Flow table abnormal behavior monitoring method and system

Abstract

The invention discloses a flow table abnormal behavior monitoring method and system. The method comprises the steps of flow table abnormal state description, preliminary control input, unknown dynamic approximation, adaptive parameter updating law, final control input and flow table abnormal behavior monitoring. The invention belongs to the field of data processing, and particularly relates to a flow meter abnormal behavior monitoring method and system.According to the scheme, through preliminary control input design, unknown disturbance is compensated through self-adaptive switching control gain, and unknown nonlinear dynamics, caused by temperature change and medium fluctuation factors, of a flow meter is approximated through a logic system; the abnormity monitoring precision is improved; online adaptive updating of parameters of the fuzzy approximator is adopted, a fusion updating mechanism is set in combination with a fusion coefficient, and continuous learning of dynamic characteristics of the flow meter in normal and abnormal states is achieved; and a self-adaptive dynamic abnormal threshold mechanism is introduced to dynamically adjust an abnormal threshold, so that the problems of false alarm and missing alarm caused by a fixed threshold are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically refers to a method and system for monitoring abnormal behaviors of a flow meter. Background Art

[0002] The method for monitoring abnormal behaviors of a flow meter aims to identify and analyze abnormal patterns in flow data to discover potential security threats, system failures or other abnormal situations. However, general methods for monitoring abnormal behaviors of a flow meter have problems such as slow response speed when an abnormality occurs due to environmental interference or hardware aging, which may lead to insufficient or excessive control responses and easy misjudgment; general methods for monitoring abnormal behaviors of a flow meter lack dynamic adaptability and cannot dynamically adjust the monitoring strategy according to the changes of the flow meter in different working environments and states, resulting in a lag in response when external interference and abnormalities occur. Summary of the Invention

[0003] In view of the above situation, to overcome the defects of the prior art, the present invention provides a method and system for monitoring abnormal behaviors of a flow meter. Aiming at the problem that general methods for monitoring abnormal behaviors of a flow meter have slow response speed when an abnormality occurs due to environmental interference or hardware aging, which may lead to insufficient or excessive control responses and easy misjudgment, this solution designs a preliminary control input, uses an adaptive switching control gain to compensate for unknown disturbances, ensures smooth convergence of the error, and uses a logic system to approximate the unknown nonlinear dynamics of the flow meter caused by temperature changes and medium fluctuations, thereby improving the accuracy of abnormal monitoring; aiming at the problem that general methods for monitoring abnormal behaviors of a flow meter lack dynamic adaptability and cannot dynamically adjust the monitoring strategy according to the changes of the flow meter in different working environments and states, resulting in a lag in response when external interference and abnormalities occur, this solution adopts online adaptive updating of the parameters of the fuzzy approximator, combines with the setting of a fusion coefficient to establish a fusion update mechanism, and realizes continuous learning of the dynamic characteristics of the flow meter in normal and abnormal states; an adaptive dynamic abnormal threshold mechanism is introduced to dynamically adjust the abnormal threshold and adjust it according to the real-time state of the flow meter to avoid false alarms and missed alarms caused by a fixed threshold.

[0004] The technical solution adopted by the present invention is as follows: A method for monitoring abnormal behaviors of a flow meter provided by the present invention includes the following steps:

[0005] Step S1: Description of the abnormal state of the flow meter;

[0006] Step S2: Preliminary control input;

[0007] Step S3: Approximation of unknown dynamics;

[0008] Step S4: Adaptive parameter update law;

[0009] Step S5: Final control input;

[0010] Step S6: Monitoring of abnormal behavior of the flow meter.

[0011] Furthermore, in Step S1, the description of the abnormal state of the flow meter is to define the tracking error e(t) as e(t) = x(t) - y(t); where x(t) is the flow signal actually collected by the flow meter; t is the sampling time; y(t) is the ideal reference signal; using the error between the collected flow data and the reference signal, two state error components and are taken, respectively taking the error and the derivative of the error; a sliding mode surface is constructed to describe the comprehensive deviation situation, expressed as: ; where and are design coefficients.

[0012] Furthermore, in Step S2, the preliminary control input is designed as: ; where is the fractional derivative used to capture the memory effect of the flow meter; and respectively represent the unknown nonlinear parts in the actual dynamics and the reference dynamics of the flow meter, x is the flow data; sgn(·) is the sign function, and its role is to reach control; is the switching control gain; set ; is the switching control gain at t - 1 sampling; and are the gain increase coefficient and the attenuation coefficient respectively.

[0013] Furthermore, in Step S3, the approximation of the unknown dynamics is to use a logical system to approximate the unknown function; the unified parameter vector and are used to represent the uncertain dynamics of the flow meter, defined as: ; ; where , , and are basis functions constructed by Gaussian membership functions, and are the approximation of the actual dynamics of the flow meter, and are the approximation of the reference signal dynamics, , , and are the parameter vectors corresponding to the basis functions; T is the transpose operation.

[0014] Furthermore, in step S4, the adaptive parameter update law updates the parameters of the fuzzy approximator online adaptively and introduces a fusion coefficient to set up a fusion update mechanism; the adaptive update law is expressed as: ; ; where n(x) and n(y) are fuzzy membership functions constructed based on traffic data; , , and are positive regulation gain parameters; y is the ideal output of the flow meter under normal working conditions.

[0015] Furthermore, in step S5, the final control input generates a smooth switching control signal using fuzzy rules , and defines the reaching control part as: ; combining the equivalent control and the reaching control, the final control input is expressed as: ; where is a normalization parameter.

[0016] Furthermore, in step S6, the abnormal behavior monitoring of the flow meter preset a basic abnormal threshold , and introduces an adaptive dynamic abnormal threshold, expressed as: ; when , it is determined that the flow meter has abnormal behavior and a warning is given to relevant personnel; where is the abnormal threshold at time t; is an adjustment coefficient; TR is the sliding window duration; is the mean value within the window; is an integral variable.

[0017] A flow meter abnormal behavior monitoring system provided by the present invention includes a flow meter abnormal state description module, a preliminary control input module, an unknown dynamic approximation module, an adaptive parameter update law module, a final control input module, and a flow meter abnormal behavior monitoring module;

[0018] The flow meter abnormal state description module constructs a sliding mode surface to describe the abnormal state of the flow meter by comparing the actually collected flow with the ideal reference signal;

[0019] The preliminary control input module designs a preliminary control input by compensating for interference;

[0020] The unknown dynamic approximation module approximates the unknown non-linear dynamics in the flow meter and the reference signal;

[0021] The adaptive parameter update law module updates the parameters of the fuzzy approximator through the adaptive parameter update law and the fusion coefficient;

[0022] The final control input module forms a final control input based on the smooth switching control signal;

[0023] The flow meter abnormal behavior monitoring module monitors the abnormal behavior of the flow meter based on an adaptive dynamic abnormal threshold.

[0024] The beneficial effects achieved by the present invention using the above solution are as follows:

[0025] (1) Aiming at the problem that the general flow meter abnormal behavior monitoring method is affected by environmental interference or hardware aging, resulting in a slow response speed when an abnormality occurs, which in turn leads to insufficient or excessive control response and is prone to misjudgment. This solution designs a preliminary control input, uses an adaptive switching control gain to compensate for unknown disturbances, ensures smooth convergence of errors, and uses a logical system to approximate the unknown nonlinear dynamics of the flow meter caused by temperature changes and medium fluctuations, thereby improving the accuracy of abnormal monitoring.

[0026] (2) Aiming at the problem that the general flow meter abnormal behavior monitoring method lacks dynamic adaptability and cannot dynamically adjust the monitoring strategy as the flow meter changes in different working environments and states, resulting in a lag in response when external interference and abnormalities occur. This solution uses online adaptive update of the fuzzy approximator parameters, combines the fusion coefficient to set a fusion update mechanism, and realizes continuous learning of the dynamic characteristics of the flow meter in normal and abnormal states; introduces an adaptive dynamic abnormal threshold mechanism, dynamically adjusts the abnormal threshold, and adjusts according to the real-time state of the flow meter to avoid false alarms and missed alarms caused by fixed thresholds. Description of the Drawings

[0027] Figure 1 is a schematic flow chart of a method for monitoring abnormal behavior of a flow meter provided by the present invention;

[0028] Figure 2 is a schematic diagram of a system for monitoring abnormal behavior of a flow meter provided by the present invention.

[0029] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. Detailed Embodiments

[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0031] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc. indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the system or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0032] Embodiment 1. Refer to Figure 1 , a flow meter abnormal behavior monitoring method provided by the present invention, the method includes the following steps:

[0033] Step S1: Description of the abnormal state of the flow meter; by comparing the actually collected flow rate with the ideal reference signal, a sliding mode surface is constructed to describe the abnormal state of the flow meter;

[0034] Step S2: Initial control input; the initial control input is designed by compensating for interference;

[0035] Step S3: Approximation of unknown dynamics; approximate the unknown non-linear dynamics in the flow meter and the reference signal;

[0036] Step S4: Adaptive parameter update law; update the parameters of the fuzzy approximator through the adaptive parameter update law and the fusion coefficient;

[0037] Step S5: Final control input; form the final control input based on the smooth switching control signal;

[0038] Step S6: Monitoring of abnormal behavior of the flow meter; monitor the abnormal behavior of the flow meter based on the adaptive dynamic abnormal threshold.

[0039] Embodiment 2. Refer to Figure 1 , this embodiment is based on the above embodiment. In step S1, the description of the abnormal state of the flow meter is that in the monitoring task of the liquid flow meter, the actually collected liquid flow data is affected by various factors such as sensor error, environmental interference, and hardware aging, and there are unknown non-linear dynamics and disturbances; define the tracking error e(t) as e(t) = x(t) - y(t); where, x(t) is the flow signal actually collected by the flow meter; t is the sampling time; y(t) is the ideal reference signal, which reflects the expected output of the flow meter in the normal state; to achieve abnormal tracking, using the error between the collected flow data and the reference signal, introduce two state error components and , respectively take the error and the derivative of the error; construct the sliding mode surface , used to describe the comprehensive deviation situation, expressed as: ; where, and is the design coefficient, which is used to weight the error component so that the sliding mode surface can comprehensively reflect the abnormal change situation.

[0040] Embodiment 3, refer to Figure 1 , based on the above embodiment, in step S2, the preliminary control input is to compensate for the interference and ensure the smooth convergence of the error. The preliminary control input is designed as: ; where is the fractional derivative, which is used to capture the memory effect of the flow meter; and respectively represent the unknown non-linear parts in the actual dynamics and the reference dynamics of the flow meter. x is the flow data; sgn(·) is the sign function, and its function is to reach the control; is the switching control gain; to ensure sufficient interference compensation, set ; in the abnormal monitoring of the flow meter is used to drive the actuator, thereby adjusting the physical output of the liquid flow meter; is the switching control gain at t-1 sampling; and are the gain boosting coefficient and the attenuation coefficient respectively.

[0041] Embodiment 4, refer to Figure 1 , based on the above embodiment, in step S3, the unknown dynamic approximation is to handle the unknown non-linear dynamics caused by temperature changes and medium fluctuations in the liquid flow meter. A logical system is used to approximate the unknown function to enhance the ability to represent uncertain information; the unified parameter vector and represent the uncertain dynamics of the liquid flow meter, and are defined as: ; ; where , , and are the basis functions constructed by Gaussian membership functions, and are the approximation of the actual dynamics of the liquid flow meter, and are the approximation of the reference signal dynamics, , , and are the parameter vectors corresponding to the basis functions; T is the transpose operation.

[0042] By performing the above operations, for the general flow meter abnormal behavior monitoring method, due to being affected by environmental interference or hardware aging, the response speed is slow when an abnormality occurs, which may lead to insufficient or excessive control response and is prone to misjudgment. In this solution, through preliminary control input design, an adaptive switching control gain is used to compensate for unknown disturbances to ensure smooth convergence of the error. A logic system is used to approximate the unknown nonlinear dynamics of the flow meter caused by temperature changes and medium fluctuations, thereby improving the accuracy of abnormal monitoring.

[0043] Embodiment 5, refer to Figure 1 , based on the above embodiment, in step S4, the adaptive parameter update law is to update the parameters of the fuzzy approximator online adaptively, enabling the system to continuously learn the dynamic characteristics of the flow meter in normal and abnormal states, so as to quickly respond to external disturbances; a fusion coefficient is set to define a fusion update mechanism, which can not only update directly with a fast response but also capture uncertainties through indirect adaptation, ensuring that the monitoring task has higher sensitivity; the adaptive update law is expressed as: ; ; where, n(x) and n(y) are fuzzy membership functions constructed based on flow data; , , and are positive regulation gain parameters; y is the ideal output of the flow meter under normal operating conditions.

[0044] Embodiment 6, refer to Figure 1 , based on the above embodiment, in step S5, the final control input is to quickly correct the output of the liquid flow meter in an abnormal state while avoiding the chattering problem caused by the sign function, and a fuzzy rule is used to generate a smooth switching control signal , and the reaching control part is defined as: ; by combining the equivalent control and the reaching control, the final control input is expressed as: ; where, is the normalization parameter.

[0045] Embodiment 7, refer to Figure 1 , based on the above embodiment, in step S6, for the abnormal behavior monitoring of the flow meter, a basic abnormal threshold is preset, and an adaptive dynamic abnormal threshold is introduced, expressed as: ; when , it is determined that the liquid flow meter has abnormal behavior and a warning is given to relevant personnel; where, is the abnormal threshold at time t; is the adjustment coefficient; TR is the sliding window duration; is the mean value within the window; is the integration variable.

[0046] By performing the above operations, for the general flow meter abnormal behavior monitoring method, there is a lack of dynamic adaptability, and it is unable to dynamically adjust the monitoring strategy as the flow meter changes in different working environments and states, resulting in a lag in response when external interference and anomalies occur. In this solution, the parameters of the online adaptive update fuzzy approximator are adopted, and the fusion update mechanism is set by combining the fusion coefficient to realize the continuous learning of the dynamic characteristics of the flow meter in normal and abnormal states; the adaptive dynamic abnormal threshold mechanism is introduced to dynamically adjust the abnormal threshold and adjust it according to the real-time state of the flow meter to avoid false alarms and missed alarms caused by fixed thresholds.

[0047] Embodiment VIII, refer to Figure 2 Based on the above embodiment, a flow meter abnormal behavior monitoring system provided by the present invention includes a flow meter abnormal state description module, a preliminary control input module, an unknown dynamic approximation module, an adaptive parameter update law module, a final control input module, and a flow meter abnormal behavior monitoring module;

[0048] The flow meter abnormal state description module constructs a sliding mode surface to describe the abnormal state of the flow meter by comparing the actually collected flow rate with the ideal reference signal;

[0049] The preliminary control input module designs the preliminary control input by compensating for interference;

[0050] The unknown dynamic approximation module approximates the unknown nonlinear dynamics in the flow meter and the reference signal;

[0051] The adaptive parameter update law module updates the parameters of the fuzzy approximator through the adaptive parameter update law and the fusion coefficient;

[0052] The final control input module forms the final control input based on the smooth switching control signal;

[0053] The flow meter abnormal behavior monitoring module monitors the abnormal behavior of the flow meter based on the adaptive dynamic abnormal threshold.

[0054] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprises", "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0055] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

[0056] The present invention and its embodiments have been described above. Such description is not restrictive. What is shown in the drawings is only one of the embodiments of the present invention, and the actual structure is not limited thereto. In general, if those of ordinary skill in the art are inspired by it and, without departing from the purpose of the present invention, design similar structural modes and embodiments to this technical solution without creative efforts, they should fall within the protection scope of the present invention.

Claims

1. A method for monitoring abnormal behavior of a flow meter, characterized in that: The method comprises the following steps: Step S1: Description of abnormal state of flow meter; Step S2: preliminary control input; Step S3: unknown dynamic approximation; Step S4: Adaptive parameter update law; Step S5: final control input; Step S6: Monitoring abnormal behavior of the flow meter.

2. A flow meter abnormal behavior monitoring method according to claim 1, characterized in that: In step S1, the abnormal state description of the flow meter is to define the tracking error e(t) as e(t) = x(t)- y(t); where x(t) is the flow signal actually collected by the flow meter; t is the sampling time; y(t) is the ideal reference signal; using the error between the collected flow data and the reference signal, two state error components are introduced and , respectively take the error and error derivative; construct the sliding surface , used to describe the comprehensive deviation, expressed as: ;in, and is the design factor.

3. A flow meter abnormal behavior monitoring method according to claim 2, characterized in that: In step S2, the preliminary control input Designed for: ;in, is a fractional derivative used to capture the flow meter memory effect; and They represent the unknown nonlinear parts in the actual dynamics and reference dynamics of the flow meter, respectively; x is the flow data; sgn(·) is the sign function, which acts as the arrival control; is the switching control gain; setting ; is the switching control gain at sampling time t-1; and They are the gain boost factor and the attenuation factor respectively.

4. A flow meter abnormal behavior monitoring method according to claim 3, characterized in that: In step S3, the unknown dynamic approximation is to use a logic system to approximate the unknown function; using a unified parameter vector and represents the uncertain dynamics of the flow table and is defined as: ; ;in, , , and is the basis function constructed by Gaussian membership function, and It is an approximation of the actual dynamics of the flow meter. and is a dynamic approximation of the reference signal, , , and is the parameter vector of the corresponding basis function; T is the transpose operation.

5. A flow meter abnormal behavior monitoring method according to claim 4, characterized in that: In step S4, the adaptive parameter update law is to introduce the fusion coefficient by online adaptively updating the fuzzy approximator parameters Set the fusion update mechanism; the adaptive update law is expressed as: ; ; Among them, n(x) and n(y) are fuzzy membership functions constructed based on flow data; , , and is the positive adjustment gain parameter; y is the ideal output of the flow meter under normal working conditions.

6. A flow meter abnormal behavior monitoring method according to claim 1, characterized in that: In step S5, the final control input is a smooth switching control signal generated by using fuzzy rules , define the arrival control part as: ; Combine equivalent control and arrival control, and the final control input is expressed as: ;in, is the normalization parameter.

7. A flow meter abnormal behavior monitoring method according to claim 1, characterized in that: In step S6, the flow meter abnormal behavior monitoring is to pre-set the basic abnormal threshold , introducing an adaptive dynamic abnormal threshold, expressed as: ;when , determine if the flow meter has abnormal behavior and issue an early warning to relevant personnel; among them, is the abnormal threshold at time t; is the adjustment coefficient; TR is the sliding window length; is the mean within the window; is the integration variable.

8. A flow meter abnormal behavior monitoring system, used to implement a flow meter abnormal behavior monitoring method according to any one of claims 1 to 7, characterized in that: It includes a flow meter abnormal state description module, a preliminary control input module, an unknown dynamic approximation module, an adaptive parameter update law module, a final control input module and a flow meter abnormal behavior monitoring module; The flow meter abnormal state description module constructs a sliding surface to describe the abnormal state of the flow meter by comparing the actual collected flow with the ideal reference signal; The preliminary control input module designs a preliminary control input by compensating for disturbances; The unknown dynamics approximation module approximates the unknown nonlinear dynamics in the flow meter and the reference signal; The adaptive parameter update law module updates the fuzzy approximator parameters through the adaptive parameter update law and the fusion coefficient; The final control input module forms a final control input based on the smooth switching control signal; The flow meter abnormal behavior monitoring module performs abnormal behavior monitoring of the flow meter based on an adaptive dynamic abnormal threshold.

Images