CN113128832A - Operation state online diagnosis method and system for auxiliary system of large phase modulator - Google Patents

Operation state online diagnosis method and system for auxiliary system of large phase modulator Download PDF

Info

Publication number
CN113128832A
CN113128832A CN202110281437.7A CN202110281437A CN113128832A CN 113128832 A CN113128832 A CN 113128832A CN 202110281437 A CN202110281437 A CN 202110281437A CN 113128832 A CN113128832 A CN 113128832A
Authority
CN
China
Prior art keywords
diagnosed
index
indexes
auxiliary system
matrix
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110281437.7A
Other languages
Chinese (zh)
Inventor
熊富强
赵鹏
张超峰
闾昊辉
董卓
杨建明
刘亚楠
刘昊然
刘源
张阳
王天一
周挺
李佐胜
康文
蒋久松
瞿旭
雷云飞
彭舟
于艺盛
杨洋
王卿卿
严宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Corp of China SGCC
State Grid Hunan Electric Power Co Ltd
Maintenance Co of State Grid Hunan Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
State Grid Hunan Electric Power Co Ltd
Maintenance Co of State Grid Hunan Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Corp of China SGCC, State Grid Hunan Electric Power Co Ltd, Maintenance Co of State Grid Hunan Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202110281437.7A priority Critical patent/CN113128832A/en
Publication of CN113128832A publication Critical patent/CN113128832A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393Score-carding, benchmarking or key performance indicator [KPI] analysis
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01DMEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00Measuring or testing not otherwise provided for
    • G01D21/02Measuring two or more variables by means not covered by a single other subclass
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/16Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/18Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • G06N3/084Backpropagation, e.g. using gradient descent
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06Energy or water supply

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Strategic Management (AREA)
  • Artificial Intelligence (AREA)
  • Development Economics (AREA)
  • Algebra (AREA)
  • General Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Databases & Information Systems (AREA)
  • Biophysics (AREA)
  • Game Theory and Decision Science (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Cheminformatics (AREA)

Abstract

The invention discloses an online diagnosis method and system for the running state of a large phase modulator auxiliary system, which comprises the following steps: inputting original data of n indexes to be diagnosed of a large phase modifier auxiliary system, calculating degradation degrees to judge faults, solving membership degrees between the n indexes to be diagnosed and m running states of the large phase modifier auxiliary system under the condition of no faults, and constructing a fuzzy judgment matrix A between the n indexes to be diagnosed and the m running states; respectively solving weight matrixes S of n to-be-diagnosed indexes; and carrying out fuzzy operation on the S and the A to obtain a diagnosis matrix B, and selecting a final operation state as the operation state of the large phase modulator auxiliary system obtained by final evaluation according to a diagnosis result in the diagnosis matrix B. The invention can not only carry out on-line state diagnosis on the large phase modulator auxiliary system, but also provide a specific fault subsystem when the large phase modulator auxiliary system has slight fault and fault, thereby being convenient for finding out fault parts and implementing maintenance.

Description

Operation state online diagnosis method and system for auxiliary system of large phase modulator
Technical Field
The invention relates to an operation state detection technology of a large phase modulator auxiliary system, in particular to an operation state online diagnosis method and system for the large phase modulator auxiliary system.
Background
In recent years, with continuous planning and production of extra-high voltage direct current transmission projects, a converter needs to consume a large amount of reactive power in the transmission process. If the power grid fails, reactive power to be absorbed is greatly increased in the process of rapid dynamic adjustment; when the voltage of the power grid is too high, reactive power needs to be released so as to meet the requirement of the power grid on the voltage stability. It can be seen that the lack or excess of reactive power will cause instability of ac voltage, and in severe cases, the safe and stable operation of the whole ac and dc system may be endangered. The synchronous phase modulator is easy to adjust, and can emit reactive power and absorb the reactive power, so that the synchronous phase modulator is applied to extra-high voltage alternating current and direct current transmission projects. In view of the small number of synchronous phase modulators of the new generation, the short running time, and the complex and variable running environment and the load, the reliability needs to be checked in time. Therefore, in order to ensure the efficient operation and safe and stable operation of a large phase modulation unit in an extra-high voltage alternating current and direct current transmission project, the health state of the phase modulation unit needs to be correctly diagnosed, so that the maintenance and overhaul are reasonably arranged, and the service life of the phase modulation unit is prolonged.
The large capacity synchronous phase modifier has complicated structure and has some auxiliary systems besides the phase modifier for energy exchange. The normal work of the auxiliary system is an important basis for the normal operation of the phase modulation unit, the fault of the phase modulation unit can be found in time by carrying out state diagnosis on the auxiliary system, a judgment basis is provided for operation and maintenance decisions made by maintainers, the service life of the phase modulation unit is prolonged, and the stability and the safety of a power grid are guaranteed. Regular maintenance of the camera adjusting group can effectively avoid accidents, but the camera adjusting group is lack of flexibility and easy to cause excessive maintenance or insufficient maintenance, so that a large amount of manpower and material resources are consumed. The maintenance decision making according to the state diagnosis is different from the traditional maintenance, and is a mode of diagnosing the operation state of the phase modulation unit by monitoring the auxiliary system of the phase modulation unit on line, collecting operation data and obtaining the health state of each auxiliary part after data analysis. According to the diagnosis result, the deterioration degree of the phase modulation unit can be known, and a corresponding maintenance strategy is adopted. When the deterioration degree of the phase modulator is higher, the frequency of inspection and test is properly improved, and for the phase modulator with lower deterioration degree, the maintenance frequency is properly reduced, so that the phase modulator has better economical efficiency and reliability.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the invention can not only carry out online state diagnosis on the large phase modifier auxiliary system, but also provide a specific fault subsystem when the large phase modifier auxiliary system has slight fault and fault, thereby being convenient for finding out fault positions and implementing maintenance.
In order to solve the technical problems, the invention adopts the technical scheme that:
an operation state online diagnosis method for a large-scale phase modifier auxiliary system comprises the following steps:
1) inputting n original data of indexes to be diagnosed in a large phase modifier auxiliary system, wherein the n original data relate to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system;
2) respectively calculating the degradation degrees of the original data of n indexes to be diagnosed, and if the degradation degree of any index to be diagnosed is 1, judging that the auxiliary system of the large phase modulator fails and quits; otherwise, executing the next step;
3) respectively solving the membership degrees between n indexes to be diagnosed and m running states of the large phase modulator auxiliary system;
4) constructing a fuzzy judgment matrix A among n indexes to be diagnosed and m running states according to the membership degree;
5) respectively solving weight matrixes S of n to-be-diagnosed indexes;
6) carrying out fuzzy operation on the weight matrix S and the fuzzy evaluation matrix A to obtain a diagnosis matrix B, wherein the diagnosis matrix B comprises diagnosis results corresponding to m running states one by one;
7) and selecting a final operation state as the operation state of the large phase modulator auxiliary system obtained by final evaluation according to the diagnosis result in the diagnosis matrix B by adopting a maximum membership principle.
Optionally, in the n indexes to be diagnosed related to the stator cooling water system, the rotor cooling water system, the lubricating oil system and the high-pressure jacking oil system in the step 1): the indicators to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and water pH value; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level.
Optionally, when the degradation degree is calculated for the raw data of n items of indicators to be diagnosed in step 2), the step of calculating the degradation degree for the raw data of any one of the items of indicators to be diagnosed includes:
2.1) judging the type of the index to be diagnosed, wherein the related types comprise a bipolar limit type, an upper limit type, a lower limit type and a rated value type, the bipolar limit type means that the index to be diagnosed fails when exceeding the upper limit and the lower limit of a normal interval, the upper limit type means that the index to be diagnosed fails when exceeding the normal upper limit, the lower limit type means that the index to be diagnosed fails when exceeding the normal lower limit, the rated value type means that the index to be diagnosed fails when exceeding the specified margin range of the normal rated value, and if the index to be diagnosed is the bipolar limit type or the rated value type, skipping to execute the step 2.2); if the type of the upper limit value is the upper limit value type, skipping to execute the step 2.3); if the type of the lower limit value is the lower limit value type, skipping to execute the step 2.4);
2.2) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000031
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xmax、xminRespectively are the upper limit value and the lower limit value of the index to be diagnosed, and alpha-beta are respectively the allowable value range of the index to be diagnosed;
2.3) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000032
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, alpha is the allowable value of the normal operation of the index, and xmaxIs the upper limit value of the index;
2.4) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000033
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xminRespectively, the lower limit value of the index to be diagnosed.
Optionally, step 3) comprises: the operation state change process of the large phase modulator auxiliary system is gradual, normal distribution is met, and a normal membership function is adopted to calculate any index x to be diagnosediAnd operating state sjThe membership degree between the index data and the index data is substituted into n items of index data to be diagnosed in a certain operation state, and the membership degree between the n items of index data to be diagnosed and m operation states of the large phase modifier auxiliary system is analyzed from the n items of index data to be diagnosed, so that any index x to be diagnosed is obtainediAnd operating state sjThe degree of membership between is:
Figure BDA0002978884490000034
in the above formula, AijFor the index x to be diagnosediAnd operating state sjDegree of membership between, cjIs in an operating state sjCenter of normal distribution, σ is operating state sjThe width of the normal distribution.
Optionally, the function expression of the fuzzy judgment matrix a between the n indexes to be diagnosed and the m operating states is constructed according to the membership degree in step 4) and is as follows:
Figure BDA0002978884490000041
in the above formula, any AijFor the index x to be diagnosediAnd operating state sjThe membership degree between the two groups is 1,2, …, n, n is the number of indexes to be diagnosed, and j is 1,2, …, m, m is the number of running states.
Optionally, in step 5), the function expressions of the weights S of the n to-be-diagnosed indexes are respectively obtained as:
5.1) aiming at n indexes to be diagnosed, combining expert experience to carry out pairwise comparison according to the importance of each index to set a scale value to form a series of judgment matrixes, and calculating and judging the maximum characteristic value lambda of each judgment matrixmaxAnd corresponding eigenvectors thereof, and finally determining the subjective weight W of any ith index to be diagnosed according to the judgment matrixi
5.2) respectively solving the information entropy according to the following formula aiming at the sample data matrix formed by the original data of n indexes to be diagnosed, and calculating the objective weight H of any ith index to be diagnosed by adopting the following formulai
Figure BDA0002978884490000042
In the above formula, pijThe probability value Y of the ith index data in the jth operation state interval is obtained for statisticsijThe raw data of n indexes obtained by the state monitoring platform are i equal to 1,2, …, n, n is the number of indexes to be diagnosed, j equal to 1,2, …, m, m is the number of running states.
Figure BDA0002978884490000043
In the above formula, n represents the number of indices to be diagnosed, EiThe information entropy of the ith index to be diagnosed is obtained.
5.3) calculating the weight S of any ith index to be diagnosed according to the following formulai: thereby obtaining the weight S of n indicators to be diagnosediA constructed weight matrix S;
Figure BDA0002978884490000044
in the above formula, WiIs the subjective weight of the ith index to be diagnosed, HiIs the objective weight of the ith index to be diagnosed, i is 1,2, …, n is the index to be diagnosedThe number of the cells.
Optionally, in step 6), the function expression of the diagnosis matrix B obtained by performing fuzzy operation on the weight S and the fuzzy evaluation matrix a is as follows:
B=S×A=[b1,b2,b3,...,bm]
in the above formula, b1~bmThe operation states included in the diagnosis matrix B and m operation states are respectively, and selecting the final operation state for the diagnosis result in the diagnosis matrix B in step 7) specifically means selecting the operation state with the maximum membership value included in the diagnosis matrix B and m operation states as the operation state of the large phase modulator auxiliary system obtained by final evaluation.
In addition, the invention also provides an online diagnosis method for the running state of the auxiliary system of the large phase modulator, which comprises the following steps:
s1) inputting n original data of indexes to be diagnosed related to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system in the large phase modulator auxiliary system, wherein the indexes to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and pH value of water; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level;
s2) respectively carrying out normalization processing and principal component analysis to extract key feature vectors aiming at the original data of n indexes to be diagnosed, and combining the original data of the n indexes to be diagnosed to obtain comprehensive key feature vectors;
s3) inputting the comprehensive key feature vector into a pre-trained machine learning model to obtain the running state of the large phase modulator auxiliary system, wherein the machine learning model is pre-trained to establish a mapping relation between the comprehensive key feature vector and the running state of the large phase modulator auxiliary system.
In addition, the invention also provides an operation state online diagnosis system for a large phase modulation auxiliary system, which comprises a microprocessor and a memory which are connected with each other, wherein the microprocessor is programmed or configured to execute the steps of the operation state online diagnosis method for the large phase modulation auxiliary system.
Furthermore, the present invention also provides a computer-readable storage medium having stored therein a computer program programmed or configured to execute the method for online diagnosis of an operating state of a large phase modulator auxiliary system.
Compared with the prior art, the invention has the following advantages: the method adopts a fuzzy analytic hierarchy process combined with an entropy weight method to calculate the weight of each evaluation index and establish a weight matrix. Because the degradation tendency of each evaluation index is different along with the extension of the operation time, and the influence degree on the system operation is also different, the different influence degrees of each index on the evaluation result are described by adopting the combined weight so as to eliminate the uncertainty of the subjective factor. The method comprises the steps of firstly determining the subjective weight of each index by adopting an analytic hierarchy process, then determining the objective weight of each index by adopting an entropy weight process, finally combining the subjective weight and the objective weight to serve as the combined weight of the evaluation indexes, and finally giving an evaluation result by utilizing a fuzzy theory, so that the influence of subjective factors on the evaluation process can be reduced, the comprehensiveness and objectivity of the method are improved, and the engineering practicability of the method is improved by comprehensively considering an algorithm with multiple indexes.
Drawings
FIG. 1 is a schematic diagram of a basic flow of a method according to an embodiment of the present invention.
Fig. 2 is a schematic architectural diagram of n indicators to be diagnosed in the embodiment of the present invention.
Detailed Description
As shown in fig. 1, the method for online diagnosing the operating state of the auxiliary system of the large phase modulation machine in the embodiment includes:
1) inputting n original data of indexes to be diagnosed in a large phase modifier auxiliary system, wherein the n original data relate to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system;
2) respectively calculating the degradation degrees of the original data of n indexes to be diagnosed, and if the degradation degree of any index to be diagnosed is 1, judging that the auxiliary system of the large phase modulator fails and quits; otherwise, executing the next step;
3) respectively solving the membership degrees between n indexes to be diagnosed and m running states of the large phase modulator auxiliary system;
4) constructing a fuzzy judgment matrix A among n indexes to be diagnosed and m running states according to the membership degree;
5) respectively solving weight matrixes S of n to-be-diagnosed indexes;
6) carrying out fuzzy operation on the weight matrix S and the fuzzy evaluation matrix A to obtain a diagnosis matrix B, wherein the diagnosis matrix B comprises diagnosis results corresponding to m running states one by one;
7) and selecting a final operation state as the operation state of the large phase modulator auxiliary system obtained by final evaluation according to the diagnosis result in the diagnosis matrix B by adopting a maximum membership principle.
The oil system of the synchronous phase modifier is a self-circulation system, and a complete loop is formed by a lubricating oil module, an external pipeline and a phase modifier bearing bush. The oil system of the synchronous phase modifier mainly comprises a lubricating oil system and a high-pressure top shaft oil system. The lubricating oil system provides forced cooling lubricating oil for the phase modifier bearing, the lubricating oil is pumped from the main oil tank through the lubricating oil pump and then sent to the lubricating oil cooler to cool the lubricating oil, and the cooled oil is filtered through the filter to filter out micro particles in the oil and then is supplied to the bearing of the phase modifier. When the phase modifier is started and the rotor stalls, the top shaft oil pump absorbs lubricating oil from the lubricating oil main pipe, high-pressure top shaft oil is provided for the bearing bush after the lubricating oil passes through the filtering device, the rotor is forcibly jacked up, an oil film is formed in the bearing bush, friction between the bearing bush and the shaft is eliminated, and the bearing bush is prevented from being burnt. According to the structure of the synchronous phase modifier oil system, the synchronous phase modifier oil system is divided into a low-pressure lubricating oil system and a high-pressure top shaft oil system. In the operation of the phase modifier, the quality, the oil inlet pressure, the oil quantity and the oil temperature of lubricating oil are decisive factors influencing whether the bearing can normally run or not. The water system of the double-water internal cooling synchronous motor comprises an internal cooling water system, a desalting water system and an external cooling water system. The internal cooling water system comprises a stator cooling water system and a rotor cooling water system. The stator cooling water system is a closed self-circulation system. The cooling water in the stator water tank is pushed by the water pump to enter the heat exchanger, then flows into the stator coil and the outgoing line of the synchronous motor through the water filter, the water cut-off protection device and the like, takes away the heat generated by the stator winding, and flows back to the stator water tank to complete a water circulation. In order to ensure the quality of the stator cooling water, bypass cooling water is adopted for control. And one part of the cooling water after passing through the water filter enters the ion exchanger and the alkali adding device and then returns to the stator water tank, so that the conductivity and the pH value of the cooling water are controlled, and the corrosion of the cooling water on the copper wire is avoided. The rotor cooling water system is an open self-circulation system. The cooling water in the rotor water tank is pushed by a water pump to enter a heat exchanger, a water filter, a water break protection device and the like, flows into a central hole of a wire outlet end to enter a rotor water inlet chamber, then flows into a hollow lead in a rotor bar through an insulating water conduit under the action of centrifugal force, flows out from the other end (non-wire outlet end) of a coil through the insulating water conduit, takes away heat generated by a rotor winding, is thrown into a water collecting tank, and finally returns to the rotor water tank under the action of gravity. According to the composition of the cooling water system in the synchronous motor, in the operation of the phase modifier, the temperature, the flow and the pressure are main indexes for measuring the health state of the cooling water system. In the step 1) of the embodiment, the n indexes to be diagnosed related to the stator cooling water system, the rotor cooling water system, the lubricating oil system and the high-pressure jacking oil system are as follows: the indicators to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and water pH value; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level. In this embodiment, an evaluation index system is established for each subsystem, a physical quantity capable of reflecting the operating state of the subsystem is selected, and the established evaluation index system is shown in fig. 2 in combination with a sensor installed in the synchronous phase modulator.
In the method of the embodiment, each evaluation index belongs to different subsystems, has different physical meanings and dimensions, has different normal operation ranges, and needs to be preprocessed and converted into a [0,1] interval, namely normalization processing, for convenient evaluation. According to the data characteristics of the evaluation indexes, the evaluation indexes are divided into two types, namely a bipolar limit type, an upper limit type, a lower limit type and a rated value type, and different methods are adopted for de-dimensionalizing different types of indexes; wherein the bipolar limit type means: when the index value is in a certain interval, the indexes are normal; when the upper limit and the lower limit are exceeded, a fault is indicated; the upper limit value type index means that the index value cannot exceed a certain value, otherwise, the shutdown processing is required; the lower limit value type means that the index value cannot be lower than a certain value, otherwise, the alarm processing is carried out; the nominal value type means that the index data performs optimally in a nominal state, but allows operation within a certain range deviating from the nominal value.
In this embodiment, when the degradation degree is calculated for the raw data of n items of indicators to be diagnosed in step 2), the step of calculating the degradation degree for the raw data of any one of the items of indicators to be diagnosed includes:
2.1) judging the type of the index to be diagnosed, wherein the related types comprise a bipolar limit type, an upper limit type, a lower limit type and a rated value type, the bipolar limit type means that the index to be diagnosed fails when exceeding the upper limit and the lower limit of a normal interval, the upper limit type means that the index to be diagnosed fails when exceeding the normal upper limit, the lower limit type means that the index to be diagnosed fails when exceeding the normal lower limit, the rated value type means that the index to be diagnosed fails when exceeding the specified margin range of the normal rated value, and if the index to be diagnosed is the bipolar limit type or the rated value type, skipping to execute the step 2.2); if the type of the upper limit value is the upper limit value type, skipping to execute the step 2.3); if the type of the lower limit value is the lower limit value type, skipping to execute the step 2.4);
2.2) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000071
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xmax、xminRespectively are the upper limit value and the lower limit value of the index to be diagnosed, and alpha-beta are respectively the allowable value range of the index to be diagnosed;
2.3) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000081
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, alpha is the allowable value of the normal operation of the index, and xmaxIs the upper limit value of the index;
2.4) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure BDA0002978884490000082
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xminRespectively, the lower limit value of the index to be diagnosed.
In this embodiment, step 3) includes: the operation state change process of the large phase modulator auxiliary system is gradual, normal distribution is met, and a normal membership function is adopted to calculate any index x to be diagnosediAnd operating state sjThe membership degree between the index data and the index data is substituted into n items of index data to be diagnosed in a certain operation state, and the membership degree between the n items of index data to be diagnosed and m operation states of the large phase modifier auxiliary system is analyzed from the n items of index data to be diagnosed, so that any index x to be diagnosed is obtainediAnd operating state sjThe degree of membership between is:
Figure BDA0002978884490000083
in the above formula, AijFor the index x to be diagnosediAnd operating state sjDegree of membership between, cjIs in an operating state sjCenter of normal distribution, σ is operating state sjThe width of the normal distribution. In this example, c is takenj={0,0.33,0.66,1},σ=0.1。
In this embodiment, the function expression of the fuzzy judgment matrix a between the n indexes to be diagnosed and the m operating states, which is constructed according to the membership in step 4), is:
Figure BDA0002978884490000084
in the above formula, any AijFor the index x to be diagnosediAnd operating state sjThe membership degree between the two groups is 1,2, …, n, n is the number of indexes to be diagnosed, and j is 1,2, …, m, m is the number of running states.
In this embodiment, the function expressions of the weights S of the n to-be-diagnosed indexes obtained in step 5) are respectively:
5.1) aiming at n indexes to be diagnosed, combining expert experience to carry out pairwise comparison according to the importance of each index to set a scale value to form a series of judgment matrixes, and calculating and judging the maximum characteristic value lambda of each judgment matrixmaxAnd corresponding eigenvectors thereof, and finally determining the subjective weight W of any ith index to be diagnosed according to the judgment matrixi
5.2) respectively solving the information entropy according to the following formula aiming at the sample data matrix formed by the original data of n indexes to be diagnosed, and calculating the objective weight H of any ith index to be diagnosed by adopting the following formulai
Figure BDA0002978884490000091
In the above formula, pijThe probability value Y of the ith index data in the jth operation state interval is obtained for statisticsijFor n acquired by the condition monitoring platformThe original data of the indexes, i is 1,2, …, n, n is the number of the indexes to be diagnosed, j is 1,2, …, m, m is the number of the running states.
Figure BDA0002978884490000092
In the above formula, n represents the number of indices to be diagnosed, EiThe information entropy of the ith index to be diagnosed is obtained.
5.3) calculating the weight S of any ith index to be diagnosed according to the following formulaiThereby obtaining the weight S of n indicators to be diagnosediA constructed weight matrix S;
Figure BDA0002978884490000093
in the above formula, WiIs the subjective weight of the ith index to be diagnosed, HiThe objective weight of the ith index to be diagnosed is, i is 1,2, …, n, n is the number of indexes to be diagnosed.
In this embodiment, in step 6), the function expression of the diagnostic matrix B obtained by performing fuzzy operation on the weight S and the fuzzy evaluation matrix a is as follows:
B=S×A=[b1,b2,b3,...,bm] (9)
in the above formula, b1~bmThe operation states included in the diagnosis matrix B and m operation states are respectively, and selecting the final operation state for the diagnosis result in the diagnosis matrix B in step 7) specifically means selecting the operation state with the maximum membership value included in the diagnosis matrix B and m operation states as the operation state of the large phase modulator auxiliary system obtained by final evaluation.
The operation state of the auxiliary system of the large phase modulator is a gradual degradation process with ambiguity, and in order to facilitate the application of the evaluation result to the operation and maintenance of the system, the gradual degradation process is described by adopting a qualitative method. The method of the embodiment divides the state grade S of the phase modulator auxiliary system into four grades of good, normal, early warning and fault, which are S respectively1,S2,S3,S4I.e. expressed as: s ═ S1,S2,S3,S40,0.33,0.66,1, the definitions of which are shown in table 1.
Table 1: operation state grading and meaning of large phase modulator auxiliary system
Status rating Means of
Good effect Can be continuously operated for a long time
Qualified The operation is in a normal range, and the system operation is not influenced
Early warning The system can run for a short time due to slight fault, and needs to be scheduled for maintenance
Fault of When the system is in failure, the system should be stopped for maintenance
Because the running state of the large phase modulator auxiliary system has ambiguity, the method combines the fuzzy theory to realize the health state evaluation of the phase modulator auxiliary system. When the weight of each evaluation index is calculated, firstly, the subjective weight of each evaluation index is calculated by adopting an analytic hierarchy process, then, the objective weight is calculated by adopting an entropy weight method, and the combined weight obtained by combining the subjective weight and the objective weight is used as the final weight value of each index.
Firstly, combining expert experience and according to each indexComparing the importance with each other to set scale values to form a series of judgment matrixes, and calculating and judging the maximum eigenvalue lambda of each judgment matrixmaxAnd corresponding eigenvectors thereof, and finally determining the weight of each index according to the judgment matrix. In the evaluation system proposed by the method of the present embodiment, the weights of the evaluation indexes are obtained by combining expert experience analysis.
Under the nine-scale method, for the evaluation targets with n evaluation indexes, a comparison matrix can be formed by pairwise comparison.
For an oil system, the importance degree of a lubricating oil system and a jacking oil system with two indexes is B3:B4The comparison matrix formed after comparing the importance degrees of all indexes of the lubricating oil system is P (0.6: 0.4)21(ii) a The comparison matrix formed after the importance degrees of all indexes of the jacking oil system are compared is P22
Figure BDA0002978884490000101
Figure BDA0002978884490000102
For the water system, the importance degree of the stator cooling water system and the rotor cooling water system is B1:B2The comparison matrix formed after comparing the importance degrees of various indexes of the stator cooling water system is P (0.6: 0.4)11(ii) a The comparison matrix formed after the importance degrees of all indexes of the rotor cooling water system are compared is P12
Figure BDA0002978884490000111
Figure BDA0002978884490000112
Obtaining objective weight H by entropy weight methodjThe entropy method uses heat as referenceThe entropy concept in mechanics describes the size of information quantity in an event, and the dispersion degree of a certain index can be judged in evaluation, namely, the larger the entropy value is, the larger the dispersion degree is, the smaller the information quantity is, and the smaller the weight value is. According to the definition, the information entropy is solved for the sample data matrix, and the following can be obtained:
Figure BDA0002978884490000113
Figure BDA0002978884490000114
wherein if p isijWhen equal to 0, then ∑ pijlnpij=0。
Information entropy E according to each indexj(j ═ 1,2, …, k) the index weight can be found as:
Figure BDA0002978884490000115
3) calculating combining weights
And (3) solving a final weight value by combining an entropy weight method and an analytic hierarchy process as follows:
Figure BDA0002978884490000121
4) obtaining the evaluation results
And selecting a proper fuzzy operator, carrying out fuzzy operation on the weight S and the fuzzy evaluation matrix A of each evaluation target, and then adopting a maximum membership principle to obtain an evaluation result.
B=S×A=[b1,b2,b3,b4] (14)
Take bmax as max (b)j) J is 1,2,3,4 as the final evaluation result.
Taking a synchronous phase modulator auxiliary system of a certain converter station as an example, the implementation process and the evaluation effect of the evaluation method are described.
Firstly, collecting the operation data of the indexes in the month 1 of 2019, wherein the month system operates normally, selecting the data in the time period as the original data of state evaluation, and evaluating according to an evaluation flow. The information entropy and the objective weight of each index are obtained according to the normalized original data, the combination weight is calculated by combining the subjective weight, and the subjective weight and the objective weight of each index are obtained and are shown in tables 2-4.
Table 2: weighted value of each evaluation index under hierarchical analysis method
Figure BDA0002978884490000122
Table 3: weighted value of each evaluation index under entropy weight method
Figure BDA0002978884490000123
Table 4: combined weight of each evaluation index
Figure BDA0002978884490000124
Figure BDA0002978884490000131
In tables 2 to 4, the symbol definitions of the evaluation indexes are shown in FIG. 2. And performing fuzzy operation on the weight S and the fuzzy evaluation matrix A of each evaluation target to obtain B which is [ 0.28450.08660.40540.0478 ], taking bmax which is 0.4054, and classifying according to the state grades, wherein the health state is qualified, namely, the operation is in a normal range.
In conclusion, the method solves the problem of real-time evaluation of the health state of auxiliary equipment (including a water system and an oil system) of the large synchronous phase modifier. The method mainly comprises the steps of establishing a state evaluation index system, preprocessing index data, dividing state grades and determining index weight. The method adopts an analytic hierarchy process, determines subjective weight through expert experience, determines objective weight through an entropy weight method, and combines a fuzzy rule to realize comprehensive evaluation of the health state of the phase modulator auxiliary system. The method can reduce the influence of subjective factors on the evaluation process, improve the comprehensiveness and objectivity of the method, and improve the engineering practicability of the method by comprehensively considering the multi-index algorithm.
In addition, the embodiment also provides an online diagnosis method for the operating state of the auxiliary system of the large phase modulator, which comprises the following steps:
s1) inputting n original data of indexes to be diagnosed related to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system in the large phase modulator auxiliary system, wherein the indexes to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and pH value of water; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level;
s2) respectively carrying out normalization processing and principal component analysis to extract key feature vectors aiming at the original data of n indexes to be diagnosed, and combining the original data of the n indexes to be diagnosed to obtain comprehensive key feature vectors;
s3) inputting the comprehensive key feature vector into a pre-trained machine learning model to obtain the running state of the large phase modulator auxiliary system, wherein the machine learning model is pre-trained to establish a mapping relation between the comprehensive key feature vector and the running state of the large phase modulator auxiliary system.
It should be noted that, extracting the key feature vector by principal component analysis is an existing analysis method, and this embodiment is only an application of the principal component analysis method and relates to an improvement on the principal component analysis method, and therefore details are not described herein again. In addition, the machine learning model training is utilized to establish the mapping relation between the comprehensive key feature vector and the operation state of the large phase modulation auxiliary system, so that the operation state of the large phase modulation auxiliary system obtained by inputting the comprehensive key feature vector into the pre-trained machine learning model is also the basic application of the machine learning model, the key is to utilize the mapping classification between the original data of n indexes to be diagnosed and the operation state of the large phase modulation auxiliary system, and the specific realization of the machine learning model can adopt various existing machine learning models such as a BP network and various deep learning networks according to the needs, so the specific realization details of the machine learning model are not repeated.
In addition, the embodiment also provides an operation state online diagnosis system for a large phase modulation auxiliary system, which comprises a microprocessor and a memory which are connected with each other, wherein the microprocessor is programmed or configured to execute the steps of the operation state online diagnosis method for the large phase modulation auxiliary system.
Furthermore, the present embodiment also provides a computer-readable storage medium having stored therein a computer program programmed or configured to execute the aforementioned method for online diagnosis of an operating state of a large-scale phase modulation auxiliary system.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The present application is directed to methods, apparatus (systems), and computer program products according to embodiments of the application wherein instructions, which execute via a flowchart and/or a processor of the computer program product, create means for implementing functions specified in the flowchart and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.

Claims (10)

1. An online diagnosis method for the operation state of a large-scale phase modifier auxiliary system is characterized by comprising the following steps:
1) inputting n original data of indexes to be diagnosed in a large phase modifier auxiliary system, wherein the n original data relate to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system;
2) respectively calculating the degradation degrees of the original data of n indexes to be diagnosed, and if the degradation degree of any index to be diagnosed is 1, judging that the auxiliary system of the large phase modulator fails and quits; otherwise, executing the next step;
3) respectively solving the membership degrees between n indexes to be diagnosed and m running states of the large phase modulator auxiliary system;
4) constructing a fuzzy judgment matrix A among n indexes to be diagnosed and m running states according to the membership degree;
5) respectively solving weight matrixes S of n to-be-diagnosed indexes;
6) carrying out fuzzy operation on the weight matrix S and the fuzzy evaluation matrix A to obtain a diagnosis matrix B, wherein the diagnosis matrix B comprises diagnosis results corresponding to m running states one by one;
7) and selecting a final operation state as the operation state of the large phase modulator auxiliary system obtained by final evaluation according to the diagnosis result in the diagnosis matrix B by adopting a maximum membership principle.
2. The on-line diagnosis method for the operation state of the large phase modifier auxiliary system according to claim 1, wherein the n indexes to be diagnosed related to the stator cooling water system, the rotor cooling water system, the lubricating oil system and the high-pressure top shaft oil system in the step 1) comprise: the indicators to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and water pH value; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level.
3. The on-line diagnosis method for the operation state of the large-scale phase modulation auxiliary system according to claim 1, wherein when the degradation degree is calculated for the raw data of n items of the index to be diagnosed in step 2), the step of calculating the degradation degree for the raw data of any one item of the index to be diagnosed comprises:
2.1) judging the type of the index to be diagnosed, wherein the related types comprise a bipolar limit type, an upper limit type, a lower limit type and a rated value type, the bipolar limit type means that the index to be diagnosed fails when exceeding the upper limit and the lower limit of a normal interval, the upper limit type means that the index to be diagnosed fails when exceeding the normal upper limit, the lower limit type means that the index to be diagnosed fails when exceeding the normal lower limit, the rated value type means that the index to be diagnosed fails when exceeding the specified margin range of the normal rated value, and if the index to be diagnosed is the bipolar limit type or the rated value type, skipping to execute the step 2.2); if the type of the upper limit value is the upper limit value type, skipping to execute the step 2.3); if the type of the lower limit value is the lower limit value type, skipping to execute the step 2.4);
2.2) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure FDA0002978884480000021
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xmax、xminRespectively are the upper limit value and the lower limit value of the index to be diagnosed, and alpha-beta are respectively the allowable value range of the index to be diagnosed;
2.3) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure FDA0002978884480000022
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, alpha is the allowable value of the normal operation of the index, and xmaxIs the upper limit value of the index;
2.4) calculating the deterioration degree according to the following formula, and skipping to execute the step 3);
Figure FDA0002978884480000023
in the above formula, g (x) is the deterioration degree, x is the original data of the index to be diagnosed, xminRespectively, the lower limit value of the index to be diagnosed.
4. The on-line diagnosis method for the operation state of the large-scale phase modulation auxiliary system according to claim 1, wherein the step 3) comprises: because the change process of the running state of the auxiliary system of the large phase modulator is gradual, normal distribution is satisfied,calculating any index x to be diagnosed by adopting normal membership functioniAnd operating state sjThe membership degree between the index data and the index data is substituted into n items of index data to be diagnosed in a certain operation state, and the membership degree between the n items of index data to be diagnosed and m operation states of the large phase modifier auxiliary system is analyzed from the n items of index data to be diagnosed, so that any index x to be diagnosed is obtainediAnd operating state sjThe degree of membership between is:
Figure FDA0002978884480000024
in the above formula, AijFor the index x to be diagnosediAnd operating state sjDegree of membership between, cjIs in an operating state sjCenter of normal distribution, σ is operating state sjThe width of the normal distribution.
5. The on-line diagnosis method for the operating state of the large phase modulation machine auxiliary system according to claim 1, wherein the functional expression of the fuzzy judgment matrix A between the n indexes to be diagnosed and the m operating states constructed according to the membership in the step 4) is as follows:
Figure FDA0002978884480000031
in the above formula, any AijFor the index x to be diagnosediAnd operating state sjThe membership degree between the two groups is 1,2, …, n, n is the number of indexes to be diagnosed, and j is 1,2, …, m, m is the number of running states.
6. The on-line diagnosis method for the operation state of the large phase modulator auxiliary system according to claim 1, wherein the function expression of the weight S of n items of the index to be diagnosed in step 5) is respectively obtained as:
5.1) aiming at n indexes to be diagnosed, pairwise comparison is carried out according to the importance of each index by combining with expert experience to set scalesForming a series of judgment matrixes, and calculating the maximum eigenvalue lambda of each judgment matrixmaxAnd corresponding eigenvectors thereof, and finally determining the subjective weight W of any ith index to be diagnosed according to the judgment matrixi
5.2) respectively solving the information entropy according to the following formula aiming at the sample data matrix formed by the original data of n indexes to be diagnosed, and calculating the objective weight H of any ith index to be diagnosed by adopting the following formulai
Figure FDA0002978884480000032
In the above formula, pijThe probability value Y of the ith index data in the jth operation state interval is obtained for statisticsijThe raw data of n indexes obtained by the state monitoring platform are i equal to 1,2, …, n, n is the number of indexes to be diagnosed, j equal to 1,2, …, m, m is the number of running states.
Figure FDA0002978884480000033
In the above formula, n represents the number of indices to be diagnosed, EiThe information entropy of the ith index to be diagnosed is obtained.
5.3) calculating the weight S of any ith index to be diagnosed according to the following formulai: thereby obtaining the weight S of n indicators to be diagnosediA constructed weight matrix S;
Figure FDA0002978884480000034
in the above formula, WiIs the subjective weight of the ith index to be diagnosed, HiThe objective weight of the ith index to be diagnosed is, i is 1,2, …, n, n is the number of indexes to be diagnosed.
7. The on-line diagnosis method for the operation state of the large phase modulation auxiliary system according to claim 1, wherein the function expression of the diagnosis matrix B obtained by fuzzy operation of the weight S and the fuzzy evaluation matrix A in step 6) is as follows:
B=S×A=[b1,b2,b3,...,bm]
in the above formula, b1~bmThe operation states included in the diagnosis matrix B and m operation states are respectively, and selecting the final operation state for the diagnosis result in the diagnosis matrix B in step 7) specifically means selecting the operation state with the maximum membership value included in the diagnosis matrix B and m operation states as the operation state of the large phase modulator auxiliary system obtained by final evaluation.
8. An online diagnosis method for the operation state of a large-scale phase modifier auxiliary system is characterized by comprising the following steps:
s1) inputting n original data of indexes to be diagnosed related to a stator cooling water system, a rotor cooling water system, a lubricating oil system and a high-pressure top shaft oil system in the large phase modulator auxiliary system, wherein the indexes to be diagnosed of the stator cooling water system comprise water inlet pressure, water inlet temperature, inlet and outlet pressure difference, water outlet temperature, flow, liquid level, water inlet conductivity and pH value of water; the indicators to be diagnosed of the rotor cooling water system comprise water inlet pressure, water inlet temperature, flow, liquid level, inlet and outlet pressure difference and water outlet temperature; the index to be diagnosed of the lubricating oil system comprises an inlet oil temperature, an outlet oil temperature, an oil inlet pressure and an oil tank liquid level; the index to be diagnosed of the high-pressure top shaft oil system comprises inlet oil temperature, outlet oil temperature, top shaft oil outlet main pipe pressure and oil tank liquid level;
s2) respectively carrying out normalization processing and principal component analysis to extract key feature vectors aiming at the original data of n indexes to be diagnosed, and combining the original data of the n indexes to be diagnosed to obtain comprehensive key feature vectors;
s3) inputting the comprehensive key feature vector into a pre-trained machine learning model to obtain the running state of the large phase modulator auxiliary system, wherein the machine learning model is pre-trained to establish a mapping relation between the comprehensive key feature vector and the running state of the large phase modulator auxiliary system.
9. An on-line diagnostic system for the operating state of a large-scale phase modulation auxiliary system, which comprises a microprocessor and a memory which are connected with each other, and is characterized in that the microprocessor is programmed or configured to execute the steps of the on-line diagnostic method for the operating state of the large-scale phase modulation auxiliary system according to any one of claims 1-8.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored therein a computer program programmed or configured to execute the method for online diagnosis of the operating state of a large-scale phase modulation auxiliary system according to any one of claims 1 to 8.
CN202110281437.7A 2021-03-16 2021-03-16 Operation state online diagnosis method and system for auxiliary system of large phase modulator Pending CN113128832A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110281437.7A CN113128832A (en) 2021-03-16 2021-03-16 Operation state online diagnosis method and system for auxiliary system of large phase modulator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110281437.7A CN113128832A (en) 2021-03-16 2021-03-16 Operation state online diagnosis method and system for auxiliary system of large phase modulator

Publications (1)

Publication Number Publication Date
CN113128832A true CN113128832A (en) 2021-07-16

Family

ID=76773302

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110281437.7A Pending CN113128832A (en) 2021-03-16 2021-03-16 Operation state online diagnosis method and system for auxiliary system of large phase modulator

Country Status (1)

Country Link
CN (1) CN113128832A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108536945A (en) * 2018-04-02 2018-09-14 国网湖南省电力有限公司 A kind of fault diagnosis method and system for large-scale phase modifier
CN109872077A (en) * 2019-03-08 2019-06-11 国网山东省电力公司电力科学研究院 The method for evaluating state and device of double-water internal cooling synchronous capacitor cooling water system
CN110428151A (en) * 2019-07-18 2019-11-08 国网山东省电力公司电力科学研究院 A kind of method for evaluating state of synchronous capacitor oil system
CN110907732A (en) * 2019-12-04 2020-03-24 江苏方天电力技术有限公司 Phase modulator fault diagnosis method based on PCA-RBF neural network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108536945A (en) * 2018-04-02 2018-09-14 国网湖南省电力有限公司 A kind of fault diagnosis method and system for large-scale phase modifier
CN109872077A (en) * 2019-03-08 2019-06-11 国网山东省电力公司电力科学研究院 The method for evaluating state and device of double-water internal cooling synchronous capacitor cooling water system
CN110428151A (en) * 2019-07-18 2019-11-08 国网山东省电力公司电力科学研究院 A kind of method for evaluating state of synchronous capacitor oil system
CN110907732A (en) * 2019-12-04 2020-03-24 江苏方天电力技术有限公司 Phase modulator fault diagnosis method based on PCA-RBF neural network

Similar Documents

Publication Publication Date Title
CN109685340B (en) Power distribution equipment health state assessment method and system
CN105117602B (en) A kind of metering device running status method for early warning
CN113792762A (en) Water chilling unit fault diagnosis method, system and medium based on Bayesian optimization LightGBM
CN111648992B (en) Gas turbine compressor fault identification early warning method
CN107274067B (en) Distribution transformer overload risk assessment method
CN107728059B (en) Pitch system state evaluation method
CN110674120A (en) Wind power plant data cleaning method and device
CN112179655A (en) Turbo generator fault early warning method based on threshold classification
CN108304931A (en) A kind of Condition-based Maintenance of Substation Equipment method for diagnosing faults
CN109872077B (en) Method and device for evaluating state of cooling water system of double-water internal cooling synchronous phase modifier
CN112990627B (en) Power quality evaluation method
CN109188227A (en) Double-fed wind driven generator insulation state evaluation method and system
CN112613554A (en) Fault prediction method and system for variable pitch system of wind driven generator
CN107169655A (en) A kind of method and device of preferred power distribution network project compatibility
Zhu et al. Operational state assessment of wind turbine gearbox based on long short-term memory networks and fuzzy synthesis
CN111563693B (en) Scoring method, scoring equipment and scoring storage medium for health value of rail transit equipment
CN116928038A (en) Fault early warning method and device based on main bearing temperature of offshore wind turbine generator
CN110033181B (en) Power generation equipment state evaluation method based on self-encoder
CN114169718A (en) Method for improving reliability of wind turbine generator based on state evaluation of wind turbine generator
CN115730228A (en) Central air conditioner energy consumption analysis method based on BP neural network
CN115204658A (en) Method and system for assessing health state of oil supply equipment of hydroelectric generating set
CN113722656B (en) Real-time health evaluation method and system for thermal generator set
CN105426665B (en) Method is determined based on the DYNAMIC RELIABILITY of status monitoring
CN110702438B (en) Centrifugal water chilling unit and fresh air system combined fault diagnosis method
CN113128832A (en) Operation state online diagnosis method and system for auxiliary system of large phase modulator

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210716