time domain data into frequency domain data through the finite Fourier transform model.

[0098] As a specific example, the finite Fourier transform model is expressed as:

[0099]

[0100] Wherein, X(k) represents the complex value of the kth frequency component in the frequency domain. In the above embodiments of the present application, X(k) can be understood as the signal change value at a certain specific frequency.

[0101] Assume that after this step, the first monitoring data [IM1, IM2, IM3, IM4] is converted into a frequency domain data set [xM1, xM2, xM3, xM4] (for example, represented by the modulus value of the complex value X(k)).

[0102] Next, proceed to step S400: construct an N-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set; N is an integer greater than 2.

[0103] At this time, based on the unconverted time domain data set [sM1, sM2, sM3, sM4] and the frequency domain data set [xM1, xM2, xM3, xM4], the N-order data analysis matrix NMatrix is ​​constructed as follows:

[0104]

[0105] The N-order data analysis matrix NMatrix is ​​a semi-sparse matrix, whose block diagonal elements are combinations of the unconverted time domain data set and the frequency domain data set of the corresponding indexes, and the other elements are 0.

[0106] Next, the process proceeds to step S500: performing stability analysis on the N-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0107] Specifically, calculating eigenroots, eigenvectors, and / or matrix traces based on the N-order data analysis matrix;

[0108] Based on the calculation results of the characteristic roots, the characteristic vectors, and / or the matrix traces, it is determined whether the reaction process of the laboratory reaction device is stable.

[0109] Since only the block diagonal elements of the N-order data analysis matrix are non-zero, at this time, it is only necessary to calculate the eigenvalues, eigenvalues, and / or matrix traces of the block matrix composed of the block diagonal elements.

[0110] Taking the above N-order data analysis matrix NMatrix as an example, there are two block matrices composed of block diagonal elements:

[0111]

[0112] The standard for stability judgment in the field of automatic control is the Hurwitz stability criterion, and the present invention applies it to the stability judgment of laboratory equipment for the first time.

[0113] According to the Hurwitz stability criterion, when the main determinant of the matrix and all its diagonal subdeterminants have positive values, the system is stable and effective.

[0114] Combined with this argument, step S500 of the present invention can be further evolved into a question of determining whether the absolute value of the characteristic root is less than 1. Actual evaluation results show that the criterion is simple and effective, and is consistent with the Hurwitz stability criterion.

[0115] Since the matrix eigenvalues, eigenvectors, and matrix traces are actually related to each other, it is sufficient to calculate one of the eigenvalues, eigenvectors, and matrix traces.

[0116] To avoid repetition, the embodiment of the present invention is introduced by taking the calculation of matrix characteristic roots as an example, that is, calculating whether the absolute values ​​of the characteristic roots of the block matrix composed of block diagonal elements are all less than 1. If so, the system is stable.

[0117] When the absolute value of the characteristic root of the data analysis matrix (ie, the characteristic root of the block matrix composed of block diagonal elements) is less than 1, it is determined that the reaction process of the laboratory reaction equipment is stable.

[0118] The above embodiments introduce the method embodiments of the present invention by taking the data obtained by two types of monitoring sensors as an example. It not only takes into account multiple domain data, but also combines the matrix stability criterion to comprehensively perform stability judgment on multiple sources and multiple dimensions, thereby avoiding the errors existing in the sensor equipment itself and the contingency and randomness of the test collection time points, which lead to the problem that the stability judgment results obtained by the data analysis method of a single time domain and a single trend are relatively rough.

[0119] Figure 1 The embodiment of the present invention only takes two types of sensors (two groups of monitoring data of different categories) as an example to introduce a simple embodiment. However, in actual applications, due to the complexity of the reaction scene, more groups of monitoring data need to be collected. Therefore, Figure 2 The embodiment of the present invention takes a more general scenario as an example to introduce a multi-source data monitoring and management method for a laboratory according to another embodiment.

[0120] exist Figure 2 In the method, the method is used to determine whether the reaction process of a laboratory reaction device is stable, and the laboratory reaction device corresponds to K different types of monitoring sensors {Sen1, Sen2, ..., Sen K}, K ≥ 2;

[0121] at this time, Figure 2 The method comprises the following steps:

[0122] S110: Get each monitoring sensor Sen i The P sensing data {S i1 ,S i2 ,……,S iP},P≥2,i=1,2,……,K;

[0123] S210: Determine a time domain data set to be converted and a time domain data set not to be converted;

[0124] S310: Convert the to-be-converted time domain data set into a frequency domain data set;

[0125] S410: constructing an M-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set, where M is an integer greater than 2;

[0126] S510: Perform stability analysis on the M-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0127] Understandably, Figure 2 The method embodiment is simply to Figure 1 The sensor types (number) in the example of the method are changed from 2 to K (K>2). In addition, Figure 2 The method steps S110-S510 actually correspond to steps S100-S500 one by one. Figure 1 The principles and specific implementation methods of each step of the embodiment are also applicable to Figure 2 Embodiment of the invention.

[0128] Specifically, Figure 2 In the embodiment of step S210, determining the time domain data set to be converted and the time domain data set not to be converted may be:

[0129] Calculate each monitoring sensor Sen i The corresponding P sensor data {S i1 ,S i2 ,……,S iP}'s statistical index value index i ;

[0130] Determine the average of all statistical metric values

[0131] When index i ≥avgindex, determine the sensor data {S i1 ,S i2 ,……,S iP} is the time domain data set to be converted;

[0132] The statistical indicators are variance and standard deviation. Preferably, the statistical indicator may be a maximum rate of change.

[0133] The time domain-frequency domain conversion method of step S310 can also be referred to the aforementioned Figure 1 Step S300 of the embodiment.

[0134] Step 410 is based on the unconverted time domain data set and the frequency domain data set. At this time, the constructed M-order data analysis matrix is ​​also a sparse matrix or a semi-sparse matrix, whose block diagonal elements are the combination of the unconverted time domain data set and the frequency domain data set of the corresponding index, and the other elements are 0.

[0135] The method of constructing the M-order data analysis matrix can also be seen in the above Figure 1 Step S400 of the embodiment merely involves the corresponding changes in the order of the matrix and the diagonal block elements, and thus will not be described in detail.

[0136] For example, when k=4, taking the first monitoring data [IM1, IM2, IM3, IM4, IM5, IM6] as an example,

[0137] The corresponding form of the M-order data analysis matrix can be:

[0138]

[0139] Step S510 is also based on calculating whether the absolute values ​​of all characteristic roots of all matrices of the M-order data analysis matrix (i.e., the characteristic roots of the block matrix composed of block diagonal elements) are less than 1, to determine whether the reaction process of the laboratory reaction equipment is stable.

[0140] Of course, according to the Hurwitz stability criterion, the constructed data analysis matrix only needs to be an A×A matrix, A>2; when the non-zero elements contained in the known matrix (that is, the time domain data set and the frequency domain data set are not converted), technical personnel in this field can construct different analysis matrices according to different arrangements of the elements, including those based on the original unconverted time domain data set and the frequency domain data set, and those based on the expanded unconverted time domain data set and the frequency domain data set (for example, copying part of the data, repeating part of the data, etc.).

[0141] In the construction method provided in this embodiment, since the diagonal element form is adopted, the vector of the indicators such as eigenvalues ​​and eigenroots that need to be calculated finally is the lowest 2×2 matrix, which reduces the amount of calculation.

[0142] Figure 2 The embodiment also embodies the data value utilization method in the frequency domain (multi-domain) and multi-dimensional (block matrix composed of block diagonal elements), so as to comprehensively perform stability judgment on multiple sources and multiple dimensions.

[0143] However, it should be noted that if in certain scenarios, for certain time domain data, the time domain-frequency domain transformation is not feasible or meaningless or requires higher computing resource support, then the present invention can also directly construct an analysis matrix based on the existing time domain data to directly perform stability judgment, that is, omitting multi-domain considerations and only considering multi-dimensional factors. Such an improvement scheme is also a reflection of the value of multi-dimensional data utilization of the present invention.

[0144] therefore, Figure 3 The method proposed in the embodiment is based on this consideration. Although the time-frequency transformation is omitted, Figure 1 and Figure 2 They also belong to the same inventive concept, because based on the understanding of the prior art, multi-dimensional (analysis matrix) and multi-domain (time domain) transformations, whether alone or in combination, are creative contributions to the prior art in terms of laboratory equipment environment management, and are specific technical features that improve the prior art. Therefore Figure 1-Figure 3 The different implementation methods belong to the same general inventive concept and can solve the same technical problem (only the technical effects are different).

[0145] See also Figure 3 , shows a laboratory multi-source data monitoring management method, the method is used to determine whether the reaction process of laboratory reaction equipment is stable, the laboratory reaction equipment corresponds to K different types of monitoring sensors {Sen1, Sen2, ..., Sen K}, K ≥ 2;

[0146] The method comprises the following steps:

[0147] S120: Get each monitoring sensor Sen i K sensor data {S i1 ,S i2 ,……,S iK},i=1,2,……,K;

[0148] S220: Based on each monitoring sensor Sen i Based on the K sensor data in the first time period, a K-order data analysis matrix is ​​constructed;

[0149] S320: Perform stability analysis on the K-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

[0150] Likewise, Figure 3 Steps S120-S320 may correspond to reference Figure 2 Steps S110, S41, and S510 of the embodiment (omitting the steps of determining a time domain data set to be converted and a time domain data set not to be converted, and converting the time domain data set to be converted into a frequency domain data set) are not described in detail here.

[0151] Although multiple embodiments are given above, each embodiment can constitute an independent technical solution and may contribute to the existing technology and solve corresponding technical problems. However, it should be pointed out that different embodiments can be combined with each other without violating logic; at the same time, each embodiment can solve at least one technical problem, but it is not required that each individual embodiment solves multiple or all technical problems.

[0152] After introducing different implementations of the method embodiment, Figure 4 A schematic diagram of the functional modules of a laboratory multi-source data monitoring and management system according to an embodiment of the present invention is provided.

[0153] exist Figure 4 In FIG. 1 , a multi-source data monitoring and management system for a laboratory is shown, wherein the system includes K monitoring sensors of different categories corresponding to laboratory reaction equipment;

[0154] The K monitoring sensors of different categories {Sen1, Sen2, ..., Sen K}, K≥2, including at least two of the image monitoring sensors, noise monitoring sensors, vibration monitoring sensors, temperature monitoring sensors, and humidity monitoring sensors; multiple sensor data are collected through the K different types of monitoring sensors, and sent to the control management system of the electronic device through the data collection gateway node.

[0155] The control management system of the electronic equipment includes a matrix construction unit, a characteristic calculation unit and a stability analysis unit;

[0156] The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data;

[0157] The feature calculation unit calculates at least one matrix feature based on the data analysis matrix;

[0158] The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics;

[0159] Wherein, the matrix characteristics include characteristic roots of the data analysis matrix;

[0160] The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics, specifically including:

[0161] When the absolute value of the characteristic root of the data analysis matrix is ​​less than 1, it is determined that the reaction process of the laboratory reaction equipment is stable.

[0162] The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data, specifically comprising:

[0163] Get each monitoring sensor Sen i K sensor data {S i1 ,S i2 ,……,S iK},i=1,2,……,K;

[0164] Determine the time domain data set to be converted and the time domain data set not to be converted;

[0165] Converting the to-be-converted time-domain data set into a frequency-domain data set;

[0166] Based on the unconverted time domain data set and frequency domain data set, an M-order data analysis matrix is ​​constructed; M is an integer greater than 2.

[0167] The determining of the time domain data set to be converted and the time domain data set not to be converted specifically includes:

[0168] Calculate each monitoring sensor Sen i The maximum rate of change corresponding to the K sensor data collected in the same time period;

[0169] Based on each monitoring sensor Sen i The corresponding maximum rate of change determines the monitoring sensor Sen i Whether the corresponding K sensor data are time domain data sets to be converted.

[0170] The implementation principles of the corresponding functional modules of the system embodiment can refer to the aforementioned method embodiment.

[0171] For example, based on each monitoring sensor Sen i The corresponding maximum rate of change determines the monitoring sensor Sen i Whether the corresponding K sensor data are time domain data sets to be converted, specifically, it can be:

[0172] Calculate each monitoring sensor Sen i The corresponding P sensor data {S i1 ,S i2 ,……,S iP}Maximum rate of change value max V i ;

[0173] Determine the average of all statistical metric values

[0174] When max V i ≥avgV, determine the sensor data {S i1 ,S i2 ,……,S iP} is the time domain data set to be converted.

[0175] To sum up, the technical solution of the present invention applies the Hurwitz stability criterion to the stability judgment of laboratory equipment for the first time. It not only takes into account multiple domain data, but also combines the matrix stability criterion, so as to comprehensively perform stability judgment in multiple sources and dimensions, avoiding the errors existing in the sensor equipment itself and the contingency and randomness of the test acquisition time points, which lead to the problem that the stability judgment results obtained by the data analysis method of a single time domain and a single trend are relatively rough. It can accurately judge whether the reaction process of laboratory reaction equipment is stable based on real-time sensor data in different domains of multiple sources and multiple dimensions, thereby providing a stable reaction environment for some precision experiments, chemical testing experiments, etc.

[0176] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0177] Figure 5 Is implemented Figure 1 or Figure 2 or Figure 3 Schematic diagram of computer-readable storage media and electronic devices for the method steps / processes.

[0178] Figure 5 The computer-readable storage medium provided in the embodiment has a computer program stored thereon, and when the computer program is executed by an electronic device including a processor, the steps of a laboratory multi-source data monitoring and management method as described in the above embodiment are implemented.

[0179] For other technologies, principles, algorithms or models not elaborated in detail in this application, please refer to the prior art.

[0180] The foregoing has shown and described the method embodiments and system of the present invention, but it is understandable to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A laboratory multi-source data monitoring and management method, the method is applied to process monitoring of laboratory reaction equipment, the laboratory reaction equipment corresponds to multiple types of monitoring sensors; It is characterized in that The method comprises the following steps: S100: Acquire a plurality of first monitoring data of a first type of monitoring sensor in a first time period, and simultaneously acquire a plurality of second monitoring data of a second type of monitoring sensor in the first time period; S2 00: determining a time domain data set to be converted and a time domain data set not to be converted according to a comparison result of statistical indicators of the plurality of first monitoring data and the plurality of second monitoring data; S300: Convert the to-be-converted time domain data set into a frequency domain data set; S400: constructing an N-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set; N is an integer greater than 2; S500: Perform stability analysis on the N-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

2. A laboratory multi-source data monitoring and management method as claimed in claim 1, characterized in that: The multiple types of monitoring sensors include at least two of image monitoring sensors, noise monitoring sensors, vibration monitoring sensors, temperature monitoring sensors, and humidity monitoring sensors.

3. A laboratory multi-source data monitoring and management method as claimed in claim 1, characterized in that: The step S200 specifically includes: S201: Calculate a first statistical indicator of the plurality of first monitoring data and a second statistical indicator of the plurality of second monitoring data; the statistical indicator is one of variance, standard deviation, and maximum rate of change; S202: When the first statistical indicator is greater than the second statistical indicator, determine that a plurality of first monitoring data are time domain data sets to be converted, and determine that a plurality of second monitoring data are time domain data sets not to be converted.

4. A laboratory multi-source data monitoring and management method as claimed in claim 1, characterized in that: The step S500 specifically includes: Calculating eigenvalues, eigenvectors, and / or matrix traces based on the N-order data analysis matrix; Based on the calculation results of the characteristic roots, the characteristic vectors, and / or the matrix traces, it is determined whether the reaction process of the laboratory reaction device is stable.

5. A laboratory multi-source data monitoring and management method, the method is used to determine whether the reaction process of a laboratory reaction device is stable, the laboratory reaction device corresponds to K different types of monitoring sensors; It is characterized in that The method comprises the following steps: S110: Get each monitoring sensor Sen i P sensor data in a first time period; S210: Determine a time domain data set to be converted and a time domain data set not to be converted; S310: Convert the to-be-converted time domain data set into a frequency domain data set; S410: constructing an M-order data analysis matrix based on the unconverted time domain data set and the frequency domain data set; M is an integer greater than 2; S510: Perform stability analysis on the M-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

6. A laboratory multi-source data monitoring and management method, the method is used to determine whether the reaction process of a laboratory reaction device is stable, the laboratory reaction device corresponds to K different types of monitoring sensors; It is characterized in that The method comprises the following steps: S120: Get each monitoring sensor Sen i K sensor data in the first time period; S220: Based on each monitoring sensor Sen i Based on the K sensor data in the first time period, a K-order data analysis matrix is ​​constructed; S320: Perform stability analysis on the K-order data analysis matrix to obtain process analysis results of the laboratory reaction equipment.

7. A laboratory multi-source data monitoring and management system, the system comprising K monitoring sensors of different categories corresponding to laboratory reaction equipment; The K monitoring sensors of different categories include at least two of an image monitoring sensor, a noise monitoring sensor, a vibration monitoring sensor, a temperature monitoring sensor, and a humidity monitoring sensor; and a plurality of sensor data are collected through the K monitoring sensors of different categories; It is characterized in that The system also includes a matrix construction unit, a feature calculation unit, and a stability analysis unit; The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data; The feature calculation unit calculates at least one matrix feature based on the data analysis matrix; The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics; Wherein, the matrix characteristics include characteristic roots of the data analysis matrix; The stability analysis unit determines whether the reaction process of the laboratory reaction equipment is stable based on the matrix characteristics, specifically including: When the absolute value of the characteristic root of the data analysis matrix is ​​less than 1, it is determined that the reaction process of the laboratory reaction equipment is stable.

8. A laboratory multi-source data monitoring and management system as claimed in claim 7, characterized in that: The matrix construction unit constructs a data analysis matrix based on the plurality of sensor data, specifically comprising: obtaining each monitoring sensor Sen i K sensor data collected in the same time period; Determine the time domain data set to be converted and the time domain data set not to be converted; Converting the to-be-converted time-domain data set into a frequency-domain data set; Based on the unconverted time domain data set and frequency domain data set, an M-order data analysis matrix is ​​constructed; M is an integer greater than 2.

9. A laboratory multi-source data monitoring and management system as claimed in claim 8, characterized in that: The determining of the time domain data set to be converted and the time domain data set not to be converted specifically includes: Calculate each monitoring sensor Sen i The maximum rate of change corresponding to the K sensor data collected in the same time period; Based on each monitoring sensor Sen i The corresponding maximum rate of change determines the monitoring sensor Sen i Whether the corresponding K sensor data are time domain data sets to be converted.

10. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the steps of the multi-source data monitoring and management method for a laboratory as described in any one of claims 1 to 6 are implemented.

Images