Online monitoring method and system for state of roller lead screw control rod driving mechanism and medium

Abstract

The invention discloses an online monitoring method and system for the state of a roller screw control rod driving mechanism and a medium. Relates to the technical field of online monitoring. The method is improved on the basis of a traditional control rod driving mechanism state monitoring technology, and for a roller lead screw type control rod driving mechanism, a first effective value signal and a second effective value signal are calculated based on an acceleration signal; the time points of action starting and action ending of the roller lead screw type control rod driving mechanism are determined, action time periods with the action time reaching the standard are screened out for feature discrimination, and finally the state of the roller lead screw type control rod driving mechanism is obtained by combining all feature discrimination results. The problem that the state of the roller lead screw type control rod driving mechanism cannot be monitored on line is solved, the early failure of the roller lead screw type control rod driving mechanism is found in advance, and the safety and reliability of the control rod driving mechanism are powerfully guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of online monitoring, and in particular to a method, system and medium for online monitoring of the state of a roller screw control rod driving mechanism. Background Art

[0002] The control rod drive mechanism is an important part of the reactor to control the core power. It is also an important safety facility for the safe shutdown of the reactor. Its safety and reliability directly affect the functional integrity and safety and reliability of the reactor. In order to achieve the core power regulation task, the control rod drive mechanism needs to perform rod lifting and rod insertion actions for different lengths of time, which will generate cyclic loads on the control rod drive mechanism components, causing damage to the control rod drive mechanism components, causing bearing wear, roller jamming and other faults in the drive mechanism, affecting the normal operation of the control rod, and burying hidden dangers for the stable operation of the reactor. In the past, the fault monitoring method for the roller screw type control rod drive mechanism was to conduct continuous lifting and lowering tests on the control rod drive mechanism during the overhaul. This method is a kind of "regular maintenance". This method will cause excessive maintenance, and additional tests will also introduce additional faults to the control rod drive mechanism. Therefore, it is necessary to propose a method for online fault monitoring of the roller screw type control rod drive mechanism.

[0003] During the normal operation of the reactor, the roller screw type control rod drive mechanism is not in real-time operation state, and the length of time of the real-time operation state of the roller screw type control rod drive mechanism is uncertain. When the control rod drive mechanism is in static state, the operation state of the drive mechanism cannot be identified. Only when the drive mechanism is in operation state can its state be identified. Therefore, it is necessary to capture the time points when the control rod drive mechanism starts and ends the action. However, since there is no other trigger signal to determine the start and end points of the signal, it is necessary to start from the waveform itself. At the same time, the continuous acquisition time of the control rod drive mechanism is inconsistent, and the stable characteristic quantity cannot be used to identify the state of the control rod drive mechanism. Summary of the invention

[0004] The technical problem to be solved by the present invention is that during the normal operation of the reactor, the roller screw type control rod drive mechanism is not in a real-time operation state, and the length of time of the real-time operation state of the roller screw type control rod drive mechanism is uncertain, so it is difficult to accurately identify the operation state of the drive mechanism; the purpose of the present invention is to provide a method, system and medium for online monitoring of the state of a roller screw control rod drive mechanism, and on the basis of the traditional control rod drive mechanism state monitoring technology, the method is improved, and for the roller screw type control rod drive mechanism, a first effective value signal and a second effective value signal are calculated based on the acceleration signal, so as to determine the time points of the start and end of the action of the roller screw type control rod drive mechanism, screen out the action time period that meets the action time standard for feature identification, and finally combine all feature identification results to obtain the state of the roller screw control rod drive mechanism; the problem that the state of the roller screw type control rod drive mechanism cannot be monitored online is solved, and the early failure of the roller screw type control rod drive mechanism is discovered in advance, which effectively guarantees the safety and reliability of the control rod drive mechanism.

[0005] The present invention is achieved through the following technical solutions:

[0006] This solution provides an online monitoring method for the state of a roller screw control rod drive mechanism, including:

[0007] Collecting the acceleration signal of the roller screw control rod driving mechanism;

[0008] Calculating a first effective value signal and a second effective value signal based on the acceleration signal, analyzing the first effective value signal and the second effective value signal to determine a start action time of the roller screw control rod drive mechanism, and analyzing the second effective value signal to determine an end action time of the roller screw control rod drive mechanism;

[0009] Counting the start action time and the end action time of the roller screw control rod drive mechanism, determining the action time period of the roller screw control rod drive mechanism, and screening out the action time period that meets the action time standard;

[0010] The characteristics of the action time periods that meet the action time standards are identified, and the state of the roller screw control rod drive mechanism is obtained by combining all the characteristic identification results.

[0011] Working principle of the scheme: During the normal operation of the reactor, the roller screw type control rod drive mechanism is not in a real-time operation state, and the length of time of the real-time operation state of the roller screw type control rod drive mechanism is uncertain, making it difficult to accurately identify the operation state of the drive mechanism; the purpose of the present invention is to provide a method, system and medium for online monitoring of the state of a roller screw control rod drive mechanism, and to improve the method on the basis of the traditional control rod drive mechanism state monitoring technology. For the roller screw type control rod drive mechanism, a first effective value signal and a second effective value signal are calculated based on the acceleration signal to determine the time points of the start and end of the action of the roller screw type control rod drive mechanism, and the action time period with the action time meeting the standard is screened out for feature identification, and finally the state of the roller screw control rod drive mechanism is obtained by combining all feature identification results; the problem that the state of the roller screw type control rod drive mechanism cannot be monitored online is solved, and the early failure of the roller screw type control rod drive mechanism is discovered in advance, which effectively guarantees the safety and reliability of the control rod drive mechanism.

[0012] Further optimization plans also include:

[0013] An abnormal result threshold is set, all feature identification results are obtained, each feature identification result is compared with the abnormal result threshold, and the number N of feature identification results exceeding the abnormal result threshold is counted. When the number N accounts for more than 75% of the total number of feature identification results, the roller screw control rod drive mechanism is judged to be faulty; otherwise, the roller screw control rod drive mechanism is judged to be normal.

[0014] A further optimization scheme is that the acquisition of the acceleration signal of the roller screw control rod driving mechanism includes the following method:

[0015] An acceleration sensor is arranged on the bearing seat of the roller screw control rod driving mechanism to continuously collect acceleration signals. The data length of each collected acceleration signal is L, and the time of each collection of the acceleration signal is recorded.

[0016] A further optimization scheme is that the first effective value signal and the second effective value signal are calculated based on the acceleration signal, including a method:

[0017] Get acceleration signal;

[0018] Calculate the square of the effective value of the acceleration signal collected each time to obtain a first effective value signal;

[0019] Set the number of elements N in the storage queue A to store the first valid value signal in the storage queue A; update the storage queue A when the storage queue A is full of elements: the next element is arranged at the end of the storage queue A, and the element that entered the queue first is squeezed out;

[0020] After the storage queue A is full of elements and after updating, the mean is calculated to obtain a second effective value signal.

[0021] A further optimization scheme is that the analyzing the first effective value signal and the second effective value signal to determine the start action time of the roller screw control rod driving mechanism includes:

[0022] Calculating a quotient of the first effective value signal and the second effective value signal;

[0023] A threshold value x1 is preset, and the quotient is compared with the threshold value x1; among the quotients corresponding to the acceleration signals collected three times in succession, when the quotients corresponding to two times of the acceleration signals collected are both less than the threshold value x1, the time of the last acceleration signal collection is determined to be the start action time of the roller screw control rod drive mechanism.

[0024] A further optimization scheme is that the analysis of the second effective value signal to determine the end action time of the roller screw control rod driving mechanism includes:

[0025] A preset threshold value x2 is set, and the second effective value signal is compared with the preset threshold value x2. Among the second effective value signals corresponding to the acceleration signals collected three times in a row, when the second effective value signals corresponding to two of the acceleration signals collected are both less than the threshold value x2, the time of the first acceleration signal collection is determined to be the end action time of the roller screw control rod drive mechanism.

[0026] A further optimization scheme is that the method of screening out the action time periods that meet the action time requirements includes:

[0027] Get the total data length of all action time periods;

[0028] Set the window length w, compare the total data length in each action time period with the window length w, and determine the action time period with a total data length greater than the window length w as the action time standard.

[0029] A further optimization scheme is to perform feature identification on the action time period that meets the action time standard, and combine all feature identification results to obtain the state of the roller screw control rod driving mechanism, including the method:

[0030] Configure the number of data sub-segments m, determine the sliding window step s according to the window length w and the number of data sub-segments m, and split the data corresponding to the action time period that meets the action time standard into multiple data sub-segments with the sliding window step s; Where OL is the data length between the start and end time points of the control rod drive mechanism.

[0031] Calculate the characteristic value of each data sub-segment, and input the characteristic value into a pre-trained support vector machine model to obtain a characteristic identification result;

[0032] The state of the roller screw control rod drive mechanism is obtained by combining all feature identification results.

[0033] The present solution also provides a roller screw control rod drive mechanism state online monitoring system, which is used to implement the above-mentioned roller screw control rod drive mechanism state online monitoring method, and the system includes:

[0034] An acquisition module, used for acquiring acceleration signals of a roller screw control rod driving mechanism;

[0035] an analysis module, configured to calculate a first effective value signal and a second effective value signal based on the acceleration signal, analyze the first effective value signal and the second effective value signal to determine a start action time of the roller screw control rod drive mechanism, and analyze the second effective value signal to determine an end action time of the roller screw control rod drive mechanism;

[0036] A statistical screening module is used to count the start action time and the end action time of the roller screw control rod drive mechanism, determine the action time period of the roller screw control rod drive mechanism, and screen out the action time period that meets the action time standard;

[0037] The identification module is used to perform feature identification on the action time period that meets the action time standard, and obtain the state of the roller screw control rod driving mechanism by combining all feature identification results.

[0038] The present solution also provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned method for online monitoring of the state of a roller screw control rod drive mechanism.

[0039] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0040] 1. The present invention provides a method, system and medium for online monitoring the state of a roller screw control rod drive mechanism. Based on the traditional control rod drive mechanism state monitoring technology, the present invention makes improvements in the method. For a roller screw type control rod drive mechanism, a first effective value signal and a second effective value signal are calculated based on the acceleration signal to determine the time points of the start and end actions of the roller screw type control rod drive mechanism, screen out the action time periods that meet the action time standards for feature identification, and finally obtain the state of the roller screw control rod drive mechanism by combining all feature identification results. The present invention solves the problem that the state of the roller screw type control rod drive mechanism cannot be monitored online, and early faults of the roller screw type control rod drive mechanism are discovered in advance, which effectively ensures the safety and reliability of the control rod drive mechanism.

[0041] 2. The present invention provides a method, system and medium for online monitoring of the state of a roller screw control rod drive mechanism; in order to solve the problem of inconsistent operation time of a roller screw control rod drive mechanism, the data of the action time period whose action time meets the standard is screened out by a sliding window method, and the data is split into data sub-segments, and the characteristic value of each data sub-segment is calculated, and the characteristic value is input into a pre-trained support vector machine model to obtain a feature identification result; the data of the action time period whose action time meets the standard is screened out for identification, thereby ensuring the stability of the characteristic value and improving the identification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0043] Figure 1 It is a flow chart of the method for online monitoring of the state of the roller screw control rod driving mechanism;

[0044] Figure 2 It is a structural diagram of the online monitoring system for the roller screw control rod driving mechanism status;

[0045] Figure 3 Schematic diagram of the roller screw control rod drive mechanism structure.

[0046] In the attached picture:

[0047] 1- rotor component; 2- pressure-resistant shell; 3- counterweight; 4- acceleration sensor. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0049] During the normal operation of the reactor, the roller screw type control rod drive mechanism is not in a real-time operating state, and the length of time for the real-time operating state of the roller screw type control rod drive mechanism is uncertain. When the roller screw type control rod drive mechanism is in a static state, the operating state of the drive mechanism cannot be identified. Only when the drive mechanism is in an operating state can its state be identified. Therefore, it is necessary to capture the time points when the roller screw type control rod drive mechanism starts and ends the action. However, since there are no other trigger signals to determine the start and end points of the signal, it is necessary to start from the waveform itself; at the same time, the continuous acquisition time of the roller screw type control rod drive mechanism is inconsistent, and the stable characteristic quantity cannot be used to identify the state of the roller screw type control rod drive mechanism. In view of this, the present invention provides the following embodiments to solve the above technical problems:

[0050] Example 1

[0051] This embodiment provides a method for online monitoring of the state of a roller screw control rod driving mechanism, such as Figure 1 As shown, including:

[0052] Step 1: Collect the acceleration signal of the roller screw control rod driving mechanism; specifically including the method:

[0053] An acceleration sensor is arranged on the bearing seat of the roller screw control rod driving mechanism to continuously collect acceleration signals. The data length of each collected acceleration signal is L, and the time of each collection of the acceleration signal is recorded.

[0054] Step 2: Calculate a first effective value signal and a second effective value signal based on the acceleration signal, analyze the first effective value signal and the second effective value signal to determine the start action time of the roller screw control rod drive mechanism, and analyze the second effective value signal to determine the end action time of the roller screw control rod drive mechanism;

[0055] The method of calculating the first effective value signal and the second effective value signal based on the acceleration signal includes:

[0056] Get acceleration signal;

[0057] Calculate the square of the effective value of the acceleration signal collected each time to obtain a first effective value signal;

[0058] Set the number of elements N in the storage queue A to store the first valid value signal in the storage queue A; update the storage queue A when the storage queue A is full of elements: the next element is arranged at the end of the storage queue A, and the element that entered the queue first is squeezed out;

[0059] After the storage queue A is full of elements and after updating, the mean is calculated to obtain a second effective value signal.

[0060] The acceleration signal collected for the nth time calculates a first effective value signal q1n (when the storage queue A is full of elements, it is q11, q12, q13, ..., q1n), and a second effective value signal q2n is calculated accordingly based on the storage queue A; the acceleration signal collected for the n+1th time calculates a first effective value signal q1(n+1), (when the storage queue A is full of elements, it is q12, q13, q14, ..., q1(n+1)), and there will be a corresponding second effective value signal q2n.

[0061] Since the roller screw type control rod drive mechanism does not produce any action when the reactor is operating normally, the signal measured by the acceleration sensor arranged in the drive mechanism is a background noise signal with a low amplitude. When the reactor needs power regulation, the control rod drive mechanism motor moves, thereby forcing the drive mechanism to vibrate, and the acceleration sensor captures a signal with a high vibration amplitude. At this time, the second effective value of the signal changes little, and the first effective value changes greatly. The starting time of the signal can be determined by the ratio, and the first effective value signal and the second effective value signal are analyzed to determine the starting action time of the roller screw control rod drive mechanism, including:

[0062] Calculating a quotient of the first effective value signal and the second effective value signal;

[0063] A threshold value x1 is preset, and the quotient is compared with the threshold value x1; among the quotients corresponding to the acceleration signals collected three times in succession, when the quotients corresponding to two times of the acceleration signals collected are both less than the threshold value x1, the time of the last acceleration signal collection is determined to be the start action time of the roller screw control rod drive mechanism.

[0064] Similarly, when the control rod drive mechanism ends, the acceleration measurement amplitude decreases significantly, and the more robust second effective value is selected as the judgment criterion. The smaller the signal, the more obvious the end of the control rod drive mechanism action.

[0065] Therefore, analyzing the second effective value signal to determine the end action time of the roller screw control rod driving mechanism includes:

[0066] A preset threshold value x2 is set, and the second effective value signal is compared with the preset threshold value x2. Among the second effective value signals corresponding to the acceleration signals collected three times in a row, when the second effective value signals corresponding to two of the acceleration signals collected are both less than the threshold value x2, the time of the first acceleration signal collection is determined to be the end action time of the roller screw control rod drive mechanism.

[0067] Step 3, counting the start action time and the end action time of the roller screw control rod driving mechanism, determining the action time period of the roller screw control rod driving mechanism, and screening out the action time period that meets the action time standard;

[0068] The method of screening out the action time period that meets the action time standard includes:

[0069] Get the total data length of all action time periods;

[0070] Set the window length w, compare the total data length in each action time period with the window length w, and determine the action time period with a total data length greater than the window length w as the action time standard.

[0071] Step 4: Perform feature identification on the action time period that meets the action time standard, and combine all feature identification results to obtain the state of the roller screw control rod driving mechanism.

[0072] Included methods:

[0073] Configure the number of data sub-segments m, determine the sliding window step s according to the window length w and the number of data sub-segments m, and split the data corresponding to the action time period that meets the action time standard into multiple data sub-segments with the sliding window step s; Where OL is the data length between the start and end time points of the control rod drive mechanism.

[0074] Calculate the characteristic value of each data sub-segment, and input the characteristic value into a pre-trained support vector machine model to obtain a characteristic identification result;

[0075] The state of the roller screw control rod drive mechanism is obtained by combining all feature identification results.

[0076] Also includes step five:

[0077] An abnormal result threshold is set, all feature identification results are obtained, each feature identification result is compared with the abnormal result threshold, and the number N of feature identification results exceeding the abnormal result threshold is counted. When the number N accounts for more than 75% of the total number of feature identification results, the roller screw control rod drive mechanism is judged to be faulty; otherwise, the roller screw control rod drive mechanism is judged to be normal.

[0078] The present embodiment provides a method for online monitoring of the state of a roller screw control rod drive mechanism. Based on the traditional control rod drive mechanism state monitoring technology, an improvement is made in the method. For the roller screw type control rod drive mechanism, a first effective value signal and a second effective value signal are calculated based on the acceleration signal, so as to determine the time points of the start and end actions of the roller screw type control rod drive mechanism, screen out the action time periods that meet the action time standards for feature identification, and finally obtain the state of the roller screw control rod drive mechanism by combining all feature identification results. The problem that the state of the roller screw type control rod drive mechanism cannot be monitored online is solved, and the early faults of the roller screw type control rod drive mechanism are discovered in advance, which effectively ensures the safety and reliability of the control rod drive mechanism.

[0079] At the same time, in order to solve the problem of inconsistent operation time of the roller screw control rod drive mechanism, the data of the action time period that meets the action time standard is screened out through a sliding window method, and split into data sub-segments, the eigenvalue of each data sub-segment is calculated, and the eigenvalue is input into the pre-trained support vector machine model to obtain the feature identification result; the data of the action time period that meets the action time standard is screened out for identification, which ensures the stability of the eigenvalue and improves the identification accuracy.

[0080] Example 2

[0081] This embodiment provides an online monitoring system for the state of a roller screw control rod driving mechanism, which is used to implement the online monitoring method for the state of a roller screw control rod driving mechanism described in Embodiment 1. Figure 2 As shown, the system comprises:

[0082] The acquisition module is used to acquire the acceleration signal of the roller screw control rod driving mechanism; the acquisition module includes an acceleration sensor, and the acceleration sensor is mounted on the rotor of the roller screw control rod driving mechanism to acquire the acceleration signal.

[0083] an analysis module, configured to calculate a first effective value signal and a second effective value signal based on the acceleration signal, analyze the first effective value signal and the second effective value signal to determine a start action time of the roller screw control rod drive mechanism, and analyze the second effective value signal to determine an end action time of the roller screw control rod drive mechanism;

[0084] A statistical screening module is used to count the start action time and the end action time of the roller screw control rod drive mechanism, determine the action time period of the roller screw control rod drive mechanism, and screen out the action time period that meets the action time standard;

[0085] The identification module is used to perform feature identification on the action time period that meets the action time standard, and obtain the state of the roller screw control rod driving mechanism by combining all feature identification results.

[0086] Example 3

[0087] This embodiment provides a computer-readable medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method for online monitoring of the state of the roller screw control rod drive mechanism as described in Embodiment 1;

[0088] Step 1, collecting the acceleration signal of the roller screw control rod driving mechanism;

[0089] Step 2: Calculate a first effective value signal and a second effective value signal based on the acceleration signal, analyze the first effective value signal and the second effective value signal to determine the start action time of the roller screw control rod drive mechanism, and analyze the second effective value signal to determine the end action time of the roller screw control rod drive mechanism;

[0090] Step 3, counting the start action time and the end action time of the roller screw control rod driving mechanism, determining the action time period of the roller screw control rod driving mechanism, and screening out the action time period that meets the action time standard;

[0091] Step 4: Perform feature identification on the action time period that meets the action time standard, and combine all feature identification results to obtain the state of the roller screw control rod driving mechanism.

[0092] Step five, set the abnormal result threshold, obtain all feature identification results, compare each feature identification result with the abnormal result threshold, and count the number N of feature identification results that exceed the abnormal result threshold. When the number N accounts for more than 75% of the total number of feature identification results, it is determined that the roller screw control rod drive mechanism is faulty; otherwise, it is determined that the roller screw control rod drive mechanism is normal.

[0093] Example 4

[0094] This embodiment describes the present invention in detail with reference to the accompanying drawings and an embodiment of online monitoring of a roller screw type control rod drive mechanism.

[0095] In this embodiment, the method for online monitoring and evaluation of the state of the roller screw type control rod drive mechanism using an acceleration sensor specifically includes:

[0096] First, determine the time points for the start and end of the roller screw type control rod drive mechanism. For the acceleration sensor data arranged in the roller screw type control rod drive mechanism, the acceleration sensor arrangement position is as shown in the attached figure. Figure 3 As shown, the roller screw type control rod drive mechanism includes a rotor component 1, a pressure-resistant shell 2 and a counterweight 3. In this embodiment, the acceleration sensor 4 is arranged on the rotor component 1 to collect acceleration data in real time. The data length L is set to 100 points, and the square of the effective value of the 100 data points is calculated as the first effective value signal q1 j :

[0097]

[0098] where a ij represents the data point i collected for the jth time, i = 1, 2, 3, ..., 100;

[0099] The calculated first effective value signal is sent to the storage queue A. The storage queue A adopts the first-in-first-out strategy. The maximum capacity N of the storage queue A is 10. The mean of all elements in the queue is calculated to obtain the second effective value signal q2:

[0100]

[0101] The acceleration signal collected for the nth time calculates a first effective value signal q1n (when the storage queue A is full of elements, it is q11, q12, q13, ..., q1n), and a second effective value signal q2n is calculated accordingly based on the storage queue A; the acceleration signal collected for the n+1th time calculates a first effective value signal q1(n+1), (when the storage queue A is full of elements, it is q12, q13, q14, ..., q1(n+1)), and there will be a corresponding second effective value signal q2n.

[0102] The quotient of the second effective value signal and the first effective value signal is calculated and compared with the threshold value x1. If the quotient is less than the threshold value twice out of three consecutive times, it is determined that this is the time for the control rod drive mechanism to start action.

[0103] At the same time, the second effective value signal and the first effective value signal continue to be calculated, and the second effective value signal is compared with the threshold value x2. If two out of three consecutive times are less than the threshold value x2, it is determined that the start time of the first long effective value calculation is the end action time of the control rod drive mechanism.

[0104] Secondly, the data is split by sliding window. Assuming that the total data length OL obtained from the start to the end of the control rod drive mechanism is 5300 points, the length w of a single data for state identification calculation is 3000, and the number of data to be split is 10, so the step length s of the sliding window can be calculated as:

[0105]

[0106] Third, each piece of data after splitting is screened to obtain the screening results. The characteristic values ​​of each piece of data are calculated, including acceleration level, kurtosis, spectrum peaks at low frequency bands such as 1x, 2x, 4x and 8x of the signal, and frequency band energy at high frequency bands such as 800-1000Hz and 1000Hz-2000Hz. The characteristic values ​​obtained by the above calculations are used in the pre-trained support vector machine model to implement binary classification calculation and obtain the screening results.

[0107] Finally, the analysis results of each piece of data are synthesized to determine the final state of the equipment. The identification results of each piece of data obtained in step 3 are summarized. In order to consider a certain margin against false alarms, when the number of identified abnormalities accounts for more than 75% of the total number, it is determined to be abnormal, otherwise, it is determined to be normal.

[0108] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for online monitoring of the state of a roller screw control rod driving mechanism, characterized in that: include: Collecting the acceleration signal of the roller screw control rod driving mechanism; Calculating a first effective value signal and a second effective value signal based on the acceleration signal, analyzing the first effective value signal and the second effective value signal to determine a start action time of the roller screw control rod drive mechanism, and analyzing the second effective value signal to determine an end action time of the roller screw control rod drive mechanism; Counting the start action time and the end action time of the roller screw control rod drive mechanism, determining the action time period of the roller screw control rod drive mechanism, and screening out the action time period that meets the action time standard; The characteristics of the action time periods that meet the action time standards are identified, and the state of the roller screw control rod drive mechanism is obtained by combining all the characteristic identification results.

2. The method for online monitoring of the state of the roller screw control rod driving mechanism according to claim 1, characterized in that: Also includes: An abnormal result threshold is set, all feature identification results are obtained, each feature identification result is compared with the abnormal result threshold, and the number N of feature identification results exceeding the abnormal result threshold is counted. When the number N accounts for more than 75% of the total number of feature identification results, the roller screw control rod drive mechanism is judged to be faulty; otherwise, the roller screw control rod drive mechanism is judged to be normal.

3. The method for online monitoring of the state of a roller screw control rod driving mechanism according to claim 1, characterized in that: The collecting of the acceleration signal of the roller screw control rod driving mechanism; Included methods: An acceleration sensor is arranged on the bearing seat of the roller screw control rod driving mechanism to continuously collect acceleration signals. The data length of each collected acceleration signal is L, and the time of each collection of the acceleration signal is recorded.

4. The method for online monitoring of the state of a roller screw control rod driving mechanism according to claim 1, characterized in that: The method of calculating the first effective value signal and the second effective value signal based on the acceleration signal includes: Get acceleration signal; Calculate the square of the effective value of the acceleration signal collected each time to obtain a first effective value signal; Set the number of elements N in the storage queue A to store the first valid value signal in the storage queue A; update the storage queue A when the storage queue A is full of elements: the next element is arranged at the end of the storage queue A, and the element that entered the queue first is squeezed out; After the storage queue A is full of elements and after updating, the mean is calculated to obtain a second effective value signal.

5. The method for online monitoring of the state of the roller screw control rod driving mechanism according to claim 4, characterized in that: The step of analyzing the first effective value signal and the second effective value signal to determine the start action time of the roller screw control rod driving mechanism includes: Calculating a quotient of the first effective value signal and the second effective value signal; A threshold value x1 is preset, and the quotient is compared with the threshold value x1; among the quotients corresponding to the acceleration signals collected three times in succession, when the quotients corresponding to two times of the acceleration signals collected are both less than the threshold value x1, the time of the last acceleration signal collection is determined to be the start action time of the roller screw control rod drive mechanism.

6. The method for online monitoring of the state of the roller screw control rod driving mechanism according to claim 5, characterized in that: The step of analyzing the second effective value signal to determine the end action time of the roller screw control rod driving mechanism includes: A preset threshold value x2 is set, and the second effective value signal is compared with the preset threshold value x2. Among the second effective value signals corresponding to the acceleration signals collected three times in a row, when the second effective value signals corresponding to two of the acceleration signals collected are both less than the threshold value x2, the time of the first acceleration signal collection is determined to be the end action time of the roller screw control rod drive mechanism.

7. The method for online monitoring of the state of a roller screw control rod driving mechanism according to claim 1, characterized in that: The method of screening out the action time period that meets the action time standard includes: Get the total data length of all action time periods; Set the window length w, compare the total data length in each action time period with the window length w, and determine the action time period with a total data length greater than the window length w as the action time standard.

8. The method for online monitoring of the state of a roller screw control rod driving mechanism according to claim 8, characterized in that: The method of performing feature identification on the action time period that meets the action time standard and combining all feature identification results to obtain the state of the roller screw control rod driving mechanism includes: Configure the number of data sub-segments m, determine the sliding window step s according to the window length w and the number of data sub-segments m, and split the data corresponding to the action time period that meets the action time standard into multiple data sub-segments using the sliding window step s; Calculate the characteristic value of each data sub-segment, and input the characteristic value into a pre-trained support vector machine model to obtain a characteristic identification result; The state of the roller screw control rod drive mechanism is obtained by combining all feature identification results.

9. The roller screw control rod drive mechanism state online monitoring system is characterized by: The system is used to implement the method for online monitoring of the state of the roller screw control rod drive mechanism according to any one of claims 1 to 8, and comprises: An acquisition module, used for acquiring acceleration signals of a roller screw control rod driving mechanism; an analysis module, configured to calculate a first effective value signal and a second effective value signal based on the acceleration signal, analyze the first effective value signal and the second effective value signal to determine a start action time of the roller screw control rod drive mechanism, and analyze the second effective value signal to determine an end action time of the roller screw control rod drive mechanism; A statistical screening module is used to count the start action time and the end action time of the roller screw control rod drive mechanism, determine the action time period of the roller screw control rod drive mechanism, and screen out the action time period that meets the action time standard; The identification module is used to perform feature identification on the action time period that meets the action time standard, and obtain the state of the roller screw control rod driving mechanism by combining all feature identification results.

10. A computer readable medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the method for online monitoring of the state of a roller screw control rod drive mechanism as described in any one of claims 1 to 8.

Images