CN-121980254-A - Extra-high voltage converter valve overhaul scheme recommendation method, system, equipment and medium

CN121980254ACN 121980254 ACN121980254 ACN 121980254ACN-121980254-A

Abstract

The invention provides a recommendation method, a system, equipment and a medium for an extra-high voltage converter valve maintenance scheme, which relate to the field of extra-high voltage transmission, and are characterized by acquiring historical maintenance data and real-time monitoring data of a plurality of extra-high voltage converter valves in a power grid, calculating a comprehensive health index of each extra-high voltage converter valve by using a fuzzy analytic hierarchy process based on the historical maintenance data and the real-time monitoring data of the plurality of extra-high voltage converter valves, carrying out maintenance priority ranking on each extra-high voltage converter valve based on the comprehensive health index of each extra-high voltage converter valve combined with the power grid operation requirement to obtain a priority ranking result, and generating a maintenance scheme by using a maintenance scheme recommendation model based on the comprehensive health index of each extra-high voltage converter valve and the priority ranking result. The method improves the scientificity and the effectiveness of maintenance, ensures the safe and stable operation of the extra-high voltage converter valve, and reduces the maintenance cost.

Inventors

LI TIANQI
ZHAO XINSHUANG
REN MENGGAN
YANG HUAFEI
CHAI BIN
LUO JIAN
WU BAOYI
CHEN PENG
XU DA
CUI XUESONG

Assignees

中电普瑞电力工程有限公司

Dates

Publication Date: 20260505
Application Date: 20251203

Claims (14)

1. The method for recommending the extra-high voltage converter valve overhaul scheme is characterized by comprising the following steps of: Acquiring historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves in a power grid; Based on historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves, calculating the comprehensive health index of each extra-high voltage converter valve by using a fuzzy analytic hierarchy process; Based on the comprehensive health index of each extra-high voltage converter valve and the operation requirement of a power grid, carrying out maintenance priority sequencing on each extra-high voltage converter valve to obtain a priority sequencing result; And generating an overhaul scheme by using an overhaul scheme recommendation model based on the comprehensive health index of each extra-high voltage converter valve and the priority ranking result.
2. The method of claim 1, wherein calculating the integrated health index for each of the extra-high voltage converter valves using fuzzy analytic hierarchy process based on historical service data and real-time monitoring data for the plurality of extra-high voltage converter valves comprises: respectively extracting characteristics of historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves to obtain a historical health baseline index and a real-time abnormality index of each extra-high voltage converter valve; Based on each extra-high voltage converter valve, carrying out index fusion on the historical healthy baseline index and the real-time abnormal index of the extra-high voltage converter valve by using a fuzzy analytic hierarchy process to obtain the index of the historical healthy baseline index and the index of the real-time abnormal index; and obtaining the comprehensive health index of the extra-high voltage converter valve based on the historical health baseline index and the index thereof and the real-time abnormal index and the index thereof.
3. The method according to claim 2, wherein the performing index fusion on the historical healthy baseline index and the real-time abnormality index of the extra-high voltage converter valve by using a fuzzy analytic hierarchy process to obtain the index of the historical healthy baseline index and the index of the real-time abnormality index includes: constructing a three-level hierarchical structure with definite membership among various levels according to the historical health baseline index and the real-time abnormality index of the extra-high voltage converter valve and the health evaluation target of the converter valve, wherein the three-level hierarchical structure comprises a target layer, a criterion layer and an index layer, the data of the target layer is a comprehensive health evaluation target, the data of the criterion layer is the historical health baseline index and the real-time abnormality index, and the data of the index layer is specific characteristics of the historical health baseline index and the real-time abnormality index; Comparing importance of each level of data in the three-level hierarchical structure in pairs based on a scale method to construct a fuzzy judgment matrix; consistency test of judgment rationality is carried out on the fuzzy judgment matrix, and weights of data of all levels are calculated by adopting a characteristic value method, so that weights of the historical healthy baseline indexes and weights of the real-time abnormal indexes are obtained; Dynamically adjusting the weight of the historical healthy baseline index and the weight of the real-time abnormality index according to the running years of the power grid to obtain the index of the historical healthy baseline index and the index of the real-time abnormality index.
4. The method according to claim 2, wherein the step of respectively extracting the historical overhaul data and the real-time monitoring data of the plurality of extra-high voltage converter valves to obtain a historical health baseline index and a real-time abnormality index of each of the extra-high voltage converter valves includes: The method comprises the steps of carrying out abnormal value elimination and missing value processing on historical maintenance data of a plurality of extra-high voltage converter valves to obtain processed historical maintenance data, inputting the processed historical maintenance data into a pre-constructed historical health baseline characteristic extraction model to carry out performance attenuation characteristic extraction to obtain a historical health baseline index of each extra-high voltage converter valve, wherein the historical health baseline index comprises an accumulated loss coefficient, a fault recurrence probability and a residual life prediction value; Carrying out Kalman filtering denoising treatment on the real-time monitoring data of the plurality of extra-high voltage converter valves to obtain treated real-time data, and converting the treated real-time data into standardized real-time abnormality indexes, wherein the real-time abnormality indexes comprise instantaneous overvoltage frequency, local discharge capacity exceeding time length, cooling system flow deviation rate and IGBT junction temperature exceeding frequency.
5. The method of claim 4, wherein the historical health baseline index has a functional expression of: Wherein, HBI is a historical health baseline index, n is a historical feature dimension; Attenuation coefficients of the i-th historical feature in the historical overhaul data; A normalized value for the i-th historical feature; The function expression of the real-time abnormality index is as follows: wherein RAI is real-time abnormality index, m is real-time monitoring characteristic dimension; Abnormal weight of the j-th real-time feature in the real-time monitoring data; normalized values for the degree of abnormality of the jth real-time feature.
6. The method of claim 2, wherein the functional expression of the integrated health index is: Wherein CHI is the comprehensive health index; Is the weight of the historical healthy baseline index, HBI is the historical healthy baseline index, and RAI is the real-time abnormality index.
7. The method of claim 1, wherein the prioritizing the maintenance of each of the extra-high voltage converter valves based on the integrated health index of each of the extra-high voltage converter valves in combination with the grid operation requirement to obtain a prioritized result comprises: Obtaining an optimization target value of each extra-high voltage converter valve according to an optimization target defined by power grid operation requirements, wherein the optimization targets comprise a safety target, an economical target and a feasibility target; Based on each extra-high voltage converter valve, weighting and summing according to the optimization target value of the extra-high voltage converter valve through the multi-target optimization function to obtain a comprehensive optimization target value of the extra-high voltage converter valve; And carrying out priority maintenance sequencing on each extra-high voltage converter valve according to each comprehensive health index and the comprehensive optimization target value to obtain a priority sequencing result.
8. The method of claim 7, wherein prioritizing each of the uhv converter valves according to each of the integrated health index and the integrated optimization target value, after obtaining a prioritization result, further comprises: And performing scene adjustment on the overhaul priority list according to the power grid cross-region power transmission plan and the load peak-valley period to generate a priority ordering result applicable to the scene change of the power grid.
9. The method of claim 1, wherein generating a service plan using a service plan recommendation model based on the integrated health index and the prioritization results for each of the extra-high voltage converter valves comprises: constructing an overhaul scheme knowledge base; inputting the comprehensive health index of each extra-high voltage converter valve into a pre-trained overhaul scheme recommendation model according to the sequence of the priority ordering result, and generating an overhaul scheme by combining power grid load rate grade data and spare part inventory grade data; When the comprehensive health index is greater than or equal to a first preset value, the corresponding overhaul scheme is of a conventional maintenance type; when the comprehensive health index is larger than or equal to a second preset value and smaller than the first preset value, the corresponding overhaul scheme is an overhaul scheme of a special overhaul type; When the comprehensive health index is smaller than a second preset value, the corresponding overhaul scheme is an urgent repair type overhaul scheme.
10. The method of claim 9, wherein the training process of the overhaul plan recommendation model comprises: Acquiring historical maintenance data of a converter valve, wherein the historical maintenance data of the converter valve comprises a historical comprehensive health index grade, a power grid load rate grade and a spare part inventory grade of the converter valve; The historical comprehensive health index grade, the power grid load rate grade and the spare part inventory grade of the converter valve are used as state space parameters, and a strategy gradient algorithm is adopted to calculate and obtain action space parameters of an intelligent agent according to the state space parameters, wherein the action space parameters comprise conventional maintenance and overhaul scheme actions, special overhaul scheme actions and emergency repair scheme actions; And calculating a reward value through a reward function according to the state space parameter and the action space parameter, and performing iterative training on the intelligent body based on the reward value to obtain a maintenance scheme recommendation model.
11. The method of claim 1, wherein after generating a service plan using a service plan recommendation model based on the integrated health index and the prioritization result for each of the extra-high voltage converter valves, further comprising: the method comprises the steps of collecting effect data after the maintenance scheme is implemented, wherein the effect data comprises the failure removal rate, the cost deviation rate and the operation stability of the extra-high voltage converter valve; and reversely correcting parameters extracted by the features, weight coefficients of the fuzzy analytic hierarchy process and parameters of the overhaul scheme recommended model according to the effect data to form a self-optimization closed loop.
12. An extra-high voltage converter valve overhaul scheme recommendation system, comprising: the data acquisition module is used for acquiring historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves in the power grid; The health degree evaluation module is used for calculating the comprehensive health degree index of each extra-high voltage converter valve by using a fuzzy analytic hierarchy process based on the historical overhaul data and the real-time monitoring data of the plurality of extra-high voltage converter valves; the priority ranking module is used for ranking the overhaul priorities of the extra-high voltage converter valves based on the comprehensive health index of each extra-high voltage converter valve and the power grid operation requirement to obtain a priority ranking result; And the overhaul decision module is used for generating an overhaul scheme by using an overhaul scheme recommendation model based on the comprehensive health index of each extra-high voltage converter valve and the priority ranking result.
13. A computer device, comprising: one or more processors; The processor is used for storing one or more programs; When the one or more programs are executed by the one or more processors, an extra-high voltage converter valve service scheme recommendation method according to any one of claims 1 to 11 is implemented.
14. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when executed, implements a method for recommending an extra-high voltage converter valve service solution according to any one of claims 1 to 11.

Description

Extra-high voltage converter valve overhaul scheme recommendation method, system, equipment and medium Technical Field The invention relates to the field of extra-high voltage transmission, in particular to a method, a system, equipment and a medium for recommending an extra-high voltage converter valve overhaul scheme. Background The extra-high voltage converter valve is a key device in an extra-high voltage direct current transmission system, and the running state of the extra-high voltage converter valve is directly related to the stability and reliability of the whole transmission system. With the continuous expansion of the ultra-high voltage network scale, the operation environment of the converter valve is increasingly complex, the traditional overhaul scheme either only depends on historical overhaul records and fault data, ignores the abrupt change of the real-time operation state of equipment (such as instantaneous overvoltage and partial discharge amount surge), easily causes 'overhaul lag' to cause equipment fault, or only depends on static real-time monitoring data, does not consider the long-term performance attenuation trend (such as accumulated loss and ageing degree) of the equipment, and possibly misjudges occasional abnormal signals to cause 'excessive overhaul' to increase cost. In addition, the traditional method adopts a fixed threshold value to divide the health state of the equipment, and the evaluation weight cannot be dynamically adjusted according to the running years of the equipment, the load fluctuation of the power grid and the change of the environmental conditions. The traditional method uniformly adopts fixed weight, and the deviation between the evaluation result and the actual health state is larger. Meanwhile, the traditional overhaul scheme is formulated by depending on expert experience, a standardized knowledge base and an intelligent model support are lacked, and different experts may have differences in judging the same equipment state, so that scheme suitability is uneven. Meanwhile, the economical efficiency (such as maintenance cost and power outage loss) and feasibility (such as spare part inventory and resource scheduling) of the power grid are not fully considered, and the 'safe-economical-feasible' multi-objective balance is difficult to achieve. In summary, the conventional extra-high voltage converter valve maintenance scheme in the prior art has the problems of dependence on single historical data or static real-time data, insufficient dynamic adaptability and difficulty in balancing the safety and the economy of the power grid. Disclosure of Invention The method aims to solve the problems that the traditional extra-high voltage converter valve overhaul scheme in the prior art depends on single historical data or static real-time data, has insufficient dynamic adaptability and is difficult to balance the safety and the economy of a power grid. In a first aspect, the invention provides a method for recommending an extra-high voltage converter valve overhaul scheme, which comprises the following steps: Acquiring historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves in a power grid; Based on historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves, calculating the comprehensive health index of each extra-high voltage converter valve by using a fuzzy analytic hierarchy process; Based on the comprehensive health index of each extra-high voltage converter valve and the operation requirement of a power grid, carrying out maintenance priority sequencing on each extra-high voltage converter valve to obtain a priority sequencing result; And generating an overhaul scheme by using an overhaul scheme recommendation model based on the comprehensive health index of each extra-high voltage converter valve and the priority ranking result. Preferably, the calculating, based on the historical overhaul data and the real-time monitoring data of the plurality of extra-high voltage converter valves, the comprehensive health index of each extra-high voltage converter valve by using a fuzzy analytic hierarchy process includes: respectively extracting characteristics of historical overhaul data and real-time monitoring data of a plurality of extra-high voltage converter valves to obtain a historical health baseline index and a real-time abnormality index of each extra-high voltage converter valve; Based on each extra-high voltage converter valve, carrying out index fusion on the historical healthy baseline index and the real-time abnormal index of the extra-high voltage converter valve by using a fuzzy analytic hierarchy process to obtain the index of the historical healthy baseline index and the index of the real-time abnormal index; and obtaining the comprehensive health index of the extra-high voltage converter valve based on the historical health baseline index and the ind