A cluster operation status determination method, device and medium based on model
Through the model-based cluster operation state determination method, the Markov model and historical data standard deviation ellipse are used to predict the target device cluster state, which solves the problem of unpredictable future status in the existing technology and improves the timeliness of monitoring and accuracy reduces the risk of safety accidents.
Patent Information
- Application Number
- CN202410153101.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-01
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-02-01
Smart Images

Figure CN117972348B_ABST
Abstract
Description
Background Art
[0002] At present, the status monitoring of asset equipment is mainly achieved by collecting the operating parameters of each asset equipment, and comparing the collected operating parameters of each asset equipment with the preset parameter threshold in real time. If the collected operating parameters of the asset equipment are greater than the corresponding parameter threshold, it means that the asset equipment has an abnormality at a historical moment and needs to be maintained. However, this status monitoring method can only monitor asset equipment that has experienced abnormal conditions at a historical moment, and cannot predict the status of the asset equipment at a future moment. It has a lag and is prone to safety accidents. Summary of the invention
[0003] In view of the above technical problems, the technical solution adopted by the present invention is:
[0004] According to one aspect of the present application, a model-based cluster operation status determination method is provided, which is applied to a cluster operation status determination system, wherein the cluster operation status determination system is connected to a target device cluster, the target device cluster includes n target devices, and each target device corresponds to k operation parameter identifiers;
[0005] The cluster operation status determination method based on the model comprises the following steps:
[0006] Step S100: every first target time period, obtain a number of operating parameters corresponding to each operating parameter identifier of each target device in a second target time period; the end time of the second target time period is the current time; the length of the second target time period is greater than the length of the first target time period;
[0007] Step S200, determining an operation vector to be identified of a target device cluster in the second target time period according to a number of operation parameters of each target device in the second target time period;
[0008] Step S300, obtaining the observation value and hidden state value of the feature point corresponding to the last feature information in the running vector to be identified, and determining them as the target observation value and target hidden state value respectively; the target observation value is the feature information of the corresponding feature point; the target hidden state value is the direction and length of the connection between the corresponding feature point and the feature point before the feature point;
[0009] Step S400, repeatedly inputting the target observation value and the target hidden state value into the preset Markov model X times, and obtaining X predicted observation sequences output by the Markov model; the Markov model is obtained by the observation value and hidden state value of the feature point corresponding to each feature information in the running vector to be identified;
[0010] Step S500: input X prediction observation sequences into a preset historical data standard deviation ellipse, and determine the number of predictions Y of the Markov model by the distance from the prediction feature point corresponding to each prediction observation sequence to the focus of the historical data standard deviation ellipse; wherein Y ≥ X; the historical data standard deviation ellipse is obtained according to the operating parameters of the target device cluster when it is in a safe operating state during the historical period;
[0011] Step S600, repeatedly inputting the target observation value and the target hidden state value into the preset Markov model Y times, to obtain Y predicted observation sequences output by the preset Markov model;
[0012] Step S700, inputting Y prediction observation sequences into a preset historical data standard deviation ellipse, and determining the number Z of prediction observation sequences that exist outside the historical data standard deviation ellipse among the Y prediction observation sequences;
[0013] Step S800: If Z / Y is greater than a preset observation sequence ratio threshold, it is determined that the target device cluster is in an abnormal operation state within a third target time period; the start time of the third target time period is the end time of the second target time period.
[0014] In an exemplary embodiment of the present application, step S200 includes:
[0015] Step S201: Determine each average operating parameter of each target device in the second target time period according to the operating parameters corresponding to each operating parameter identifier of each target device in the second target time period, and obtain a target average operating parameter list set A=(A1, A2, ..., A i ,...,A n );A i =(A i1 ,A i2 ,...,A ij ,...,A ig );A ij =(A ij1 ,A ij2 ,...,A ijh ,...,A ijk ); where, i=1,2,...,n; A i is a list of average operating parameters of the i-th target device in the second target time period; j = 1, 2, ..., g; g is the number of sub-time periods in the second target time period, and the g sub-time periods are connected in sequence; A ij is a list of average operating parameters of the i-th target device in the j-th sub-time period of the second target time period; h = 1, 2, ..., k; A ijhis the hth average operating parameter of the i-th target device in the j-th sub-time period of the second target time period; A ijh The operation parameter corresponding to the h-th operation parameter identifier of the i-th target device in the j-th sub-time period of the second target time period is averaged to obtain the operation parameter;
[0016] Step S202: According to the target average operating parameter list set A, a target device risk value list set B is obtained for each target device in the second target time period = (B1, B2, ..., B i ,...,B n );B i =(B i1 ,B i2 ,...,B ij ,...,B ig );B ij =(∑ k h=1 A ijh ) / k; where B i is a list of device risk values of the i-th target device in the second target time period; B ij is the device risk value of the i-th target device in the j-th sub-time period of the second target time period;
[0017] Step S203: According to the target device risk value list set B, a target cluster risk value list D of the target device cluster in the second target time period is obtained, which is equal to (D1, D2, ..., D j ,...,D g );D j =(∑ n i=1 B ij ) / n; where D j is the cluster risk value of the target device cluster in the jth sub-time period of the second target time period;
[0018] Step S204: According to the target average operating parameter list set A and the target device risk value list set B, a target device fluctuation value list set C = (C1, C2, ..., C i ,...,C n );C i =(C i1 ,C i2 ,...,C ij ,...,C ig );C ij =(∑ k h=1 (A ijh -B ij ) 2) / k; where C i is a list of device fluctuation values of the i-th target device in the second target time period; C ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the second target time period;
[0019] Step S205: According to the target device fluctuation value list set C, a target cluster fluctuation value list E of the target device cluster in the second target time period is obtained = (E1, E2, ..., E j ,...,E g );E j =(∑ n i=1 C ij ) / n; where E j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the second target time period;
[0020] Step S206: Determine the target device cluster's operation vector to be identified in the second target time period according to the target cluster risk value list D and the target cluster fluctuation value list E. j ,...,F g );F j =(D j ,E j );F j It is the characteristic information corresponding to the target device cluster in the jth sub-time period of the second target time period.
[0021] In an exemplary embodiment of the present application, after step S200, the model-based cluster operation status determination method further includes:
[0022] Step S210, compare the operation vector to be identified with several preset target operation vectors. If the matching degree between the operation vector to be identified and each preset target operation vector is less than the preset matching degree threshold, execute step S300; otherwise, determine that the target device cluster is in a safe operation state within the third target time period; the target operation vector is obtained according to several operation parameters of the target device cluster within the historical safety period; the length of the historical safety period is equal to the length of the second target time period; and the target device cluster is in a safe operation state within the historical critical period; the start time of the historical critical period is the end time of the historical safety period; the length of the historical critical period is equal to the length of the third target time period.
[0023] In an exemplary embodiment of the present application, the target operation vector is determined by the following steps:
[0024] Step S211: Obtain each average operating parameter of each target device in the historical safety period, and obtain a historical average operating parameter list set G = (G1, G2, ..., G i ,...,G n );G i =(G i1 ,G i2 ,...,G ij ,...,G ig );G ij =(G ij1 ,G ij2 ,...,G ijh ,...,G ijk ), where G i is the average operating parameter list of the i-th target device in the historical safety period; G ij is a list of average operating parameters of the i-th target device in the j-th sub-period of the historical safety period; G ijh is the hth average operating parameter of the i-th target device in the j-th sub-time period of the historical safety period; the historical safety period includes g sub-time periods, and the length of the sub-time period included in the historical safety period is equal to the length of the sub-time period included in the second target time period;
[0025] Step S212: According to the historical average operating parameter list set G, obtain the historical equipment risk value list set H = (H1, H2, ..., H i ,...,H n );H i =(H i1 ,H i2 ,...,H ij ,...,H ig );H ij =(∑ k h=1 G ijh ) / k; where H i is the device risk value list of the i-th target device in the historical security period; H ij is the device risk value of the i-th target device in the j-th sub-period of the historical safety period;
[0026] Step S213: According to the historical equipment risk value list set H, obtain the historical cluster risk value list I of the target equipment cluster in the historical safety period = (I1, I2, ..., I j ,...,I g );I j =(∑ n i=1 H ij ) / n; where Ij is the cluster risk value of the target device cluster in the jth sub-period of the historical safety period;
[0027] Step S214: According to the historical average operating parameter list set G and the historical equipment risk value list set H, obtain the historical equipment fluctuation value list set J of each target equipment in the historical safety period = (J1, J2, ..., J i ,...,J n );J i =(J i1 ,J i2 ,...,J ij ,...,J ig );J ij =(∑ k h=1 (G ijh -H ij ) 2 ) / k; where J i is the device fluctuation value list of the i-th target device in the historical safety period; J ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the historical safety period;
[0028] Step S215: According to the historical equipment fluctuation value list set J, obtain the historical cluster fluctuation value list K of the target equipment cluster in the historical safety period = (K1, K2, ..., K j ,...,K g );K j =(∑ n i=1 J ij ) / n; where K j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the historical safety period;
[0029] Step S216: Determine the historical operation vector L of the target device cluster in the historical safety period according to the historical cluster risk value list I and the historical cluster fluctuation value list K. j ,...,L g );L j =(I j ,K j );L j The characteristic information corresponding to the target device cluster in the jth sub-time period of the historical security period;
[0030] Step S217: If the number of the historical safety period is one, the historical operation vector corresponding to the historical safety period is determined as the target operation vector.
[0031] In an exemplary embodiment of the present application, step S217 further includes:
[0032] Step S2171: If there are multiple historical safety periods, cluster the historical operation vectors corresponding to each historical safety period to obtain a number of historical operation vector clusters; each historical operation vector cluster includes a number of historical operation vectors;
[0033] Step S2172: average a number of historical operation vectors in each historical operation vector group to obtain a historical center vector corresponding to each historical operation vector group;
[0034] Step S2173: determine each historical center vector as the target operation vector.
[0035] In an exemplary embodiment of the present application, the Markov model is determined by the following steps:
[0036] Step S401: F1, F2, ..., F j ,...,F g The corresponding feature points are connected in sequence;
[0037] Step S402: F m-1 The corresponding feature points and F m The angle between the connecting line between the corresponding feature points and the horizontal coordinate axis and F m-1 The corresponding feature points and F m The length of the connecting line between the corresponding feature points is determined as F m The corresponding hidden state value; where m = 2, ..., g;
[0038] Step S403: m ,E m ) is determined as F m The corresponding observed value;
[0039] Step S404: input the observation value and hidden state value corresponding to each feature point into a preset mathematical model for training to obtain a Markov model.
[0040] In an exemplary embodiment of the present application, the historical data standard deviation ellipse is determined by the following steps:
[0041] Step S501, obtaining a number of historical operating parameters of each target device when it is in a safe operating state in each sub-time period of the historical period;
[0042] Step S502: obtaining, according to a number of historical operating parameters, a number of historical period equipment risk values and a number of historical period equipment fluctuation values of each target equipment in each sub-time period of the historical period;
[0043] Step S503: Obtain the historical period cluster risk value and historical period cluster fluctuation value of the target device cluster in each sub-period of the historical period according to the several historical period device risk values and several historical period device fluctuation values corresponding to each target device;
[0044] Step S504: taking the historical period cluster risk value and the historical period cluster volatility value corresponding to the same sub-period as the horizontal coordinate and vertical coordinate of a historical feature point respectively, to obtain a number of historical feature points; the number of historical feature points is equal to the number of sub-periods of the historical period;
[0045] Step S505: input each historical feature point into the standard deviation ellipse tool to obtain a standard deviation ellipse of historical data.
[0046] In an exemplary embodiment of the present application, step S500 includes:
[0047] Step S510, inputting X prediction observation sequences into a preset historical data standard deviation ellipse to obtain X prediction observation sequences output by a preset Markov model;
[0048] Step S520: Determine the sum of the distances from the predicted feature point corresponding to each predicted observation sequence to the two foci of the standard deviation ellipse of the historical data, and obtain the focal length and list P = (P1, P2, ..., P q ,...,P X ), where q = 1, 2, ..., X; P q It is the sum of the distances from the predicted feature point corresponding to the qth predicted observation sequence to the two foci of the standard deviation ellipse of the historical data;
[0049] Step S530: Determine F g The sum of the distances from the corresponding feature point to the two foci of the standard deviation ellipse of the historical data R g ;
[0050] Step S540, traverse the focal length and list P, if P q <R g , then P q Determine as the target focal length and, obtain u target focal lengths and;
[0051] Step S550: If u=X, then determine the prediction number Y=X of the Markov model; if u<X / a, then determine the prediction number Y=a×X of the Markov model; if u<X / b, then determine the prediction number Y=b×X of the Markov model; wherein 1<a<b; a and b are both preset coefficient factors.
[0052] According to one aspect of the present application, a non-transitory computer-readable storage medium is provided, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the aforementioned model-based cluster operation status determination method.
[0053] According to one aspect of the present application, an electronic device is provided, including a processor and the aforementioned non-transitory computer-readable storage medium.
[0054] The present invention has at least the following beneficial effects:
[0055] The present invention determines the target device cluster in the second target time period according to the several operating parameters corresponding to each operating parameter identification of each target device in the second target time period obtained every first target time period, determines the target device cluster in the second target time period to be identified operating vector according to the target observation value and target hidden state value corresponding to the last feature information in the operating vector to be identified, determines the number of predictions Y of the Markov model, repeatedly inputs the target observation value and the target hidden state value into the preset Markov model Y times, obtains Y predicted observation sequences output by the preset Markov model, inputs the Y predicted observation sequences into the preset historical data standard deviation ellipse, determines the number Z of predicted observation sequences that exist outside the historical data standard deviation ellipse in the Y predicted observation sequences, and if Z / Y is greater than the preset observation sequence ratio threshold, determines that the target device cluster is in an abnormal operating state in the third target time period. The Markov model is calculated by the target observation value and the target hidden state value corresponding to the target device cluster, and the corresponding number of predictions is determined to reduce the calculation amount of the Markov model, and the target observation value and the target hidden state value are calculated by the Markov model to predict the operating state of the target device cluster at a future moment. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.
[0057] Figure 1 A flowchart of a cluster operation status determination method based on a model provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0059] A cluster operation state determination method based on a model is applied to a cluster operation state determination system. The cluster operation state determination system is connected to a target device cluster, the target device cluster includes n target devices, and each target device corresponds to k operation parameter identifiers.
[0060] The several target devices included in the target device cluster can be production equipment or asset equipment for network communication, etc. Each operating parameter identifier corresponds to a unique type of operating parameter. The operating parameter identifier is used to indicate the type of operating parameter monitored in real time for the target device. The operating parameter can be the operating temperature of the target device, ambient humidity, number of data packets received, number of data packets sent, etc.
[0061] like Figure 1 As shown, the cluster operation status determination method based on the model proposed in the present invention comprises the following steps:
[0062] Step S100: every first target time period, obtain a number of operating parameters corresponding to each operating parameter identifier of each target device in a second target time period; the end time of the second target time period is the current time; the length of the second target time period is greater than the length of the first target time period;
[0063] In one embodiment of the present application, the first target time period can be set to ten minutes, and the second target time period can be set to the past three hours, that is, every ten minutes, the average operating parameter corresponding to each operating parameter identifier of each target device in the past three hours is obtained. The first target time period and the second target time period can be determined according to the number of monitored target devices in the target device cluster or the number of types of monitored operating parameters. The more complex the target device cluster is, the shorter the length of the corresponding target time period is set.
[0064] The number of sub-time periods within the second target time period can also be determined according to the frequency of the operating parameters that the user wants to obtain. If the user wants to collect operating parameters with a higher frequency, the number of sub-time periods will be increased. Conversely, the number of sub-time periods will be reduced. The length of each sub-time period within the second target time period is the same, that is, the second target time period is equally divided into several sub-time periods to ensure that the difference in the average operating parameters obtained is not too large.
[0065] Step S200, determining an operation vector to be identified of a target device cluster in the second target time period according to a number of operation parameters of each target device in the second target time period;
[0066] Further, the method for determining the operation vector to be identified includes steps S201 to S206:
[0067] Step S201: Determine each average operating parameter of each target device in the second target time period according to the operating parameters corresponding to each operating parameter identifier of each target device in the second target time period, and obtain a target average operating parameter list set A=(A1, A2, ..., A i ,...,A n );A i =(A i1 ,A i2 ,...,A ij ,...,A ig );A ij =(A ij1 ,A ij2 ,...,A ijh ,...,A ijk ); where, i=1,2,...,n; A i is a list of average operating parameters of the i-th target device in the second target time period; j = 1, 2, ..., g; g is the number of sub-time periods in the second target time period, and the g sub-time periods are connected in sequence; A ij is a list of average operating parameters of the i-th target device in the j-th sub-time period of the second target time period; h = 1, 2, ..., k; A ijh is the hth average operating parameter of the i-th target device in the j-th sub-time period of the second target time period; A ijh The operation parameter corresponding to the h-th operation parameter identifier of the i-th target device in the j-th sub-time period of the second target time period is averaged to obtain the operation parameter;
[0068] The average operating parameter is the average value of several operating parameters corresponding to each operating parameter identifier of the target device within the second target time period, that is, each operating parameter identifier corresponds to an average operating parameter, and the average operating parameter is the average value of several operating parameters of the same type.
[0069] Step S202: According to the target average operating parameter list set A, a target device risk value list set B is obtained for each target device in the second target time period = (B1, B2, ..., B i ,...,B n );B i =(B i1 ,B i2,...,B ij ,...,B ig );B ij =(∑ k h=1 A ijh ) / k; where B i is a list of device risk values of the i-th target device in the second target time period; B ij is the device risk value of the i-th target device in the j-th sub-time period of the second target time period;
[0070] Step S203: According to the target device risk value list set B, a target cluster risk value list D of the target device cluster in the second target time period is obtained, which is equal to (D1, D2, ..., D j ,...,D g );D j =(∑ n i=1 B ij ) / n; where D j is the cluster risk value of the target device cluster in the jth sub-time period of the second target time period;
[0071] Step S204: According to the target average operating parameter list set A and the target device risk value list set B, a target device fluctuation value list set C = (C1, C2, ..., C i ,...,C n );C i =(C i1 ,C i2 ,...,C ij ,...,C ig );C ij =(∑ k h=1 (A ijh -B ij ) 2 ) / k; where C i is a list of device fluctuation values of the i-th target device in the second target time period; C ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the second target time period;
[0072] Step S205: According to the target device fluctuation value list set C, a target cluster fluctuation value list E of the target device cluster in the second target time period is obtained = (E1, E2, ..., E j ,...,E g );E j =(∑ n i=1 C ij) / n; where E j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the second target time period;
[0073] The fluctuation value indicates the fluctuation of the operating parameters of the target device cluster in the corresponding time period, and determines whether the target device cluster is in a normal and stable operating state in the corresponding time period. The larger the fluctuation value, the more unstable the operation of the target device cluster in the corresponding time period.
[0074] The risk value indicates the risk level of the operation status of the target device cluster in the corresponding time period. The larger the risk value, the more likely the target device cluster is to have abnormal operation in the corresponding time period.
[0075] First, the target device risk value and the target device fluctuation value of each target device in the corresponding time period are determined, and then the target cluster risk value and the target cluster fluctuation value of the target device cluster in the corresponding time period are determined.
[0076] Step S206: Determine the target device cluster's operation vector to be identified in the second target time period according to the target cluster risk value list D and the target cluster fluctuation value list E. j ,...,F g );F j =(D j ,E j );F j It is the characteristic information corresponding to the target device cluster in the jth sub-time period of the second target time period.
[0077] Each feature information of the operation vector to be identified corresponding to the target device cluster can be represented as a two-dimensional coordinate point, the horizontal coordinate of the two-dimensional coordinate point is the target cluster risk value of the target device cluster in the corresponding time period, and the vertical coordinate of the two-dimensional coordinate point is the target cluster fluctuation value of the target device cluster in the corresponding time period.
[0078] Through the operation vector to be identified, the target cluster risk value and the target cluster fluctuation value of the target device cluster in each sub-time period of the corresponding time period can be obtained.
[0079] Step S210, comparing the operation vector to be identified with a plurality of preset target operation vectors, if the matching degree between the operation vector to be identified and each preset target operation vector is less than a preset matching degree threshold, executing step S300;
[0080] If the matching degree between the operation vector to be identified and any preset target operation vector is greater than or equal to a preset matching degree threshold, it is determined that the target device cluster is in a safe operation state within the third target time period;
[0081] Among them, the target operation vector is obtained according to several operation parameters of the target device cluster during the historical safety period; the length of the historical safety period is equal to the length of the second target time period; and the target device cluster is in a safe operation state during the historical critical period; the start time of the historical critical period is the end time of the historical safety period; the start time of the third target time period is the end time of the second target time period, and the length of the third target time period is equal to the length of the historical critical period.
[0082] In order to ensure that the matching degree between the running vector to be identified and the target running vector is more accurate, the length of the historical safety period is set to be equal to the length of the second target time period, and the number of sub-time periods in the historical safety period is equal to the number of sub-time periods in the second target time period, so as to ensure that the dimension of each target running vector is equal to the dimension of the running vector to be identified, so as to facilitate comparison between the two.
[0083] In addition, there are specific rules for limiting the historical safety period, that is, the target device cluster is in a safe operating state during the historical critical period after the historical safety period, so as to facilitate the subsequent prediction of the operating state of the target device cluster at a future moment. The future moment is the third target time period, and the length of the third target time period is equal to the length of the historical critical period. The historical safety period and the historical critical period are both historical moments.
[0084] Through the operating parameters of the target device cluster at historical moments, several target operating vectors are obtained, and then the operating vector to be identified is compared with several target operating vectors. If the matching degree between the operating vector to be identified and any target operating vector is greater than or equal to a preset matching degree threshold, it means that the operating state of the target device cluster in the second target time period is similar to the operating state of the target device cluster in the historical safety period corresponding to the similar target operating vector. Since the target device cluster is in a safe operating state in the historical critical time period after the historical safety period, it can be determined that the target device cluster is also in a safe operating state in the third target time period after the second target time period, so as to predict the operating state of the target device cluster at future moments.
[0085] If the matching degree between the operation vector to be identified and each target operation vector is less than the preset matching degree threshold, it means that the operation status of the target device cluster in the second target time period is different from the operation status of the target device cluster in the historical safety period corresponding to each target operation vector, and it cannot be determined whether the operation status of the target device cluster in the third target time period is normal. Further judgment is required, that is, executing step S300 to determine the operation status of the target device cluster in the third target time period.
[0086] Further, the target operation vector is determined through steps S211 to S2173:
[0087] Step S211: Obtain each average operating parameter of each target device in the historical safety period, and obtain a historical average operating parameter list set G = (G1, G2, ..., G i ,...,G n );G i =(G i1 ,G i2 ,...,G ij ,...,G ig );G ij =(G ij1 ,G ij2 ,...,G ijh ,...,G ijk ), where G i is the average operating parameter list of the i-th target device in the historical safety period; G ij is a list of average operating parameters of the i-th target device in the j-th sub-period of the historical safety period; G ijh is the hth average operating parameter of the i-th target device in the j-th sub-time period of the historical safety period; the historical safety period includes g sub-time periods, and the length of the sub-time period included in the historical safety period is equal to the length of the sub-time period included in the second target time period;
[0088] Step S212: According to the historical average operating parameter list set G, obtain the historical equipment risk value list set H = (H1, H2, ..., H i ,...,H n );H i =(H i1 ,H i2 ,...,H ij ,...,H ig );H ij =(∑ k h=1 G ijh ) / k; where H i is the device risk value list of the i-th target device in the historical security period; H ij is the device risk value of the i-th target device in the j-th sub-period of the historical safety period;
[0089] Step S213: According to the historical equipment risk value list set H, obtain the historical cluster risk value list I of the target equipment cluster in the historical safety period = (I1, I2, ..., I j ,...,I g );I j =(∑ n i=1 Hij ) / n; where I j is the cluster risk value of the target device cluster in the jth sub-period of the historical safety period;
[0090] Step S214: According to the historical average operating parameter list set G and the historical equipment risk value list set H, obtain the historical equipment fluctuation value list set J of each target equipment in the historical safety period = (J1, J2, ..., J i ,...,J n );J i =(J i1 ,J i2 ,...,J ij ,...,J ig );J ij =(∑ k h=1 (G ijh -H ij ) 2 ) / k; where J i is the device fluctuation value list of the i-th target device in the historical safety period; J ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the historical safety period;
[0091] Step S215: According to the historical equipment fluctuation value list set J, obtain the historical cluster fluctuation value list K of the target equipment cluster in the historical safety period = (K1, K2, ..., K j ,...,K g );K j =(∑ n i=1 J ij ) / n; where K j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the historical safety period;
[0092] Step S216: Determine the historical operation vector L of the target device cluster in the historical safety period according to the historical cluster risk value list I and the historical cluster fluctuation value list K. j ,...,L g );L j =(I j ,K j );L j The characteristic information corresponding to the target device cluster in the jth sub-time period of the historical security period;
[0093] Step S217: If the number of historical safety periods is one, determine the historical operation vector corresponding to the historical safety period as the target operation vector;
[0094] Step S2171: If there are multiple historical safety periods, cluster the historical operation vectors corresponding to each historical safety period to obtain a number of historical operation vector clusters; each historical operation vector cluster includes a number of historical operation vectors;
[0095] Step S2172: average a number of historical operation vectors in each historical operation vector group to obtain a historical center vector corresponding to each historical operation vector group;
[0096] Step S2173: determine each historical center vector as the target operation vector.
[0097] The method for determining the target operation vector is the same as the method for determining the operation vector to be identified, except that the time period of the collected operation parameters is different.
[0098] Step S300, obtaining the observation value and hidden state value of the feature point corresponding to the last feature information in the running vector to be identified, and determining them as the target observation value and target hidden state value respectively;
[0099] The target observation value is the feature information of the corresponding feature point (i.e., the two-dimensional coordinate point corresponding to the feature information mentioned in step S206), and the target hidden state value is the direction and length of the line between the corresponding feature point and the feature point before the feature point (i.e., connecting every two adjacent feature points, the angle between the line and the horizontal direction and the length of the line segment are the target hidden state value corresponding to the next feature point).
[0100] Step S400, repeatedly inputting the target observation value and the target hidden state value into the preset Markov model X times to obtain X predicted observation sequences output by the Markov model;
[0101] The Markov model is obtained by the observed value and hidden state value of the feature point corresponding to each feature information in the running vector to be identified. The observed value and hidden state value are factors for determining the Markov model. The method for determining the Markov model is a prior art and will not be repeated here.
[0102] The predicted observation sequence output by the Markov model represents the moving direction and moving distance corresponding to the next feature point of the target device cluster. The two-dimensional coordinate point (feature information) of the next feature point can be determined through the moving direction and moving distance.
[0103] Among them, an embodiment of the method for determining the Markov model is the method described in step S401 to step S404:
[0104] Step S401: F1, F2, ..., F j ,...,F gThe corresponding feature points are connected in sequence;
[0105] Step S402: F m-1 The corresponding feature points and F m The angle between the connecting line between the corresponding feature points and the horizontal coordinate axis and F m-1 The corresponding feature points and F m The length of the connecting line between the corresponding feature points is determined as F m The corresponding hidden state value; where m = 2, ..., g;
[0106] Step S403: m ,E m ) is determined as F m The corresponding observed value;
[0107] Step S404: input the observation value and hidden state value corresponding to each feature point into a preset mathematical model for training to obtain a Markov model.
[0108] Step S500: input X prediction observation sequences into a preset historical data standard deviation ellipse, and determine the number of predictions Y of the Markov model by the distance from the prediction feature point corresponding to each prediction observation sequence to the focus of the historical data standard deviation ellipse; wherein Y ≥ X; the historical data standard deviation ellipse is obtained according to the operating parameters of the target device cluster when it is in a safe operating state during the historical period;
[0109] The historical data standard deviation ellipse is obtained based on the operating parameters of the target device cluster when it is in a safe operating state during the historical period.
[0110] The historical data standard deviation ellipse is a standard deviation ellipse obtained based on several safety operation parameters of the target device cluster in the historical period. The historical period can be determined according to user needs. There is no limited relationship between the historical period and the second target time period and the historical safety period. The two may overlap or not. The method of determining the standard deviation ellipse through data is an existing technology and will not be repeated here.
[0111] Further, step S500 includes:
[0112] Step S510, inputting X prediction observation sequences into a preset historical data standard deviation ellipse to obtain X prediction observation sequences output by a preset Markov model;
[0113] Step S520: Determine the sum of the distances from the predicted feature point corresponding to each predicted observation sequence to the two foci of the standard deviation ellipse of the historical data, and obtain the focal length and list P = (P1, P2, ..., P q ,...,P X ), where q = 1, 2, ..., X; Pq It is the sum of the distances from the predicted feature point corresponding to the qth predicted observation sequence to the two foci of the standard deviation ellipse of the historical data;
[0114] Step S530: Determine F g The sum of the distances from the corresponding feature point to the two foci of the standard deviation ellipse of the historical data R g ;
[0115] Step S540, traverse the focal length and list P, if P q <R g , then P q Determine as the target focal length and, obtain u target focal lengths and;
[0116] If P q <R g , then it means that the predicted feature point corresponding to the qth predicted observation sequence is g The corresponding feature points may exist in the same historical data standard deviation ellipse, or the movement trends of the two points are still moving in the same historical data standard deviation ellipse.
[0117] Step S550: If u=X, then determine the prediction number Y=X of the Markov model; if u<X / a, then determine the prediction number Y=a×X of the Markov model; if u<X / b, then determine the prediction number Y=b×X of the Markov model; wherein 1<a<b; a and b are both preset coefficient factors.
[0118] The number of predictions of the Markov model is determined by using different numbers of target focal lengths and corresponding prediction feature points. The larger the number of target focal lengths and corresponding prediction feature points, the more likely it is that the target focal lengths and corresponding prediction feature points are moving within the standard deviation ellipse of the historical data, indicating that the target device cluster will be in a normal operating state in the future, and there is no need for additional multiple predictions. Conversely, the smaller the number of target focal lengths and corresponding prediction feature points, the more likely it is that the target focal lengths and corresponding prediction feature points are moving outside the standard deviation ellipse of the historical data, indicating that the target device cluster is likely to be in an abnormal operating state in the future. Therefore, in order to further improve the accuracy of the prediction, it is necessary to increase the number of predictions to determine whether the target device cluster will be in an abnormal operating state in the future.
[0119] The historical data standard deviation ellipse is determined through steps S501 to S505:
[0120] Step S501, obtaining a number of historical operating parameters of each target device when it is in a safe operating state in each sub-time period of the historical period;
[0121] Step S502: obtaining, according to a number of historical operating parameters, a number of historical period equipment risk values and a number of historical period equipment fluctuation values of each target equipment in each sub-time period of the historical period;
[0122] Step S503: Obtain the historical period cluster risk value and historical period cluster fluctuation value of the target device cluster in each sub-period of the historical period according to the several historical period device risk values and several historical period device fluctuation values corresponding to each target device;
[0123] Step S504: taking the historical period cluster risk value and the historical period cluster volatility value corresponding to the same sub-period as the horizontal coordinate and vertical coordinate of a historical feature point respectively, to obtain a number of historical feature points; the number of historical feature points is equal to the number of sub-periods of the historical period;
[0124] Step S505: input each historical feature point into the standard deviation ellipse tool to obtain a standard deviation ellipse of historical data.
[0125] Step S600, repeatedly inputting the target observation value and the target hidden state value into the preset Markov model Y times, to obtain Y predicted observation sequences output by the preset Markov model;
[0126] Step S700, inputting Y prediction observation sequences into a preset historical data standard deviation ellipse, and determining the number Z of prediction observation sequences that exist outside the historical data standard deviation ellipse among the Y prediction observation sequences;
[0127] Since the historical data standard deviation ellipse is obtained based on the safe operation parameters of the target device cluster, it means that the feature points within the historical data standard deviation ellipse are all feature points of the target device cluster in a safe operation state. The predicted observation sequence is input into the historical data standard deviation ellipse. If the feature points determined by the predicted observation sequence are within the historical data standard deviation ellipse, it means that the feature points determined by the predicted observation sequence are feature points of the target device cluster in a safe operation state at a historical moment. Conversely, if the feature points determined by the predicted observation sequence are outside the historical data standard deviation ellipse, it means that the feature points determined by the predicted observation sequence are not feature points of the target device cluster in a safe operation state at a historical moment.
[0128] Step S800: If Z / Y is greater than a preset observation sequence ratio threshold, it is determined that the target device cluster is in an abnormal operation state within a third target time period.
[0129] If Z / Y is greater than the preset observation sequence ratio threshold, it means that among the predicted Y feature points, the proportion of feature points existing outside the standard deviation ellipse of historical data is relatively large, indicating that the next feature point of the target device cluster is likely to be outside the standard deviation ellipse of historical data. If it is outside the standard deviation ellipse of historical data, it means that the target device cluster is in an abnormal operating state, and it is determined that the target device cluster is in an abnormal operating state within the third target time period.
[0130] The present invention determines the target device cluster in the second target time period according to the several operating parameters corresponding to each operating parameter identification of each target device in the second target time period obtained every first target time period, determines the target device cluster in the second target time period to be identified operating vector according to the target observation value and target hidden state value corresponding to the last feature information in the operating vector to be identified, determines the number of predictions Y of the Markov model, repeatedly inputs the target observation value and the target hidden state value into the preset Markov model Y times, obtains Y predicted observation sequences output by the preset Markov model, inputs the Y predicted observation sequences into the preset historical data standard deviation ellipse, determines the number Z of predicted observation sequences that exist outside the historical data standard deviation ellipse in the Y predicted observation sequences, and if Z / Y is greater than the preset observation sequence ratio threshold, determines that the target device cluster is in an abnormal operating state in the third target time period. The Markov model is calculated by the target observation value and the target hidden state value corresponding to the target device cluster, and the corresponding number of predictions is determined to reduce the calculation amount of the Markov model, and the target observation value and the target hidden state value are calculated by the Markov model to predict the operating state of the target device cluster at a future moment.
[0131] An embodiment of the present invention further provides a computer program product, which includes program code. When the program product is run on an electronic device, the program code is used to enable the electronic device to execute the steps of the method according to various exemplary embodiments of the present invention described above in this specification.
[0132] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.
[0133] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.
[0134] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
[0135] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as a system, method or program product. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as a "circuit", "module" or "system".
[0136] The electronic device according to this embodiment of the present invention is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0137] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the at least one processor mentioned above, the at least one storage device mentioned above, and a bus connecting different system components (including storage devices and processors).
[0138] The storage stores program codes, which can be executed by the processor, so that the processor executes the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification.
[0139] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read only memory (ROM).
[0140] The storage may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.
[0141] The bus may represent one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.
[0142] The electronic device can also communicate with one or more external devices (such as keyboards, pointing devices, Bluetooth devices, etc.), and can also communicate with one or more devices that enable users to interact with the electronic device, and / or communicate with any device (such as routers, modems, etc.) that enables the electronic device to communicate with one or more other computing devices. This communication can be carried out through an input / output (I / O) interface. In addition, the electronic device can also communicate with one or more networks (such as local area networks (LANs), wide area networks (WANs) and / or public networks, such as the Internet) through a network adapter. As shown in the figure, the network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.
[0143] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.
[0144] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.
[0145] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0146] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0147] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
[0148] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).
[0149] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.
[0150] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.
[0151] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims
1. A cluster operation status determination method based on a model, characterized in that: Applied to a cluster operation status determination system, the cluster operation status determination system is connected to a target device cluster, the target device cluster includes n target devices, and each of the target devices corresponds to k operation parameter identifiers; The method comprises the following steps: Step S100: every first target time period, obtain a number of operating parameters corresponding to each operating parameter identifier of each target device in a second target time period; the end time of the second target time period is the current time; the length of the second target time period is greater than the length of the first target time period; Step S200, determining an operation vector to be identified of the target device cluster in the second target time period according to a plurality of operation parameters of each of the target devices in the second target time period; Step S300, obtaining the observation value and hidden state value of the feature point corresponding to the last feature information in the to-be-identified running vector, and determining them as the target observation value and target hidden state value respectively; the target observation value is the feature information of the corresponding feature point; the target hidden state value is the direction and length of the connection between the corresponding feature point and the feature point before the feature point; Step S400, repeatedly inputting the target observation value and the target hidden state value into a preset Markov model X times to obtain X predicted observation sequences output by the Markov model; the Markov model is obtained by the observation value and hidden state value of the feature point corresponding to each feature information in the running vector to be identified; Step S500, inputting X prediction observation sequences into a preset historical data standard deviation ellipse, and determining the prediction number Y of the Markov model by the distance from the prediction feature point corresponding to each prediction observation sequence to the focus of the historical data standard deviation ellipse; wherein Y≥X; the historical data standard deviation ellipse is obtained according to the operating parameters of the target device cluster when it is in a safe operating state during a historical period; Step S600, repeatedly inputting the target observation value and the target hidden state value into a preset Markov model Y times to obtain Y predicted observation sequences output by the preset Markov model; Step S700, inputting Y prediction observation sequences into a preset historical data standard deviation ellipse, and determining the number Z of prediction observation sequences that are outside the historical data standard deviation ellipse among the Y prediction observation sequences; Step S800: If Z / Y is greater than a preset observation sequence ratio threshold, it is determined that the target device cluster is in an abnormal operation state within a third target time period; the start time of the third target time period is the end time of the second target time period; Wherein, the step S200 includes: Step S201: Determine each average operating parameter of each target device in the second target time period according to the operating parameters corresponding to each operating parameter identifier of each target device in the second target time period, and obtain a target average operating parameter list set A=(A1, A2, ..., A i ,...,A n );A i =(A i1 ,A i2 ,...,A ij ,...,A ig );A ij =(A ij1 ,A ij2 ,...,A ijh ,...,A ijk ); where, i=1,2,...,n; A i is a list of average operating parameters of the i-th target device in the second target time period; j=1,2,...,g; g is the number of sub-time periods in the second target time period, and the g sub-time periods are connected in sequence; A ij is a list of average operating parameters of the i-th target device in the j-th sub-time period of the second target time period; h=1,2,...,k; A ijh is the hth average operating parameter of the i-th target device in the j-th sub-time period of the second target time period; A ijh The operation parameter corresponding to the h-th operation parameter identifier of the i-th target device in the j-th sub-time period of the second target time period is averaged to obtain the operation parameter; Step S202: According to the target average operating parameter list set A, a target device risk value list set B = (B1, B2, ..., B i ,...,B n );B i =(B i1 ,B i2 ,...,B ij ,...,B ig );B ij =(∑ k h=1 A ijh ) / k; where B i is a list of device risk values of the i-th target device in the second target time period; B ij is the device risk value of the i-th target device in the j-th sub-time period of the second target time period; Step S203: According to the target device risk value list set B, a target cluster risk value list D=(D1, D2, ..., D j ,...,D g );D j =(∑ n i=1 B ij ) / n; where D j is the cluster risk value of the target device cluster in the jth sub-time period of the second target time period; Step S204: According to the target average operating parameter list set A and the target device risk value list set B, a target device fluctuation value list set C = (C1, C2, ..., C i ,...,C n );C i =(C i1 ,C i2 ,...,C ij ,...,C ig );C ij =(∑ k h=1 (A ijh -B ij ) 2 ) / k; where C i is a list of device fluctuation values of the i-th target device in the second target time period; C ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the second target time period; Step S205: According to the target device fluctuation value list set C, a target cluster fluctuation value list E=(E1, E2, ..., E j ,...,E g );E j =(∑ n i=1 C ij ) / n; where E j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the second target time period; Step S206: Determine the target device cluster's operation vector to be identified within the second target time period according to the target cluster risk value list D and the target cluster fluctuation value list E. j ,...,F g );F j =(D j ,E j );F j It is the characteristic information corresponding to the target device cluster in the jth sub-time period of the second target time period.
2. The method according to claim 1, characterized in that After step S200, the method further includes: Step S210, compare the operation vector to be identified with several preset target operation vectors. If the matching degree between the operation vector to be identified and each preset target operation vector is less than the preset matching degree threshold, execute step S300; otherwise, determine that the target device cluster is in a safe operation state within the third target time period; the target operation vector is obtained according to several operation parameters of the target device cluster within the historical safety time period; the length of the historical safety time period is equal to the length of the second target time period; and the target device cluster is in a safe operation state within the historical critical time period; the start time of the historical critical time period is the end time of the historical safety time period; the length of the historical critical time period is equal to the length of the third target time period.
3. The method according to claim 2, characterized in that The target operation vector is determined by the following steps: Step S211: Obtain each average operating parameter of each target device in the historical security period, and obtain a historical average operating parameter list set G=(G1, G2, ..., G i ,...,G n );G i =(G i1 ,G i2 ,...,G ij ,...,G ig );G ij =(G ij1 ,G ij2 ,...,G ijh ,...,G ijk ), where G i is a list of average operating parameters of the i-th target device in the historical safety period; G ij is a list of average operating parameters of the i-th target device in the j-th sub-time period of the historical safety period; G ijh is the hth average operating parameter of the i-th target device in the j-th sub-time period of the historical safety period; the historical safety period includes g sub-time periods, and the length of the sub-time period included in the historical safety period is equal to the length of the sub-time period included in the second target time period; Step S212: According to the historical average operating parameter list set G, obtain the historical device risk value list set H = (H1, H2, ..., H i ,...,H n );H i =(H i1 ,H i2 ,...,H ij ,...,H ig );H ij =(∑ k h=1 G ijh ) / k; where H i is a list of device risk values of the i-th target device in the historical security period; H ij is the device risk value of the i-th target device in the j-th sub-time period of the historical safety period; Step S213: According to the historical device risk value list set H, obtain the historical cluster risk value list I=(I1, I2, ..., I j ,...,I g );I j =(∑ n i=1 H ij ) / n; where I j is the cluster risk value of the target device cluster in the jth sub-time period of the historical safety period; Step S214: According to the historical average operating parameter list set G and the historical equipment risk value list set H, obtain the historical equipment fluctuation value list set J=(J1, J2, ..., J i ,...,J n );J i =(J i1 ,J i2 ,...,J ij ,...,J ig );J ij =(∑ k h=1 (G ijh -H ij ) 2 ) / k; where J i is a list of device fluctuation values of the i-th target device in the historical safety period; J ij is the device fluctuation value of the i-th target device in the j-th sub-time period of the historical security period; Step S215: According to the historical equipment fluctuation value list set J, obtain the historical cluster fluctuation value list K=(K1, K2, ..., K j ,...,K g );K j =(∑ n i=1 J ij ) / n; where K j is the cluster fluctuation value of the target device cluster in the jth sub-time period of the historical safety period; Step S216: Determine the historical operation vector L=(L1, L2, ..., L j ,...,L g );L j =(I j ,K j );L j The characteristic information corresponding to the target device cluster in the jth sub-time period of the historical security period; Step S217: If the number of the historical safety period is one, determine the historical operation vector corresponding to the historical safety period as the target operation vector.
4. The method according to claim 3, characterized in that The step S217 further includes: Step S2171: If there are multiple historical safety periods, cluster the historical operation vectors corresponding to each historical safety period to obtain a plurality of historical operation vector clusters; each of the historical operation vector clusters includes a plurality of historical operation vectors; Step S2172: average a plurality of historical operation vectors in each of the historical operation vector clusters to obtain a historical center vector corresponding to each of the historical operation vector clusters; Step S2173: determine each of the historical center vectors as a target operation vector.
5. The method according to claim 4, characterized in that The Markov model is determined by the following steps: Step S401: F1, F2, ..., F j ,...,F g The corresponding feature points are connected in sequence; Step S402: F m-1 The corresponding feature points and F m The angle between the connecting line between the corresponding feature points and the horizontal coordinate axis and F m-1 The corresponding feature points and F m The length of the connecting line between the corresponding feature points is determined as F m The corresponding hidden state value; where m=2,...,g; Step S403: m ,E m ) is determined as F m The corresponding observed value; Step S404: input the observation value and hidden state value corresponding to each feature point into a preset mathematical model for training to obtain the Markov model.
6. The method according to claim 5, characterized in that The historical data standard deviation ellipse is determined by the following steps: Step S501, obtaining a number of historical operating parameters of each target device when it is in a safe operating state in each sub-time period of the historical period; Step S502: obtaining, according to the historical operating parameters, a plurality of historical period equipment risk values and a plurality of historical period equipment fluctuation values of each target equipment in each sub-period of the historical period; Step S503: Obtain the historical period cluster risk value and historical period cluster fluctuation value of the target device cluster in each sub-period of the historical period according to the several historical period device risk values and several historical period device fluctuation values corresponding to each of the target devices; Step S504: taking the historical period cluster risk value and the historical period cluster volatility value corresponding to the same sub-period as the horizontal coordinate and vertical coordinate of a historical feature point respectively, to obtain a number of historical feature points; the number of the historical feature points is equal to the number of sub-periods of the historical period; Step S505: input each of the historical feature points into a standard deviation ellipse tool to obtain a historical data standard deviation ellipse.
7. The method according to claim 6, characterized in that The step S500 includes: Step S510, inputting X prediction observation sequences into a preset historical data standard deviation ellipse to obtain X prediction observation sequences output by a preset Markov model; Step S520: Determine the sum of the distances from the predicted feature point corresponding to each predicted observation sequence to the two foci of the standard deviation ellipse of the historical data, and obtain the focal length and list P=(P1, P2, ..., P q ,...,P X ), where q=1,2,...,X; P q is the sum of the distances from the predicted feature point corresponding to the qth predicted observation sequence to the two foci of the standard deviation ellipse of the historical data; Step S530: Determine F g The sum of the distances R from the corresponding feature point to the two foci of the standard deviation ellipse of the historical data g ; Step S540, traverse the focal length and list P, if P q <R g , then P q Determine as the target focal length and, obtain u target focal lengths and; Step S550: If u=X, then determine the prediction number Y=X of the Markov model; if u<X / a, then determine the prediction number Y=a×X of the Markov model; if u<X / b, then determine the prediction number Y=b×X of the Markov model; wherein 1<a<b; a and b are both preset coefficient factors.
8. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the method according to any one of claims 1 to 7.
9. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 8.
