SF6 gas density relay calibration signal anti-jitter processing method and system

Abstract

The present application relates to a kind of SF6 gas density relay verification signal anti-jitter processing method and system, comprising: first, the switching quantity signal of relay contact in verification is collected in real time, and the original sequence containing jitter noise is obtained;Then initialize Kalman filtering algorithm, while configuring Snow Ablation Optimizer (Snow Ablation Optimizer, SAO);Then build fitness function containing smoothness, real-time, accuracy index, with its minimization as goal drive Snow Ablation Optimizer optimization, obtain optimal covariance matrix;Optimal parameter is injected into Kalman filtering and resets state, then input original signal point by point, and output anti-jitter signal by recursive optimal estimation;Finally, through threshold value judgment identifies contact action time, records corresponding pressure / density value and completes verification.Compared with prior art, the present application has the advantages of high anti-jitter adaptability and universality, accurate verification result, intelligent verification process and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of SF6 gas density relay calibration, in particular to a SF6 gas density relay calibration signal anti-jitter processing method and system. BACKGROUND

[0002] SF6 gas density relay is a key safety protection element in gas insulated switchgear (GIS), and its performance state is directly related to the reliability of power system operation and the insulation safety of the device itself. In the long-term operation of GIS equipment, due to factors such as mechanical stress, material aging, environmental temperature change, and accidental vibration impact, the performance of the density relay itself may drift or deviate. Therefore, periodic on-site calibration of the density relay has become an important part of power operation and maintenance. During the calibration process, the calibration device needs to simulate the pressure change of the gas in the device, and accurately record the pressure-density value corresponding to the relay contact action, in order to determine whether the alarm, locking and other functions are accurate and reliable.

[0003] However, in actual calibration operation, due to the wear and tear of the internal mechanical structure of the density relay, spring fatigue, contact surface oxidation, and inevitable mechanical collision and rebound during the action process, combined with complex working conditions such as on-site electromagnetic interference and equipment vibration, the contact often produces continuous and rapid on-off jitter at the moment of opening and closing. This contact bounce phenomenon will cause the collected switching signals to produce serious high-frequency oscillation, rather than clear step changes. If not handled, it will cause the following problems: (1) misjudgment of the contact state, affecting the accuracy of the calibration action value; (2) threshold drift, making the action time record unstable; (3) poor repeatability of measurement; (4) false triggering of the logic judgment of the calibration software.

[0004] To solve the above problems, the traditional method mainly starts from the hardware and software two levels. Hardware anti-jitter often uses RC low-pass filter circuit, Schmidt trigger shaping and other ways, which smooths and delays the signal through analog circuit to suppress the burr. While software anti-jitter relies on delay judgment method, window average filtering or setting fixed dead time algorithm, after detecting the contact state change, delaying a fixed time to sample the signal to avoid the jitter period.

[0005] However, with the increase of GIS device voltage level and structural diversification, the type, action mechanism and mechanical response speed of density relay are significantly different, and the traditional anti-shaking method gradually shows its limitations: fixed parameters are difficult to adapt to the action speed difference of different density relays; it cannot track the nonlinear shaking characteristics in real time; the performance decreases when the noise level changes or the action speed changes; it cannot balance the response speed and stability. Patent CN218630097U discloses an improved SF6 gas density relay calibration device, which comprises a to-be-tested SF6 gas density relay and a calibration device host, and further comprises an integrated signal acquisition device; the to-be-tested SF6 gas density relay is connected with the calibration device host, and the calibration device host is connected with the integrated signal acquisition device, which can replace the existing 6-time repeated plug-in operation mode. However, there are still problems such as only optimizing the connection without solving the core precision problem, limited adaptability without intelligent optimization, not realizing intelligent upgrading of the process, and not substantially reducing the long-term operation and maintenance cost.

[0006] Therefore, under the background of the continuous improvement of the current smart grid construction and equipment condition-based maintenance system, there is an urgent need for a new anti-shaking algorithm that can intelligently adapt to different relay characteristics, dynamically adjust the shaking elimination strategy, and has high response speed and high stability. The algorithm should have the ability to learn online from signal characteristics, dynamically identify shaking patterns and noise levels based on real-time acquisition of contact signals, and adaptively optimize shaking elimination parameters based on this to achieve global performance optimization. This not only has a direct significance for improving the accuracy and efficiency of density relay calibration, but also is an important technical support for ensuring the safe operation of GIS equipment and promoting the intelligent operation and maintenance level of power equipment. SUMMARY

[0007] The purpose of the present application is to solve the problems of fixed parameters in traditional anti-shaking methods, such as difficulty in adapting to different density relay action speed differences, inability to track nonlinear shaking characteristics in real time, performance degradation when noise or action speed changes, and difficulty in balancing response speed and stability, while overcoming the deficiencies of related patents such as only optimizing the connection without solving the core precision problem, to provide an SF6 gas density relay calibration signal anti-shaking processing method and system.

[0008] The purpose of the present application can be achieved by the following technical solutions:

[0009] An SF6 gas density relay calibration signal anti-shaking processing method, comprising the following steps:

[0010] Real-time acquisition of relay contact switch quantity signals during SF6 gas density relay calibration process to obtain original signal sequence containing shaking noise, initialization of parameters of Kalman filter algorithm and parameters of snow melting optimizer;

[0011] The fitness function including the smoothness index, the real-time index and the accuracy index is constructed, and the snow melt optimizer is driven to iteratively optimize in a preset search range to obtain the optimal process noise covariance matrix and the optimal observation noise covariance matrix, with the value of the fitness function being minimized as the goal.

[0012] The obtained optimal process noise covariance matrix and optimal observation noise covariance matrix are injected into the Kalman filtering algorithm, and the state of the Kalman filtering algorithm is reset.

[0013] The collected original signal sequence is input point by point into the Kalman filtering algorithm with updated parameters, and the original signal sequence is recursively optimally estimated by the Kalman filtering algorithm to output the processed anti-jitter signal.

[0014] Further, the expression of the fitness function is:

[0015]

[0016] Wherein, F(Q, R) is the fitness function value, Q is the process noise covariance matrix, R is the observation noise covariance matrix, and α, β, γ are weight coefficients and satisfy α+β+γ=1, J smoothness The smoothness index is: J lag The real-time index is: J accuracy The accuracy index is:

[0017] Further, the expression of the smoothness index is:

[0018]

[0019] Wherein, y kalman The output signal sequence of the Kalman filtering after the parameter processing in a preset time window is: Variance The variance calculation function is:

[0020] Further, the expression of the real-time index is:

[0021]

[0022] Wherein, t rise_start And t rise_end The start and end time of the signal jump edge time window is: y kalman The data segment of the Kalman filtering output signal in the jump edge time window is: y reference[...] is the root mean square error calculation function for the data segment of the reference signal in the same jump along the time window.

[0023] Further, the reference signal is a signal obtained by performing strong low-pass filtering on the original signal sequence, or is an ideal non-jittered step signal.

[0024] Further, the expression of the accuracy index is:

[0025]

[0026] Wherein, t steady is the steady-state time window after the signal jump is completed, Mean( y kalman [ t steady ]) is the average value of the Kalman filter output signal in the steady-state time window, y ideal is the ideal logic level value in the steady-state time window, and Mean is the mean calculation function.

[0027] Further, the Kalman filtering algorithm is based on optimal parameters, and recursively optimally estimates the input signal at each time, and outputs a contact state estimation curve with minimum lag.

[0028] Further, the method further comprises:

[0029] The output anti-jitter signal is subjected to threshold judgment, the relay contact action time is identified, the pressure value or density value corresponding to the time is recorded, and SF6 gas density relay calibration is completed.

[0030] The application also provides an SF6 gas density relay calibration signal anti-jitter processing system, comprising:

[0031] A signal acquisition module is configured to acquire the relay contact on-off signal in the SF6 gas density relay calibration process in real time, and output an original signal sequence containing jitter noise;

[0032] A parameter initialization module is configured to initialize the parameters of the Kalman filtering algorithm;

[0033] An optimizer configuration module is configured to initialize the parameters of the snow melting optimizer;

[0034] An adaptability function construction module is configured to construct an adaptability function containing a smoothness index, a real-time index and an accuracy index, and to minimize the value of the adaptability function as an objective, so as to drive the snow melting optimizer to perform iterative optimization in a preset search range, and obtain an optimal process noise covariance matrix and an optimal observation noise covariance matrix.

[0035] a parameter optimization module configured to drive the snowmelt optimizer to iteratively optimize with the objective of minimizing the fitness function, and output an optimal process noise covariance matrix and an optimal observation noise covariance matrix;

[0036] a filter configuration module configured to inject the optimal process noise covariance matrix and the optimal observation noise covariance matrix into the Kalman filter algorithm, and reset the Kalman filter algorithm;

[0037] a filtering processing module configured to input the original signal sequence into the reset Kalman filter algorithm, and perform recursive optimal estimation, and output an anti-jitter signal after processing.

[0038] Further, the application further comprises:

[0039] an action value recording module configured to perform threshold value judgment on the contact state estimation curve based on a preset logic level threshold value, identify a contact action time, and record a corresponding standard pressure value or density value.

[0040] Compared with the prior art, the application has the following beneficial effects:

[0041] 1. High anti-jitter adaptability and universality: the snowmelt optimizer is used to adaptively optimize the core parameters of the Kalman filter, which breaks through the limitation of the traditional fixed parameter filtering method that cannot adapt to different types of SF6 gas density relays and different on-site interference conditions, can dynamically adjust the filtering parameters according to the mechanical characteristics of the relay and the on-site noise intensity, can be compatible with multiple types of density relay calibration scenes, and greatly improves the universality of the algorithm.

[0042] 2. Signal processing considering stability and real-time performance: a multi-objective fitness function including smoothness, real-time performance and accuracy is constructed, and the snowmelt optimizer is driven to optimize the function, so that the stability and real-time performance of signal processing are considered, the high-frequency jitter noise is suppressed, the signal processing lag is reduced to the greatest extent, the identification stability of the contact action time is ensured, and the timeliness of the calibration data is ensured.

[0043] 3. High accuracy of calibration results: the Kalman filter with optimal parameters is used to perform recursive optimal estimation on the jitter signal, which can effectively filter out signal oscillation caused by relay contact bounce and on-site electromagnetic interference, accurately restore the real contact action trajectory, avoid the contact state misjudgment and threshold drift caused by signal jitter in the traditional method, significantly improve the measurement accuracy of the calibration action value and the result repeatability of multiple calibrations, and provide reliable data support for safe operation of GIS equipment.

[0044] 4. The intelligence and automation level of the verification process is high: the full-process technical closed loop from automatic parameter optimization to real-time signal filtering and action value automatic determination is realized, manual filtering parameter adjustment is not needed, the interference of manual operation on the verification result is reduced, and the intelligence and automation degree of the power equipment state maintenance is improved. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 The flowchart of the present application is shown in the figure.

[0046] Figure 2 The structural schematic diagram of the system of the present application is shown in the figure. DETAILED DESCRIPTION

[0047] The present application will be described in detail below in combination with the drawings and specific embodiments. The embodiments are implemented on the premise of the technical scheme of the present application, and detailed implementation modes and specific operation processes are given, but the protection scope of the present application is not limited to the following embodiments.

[0048] Embodiment 1

[0049] The embodiment provides a signal anti-jitter processing method for SF6 gas density relay verification, which is a signal anti-jitter processing method based on SnowAblation Optimizer (SAO) optimization Kalman filter algorithm parameters, applied to the SF6 gas density relay field verification scene, and specifically includes the following steps, and the overall flow is shown in the figure. Figure 1

[0050] First, the verification signal collection and parameter initialization are performed:

[0051] The noise-containing contact signal is collected: in the density relay verification process, the on-off signal of the contact is collected in real time to obtain the original signal sequence containing jitter noise;

[0052] Initialization: the initial values of the state vector and the observation vector of the Kalman filter are initialized, and the initial estimated values of the filter parameter process noise covariance matrix Q and the observation noise covariance matrix R are given, and the parameters of the SnowAblation Optimizer (SAO) are simultaneously initialized, including the population size, the maximum number of iterations and the optimization search range of Q and R.

[0053] Then, the fitness function is constructed and the SnowAblation Optimizer is driven for optimization (SAO optimization):

[0054] Fitness function construction: taking the to-be-evaluated (Q, R) parameters as input, the fitness value representing the pros and cons of the filtering effect is output, and the specific calculation expression is as follows:

[0055]

[0056] ​Wherein, F(Q, R) is the fitness function value, the smaller the value, the better the filtering performance of the group (Q, R) parameters. Alpha, beta, gamma are weight coefficients, used to balance the relative importance of the three optimization objectives, satisfying alpha+beta+gamma=1; J smoothness is the smoothness index, J lag is the real-time index, J accuracy is the accuracy index.

[0057] The specific calculation expression of the smoothness index is as follows:

[0058]

[0059] Wherein, y kalman is the output signal sequence in a time window after the Kalman filtering processing under the (Q, R) parameters to be evaluated; Variance is the variance calculation function, and the index represents the signal smoothness by calculating the variance of the filtered signal. The smaller the variance, the less the signal jitter and the stronger the smoothness.

[0060] The specific calculation expression of the real-time index is as follows:

[0061]

[0062] Wherein, t rise_start , t rise_end The jump edge window is defined, in which the signal jumps from one steady state to another steady state. y kalman [...] is the data segment of the Kalman filtering output signal in the jump edge window; y reference [...] is the data segment of the reference signal in the same time window; This reference signal can be obtained by strong low-pass filtering on the original signal, or an ideal, non-jittered step signal; RMSE is the root mean square error. The index measures the filtering lag by calculating the root mean square error of the filtered signal and the reference signal. The smaller the error, the stronger the real-time.

[0063] The specific calculation expression of the accuracy index is as follows:

[0064]

[0065] Wherein, t steady is a steady state time window after the signal jump is completed; Mean( y kalman [t steady ]) is the average value of Kalman filter output within the steady state window; y ideal is the ideal logic level value expected within the steady state time period, which represents the filtering accuracy by calculating the absolute error between the filtered steady state signal and the ideal level, and the smaller the error is, the closer the filtered signal is to the true state.

[0066] The fitness function containing the smoothness, real-time and accuracy indicators is constructed by the above expressions, wherein the smoothness indicator is calculated by the variance of the filtered signal, the real-time indicator is the root mean square error between the filtered signal and the reference signal within the jump edge window, and the accuracy indicator is the absolute error between the average value of the filtered signal within the steady state window and the ideal level value, and the fitness function value is obtained after balancing each indicator by the weight coefficient.

[0067] After constructing the fitness function, the driving snow ablation optimizer is optimized, and the driving snow ablation optimizer iteratively optimizes in the search space to minimize the fitness function, to obtain the optimal process noise covariance matrix Q opt and the optimal observation noise covariance matrix R opt . The optimization process includes:

[0068] (1) generating a set of parameters (Q, R);

[0069] (2) running Kalman filter using (Q, R);

[0070] (3) calculating the current fitness function according to the above formula;

[0071] (4) determining whether the optimal solution is reached, if yes, outputting the optimal process noise covariance matrix Q opt and the optimal observation noise covariance matrix R opt , if not, returning to (1).

[0072] Then the optimal parameters are injected into and the model of Kalman filter is restarted, and the Kalman filter is run: injecting Q opt and R opt obtained by the snow ablation optimizer into the Kalman filter model, resetting the filter state, completing the switching from the parameter learning optimization phase to the optimal filtering phase, and preparing for the subsequent accurate filtering of real-time signals. This step completes the switching from the "parameter learning optimization phase" to the "fixed parameter optimal filtering phase".

[0073] Then, real-time anti-jitter processing based on the optimal parameters is performed, and the smoothed signal after anti-jitter processing is output: inputting the original signal sequence point by point into the model configured with the optimal parameters (Q opt , R optThe Kalman filter outputs a smooth, stable and lag-minimized contact state estimation curve through recursive optimal estimation of the input signal at each time point. This process can effectively suppress the high-frequency jitter caused by contact bounce and restore the true action trajectory of the contact.

[0074] Reference Figure 1 As shown in the preferred embodiment further comprises:

[0075] Complete anti-jitter signal output and action value determination and record: with the smooth signal output by the Kalman filter as the result, the precise time of contact action is identified through threshold value determination, and the standard pressure or density value corresponding to the time is recorded as the calibration action value of the density relay, completing the calibration.

[0076] Specifically, when performing threshold value determination, a threshold value of 50% logic level is set to accurately identify the precise time of contact action (from open to close or from close to open), and the standard pressure value or density value corresponding to the precise time is recorded as the final calibration action value of the density relay.

[0077] The above method, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0078] Example 2

[0079] The present embodiment provides an SF6 gas density relay calibration signal anti-jitter processing system, which is integrated in an SF6 gas density relay intelligent calibration device, as shown in Figure 2 As shown, it specifically includes a signal acquisition module 1, a parameter initialization module 2, an optimizer configuration module 3, a fitness function construction module 4, a parameter optimization module 5, a filter configuration module 6, and a filter processing module 7. Each module cooperates to realize signal anti-jitter processing and accurate calibration action value recording, and the specific steps are as follows:

[0080] Signal acquisition module 1: Its core function is to collect the on-off signal of the relay contact in real time. The signal acquisition technology relied on is the basic signal acquisition technology in the field of industrial measurement and control. This technology can convert the on-off state of the mechanical contact at the physical level into a digital logic level signal (such as high level representing on and low level representing off). Its basic principle is to realize the isolation of strong and weak electricity through photoelectric coupling or electromagnetic isolation circuit to avoid the influence of on-site electromagnetic interference on the signal. The original signal sequence containing jitter noise is output to provide data basis for subsequent anti-jitter processing.

[0081] Parameter initialization module 2: The core technology of this module is the Kalman filter parameter initialization technology. Kalman filter is a linear filtering algorithm based on recursive optimal estimation, which depends on core parameters such as state vector, observation vector and noise covariance matrix. The state vector is used to represent the internal state of the system, the observation vector corresponds to the external observation value collected by the sensor, the process noise covariance matrix Q is used to describe the uncertainty of the system model itself, and the observation noise covariance matrix R is used to represent the noise intensity in the observation process. The function of this module is to assign initial values to the state vector and observation vector of Kalman filter, and to make initial estimates of Q and R to provide parameter support for the initial operation of Kalman filter.

[0082] Optimizer configuration module 3: This module involves the snow ablation optimizer initialization technology. Snow Ablation Optimizer (SAO) is an intelligent optimization algorithm that simulates the melting, migration and refreezing process of snow under temperature changes. Its basic principle is to search for the optimal solution by simulating the melting, migration and refreezing process of snow under temperature changes, with the characteristics of fast convergence speed and strong global search ability. This module is used to initialize the population size of snow ablation optimizer (i.e. the number of parameter combinations participating in search each iteration), the maximum number of iterations (the upper limit of iterations for algorithm termination) and the optimization search range of Q and R, to define the parameter boundary and operation rules for the subsequent parameter optimization link, and to ensure that the optimization process is carried out within a reasonable parameter range.

[0083] Fitness function construction module 4: It is a software algorithm module, which generates a fitness function with Q and R parameters as input and filter effect fitness value as output. This function integrates smoothness, real-time performance and accuracy, which can realize the comprehensive evaluation of filter performance.

[0084] Parameter optimization module 5: It is the core algorithm module of the system, which is used to drive the snow ablation optimizer to iteratively optimize the fitness function with the goal of minimizing the fitness function, and finally output the parameter combination that can achieve the optimal filter performance: the optimal process noise covariance matrix Q opt and the optimal observation noise covariance matrix R opt .

[0085] Filter configuration module 6: responsible for configuring Q opt and R opt injection Kalman filter algorithm model, while resetting and initializing the Kalman filter, realizing the dynamic update of the filter parameters, and completing the parameter optimization configuration of the filter algorithm.

[0086] Filter processing module 7: the core function is to input the original signal sequence into the Kalman filter configured with Q opt and R opt , and through the recursive optimal estimation logic of the Kalman filter, the original signal containing jitter noise is processed, and finally the smooth contact state estimation curve is output.

[0087] In the preferred embodiment, the system further comprises:

[0088] Action value recording module 8: used for threshold judgment of the contact state estimation curve, the threshold judgment is realized based on a preset logic level threshold, the contact action time can be automatically identified, and at the same time, the pressure / density data synchronously collected by the verification device is called to record the standard pressure value or density value corresponding to the time, and the record retention of the verification data is completed.

[0089] The rest is the same as example 1.

[0090] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes. The solutions in the embodiments of the present application can be implemented in various computer languages, such as object-oriented programming language Java and interpreted scripting language JavaScript.

[0091] The present application is described with reference to flowcharts and / or block diagrams according to the methods, systems and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one flow or multiple flows and / or one block or multiple blocks. Figure 1 The device that implements the functions specified in one flow or multiple flows and / or one block or multiple blocks.

[0092] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or the functions specified in the block Figure 1 one flow or multiple flows and / or the functions specified in the block

[0093] The above describes the specific embodiments of the present application in conjunction with the drawings, but is not a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications or variations made by those skilled in the art on the basis of the technical solutions of the present application without creative labor are still within the scope of protection of the present application.

Claims

1. A method for anti-jitter processing of a calibration signal for an SF6 gas density relay, characterized in that The method comprises the following steps: Real-time collection of relay contact switch quantity signals in the SF6 gas density relay calibration process to obtain an original signal sequence containing jitter noise, initialization of parameters of a Kalman filter algorithm and parameters of a snow-melting optimizer; A fitness function containing a smoothness index, a real-time index and an accuracy index is constructed, and the value of the fitness function is minimized as the goal to drive the snow-melting optimizer to iteratively optimize in a preset search range to obtain an optimal process noise covariance matrix and an optimal observation noise covariance matrix; The obtained optimal process noise covariance matrix and optimal observation noise covariance matrix are injected into the Kalman filter algorithm, and the state of the Kalman filter algorithm is reset; The original signal sequence collected is input point by point into the Kalman filter algorithm with updated parameters, and the original signal sequence is recursively optimally estimated by the Kalman filter algorithm to output an anti-jitter signal after processing. The expression of the fitness function is: Wherein, F(Q, R) is fitness function value, Q is process noise covariance matrix, R is observation noise covariance matrix, α, β, γ are weight coefficients and satisfy α+β+γ=1, J smoothness is a smoothness index, J lag is a real-time index, J accuracy is an accuracy index.

2. The method for anti-jitter processing of the SF6 gas density relay calibration signal according to claim 1, characterized in that, The expression of the smoothness index is: wherein y kalman is the output signal sequence of the Kalman filter for the parameter to be evaluated within the preset time window, Variance is a variance calculation function.

3. The method for anti-jitter processing of the SF6 gas density relay calibration signal according to claim 1, characterized in that, The expression of the real-time index is: wherein t rise_start and t rise_end tstart and tend are the start and end time instants of the signal jump edge time window, y kalman [...] is the data segment of the Kalman filter output signal within the jump edge time window, y reference [...] is the data segment of the reference signal within the same jump edge time window, and RMSE is the root mean square error calculation function.

4. The method of claim 3, wherein the method further comprises: The reference signal is a signal obtained by performing strong low-pass filtering on the original signal sequence, or an ideal step signal without jitter.

5. The method for anti-jitter processing of the SF6 gas density relay calibration signal according to claim 1, characterized in that, The expression of the accuracy index is: wherein, t steady is a steady state time window after the signal transition is completed, Mean( y kalman [ t steady ]) is the average value of the Kalman filter output signal within the steady state time window, y ideal is the ideal logic level value within the steady state time window, Mean is the average value calculation function.

6. The method of claim 1, wherein the anti-jitter processing of the SF6 gas density relay calibration signal is performed by a microprocessor. The Kalman filter algorithm based on optimal parameters recursively optimally estimates the input signal at each time to output a contact state estimation curve with minimum lag.

7. The method for anti-jitter processing of the SF6 gas density relay calibration signal according to claim 1, characterized in that, The method further comprises: Threshold value judgment is performed on the output anti-jitter signal to identify the relay contact action time, and the pressure value or density value corresponding to the time is recorded to complete SF6 gas density relay calibration.

8. A system for anti-jitter processing of a calibration signal for an SF6 gas density relay, characterized in that It comprises: A signal collection module is configured to collect relay contact switch quantity signals in the SF6 gas density relay calibration process in real time to output an original signal sequence containing jitter noise; A parameter initialization module is configured to initialize parameters of a Kalman filter algorithm; An optimizer configuration module is configured to initialize parameters of a snow-melting optimizer; A fitness function construction module is configured to construct a fitness function containing a smoothness index, a real-time index and an accuracy index, and to minimize the value of the fitness function as the goal to drive the snow-melting optimizer to iteratively optimize in a preset search range to obtain an optimal process noise covariance matrix and an optimal observation noise covariance matrix; A parameter optimization module is configured to drive the snow-melting optimizer to iteratively optimize to minimize the fitness function as the goal to output an optimal process noise covariance matrix and an optimal observation noise covariance matrix; A filter configuration module is configured to inject the optimal process noise covariance matrix and the optimal observation noise covariance matrix into a Kalman filter algorithm and reset the Kalman filter algorithm; A filtering processing module is configured to input the original signal sequence into the reset Kalman filter algorithm to perform recursive optimal estimation and output an anti-jitter signal after processing. The expression of the fitness function is: Wherein, F(Q, R) is fitness function value, Q is process noise covariance matrix, R is observation noise covariance matrix, and a, β, γ are weight coefficients and satisfy a+β+γ=1, J smoothness is a smoothness index, J lag is a real-time index, J accuracy is an accuracy index.

9. The anti-jitter processing system for SF6 gas density relay calibration signal according to claim 8, characterized in that, It further comprises: An action value recording module is configured to perform threshold value judgment on the contact state estimation curve based on a preset logic level threshold value, identify the contact action time and record the corresponding standard pressure value or density value.

Images