Laboratory equipment maintenance management methods, devices, equipment and media

Abstract

The present invention provides a method, device, equipment and medium for maintenance management of experimental equipment, wherein the method comprises: obtaining a first performance indicator set of a target equipment in a first time period; obtaining a fault category of the target equipment according to the first performance indicator set, wherein the fault category is divided into a first category and a second category; extracting features from the first performance indicator set to obtain a first feature indicator set; constructing a first model according to a first sample, wherein the first sample is used to characterize a data set consisting of a first feature element and a fault category in the first feature indicator set; obtaining a second feature indicator set of the target equipment in a second time period; inputting the second feature indicator set into the first model to obtain a fault prediction probability of the target equipment, and determining the faulty equipment by the fault prediction probability. The present invention reduces the maintenance cost of equipment by proposing a method, device, equipment and medium for maintenance management of experimental equipment.

Description

Technical Field

[0001] The present invention relates to the field of equipment maintenance management, and in particular to a method, device, equipment and medium for maintenance management of experimental equipment. Background Art

[0002] At present, most maintenance management methods for laboratory equipment can only record, process, monitor and archive the equipment maintenance process. Therefore, the abnormality can only be recorded after the failure occurs and wait for relevant technicians to handle it. This will greatly increase the maintenance cost, delay the progress of teaching and scientific research, and even more seriously affect the safety of students and teachers. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the embodiment of the present invention provides a maintenance management method, device, equipment and medium for experimental equipment to reduce the maintenance cost of the equipment.

[0004] On the one hand, an embodiment of the present invention provides a method for maintenance management of experimental equipment, including: obtaining a first performance indicator set for a first time period of a target device; obtaining a fault category of the target device based on the first performance indicator set, wherein the fault category is divided into a first category and a second category, the first category being a performance element in the first performance indicator set indicating that a fault has occurred, and the second category being a performance element in the first performance indicator set indicating that no fault has occurred; performing feature extraction on the first performance indicator set to obtain a first feature indicator set; constructing a first model based on a first sample, wherein the first sample is used to characterize a data set consisting of a first feature element in the first feature indicator set and the fault category; obtaining a second feature indicator set for a second time period of the target device; inputting the second feature indicator set into the first model to obtain a fault prediction probability of the target device, and determining the faulty device through the fault prediction probability.

[0005] The experimental equipment maintenance management method according to the embodiment of the present invention has at least the following beneficial effects: feature processing is performed on the first performance indicator set of the target equipment to obtain the first feature indicator set, a first model is constructed based on the first feature indicator set, the second feature indicator set of the target equipment is input into the first model to obtain the faulty equipment, and the distance between the faulty equipment and the maintenance object category is used to assign the faulty equipment to the corresponding maintenance object for maintenance. In this way, intelligent fault prediction is performed on the equipment, thereby reducing the maintenance cost of the equipment.

[0006] According to some embodiments of the present invention, obtaining the fault category of the target device according to the first performance indicator set includes: the target device whose fault category is the first category is the first device; determining that the proportion of the first device in the target devices is less than a first threshold, randomly sampling the target devices of the second category so that the proportion is greater than or equal to the first threshold; determining that the performance element is a timing indicator, statistically processing the timing indicator to obtain at least one of a continuity indicator and a discontinuity indicator; determining that the performance element is a continuity indicator, normalizing the continuity indicator as a statistical element in the statistical indicator set; determining that the performance element is a discontinuity indicator, encoding the discontinuity indicator as a statistical element in the statistical indicator set.

[0007] According to some embodiments of the present invention, extracting features from the first performance indicator set to obtain a first feature indicator set includes: obtaining a target correlation based on a first average correlation and a second average correlation, wherein the first average correlation is an average correlation between the statistical element and the fault category, and the second average correlation is an average correlation between two random different statistical elements; the correlation formula for obtaining the target correlation is: Among them, Merit s For the target relevance, is the first average correlation value, is the second average correlation value, and k is the number of the statistical elements.

[0008] According to some embodiments of the present invention, extracting features from the first performance indicator set to obtain a first feature indicator set includes: randomly extracting two different statistical elements from the statistical indicator set as a first test indicator and a second test indicator, wherein the average value of the first correlation is the first average correlation, and the average value of the second correlation is the second average correlation; ff =r XY , r XY is the second correlation, and the first test indicator is expressed as: X = (x1, x2, ..., x n ); the second test index is expressed as: Y = (y1, y2, ..., y n ), n is the total number of the statistical elements in the statistical indicator set; it is determined that the first test indicator and the second test indicator are both continuity indicators, then the first formula for obtaining the second correlation is: in, For x i The mean of for y iThe mean of; determining that the first test indicator and the second test indicator are both discontinuous indicators, then the second formula for obtaining the second correlation is: Among them, n x is the number of discrete values ​​in the first test indicator, and the discrete value is 1...n x , x ki The value of is: if x k =i, then x ki =1, if x k ≠i, then x ki =0;n y is the number of discrete values ​​in the second test indicator, and the discrete value is 1...n y ,y kj The value of is: if y k =j, then y kj =1, if y k ≠j, then y kj =0;P xi =Sum(x k =i) / n,P yj =Sum(y k =j) / n,P ij =Sum(x k =i,y k =j) / n; determining that one of the first test indicator and the second test indicator is a continuity indicator and the other is a discontinuity indicator, then the third formula for obtaining the second correlation is: Among them, n x is the number of discrete values ​​in the first test indicator, and the discrete value is 1...n x , x ki The value of is: if x k =i, then x ki =1, if x k ≠i, then x ki =0; for y i The mean value, P xi =Sum(x k =i) / n.

[0009] According to some embodiments of the present invention, extracting features from the first performance indicator set to obtain a first feature indicator set includes:

[0010] The first step is to determine that the number of statistical elements is 1, obtain the target correlation of the statistical elements according to the correlation formula, obtain the statistical indicator corresponding to the reference value as the first characteristic element in the first characteristic indicator set, mark the statistical element, and update the number of the statistical elements, wherein the reference value is the maximum value of the target correlation obtained in the statistical indicator set; the second step is to determine that the number of statistical elements is not 1, obtain the target correlation of the unmarked statistical elements in the statistical indicator set according to the correlation formula, if the maximum value of the target correlation is greater than the reference value, obtain the corresponding statistical element as the first characteristic element in the first characteristic indicator set, mark the statistical element, and update the number of the statistical elements; the third step is to repeat the second step until the maximum value of the target correlation of the statistical element is less than or equal to the reference value, or the number of the statistical elements is equal to the number of the statistical elements in the statistical indicator set.

[0011] According to some embodiments of the present invention, constructing a first model based on the first sample includes: obtaining a first-order derivative and a second-order derivative through the first sample and a loss function; determining that the first characteristic element is a continuity indicator, then discretizing the first characteristic element to obtain a discontinuity indicator; obtaining a gain value of the first sample, and selecting the first sample with the largest gain value as a splitting point; splitting according to the splitting point to obtain a target tree, determining that the depth of the target tree is equal to a second threshold, stopping splitting, and obtaining the target tree as the first model; obtaining the weights of the nodes of the first model, and obtaining prediction values ​​according to the weights.

[0012] According to some embodiments of the present invention, it also includes: the target device whose fault category is the first category is the first device; obtaining a third characteristic indicator set for the first device, and forming a data set as a second sample based on a third characteristic element of the third characteristic indicator set and a maintenance object category; obtaining a fourth characteristic indicator set for the faulty device; obtaining a target distance based on the third characteristic indicator set and the fourth characteristic indicator set; obtaining the maintenance object category corresponding to the third characteristic element with the smallest target distance, and assigning the faulty device to the maintenance object category for maintenance.

[0013] On the other hand, an embodiment of the present invention provides an experimental equipment maintenance management device, including: a first module, used to obtain a first performance indicator set of a target device in a first time period; a second module, used to obtain a fault category of the target device based on the first performance indicator set, wherein the fault category is divided into a first category and a second category, the first category is that the performance element in the first performance indicator set shows that a fault has occurred, and the second category is that the performance element in the first performance indicator set shows that no fault has occurred; a third module, used to extract features from the first performance indicator set to obtain a first feature indicator set; a fourth module, used to construct a first model based on a first sample, wherein the first sample is used to characterize a data set consisting of a first feature element in the first feature indicator set and the fault category; a fifth module, used to obtain a second feature indicator set of the target device in a second time period; and a sixth module, used to input the second feature indicator set into the first model to obtain a fault prediction probability of the target device, and determine the faulty device by the fault prediction probability.

[0014] The experimental equipment maintenance management device according to the embodiment of the present invention has at least the following beneficial effects: the first performance indicator set of the target equipment is processed to obtain the first characteristic indicator set, the first model is constructed according to the first characteristic indicator set, the second characteristic indicator set of the target equipment is input into the first model to obtain the faulty equipment, and the distance between the faulty equipment and the maintenance object category is used to allocate the faulty equipment to the corresponding maintenance object for maintenance. In this way, the equipment is intelligently predicted for faults, thereby reducing the maintenance cost of the equipment.

[0015] On the other hand, an embodiment of the present invention also provides a computer device, comprising: at least one processor; at least one memory for storing at least one program; when the at least one program is executed by the at least one processor, the at least one processor implements the experimental equipment maintenance management method as described in any one of the above.

[0016] The computer equipment according to the embodiment of the present invention has at least the same beneficial effects as the above-mentioned experimental equipment maintenance management method.

[0017] On the other hand, an embodiment of the present invention provides a storage medium, wherein program instructions are stored in the storage medium, and when the program instructions are executed by a processor, the experimental equipment maintenance management method described in any one of the above can be implemented.

[0018] The storage medium according to the embodiment of the present invention has at least the same beneficial effects as the above-mentioned experimental equipment maintenance management method.

[0019] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flowchart of the steps of the experimental equipment maintenance management method provided by an embodiment of the present invention;

[0021] Figure 2 is a flowchart of the steps of a maintenance management method for experimental equipment provided by another embodiment of the present invention;

[0022] Figure 3 is a schematic block diagram of an experimental equipment maintenance management device provided by an embodiment of the present invention;

[0023] Figure 4 is a schematic block diagram of a device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. In the subsequent description, the use of suffixes such as "module", "component" or "unit" used to represent elements is only for the purpose of facilitating the description of the present invention, and has no specific meaning in itself. Therefore, "module", "component" or "unit" can be used in a mixed manner. "First", "second" and the like are only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the order of the indicated technical features. In this subsequent description, the continuous numbering of the method steps is for the convenience of review and understanding. In combination with the overall technical solution of the present invention and the logical relationship between the various steps, adjusting the implementation order between the steps will not affect the technical effect achieved by the technical solution of the present invention. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0025] Reference Figure 1 , the method of the embodiment of the present invention includes:

[0026] Step S100, obtaining a first performance indicator set of a target device in a first time period;

[0027] Step S200, obtaining a fault category of the target device according to the first performance indicator set, wherein the fault category is divided into a first category and a second category, the first category is when the performance element in the first performance indicator set shows that a fault has occurred, and the second category is when the performance element in the first performance indicator set shows that no fault has occurred;

[0028] According to some embodiments of the present invention, step S200 further includes the following steps:

[0029] The target device whose fault category is the first category is the first device;

[0030] Determining that the proportion of the first device in the target devices is less than a first threshold, randomly sampling the target devices of the second category so that the proportion is greater than or equal to the first threshold;

[0031] If the performance element is determined to be a time series indicator, the time series indicator is statistically processed to obtain at least one of a continuous indicator and a discontinuous indicator;

[0032] If the performance element is determined to be a continuous indicator, the continuous indicator is normalized and used as a statistical element in the statistical indicator set;

[0033] If the performance element is determined to be a discontinuous indicator, the discontinuous indicator is coded and processed as a statistical element in the statistical indicator set.

[0034] Specifically, a device performance indicator capture tool is used to obtain a first performance indicator set of a target device in a first time period, wherein the first performance indicator set contains many performance elements, including historical fault conditions of the target device, and the historical fault conditions include predicted and manually detected faults. The target device is classified according to the performance elements in the first performance indicator set, wherein the device that has failed is in the first category, and the device that has not failed is in the second category. Considering that the frequency of occurrence of faulty devices is not high, it will affect the accuracy of the model when establishing and using the model. Therefore, the first category of faulty devices, that is, the proportion of the first device in the faulty devices, is obtained according to the fault category. If the proportion is less than the first threshold, the target devices in the second category are randomly selected so that the proportion of the first device and the target device is equal to the first threshold. The randomly selected devices become a new target device set for subsequent processing. It can be understood that the first threshold can be selected based on prior knowledge or actual needs. According to this method, the samples are balanced, thereby improving the accuracy of model prediction. In the first performance indicator set, the performance element may be a time series performance indicator or a non-time series performance indicator. The time series points in the time series performance indicators are obtained. For example, the performance data of each hour in the first time period can be selected, and each hour of performance data corresponds to a time series point. The obtained time series points are statistically processed to obtain non-continuous indicators such as maximum value, minimum value, mean, variance, trend, peak value, skewness, etc. It should be noted that the missing time series points are supplemented with 0. For example, Monday to Friday is taken as the first time period, and the performance indicator data of 12 hours in the two time periods of 8-12 o'clock and 14-22 o'clock every day is selected as 60 time series points. There may be time series points without performance indicator data. At this time, these missing values ​​are supplemented with 0. The non-temporal indicators are divided into continuous indicators and discontinuous indicators, wherein all discontinuous indicators, including the discontinuous indicators in the non-temporal indicators and the discontinuous indicators obtained after processing the temporal indicators, are encoded to obtain statistical elements; after the continuous indicators are normalized, statistical elements are also obtained, wherein the missing values ​​in the continuous indicators are also filled; the obtained statistical elements are aggregated to obtain a statistical indicator set. It can be understood that the discontinuous indicators are discrete indicators. This method is used to preprocess the first performance indicator set so that the indicators can be used in the subsequent execution environment.

[0035] Step S300, extracting features from the first performance indicator set to obtain a first feature indicator set;

[0036] Specifically, the target correlation is obtained according to the first average correlation and the second average correlation, wherein the first average correlation is the average correlation between the statistical element and the fault category, and the second average correlation is the average correlation between two random different statistical elements; it should be noted that the first category is marked as 1, and the second category is marked as 0. The statistical elements in the statistical indicator set are screened by the target correlation, wherein the correlation formula for obtaining the target correlation is: in, is the first average correlation value, is the second average correlation value, k is the number of statistical elements. Two different statistical elements are randomly selected from the statistical indicator set as the first test indicator and the second test indicator, wherein the average value of the first correlation is the first average correlation, and the average value of the second correlation is the second average correlation; r ff =r XY , r XY For the second correlation, the first test indicator is expressed as: X = (x1, x2, ..., x n ); The second test indicator is expressed as: Y = (y1, y2, ..., y n ), n is the total number of statistical elements in the statistical indicator set; it can be understood that r ff =r XY , although expressed in different forms, they both represent the second correlation of two different statistical elements, r XY The x and y in the formula refer to the second correlation between the first test index and the second test index, and r ff The f in the table generally refers to the statistical elements in the statistical indicator set.

[0037] Here are the methods to obtain the second correlation for different statistical elements:

[0038] It is determined that both the first test index and the second test index are continuity indexes, and the first formula for obtaining the second correlation is: in, For x i The mean of for y i The mean of

[0039] It is determined that both the first test indicator and the second test indicator are discontinuous indicators, and the second formula for obtaining the second correlation is: Among them, n x is the number of discrete values ​​in the first test indicator, and the discrete value is 1...n x , x ki The value of is: if x k =i, then x ki =1, if x k ≠i, then xki =0;n y is the number of discrete values ​​in the second test indicator, and the discrete value is 1...n y ,y kj The value of is: if y k =j, then y kj =1, if y k ≠j, then y kj =0;P xi =Sum(x k =i) / n,P yj =Sum(y k =j) / n,P ij =Sum(x k =i,y k =j) / n;

[0040] It is determined that one of the first test indicator and the second test indicator is a continuous indicator and the other is a discontinuous indicator, then the third formula for obtaining the second correlation is: Among them, n x is the number of discrete values ​​in the first test indicator, and the discrete value is 1...n x , x ki The value of is: if x k =i, then x ki =1, if x k ≠i, then x ki =0; for y i The mean of .

[0041] It should be noted that, because the fault category is discrete, when obtaining the first correlation, if the selected statistical element is determined to be a continuous indicator, the third formula is used for calculation, and if the selected statistical element is determined to be a discontinuous indicator, the second formula is used for calculation.

[0042] The target correlation is used to extract features from the statistical indicator set to obtain the first feature indicator set:

[0043] The first step is to determine that the number of statistical elements is 1, that is, k in the correlation formula is 1, then obtain the target correlation of all statistical elements in the statistical indicator set, obtain the statistical element with the largest target correlation as the first characteristic element in the first characteristic indicator set, mark the selected statistical elements, and increase the number of statistical elements, that is, the k value; wherein the maximum value of the target correlation is used as a reference value.

[0044] The second step is to determine that the number of statistical elements is not 1, that is, k in the correlation formula is not 1, then obtain the unmarked statistical elements in the statistical indicator set, obtain their target correlation sizes, select the maximum value of the target correlation, compare the maximum value with the reference value, and if it is greater than the reference value, obtain the statistical element corresponding to the maximum value as the first characteristic element in the first characteristic indicator set, mark the statistical element selected as the first characteristic element, and increase the number of statistical elements;

[0045] The third step is to repeat the second step until the target correlation of the statistical element is determined to be less than or equal to the reference value, or the number of statistical elements is determined, that is, the k value in the correlation formula is equal to the number of statistical elements in the statistical indicator set, then terminate the feature selection.

[0046] It can be understood that when the second step is repeated, the k value is continuously increased, that is, each time it is repeated, the k value is updated, and 1 is added to the original value to obtain a new value.

[0047] It should be noted that after obtaining the first characteristic indicator set, the first characteristic elements can be sorted according to the size of the first correlation, so as to facilitate the subsequent prediction of the target device and the fault location. After the performance indicators are processed into statistical indicators, there will be many indicator features, which not only generate redundancy between the indicators, but also too many indicators will affect the efficiency of model training. This method can better capture the feature correlation and select the first characteristic element in the first characteristic indicator set from the statistical indicator set.

[0048] Step S400, constructing a first model according to a first sample, wherein the first sample is used to characterize a data set consisting of a first characteristic element in a first characteristic indicator set and a fault category;

[0049] Specifically, the first-order derivative and the second-order derivative are obtained according to the first sample and the loss function, wherein the first sample is obtained according to the data set consisting of the first characteristic element in the first characteristic indicator set and the fault category, and the first sample is expressed as: I = {(x1, y1), (x2, y2), ..., (x n ,y n )},x i (i=1,...,n) is the attribute value of the first feature element, y i (i=1,...,n) is the fault category of the target device; the loss function is: Get each sample, that is, I = {(x1, y1), (x2, y2), ..., (x n ,y n )} in i (i=1,...,n), through each sample in this round of loss function based on The first and second derivatives of: g ti =p ti -y i ,h ti =p ti *(1-p ti ), where g ti is the first-order derivative, h ti is the second-order derivative, The first characteristic element in the first characteristic index set is numbered. If the first characteristic element is a continuous index, the first characteristic element is discretized to obtain a discontinuous index. The gain value of the characteristic of the first characteristic element is obtained, and the first sample with the largest gain value is selected as the splitting point, where the formula of the gain value is: Among them, gain is the gain value, i∈L is the first sample split to the left subtree, i∈R is the first sample split to the right subtree, and λ and γ are regularization coefficients. For the first samples of the left subtree and the right subtree, continue to select the first sample with the largest gain value to split to obtain the target tree until it is confirmed that the depth of the target tree is equal to the second threshold, stop splitting, and obtain the target tree as the first model. It can be understood that the second threshold can be selected based on prior knowledge or actual needs. The weight of each node is obtained through the first model, where the node is the last leaf node of the target tree, and the formula for calculating the weight is: where w tj is the weight, i∈j is the first sample of node j assigned to the tth tree, and η is the learning rate. It can be understood that the regularization coefficient and the learning rate can be selected according to prior knowledge or actual needs, where The samples in the first sample are calculated to construct a target tree of the target value, the target tree is obtained as the first model, and the predicted value is obtained according to the first model: in, is the predicted value, w tj is the weight of the node, T is the number of target trees, and t is the number of target trees. This method can prevent data overfitting, and the accuracy of the first model obtained is higher than that of the ordinary model. It can be understood that the third threshold can be selected according to prior knowledge or actual needs.

[0050] Step S500, obtaining a second characteristic indicator set of the target device in a second time period;

[0051] Step S600: input the second characteristic indicator set into the first model to obtain the fault prediction probability of the target device, and determine the faulty device according to the fault prediction probability.

[0052] Specifically, a second characteristic indicator set of the target device in the second time period is obtained, and the second characteristic indicator set is also preprocessed with reference to the preprocessing of the first characteristic indicator set to obtain indicators that are convenient for subsequent operations, and the preprocessed indicators are input into the first model to obtain the predicted value of device i. The predicted value is the predicted fault probability. If the predicted fault probability is greater than the third threshold, it is determined to be a faulty device.

[0053] According to some embodiments of the present invention, the present invention further comprises the following steps: Figure 2 As shown:

[0054] Step S710, obtaining a third characteristic indicator set of the first device, and forming a data set as a second sample according to a third characteristic element of the third characteristic indicator set and a maintenance object category;

[0055] Step S720, obtaining a fourth characteristic indicator set of the faulty device;

[0056] Step S730, obtaining a target distance according to the third characteristic indicator set and the fourth characteristic indicator set;

[0057] Step S740: Obtain the maintenance object category corresponding to the third characteristic element with the smallest target distance, and assign the faulty equipment to the maintenance object category for maintenance.

[0058] Specifically, the third characteristic index set of the first category of the target equipment is obtained, and a data set is formed as a second sample according to the third characteristic element in the third index set and the maintenance object category. The second sample is expressed as: I = {(x1, z1), (x2, z2), ..., (x n ,z n )},x i (i=1,...,n) is the attribute value of the third characteristic element, z i (i=1,…,n) is the maintenance object category, for example, the maintenance object category can be 1, 2, 3, where 1 is the supplier, 2 is the laboratory teacher, and 3 is the software and hardware major student. Get the fourth characteristic indicator set of the faulty equipment, where the third characteristic indicator set is expressed as: S i =(S i1 ,...,S iq ), S ia (a=1,...,q) is the value of the attribute of the ath third characteristic element; the fourth characteristic index set is expressed as: x j =(x j1 ,...,x jq ), x ja (a=1,...,q) is the value of the attribute of the ath fourth characteristic element. Get the target distance of the third characteristic index set and the fourth characteristic index set: Among them, C p For the maintenance object category, it can be understood that the third characteristic indicator set has a different form of expression in the second sample and in the target distance, but the content represented is the same, and different forms of expression are used to distinguish during calculation. Obtain the third characteristic indicator element with the smallest target distance from the fourth characteristic element, and obtain the maintenance object category corresponding to the selected third characteristic indicator element according to the second sample, so as to obtain the maintenance object category corresponding to the fourth characteristic element, and assign the faulty equipment to the maintenance object category for processing. When the maintenance object is troubleshooting and repairing the fault, if it is found that it is not the corresponding fault, the maintenance object corresponding to the third characteristic element is modified, or if it is found that the fault cannot be solved by the assigned maintenance object, the maintenance object corresponding to the third characteristic element is modified. This method can reduce the impact on the progress of teaching and scientific research, reduce human resources, and improve maintenance efficiency. At the same time, the maintenance objects include students, and it can also exercise students' ability to handle faults.

[0059] On the one hand, referring to Figure 3 This embodiment provides an experimental equipment maintenance management device, which at least includes: a first module 310, a second module 320, a third module 330, a fourth module 340, a fifth module 350 and a sixth module 360.

[0060] Specifically, the first module 310 obtains the first performance indicator set of the target device in the first time period, the second module 320 obtains the fault category of the target device according to the first performance indicator and inputs it into the third module 330, the third module 330 extracts features from the first performance indicator set to obtain a first feature indicator set, the fourth module 340 constructs a first model based on the first feature element in the first feature indicator set and the first sample composed of the fault category, the fifth model 350 obtains the first feature indicator set of the target device in the second time period, the sixth model 360 inputs the first feature indicator set of the fifth model 350 into the first model, obtains the fault prediction concept of the target device, and obtains the faulty device among the target devices according to the fault prediction probability.

[0061] Reference Figure 4, this embodiment provides an electronic device, including a processor 410 and a memory 420 coupled to the processor 410, the memory 420 stores program instructions that can be executed by the processor 410, and the processor 410 implements the above-mentioned experimental equipment maintenance management method when executing the program instructions stored in the memory 420. Among them, the processor 410 can also be called a CPU (Central Processing Unit). The processor 410 may be an integrated circuit chip with signal processing capabilities. The processor 410 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor, or the general-purpose processor can also be any conventional processor, etc. The memory 420 may include various components (e.g., machine-readable media), including but not limited to random access memory components, read-only components and any combination thereof. The memory 420 may also include: instructions (e.g., software) (e.g., stored in one or more machine-readable media); the instructions implement the experimental equipment maintenance management method in the above-mentioned embodiment. The electronic device has the function of carrying and running the software system for experimental equipment maintenance management provided in the embodiment of the present invention, for example, a personal computer (PC), a mobile phone, a smart phone, a personal digital assistant (PDA), a wearable device, a Pocket PC (PPC), a tablet computer, etc.

[0062] This embodiment provides a computer-readable storage medium, which stores a program, and the program is executed by a processor to implement the above-mentioned experimental equipment maintenance management method.

[0063] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0064] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the aforementioned experimental equipment maintenance management method.

[0065] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0066] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0067] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0068] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0069] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0070] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0071] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0072] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for maintenance management of experimental equipment, characterized in that: include: Obtain a first performance indicator set of a target device in a first time period; Obtaining a fault category of the target device according to the first performance indicator set, wherein the fault category is divided into a first category and a second category, the first category being that the performance element in the first performance indicator set shows that a fault has occurred, and the second category being that the performance element in the first performance indicator set shows that no fault has occurred; Performing feature extraction on the first performance indicator set to obtain a first feature indicator set; Constructing a first model according to a first sample, wherein the first sample is used to characterize a data set consisting of a first characteristic element in the first characteristic indicator set and the fault category; constructing the first model according to the first sample includes: obtaining a first-order derivative and a second-order derivative through the first sample and a loss function; determining that the first characteristic element is a continuity indicator, discretizing the first characteristic element to obtain a discontinuity indicator; obtaining a gain value of the first sample, selecting the first sample with the largest gain value as a splitting point; splitting according to the splitting point to obtain a target tree, determining that the depth of the target tree is equal to a second threshold, stopping the splitting, and obtaining the target tree as the first model; obtaining the weights of the nodes of the first model, and obtaining a prediction value according to the weights; Acquire a second characteristic indicator set of the target device in a second time period; Inputting a second characteristic indicator set into the first model to obtain a fault prediction probability of the target device, and determining a faulty device according to the fault prediction probability; The target device whose fault category is the first category is the first device; the third characteristic indicator set of the first device is obtained, and a data set is composed as a second sample according to the third characteristic element of the third characteristic indicator set and the maintenance object category; the fourth characteristic indicator set of the faulty device is obtained; the target distance is obtained according to the third characteristic indicator set and the fourth characteristic indicator set; the maintenance object category corresponding to the third characteristic element with the smallest target distance is obtained, and the faulty device is assigned to the maintenance object category for maintenance.

2. The experimental equipment maintenance management method according to claim 1, characterized in that: The obtaining the fault category of the target device according to the first performance indicator set includes: The target device whose fault category is the first category is the first device; Determining that the proportion of the first devices in the target devices is less than a first threshold, randomly sampling the target devices of the second category so that the proportion is greater than or equal to the first threshold; Determining that the performance element is a time series indicator, statistically processing the time series indicator to obtain at least one of a continuous indicator and a discontinuous indicator; Determining that the performance element is a continuity indicator, normalizing the continuity indicator as a statistical element in a statistical indicator set; If it is determined that the performance element is a discontinuity indicator, the discontinuity indicator is encoded and processed as a statistical element in the statistical indicator set.

3. The experimental equipment maintenance management method according to claim 2, characterized in that: The step of extracting features from the first performance indicator set to obtain a first feature indicator set includes: Obtaining a target correlation according to a first average correlation and a second average correlation, wherein the first average correlation is an average correlation between the statistical element and the fault category, and the second average correlation is an average correlation between two random different statistical elements; The correlation formula for obtaining the target correlation is: ， in, For the target relevance, is the first average correlation, is the second average correlation, is the number of the statistical elements.

4. The experimental equipment maintenance management method according to claim 3, characterized in that: The step of extracting features from the first performance indicator set to obtain a first feature indicator set includes: Randomly extracting two different statistical elements from the statistical indicator set as a first test indicator and a second test indicator, wherein the average value of the first correlation is the first average correlation, and the average value of the second correlation is the second average correlation; , For the second correlation, the first test indicator is expressed as: ; The second test indicator is expressed as: , is the total number of the statistical elements in the statistical indicator set; It is determined that both the first test index and the second test index are continuity indexes, and the first formula for obtaining the second correlation is: ， in, for The mean of for The mean of It is determined that both the first test indicator and the second test indicator are discontinuous indicators, and the second formula for obtaining the second correlation is: ， in, is the number of discrete values ​​in the first test indicator, and the value of the discrete value is 1 , The value of is: ,but ,like ,but ; The number of discrete values ​​in the second test indicator, the value of discrete value is 1 , The value of is: ,but ,like ,but ; , , ; It is determined that one of the first test indicator and the second test indicator is a continuous indicator and the other is a discontinuous indicator, then the third formula for obtaining the second correlation is: ， in, is the number of discrete values ​​in the first test indicator, and the value of the discrete value is 1 , The value of is: ,but ,like ,but ; for The mean of .

5. The experimental equipment maintenance management method according to claim 4, characterized in that: The step of extracting features from the first performance indicator set to obtain a first feature indicator set includes: The first step is to determine that the number of the statistical elements is 1, obtain the target correlation of the statistical elements according to the correlation formula, obtain the statistical index corresponding to the reference value as the first characteristic element in the first characteristic index set, mark the statistical element, and update the number of the statistical elements, wherein the reference value is the maximum value of the target correlation obtained in the statistical index set; The second step is to determine that the number of the statistical elements is not 1, obtain the target correlation of the unmarked statistical elements in the statistical indicator set according to the correlation formula, and if the maximum value of the target correlation is greater than the reference value, obtain the corresponding statistical element as the first feature element in the first feature indicator set, mark the statistical element, and update the number of the statistical elements; The third step is to repeat the second step until the maximum value of the target correlation of the statistical element is less than or equal to the reference value, or the number of the statistical elements is equal to the number of the statistical elements in the statistical indicator set.

6. An experimental equipment maintenance management device, characterized in that: include: A first module is used to obtain a first performance indicator set of a target device in a first time period; A second module is used to obtain a fault category of the target device according to the first performance indicator set, wherein the fault category is divided into a first category and a second category, the first category is that the performance element in the first performance indicator set shows that a fault has occurred, and the second category is that the performance element in the first performance indicator set shows that no fault has occurred; A third module is used to extract features from the first performance indicator set to obtain a first feature indicator set; The fourth module is used to construct a first model according to a first sample, wherein the first sample is used to characterize a data set consisting of a first characteristic element in the first characteristic indicator set and the fault category; the construction of the first model according to the first sample includes: obtaining a first-order derivative and a second-order derivative through the first sample and a loss function; determining that the first characteristic element is a continuity indicator, discretizing the first characteristic element to obtain a discontinuity indicator; obtaining a gain value of the first sample, and selecting the first sample with the largest gain value as a splitting point; splitting according to the splitting point to obtain a target tree, determining that the depth of the target tree is equal to a second threshold, stopping the splitting, and obtaining the target tree as the first model; obtaining the weights of the nodes of the first model, and obtaining a prediction value according to the weights; A fifth module is used to obtain a second characteristic indicator set of the target device in a second time period; The sixth module is used to input the second characteristic indicator set into the first model to obtain the fault prediction probability of the target device, and determine the faulty device through the fault prediction probability; the target device whose fault category is the first category is the first device; obtain the third characteristic indicator set of the first device, and form a data set as a second sample based on the third characteristic element of the third characteristic indicator set and the maintenance object category; obtain the fourth characteristic indicator set of the faulty device; obtain the target distance based on the third characteristic indicator set and the fourth characteristic indicator set; obtain the maintenance object category corresponding to the third characteristic element with the smallest target distance, and assign the faulty device to the maintenance object category for maintenance.

7. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program is executed by a processor to implement the method according to any one of claims 1 to 5.

Images