Failure detection method and system for sensor of petroleum refining device

Abstract

The invention discloses a failure detection method and system for a sensor of a petroleum refining device, and belongs to the technical field of sensor detection. A to-be-detected sensor of a petroleum refining device is represented by adopting a partial differential equation model, and a nonlinear observer of the to-be-detected sensor is constructed based on the partial differential equation model of the to-be-detected sensor; the output is observed through an observer, and failure detection is carried out based on a detection residual error between an observed value and the corresponding actual output; the constructed nonlinear observer can cope with a multi-input multi-output condition with position interference, and the detection residual error is finally bounded under the condition that the to-be-detected sensor is healthy, so that a preset threshold value for failure judgment based on the detection residual error is easily determined, and the accuracy of failure judgment is improved. And meanwhile, a partial differential equation model does not need to be reduced into an ordinary differential equation for solving, so that false alarm and missing alarm possibly caused by model reduction are avoided, and failure detection can be accurately carried out on the sensor of the petroleum refining device.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensor detection, and more specifically, relates to a failure detection method and system for a sensor of a petroleum refining device. Background Art

[0002] In petrochemical production, sensors are used to measure and collect process parameters of each production process and upload data to the control center to ensure the safe operation and efficient production of the equipment. The petrochemical production environment is usually exposed to the outdoors and has the characteristics of high temperature, dust, and high electromagnetic interference. During long-term operation, it is easy to cause sensor failure, such as increased measurement data errors, inability to detect and report equipment abnormalities in a timely manner, and thus reduce the reliability and service life of the equipment. At the same time, due to cost constraints, the safety standards of sensors in petrochemical production are usually low, and they are prone to cyber attacks and functional failure. Therefore, it is an important means to ensure the safe operation of petroleum refining equipment to perform failure detection on sensors of petroleum refining equipment, so as to timely discover failed or about to fail sensors, replace them or reinforce them safely to reduce adverse effects.

[0003] The petroleum refining unit is a typical hybrid system, which conforms to the characteristics of lumped parameter system (LPS) as a whole, and conforms to the characteristics of distributed parameter system (DPS) in some specific sensor sensing processes, such as heat conduction, mass transfer and fluid flow. Existing methods usually represent the partial differential equation model of DPS with a set of infinite ordinary differential equations. In the process of judging whether the sensor has failed, in order to adapt to the existing linear system observer, the partial differential equation model is reduced to derive its ordinary differential equation under finite dimensions for solution, which has representation errors, so it is impossible to accurately detect the failure of the sensors in the petroleum refining unit. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a failure detection method and system for a sensor of a petroleum refining device, so as to solve the technical problem that the prior art cannot accurately perform failure detection on a sensor of a petroleum refining device.

[0005] In order to achieve the above-mentioned object, in a first aspect, the present invention provides a failure detection method of a sensor of a petroleum refining device, comprising:

[0006] Acquire current detection data of the sensor to be detected in the petroleum refining device; wherein the current detection data includes: state data of the sensor to be detected at the current moment and at all moments before the current moment;

[0007] Construct a non - linear observer for the sensor to be detected based on the partial differential equation model of the sensor to be detected, input the current detection data into the observer for solution to obtain the observed value; calculate the difference between the observed value and the actual output of the sensor to be detected to obtain the detection residual;

[0008] Judge whether the detection residual is greater than the preset threshold. If so, determine that the sensor to be detected fails; otherwise, determine that the sensor to be detected does not fail;

[0009] Among them, the equation of the observer is:

[0010]

[0011] Its boundary conditions are:

[0012]

[0013] Its observed value is:

[0014]

[0015] Among them, is the state - estimation vector at position x of the sensor to be detected at time t; 0 < x < l; l is the spatial length of the sensor to be detected; c is a constant in the partial differential equation model of the sensor to be detected; is a Lipschitz - continuous non - linear vector function used to describe the dynamic characteristics of the observed sensor to be detected; is the error - feedback control term; P1(x,t) is the first gain matrix of the observer, used to adjust the feedback of the sensor to be detected on the state - estimation error; is the actual output of the sensor to be detected at time t; is a full - rank square matrix; n is the dimension of the state data of the sensor to be detected; represents the state derivative at position 0; P 10 (t) is the second gain matrix of the sensor to be detected, used to adjust the boundary - error feedback of the observer at the boundary; u(t) = [u1(,t), u2(x,t), …, u n (x,t)] T ∈R n represents the control - input vector; u i (x,t) is the control input of the i - th state data acting on position x of the sensor to be detected at time t.

[0016] Further preferably, the preset threshold is αρ; among them, ρ is the detection threshold, and its expression is:

[0017]

[0018] Wherein, 1.5≤α≤2.5; l is the spatial length of the sensor to be detected; is the upper limit of the error of the partial differential equation model of the sensor to be detected; is the upper limit of the disturbance of the partial differential equation model of the sensor to be detected; c is a constant in the partial differential equation model of the sensor to be detected.

[0019] More preferably, α=2.

[0020] Further preferably, the partial differential equation model of the sensor to be detected is:

[0021]

[0022] Its boundary conditions and output are defined as:

[0023] v x (0,t)=Qv(0,t), v(l,t)=u(t)

[0024] y(t)=v(0,t)

[0025] Among them, v(x,t)=[v1(x,t),v2(x,t),…,v n (x,t)] T ∈[L2(0,l)] n represents the state vector of the sensor to be detected; v i (x, t) is the i-th state data of the sensor to be detected at position x at time t; i = 1, 2, ..., n; n is the state data dimension of the sensor to be detected; L2(0, l) is the quadratic integrable function space defined on (0, l); d(x, t) = [d1(x, t), d2(x, t), ..., d n (x,t)]∈R n is the disturbance vector; d i (x, t) is the disturbance of the i-th state at the position x of the sensor to be detected at time t; u(t) = [u1(x, t), u2(x, t), …, u n (x,t)] T ∈R n represents the control input vector; u i (x, t) is the control input of the i-th state data acting at the position x of the sensor to be detected at time t; is a full-rank square matrix; It is a Lipschitz continuous nonlinear vector function, which is used to describe the dynamic characteristics of the actual sensor to be detected; y(t) is the actual output of the sensor to be detected at time t.

[0026] Further preferably, the constraint conditions of the partial differential equation model of the sensor to be detected include: for all x and t≥0, the following is satisfied: in, is a preset constant.

[0027] Further preferably, the above-mentioned obtaining of current detection data of the sensor to be detected in the petroleum refining device includes: collecting the current detection data of the sensor to be detected from the petroleum refining device and performing preprocessing; wherein the preprocessing includes: denoising operation and normalization operation.

[0028] In a second aspect, the present invention provides a failure detection system for a sensor of a petroleum refining device, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the failure detection method provided in the first aspect of the present invention when executing the computer program.

[0029] In a third aspect, the present invention provides a petroleum refining device, comprising: a sensor and the failure detection system provided in the second aspect of the present invention.

[0030] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is executed by a processor, the device where the storage medium is located is controlled to execute the failure detection method provided in the first aspect of the present invention.

[0031] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0032] 1. The present invention provides a failure detection method for a sensor of a petroleum refining device, wherein a sensor to be detected of the petroleum refining device is represented by a partial differential equation model, and a nonlinear observer of the sensor to be detected is constructed based on the partial differential equation model of the sensor to be detected; then, the output is observed by the observer, and failure detection is performed based on the detection residual between the observed value and the corresponding actual output; the nonlinear observer constructed by the present invention compensates for disturbances at different positions through a gain matrix, and even if different positions of the sensor are disturbed, these position differences can be responded to through feedback gain, thereby maintaining residual stability, and being able to cope with multi-input and multi-output situations with position interference, and it satisfies the condition that under the condition that the sensor to be detected is healthy, the detection residual is ultimately bounded, and it is easy to determine a preset threshold for failure judgment based on the detection residual, and at the same time, there is no need to reduce the partial differential equation model to an ordinary differential equation for solving, thereby avoiding false alarms and omissions that may be caused by model reduction, and being able to accurately perform failure detection on the sensor of the petroleum refining device.

[0033] 2. Furthermore, the failure detection method provided by the present invention determines the preset threshold as α based on the boundedness analysis of the system error and the stability requirement.ρ ;in, The failure detection accuracy can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flow chart of a failure detection method for a sensor of a petroleum refining device provided by an embodiment of the present invention;

[0035] Figure 2 A detailed flow chart of a failure detection method for a sensor of a petroleum refining device provided by an embodiment of the present invention;

[0036] Figure 3 A diagram of the construction process of the partial differential equation model and nonlinear observer of the sensor to be detected provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0038] In order to achieve the above objectives, firstly, Figure 1 As shown, the present invention provides a failure detection method for a sensor of a petroleum refining device, comprising:

[0039] a. Obtaining the current detection data of the sensor to be detected in the petroleum refining device; wherein the current detection data includes: the device information of the sensor to be detected, and the status data at the current time and all times before the current time;

[0040] In an alternative embodiment, Figure 2 As shown, the above-mentioned acquisition of current detection data of the sensor to be detected in the petroleum refining device includes: collecting the current detection data of the sensor to be detected from the petroleum refining device and performing preprocessing; wherein, the preprocessing includes: denoising operation and normalization operation.

[0041] Specifically, the equipment information of the sensors of the oil refining unit and the status data at the current moment and all moments before the current moment are collected, and pre-processing operations such as cleaning, normalization and smoothing are performed on them to ensure the reliability and consistency of the data. The specific process is as follows:

[0042] Data collection: The equipment monitoring system or data acquisition system in the oil refining unit is usually connected to various sensors and equipment to monitor and record the equipment's operating status, temperature, pressure, flow and other parameters in real time. When a sensor fails or is abnormal, the monitoring system will record this data for subsequent analysis and processing. Therefore, the collected data comes from the sensor monitoring system automatically recording failure events and generating corresponding logs or alarms, the local database or remote server where the monitoring system stores failure data, and the regular maintenance table of data engineers or maintenance personnel. The collected data includes normal operation data before, after and without failure. These data reflecting the failure factors are related to equipment attributes, failure types, operating environment, etc.

[0043] Specifically, sensor equipment information (mainly including equipment attributes, such as equipment ID, equipment type, etc.) and status data at the current moment and all moments before the current moment (such as operating status, temperature, pressure, flow, etc.) are collected according to the sensors to be detected. The data sources used for collection include sensor detection systems, data loggers, data acquisition equipment, databases and historical records, SCADA systems (monitoring, control and data acquisition systems), third-party data suppliers, etc. In order to analyze the failure component data and the influencing component data, the above-mentioned equipment selected for data collection needs to be accompanied by sufficient data to obtain the required statistical confidence, and the time scale is greater than a complete equipment life cycle.

[0044] Data preprocessing: In order to avoid invalid data from having an invalid impact on the overall results, the data is preprocessed by cleaning, normalizing, and smoothing, thereby removing duplicate and invalid data caused by some specific problems. The equipment classification type also needs to be determined in advance based on expert advice. The purpose of data cleaning and denoising is to check data quality, identify and process missing values, outliers, and noise. Missing values ​​can be filled by interpolation, and outliers and noise can be removed or corrected by filters or anomaly detection algorithms. Data normalization aims to normalize the collected data from different sensors to eliminate the differences between different dimensions and orders of magnitude. Commonly used normalization methods include minimum-maximum normalization and standardization. Data smoothing smoothes the data by applying methods such as moving average and exponentially weighted moving average to reduce noise and volatility.

[0045] b. Construct a nonlinear observer of the sensor to be detected based on the partial differential equation model of the sensor to be detected, input the current detection data into the observer of the sensor to be detected for solving, and obtain the observed value; calculate the difference between the observed value and the actual output of the sensor to be detected to obtain the detection residual;

[0046] like Figure 3As shown in the figure, according to the physical characteristics of the oil refining device and the response principle of the sensor to be detected, the partial differential equation of the sensor to be detected is established to describe the dynamic behavior of the system and the sensor characteristics, and then the detection observer is designed to produce the estimated state and output of the established system; specifically, the following processes are included:

[0047] 1) Sensor model establishment: In the present invention, nonlinear distributed parameter systems (DPS) are used to represent the sensors to be detected in the petroleum refining device, wherein the DPS can be described by partial differential equations (PDE). Therefore, according to the physical characteristics of the petroleum refining device under study and the response principle of the sensor, a suitable PDE is derived to describe it;

[0048] In an optional implementation, the sensor to be detected is regarded as a multi-input multi-output (MIMO) nonlinear distributed parameter system (DPS), and its partial differential equation model (PDE) is:

[0049]

[0050] Its boundary conditions and output are defined as:

[0051] v x (0,t)=Qv(0,t), v(l,t)=u(t)

[0052] y(t)=v(0,t)

[0053] Among them, v(x,t)=[v1(x,t),v2(x,t),…,v n (x,t)] T ∈[L2(0,l)] n represents the state vector of the sensor to be detected; v i (x, t) is the i-th state data of the sensor to be detected at position x at time t; i = 1, 2, ..., n; n is the state data dimension of the sensor to be detected; L2(0, l) is the quadratic integrable function space defined on (0, l); d(x, t) = [d1(x, t), d2(x, t), ..., d n (x,t)]∈R n is the disturbance vector; d i (x, t) is the disturbance of the i-th state at the position x of the sensor to be detected at time t; u(t) = [u1(x, t), u2(x, t), …, u n (x,t)] T ∈R n represents the control input vector; u i(x, t) is the control input of the i-th state data at the position x of the sensor to be detected at time t; is a full-rank square matrix; is a Lipschitz continuous nonlinear vector function used to describe the dynamics of the actual sensor to be detected; is the actual output of the sensor to be detected at time t, which is used to design the observer and generate the detection residual.

[0054] In an alternative embodiment, the constraint conditions of the partial differential equation model of the sensor include: for all x and t ≥ 0, it satisfies: where, is a preset constant.

[0055] In an alternative embodiment, the spatial length l = 1 can be selected, the constant (i.e., the system diffusion coefficient) c = 1 in the partial differential equation model of the sensor to be detected, and the upper bound of the interference

[0056] 2) Design of the observer: This observer is used to monitor the health status of the nonlinear DPS;

[0057] The present invention designs an adaptive observer to generate the estimated state and output of the sensor to be detected (DPS system), which shows that under healthy conditions, the detection residual is ultimately bounded (upper bound, UB). Instead of converting the DPS into an infinite set of ODEs to define the failure detection observer and boundary conditions, based directly on the conditions represented by the PDE, the equation of the observer is obtained as:

[0058]

[0059] Its boundary conditions are:

[0060]

[0061] Its observed value is:

[0062]

[0063] where, is the state estimation vector at the position x of the sensor to be detected at time t; 0 < x < l; l is the spatial length of the sensor to be detected; L2(0, l) is the space of square-integrable functions defined on (0, l); is the i-th state estimation data at the position x of the sensor to be detected at time t; i = 1, 2,..., n; n is the dimension of the state data of the sensor to be detected; c is the constant in the partial differential equation model of the sensor to be detected; It is a Lipschitz continuous nonlinear vector function, which is used to describe the dynamic characteristics of the observed sensor to be detected; is the error feedback control term; is the actual output of the sensor to be detected at time t; is a full-rank square matrix; n is the state data dimension of the sensor to be detected; P1(x, t) is the first gain matrix of the observer, which is used to adjust the feedback of the sensor to be detected on the state estimation error; represents the state derivative at position 0; P 10 (t) is the second gain matrix of the sensor to be detected, which is used to adjust the boundary error feedback of the observer at the boundary; u(t) = [u1(x, t), u2(x, t), …, u n (x,t)] T ∈R n represents the control input vector; u i (x, t) is the control input of the i-th state data acting at the position x of the sensor to be detected at time t.

[0064] It should be noted that P1(x, t) is responsible for adjusting the feedback of the system observer to the state estimation error to improve the adaptability of the observer in space and time. 10 (t) The feedback gain matrix used for boundary conditions is mainly used to adjust the boundary error feedback of the observer at the boundary (such as x=0).

[0065] It should be noted that the above observer may be a nonlinear Luenberger observer or other observers, such as a Kalman filter, which is not limited here.

[0066] It can be regarded as an error correction mechanism. Taking the above observer as a nonlinear Luenberger observer as an example, P1(x, t) directly affects the performance of the observer, including the convergence speed and system stability. Its design considers error feedback control, error convergence speed, matching of system dynamic characteristics, stability and robustness.

[0067] Specifically, the detection residual is It can also be used to correct the state estimation error caused by the initial conditions By comparing the observer dynamics difference (observer) and the difference between the actual system dynamics (partial differential equation model) v x (0, t) = Qv(0, t), v(l, t) = u(t), y(t) = v(0, t); where the nonlinear vector function f(v(x, t), x) satisfies the following conditions:

[0068] 1) For x∈[0,l],t≥0,v(x,t)∈L2(0,l),f(v(x,t),x) is Lipschitz continuous in v and is Continuous, at t continuous;

[0069] 2) f(v(x,t),x) should satisfy where Δv represents a small change in v(x,t); ε f (Δv,x) is the approximation error, which satisfies Where ||·|| represents the Frobenius norm or Euclidean norm of the matrix;

[0070] Therefore, the state estimation error dynamic equation and boundary conditions under healthy conditions are obtained as follows:

[0071]

[0072] in, From When A(x, t) becomes positive, the system will become unstable. In order to eliminate the term A(x, t) that may make the system unstable, an appropriate observer gain is selected and the Volterra integral transformation is used to obtain: And choose the observer gains P1(x,t) and P 10 (t) is: P 10 (t) = L(0,0,t) converts the observer error dynamics formula into:

[0073] L(x,σ,t) is the unique solution to the well-defined PDE defined by the following formula:

[0074]

[0075] b>0 is an arbitrary scalar;

[0076] In the above formula and is the inverse transform The kernel matrix.

[0077] Observer Error Dynamics Equation The observer gain P1(x, t) is used to adjust the error e(t) between the observer state and the actual state, thereby ensuring the boundedness and asymptotic convergence of the error dynamics. The constructed observer satisfies:

[0078] 1. Lyapunov stability: P1(x, t) is used to control the feedback strength of the observer error detection residual e(t) to adjust the convergence speed, thereby ensuring the stability of the observer error, that is, the Lyapunov derivative is negative definite. Based on the gain matrix P1(x, t), the Lyapunov derivative of the error dynamic equation can remain negative definite, thereby ensuring that the error converges to 0 under non-failure conditions.

[0079] 2. Error boundedness: In the presence of nonlinear approximation errors and disturbances, the observer can remain bounded, thus avoiding infinite error growth and ensuring the accuracy of the observer.

[0080] When the above observer is a Kalman filter, its gain matrix is ​​usually expressed as K(x, t), which has a similar function to P1(x, t) and will not be described in detail here.

[0081] The observer compensates for disturbances at different positions through the gain matrix. Even if the sensor is disturbed at different positions, it can respond to these position differences through feedback gain to keep the residual stable. The gain matrix is ​​a multi-dimensional state matrix. The observer adapts to the complex dynamics of the MIMO system by setting independent estimation and feedback correction terms for each output dimension, enabling it to track and feedback the multi-dimensional variables of the MIMO system.

[0082] c. Determine whether the detection residual is greater than a preset threshold. If so, determine that the sensor to be detected is failed; otherwise, determine that the sensor to be detected is not failed.

[0083] In an optional implementation manner, the preset threshold is α ρ ; Among them, ρ is the detection threshold, and its expression is:

[0084]

[0085] Wherein, 1.5≤α≤2.5; l is the spatial length of the sensor to be detected; is the upper limit of the error of the partial differential equation model of the sensor to be detected; is the upper limit of the disturbance of the partial differential equation model of the sensor to be detected; c is a constant in the partial differential equation model of the sensor to be detected.

[0086] In an optional implementation, when the sensor system is stable, 1.5≤α≤2; when the disturbance is large, 2≤α≤2.5. Preferably, α=2.

[0087] It should be noted that the preset threshold is determined based on the boundedness analysis of the system error and the stability requirements. The Lyapunov function V(t) is used to study the variation of the error over time and space, so as to obtain the boundedness of the error. The derivative of the Lyapunov function is It expresses the convergence trend of the system state and is usually required to be negative definite in the non-failure state.

[0088] The steps for determining the preset threshold are as follows:

[0089] 1. Select Lyapunov function: Define a Lyapunov function V(t) to analyze the stability of the system. This function represents the sum of squares of errors and integrates the partial derivatives of the spatial position:

[0090]

[0091] Among them, Ξ(x,t) is the representation of the observer error in space and is the observer error Through the result of integral transformation, its dynamic equation becomes stable under the appropriate kernel function selection, which helps to keep the error bounded in the non-failure case. x (t) is the partial derivative of Ξ(x,t) with respect to the control variable x.

[0092] 2. Derivative and apply integration by parts: Derivative V(t) with respect to time, and integrate by parts and substitute the boundary conditions:

[0093]

[0094] In the partial differential equations of distributed parameter systems, c represents the diffusion coefficient, which is usually related to the spatial dynamic characteristics of the system. b is the damping or attenuation coefficient in the observer error dynamics equation, which is used to control the convergence rate of the error in time t.

[0095] 3. Apply Poincare inequality: Combined with the upper bound expression of the error, use Poincare inequality to relate the error to its first-order derivative, and we get:

[0096]

[0097] 4. Derive the threshold formula: Combining all the results, we can finally get the upper bound of the residual under the non-failure condition, that is, the detection threshold ρ:

[0098]

[0099] The final preset threshold is 2ρ, which is twice the value to reduce the risk of false alarms. When the threshold is derived through Poincara and Agmon inequalities, a conservative estimate is obtained, which means that it is often lower than the actual value. More importantly, it is obtained by testing the selected threshold with historical data. The following is the proof process of sensor detectability:

[0100] When the sensor fails, the observer error changes dynamically as follows:

[0101]

[0102] For t ≥ t s ,Will Applying the above formula, we can get Ξ(l,t)=0;

[0103] Solving the partial differential equation represented by the formula gives

[0104] The obtained detection residual is:

[0105] in,

[0106]

[0107] when

[0108] holds and the selected detection threshold satisfies:

[0109]

[0110] This will result in:

[0111]

[0112] Therefore, it can be ensured that sensor failures that meet given conditions are detected.

[0113] In a second aspect, the present invention provides a failure detection system for a sensor of a petroleum refining device, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the failure detection method provided in the first aspect of the present invention when executing the computer program.

[0114] The related technical solution is the same as the failure detection method provided in the first aspect of the present invention, and will not be described in detail here.

[0115] In a third aspect, the present invention provides a petroleum refining device, comprising: a sensor and the failure detection system provided in the second aspect of the present invention.

[0116] The related technical solution is the same as the failure detection system provided in the second aspect of the present invention, and will not be described in detail here.

[0117] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is executed by a processor, the device where the storage medium is located is controlled to execute the failure detection method provided in the first aspect of the present invention.

[0118] The related technical solution is the same as the failure detection method provided in the first aspect of the present invention, and will not be described in detail here.

[0119] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting failure of a sensor in a petroleum refining device, characterized in that: include: Obtain the current detection data of the sensors to be detected in the oil refining device; The current detection data includes: the status data of the sensor to be detected at the current moment and all moments before the current moment; A nonlinear observer of the sensor to be detected is constructed based on the partial differential equation model of the sensor to be detected, and current detection data is input into the observer for solving to obtain an observation value; the difference between the observation value and the actual output of the sensor to be detected is calculated to obtain a detection residual; Determine whether the detection residual is greater than a preset threshold, if so, determine that the sensor to be detected is failed, otherwise, determine that the sensor to be detected is not failed; The observer equation is: Its boundary conditions are: Its observed values ​​are: Among them, is the state estimation vector at the position x of the sensor to be detected at time t; 0 < x < l; l is the spatial length of the sensor to be detected; c is a constant in the partial differential equation model of the sensor to be detected; is a Lipschitz continuous non-linear vector function used to describe the dynamic characteristics of the observed sensor to be detected; is the error feedback control term; P1(x,t) is the first gain matrix of the observer, used to adjust the feedback of the sensor to be detected on the state estimation error; is the actual output of the sensor to be detected at time t; is a full-rank square matrix; n is the dimension of the state data of the sensor to be detected; represents the state derivative at position 0; P 10 (t) is the second gain matrix of the sensor to be detected, used to adjust the boundary error feedback of the observer at the boundary; u(t) = [u1(x,t), u2(x,t), …, u n (x,t)] T ∈R n represents the control input vector; u i (x,t) is the control input of the i-th state data acting on the position x of the sensor to be detected at time t.

2. The failure detection method of a petroleum refining device sensor according to claim 1, characterized in that: The preset threshold is αρ; where ρ is the detection threshold, and its expression is: Wherein, 1.5≤α≤2.5; l is the spatial length of the sensor to be detected; is the upper limit of the error of the partial differential equation model of the sensor to be detected; is the upper limit of the disturbance of the partial differential equation model of the sensor to be detected; c is a constant in the partial differential equation model of the sensor to be detected.

3. The failure detection method of a petroleum refining device sensor according to claim 2, characterized in that: α＝2。 4. The failure detection method of a petroleum refining device sensor according to claim 1, characterized in that: The obtaining of the current detection data of the sensor to be detected in the petroleum refining device comprises: collecting the current detection data of the sensor to be detected from the petroleum refining device and performing preprocessing; The preprocessing includes: denoising operation and normalization operation.

5. The method for detecting failure of a sensor in a petroleum refining device according to any one of claims 1 to 4, characterized in that: The partial differential equation model of the sensor to be detected is: Its boundary conditions and output are defined as: v x (0,t)=Qv(t0,t),v(l,t)=u(t) y(t)=v(0,t) Among them, v(x,t)=[v1(x,t),v2(x,t),…,v n (x,t)] T ∈[L2(0,l)] n represents the state vector of the sensor to be detected; v i (x, t) is the i-th state data of the sensor to be detected at position x at time t; i = 1, 2, ..., n; n is the state data dimension of the sensor to be detected; L2(0, l) is the quadratic integrable function space defined on (0, l); d(x, t) = [d1(x, t), d2(x, t), ..., d n (x,t)]∈R n is the disturbance vector; d i (x, t) is the disturbance of the i-th state at the position x of the sensor to be detected at time t; u(t) = [u1(x, t), u2(x, t), …, u n (x,t)] T ∈R n represents the control input vector; u i (x, t) is the control input of the i-th state data acting at the position x of the sensor to be detected at time t; is a full-rank square matrix; It is a Lipschitz continuous nonlinear vector function, which is used to describe the dynamic characteristics of the actual sensor to be detected; y(t) is the actual output of the sensor to be detected at time t.

6. The method for detecting failure of a sensor in a petroleum refining device according to claim 5, characterized in that: The constraint conditions of the partial differential equation model of the sensor to be detected include: for all x and t≥0, the following conditions are satisfied: in, is a preset constant.

7. A failure detection system for a sensor in a petroleum refining device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor executes the failure detection method for a sensor of a petroleum refining device according to any one of claims 1 to 6 when executing the computer program.

8. A petroleum refining device, characterized in that: include: A sensor and a failure detection system as claimed in claim 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed by a processor, the device where the storage medium is located is controlled to execute the failure detection method for a sensor of a petroleum refining device according to any one of claims 1 to 6.

Images