Sensor state detection method and apparatus, electronic device, and storage medium

Abstract

The application provides a sensor state detection method and device, electronic equipment and storage medium, comprising: constructing a sliding mode fault observer based on a fan linear variable parameter model; inputting sensor measurement data and a control variable output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal; and correcting the control variable according to the sensor fault compensation signal. The application designs a fault observer based on a fan linear variable parameter model, extends a sensor at risk of failure to the fan linear variable parameter model as a fan component, and then detects the fault, thereby improving the precision and application range of sensor fault diagnosis, providing the ability of fault detection, isolation and fault-tolerant control after sensor failure or failure, reducing the influence of model uncertainty on fault diagnosis, and having strong physical concept, while meeting safety and increasing economy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromechanical fault diagnosis, and particularly relates to a sensor state detection method and device, electronic equipment and a storage medium. BACKGROUND

[0002] For fault detection and isolation (FDI) and fault tolerant control (FTC) after sensor failure or invalidation, the existing technology can be divided into three categories, which are hardware redundancy, information method and soft redundancy method.

[0003] The traditional hardware redundancy method will cause the cost to rise, and the double redundancy (i.e. configuring two sensors for the same signal) solution can only detect the fault, and it is difficult to locate the fault.

[0004] The information method, i.e. the data-driven and signal-based method, relies too much on the analytical redundancy of the system and the sensor redundancy, and its effect cannot be guaranteed when the data sample size of a specific fan is insufficient, and more research is needed in reducing the sensitivity to noise and interference.

[0005] The model-based soft redundancy method can solve the above problems, which relies on mathematical models to generate observation values online, but the wind turbine is a relatively complex nonlinear system, and the model generated by the piecewise linearization method can only guarantee the accuracy within a specific range, and the accuracy will decrease significantly once it exceeds the specific range, so the use range of the method is limited.

[0006] Therefore, it is urgent to improve the existing sensor state detection method to overcome the shortcomings of the existing technology. SUMMARY

[0007] The present application provides a sensor state detection method, device, electronic equipment and storage medium, which solves or improves the defects of the prior art in implementing sensor state detection and fault tolerant control, so as to meet the safety of sensor state detection and improve the economy.

[0008] In a first aspect, the present application provides a sensor state detection method, comprising: constructing a sliding mode fault observer based on a wind turbine linear variable parameter model;

[0009] Inputting the sensor measurement data and the control amount output by the wind turbine controller into the sliding mode fault observer to obtain a sensor fault compensation signal;

[0010] According to the sensor fault compensation signal, the control amount is corrected.

[0011] The sensor state detection method provided by the application further comprises the following steps before constructing a sliding mode fault observer based on a fan linear variable parameter model:

[0012] A fan linear variable parameter model is constructed based on a control variable, a sensor fault signal and a system state variable.

[0013] The sensor state detection method provided by the application, wherein the step of constructing a sliding mode fault observer based on a fan linear variable parameter model comprises the following steps:

[0014] A sensor fault signal matrix is constructed.

[0015] The sensor fault signal matrix is augmented to the fan linear variable parameter model to obtain an augmented fault signal matrix.

[0016] The sliding mode fault observer is constructed based on the augmented fault signal matrix.

[0017] The sensor state detection method provided by the application further comprises the following steps before inputting sensor measurement data and a control variable output by a fan controller into the sliding mode fault observer:

[0018] The coefficient matrix of an output vector error of the sliding mode fault observer, the coefficient matrix of an injected fault signal and the gain of the injected fault signal are solved by optimizing a linear matrix inequality among the three as a constraint and taking the L2 gain between the model uncertainty and a reconstruction difference value being less than a preset threshold as an objective, to obtain a current sliding mode fault observer.

[0019] The reconstruction difference value is the difference between a sensor fault reconstruction signal and a sensor fault signal.

[0020] The sensor state detection method provided by the application, wherein the step of inputting sensor measurement data and a control variable output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal comprises the following steps after the coefficient matrix of the output vector error, the coefficient matrix of the injected fault signal and the gain of the injected fault signal are solved:

[0021] The injected fault signal is determined.

[0022] The augmented fault signal matrix is converted according to the coefficient matrix of the output vector error, the coefficient matrix of the injected fault signal and the gain of the injected fault signal to obtain a current fault signal matrix.

[0023] A current reconstruction difference value is determined according to the current fault signal matrix and a current model uncertainty.

[0024] determining a current sensor fault reconstruction signal according to the current sensor fault signal and the current reconstruction difference;

[0025] determining the sensor fault compensation signal according to the current sensor fault reconstruction signal.

[0026] According to the sensor state detection method provided by the application, the sensor fault signal matrix is constructed, comprising:

[0027] dividing the sensor measurement data into fault risk data and non-fault risk data;

[0028] creating the sensor fault signal matrix based on the fault risk data.

[0029] According to the sensor state detection method provided by the application, after the sliding mode fault observer is constructed based on the augmented fault signal matrix, the method further comprises:

[0030] determining the output error of the sliding mode fault observer;

[0031] designing the sliding mode surface of the sliding mode fault observer according to the derivative result of the output error;

[0032] on the sliding mode surface, the output vector error of the sliding mode fault observer tends to 0 in a limited time.

[0033] In a second aspect, the application further provides a sensor state detection device, mainly comprising:

[0034] a model construction unit configured to construct a sliding mode fault observer based on a linear variable parameter model of a fan;

[0035] a signal compensation unit configured to input sensor measurement data and a control amount output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal;

[0036] a signal correction unit configured to correct the control amount according to the sensor fault compensation signal.

[0037] In a third aspect, the application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the sensor state detection method according to any of the above aspects when executing the program.

[0038] In a fourth aspect, the application further provides a non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the steps of the sensor state detection method according to any of the above aspects.

[0039] The sensor state detection method, device, electronic equipment and storage medium provided by the application, based on a fan linear variable parameter model, design a fault observer, and the sensor with a risk of failure is extended to the fan linear variable parameter model as a fan component, and then the fault is diagnosed, the precision and application range of the sensor fault diagnosis are improved, the ability of fault detection, isolation and fault-tolerant control after the failure of the sensor is provided, the influence of model uncertainty on fault diagnosis is reduced, the physical concept is strong, the safety is met, and the economy is increased. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0041] Figure 1 is one of the flowcharts of the sensor state detection method provided by the application;

[0042] Figure 2 is a structural schematic diagram of the fan state detection and signal reconstruction system provided by the application;

[0043] Figure 3 is the second flowchart of the sensor state detection method provided by the application;

[0044] Figure 4 is a structural schematic diagram of the sensor state detection system provided by the application;

[0045] Figure 5 is a structural schematic diagram of the electronic equipment provided by the application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical scheme and advantages of the application more clear, the technical scheme in the application will be clearly and completely described below in combination with the drawings in the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0047] It should be noted that in the description of the embodiments of the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article or device comprising the element. The terms "upper", "lower" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0048] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually a class, and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / ", generally indicates that the front and rear associated objects are in an "or" relationship.

[0049] The following will be described in conjunction with Figures 1-5 The sensor state detection method and device provided by the embodiments of the present application are described.

[0050] Figure 1 is one of the flowcharts of the sensor state detection method provided by the present application, as Figure 1 shown, including but not limited to the following steps:

[0051] Step 101: Construct a sliding mode observer (SMO) based on a fan linear parameter varying (LPV) model.

[0052] The main function of the sliding mode fault observer is to estimate the induced electromotive force of the fan motor and the detection signal of the sensor such as speed and position, and the biggest advantage is that the system disturbance such as uncertainty and parameter variation meeting the matching condition is invariant after the fan system enters the sliding mode.

[0053] The soft margin method adopted in the prior art is to directly generate the sensor-related observation value on line by using the fan LPV model, and the fan system is a relatively complex nonlinear system, so the accuracy of the sensor state detection cannot be guaranteed.

[0054] Therefore, the sensor state detection method provided by the application considers the advantages of the fan LPV model and the sliding mode fault observer, designs a fault observer based on the fan LVP model, can effectively improve the accuracy and application range of sensor state detection, and reduce the influence of the uncertainty of the fan LVP model on state detection.

[0055] Step 102: input the sensor measurement data and the control amount output by the fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal.

[0056] Figure 2 The structure diagram of the fan state detection and signal reconstruction system provided by the application is shown in Figure 2 As shown in the figure, Pr refers to the received instruction of the controller, generally including the fan speed and the output power; u(t) is the control amount of the controller; β r is the target pitch angle of the wind turbine blade system; τ g,r is the target torque of the motor; β m ,v w,m ,W r,m ,τ g,m ,W g,m ,P g is the sensor measurement data, which is the measurement value of the pitch angle, the measured wind speed, the measurement speed of the transmission system, the measurement torque of the transmission system, the measurement speed of the motor and the actual output power of the motor, and in actual use, appropriate increase or decrease can be made according to the sensor configuration of the fan system.

[0057] The signal output f0 under the sensor fault condition can be understood as the fault signal superimposed on the correct signal, and f cor is the sensor fault compensation signal generated by the sliding mode fault estimator.

[0058] The sliding mode fault estimator mainly includes a sliding mode fault observer, and is also used for processing the state observation result of the sliding mode fault observer to realize sensor fault estimation and output a sensor fault compensation signal.

[0059] Step 103: modifying the control quantity according to the sensor fault compensation signal.

[0060] In the sensor state detection method provided by the application, the fan controller (hereinafter referred to as the controller) mainly modifies the control quantity originally generated by the controller according to the sensor measurement data and the sensor fault compensation signal output by the sliding mode fault observer, so as to realize the control of the wind turbine operation state by using the modified control quantity instruction.

[0061] The sensor state detection method provided by the application designs a fault observer based on a wind turbine linear parameter-varying model, and extends the sensor at risk of failure to the wind turbine linear parameter-varying model as a wind turbine component, and then performs fault diagnosis, thereby improving the accuracy and application range of sensor fault diagnosis, providing the ability of fault detection, isolation and fault-tolerant control after sensor failure or failure, reducing the influence of model uncertainty on fault diagnosis, and having strong physical concept, while meeting safety requirements, increasing economy.

[0062] Based on the content of the above embodiment, as an optional embodiment, before constructing the sliding mode fault observer based on the wind turbine linear parameter-varying model, the following steps are further included:

[0063] Based on the control quantity, the sensor fault signal and the system state quantity, the wind turbine LPV model is constructed.

[0064] In the absence of special instructions, the system mentioned in the subsequent embodiments refers to the wind turbine system, that is, the wind turbine system.

[0065] Figure 3 Fig. 2 is a flowchart of the sensor state detection method provided by the application, as shown in the figure, before the sensor state detection, the wind turbine LPV model can be constructed in advance: Figure 3

[0066]

[0067] Wherein, x p (t) represents the system state quantity at time t; u(t) is the control quantity related signal at time t; f0(t) is the sensor fault signal at time t; A p (β), B p (β) are the coefficient matrices of the LPV model when the blade angle is β; C p , N p are the coefficient matrices; represents the derivative of the system state quantity at time t, which is the change of the system state quantity in unit time in the discrete case; y p (t) is the sensor measurement data output by the wind turbine LPV model at time t. ​

[0068] Further, considering that in the existing sensor state detection method, the sensor fault signal cannot be accurately distinguished and defined from the model uncertainty and the signal offset generated by the performance degradation of the fan system, thereby causing the sensor state detection result to be greatly deviated, the fan linear variable parameter model is constructed, the model uncertainty and the signal offset generated by the performance degradation of the fan system are considered as variable factors, and the expression of the constructed fan linear variable parameter model is specifically as follows:

[0069]

[0070] Wherein, D p (t) represents the signal offset generated by the performance degradation at t moment, and is a low communication signal; ξ(t) represents the model uncertainty of the fan LPV model at t moment, and is an unknown but bounded variable; A p (β), B p (β) constitute the coefficient matrix of the LPV model when the blade angle is β, M p , C p , N p are coefficient matrices.

[0071] As shown in Figure 2 , after the fan LPV model is constructed, the sliding mode fault observer can be constructed based on the fan LPV model, mainly including:

[0072] Constructing a sensor fault signal matrix; augmenting the sensor fault signal matrix to the fan linear variable parameter model to obtain an augmented fault signal matrix; and constructing the sliding mode fault observer based on the augmented fault signal matrix.

[0073] As an optional embodiment, the above-mentioned construction of the sensor fault signal matrix mainly includes the following steps:

[0074] Dividing the sensor measurement data into fault risk data and non-fault risk data; and creating a sensor fault signal matrix based on the fault risk data.

[0075] Optionally, the sensor measurement data y p (t) output by the fan LPV model is divided into two types of non-fault risk data y p,1 (t) and fault risk data y p,2 (t), and then the sensor measurement data y p (t) can be expressed as:

[0076]

[0077] Wherein, all the subscripts 1 represent non-fault risk items, and all the subscripts 2 represent fault risk items, such as: D p,1(t) is the signal bias representing the risk of failure at time t, D p,2 (t) is the signal bias representing the risk of failure at time t.

[0078] Further, for the risk of failure data y p,2 (t), a new sensor failure signal matrix Then:

[0079]

[0080] where A f is the Hurwitz matrix.

[0081] Further, to realize the risk of failure of the sensor as a fan component augmented to the system model, and then to diagnose the failure, the sensor failure signal matrix augmented to the pre-constructed fan LPV model to obtain the augmented failure signal matrix:

[0082]

[0083]

[0084] Further, the augmented failure signal matrix can be simplified as:

[0085]

[0086] where y m (t) is the sensor measurement data output by the augmented fan LPV model.

[0087] Further, based on the augmented failure signal matrix the current sliding mode fault observer is constructed:

[0088]

[0089] where, is the state estimation matrix at time t; is the output estimation matrix of the sliding mode fault observer at time t; e y (t) is the output vector error of the sliding mode fault observer at time t, is used to weigh and correct the output estimation matrix of the sliding mode fault observer; v(t) is the injected fault signal of the sliding mode fault observer, which is an estimate of the type of sensor failure signal, such as step, sine, etc.; G l (β), G n are the coefficient matrix of the output vector error and the coefficient matrix of the injected fault signal, respectively.

[0090] Based on the content of the above embodiment, as an optional embodiment, after the augmented fault signal matrix is constructed, the following can also be included:

[0091] The output error of the sliding mode fault observer is determined, the sliding surface of the sliding mode fault observer is designed according to the derivative result of the output error, and the output vector error of the sliding mode fault observer tends to 0 in a limited time on the sliding surface.

[0092] The output error of the sliding mode fault observer can be expressed as:

[0093]

[0094] Wherein The state observation error is represented, and the following can be obtained:

[0095]

[0096] Further, the sliding surface of the sliding mode fault observer is designed, that is, the output error e ym (t) of the sliding mode fault observer tends to 0 in a limited time on the sliding surface, and the expression of the sliding surface S can be:

[0097] S={e∈R n :e ym (t)=0}.

[0098] Based on the content of the above embodiment, as an optional embodiment, before the sensor measurement data and the control amount output by the fan controller are input into the sliding mode fault observer, the following is further included: by constructing a linear matrix inequality optimization constraint among the coefficient matrix of the output vector error of the sliding mode fault observer, the coefficient matrix of the injected fault signal, and the gain of the injected fault signal, the coefficient matrix of the output vector error, the coefficient matrix of the injected fault signal, and the gain of the injected fault signal are solved to obtain the current sliding mode fault observer, with the L2 gain between the model uncertainty and the reconstruction difference being less than a preset threshold as the target.

[0099] Wherein, the reconstruction difference is the difference between the sensor fault reconstruction signal and the sensor fault signal.

[0100] Further, the sensor state detection method provided by the present application can determine the parameters of the sliding mode fault observer according to the current sensor data after the sliding mode fault observer is constructed to obtain the current sliding mode fault observer, including: defining the sensor fault reconstruction signal:

[0101]

[0102] The gain W of the fault injection signal, the coefficient matrix G of the output vector error of the sliding mode fault observer in the above formula n , the coefficient matrix G of the fault injection signal l as the to-be-solved parameter.

[0103] In the sensor state detection method provided by the application, the linear matrix inequality optimization (LMI) method can be used, and by selecting appropriate G n , G l and the gain W, the L2 gain between the system uncertainty ξ (t) and is less than the preset threshold γ, wherein the preset threshold γ is a set small value.

[0104] Then, the injection fault signal of the current sliding mode fault observer is designed as:

[0105]

[0106] Wherein, the subscript j represents the jth sensor signal, k1, k2, k3, k4 are all coefficients defined according to actual requirements, and the definition of the sign function is:

[0107]

[0108] Further, according to the designed sliding mode fault injection signal, the sensor fault is estimated in real time to reconstruct the signal, and the expression is:

[0109]

[0110] Wherein, H (s) is a transformation according to the current fan LPV model, and the transformation result depends on the selection of G l , and ξ a is the uncertainty of the current system.

[0111] Finally, the sensor fault compensation signal f cor is expressed as:

[0112]

[0113] The sensor state detection method provided by the application adopts the design of the sliding mode fault observer based on the fan LVP model, so as to realize the fault signal reconstruction by using the designed current sliding mode observer, and corrects before the controller reads the sensor signal. Without additional design of fault-tolerant controller, the function is added to the original control system, so as to improve the robustness and fault-tolerant ability of the system.

[0114] Figure 4is a structural schematic diagram of the sensor state detection system provided by the application, as shown in Figure 4 The model construction unit 41, the signal compensation unit 42 and the signal correction unit 43 are mainly included.

[0115] The model construction unit 41 is mainly used for constructing a sliding mode fault observer based on a fan linear variable parameter model.

[0116] The signal compensation unit 42 is mainly used for inputting sensor measurement data and a control quantity output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal.

[0117] The signal correction unit 43 is mainly used for correcting the control quantity according to the sensor fault compensation signal.

[0118] The sensor state detection device provided by the application can effectively improve the precision and application range of sensor state detection and reduce the influence of the uncertainty of the fan LVP model on state detection.

[0119] The signal compensation unit 42 is in communication connection with each sensor in the wind turbine and is mainly used for receiving sensor measurement data uploaded by each sensor in real time. Meanwhile, the signal compensation unit 42 is in communication connection with the fan controller and is used for receiving sensor measurement data uploaded by each sensor in real time and a control quantity output by the controller. Finally, the collected sensor measurement data and the control quantity are input into the sliding mode fault observer to obtain a sensor fault compensation signal.

[0120] Finally, in the signal correction unit 43, the control quantity is corrected by using the sensor fault compensation signal, and the corrected control quantity is used to realize operation control of the fan generator.

[0121] In the sensor state detection device provided by the application, the fan controller (hereinafter referred to as the controller) mainly corrects a control quantity originally generated by the controller according to sensor measurement data and a sensor fault compensation signal output by the sliding mode fault observer, so as to realize control of the operation state of the wind turbine by using the corrected control quantity instruction.

[0122] The sensor state detection method provided by the application can design a fault observer based on a fan linear variable parameter model, extend a sensor with a risk of fault to the fan linear variable parameter model as a fan component, and then diagnose the fault, thereby improving the precision and application range of sensor fault diagnosis, providing the ability of fault detection, isolation and fault-tolerant control after sensor fault or failure, reducing the influence of model uncertainty on fault diagnosis, and being strong in physical concept, safe and economical.

[0123] It should be noted that the sensor state detection device provided by the embodiments of the present application can execute the sensor state detection method of any of the above-mentioned embodiments in specific operation, and the present embodiment will not be repeated.

[0124] Figure 5 is a structural schematic diagram of an electronic device provided by the present application, as Figure 5 shown, the electronic device can include a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 complete mutual communication through the communications bus 540. The processor 510 can invoke the logic instructions in the memory 530 to execute a sensor state detection method, which includes: constructing a sliding mode fault observer based on a fan linear variable parameter model; inputting sensor measurement data and a control quantity output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal; and correcting the control quantity according to the sensor fault compensation signal.

[0125] In addition, the logic instructions in the memory 530 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0126] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer readable storage medium, and the computer program includes program instructions, when the program instructions are executed by a computer, the computer can execute the sensor state detection method provided by the above-mentioned methods, which includes: constructing a sliding mode fault observer based on a fan linear variable parameter model; inputting sensor measurement data and a control quantity output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal; and correcting the control quantity according to the sensor fault compensation signal.

[0127] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a sensor state detection method provided by any of the above embodiments, the method comprising: constructing a sliding mode fault observer based on a linear variable parameter model of a fan; inputting sensor measurement data and a control quantity output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal; and correcting the control quantity according to the sensor fault compensation signal.

[0128] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed on a plurality of network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0129] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus necessary universal hardware platforms, and of course can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some part of the embodiments.

[0130] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A sensor state detection method characterized by, The method comprises the following steps: A sliding mode fault observer is constructed based on a linear variable parameter model of a fan, comprising the following steps: a sensor fault signal matrix is constructed; the sensor fault signal matrix is augmented to the linear variable parameter model of the fan to obtain an augmented fault signal matrix; and the sliding mode fault observer is constructed based on the augmented fault signal matrix; A linear matrix inequality optimization is performed among a coefficient matrix of an output vector error of the sliding mode fault observer, a coefficient matrix of an injected fault signal, and a gain of the injected fault signal, with a L2 gain between model uncertainty and a reconstruction difference being less than a preset threshold as an objective, to obtain a current sliding mode fault observer; the reconstruction difference is a difference between a sensor fault reconstruction signal and a sensor fault signal; Sensor measurement data and a control quantity output by a fan controller are input into the sliding mode fault observer to obtain a sensor fault compensation signal, comprising the following steps: the injected fault signal v(t) is determined, wherein an expression of v(t) is: According to the coefficient matrix G of the output vector error l (β), the coefficient matrix G of the injected fault signal n And the gain W of the injected fault signal, the augmented fault signal matrix is converted to obtain the current fault signal matrix. Specifically, the fault signal characteristics are extracted by the following error equation to obtain the current fault signal matrix: A current reconstruction difference is determined by calculating a difference between a sensor fault reconstruction signal and a sensor fault signal according to the current fault signal matrix and current model uncertainty ξ(t); A current sensor fault reconstruction signal is determined according to the current sensor fault signal and the current reconstruction difference; and the sensor fault compensation signal is determined according to the current sensor fault reconstruction signal; The control quantity is corrected according to the sensor fault compensation signal. wherein v j (t) represents the injection signal designed for the jth sensor in the sliding mode fault observer; k1, k2, k3, k4 are coefficients defined according to actual needs; z j (t) represents the fault state of the jth sensor fault signal at time t; represents the rate of change of the jth sensor fault signal with time; e ym,j (t) is the jth component of the output vector error of the sliding mode fault observer, representing the error between the measurement value of the jth sensor and the estimated value of the observer, y represents the sensor measurement value, m is the sensor measurement value output by the augmented wind turbine linear variable parameter model; F is the injection matrix of the sensor fault signal; M is the model uncertainty coefficient matrix; f0(t) is the sensor fault signal; the coefficient matrix A(β) is the coefficient matrix of the wind turbine linear variable parameter model when the blade angle is β; represents the rate of change of the output vector error of the sliding mode fault observer with time.

2. The sensor state detection method according to claim 1, characterized by, Before the sliding mode fault observer is constructed based on the linear variable parameter model of the fan, the method further comprises the following steps: The linear variable parameter model of the fan is constructed based on a control quantity, a sensor fault signal, and a system state quantity.

3. The sensor state detection method according to claim 1, characterized by, The sensor fault signal matrix is constructed, comprising the following steps: The sensor measurement data is divided into fault risk data and non-fault risk data; The sensor fault signal matrix is created based on the fault risk data.

4. The sensor state detection method according to claim 1, characterized by, After the sliding mode fault observer is constructed based on the augmented fault signal matrix, the method further comprises the following steps: An output error of the sliding mode fault observer is determined; A sliding surface of the sliding mode fault observer is designed according to a derivative result of the output error; An output vector error of the sliding mode fault observer tends to 0 in a limited time on the sliding surface.

5. A sensor state detection device characterized by comprising: The method comprises the following steps: The model construction unit is configured to construct a sliding mode fault observer based on a fan linear variable parameter model, and includes: constructing a sensor fault signal matrix; augmenting the sensor fault signal matrix to the fan linear variable parameter model to obtain an augmented fault signal matrix; constructing the sliding mode fault observer based on the augmented fault signal matrix; solving a coefficient matrix of an output vector error of the sliding mode fault observer, a coefficient matrix of an injected fault signal, and a gain of the injected fault signal, to obtain a current sliding mode fault observer, by taking the L2 gain between model uncertainty and a reconstruction difference being less than a preset threshold as an objective, and taking a linear matrix inequality optimization between the three as a constraint; the reconstruction difference is a difference between a sensor fault reconstruction signal and a sensor fault signal; The signal compensation unit is configured to input sensor measurement data and a control quantity output by a fan controller into the sliding mode fault observer to obtain a sensor fault compensation signal, and includes: determining the injected fault signal v(t), where an expression of v(t) is: According to the coefficient matrix G of the output vector error l (β), the coefficient matrix G of the injected fault signal n And the gain W of the injected fault signal, the augmented fault signal matrix is converted to obtain the current fault signal matrix. Specifically, the fault signal characteristics are extracted by the following error equation to obtain the current fault signal matrix: According to the current fault signal matrix and the current model uncertainty ξ(t), a current reconstruction difference is determined by calculating a difference between a sensor fault reconstruction signal and a sensor fault signal; According to the current sensor fault signal and the current reconstruction difference, a current sensor fault reconstruction signal is determined; and according to the current sensor fault reconstruction signal, the sensor fault compensation signal is determined; The signal correction unit is configured to correct the control quantity according to the sensor fault compensation signal. wherein v j (t) represents the injection signal designed in the sliding mode fault observer for the jth sensor; k1, k2, k3, k4 are coefficients defined according to actual needs; z j (t) represents the fault state of the jth sensor fault signal at time t; represents the rate of change of the jth sensor fault signal with time; e ym,j (t) is the jth component of the output vector error of the sliding mode fault observer, representing the error between the measurement value of the jth sensor and the estimated value of the observer, y represents the output of the system, m is the sensor measurement value output by the augmented fan linear variable parameter model; F is the injection matrix of the sensor fault signal; M is the model uncertainty coefficient matrix; f0(t) is the sensor fault signal.

6. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the sensor state detection method steps in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the sensor state detection method steps in any one of claims 1 to 4.

Images