A Fault Detection Method and System for a Reluctance Type Speed Sensor of a Gas Turbine

Abstract

The present invention relates to a method and system for fault detection of a reluctance type speed sensor of a gas turbine, and relates to the technical field of gas turbines. This method uses the operating data of the gas turbine when there is no fault to establish a fuzzy observer with fuel flow as the input and gas turbine speed as the output; inputs the fuel flow of the actually operating gas turbine into the fuzzy observer, compares the actually measured speed with the estimated speed output by the fuzzy observer to obtain the estimated relative difference; compares the estimated relative difference with the adaptive fault threshold, and when the relative difference is greater than the threshold, the control system issues a fault alarm for the speed sensor. The present invention can perform real-time fault detection on the reluctance type speed sensor of the gas turbine by establishing a fuzzy observer, and when a fault occurs, it alarms to prompt the gas turbine operator to repair the speed measurement circuit in time. It can eliminate the modeling uncertainty and the disturbance caused by the characteristics of the speed sensor itself, which is of great significance for the stable operation of the gas turbine.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas turbines, and particularly relates to a method and system for detecting faults of a magnetoresistive speed sensor of a gas turbine. Background Art

[0002] The rotational speed of a gas turbine is an important monitored input quantity for a gas turbine control system. Since a gas turbine operates in harsh environments such as high temperature, oil mist, and vibration, a magnetoresistive speed sensor is usually selected, and it is also one of the most prone-to-failure links in the gas turbine control system. After a speed sensor fails, drift faults, step faults, degradation faults, etc. that occur in its measured value will all cause the control loop to be inaccurate, and it is extremely easy to occur protection actions such as reducing the working condition and emergency shutdown, affecting the safe operation of the unit. Therefore, quickly and accurately detecting the faults of the magnetoresistive speed sensor of the gas turbine and timely repairing the sensor circuit is of great significance. Summary of the Invention

[0003] To solve the above technical problems, the present disclosure provides a method and system for detecting faults of a magnetoresistive speed sensor of a gas turbine.

[0004] To achieve the above object, some embodiments of the present application provide a method for detecting faults of a magnetoresistive speed sensor of a gas turbine, the method including the following steps:

[0005] Step 1, using the operation data of the gas turbine when there is no fault, setting fuzzy rules through a fuzzy inference base, and establishing a fuzzy observer;

[0006] Step 2, when the gas turbine is running, performing real-time rotational speed measurement through a rotational speed acquisition module;

[0007] Step 3, inputting the real-time fuel flow rate and rotational speed measurement value into the fuzzy observer, the fuzzy observer estimating the current rotational speed value according to the fuel flow rate, and using the input rotational speed of the previous moment to calculate a compensation gain to estimate the estimated value of the current rotational speed value;

[0008] Step 4, comparing the real-time rotational speed measurement value with the estimated rotational speed output by the fuzzy observer in Step 3 to obtain an estimated relative difference, and if the absolute value of the estimated relative difference is greater than an adaptive fault threshold, the control system issues a fault alarm for the magnetoresistive speed sensor.

[0009] Further, Step 1 further includes: representing the actual operating state of the gas turbine through the following gas turbine model:

[0010] ;

[0011] Wherein, u is the system input, representing the fuel flow rate, is a measurable state variable, N h , N p respectively represent the high-pressure compressor and power turbine speeds; f(x) is an unknown non-linear smooth function; y is of the same dimension as x and is the system output; M(t) is an unknown bounded fault function; A , B , C , D are known invertible matrices of appropriate dimensions.

[0012] Furthermore, the specific steps for establishing a fuzzy observer using the operating data of the gas turbine without faults in Step 1 include: collecting the test data of multiple gas turbines without faults, and using the actual fuel flow rate u and the real-time speed of the gas turbine x as the inputs of the fuzzy observer. The estimated value u of the non-linear part corresponding to the fuel flow rate x 1 is used as the output of the fuzzy observer. The real-time speed of the gas turbine x is used to calculate the compensation amount of the fuzzy observer, and the compensation amount is used to correct the output speed. The specific formula is:

[0013] ;

[0014] where, A, B, D is a known invertible matrix of appropriate dimensions, L is the observer gain; is a fuzzy function used to represent the non-linear part in the model f(x) , and through defuzzification, there is ; In the formula, is the weight vector, is the fuzzy basis function vector, g i (x) is a Gaussian form of fuzzy membership function, and its expression is as follows:

[0015] ;

[0016] where, is of the same dimension as x and is the center vector; is the width of the Gaussian function.

[0017] Furthermore, Step 1 further includes: based on the fuzzy observer, the following Riccati equation is given:

[0018] ;

[0019] where the symmetric matrix , β is a given positive constant; then its positive definite symmetric solution P can be obtained; thus, the relative difference in rotational speed estimation is defined, and the observer gain is , and the update rate of the weight vector w is:

[0020] .

[0021] Furthermore, step four further includes: calculating the relative estimation difference based on the estimated value obtained from the fuzzy observer, comparing the relative estimation difference with the fault threshold through a comparator to achieve fault detection, and selecting the adaptive fault threshold as , and we can get:

[0022] .

[0023] Furthermore, step four further includes: the calculation method of the adaptive fault threshold Y 1 is:

[0024] ;

[0025] where Z g1 is calculated as follows:

[0026] ;

[0027] where, x(t) is the current rotational speed sampling value, x L and x H are the given gas turbine rotational speed intervals, a 0 , a 1 , a 2 , a 3 and a 4 are adjustment coefficients; Z g0 is the reference threshold under ideal conditions, taking the maximum difference between the rotational speed estimation value and the actual sampling value under fault-free conditions; Z g1 is the adaptive adjustment part; if the absolute value of the relative estimation difference is greater than the adaptive fault threshold, the control system issues a fault alarm for the reluctance rotational speed sensor.

[0028] On the other hand, the present application also provides a fault detection system for a reluctance type speed sensor of a gas turbine. The system includes a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the system is triggered to execute the above-mentioned fault detection method for the reluctance type speed sensor of the gas turbine.

[0029] Compared with the prior art, the fault detection method and system for the reluctance type speed sensor of the gas turbine provided by the present application utilize the operating data of the gas turbine when there is no fault to establish a fuzzy observer, estimate the real-time speed of the gas turbine by using the fuel flow rate input to the gas turbine, and correct the estimated speed by using the speed acquisition value at the previous moment. The real-time acquired speed of the gas turbine is compared with the estimated speed output by the fuzzy observer, and the absolute value of the estimated relative difference is compared with the adaptive fault threshold. If the absolute value of the estimated relative difference is greater than the adaptive fault threshold, the control system issues an alarm of "speed sensor fault". By establishing a fuzzy observer, the present application can perform real-time fault detection on the reluctance type speed sensor of the gas turbine, and when a fault occurs, it alarms to prompt the gas turbine operator to repair the speed measurement circuit in time, which is of great significance for the stable operation of the gas turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flowchart of the fault detection method for the reluctance type speed sensor of the gas turbine provided by an embodiment of the present disclosure.

[0031] In the figure, u is the fuel flow rate input to the gas turbine, x is the measured value of the gas turbine speed, x 1 is the estimated value of the speed by the fuzzy observer, y is the estimated relative difference, Y 1 is the fault threshold. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] In order to more clearly understand the purpose, features and advantages of the present disclosure, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It should be noted that, without conflict, the embodiments of the present disclosure and their features can be combined with each other. Many specific details are provided in the following description for a comprehensive understanding of the present disclosure, but actually the present disclosure can be implemented in other ways, and these ways may be different from the following description. Obviously, the embodiments in the specification only represent some embodiments of the present disclosure, rather than all embodiments.

[0033] Since gas turbines operate in harsh environments such as high temperatures, oil mists, and vibrations, magnetoresistive speed sensors are usually selected. As the operating time of magnetoresistive speed sensors increases, the probability of their failure gradually increases, which in turn poses major potential faults to the gas turbine control system. The purpose of the present invention is to quickly and accurately detect the faults of magnetoresistive speed sensors, avoid protection actions such as load reduction and emergency shutdown of the gas turbine control system caused by abnormal speeds, and thus issue an alarm to remind the gas turbine operators to promptly repair the speed measurement circuit and reduce or eliminate the impact of the faults.

[0034] Magnetoresistive speed sensors of gas turbines are generally installed on the gas turbine casing. When the gas turbine rotor rotates, the sensor generates an induced pulse electromotive force with a certain frequency. By measuring the pulse frequency, the actual speed of the gas turbine can be calculated.

[0035] Since speed sensors operate in harsh environments such as high temperatures, oil mists, and vibrations, they are components that are extremely prone to failure. For magnetoresistive speed sensors, due to their own characteristics, the signal strength is weak at low speeds and is easily affected by external interference, resulting in speed fluctuations. Through a fault detection method, a fault threshold can be designed to distinguish between sensor faults and speed fluctuations caused by interference, avoiding unnecessary shutdown losses.

[0036] In view of the above problems, the embodiments of the present disclosure provide a method for detecting faults of magnetoresistive speed sensors of gas turbines. Next, this method will be introduced in combination with specific embodiments.

[0037] The following Figure 1 introduces the method for detecting faults of magnetoresistive speed sensors of gas turbines shown below. The specific steps included in this method are as follows:

[0038] Step 1: Using the operating data of multiple gas turbines when they are fault-free, taking the fuel flow rate input to the gas turbine u and the measured gas turbine speed as the inputs of the fuzzy observer. Among them, Nh and Np represent the speeds of the high-pressure compressor and the power turbine respectively; the estimated speed value x1 of the fuzzy observer is used as the output of the fuzzy observer. Through the fuzzy inference base, fuzzy rules are set to establish a fuzzy observer. The fuzzy inference base is a description tool based on fuzzy set theory and is used to perform mathematical logic expansion on the described object, that is, the test run data is normalized and fuzzified to give corresponding fuzzy rules. Fuzzy rules refer to the fuzzy relationships defined in the binary domain. It is a knowledge representation method based on fuzzy logic and is used to describe the knowledge or rules in the field of object control.

[0039] The fuzzy observer model is as follows:

[0040] ; (1)

[0041] Among them,A, B, D is a known dimensionally observable matrix, L is the observer gain; is a fuzzy function used to represent the nonlinear part in the model f(x) , and there is ; In the formula, is the weight vector, is the fuzzy basis function vector, g i (x) is a Gaussian - shaped fuzzy membership function, and its expression is as follows:

[0042] ; (2)

[0043] where, and x are of the same dimension and are the center vectors; is the width of the Gaussian function.

[0044] Based on the Gaussian fuzzy membership function, the appropriate number of fuzzy sets and inference rules can be selected through multiple experiments. Taking the high - pressure compressor speed and actual fuel flow of the gas turbine as inputs, and taking the estimated value of the fuzzy function for the nonlinear part of the high - pressure compressor speed in the model as the output, the following high - pressure compressor speed fuzzy rule table is adopted, as shown in Table 1:

[0045] ;

[0046] Taking the power turbine speed and actual fuel flow of the gas turbine as inputs, and taking the estimated value of the fuzzy function for the nonlinear part of the power turbine speed in the model as the output, the following power turbine speed fuzzy rule table is adopted, as shown in Table 2:

[0047] ;

[0048] Based on the above - mentioned fuzzy observer, the following Riccati equation is given,

[0049] ; (3)

[0050] In the formula, the symmetric matrix , β is a given positive constant; then its positive - definite symmetric solution P can be obtained; thus, the relative difference in speed estimation is defined, and then the observer gain is , and the update rate of the weight vector w is:

[0051] ; (4)

[0052] In some embodiments, the following gas turbine model represents the actual operating state of the gas turbine:

[0053] ;

[0054] wherein, u is the system input, representing the fuel flow rate, is a measurable state variable, N h and N p respectively represent the rotational speeds of the high-pressure compressor and the power turbine; f(x) is an unknown non-linear smooth function; y is of the same dimension as x and is the system output; M(t) is an unknown bounded fault function; A , B , C , D are known invertible matrices of appropriate dimensions.

[0055] Step 2: When the gas turbine is operating, obtain the current fuel flow rate through a fuel flow meter, and obtain the rotational speed of the high-pressure compressor N h and the rotational speed of the power turbine N p through a rotational speed acquisition module;

[0056] Step 3: When the gas turbine is operating, input the real-time fuel flow rate and rotational speed measurement values into the fuzzy observer established in Step 1. As described in Equation (1), the fuzzy observer estimates the current rotational speed value based on the fuel flow rate and the rotational speed value at the previous moment, and calculates the compensation gain using the input rotational speed at the previous moment to correct the current rotational speed estimate value;

[0057] Step 4: Compare the real-time measured rotational speed with the estimated rotational speed output by the fuzzy observer in Step 3 to obtain the estimated relative difference. If the absolute value of the estimated relative difference is greater than the adaptive fault threshold, the control system issues an alarm of "rotational speed sensor failure".

[0058] Furthermore, in Step 4, according to the estimated value obtained by the fuzzy observer, calculate the relative estimation difference , and through comparison with the fault threshold, realize the detection of the fault. Select the adaptive fault threshold as , and it can be obtained that

[0059] ;

[0060] In practice, for the reluctance type speed sensor of a gas turbine, the signal strength is weak at low speeds, and there is even an unstable sampling interval at the initial stage of startup. In this interval, the speed signal is prone to distortion due to external disturbances. As the speed increases, the signal strength and stability gradually increase, and the disturbance to the measured signal decreases. When the high-speed region is reached, a small change in speed will cause a large change in the operating conditions, and corresponding micro-compensation is required at this time. Therefore, using the traditional fixed-threshold detection not only makes it difficult to determine the threshold, but also prone to misdetection and missed detection. Therefore, the present invention proposes an adaptive fault threshold design method.

[0061] Specifically, due to the characteristics of the reluctance type speed sensor, the signal is weak and unstable at low speeds, and as the speed increases, its signal strength and stability will gradually increase. Thus, the adaptive fault threshold in step four is defined. Y 1 The calculation method is as follows:

[0062] ;

[0063] Among them Z g1 The calculation expression of

[0064] ;

[0065] Among them, x(t) is the current speed sampling value, x L and x H are the given gas turbine speed intervals, a 0 , a 1 , a 2 , a 3 and a 4 are adjustment coefficients; Z g0 is the reference threshold under ideal conditions, taking the maximum difference between the speed estimated value and the actual sampling value under fault-free conditions; Z g1 is the adaptive adjustment part; if the absolute value of the estimated relative difference is greater than the adaptive fault threshold, the control system issues a fault alarm for the reluctance type speed sensor.

[0066] After the alarm occurs, the gas turbine operator should stop the machine for inspection in a timely manner according to the actual situation and repair the speed measurement circuit to prevent more serious faults from occurring.

[0067] The present invention also provides a fault detection system for a reluctance type speed sensor of a gas turbine. The system includes a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the system is triggered to execute the foregoing fault detection method for the reluctance type speed sensor of the gas turbine.

[0068] In summary, for the fault detection method and system for a reluctance type speed sensor of a gas turbine provided by the present invention, a fuzzy observer with fuel flow as the input and gas turbine speed as the output is established by using the operation data of the gas turbine when it is fault-free; the fuel flow of the actually operating gas turbine is input into the observer, and by comparing the actually measured speed with the estimated speed output by the fuzzy observer, an estimated relative difference is obtained; by setting an adaptive threshold, the uncertainty in modeling and the disturbance caused by the characteristics of the reluctance type speed sensor itself are eliminated, and the estimated relative difference is compared with the threshold to detect the reluctance type speed sensor of the gas turbine. When the relative difference is greater than the threshold, the control system issues an alarm of "speed sensor fault" to remind the gas turbine operator to timely repair the speed measurement circuit and reduce or eliminate the impact of the fault.

[0069] It should be noted that the system of the present invention may further include one or more of the following components: a processor and a memory.

[0070] Optionally, the processor is connected to various parts within the entire autonomous operation device through various interfaces and lines. By running or executing the instructions, programs, code sets or instruction sets stored in the memory, and by calling the data stored in the memory, the processor executes various functions of the autonomous operation device and processes data. Optionally, the processor may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a neural-network processing unit (NPU). Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for the rendering and drawing of the content required to be displayed on the touch display screen; the NPU is used to implement artificial intelligence (AI) functions.

[0071] The memory may include a Random Access Memory (RAM), and may also include a Read-Only Memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the above method embodiments, etc.; the data storage area may store data created according to the use of the autonomous operation device, etc.

[0072] The present disclosure also provides a computer-readable storage medium storing a computer program, which is used to be executed by a processor to implement the positioning control method of the autonomous operation device as described in the above embodiments.

[0073] Although the present invention has been described in detail by referring to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions should all be within the scope covered by the present invention. Or any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims

1. A method for detecting faults of a gas turbine reluctance speed sensor, characterized in that: The method comprises the following steps: Step 1: Using the gas turbine running data without fault, through the fuzzy reasoning base, setting fuzzy rules, and establishing a fuzzy observer; Step 2: When the gas turbine is running, real-time speed measurement is performed through a speed acquisition module; Step 3, inputting the real-time fuel flow and speed measurement values ​​into the fuzzy observer, the fuzzy observer estimates the current speed value according to the fuel flow, and uses the input speed at the previous moment to calculate the compensation gain, and estimates the estimated value of the current speed value; Step 4: compare the real-time measured speed with the estimated speed output by the fuzzy observer in step 3 to obtain an estimated relative difference. If the absolute value of the estimated relative difference is greater than the adaptive fault threshold Y1, the control system issues a magnetoresistive speed sensor fault alarm. The calculation method of the adaptive fault threshold Y1 is as follows: Y1=Z g0 +Z g1 Where Z g1 The calculation expression is as follows: Among them, x(t) is the current speed sampling value, x L and x H is a given gas turbine speed range, a0, a1, a2, a3 and a4 are adjustment coefficients; Z g0 is the reference threshold under ideal conditions, taking the maximum difference between the estimated speed value and the actual sampling value under fault-free conditions; Z g1 This is the adaptive adjustment part.

2. The method according to claim 1, characterized in that The step 1 further comprises: representing the actual operating state of the gas turbine by the following gas turbine model: Where u is the system input, representing the fuel flow rate, x = [N h N p ] T is the measurable state quantity, N h 、N p They represent the speeds of the high-pressure compressor and the power turbine respectively; f(x) is an unknown nonlinear smooth function; y has the same dimension as x and is the system output; M(t) is an unknown bounded fault function; A, B, C, and D are reversible matrices of known appropriate dimensions.

3. The method according to claim 1, characterized in that The step 1 in which the fuzzy observer is established with the operating data of the gas turbine when it is fault-free specifically includes: collecting test data of multiple gas turbines when they are fault-free, using the actual fuel flow u of the gas turbine and the real-time speed x of the gas turbine as the input of the fuzzy observer, using the nonlinear part estimation value x1 corresponding to the fuel flow u as the output of the fuzzy observer, and the real-time speed x of the gas turbine is used to calculate the compensation amount of the fuzzy observer, and the compensation amount is used to correct the output speed. The specific formula is: x1(t+1)=Ax1(t)+Bu(t)+Dξ(x,w|u)+L(x(t)-x1(t)) Where A, B, and D are known dimensionally observable matrices, L is the observer gain, ξ(x, w|u) is a fuzzy function used to represent the nonlinear part f(x) in the model, and ξ(x, w|u) = w T G(x|u); where w=[ω1,ω2,…,ω n ] T is the weight vector, is the fuzzy basis function vector, g i (x) is the Gaussian fuzzy membership function, which is expressed as follows: Among them, C i =[c1, c2, …, c n ] T Same dimension as x, central vector; σ i is the width of the Gaussian function.

4. The method according to claim 3, characterized in that The step 1 further includes: based on the fuzzy observer, the following Riccati equation is given: In the formula, the symmetric matrix Ψ>0, β is a given positive constant; then its positive symmetric solution P can be obtained; thus, the relative difference of speed estimation is defined as e(t)=x1(t)-x(t), then the observer gain is The update rate of the weight vector w is:

5. The method according to claim 2, characterized in that: The step 4 further includes: calculating an estimated relative difference Y(t)=C(x1(t)-x(t)) according to the estimated value obtained by the fuzzy observer, and performing fault detection by comparing the absolute value of the estimated relative difference Y with the adaptive fault threshold Y1, which is expressed as:

6. A gas turbine reluctance speed sensor fault detection system, the system comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein: When the computer program instructions are executed by the processor, the system is triggered to execute the gas turbine reluctance speed sensor fault detection method according to any one of claims 1 to 5.

Images