CN-122016618-A - Monitoring and evaluating method for rust condition of ball valve

CN122016618ACN 122016618 ACN122016618 ACN 122016618ACN-122016618-A

Abstract

The invention discloses a monitoring and evaluating method for the corrosion condition of a ball valve, which comprises the following steps of obtaining data parameters of the ball valve to be detected, obtaining sampling time points of each monitoring period according to sampling frequency coefficients of each monitoring period, obtaining corrosion categories and pollution factors obtained at each sampling time point, taking samples with the pollution factors exceeding pollution factor thresholds of corresponding categories as corrosion samples, establishing corrosion records, wherein the corrosion records are a set of continuous corrosion samples in the same corrosion category, the corrosion records comprise sampling time points of the corrosion samples and pollution factors of the corrosion samples, obtaining corrosion scores of all the corrosion records, evaluating the corrosion condition of the ball valve according to the corrosion scores, evaluating the corrosion condition according to the corrosion scores, and improving the accuracy of prediction.

Inventors

WANG ZHIYONG
WANG TING
Gu Dang
CHEN WEI
YOU LIHONG
JIN ZHENGHAO
YU CONG
WANG JUNHUA
LIU ZUO
WANG MIN
Xu Zongce
YU YOUPENG
Qu Xiaoshuai
YU JINWEI
ZHAO DONGDONG

Assignees

浙江超众阀门制造有限公司

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (9)

1. The monitoring and evaluating method for the rust condition of the ball valve is characterized by comprising the following steps of: Obtaining data parameters of a ball valve to be detected, wherein the data parameters comprise sampling positions of rust points, rust types, pollution factor values corresponding to the rust types, monitoring time periods, basic sampling frequencies of the monitoring time periods and sampling frequency coefficients of each monitoring time period; obtaining a sampling time point of each monitoring period according to the sampling frequency coefficient of each monitoring period; The rust category and the pollution factor acquired at each sampling time point are obtained, wherein the samples with the pollution factor exceeding the pollution factor threshold value of the corresponding category are taken as rust samples; Establishing a corrosion record, wherein the corrosion record is a set of continuous corrosion samples under the same corrosion category, and comprises a sampling time point of the corrosion samples and a pollution factor of the corrosion samples; and (5) obtaining the corrosion scores of all corrosion records, and evaluating the corrosion condition of the ball valve according to the corrosion scores.
2. The method for monitoring and evaluating the rust condition of a ball valve according to claim 1, wherein the setting of the data parameters comprises the steps of: The data parameters comprise m corrosion categories, wherein the corrosion category S= [ S 1 ,S 2 ,S 3 ,…,S m ], and the weight of the ith corrosion category S i is gamma i ; Setting data parameters including n monitoring periods AT, wherein the monitoring period at= [ AT 1 ,AT 2 ,AT 3 ,…,AT n ] and the sampling frequency coefficient of the j-th monitoring period AT j is delta j , setting the basic sampling frequency to be H 0 , obtaining the sampling frequency H j =δ j H 0 of the monitoring period AT j , and obtaining all sampling time points of the monitoring period AT j according to the sampling frequency H j of the monitoring period AT j ; The fouling factor of the S ω rust-like class included in the sampling data obtained AT the sampling time point ω of the monitoring period AT j is Q ω ; The step of establishing a rust record includes: Setting a pollution-causing factor threshold of the S ω type corrosion category as Q 0 , obtaining a pollution-causing factor Q ω of the S ω type corrosion category according to a sampling time point omega, taking the sub-sampling as a corrosion sampling when Q ω ≥Q 0 , adding the sub-sampling into a corrosion record L z of the S ω type corrosion category, setting the sampling time point omega as the first time of the corrosion record L z , setting a pollution-causing factor Q ω+ζ ＜Q 0 obtained by a sampling time point omega+zeta after the sampling time point omega, and taking the sampling time point omega+zeta as the last time of the corrosion record L z ; the step of calculating the rust fraction of the rust record comprises: Setting a rust record L z of a rust class of S ω to include r rust samples, wherein the peak value of a pollution causing factor is Q max , the valley value of the pollution causing factor is Q min , setting a pollution increasing interval LZ 1 to include r 1 rust samples, a pollution leveling interval LZ 2 to include r 2 samples, and a pollution reducing interval to include r 3 rust samples, wherein r=r 1 +r 2 +r 3 ,Q max ＞Q min ; obtaining a pollution increase interval LZ 1 , a pollution leveling interval LZ 2 and a pollution reduction interval LZ 3 of a corrosion record L z according to sampling time points of r corrosion samples and a pollution factor curve, quantifying time distribution and corrosion type lifting rates of the pollution increase interval LZ 1 , the pollution leveling interval LZ 2 and the pollution reduction interval LZ 3 , and obtaining an increase interval coefficient psi 1 of the pollution increase interval LZ 1 , an leveling interval coefficient psi 2 of the pollution leveling interval LZ 2 and a reduction interval coefficient psi 3 of the pollution reduction interval LZ 3 ; obtaining the rust score of the rust record L z of the S ω rust type Wherein the weight of the S ω rust-like class is gamma ω ; the step of obtaining the rust fraction includes: According to the selected preset time period T Pre-preparation , setting k corrosion records in the preset time period T Pre-preparation to obtain the score of the S ω corrosion category ; The step of evaluating the rust condition of the ball valve comprises the following steps: Setting S ω rust class in a preset time period T Pre-preparation fraction threshold F 0 ; When (when) When F 0 is not met, judging that the rust class of S ω does not cause destructive damage; When (when) And when the rust class is not less than F 0 , judging that the rust class of S ω is harmful, giving a warning, and uploading the analysis result of the sampling data.
3. The method for monitoring and evaluating the rust condition of a ball valve according to claim 2, wherein in the step of establishing the rust record, the sampling frequency coefficient δ j of the monitoring period AT j corresponding to the sampling time point ω is adjusted when Q ω ≥Q 0 is based on the fouling factor Q ω of the rust class S ω obtained AT the sampling time point ω, and the sampling frequency coefficient δ j ' of the adjusted monitoring period AT j is= γ ω δ j Wherein gamma ω is the weight of the S ω rust-like class, 0.5< gamma ω <1, Representative is rounded up to obtain the sampling frequency H of the adjustment monitoring period AT j j '=H j δ j '。
4. A method for monitoring and evaluating the rust condition of a ball valve according to claim 3, wherein after the sampling frequency is adjusted to H j ', the next sampling time point ω+1 is sampled to obtain a pollution factor Q ω+1 , and if the pollution factor Q ω+1 ≥Q ω is obtained, the sampling frequency coefficient δ j ' of the monitoring period AT j corresponding to the sampling time point ω is adjusted, and the adjusted sampling frequency δ j ″ = is adjusted γ ω δ j ' 。
5. The method for monitoring and evaluating the rust condition of a ball valve according to claim 4, wherein a lowest sampling frequency threshold H min is set, and when the adjusted sampling frequency reaches the lowest threshold H min , sampling is performed with H min as the sampling frequency.
6. The method for monitoring and evaluating the rust condition of the ball valve according to claim 2, wherein the method for obtaining the pollution increase interval coefficient ψ 1 in the rust record L z is as follows: Setting the duration of a pollution increasing interval LZ 1 as T 1 and the total duration of a rust record L z as T 0 to obtain the time distribution lambda 1 = T 1 / T 0 of the pollution increasing interval; The pollution increase rate epsilon 1 =（Q max -Q ω ）/ T 1 ; The increment interval coefficient psi 1 =λ 1 ε 1 of the pollution increment interval.
7. The method for monitoring and evaluating the rust condition of the ball valve according to claim 2, wherein the method for obtaining the leveling interval coefficient ψ 3 of the pollution reduction interval is as follows: setting the duration of a pollution leveling interval as T 2 and the total duration in the corrosion record as T 0 , so as to obtain the time distribution lambda 2 = T 2 / T 0 of a pollution growth interval; The leveling amplitude epsilon 1 =Q ω / 2T 2 ; The increment interval coefficient psi 2 =λ 2 ε 2 of the pollution increment interval.
8. The method for monitoring and evaluating the rust condition of the ball valve according to claim 2, wherein the method for obtaining the pollution reduction coefficient ψ 3 of the pollution reduction interval is as follows: setting the duration of a pollution reduction interval as T 3 and the total duration in the rust record as T 0 , so as to obtain the time distribution lambda 3 = T 3 / T 0 of a pollution increase interval; the pollution increase rate epsilon 3 =（Q ω -Q mix ）/ T 2 ; The increment interval coefficient psi 3 =λ 3 ε 3 of the pollution increment interval.
9. The method for monitoring and evaluating the rust condition of the ball valve according to claim 8, wherein the occurrence of rust sampling is predicted based on historical quality data by using a deep learning neural network, and the adjustment of the monitoring period setting is performed based on the prediction result.

Description

Monitoring and evaluating method for rust condition of ball valve Technical Field The application relates to the field of ball valve corrosion monitoring, in particular to a monitoring and evaluating method for ball valve corrosion conditions. Background Ball valves are often used in petroleum, chemical or water treatment industries, and due to medium problems, valves can generate corrosion problems, such as chemical corrosion, chemical reaction of metal surfaces with surrounding media (such as corrosive substances of acid, alkali, salt and the like), or electrochemical corrosion (caused by electrolyte formed on metal surfaces by water films in a wet environment), biological corrosion (caused by growth of microorganisms on the metal surfaces) and the like, the sealing structure of the valves can be damaged, so that the sealing performance is reduced, leakage phenomena occur, the leakage caused by the valve corrosion can cause environmental pollution, and serious potential safety hazards such as explosion, fire and damage to personnel health can be brought if the leakage is flammable and explosive or toxic and harmful fluid. In a chemical plant, many devices are made of metal, leakage caused by valve corrosion can occur due to oil leakage, gas leakage, leakage and other phenomena of the devices, and serious accidents such as explosion, fire and the like are caused, so that life and property damage is caused to the environment and personnel around the plant area. Therefore, it is necessary to consider monitoring and evaluating the rust damage condition of the ball valve, the currently used method includes sensor technology (such as pressure sensor and flow sensor), visual detection (camera or laser scanning), acoustic detection (by sound analysis to judge blockage), vibration analysis, and prediction model based on data analysis, when the aspects of accuracy, real-time performance, cost, maintenance difficulty, environmental adaptability and the like are involved, the sensor may be easily interfered by environment, visual detection is poor in dark place or serious pollution, high-precision equipment is required for the acoustic detection, the cost is high, a large amount of data support is required for the data analysis model, and fitting problems may exist, therefore, it is necessary to find a more reliable and economical detection scheme, and it is necessary to integrate various technologies to improve the effect, and a design capable of ensuring data accuracy, reducing false alarm rate and performing full-scale optimization on the ball valve and its application environment (pipeline) is required. Disclosure of Invention In order to solve the problem, the application provides a monitoring and evaluating method for the rust condition of a ball valve, which comprises the following steps: Obtaining data parameters of a ball valve to be detected, wherein the data parameters comprise sampling positions of rust points, rust types, pollution factor values corresponding to the rust types, monitoring time periods, basic sampling frequencies of the monitoring time periods and sampling frequency coefficients of each monitoring time period; obtaining a sampling time point of each monitoring period according to the sampling frequency coefficient of each monitoring period; The rust category and the pollution factor acquired at each sampling time point are obtained, wherein the samples with the pollution factor exceeding the pollution factor threshold value of the corresponding category are taken as rust samples; Establishing a corrosion record, wherein the corrosion record is a set of continuous corrosion samples under the same corrosion category, and comprises a sampling time point of the corrosion samples and a pollution factor of the corrosion samples; and (5) obtaining the corrosion scores of all corrosion records, and evaluating the corrosion condition of the ball valve according to the corrosion scores. Wherein, the setting of the data parameters comprises the following steps: The data parameters comprise m corrosion categories, wherein the corrosion category S= [ S 1,S2,S3,…,Sm ], and the weight of the ith corrosion category S i is gamma i; Setting data parameters including n monitoring periods AT, wherein the monitoring period at= [ AT 1,AT2,AT3,…,ATn ] and the sampling frequency coefficient of the j-th monitoring period AT j is delta j, setting the basic sampling frequency to be H 0, obtaining the sampling frequency H j=δjH0 of the monitoring period AT j, and obtaining all sampling time points of the monitoring period AT j according to the sampling frequency H j of the monitoring period AT j; The fouling factor of the S ω rust-like class included in the sampling data obtained AT the sampling time point ω of the monitoring period AT j is Q ω; The step of establishing a rust record includes: Setting a pollution-causing factor threshold of the S ω type corrosion category as Q 0, obtaining a pollution-causing factor