Transient identification and classification method, system, equipment and medium

Abstract

The invention relates to a transient identification and classification method, system and device and a medium. The method comprises the following steps: acquiring an operation parameter sequence of a component within preset sampling time; according to a preset amplitude change speed interval, extracting actual transient data from the operation parameter sequence, and calculating transient representation data; and according to the transient representation data, determining that the actual transient data is a classifiable transient, respectively calculating similarity values between the actual transient data and a plurality of reference transients, and taking a transient type of the reference transient with the highest similarity value as a classification result of the component. According to the method, the transient data can be identified and classified in a more accurate and efficient manner, the identification accuracy and efficiency are remarkably improved, and the labor investment is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of unit transient monitoring, and particularly relates to a transient identification and classification method, system, device and medium. BACKGROUND

[0002] In the field of nuclear power plant transient management, the existing technology mainly has the following problems: on the one hand, CPR and other nuclear power plants usually use manual identification and classification to manage transients, which consumes a lot of manpower and material resources. With the increase of operation data, the problem of low efficiency is increasingly evident, and manual identification is prone to errors and difficult to accurately remember all transient categories and corresponding thresholds. On the other hand, the existing intelligent classification technology focuses on classification based on unit state characteristics, but from the perspective of parameter envelope that leads to stress effect of important components, the rationality of this classification method is insufficient, and it cannot accurately evaluate the influence of actual transients of the power plant on fatigue accumulation effect of important components of the unit. Therefore, a transient identification and classification method, system, device and medium are needed. SUMMARY

[0003] In view of the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide a transient identification and classification method, system, device and medium, which is used to solve the problem that the existing technology CPR and other nuclear power plants rely too much on technical personnel to manually identify and classify transients, which consumes a lot of manpower and material resources. Moreover, with the increase of operation data, manual identification is inefficient and prone to errors, which is difficult to meet the growing demand for intelligent statistics of transients. The existing intelligent classification technology focuses on classification according to unit state characteristics, rather than from the perspective of key parameters such as transient process temperature and pressure change that lead to stress effect of important components. Therefore, the rationality and accuracy of its classification are questionable, and it cannot accurately identify and classify actual transients, which is difficult to meet the high-precision requirements of nuclear power plants for transient management.

[0004] To achieve the above-mentioned purposes and other related purposes, the present application provides a transient identification and classification method, system, device and medium, which is applied to the field of unit transient monitoring, and obtains a running parameter sequence of a component within a preset sampling time; according to a preset amplitude change speed interval, extracts actual transient data from the running parameter sequence, and calculates transient characteristic data; according to the transient characteristic data, determines that the actual transient data is a classifiable transient, and respectively calculates similarity values of the actual transient data and a plurality of reference transients, and takes a transient type of the reference transient with the highest similarity value as a classification result of the component.

[0005] In an embodiment of the present application, when the operation parameter sequence is one, the step of extracting actual transient data from the operation parameter sequence and calculating transient characteristic data according to preset amplitude variation speed intervals comprises: dividing the operation parameter sequence into a plurality of parameter subsequences according to amplitude variation speed; selecting parameter subsequences with amplitude variation speed beyond the amplitude variation speed intervals as actual transient data; and calculating the maximum parameter variation amplitude and the maximum parameter variation rate of the actual transient data as the transient characteristic data.

[0006] In an embodiment of the present application, the step of determining the actual transient data as classifiable transient according to the transient characteristic data, calculating similarity values of the actual transient data and a plurality of reference transients respectively, and taking the transient type of the reference transient with the highest similarity value as the classification result of the component comprises: comparing the transient characteristic data with a plurality of preset reference transients to determine the type of the actual transient data as classifiable transient; calculating distance values of the actual transient data and each reference transient according to a DTW (Dynamic Time Warping) algorithm, and obtaining corresponding similarity values according to the distance values; screening the reference transient with the highest similarity value; and taking the transient type corresponding to the screened reference transient as the classification result of the component.

[0007] In an embodiment of the present application, the similarity value is obtained according to the distance value by the following formula: wherein η is the similarity value and d is the distance value.

[0008] In an embodiment of the present application, when the operation parameter sequence is a plurality of sequences, the step of extracting actual transient data from the operation parameter sequence and calculating transient characteristic data according to preset amplitude variation speed intervals comprises: dividing each operation parameter sequence into a plurality of parameter subsequences according to the corresponding amplitude variation speed; screening a plurality of segmentation sequences from the plurality of parameter subsequences according to the amplitude variation speed intervals corresponding to each operation parameter sequence; taking the earliest sampling time in all segmentation sequences as a segmentation starting point; judging all segmentation sequences according to sampling time sequence to determine a segmentation ending point; when the segmentation sequence with the earliest sampling time has no intersection with other segmentation sequences, taking the sampling ending point of the segmentation sequence with the earliest sampling time as the segmentation ending point; when the segmentation sequence with the earliest sampling time has intersection with other segmentation sequences, taking the sampling ending point of the last segmentation sequence with intersection with only one segmentation sequence as the segmentation ending point; extracting corresponding data from each operation parameter sequence according to the segmentation starting point and the segmentation ending point as actual transient data corresponding to each operation parameter sequence; and calculating the maximum parameter variation amplitude and the maximum parameter variation rate of each actual transient data as the transient characteristic data corresponding to each actual transient data.

[0009] In an embodiment of the present application, the step of determining the actual transient data as classifiable transient according to the transient characteristic data, and respectively calculating similarity values of the actual transient data and a plurality of reference transients, and taking the transient type of the reference transient with the highest similarity value as the classification result of the component comprises: comparing the transient characteristic data with a plurality of preset reference transients to determine the type of the actual transient data as classifiable transient; for each reference transient: respectively calculating sub-distance values of each actual transient data and corresponding reference transient data of the reference transient according to the DTW algorithm; converting each sub-distance value into a corresponding sub-similarity value; performing weighted operation on all the sub-similarity values to obtain the similarity values of the reference transient and a plurality of the actual transient data; and taking the transient type corresponding to the reference transient with the highest similarity value as the classification result of the component.

[0010] In an embodiment of the present application, the step of comparing the transient characteristic data with a plurality of preset reference transients to determine the type of the actual transient data as classifiable transient comprises: comparing the transient characteristic data with corresponding reference transient characteristic data in each of the reference transients; and when the transient characteristic data is less than all the reference transient characteristic data, determining the type of the corresponding actual transient data as classifiable transient.

[0011] The present application also provides a transient identification and classification system, which comprises: a data acquisition module configured to acquire a sequence of operating parameters of a component within a preset sampling time; a transient data determination and characteristic calculation module configured to extract actual transient data from the sequence of operating parameters according to a preset amplitude variation speed interval, and calculate transient characteristic data; and a classification result output module configured to determine the actual transient data as classifiable transient according to the transient characteristic data, respectively calculate similarity values of the actual transient data and a plurality of reference transients, and take the transient type of the reference transient with the highest similarity value as the classification result of the component.

[0012] The present application also provides an electronic device, which comprises: one or more processors; and a storage device configured to store one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the transient identification and classification method as described above.

[0013] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor of a computer, causes the computer to perform the transient identification and classification method as described above.

[0014] The transient identification and classification method, system, device and medium provided by the present application have the following beneficial effects: automatic identification of transient data is realized, manual intervention is reduced, errors caused by manual judgment negligence are avoided, and the speed of transient identification and classification is improved. The similarity between the actual transient data and the reference transient is calculated by the DTW algorithm, which can effectively process transient data sequences of different lengths and improve the accuracy of classification, thereby providing a decision basis for transient classification. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A flowchart of a transient identification and classification method provided by an embodiment of the present application is shown.

[0016] Figure 2 A structural block diagram of a transient identification and classification system provided by an embodiment of the present application is shown.

[0017] Figure 3 A statistical system framework of a transient identification and classification system provided by an embodiment of the present application is shown.

[0018] Figure 4 An automatic classification suggestion function interface of an identified power plant transient provided by an embodiment of the present application is shown.

[0019] Figure 5 A query function interface of a classified power plant transient data provided by an embodiment of the present application is shown.

[0020] Figure 6 A structural diagram of an electronic device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] The embodiments of the present application will be described in detail by specific examples. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in the specification. The present application can also be implemented or applied by different specific embodiments, and the details in the specification can be modified or changed based on different views and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] It should be noted that the diagrams provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the components related to the present application are shown in the diagrams, not the number, shape and size of the components when actually implemented. The type, number and ratio of the components when actually implemented can be arbitrarily changed, and the layout type of the components can also be more complex.

[0023] In the following description, numerous specific details are discussed in order to provide a thorough understanding of embodiments of the present application. However, those skilled in the art will recognize that the embodiments of the present application can be practiced without these specific details. In other instances, well-known structures and devices are not described in detail in order to avoid obscuring embodiments of the present application.

[0024] The present application provides a transient identification and classification method, by acquiring the operation parameter sequence of a component within a preset sampling time, first identifying and extracting the actual transient data in the operation parameter sequence, second determining the acquired data as classifiable transient, and then determining the actual transient data, screening the reference transient with the largest similarity to the actual transient data, and thus determining the transient type corresponding to the actual transient data.

[0025] Referring to Figure 1 , a flowchart of the transient identification and classification method in an exemplary embodiment of the present application is shown, including the following steps:

[0026] S100, acquiring the operation parameter sequence of a component within a preset sampling time.

[0027] In the field of nuclear power plants, the integrity of the pressure boundary of the primary loop of the reactor is crucial to the safety of the nuclear power plant. From the factory test to the end of the service life of the nuclear power plant, the equipment such as pipes and containers of the primary loop is in a high-temperature, high-pressure, and high-irradiation environment for a long time. The primary loop equipment not only bears static loads such as temperature and internal pressure, but also bears dynamic loads caused by phenomena such as medium fluid impact. Different loads acting on the equipment will cause a certain degree of mechanical fatigue or damage. Among them, transient is a situation that causes temperature and pressure changes during the operation of the nuclear power plant. These situations often occur in nuclear power plants, such as unit shutdown, power regulation, fault response, etc. These transients will have different degrees of impact on various systems, components, and equipment of the nuclear power plant, which may cause their deformation, fatigue, fracture, and other failure modes, thereby affecting the performance and safety of the nuclear power plant. Fatigue refers to the fact that the equipment and components of the nuclear power plant are subjected to various cyclic loads during operation, and the cyclic stress generated thereby may cause microphysical damage to the relevant materials of the items. Even when the stress is far below the ultimate strength of the equipment and component materials, this micro-damage can accumulate under the action of continuous cyclic load until it develops into a crack or other macro-damage, thereby causing the equipment or component to fail. This process of damage and failure of the item due to cyclic load is called fatigue.

[0028] The power plant needs to supervise the state parameters and the occurrence times of various transients in the actual operation process, carry out fatigue monitoring and transient statistical work, facilitate effective control of the state of key equipment of the power plant, tracking and evaluating the state of the power plant operation. During the entire service life of the power plant, recording the changes of important parameters related to the pressure boundary of the primary loop, recording and archiving various transients are crucial for unit life management. During the operation of the nuclear power plant unit, the technical personnel of the power plant can monitor and predict the fatigue damage of the equipment by periodically checking the unit operation parameters, identifying and counting the transients of the unit equipment, and then optimizing the operation and management of the power plant. At present, the transient fluctuation needs to be judged, classified and recorded by manual operation. The unit transients are identified by manual identification, and classified according to the design reference transient data, which consumes a lot of manpower. The present application obtains the transient data sequence of the nuclear power plant components by periodically checking the unit operation parameter sequence.

[0029] S200, extracting actual transient data from the operation parameter sequence according to the preset amplitude change speed interval, and calculating transient characteristic data.

[0030] The initially obtained operation parameter sequence needs to be judged whether a transient has occurred therein. If a transient has occurred therein, the actual transient data is screened out from the operation parameter sequence to be further classified. The transient refers to the case that the amplitude change speed of the parameter data in the power plant operation data in a period of time exceeds the corresponding amplitude change speed interval in the period of time, and it is considered that the component corresponding to the power plant parameter has experienced a transient. If no transient occurs in the operation parameter sequence, the operation parameter sequence is a non-transient sequence, and the subsequent described content is a method implemented under the condition that the operation parameter sequence has a transient. In order to more accurately screen out the judgment transient and the actual transient data, the present application is described respectively according to one (for example: only temperature operation parameter sequence) or multiple (for example: temperature, pressure and other multiple operation parameters exist at the same time, the data length of each operation parameter sequence is the same, and the corresponding actual operation parameter exists for each operation parameter sequence at the same time) operation parameter sequences.

[0031] S210, when the operation parameter sequence is one, specifically, in an embodiment of the present application, the step of extracting actual transient data from the operation parameter sequence according to the preset amplitude change speed interval and calculating transient characteristic data comprises:

[0032] S211, dividing the operation parameter sequence into a plurality of parameter subsequences according to the amplitude change speed;

[0033] For each operation parameter, the amplitude change speed is calculated according to the amplitude change at the preset length, and the calculation formula of the amplitude change speed is: wherein v is the amplitude variation speed corresponding to the current operation parameter obtained, y after is the operation parameter at the preset time length Δt, y now is the current operation parameter, and Δt is the preset time length, i.e. a fixed value. The operation parameter sequence is divided into a plurality of parameter subsequences according to the amplitude variation speed. For example, the operation parameter sequence of temperature is obtained, and the amplitude variation speed of each temperature value point, i.e. operation parameter, in the future 3 hours is calculated, i.e. the temperature value at the future 3 hours of the current temperature value minus the current temperature value divided by 3. The speed value obtained is the amplitude variation speed corresponding to the current temperature value. Then, the operation parameter sequence is divided into a plurality of parameter subsequences according to the amplitude variation speed corresponding to each temperature value point. Generally, the operation parameter sequence of temperature is divided into a plurality of parameter subsequences according to the temperature value points with the amplitude variation speed greater than or equal to the preset speed threshold.

[0034] S212, selecting the parameter subsequence with the amplitude variation speed exceeding the amplitude variation speed interval as the actual transient data.

[0035] Among the plurality of parameter subsequences, the parameter subsequence with the amplitude variation speed corresponding to each operation parameter exceeding the amplitude variation speed interval is selected, and the parameter subsequence is the actual transient data.

[0036] S213, calculating the maximum parameter variation amplitude and the maximum parameter variation rate of the actual transient data as the transient characteristic data.

[0037] The maximum parameter variation amplitude and the maximum parameter variation rate of the actual transient data are obtained, which are used as the judgment basis for judging whether the actual transient data can be classified in the next step. The calculation formula of the maximum parameter variation amplitude is as follows: Δy max =y max -y min wherein Δy max is the maximum amplitude variation amplitude, y max is the maximum value in the actual transient data, and y min is the minimum value in the actual transient data. The calculation formula of the maximum parameter variation rate is as follows: wherein v max is the maximum parameter variation rate, and t min,max is the time between y max and y min .

[0038] S220, when the operation parameter sequence is multiple, the step of extracting the actual transient data from the operation parameter sequence and calculating the transient characteristic data according to the preset amplitude variation speed interval comprises:

[0039] S221, divide each operating parameter sequence into several parameter subsequences according to its corresponding amplitude variation speed;

[0040] When the operating parameter sequence is greater than one, the actual transient judgment process is quite different from that of a single operating parameter sequence. For each operating parameter sequence, the corresponding amplitude variation speed is divided into several parameter subsequences, and the division process of the parameter subsequences for each operating parameter sequence is the same as that of the single operating parameter sequence.

[0041] S222, according to the amplitude variation speed interval corresponding to each operating parameter sequence, select multiple segmentation sequences from the several parameter subsequences.

[0042] For each operating parameter sequence: among the several parameter subsequences, select the multiple parameter subsequences in which the amplitude variation speed corresponding to each operating parameter is within the amplitude variation speed interval, which are the segmentation sequences corresponding to the operating parameter sequence.

[0043] S223, take the earliest sampling time in all segmentation sequences as the segmentation starting point.

[0044] Among the segmentation sequences of multiple operating parameters, according to the time sequence, select the segmentation sequence with the earliest segmentation starting point as the segmentation starting point.

[0045] S224, determine the segmentation endpoint by judging all segmentation sequences according to the sampling time sequence:

[0046] When the segmentation sequence with the earliest sampling time has no intersection with other segmentation sequences, the sampling endpoint of the segmentation sequence with the earliest sampling time is taken as the segmentation endpoint.

[0047] That is, at this time, only one operating parameter sequence has occurred transient, so the sampling endpoint of the segmentation sequence corresponding to it is taken as the segmentation endpoint.

[0048] When the segmentation sequence with the earliest sampling time has intersection with other segmentation sequences, the sampling endpoint of the last segmentation sequence which only has intersection with one segmentation sequence is taken as the segmentation endpoint.

[0049] If more than one operation parameter sequence has a transient during the period, the segmentation sequence that has a transient in each operation parameter sequence is screened for repetition in time until only one segmentation sequence is repeated, and the sampling end point of the last segmentation sequence that has an intersection with only one segmentation sequence is taken as the segmentation end point. Taking temperature and pressure as examples: a transient occurs during 1-1.5 hours of unit operation, the amplitude variation speed of temperature during the period exceeds the preset threshold, and the segmentation sequence of temperature is the operation parameter of the unit during 1-1.5 hours of operation. The amplitude variation speed of pressure during the period does not exceed the corresponding threshold, so only the sampling end point of the segmentation sequence corresponding to temperature is taken as the segmentation end point. However, if the amplitude variation speed of temperature exceeds the preset threshold of temperature during 1-1.5 hours, and the amplitude variation speed of pressure exceeds the preset threshold of pressure during 1.3-7 hours, because the time when the amplitude variation speed of temperature and pressure exceeds the corresponding threshold has an intersection in time, the final segmentation end point should take the sampling end point of the last segmentation sequence as the segmentation end point, that is, the sampling end point of the segmentation sequence corresponding to pressure is taken as the segmentation end point.

[0050] S225, extracting the corresponding data from each operation parameter sequence according to the segmentation start point and the segmentation end point as the actual transient data corresponding to each operation parameter sequence;

[0051] S226, calculating the maximum parameter variation amplitude and the maximum parameter variation rate of each actual transient data as the transient characteristic data corresponding thereto.

[0052] For the actual transient data corresponding to each operation parameter sequence, the maximum parameter variation amplitude and the maximum parameter variation rate of each actual transient data are calculated, and the calculation formula is the same as above when there is only one operation parameter sequence. The

[0053] S300, according to the transient characteristic data, determining that the actual transient data is a classifiable transient, and calculating the similarity values of the actual transient data and a plurality of reference transients respectively, and taking the transient type of the reference transient with the highest similarity value as the classification result of the component, comprising the steps of:

[0054] S310, when there is only one operation parameter sequence:

[0055] S311, comparing the transient characteristic data with a plurality of preset reference transients to determine that the type of the actual transient data is a classifiable transient.

[0056] Specifically, in an embodiment of the present application, the step of comparing the transient characteristic data with a plurality of preset reference transients to determine that the type of the actual transient data is a classifiable transient comprises:

[0057] First, the transient characterization data is compared with the corresponding reference transient characterization data in each of the reference transients.

[0058] Finally, when the transient characterization data is smaller than all the reference transient characterization data, the type of the corresponding actual transient data is determined to be a classifiable transient.

[0059] The obtained transient characterization data is compared with the benchmark transient characterization data in each benchmark transient. For example, the temperature transient characterization data has a maximum parameter change amplitude of 60°C and a maximum parameter change rate of 50°C / min, i.e., (60°C, 50°C / min). There are multiple benchmark transients, each corresponding to a benchmark transient characterization data. Therefore, there are multiple benchmark transient characterization data available for judgment: (80°C, 56°C / min) and (100°C, 110°C / min). Based on the above two benchmark transient characterization data, the maximum temperature change amplitude of 60°C in the temperature transient characterization data is less than the maximum parameter change amplitude of all benchmark transient characterization data (80°C and 100°C). At the same time, the maximum temperature change rate is 50°C / min, which is less than or equal to the maximum parameter change rate of all benchmark transient characterization data (56°C / min and 110°C / min). Therefore, the obtained actual temperature transient data is judged to be a classifiable transient. On the contrary, the maximum parameter change rate of 57°C / min in the transient characterization data of another temperature (60°C, 57°C / min) is greater than the maximum parameter change rate of 56°C / min in the transient characterization parameters of the benchmark transient. Therefore, the actual transient data of this temperature is an unclassifiable transient.

[0060] S312: Calculate the distance between the actual transient data and each of the reference transients according to the DTW algorithm, and obtain a corresponding similarity value based on the distance value.

[0061] Since the sequence length of the actual transient data of the obtained operating parameter sequence is often different from the sequence length of the reference transient, in order to solve this problem, the DTW algorithm is used to calculate the distance value between the actual transient data and each reference transient. The calculation process of the DTW algorithm is as follows:

[0062] First, the distance matrix between the actual transient data and the benchmark transient is calculated, that is, the local distance of each pair of points is calculated, usually using Euclidean distance or Manhattan distance. The Euclidean distance formula is as follows: d(i, j) = |x i -y j | 2 , where d(i,j) is the i-th operating parameter x in the actual transient data i and the reference transient j-th value y j The Euclidean distance between .

[0063] Secondly, a cumulative distance matrix is calculated, and the minimum cumulative distance from (1, 1) to (i, j) is recursively calculated, that is: wherein D(i, j) is the distance value between the i th operating parameter x i of the actual transient data and the j th value y j of the reference transient.

[0064] Then, the optimal alignment path is calculated according to the cumulative distance matrix, and the distance value between the actual transient data and the reference transient is calculated.

[0065] Further, in an embodiment of the present application, the similarity value is calculated according to the distance value through the following formula:

[0066] wherein η is the similarity value, and d is the distance value between the actual transient data and the reference transient.

[0067] S313, the reference transient with the highest similarity value is screened out.

[0068] The similarity value between the actual transient data and each reference transient is calculated, and the reference transient with the highest similarity is screened out.

[0069] S314, the transient type corresponding to the screened reference transient is taken as the classification result of the component.

[0070] S320, when the operating parameter sequence is only one, the actual transient data is determined to be a classifiable transient according to the transient characteristic data, the similarity values between the actual transient data and multiple reference transients are calculated respectively, and the transient type of the reference transient with the highest similarity value is taken as the classification result of the component.

[0071] S321, the transient characteristic data is compared with multiple preset reference transients, and the type of the actual transient data is determined to be a classifiable transient.

[0072] The transient characterization data of the actual transient data is compared with the benchmark transient characterization data of the benchmark transient, when the transient characterization data is less than the corresponding data in all benchmark transient characterization data, the actual transient data is the classifiable transient. For example, there are multiple operating parameter sequences, which are temperature and pressure respectively, the maximum parameter variation amplitude of temperature is 30℃, the maximum parameter variation rate of temperature is 20℃ / min, the maximum parameter variation amplitude of pressure is 1MPa, and the maximum parameter variation rate of pressure is 0.5MPa / min, that is (30℃, 20℃ / min, 1MPa, 0.5MPa / min); the transient characterization data of the benchmark transient has multiple, which are (80℃, 56℃ / min, 4MPa, 3MPa / min), (35℃, 80℃ / min, 3MPa, 1MPa / min), (100℃, 110℃ / min, 1.5MPa, 0.75MPa / min), after comparison, the transient characterization data of the actual transient data is less than or equal to the corresponding data in all benchmark transient characterization data, therefore, the actual transient data is the classifiable transient. On the contrary, when there is any one data in the transient characterization data of the actual transient data exceeding the corresponding data in the transient characterization data of the benchmark transient, the actual transient data is the unclassifiable transient.

[0073] S322, for each benchmark transient:

[0074] According to the DTW algorithm, the sub-distance values of each actual transient data and the corresponding benchmark transient data of the benchmark transient are calculated respectively.

[0075] When the operating parameter sequence is multiple, the number of the benchmark transient data of the benchmark transient is the same as the number of the operating parameter sequence, for example: the data of the benchmark transient is recorded every two minutes, which are:

[0076] temperature (280, 278, 276, 274, 272...),

[0077] pressure (15.0, 14.8, 14.6, 14.4, 14.2...);

[0078] The actual transient data of the operating parameter sequence is:

[0079] temperature (285, 283, 281, 279, 277...),

[0080] pressure (15.5, 15.3, 15.1, 14.9, 14.7...);

[0081] Among them, the length of the data sequence in the benchmark transient and the length of the actual transient data of the operating parameter sequence are not necessarily the same, therefore, the sub-distance values of the above temperature and pressure are calculated respectively by using the DTW algorithm.

[0082] convert each sub-distance value into a corresponding sub-similarity value;

[0083] Each sub-distance value is calculated into a sub-similarity value according to the following formula respectively:

[0084] wherein η i is a sub-similarity value, d i is the distance value between the i-th actual transient data and the corresponding reference transient data of the reference transient data.

[0085] All the sub-similarity values are weighted to obtain a similarity value of the reference transient data and the plurality of actual transient data;

[0086] The transient type of the reference transient data with the highest similarity value is selected as the classification result of the component.

[0087] Since there are multiple reference transients, the similarity value of the actual transient data of the operating parameter sequence and each reference transient is calculated, and finally the reference transient with the highest similarity value is selected, and the transient type of the reference transient is selected as the transient type of the monitored component.

[0088] As Figure 2 shown, the transient identification and classification system 200 includes a data acquisition module 210, a transient data determination and characterization calculation module 220, and a classification result output module 230. The data acquisition module 210 is configured to acquire an operating parameter sequence of a component within a preset sampling time. The transient data determination and characterization calculation module 220 is configured to extract actual transient data from the operating parameter sequence according to a preset amplitude change speed interval, and calculate transient characterization data. The classification result output module 230 is configured to determine that the actual transient data is a classifiable transient according to the transient characterization data, calculate similarity values of the actual transient data and a plurality of reference transients respectively, and select a transient type of a reference transient with the highest similarity value as a classification result of the component.

[0089] The specific limitations of the transient identification and classification system can be referred to the limitations of the transient identification and classification method described above, which will not be repeated here. Each module in the above transient identification and classification system can be realized by software, hardware, and combinations thereof, in whole or in part.

[0090] The above modules can be embedded in or independent of the processor in the computer device in hardware format, or stored in the memory in the computer device in software format, so that the processor can call the operations of the above modules.

[0091] It should be noted that in order to highlight the innovative part of the present application, the modules not closely related to solving the technical problems proposed by the present application are not introduced in the present embodiment, but this does not mean that there are no other modules in the present embodiment.

[0092] Further, as shown in the flow chart, Figure 3 The present application supplements the processing procedure of the transient identification and classification system 200 in combination with the actual operation and management of the power plant,

[0093] The specific data processing procedure is that the transient identification and classification system 200 is provided with a display device with display function, the power plant operation data is imported into the data acquisition module 210, and then the power plant transient automatic identification and power plant transient data analysis are performed through the transient data determination and characterization calculation module 220, that is, whether a transient occurs is identified according to the input power plant operation data, then whether it is a classifiable transient is analyzed, and finally the transient characterization data is obtained. For example, the threshold values of temperature and pressure of the device to be identified are manually input by artificial, the identification button in the system operation interface is clicked, and the interface displays that the transient identification and classification module calculates the temperature or pressure curve of each loop of the entire power plant system. Because there are multiple curves, multiple loops are corresponded, and each loop has a corresponding device, which is equivalent to calculating each curve to obtain the transient condition of each device. If a transient is identified, the identified transient condition is displayed through the display device, which specifically includes the identified power plant transient code, identification time, transient start time, transient end time, trigger threshold (trigger temperature threshold or pressure threshold), trigger column (that is, which parameters trigger the threshold), device trigger state, classification type and other information. The change curve of the viewable transient process data of the identified transient is implemented to query the duration and trigger device. The transient analysis procedure is to calculate the transient characterization data of the transient. The automatic classification process of the transient can refer to the above-mentioned transient identification and classification method. The transient characterization data (including the maximum parameter change amplitude of the temperature curve, the maximum parameter change rate, the maximum parameter change amplitude of the pressure curve, etc.) automatically identified by the module is compared with the design benchmark transient data stored in the design benchmark transient database to realize the power plant transient data analysis. Then, the transient intelligent classification is that if any parameter characteristic value of the actual transient exceeds the corresponding value of all design benchmark transient data in the design benchmark transient database, the actual transient is classified into "unclassifiable transient". If none of them exceeds the corresponding value, the DTW analysis similarity of the transient is performed in the classification result output module 230.

[0094] For example, the actual transient data is: temperature change amplitude 60℃, temperature change maximum rate 70℃ / Min, pressure change amplitude 2MPa; the design reference transient database has three design reference transient data, which are (80℃, 56℃ / Min, 4MPa), (35℃, 80℃ / Min, 3MPa), (100℃, 110℃ / Min, 1.5MPa), since the temperature change amplitude, temperature change rate, pressure change amplitude in the actual transient data cannot be simultaneously enveloped by the temperature change amplitude, temperature change rate, pressure change amplitude of the same transient data of all design reference transient data in the basic database design reference transient database, therefore, the actual transient data is classified into "unclassifiable transient".

[0095] The DTW algorithm in the DTW analysis matching is essentially a function of calculating the similarity score between two sequences and different sequence lengths. To achieve this purpose, the sequence is nonlinearly "constrained and regularized" in the time dimension, so as to evaluate the similarity on the basis of eliminating the duration difference.

[0096] After obtaining the similarity of the actual transient and temperature and pressure respectively, in order to ensure the consistency of the automatic classification suggestion and the manual classification result, the transient recognition classification module can set different weight factors for the temperature parameter sequence and the pressure parameter sequence, and can be displayed and operated on the display device, and the final similarity with the design reference transient is calculated. As shown in Figure 4 Then the transient recognition classification module will provide multiple classifiable categories according to the numerical value of the similarity, and display on the display device, the system can directly select the transient type with the highest similarity as the classification result input into the power plant transient database, or the technical personnel can operate to select the corresponding transient recognition result input into the power plant transient database.

[0097] In addition, under the premise of ensuring that the design reference transient can envelope the actual transient of the power plant, the technical personnel can manually select and classify the similarity between the design reference transient and the actual transient of the power plant, the consumption degree of the design reference transient and other factors according to the cause of the actual transient of the power plant, wherein the data of the design reference transient can be stored in the design reference transient storage module added in the data acquisition model, and the design reference transient storage module is used for storing the design reference transient data corresponding to the preset design reference transient; therefore, when the matching degree of the automatic classification suggestion and the manual classification result is poor, the technical personnel can also manually classify or divide into the "unclassifiable transient" category to record the transient statistics of the power plant, and input into the power plant transient database of the output module 230. In order to further adapt to the actual use, the present application provides the transient statistical classification query of the classified transient data of the power plant, and realizes the design of the unit transient consumption statistical of the power plant transient and the transient statistical daily report input:

[0098] Firstly, the transient statistics classification query can realize the following according to the data stored in the power plant transient database: the query of the duration of the power plant transient, the temperature amplitude and other information; the classification of the actual power plant transient and the comparison of the actual transient and the data of the classified transient; the note of the cause of the power plant transient; the reclassification of the power plant transient, and the operation interface is as shown in Figure 5

[0099] Secondly, the unit transient consumption statistics can realize the following: the consumption statistics of the transient times, which can display the cumulative consumption of the power plant transient and the remaining times, can correspond to the related equipment components, and can warn the consumption of the power plant transient.

[0100] Finally, the transient statistics daily report input can realize the following according to the data stored in the power plant transient database: the storage and management of the power plant transient statistics report, which can realize the uploading, storage and downloading of the power plant transient statistics monthly report or quarterly report, and is convenient for maintenance and management.

[0101] For the design benchmark transient data, it is stored in the design benchmark transient database, wherein each transient corresponds to a serial number, a power plant design benchmark transient code (each preset transient has a unique code, which can reflect the monitoring object in the string code), a name (the name of the running state such as start-up and shut-down), a monitoring object (equipment or system), a total number (the total number of the power plant expected to occur the transient), and an operation (the temperature and pressure expected change curve of the designed transient) parameter, and the functions thereof are as follows:

[0102] Firstly, the change of each transient characteristic parameter is displayed, including the amplitude and rate information of the transient temperature and pressure change, and by clicking the temperature graph and pressure graph icons in the operation parameter column of the design benchmark transient database main interface, the characteristic parameter change graph of each preset transient is displayed.

[0103] Secondly, the multi-dimensional query of the design benchmark transient data is realized, which can be queried according to multiple dimensions such as "transient related equipment", "transient name" or "transient code", and by clicking a certain transient, the corresponding design benchmark transient data graph can be opened.

[0104] The above transient identification and classification statistics system only takes the temperature and pressure and other running parameter sequences as examples, and other transient identification and classification according to the time sequence obtained power plant component running parameters according to this method are still within the protection scope of the present application. When multiple running parameter sequences are collected, that is, when multiple running parameter data are collected, the corresponding modification of the transient identification and classification statistics system is still within the protection scope of the present application.

[0105] As shown in Figure 6 ​As shown, the present application also discloses an electronic device 3 comprising a memory 32, a processor 31 and a bus, and can further comprise a computer program stored in the memory 32 and executable on the processor 31, such as a DTW algorithm program.

[0106] The memory 32 comprises at least one type of readable storage medium, such as a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g. an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 32 can be an internal storage unit of the electronic device 3, such as a mobile hard disk of the electronic device 3. In other embodiments, the memory 32 can also be an external storage device of the electronic device 3, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 3. Further, the memory 32 can comprise both an internal storage unit and an external storage device of the electronic device 3. The memory 32 can be used not only to store application software and various data installed on the electronic device 3, but also to temporarily store data that has been output or will be output.

[0107] The processor 31 can be composed of integrated circuits in some embodiments, such as a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more combinations of a central processing unit (CPU), a microprocessor, a digital processing chip, a graphics processor and various control chips, etc. The processor 31 is a control unit of the electronic device 3, which connects various components of the entire electronic device 3 through various interfaces and lines, and executes various functions and processes data of the electronic device 3 by running or executing programs or modules stored in the memory 32 and calling data stored in the memory 32.

[0108] The processor 31 executes an operating system and various application programs installed on the electronic device 3. The processor 31 executes the application programs to implement the steps in the transient identification and classification method described above.

[0109] For example, the computer program can be divided into one or more modules, which are stored in the memory 32 and executed by the processor 31 to complete the present application. The one or more modules can be a series of computer program instructions capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device 3.

[0110] The integrated unit in the form of the software function module can be stored in a computer readable storage medium, which can be non-volatile or volatile. The software function module is stored in a storage medium, and includes a plurality of instructions for enabling a computer device (which can be a personal computer, a computer device, or a network device, etc.) or a processor to execute part of the functions of the transient identification and classification method.

[0111] In summary, the transient identification and classification method, system, device and medium disclosed by the present application can realize automatic identification of transient data, reduce manual intervention, avoid errors caused by manual judgment negligence, and improve the speed of transient identification and classification. The similarity between the actual transient data and the reference transient data is calculated by the DTW algorithm, which can effectively process transient data sequences of different lengths and improve the accuracy of classification, providing a decision basis for transient classification. Therefore, the present application effectively overcomes the shortcomings of the prior art and has high industrial utilization value.

[0112] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical idea disclosed by the present application should be covered by the claims of the present application.

Claims

1. A transient identification and classification method, characterized in that: The method comprises: Obtain the operating parameter sequence of the component within the preset sampling time; extracting actual transient data from the operating parameter sequence according to a preset amplitude change speed interval, and calculating transient characterization data; The actual transient data is determined to be a classifiable transient according to the transient characterization data, and similarity values ​​between the actual transient data and multiple reference transients are calculated respectively, and the transient type of the reference transient with the highest similarity value is used as the classification result of the component.

2. The transient identification and classification method according to claim 1, characterized in that: When there is one operating parameter sequence, the steps of extracting actual transient data from the operating parameter sequence according to a preset amplitude change speed interval and calculating transient characterization data include: Dividing the operating parameter sequence into a plurality of parameter subsequences according to amplitude change speed; Selecting a parameter subsequence whose amplitude change speed exceeds the amplitude change speed interval as actual transient data; The maximum parameter change amplitude and the maximum parameter change rate of the actual transient data are calculated as the transient characterization data.

3. The transient identification and classification method according to claim 2, characterized in that: The steps of determining, based on the transient characterization data, that the actual transient data is a classifiable transient, respectively calculating similarity values ​​between the actual transient data and a plurality of reference transients, and taking the transient type of the reference transient with the highest similarity value as the classification result of the component include: Comparing the transient characterization data with a plurality of preset reference transients to determine that the type of the actual transient data is a classifiable transient; Calculating the distance between the actual transient data and each of the reference transients according to the DTW algorithm, and obtaining a corresponding similarity value based on the distance value; Screening out the reference transient with the highest similarity value; The transient type corresponding to the filtered reference transient is used as the classification result of the component.

4. The transient identification and classification method according to claim 3, characterized in that: The similarity value is calculated based on the distance value using the following formula: Wherein, η is the similarity value, and d is the distance value.

5. The transient identification and classification method according to claim 1, characterized in that: When there are multiple operating parameter sequences, the steps of extracting actual transient data from the operating parameter sequences according to a preset amplitude change speed interval and calculating transient characterization data include: Divide each operating parameter sequence into several parameter subsequences according to its corresponding amplitude change speed; According to the amplitude change speed interval corresponding to each operating parameter sequence, a plurality of segmentation sequences are selected from a plurality of parameter subsequences; The earliest sampling time in all segmentation sequences is used as the segmentation starting point; Determine all segmentation sequences according to the sampling time sequence and determine the segmentation endpoint: When the segmentation sequence with the earliest sampling time has no intersection with other segmentation sequences, the sampling end point of the segmentation sequence with the earliest sampling time is used as the segmentation end point; When the segmentation sequence with the earliest sampling time intersects with other segmentation sequences, the sampling end point of the last segmentation sequence that intersects with only one segmentation sequence is used as the segmentation end point; Extract corresponding data from each operating parameter sequence according to the segmentation start point and the segmentation end point as the actual transient data corresponding to each operating parameter sequence; The maximum parameter change amplitude and the maximum parameter change rate of each actual transient data are calculated as the corresponding transient characterization data.

6. The transient identification and classification method according to claim 5, characterized in that: The step of determining, based on the transient characterization data, that the actual transient data is a classifiable transient, calculating similarity values ​​between the actual transient data and a plurality of reference transients, and taking the transient type of the reference transient with the highest similarity value as the classification result of the component includes: Comparing the transient characterization data with a plurality of preset reference transients to determine that the type of the actual transient data is a classifiable transient; For each benchmark transient: According to the DTW algorithm, the sub-distance value between each actual transient data and the corresponding benchmark transient data of the benchmark transient is calculated respectively; Convert each sub-distance value into the corresponding sub-similarity value; Performing a weighted operation on all the sub-similarity values ​​to obtain similarity values ​​between a reference transient and a plurality of the actual transient data; The transient type corresponding to the reference transient with the highest similarity value is selected as the classification result of the component.

7. The transient identification and classification method according to claim 3 or 6, characterized in that: The step of comparing the transient characterization data with a plurality of preset reference transients to determine whether the type of the actual transient data is a classifiable transient includes: comparing the transient characterization data with corresponding reference transient characterization data in each of the reference transients; When the transient characterization data is smaller than all the reference transient characterization data, the type of the corresponding actual transient data is determined to be a classifiable transient.

8. A transient recognition and classification system, characterized in that: The system comprises: A data acquisition module is used to obtain the operating parameter sequence of the component within a preset sampling time; A transient data determination and characterization calculation module, configured to extract actual transient data from the operating parameter sequence according to a preset amplitude change speed interval, and calculate transient characterization data; a classification result output module, configured to determine, based on the transient characterization data, that the actual transient data is a classifiable transient, calculate similarity values ​​between the actual transient data and a plurality of reference transients, and use the transient type of the reference transient with the highest similarity value as the classification result of the component; 9. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the transient recognition and classification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed by a processor of a computer, the computer is caused to execute the transient identification and classification method according to any one of claims 1 to 7.

Images