Interference alarm identification and suppression method and system based on mixed geometric distribution

Abstract

The application provides a mixed geometric distribution-based interference alarm identification and suppression method and system. The scheme comprises obtaining a historical data sequence of a monitored variable, and obtaining an alarm duration set according to a corresponding alarm threshold; based on the alarm duration set, obtaining a sample distribution; fitting the sample distribution with a mixed geometric distribution to obtain a mixed geometric distribution with the maximum fitting goodness of the sample distribution; based on the obtained mixed geometric distribution and a preset false alarm rate, determining a time upper limit value of false alarm judgment, and taking the upper limit value as an alarm delay parameter to realize identification and suppression of an interference alarm.

Description

Technical Field

[0001] This invention belongs to the field of industrial alarm technology, and in particular relates to a method and system for interference alarm identification and suppression based on hybrid geometric distribution. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the increasing complexity of modern process industry production processes and the rapid development of information technology, industrial production monitoring systems have been widely applied in production. Alarm systems, as a mainstream industrial production monitoring system, have become essential safety monitoring systems for process industries such as power and chemicals. Alarm systems generate alarm signals by comparing the amplitude of the monitored process variables with their alarm thresholds in real time. This alerts production operators to take effective operational measures to bring the monitored process variables back to their normal fluctuation range, preventing abnormalities in the production process from escalating into production accidents and avoiding economic losses and product quality degradation caused by such accidents.

[0004] Current alarm systems commonly suffer from an excessive number of interfering alarms. The root cause of these interfering alarms is noise and random interference in process variables, leading to amplitudes exceeding alarm thresholds for short periods. A large number of interfering alarms not only significantly reduces operators' trust in the alarm system, rendering it ineffective, but also drowns out genuine alarms, preventing operators from effectively detecting anomalies in the production process and seriously jeopardizing production safety and efficiency. Therefore, identifying and eliminating interfering alarms in alarm systems is of great significance for monitoring process production in the process industry.

[0005] The inventors discovered that existing interference alarm identification methods mainly determine whether the alarm duration exceeds a given value. While this method is simple to implement, the selection of the given alarm duration value is highly subjective and its effectiveness in practical applications is unsatisfactory. Furthermore, Chinese patent CN109087490A discloses an interference alarm identification method based on alarm duration characteristics. This method assumes that the process variable has stationary characteristics and identifies interference alarms based on the geometric distribution of alarm duration. Although this method overcomes the shortcomings of the given alarm duration value to some extent, it does not fully consider the fact that industrial process variables often have non-stationary characteristics, and it does not link the interference alarm identification results with interference alarm suppression methods, thus exhibiting significant limitations in practical applications. Summary of the Invention

[0006] To address the aforementioned problems, this invention provides a method and system for interference alarm identification and suppression based on a hybrid geometric distribution. This solution resolves issues such as existing interference alarm identification methods being unsuitable for non-stationary industrial process variables, lacking a correlation with interference alarm suppression methods, and having limitations in practical applications. Compared to existing interference alarm identification methods, this invention extends the application scope of interference alarm identification from stationary industrial process variables to non-stationary working process variables, while also enabling the selection of alarm delay parameters and utilizing the alarm delay to suppress interference alarms.

[0007] According to a first aspect of the present invention, an interference alarm identification and suppression method based on a hybrid geometric distribution is provided, comprising:

[0008] Obtain the historical data sequence of the monitored variable, and obtain the alarm duration set according to the corresponding alarm threshold;

[0009] Based on the set of alarm durations, obtain its sample distribution;

[0010] The sample distribution is fitted using a mixed geometric distribution to obtain a mixed geometric distribution with the maximum goodness of fit to the sample distribution;

[0011] Based on the obtained mixed geometric distribution and the preset false alarm rate, the upper limit of the time for judging false alarms is determined, and the upper limit is used as the parameter of the alarm delay device to realize the identification and suppression of interference alarms.

[0012] Furthermore, the step of obtaining the alarm duration set based on the corresponding alarm threshold specifically involves: obtaining an alarm signal sequence based on the historical data sequence of the monitored variable, according to its alarm threshold and alarm signal generation mechanism; and obtaining the alarm duration of each alarm by using the alarm release time and alarm trigger time of each alarm, thereby obtaining the alarm duration set.

[0013] Furthermore, the step of obtaining the mixed geometric distribution with the maximum goodness of fit to the sample distribution specifically employs a genetic algorithm to optimize and solve the parameters when the mixed geometric distribution and the sample distribution have the maximum goodness of fit, and obtains the optimal mixed geometric distribution based on the obtained parameters.

[0014] Furthermore, obtaining the mixed geometric distribution with the maximum goodness of fit to the sample distribution further includes: determining the parameters of the mixed geometric distribution with the maximum goodness of fit to the sample distribution based on a genetic algorithm; calculating the goodness of fit between the mixed geometric distribution and the sample distribution based on the current parameters; if the goodness of fit is not less than a preset threshold, then the current mixed geometric distribution is taken as the optimal mixed geometric distribution; if the goodness of fit is less than the preset threshold, then the element corresponding to the maximum value is deleted from the alarm duration set, and the mixed geometric distribution parameters are re-optimized using the deleted set until the goodness of fit is not less than the preset threshold.

[0015] Furthermore, the hybrid geometric distribution is specifically represented as follows:

[0016]

[0017] Where, ω m Let ω be the weighting coefficient of the m-th geometric distribution in the mixed geometric distribution. m satisfy The vector ω is ω=[ω1,ω2,…,ω m ], q m Let q be the alarm probability of the m-th geometric distribution, and q m ∈[0,1), the vector q is q=[q1,q2,…,q m The parameter M represents the order of the mixed geometric distribution, m∈[1,M].

[0018] Furthermore, the goodness of fit is specifically expressed as follows:

[0019]

[0020] in, To optimize the mixed geometric distribution parameter values ​​obtained during the solution process, f0(n) represents the sample distribution, and N... max It is the maximum value of the elements in the set of alarm durations, that is, the element with the longest alarm duration.

[0021] Furthermore, the parameters of the alarm delay unit specifically satisfy the following constraints:

[0022]

[0023] Where z is the alarm delay parameter and α is the preset false alarm rate.

[0024] According to a second aspect of the present invention, an interference alarm identification and suppression system based on a hybrid geometric distribution is provided, comprising:

[0025] The alarm duration set acquisition unit is used to acquire the historical data sequence of the monitored variable and acquire the alarm duration set according to the corresponding alarm threshold.

[0026] A sample distribution acquisition unit is used to acquire the sample distribution based on the set of alarm durations;

[0027] A mixed geometric distribution acquisition unit is used to fit the sample distribution with a mixed geometric distribution to obtain a mixed geometric distribution with the maximum goodness of fit to the sample distribution;

[0028] The interference alarm identification and suppression unit is used to determine the upper limit of the time for judging false alarms based on the obtained mixed geometric distribution and the preset false alarm rate, and to use the upper limit as the alarm delay parameter to realize the identification and suppression of interference alarms.

[0029] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the memory, wherein the processor executes the program to implement the aforementioned interference alarm identification and suppression method based on hybrid geometric distribution.

[0030] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the aforementioned method for interference alarm identification and suppression based on a hybrid geometric distribution.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] (1) This invention provides a method and system for interference alarm identification and suppression based on hybrid geometric distribution. The solution solves the problems that existing interference alarm identification methods are not applicable to non-stationary industrial process variables, have not established a correlation with interference alarm suppression methods, and have limitations in practical applications. Compared with existing interference alarm identification methods, the solution of this invention extends the application scope of interference alarm identification methods from stationary industrial process variables to non-stationary working process variables. At the same time, it can complete the selection of alarm delay parameters and use the alarm delay to achieve interference alarm suppression.

[0033] (2) The solution described in this invention is applicable not only to stationary industrial process variables but also to non-stationary process variables, overcoming the limitation of existing interference alarm identification methods that are only applicable to stationary process variables;

[0034] (3) The solution described in this invention realizes the design of an alarm delayer based on interference alarm identification, and realizes the direct correlation between interference alarm identification and interference alarm suppression, which is of great significance for reducing the number of interference alarms in the alarm system.

[0035] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0036] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0037] Figure 1 This is a flowchart of the interference alarm identification and suppression method based on hybrid geometric distribution described in this embodiment of the invention;

[0038] Figure 2 This refers to the partial data trend curves and alarm thresholds of the alarm measurement points described in this embodiment of the invention.

[0039] Figure 3 This is a histogram of alarm duration as described in the embodiments of the present invention;

[0040] Figure 4(a) is a schematic diagram of the sample distribution and mixed geometric distribution curves under the initial conditions described in the embodiment of the present invention;

[0041] Figure 4(b) is a schematic diagram of the sample distribution and mixed geometric distribution curves under the final conditions described in the embodiment of the present invention;

[0042] Figure 5(a) is a partial schematic diagram of the sample distribution and mixed geometric distribution curve under the initial conditions described in the embodiment of the present invention;

[0043] Figure 5(b) is a partial schematic diagram of the sample distribution and mixed geometric distribution curves under the final conditions described in the embodiment of the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0045] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0046] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0047] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0048] Example 1:

[0049] The purpose of this embodiment is to provide a method for identifying and suppressing interference alarms based on a hybrid geometric distribution.

[0050] An interference alarm identification and suppression method based on hybrid geometric distribution, comprising:

[0051] Obtain the historical data sequence of the monitored variable, and obtain the alarm duration set according to the corresponding alarm threshold;

[0052] Based on the set of alarm durations, obtain its sample distribution;

[0053] The sample distribution is fitted using a mixed geometric distribution to obtain a mixed geometric distribution with the maximum goodness of fit to the sample distribution;

[0054] Based on the obtained mixed geometric distribution and the preset false alarm rate, the upper limit of the time for judging false alarms is determined, and the upper limit is used as the parameter of the alarm delay device to realize the identification and suppression of interference alarms.

[0055] In specific implementation, obtaining the alarm duration set based on the corresponding alarm threshold involves: obtaining an alarm signal sequence based on the historical data sequence of the monitored variable, according to its alarm threshold and alarm signal generation mechanism; and obtaining the alarm duration of each alarm by using the alarm release time and alarm trigger time of each alarm, thereby obtaining the alarm duration set.

[0056] In a specific implementation, the process of obtaining the mixed geometric distribution that has the maximum goodness of fit with the sample distribution is specifically achieved by using a genetic algorithm to optimize and solve for the parameters that have the maximum goodness of fit between the mixed geometric distribution and the sample distribution, and then obtaining the optimal mixed geometric distribution based on the obtained parameters.

[0057] The step of obtaining the mixed geometric distribution with the maximum goodness of fit to the sample distribution further includes: determining the parameters of the mixed geometric distribution with the maximum goodness of fit to the sample distribution based on a genetic algorithm; calculating the goodness of fit between the mixed geometric distribution and the sample distribution based on the mixed geometric distribution with the current parameters; if the goodness of fit is not less than a preset threshold, then the current mixed geometric distribution is taken as the optimal mixed geometric distribution; if the goodness of fit is less than the preset threshold, then the element corresponding to the maximum value is deleted from the alarm duration set, and the mixed geometric distribution parameters are optimized again using the deleted set until the goodness of fit is not less than the preset threshold.

[0058] Specifically, for ease of understanding, the following detailed description of the solution in this embodiment is provided in conjunction with the accompanying drawings:

[0059] like Figure 1 As shown in this embodiment, the interference alarm identification and suppression method based on hybrid geometric distribution specifically includes the following steps:

[0060] Step 1: Obtain the alarm duration sequence from the historical data of the alarm monitoring variable and the alarm threshold, that is, obtain the historical data sequence of the monitored variable. Based on its alarm threshold x tp The alarm signal sequence is obtained from the alarm signal generation mechanism. The alarm duration set corresponding to the alarm sequence is obtained.

[0061] Specifically, regarding the obtained historical data sequences Its alarm signal sequence is as follows:

[0062]

[0063] Where N0 is the number of elements in the current historical data sequence, x tp The alarm threshold is the variable for alarm monitoring. Then, the alarm signal sequence... The alarm duration sequence is obtained, namely:

[0064] T1(k)=n2-n1+1 (2)

[0065] Where n1 and n2 represent the alarm trigger time and alarm clearing time of the k-th alarm, respectively, and n1 satisfies x a (n1) = 1 and x a (n1-1)=0, n2 satisfies x a (n2)=1 and x a (n2+1)=0, and n1 and n2 satisfy n2≥n1. Using the obtained By analyzing the durations of all alarms, we can obtain the alarm duration set. Parameter K0 represents the current alarm duration set. The number of elements in the data.

[0066] Step 2: From the alarm duration set The sample distribution f0 is obtained, and a mixed geometric distribution is used to fit f0, that is, the alarm duration set. Based on this, its sample distribution f0 is calculated, and a genetic algorithm is used to find the mixed geometric distribution that has the best fit to the sample distribution f0 of the alarm duration. If the fit is greater than the given goodness-of-fit threshold, then it is judged. All alarms corresponding to the elements in the middle are interference alarms; otherwise, delete them. Take the element with the largest value from the set, and get... subset of Next, determine the goodness of fit. With goodness threshold If the goodness of fit is still less than the goodness of fit threshold, repeat the above operations of deleting the maximum element and determining the goodness of fit in the subset 1 times until the goodness of fit is greater than the goodness of fit threshold. at this time All alarms corresponding to the elements in the middle are interference alarms, and the corresponding mixed geometric distribution is the optimal mixed geometric distribution.

[0067] Specifically, the hybrid geometric distribution used in this embodiment is as follows:

[0068]

[0069] Where, ω m Let ω be the weighting coefficient of the m-th geometric distribution in the mixed geometric distribution. m satisfy The vector ω is ω=[ω1,ω2,…,ω m ], q m Let q be the alarm probability of the m-th geometric distribution, and q m ∈[0,1), the parameter q is q=[q1,q2,…,q m The parameter M represents the order of the mixed geometric distribution, m∈[1,M], and n represents the alarm duration.

[0070] Alarm duration set The sample distribution f0 is:

[0071] f0(n)=C(T1=n) / K0 (4)

[0072] Among them, C(T) i =n) means to find the set The number of elements in T1 = n. For a given set of alarm durations, the genetic algorithm determines the parameters that maximize the goodness of fit between the mixture distribution in equation (3) and the sample distribution f0. This can be further represented as being determined by a genetic algorithm. This makes the following expression reach its minimum value, namely:

[0073]

[0074] in, The obtained mixed geometric distribution parameter values. The goodness of fit between f0 and f0 is expressed as follows:

[0075]

[0076] Where, N max for The maximum value of the elements in the middle. When the goodness of fit satisfies equation (7), that is:

[0077]

[0078] in, Let be the given goodness-of-fit threshold. When equation (7) holds, then... All elements in the middle correspond to interference alarms; otherwise, from Delete the element corresponding to the maximum value to get subset of Then judge from equations (3) to (7). Check if all elements in the middle correspond to interference alarms. Repeat the above steps once, until... Satisfying equation (7), and the optimal mixed geometric distribution at this time is denoted as:

[0079]

[0080] Step 3: Determine the alarm delay device parameter value based on the optimal hybrid geometric distribution in equation (8). That is, under the condition of a given false alarm rate α, determine the upper limit of the time for false alarm judgment by the cumulative distribution of the obtained optimal fitted geometric distribution, and use this upper limit value as the alarm delay device parameter to realize the identification and suppression of interference alarms.

[0081] The alarm delay device used in this embodiment is:

[0082]

[0083] Where z represents the alarm delay parameter. Given a false alarm rate α, the alarm delay parameter z satisfies:

[0084]

[0085] To demonstrate the effectiveness of the solution described in this embodiment, the following experimental verification was conducted with specific examples:

[0086] This embodiment takes the generator frequency measurement point of a 300MW thermal power unit as an example to illustrate the specific effect of the method described in this embodiment in practical applications.

[0087] The high alarm threshold for the generator frequency measurement point is x tp =50.2999Hz, and 30 days of historical data for this measuring point were acquired with a sampling period of 1 second. The data length N = 2.592 × 10 5 Some historical data such as Figure 2As shown in the figure. The following section takes the interference alarm identification and suppression at this measurement point as an example to explain in detail the application process of the alarm prediction method disclosed in this embodiment:

[0088] Step 1: Obtain the historical data sequence according to equation (1) alarm sequence Then, the alarm duration set is obtained according to equation (2). T1(k) is distributed in the interval [1, 413];

[0089] Step 2: Given a goodness-of-fit threshold The fit between the mixed geometric distribution determined by the genetic algorithm and the sample distribution at this point is shown in Figure 4(a) (Figure 5(a) is a partially enlarged view of this figure). Therefore, the set is deleted. The largest element in the mixture is T1(k) = 413. The mixing steps above are repeated l = 383 times until... If equation (7) is satisfied, then the goodness of fit is: The fitting relationship between the mixed geometric distribution and the sample distribution is shown in Figure 4(b) (Figure 5(b) is a partially enlarged view of this figure), and the corresponding optimal mixed geometric distribution is:

[0090] f(n|ω*,q*,3)=0.7138×0.8317 n-1 ×(1-0.8317)+0.1424×0.1781 n-1 ×(1-0.1781)+0.1444×0.1807 n-1 ×(1-0.1807)

[0091] Step 3: Based on the obtained f(n|ω*,q*,M), and under the given false alarm threshold α=0.05, the alarm delay parameter z=23 is obtained from equation (10). After setting the alarm delay parameter in equation (9) to 23, The number of alarms was reduced from 28,376 to 463, achieving an interference alarm suppression rate of over 98%.

[0092] Meanwhile, to demonstrate the effectiveness of the solution described in this embodiment, an interference alarm identification method based on alarm duration characteristics disclosed in patent CN109087490A is applied to this case and compared with the solution described in this embodiment. Specifically, for the solution of patent CN109087490A, when the goodness-of-fit threshold β is 0.99, the obtained geometric distribution function is f. m (n) = 0.4647 n-1 (1-0.4647), the maximum goodness of fit is only R. 2=0.9725, which does not meet the goodness-of-fit requirement, therefore the method fails. The fitting relationship between ζ(n) and ζ(n) is shown in Figure (4)(a), where ζ(n) is the sample distribution curve of the alarm duration.

[0093] Example 2:

[0094] The purpose of this embodiment is to provide an interference alarm identification and suppression system based on a hybrid geometric distribution.

[0095] An interference alarm identification and suppression system based on hybrid geometric distribution, comprising:

[0096] The alarm duration set acquisition unit is used to acquire the historical data sequence of the monitored variable and acquire the alarm duration set according to the corresponding alarm threshold.

[0097] A sample distribution acquisition unit is used to acquire the sample distribution based on the set of alarm durations;

[0098] A mixed geometric distribution acquisition unit is used to fit the sample distribution with a mixed geometric distribution to obtain a mixed geometric distribution with the maximum goodness of fit to the sample distribution;

[0099] The interference alarm identification and suppression unit is used to determine the upper limit of the time for judging false alarms based on the obtained mixed geometric distribution and the preset false alarm rate, and to use the upper limit as the alarm delay parameter to realize the identification and suppression of interference alarms.

[0100] Furthermore, the relevant technical details of the system described in this embodiment have been described in detail in Embodiment 1, and therefore will not be repeated here.

[0101] In further embodiments, the following is also provided:

[0102] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0103] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0104] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0105] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0106] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0107] Those skilled in the art will recognize that the units, i.e., algorithm steps, of the various examples described in connection with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0108] The interference alarm identification and suppression method and system based on hybrid geometric distribution provided in the above embodiments can be implemented and has broad application prospects.

[0109] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for interference alarm identification and suppression based on hybrid geometric distribution, characterized in that, include: Obtain the historical data sequence of the monitored variable, and obtain the alarm duration set according to the corresponding alarm threshold; Based on the set of alarm durations, obtain its sample distribution; The sample distribution is fitted using a mixed geometric distribution to obtain the mixed geometric distribution with the maximum goodness of fit. Specifically, a genetic algorithm is used to optimize and solve for the parameters that maximize the goodness of fit between the mixed geometric distribution and the sample distribution. Based on the obtained parameters, the optimal mixed geometric distribution is obtained. The mixed geometric distribution is specifically represented as follows: ； in, For the mixed geometric distribution, the first Weight coefficients of a geometric distribution, satisfy ,vector for , For the first The alarm probability is given by a geometric distribution, and ,vector for ,parameter This indicates the order of the mixed geometric distribution. ; Based on the obtained mixed geometric distribution and the preset false alarm rate, an upper limit for the time of false alarm judgment is determined, and the upper limit is used as the parameter of the alarm delay device to realize the identification and suppression of interference alarms; the alarm delay device parameter specifically satisfies the following constraints: ； Where z is the alarm delay parameter, This is the preset false alarm rate.

2. The interference alarm identification and suppression method based on hybrid geometric distribution as described in claim 1, characterized in that, The step of obtaining the alarm duration set based on the corresponding alarm threshold specifically involves: obtaining an alarm signal sequence based on the historical data sequence of the monitored variable, according to its alarm threshold and alarm signal generation mechanism; and obtaining the alarm duration of each alarm by using the alarm release time and alarm trigger time of each alarm, thereby obtaining the alarm duration set.

3. The interference alarm identification and suppression method based on hybrid geometric distribution as described in claim 1, characterized in that, The step of obtaining the mixed geometric distribution with the maximum goodness of fit to the sample distribution further includes: determining the parameters of the mixed geometric distribution with the maximum goodness of fit to the sample distribution based on a genetic algorithm; calculating the goodness of fit between the mixed geometric distribution and the sample distribution based on the mixed geometric distribution with the current parameters; if the goodness of fit is not less than a preset threshold, then the current mixed geometric distribution is taken as the optimal mixed geometric distribution; if the goodness of fit is less than the preset threshold, then the element corresponding to the maximum value is deleted from the alarm duration set, and the mixed geometric distribution parameters are optimized again using the deleted set until the goodness of fit is not less than the preset threshold.

4. The interference alarm identification and suppression method based on hybrid geometric distribution as described in claim 1, characterized in that, The goodness of fit is specifically expressed as follows: in, To optimize the mixed geometric distribution parameter values ​​obtained during the solution process, For the sample distribution, It is the maximum value of the elements in the set of alarm durations, that is, the element with the longest alarm duration.

5. An interference alarm identification and suppression system based on hybrid geometric distribution, characterized in that, include: The alarm duration set acquisition unit is used to acquire the historical data sequence of the monitored variable and acquire the alarm duration set according to the corresponding alarm threshold. A sample distribution acquisition unit is used to acquire the sample distribution based on the set of alarm durations; A mixed geometric distribution acquisition unit is used to fit the sample distribution with a mixed geometric distribution to obtain the mixed geometric distribution with the maximum goodness of fit to the sample distribution. Specifically, a genetic algorithm is used to optimize and solve for the parameters when the mixed geometric distribution and the sample distribution have the maximum goodness of fit. Based on the obtained parameters, the optimal mixed geometric distribution is obtained. The mixed geometric distribution is specifically represented as follows: ； in, For the mixed geometric distribution, the first Weight coefficients of a geometric distribution, satisfy ,vector for , For the first The alarm probability is given by a geometric distribution, and ,vector for ,parameter This indicates the order of the mixed geometric distribution. ; The interference alarm identification and suppression unit is used to determine the upper limit of the time for judging false alarms based on the obtained mixed geometric distribution and the preset false alarm rate, and uses the upper limit as the alarm delay parameter to realize the identification and suppression of interference alarms; the alarm delay parameter specifically satisfies the following constraints: ； Where z is the alarm delay parameter, This is the preset false alarm rate.

6. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and running thereon, characterized in that, When the processor executes the program, it implements the interference alarm identification and suppression method based on hybrid geometric distribution as described in any one of claims 1-4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements an interference alarm identification and suppression method based on a hybrid geometric distribution as described in any one of claims 1-4.

Images