Intelligent monitoring system and method for on-off state of protection hard pressing plate

Abstract

The invention provides an intelligent monitoring system and method for the on-off state of a protection hard pressing plate, and relates to the technical field of protection device monitoring, mechanical actions of the protection hard pressing plate are converted into electric signals to be collected, the collected on-off signal of each protection hard pressing plate is pushed, and the on-off state of each protection hard pressing plate is monitored according to the obtained on-off state signal of each protection hard pressing plate. Establishing a two-dimensional pushing duration matrix according to the obtained pushing duration of each commissioning and retreating state signal, analyzing the two-dimensional pushing duration matrix, finding out a plurality of abnormal points, determining whether the corresponding protection hard pressing plate has commissioning and retreating abnormity according to the plurality of abnormal points, and if yes, sending the commissioning and retreating state signal to the corresponding protection hard pressing plate. And the corresponding protection hard strap switching abnormal condition is displayed, so that remote real-time monitoring is carried out, the detection is accurate, an operator does not need to go to the field for inspection, and the detection efficiency is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of protection device monitoring, and in particular relates to an intelligent monitoring system and method for the insertion and withdrawal status of a protection hard pressure plate. Background Art

[0002] Transformers, circuit breakers, lines and other primary equipment need to be equipped with protection devices. The protection hard pressure plate is an important part of the protection device logic and export function realization. It is the bridge and link between the protection device and the external wiring, which is related to whether the protection function and action output can function normally. When the primary equipment fails, the correct investment and withdrawal of the protection hard pressure plate can ensure the correct operation of the protection device and avoid further expansion of the accident. Therefore, the investment or withdrawal status of the protection hard pressure plate is crucial to the realization of the protection device function. At present, the detection of the action status of the protection hard pressure plate is completed by the inspection and inspection of the staff, but the actual level of the staff is uneven, and there are cases of missed inspection and wrong inspection. With the continuous development of the scale of the power grid, the inspection workload of the protection hard pressure plate investment and withdrawal is increasing. The substation accidents caused by the wrong investment or omission of the protection hard pressure plate are increasing year by year. Therefore, it is necessary to find a method for real-time monitoring of the protection hard pressure plate. Summary of the invention

[0003] In view of this, the present invention provides an intelligent monitoring method for real-time monitoring of the insertion and withdrawal status of a protective hard pressure plate.

[0004] It is also necessary to provide an intelligent monitoring system for the insertion and withdrawal status of the protective hard pressure plate.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A method for intelligently monitoring the insertion and withdrawal status of a protective hard pressure plate comprises the following steps:

[0007] Step 1: Collect the throw-in and throw-out signals of each protective hard pressure plate;

[0008] Step 2: Pushing the collected activation and deactivation signals of each protective hard pressure plate;

[0009] Step 3: Analyze the abnormal situation based on the acquired throw-in and throw-out signals of each protective hard pressure plate;

[0010] The step three specifically includes the following steps:

[0011] S1: Obtain the throw-in / out status signal of each protective hard pressure plate;

[0012] S2: establishing a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired;

[0013] S3: Analyze the two-dimensional push duration matrix to find multiple abnormal points;

[0014] S4: Determine whether the corresponding protective hard pressure plate has an abnormality in the insertion and withdrawal according to the multiple abnormal points;

[0015] S5: Displaying the abnormal situation of the corresponding protective hard pressure plate.

[0016] Preferably, the step S2 establishes a two-dimensional push duration matrix according to the acquired push duration of each investment / exit status signal, and specifically comprises the following steps:

[0017] S21: extracting the time of the throw-in / out state signal of each protective hard pressure plate according to the acquired throw-in / out state signal of each protective hard pressure plate;

[0018] S22: according to the extracted time of the throw-in / out state signal of each protective hard pressure plate, the time difference of adjacent signals is calculated to obtain the push duration;

[0019] S23: sorting the calculated push durations, and constructing a two-dimensional push duration matrix after sorting;

[0020] Among them: the row represents the number of each protective hard pressure plate, and the column represents the order of push duration;

[0021] S24: Fill the empty cells in the two-dimensional push duration matrix using a linear interpolation method.

[0022] Preferably, the step S3 analyzes the two-dimensional push duration matrix to find multiple abnormal points, specifically comprising the following steps:

[0023] S31: Setting the push cycle and probability threshold;

[0024] S32: extracting duration data within a push cycle from the two-dimensional push duration matrix, analyzing it, and calculating the mean and standard deviation of the duration data within the push cycle;

[0025] S33: Calculate the probability corresponding to each duration data, and find the abnormal point whose probability is lower than the threshold;

[0026] S34: Analyze the entire two-dimensional push duration matrix in a sliding push cycle to find all abnormal points in the two-dimensional push duration matrix.

[0027] Preferably, the step S4 of determining whether the corresponding protective hard pressing plate has any abnormality in insertion and withdrawal according to the multiple abnormal points comprises the following steps:

[0028] S41: Record the number of the protective hard pressing plate corresponding to each abnormal point in the two-dimensional push duration matrix;

[0029] S42: Counting the number of abnormal points corresponding to each protective hard pressing plate number;

[0030] S43: When the number of abnormal points exceeds the abnormality judgment threshold, it is determined that an abnormality exists.

[0031] Preferably, in S33, the probability being lower than a threshold is specifically lower than 0.9 to 0.95.

[0032] Preferably, in S43, the judgment threshold is that the number of abnormal points of the protective hard pressure plate is greater than 10%.

[0033] A protective hard pressure plate throw and withdraw status intelligent monitoring system, implements the protective hard pressure plate throw and withdraw status intelligent monitoring method as described above, the protective hard pressure plate throw and withdraw status intelligent monitoring system comprises an acquisition module, a communication module, and an abnormality analysis module, the abnormality analysis module comprises: a control unit, a display unit, the acquisition module is electrically connected to the communication module, the communication module is electrically connected to the control unit, and the control unit is electrically connected to the display unit.

[0034] Preferably, the abnormality analysis module further comprises a communication convergence unit, one end of the communication convergence unit is connected to the control unit, and the other end of the communication convergence unit is connected to the display unit.

[0035] Preferably, the acquisition module comprises an acquisition unit and a power supply unit, the acquisition unit is connected to the power supply unit, the acquisition unit is connected to the communication module, and the power supply unit is connected to the communication module.

[0036] Preferably, the acquisition unit is a micro switch.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention provides an intelligent monitoring method for the insertion and withdrawal status of a protective hard pressure plate. The method converts the mechanical action of the protective pressure plate into an electrical signal for collection, pushes the collected insertion and withdrawal signals of each protective hard pressure plate, obtains the insertion and withdrawal status signal of each protective hard pressure plate, and establishes a two-dimensional push duration matrix according to the push duration of each obtained insertion and withdrawal status signal. The two-dimensional push duration matrix is ​​analyzed to find multiple abnormal points. According to the multiple abnormal points, it is determined whether the corresponding protective hard pressure plate has insertion and withdrawal abnormalities, and the corresponding insertion and withdrawal abnormalities of the protective hard pressure plate are displayed for remote real-time monitoring. The detection is accurate, and there is no need for operators to go to the site for inspection, and the detection efficiency is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the steps of the intelligent monitoring method for protecting the insertion and withdrawal status of the hard pressure plate.

[0040] Figure 2 Schematic diagram of the intelligent monitoring system for protecting the insertion and withdrawal status of the hard pressure plate.

[0041] In the figure: an intelligent monitoring system 10 for the insertion and withdrawal status of a protective hard pressure plate, a collection module 100, a collection unit 110, a power supply unit 120, a communication module 200, an abnormality analysis module 300, a control unit 310, a communication convergence unit 320, and a display unit 330. DETAILED DESCRIPTION

[0042] The technical solutions and technical effects of the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings of the present invention.

[0043] like Figure 1 As shown, a method for intelligently monitoring the insertion and withdrawal status of a protective hard pressure plate comprises the following steps:

[0044] Step 1: Collect the throw-in and throw-out signals of each protective hard pressure plate;

[0045] Step 2: Pushing the collected activation and deactivation signals of each protective hard pressure plate;

[0046] Step 3: Analyze the abnormal situation based on the acquired throw-in and throw-out signals of each protective hard pressure plate;

[0047] The step three specifically includes the following steps:

[0048] S1: Obtain the throw-in / out status signal of each protective hard pressure plate;

[0049] S2: establishing a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired;

[0050] S3: Analyze the two-dimensional push duration matrix to find multiple abnormal points;

[0051] S4: Determine whether the corresponding protective hard pressure plate has an abnormality in the insertion and withdrawal according to the multiple abnormal points;

[0052] S5: Displaying the abnormal situation of the corresponding protective hard pressure plate.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The present invention provides an intelligent monitoring method for the insertion and withdrawal status of a protective hard pressure plate. The method converts the mechanical action of the protective pressure plate into an electrical signal for collection, pushes the collected insertion and withdrawal signals of each protective hard pressure plate, obtains the insertion and withdrawal status signal of each protective hard pressure plate, and establishes a two-dimensional push duration matrix according to the push duration of each obtained insertion and withdrawal status signal. The two-dimensional push duration matrix is ​​analyzed to find multiple abnormal points. According to the multiple abnormal points, it is determined whether the corresponding protective hard pressure plate has insertion and withdrawal abnormalities, and the corresponding insertion and withdrawal abnormalities of the protective hard pressure plate are displayed for remote real-time monitoring. The detection is accurate, and there is no need for operators to go to the site for inspection, and the detection efficiency is high.

[0055] Furthermore, the step S2 establishes a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired, and specifically includes the following steps:

[0056] S21: extracting the time of the throw-in / out state signal of each protective hard pressure plate according to the acquired throw-in / out state signal of each protective hard pressure plate;

[0057] S22: According to the extracted signal time of each protective hard pressure plate, the time difference of adjacent signals is calculated to obtain the push duration. For example, the nth signal time is T n , the n+1th time is T n+1 , then the push duration of the nth signal is T n+1 -T n ;

[0058] S23: sorting the calculated push durations of the various signals according to the numbers of the protective hard pressure plates and the monitoring time, and constructing a two-dimensional push duration matrix after sorting;

[0059] Among them: the row represents the number of each protective hard pressure plate, and the column represents the order of push duration;

[0060] S24: Filling the empty cells in the two-dimensional push duration matrix using a linear interpolation method;

[0061] S25: Save the filled two-dimensional push duration matrix to a database or a memory structure.

[0062] The push duration of each signal is extracted from the original investment and withdrawal status signal, and a complete two-dimensional push duration matrix is ​​constructed to lay the foundation for subsequent analysis and processing.

[0063] Furthermore, the S3 analyzes the two-dimensional push duration matrix to find multiple abnormal points, specifically including the following steps:

[0064] S31: Setting the push cycle and probability threshold, for example, the push cycle is 5 investment and withdrawal status signals, and the probability threshold is 3 times;

[0065] S32: extracting m duration data of a push cycle from the two-dimensional push duration matrix, performing Bayesian analysis, assuming that the duration data conforms to a normal distribution, and calculating the mean and standard deviation of the duration data within the push cycle;

[0066] S33: For each duration data, calculate its probability of occurrence under the current mean and standard deviation. If the probability corresponding to a certain duration data is lower than the threshold, calculate the probability corresponding to each duration data, and mark the data as an abnormal point.

[0067] S34: sliding the push cycle on the two-dimensional push duration matrix, repeating steps S31-S33, and finding all abnormal points in the two-dimensional push duration matrix;

[0068] The Bayesian method is used to use the probability model to judge the abnormal situation of the data, and the probability threshold is used to detect abnormal data points that obviously deviate from the normal distribution. The sliding window method can be used to comprehensively analyze the entire two-dimensional push duration matrix. It is simple and fast, and avoids affecting status monitoring. At the same time, it can judge the status false alarm caused by the failure of the micro switch itself.

[0069] Further, the step S4 determines whether the corresponding protective hard pressing plate has any abnormality in the insertion and withdrawal according to the multiple abnormal points, including the following steps:

[0070] S41: Record the number of the protective hard pressing plate corresponding to each abnormal point in the two-dimensional push duration matrix;

[0071] S42: Counting the number of abnormal points corresponding to each protective hard pressing plate number;

[0072] S43: If the number of abnormal points of a certain protective hard pressure plate exceeds the preset abnormality judgment threshold, determine whether there is an abnormality in the deployment and withdrawal of the protective hard pressure plate; for example, if the abnormal points of a certain protective hard pressure plate appear in multiple consecutive time periods of the two-dimensional push duration matrix, then there is an abnormality in the deployment and withdrawal of the protective hard pressure plate, and the abnormal deployment and withdrawal of the protective hard pressure plate will be sent to the operation and maintenance personnel; by analyzing the number and distribution of abnormal points, combined with the system topology, the abnormal protective hard pressure plate can be accurately determined, providing a repair basis for the operation and maintenance personnel.

[0073] Furthermore, a method for determining whether there is an abnormality in the corresponding protective hard pressure plate insertion and withdrawal is to use a pre-trained anomaly point-anomaly model. The specific steps for establishing and training the anomaly point-anomaly model include: constructing a sample data set corresponding to the anomaly point features and anomaly labels; performing feature engineering on the sample data to extract an effective feature subset; selecting a model that processes time series data and has good generalization capabilities, and determining the network structure and hyperparameters; using the sample data set to train the model, comparing the error between the model output result and the true label through the loss function, and updating the network parameters using the gradient descent algorithm; using the validation set to evaluate the model indicators, and adjusting the hyperparameters to improve the generalization performance; after obtaining an anomaly point-anomaly model with excellent performance, saving parameters and other information; and using the model to predict anomalies for newly input anomaly point features.

[0074] The specific steps for establishing and training the above model are as follows:

[0075] 1) Construct anomaly point-abnormal sample data set

[0076] Collect historical abnormal sample data, including abnormal point features and corresponding abnormal labels. The abnormal point features can include the value of the point, context time series and other information. The abnormal label indicates the abnormal category, such as the protection hard pressure plate is mistakenly deployed or missed.

[0077] 2) Feature Engineering

[0078] Perform feature engineering on the sample data of abnormal points to extract effective feature subsets, such as statistical features and frequency domain features of time series data, reduce redundant and invalid features, and improve model training effects.

[0079] 3) Model selection

[0080] Select a model that is suitable for processing time series data and has the ability to generalize for few-sample problems, such as recursive neural networks such as LSTM and GRU, and determine the network structure and hyperparameters.

[0081] 4) Model training

[0082] Using the sample data set after feature engineering, input the outlier features and outlier labels in turn, compare the error between the model output and the true label through the loss function, use the gradient descent algorithm to update the network parameters, and complete the model training.

[0083] 5) Model Evaluation

[0084] Use the validation set to evaluate model accuracy metrics such as precision, recall, F1 score, etc., and adjust hyperparameters to improve model generalization performance. Repeat the training and evaluation cycle.

[0085] 6) Model saving

[0086] After obtaining the outlier point-outlier model with excellent performance, save it and back up the model structure, parameters and other information.

[0087] 7) Model prediction

[0088] For the newly input abnormal points, their features are extracted and input into the trained abnormal point-anomaly model. The model outputs the abnormality prediction category and determines the abnormality type of the protective hard pressure plate.

[0089] By constructing a sample data set corresponding to anomaly points and anomalies and applying a deep learning model for feature learning, we can obtain an anomaly point-anomaly prediction model with good generalization performance and flexibility. This model can distinguish which specific protective hard pressure plate has an abnormality, greatly improving the efficiency of fault analysis. It eliminates the need for manual inspections and enables efficient and accurate anomaly diagnosis.

[0090] Furthermore, in S33, the probability being lower than a threshold is specifically lower than 0.9 to 0.95.

[0091] Furthermore, in S43, the judgment threshold is that the number of abnormal points of the protective hard pressure plate is greater than 10%.

[0092] like Figure 2 As shown, a protective hard pressure plate throw-in and throw-out state intelligent monitoring system 10 implements the protective hard pressure plate throw-in and throw-out state intelligent monitoring method as described above, the protective hard pressure plate throw-in and throw-out state intelligent monitoring system 10 comprises a collection module 100, a communication module 200, and an abnormality analysis module 300, the collection module 100 and the communication module 200 are several, the abnormality analysis module 300 comprises: a control unit 310, a display unit 330, several collection modules 100 are respectively connected to several protective hard pressure plates, several collection modules 100 are respectively electrically connected to several communication modules 200, several communication modules 200 are electrically connected to the control unit 310, the control unit 310 is electrically connected to the display unit 330, the collection module 100 collects the throw-in and throw-out state of the protective hard pressure plate, the communication module 200 transmits the information collected by the collection module 100 to the control unit 310, the control unit 310 analyzes and processes the collected information, and the display unit 330 displays the result processed by the control unit 310.

[0093] Specifically, the communication module 200 adopts the LoRa wireless transmission method to transmit the protection hard pressure plate status data, and can transmit the collected data in a timely and complete manner; the control unit 310 adopts the STM32 microcontroller as the data processor core, the program running success rate reaches 100%, and the signal output accuracy rate is 100%.

[0094] Furthermore, the abnormality analysis module 300 further includes a communication convergence unit 320 , one end of which is connected to the control unit 310 , and the other end of which is connected to the display unit 330 .

[0095] Specifically, the communication convergence unit 320 uses RS485 communication to converge data, and the data transmission accuracy reaches 100%.

[0096] Furthermore, the acquisition module 100 includes a acquisition unit 110 and a power supply unit 120 , the acquisition unit 110 is connected to the power supply unit 120 , the acquisition unit 110 is connected to the communication module 200 , and the power supply unit 120 is connected to the communication module 200 .

[0097] Furthermore, the acquisition unit 110 is a micro switch, which can continuously and stably monitor the state of the protective hard pressure plate with an accuracy of 100%. The power supply unit 120 uses a lithium-ion battery as a power source, and the voltage output is stable.

[0098] A specific example of the abnormality analysis module 300 of the present invention is introduced through the following embodiment.

[0099] Variable Description:

[0100] N: number of monitoring devices

[0101] i: monitoring device number, ranging from 1 to N

[0102] x, y: sequence number in the communication connection message

[0103] ·B i : Data receiving buffer of the i-th monitoring device

[0104] ·L i : Cache area B i Received data length

[0105] ·S i : A complete original status signal received from monitoring device i

[0106] M: Push duration matrix, the number of rows is the number of monitoring devices N, and the number of columns is the time series

[0107] ·t j : Timestamp of the jth state signal

[0108] ·Δt j : The push duration difference between two adjacent state signals j and j+1

[0109] w: time window size

[0110] ·p t : Probability threshold for judging irregularity

[0111] μ, σ: Parameters of the normal distribution that the data in the window conforms to

[0112] ·p i : The probability corresponding to the data in the i-th window

[0113] ·n i : The number of irregular points of monitoring device i

[0114] ·N t : The threshold value of the number of irregular points for judging abnormalities

[0115] Formula Description:

[0116] Three-way handshake message: indicates the three-way message exchange for establishing a TCP connection

[0117] ·B i =Array [1~10KB]: defines the buffer data structure

[0118] ·S i = Extraction (B i , L i ): Extract a complete status signal from the buffer

[0119] Status data i =Analysis(S i ): Analyze the status signal to obtain the original data

[0120] ·Δt j =t j+1 -t j :Calculate the time difference of adjacent signals

[0121] · Calculate the mean within the window

[0122] · Calculate the standard deviation within the window

[0123] ·p i =P(M i |μ,σ): Calculate the probability under normal distribution

[0124] ·n i >N t : Determine the abnormal condition of a single protective hard plate

[0125] The specific implementation process is as follows:

[0126] Step 1: The acquisition unit 110 acquires the throw-in / out signal of each protective hard pressure plate.

[0127] Step 2: The communication module 200 pushes the insertion and withdrawal signals of each protective hard pressure plate collected by the collection unit 110.

[0128] Step 3: Analyze abnormal situations based on the activation and deactivation signals of each protective hard pressure plate.

[0129] S1: Obtain the throw-in / out status signal sent by each protective hard pressure plate. Assume that the number of protective hard pressure plates is N, which are numbered 1, 2, ..., N.

[0130] 1. Establish a communication connection

[0131] The control unit 310 establishes a communication connection with each protective hard pressure plate i (i=1, 2, ..., N) through a micro switch and a communication module 200 for subsequent data transmission. The connection establishment follows the TCP three-way handshake protocol:

[0132] The control unit 310 sends a connection request message to the monitoring device (micro switch, communication module 200) ii on the protective hard pressure plate:

[0133] SYN i =(src = host computer, dst = i, seq = x)

[0134] The monitoring device on the protection hard pressure plate responds to the confirmation message: SYNACK i =(src=i, dst=host computer, ack=x+1, seq=y)

[0135] The control unit 310 replies with a confirmation message, and the three-way handshake is completed:

[0136] ACK i =(src = host computer, dst = i, ack = y + 1, seq = x + 1)

[0137] The above process is repeated until the control unit 310 establishes connections with the monitoring devices on all N protective hard pressure plates.

[0138] 2. Allocate receive buffer

[0139] The control unit 310 allocates a receiving buffer area B for each protective hard pressure plate i i , used to temporarily store received data. B i The size is generally 1-10KB.

[0140] B i = array [1~10KB], i = 1, 2, ..., N

[0141] 3. Loop read buffer area

[0142] The control unit 310 starts the parsing thread and reads the data of each protective hard pressure plate in real time. i , do the following:

[0143] Read B i Received data length L i

[0144] Extract a complete original state signal S i

[0145] S i = Extraction (B i , L i )

[0146] Parsing S i , get status data

[0147] Status data i =Analysis(S i )

[0148] Save the parsed data

[0149] Clear B i Waiting for new data

[0150] The above process is repeated to continuously obtain the latest status data of each protective hard pressure plate.

[0151] S2: establishing a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired;

[0152] Construct a two-dimensional push duration matrix M, where:

[0153] The number of rows m = N, indicating N protective hard pressure plates

[0154] The number of columns n represents a discrete time series

[0155] M m×n

[0156] For the jth state signal S j , the timing time is t j

[0157] Calculate the push duration of adjacent signals:

[0158] Δt j =t j+1 -t j

[0159] All Δt j Fill in the corresponding cells in matrix M

[0160] Fill the missing cells in the matrix with linear interpolation

[0161] Finally, the complete push duration matrix M is obtained.

[0162] S3: Analyze the two-dimensional push duration matrix to find multiple abnormal points;

[0163] Set the push cycle w and the abnormal judgment threshold p t

[0164] The sliding push cycle moves on M and judges the data in the current push cycle: Assume that the data in the push cycle conforms to the normal distribution N (μ, σ 2 )σ

[0165] Calculate the mean μ and standard deviation σ of the push cycle data

[0166]

[0167] For each data M in the push cycle i , calculate its probability p i :

[0168]

[0169] If p i <p t , then M i For abnormal points

[0170] Repeat the above process and record all abnormal points.

[0171] S4: Determine whether the corresponding protective hard pressure plate has an abnormality in the insertion and withdrawal according to the multiple abnormal points;

[0172] Record the device number i corresponding to each abnormal point

[0173] Count the number of outliers n for each number i i

[0174] Set the abnormality judgment threshold N t , if n i >N t , then the protection hard pressure plate i is judged to be abnormal

[0175] Determine the root cause of the anomaly based on the distribution characteristics of the anomaly points:

[0176] If they are concentrated in a certain time period, it is determined that there is an abnormality in the protection of the hard pressure plate.

[0177] If they are dispersed, it is judged that the protective hard pressure plate is normally deployed.

[0178] S5: Displaying the abnormal situation of the corresponding protective hard pressure plate;

[0179] Output abnormal alarm information, including number i and abnormal module.

[0180] In summary, by statistically analyzing abnormal points and combining them with system topology information, the abnormal situations of the protective hard pressure plate can be accurately determined.

[0181] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent monitoring method for the insertion and withdrawal status of a protective hard pressure plate, characterized in that: The following steps are involved: Step 1: Collect the throw-in and throw-out signals of each protective hard pressure plate; Step 2: Pushing the collected throw-in and throw-out signals of each protective hard pressure plate; Step 3: Analyze the abnormal situation based on the acquired throw-in and throw-out signals of each protective hard pressure plate; The step three specifically includes the following steps: S1: Obtain the throw-in / out status signal of each protective hard pressure plate; S2: establishing a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired; S3: Analyze the two-dimensional push duration matrix to find multiple abnormal points; S4: Determine whether the corresponding protective hard pressure plate has an abnormality in the insertion and withdrawal according to the multiple abnormal points; S5: Displaying the abnormal situation of the corresponding protective hard pressure plate.

2. The intelligent monitoring method for the insertion and withdrawal status of the protective hard pressure plate according to claim 1 is characterized in that: The step S2 establishes a two-dimensional push duration matrix according to the push duration of each of the investment and withdrawal status signals acquired, and specifically comprises the following steps: S21: extracting the time of the throw-in / out state signal of each protective hard pressure plate according to the acquired throw-in / out state signal of each protective hard pressure plate; S22: according to the extracted time of the throw-in / out state signal of each protective hard pressure plate, the time difference of adjacent signals is calculated to obtain the push duration; S23: sorting the calculated push durations, and constructing a two-dimensional push duration matrix after sorting; Among them: the row represents the number of each protective hard pressure plate, and the column represents the order of push duration; S24: Fill the empty cells in the two-dimensional push duration matrix using a linear interpolation method.

3. The intelligent monitoring method for the insertion and withdrawal status of the protective hard pressure plate according to claim 2 is characterized in that: S3 analyzes the two-dimensional push duration matrix to find multiple abnormal points, which specifically includes the following steps: S31: Setting the push cycle and probability threshold; S32: extracting duration data within a push cycle from the two-dimensional push duration matrix, analyzing it, and calculating the mean and standard deviation of the duration data within the push cycle; S33: Calculate the probability corresponding to each duration data, and find the abnormal point whose probability is lower than the threshold; S34: Analyze the entire two-dimensional push duration matrix in a sliding push cycle to find all abnormal points in the two-dimensional push duration matrix.

4. The intelligent monitoring method for the insertion and withdrawal status of the protective hard pressure plate according to claim 3 is characterized in that: The step S4 determines whether the corresponding protective hard pressing plate has an abnormality in the insertion and withdrawal according to the multiple abnormal points, and includes the following steps: S41: Record the number of the protective hard pressing plate corresponding to each abnormal point in the two-dimensional push duration matrix; S42: Counting the number of abnormal points corresponding to each protective hard pressing plate number; S43: When the number of abnormal points exceeds the abnormality judgment threshold, it is determined that an abnormality exists.

5. The intelligent monitoring method for the insertion and withdrawal status of the protective hard pressure plate according to claim 3 is characterized in that: In S33, the probability being lower than a threshold is specifically lower than 0.9 to 0.

95.

6. The intelligent monitoring method for the insertion and withdrawal status of the protective hard pressure plate according to claim 4 is characterized in that: In S43, the judgment threshold is that the number of irregular points on the protective hard pressure plate is greater than 10%.

7. An intelligent monitoring system for the insertion and withdrawal status of a protective hard pressure plate, which implements the intelligent monitoring method for the insertion and withdrawal status of a protective hard pressure plate as claimed in claim 1, characterized in that: The intelligent monitoring system for the insertion and withdrawal status of the protective hard pressure plate includes a collection module, a communication module, and an abnormality analysis module. The abnormality analysis module includes: a control unit and a display unit. The collection module is electrically connected to the communication module, the communication module is electrically connected to the control unit, and the control unit is electrically connected to the display unit.

8. The intelligent monitoring system for the insertion and withdrawal status of the protective hard pressure plate according to claim 7 is characterized in that: The abnormality analysis module further includes a communication convergence unit, one end of which is connected to the control unit, and the other end of which is connected to the display unit.

9. The intelligent monitoring system for the insertion and withdrawal status of the protective hard pressure plate according to claim 7, characterized in that: The acquisition module includes an acquisition unit and a power supply unit. The acquisition unit is connected to the power supply unit. The acquisition unit is connected to the communication module. The power supply unit is connected to the communication module.

10. The intelligent monitoring system for the insertion and withdrawal status of the protective hard pressure plate according to claim 9, characterized in that: The acquisition unit is a micro switch.

Images