Converter valve and optical CT maintenance plan adjustment method based on dynamic alarm function

Abstract

The invention relates to the technical field of flexible direct-current power transmission, and particularly discloses a converter valve and optical CT maintenance plan adjustment method based on a dynamic alarm function, and the method comprises the following steps: firstly, building a state evaluation model for a converter valve bridge arm and an optical CT; secondly, making a state opportunity maintenance strategy; establishing an alarm maintenance function, an opportunity maintenance function and a replacement threshold value of the converter valve bridge arm and the optical CT; solving the alarm maintenance threshold function and the opportunity maintenance threshold function; when the current monitoring value suddenly increases and the state monitoring value exceeds a replacement threshold value, the module is replaced; according to the calculated and solved maintenance threshold function in each period, comparing the maintenance threshold function with a state monitoring value, and finally formulating a proper combined maintenance strategy. According to the converter valve and optical CT maintenance plan adjustment method based on the dynamic alarm function, the maintenance threshold function is adjusted according to different maintenance effects and maintenance times, so that the maintenance threshold function corresponds to the real-time state of the component, and the effectiveness of the maintenance strategy is ensured.

Description

Technical Field

[0001] The invention relates to the technical field of flexible direct current transmission, and in particular to a method for adjusting a converter valve and an optical CT maintenance plan based on a dynamic alarm function. Background Art

[0002] Due to the characteristics of renewable energy generation such as small power generation capacity, unstable output power, wide distribution range and far away from the load center, the previous AC transmission technology or the more traditional DC transmission technology has certain defects in facing these problems. In this context, flexible DC transmission technology can better deal with these problems. Compared with the previous DC transmission system based on ordinary thyristors, the flexible DC transmission system can adjust the operation mode more flexibly and better control the system. The output of new energy is greatly affected by external environmental factors, and its output power is extremely unstable. The flexible DC transmission system can better combine with this unstable energy through precise control capabilities to improve energy utilization efficiency. After the flexible DC project was put into actual operation, it faced a severe challenge: the faults of the converter valve and pure optical CT (current transformer) have caused the system to shut down unexpectedly many times, posing a serious threat to the safe and stable operation of the flexible DC power grid. Due to the long-term operation characteristics of the flexible DC equipment, the unknown fault mechanism and the low efficiency of fault analysis and positioning, the maintenance cost is high, and the maintenance strategy is adjusted with the change of the operating status.

[0003] In the formulation of current converter valve maintenance strategies, most adopt the state maintenance method, that is, maintenance decisions are made based on the real-time operating status of the components, and preventive maintenance is performed on components in a degraded state, but the impact of the maintenance effect on subsequent operation is not considered, that is, the decision is made using an unchanging maintenance threshold function. As the number of maintenance increases, the misjudgment rate of the component degradation state will also increase, affecting the reliability of the decision. Therefore, there is an urgent need for a dynamic adjustment technology for maintenance plans based on the Weibull proportional intensity model. Summary of the invention

[0004] The purpose of the present invention is to provide a method for adjusting the maintenance plan of a converter valve and an optical CT based on a dynamic alarm function, taking into account the impact on the health status of the components after each maintenance, and different maintenance methods have different degrees of impact on the health status of the components. At the same time, the maintenance threshold function determined based on economic efficiency will also change accordingly and be continuously adjusted, so that the real-time status assessment is more accurate, the reliability and economy of the decision-making are improved, and the stable and safe operation of the converter station is ensured.

[0005] To achieve the above object, the present invention provides a method for adjusting the maintenance plan of a converter valve and an optical CT based on a dynamic alarm function, comprising the following steps:

[0006] S1. Establish a state assessment model for the converter valve bridge arm and optical CT;

[0007] S2, formulate state opportunity maintenance strategy;

[0008] Establish the alarm maintenance function M of the converter valve bridge arm and optical CT respectively c k , opportunity maintenance function M o k and the replacement threshold M(max), in a converter valve, including six converter valve bridge arms and two optical CTs, that is, these eight components are grouped together to formulate a combined maintenance strategy;

[0009] S3, solving the alarm maintenance threshold function and the opportunity maintenance threshold function;

[0010] S4. When the current monitoring value suddenly increases and the status monitoring value exceeds the replacement threshold, the component is replaced;

[0011] S5. Compare the calculated maintenance threshold function in each cycle with the condition monitoring value, and finally formulate a suitable combined maintenance strategy.

[0012] Preferably, in step S1, the specific process of establishing the state assessment model for the converter valve bridge arm and the optical CT is as follows:

[0013] Considering the impact of maintenance effect on the real-time health status of components, the component status assessment model is established as the Weibull proportional strength model. This model considers equipment life, maintenance activities and status information while calculating the failure rate. Its expression is as follows:

[0014]

[0015] Among them, β is the shape parameter, η is the life parameter, γ is the p-dimensional vector of the regression coefficient of Z arranged in the corresponding order, Z represents the real-time monitoring value of the covariate, and the covariate in the status evaluation of the converter valve and optical CT is the current value at each location, T n is the time required for maintenance, ε n is the improvement factor, with a value range of [0, 1]. If the minimum maintenance is adopted, ε n =1, if the replacement method is adopted, then ε n =0, β, η, γ are solved by maximum likelihood estimation based on historical fault data and current monitoring values;

[0016] ε n The calculation formula is as follows:

[0017]

[0018] Among them, u is the cost adjustment coefficient, C jxThe cost of maintenance, C gh is the cost required for replacement, v is the time coefficient, t is consistent with t in the Weibull proportional strength model, is the component operation time, n is the number of maintenance times, and w is the maintenance coefficient.

[0019] Preferably, in step S2, when the status monitoring value of one component exceeds the alarm maintenance function, the real-time status of other components is monitored, and the components whose real-time status exceeds the alarm maintenance threshold function are overhauled, and the components whose real-time status exceeds the opportunity maintenance threshold function are minor repaired;

[0020] Considering the sudden serious failure of components, a replacement threshold is set. When the component status characterization value exceeds the replacement threshold, replacement maintenance measures are taken. Considering that the number of repairs of a component should not be too many, when the number of repairs of the component exceeds the set value, replacement maintenance measures are also taken.

[0021] Preferably, in step S3, the goal of solving the alarm maintenance threshold function is:

[0022] The cost-effectiveness index of preventive overhaul is the largest within a maintenance cycle unit time. The variable is the overhaul time. The corresponding optimal preventive overhaul time and the current value under the corresponding time are obtained and substituted into the alarm maintenance function.

[0023] The goal of solving the opportunity maintenance threshold function is:

[0024] The maintenance cost of preventive minor repairs per unit time in a maintenance cycle is the lowest. The variable is the minor repair time. The corresponding optimal preventive minor repair time and the current value under the corresponding time are obtained. Substituting them into the solution is the opportunity maintenance function.

[0025] The cost-effectiveness index is the ratio of the difference between the economic benefit brought by maintenance and the cost required for maintenance to the maintenance cycle.

[0026] Preferably, in step S3, the alarm maintenance threshold function M is obtained by rewriting the proportional strength model c (t) and the opportunity maintenance threshold function M o (t) is expressed as follows:

[0027]

[0028] Among them, β and η are known, ε n T n Solve according to the maintenance situation, so the unknown quantity in the maintenance threshold function is H c and H o , H c and H o The expressions for the two unknown quantities are:

[0029]

[0030] For any component i, in the two maintenance threshold expressions of the warning maintenance threshold and the opportunity maintenance threshold, the unknown quantities are the optimal overhaul time, the optimal minor repair time and the current value at the corresponding time. Substitute them into the expressions to find H c and H o , and then solve the warning maintenance threshold function and opportunity maintenance threshold function of component i.

[0031] Preferably, in step S3, in the first maintenance cycle, since all components are not repaired, ε0T0=0, and the failure function expression in this cycle is:

[0032]

[0033] The process of solving the alarm maintenance threshold function is as follows: According to the relevant knowledge of reliability theory, let the reliability function of the component life X be R(x), the time interval from the last preventive maintenance moment to the next preventive maintenance moment is called the maintenance cycle, let the preventive maintenance cycle be T, E(Y) represents the average length of the state maintenance cycle, and we have:

[0034]

[0035] Maintenance costs within the cycle C c It consists of two parts: the cost of maintenance implementation and C m and decision misjudgment costs C n ;

[0036] Cost of maintenance implementation C m : Let the failure repair cost be C1, which includes downtime loss, fixed repair cost and equipment failure repair cost; let the preventive overhaul cost be C2, which includes downtime loss, fixed repair cost and equipment overhaul cost, then:

[0037] C m =C1N1+C2R(x)

[0038] Where N is the predicted number of failures in one cycle, expressed as follows:

[0039]

[0040] Costs of misjudgment of decision making C n : Decision misjudgment includes two situations, namely, fault misjudgment and fault omission. Assume that the subsequent cost caused by fault misjudgment is C4, the loss caused by fault omission is C5, the probability of misjudgment is P1, and the probability of omission is P2:

[0041] C n=P1·E(Y)·C4+P2·E(Y)·C5

[0042] C c =C m +C n

[0043] Assume that the economic benefit generated by maintenance during the period is W c , which is the benefit brought by the reduced downtime. Assuming the loss caused by the shutdown of the converter valve system per unit time is Q1, we have:

[0044] Q c =(TE(Y))·Q1

[0045] Assume the cost-effectiveness index is k1:

[0046]

[0047] The above equation is differentiated with respect to T and is set to zero to solve the optimal overhaul time T1. The current value at the corresponding time is the optimal overhaul state value Z. c , T1 and Z c Substitute it into the state maintenance threshold expression to calculate the alarm maintenance threshold function with the lowest cost.

[0048] Preferably, in step S3, the process of solving the opportunity maintenance threshold function is:

[0049] Assume that the minimum maintenance cost after a failure is C3, which includes downtime loss, fixed maintenance cost and equipment minor repair cost. Minimum maintenance means restoring the component to normal operation through maintenance, but the failure rate inside the component has the same trend as before the maintenance. The cost of preventive overhaul is C2. Since the failure rate of the component after the minimum maintenance has the same trend as before the maintenance, the average length of the component's opportunity maintenance cycle is T and remains constant. The cost of maintenance within the cycle is C u It also consists of two parts, namely the cost of maintenance implementation and C s and decision misjudgment costs C g ;

[0050] The cost C required for maintenance implementation within the period T s for:

[0051]

[0052] Assume that the subsequent cost caused by fault misjudgment is C6, the loss caused by fault omission is C7, the probability of misjudgment is P3, and the probability of omission is P4, then the cost of decision misjudgment is C q for:

[0053] C q=P3·T·C6+P4·T·C7

[0054] C u =C s +C q

[0055] The cost of preventive minor repairs per unit time is k2:

[0056]

[0057] Derivative the above formula with respect to T and set it to zero to solve the optimal opportunity maintenance time T o , find the current value at the corresponding moment is the optimal minor repair state value Z o ; T o and Z o Substitute it into the calculation formula of the threshold function to solve the opportunity maintenance threshold function;

[0058] After the first combined maintenance, considering the maintenance effect ε n T n is not 0; then the failure function expression in the second maintenance cycle is:

[0059]

[0060] Where T1 is the time required for the first maintenance, ε1 determines the maintenance cost according to the first maintenance method, and ε1 is regarded as a function of the running time t. The maintenance threshold function of the second cycle is:

[0061]

[0062] The above method is used to solve and obtain the maintenance threshold function of the second cycle;

[0063] After each component maintenance, the correlation coefficient of the improvement factor ε is determined according to the maintenance method, and it is iterated on the failure function of the previous maintenance cycle, and then the maintenance threshold function is updated using the method that minimizes the maintenance cost per unit time.

[0064] Therefore, the present invention adopts the above-mentioned method for adjusting the maintenance plan of the converter valve and the optical CT based on the dynamic alarm function, and the beneficial effects are as follows:

[0065] (1) Compared with the conventional fixed maintenance threshold function, the present invention can continuously adjust the maintenance threshold function as the maintenance proceeds, so as to make a more accurate judgment on whether the component status has deteriorated.

[0066] (2) The present invention takes into account that different types of maintenance methods generate different maintenance costs and maintenance times, which causes the value of the improvement factor to fluctuate accordingly. The improvement factor expression is used to accurately describe the maintenance effect, thereby accurately calculating the degree of state regression.

[0067] (3) The present invention takes into account sudden serious faults and sets a replacement threshold to ensure safe and reliable operation of the system; takes into account the impact of maintenance on component performance and sets a maximum number of component maintenance times. Components that reach the maximum number of maintenance times are replaced to ensure efficient operation of each component; at the same time, the maintenance threshold function is updated after the component is replaced, and it is iterated again with the maintenance, thereby obtaining a more complete dynamic adjustment system.

[0068] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is an overall flow chart of an embodiment of a method for adjusting a maintenance plan of a converter valve and an optical CT based on a dynamic alarm function of the present invention;

[0070] Figure 2 It is a topological structure diagram of a flexible direct current converter valve according to an embodiment of a converter valve and optical CT maintenance plan adjustment method based on a dynamic alarm function of the present invention;

[0071] Figure 3 It is a maintenance decision diagram of each component in an embodiment of a method for adjusting maintenance plans of converter valves and optical CTs based on a dynamic alarm function of the present invention. DETAILED DESCRIPTION

[0072] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0073] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0074] After a component undergoes preventive overhaul or preventive minor repair, its health status will change to a certain extent. The maintenance effect is between replacement, i.e., "restored to like new", and minimum maintenance, i.e., "restored to like old". Therefore, the maintenance threshold function initially set will not match the real-time status of the component as the number of component maintenance increases. The misjudgment rate of the component degradation status will also increase, affecting the reliability of maintenance decisions. Therefore, a dynamic adjustment technology is needed, that is, as the component is repaired, the maintenance threshold function is adjusted according to different maintenance effects and maintenance times to correspond to the real-time status of the component, ensuring the effectiveness of the maintenance strategy.

[0075] like Figure 1 As shown, a method for adjusting the maintenance plan of a converter valve and an optical CT based on a dynamic alarm function comprises the following steps:

[0076] In the combined operation and maintenance of converter valves and optical CTs, the six bridge arms of the converter valves are regarded as six independent research objects, and the optical CTs in the two DC lines are regarded as two research objects, that is, the combined maintenance strategy of the eight components is considered.

[0077] S1. Establish a state assessment model for the converter valve bridge arm and optical CT, that is, a proportional intensity model that considers the impact of maintenance effect on real-time status. The specific process is as follows:

[0078] The flexible DC converter valve is the core component of the flexible DC transmission system, which is used to realize the conversion and control of AC and DC. The flexible DC converter valve usually adopts a multi-stage rectifier bridge or inverter bridge topology, equipped with a control circuit and power switching devices to realize the conversion from AC to DC and DC to AC. The topology of the flexible DC converter valve is as follows: Figure 2 shown.

[0079] SM1, SM2, ..., SMN are half-bridge sub-modules used to construct full-bridge or multi-level converter topologies. In these topologies, multiple half-bridge sub-modules can be combined in series or in parallel to achieve higher power conversion from AC to DC or DC to AC.

[0080] In the maintenance strategy, the bridge arm of one phase of the converter valve is treated as an integral maintenance component. If its operating state deteriorates, the bridge arm is repaired. A converter valve includes six bridge arms, which means it is divided into six independent components.

[0081] Photocurrent transformer (optical CT) is a sensor device for measuring current, which uses optical principles to achieve non-contact measurement of current. Compared with traditional current transformers (CT), photocurrent transformers have higher accuracy and lower loss, so they are widely used when high-precision current measurement is required and in power system status monitoring.

[0082] In the flexible DC converter valve, each converter valve corresponds to two optical CTs, which are installed in the two DC lines led out of the converter valve respectively. They can cooperate with the hybrid circuit breaker to protect the circuit and control the circulation of the converter valve.

[0083] Considering the impact of maintenance effect on the real-time health status of components, the component status assessment model is established as the Weibull proportional strength model. This model considers equipment life, maintenance activities and status information while calculating the failure rate, thereby making a more comprehensive assessment of the equipment status. Its expression is as follows:

[0084]

[0085] Among them, β is the shape parameter, η is the life parameter, γ is the p-dimensional vector of the regression coefficient of Z arranged in the corresponding order, Z represents the real-time monitoring value of the covariate, and the covariate in the status evaluation of the converter valve and optical CT is the current value at each location, T n is the time required for maintenance, ε n is the improvement factor, with a value range of [0, 1]. If the minimum maintenance is adopted, ε n =1, if the replacement method is adopted, then ε n =0, β, η, γ are solved by maximum likelihood estimation based on historical fault data and current monitoring values.

[0086] ε n The calculation formula is as follows:

[0087]

[0088] Among them, u is the cost adjustment coefficient, C jx The cost of maintenance, C gh is the cost of replacement, v is the time coefficient, t is consistent with t in the Weibull proportional strength model, is the component operation time, n is the number of maintenance times, and w is the maintenance coefficient. Based on this, the condition assessment model of the converter valve bridge arm and optical CT can be established.

[0089] S2, formulate state opportunity maintenance strategy;

[0090] Establish the alarm maintenance function M of the converter valve bridge arm and optical CT respectively k , opportunity maintenance function M o k and the replacement threshold M(max). In a converter valve, there are six converter valve bridge arms and two optical CTs. The eight components are grouped together to formulate a combined maintenance strategy. When the status monitoring value of one component exceeds the alarm maintenance function M c k When the real-time status of other components exceeds the alarm maintenance threshold function M c k The components of the overhaul, the real-time status of other components exceeds the opportunity maintenance threshold function M o k Minor repairs are performed on the components; and maintenance measures are taken only when the status representation values ​​of other components exceed the alarm maintenance function.

[0091] At the same time, considering the sudden serious failure of the component, a replacement threshold is set. When the component status representation value exceeds the replacement threshold M(max), the replacement maintenance measure is taken immediately to prevent further affecting the normal operation of other components. In addition, considering that the number of maintenance times of a component should not be too many, when the number of maintenance times of the component exceeds the set value N, the replacement maintenance measure is also taken. The maintenance decision of each component is as follows Figure 3 shown.

[0092] S3, solving the alarm maintenance threshold function and the opportunity maintenance threshold function;

[0093] The economic efficiency is calculated, and the goal of solving the alarm maintenance threshold function is:

[0094] The cost-effectiveness index of preventive overhaul in a maintenance cycle is maximized. The variable is the overhaul time. The corresponding optimal preventive overhaul time and the state characterization value (current value) at the corresponding time are obtained. Substitute them into the solution to solve the alarm maintenance function.

[0095] The goal of solving the opportunity maintenance threshold function is:

[0096] Make the maintenance cost of preventive minor repairs per unit time in a maintenance cycle as low as possible. The variable is the minor repair time. Obtain the corresponding optimal preventive minor repair time and the state characterization value (current value) at the corresponding time, and substitute them to solve the opportunity maintenance function.

[0097] The cost-effectiveness index is the ratio of the difference between the economic benefits brought by maintenance and the cost required for maintenance to the maintenance cycle. The larger the index value, the better the maintenance effect.

[0098] Due to the existence of the improvement factor, the optimal maintenance time after each maintenance changes, and the maintenance threshold function also changes. The main factors causing fluctuations are the change of the improvement factor over time and the number of maintenance times, as well as the difference in maintenance time under different maintenance methods. In the first maintenance cycle, ε n and T n are all 0, and the two maintenance threshold functions are calculated on this basis; but after taking maintenance actions, ε n and T n The value of changes, and the corresponding two maintenance threshold functions change accordingly. By analogy, each subsequent maintenance action will cause the maintenance threshold function to change. Therefore, the algorithm will continuously iterate the maintenance threshold function until the component is replaced, so that the component state is restored to be like new, and the maintenance threshold function is calculated again. In addition, when the state representation value of a component exceeds the replacement threshold, the maintenance measure of replacement is taken immediately, and the component state is also restored to be like new.

[0099] Specifically, the alarm maintenance threshold function M is obtained by rewriting the proportional intensity model: c (t) and the opportunity maintenance threshold function M o (t) is expressed as follows:

[0100]

[0101] Among them, β and η are known, ε n T nIt can be solved according to the maintenance situation, so the unknown quantity in the maintenance threshold function is H c and H o , H c and H o The expressions for the two unknown quantities are:

[0102]

[0103] For any component i, in the two maintenance threshold expressions of the warning maintenance threshold and the opportunity maintenance threshold, the unknown quantities are the optimal overhaul time, the optimal minor repair time and the current value at the corresponding time. Substitute them into the expressions to find H c and H o , and then solve the warning maintenance threshold function and opportunity maintenance threshold function of component i.

[0104] In the first maintenance cycle, since all components are not repaired, ε0T0=0, and the failure function expression in this cycle is:

[0105]

[0106] First, the process of solving the alarm maintenance threshold function is as follows: According to the relevant knowledge of reliability theory, let the reliability function of component life X be R(x), the time interval from the last preventive maintenance moment to the next preventive maintenance moment is called the maintenance cycle, let the preventive maintenance cycle be T, E(Y) represents the average length of the state maintenance cycle, and we have:

[0107]

[0108] Maintenance costs within the cycle C c It consists of two parts: the cost of maintenance implementation and C m and decision misjudgment costs C n ;

[0109] Cost of maintenance implementation C m : Let the failure repair cost be C1, which includes downtime loss, fixed repair cost and equipment failure repair cost; let the preventive overhaul cost be C2, which includes downtime loss, fixed repair cost and equipment overhaul cost, then:

[0110] C m =C1N1+C2R(x)

[0111] Where N is the predicted number of failures in one cycle, expressed as follows:

[0112]

[0113] Costs of misjudgment of decision making C n: Decision misjudgment includes two situations, namely, fault misjudgment and fault omission. Assume that the subsequent cost caused by fault misjudgment is C4, the loss caused by fault omission is C5, the probability of misjudgment is P1, and the probability of omission is P2:

[0114] C n =P1·E(Y)·C4+P2·E(Y)·C5

[0115] C c =C m +C n

[0116] Assume that the economic benefit generated by maintenance during the period is W c , which is the benefit brought by the reduced downtime. Assuming the loss caused by the shutdown of the converter valve system per unit time is Q1, we have:

[0117] W c =(TE(Y))·Q1

[0118] Assume the cost-effectiveness index is k1:

[0119]

[0120] In order to maximize the cost-effectiveness index per unit time, the above formula is differentiated with respect to T and set to zero to solve the optimal overhaul time T1. The current value at the corresponding moment is the optimal overhaul state value Z. c , T1 and Z c Substitute it into the state maintenance threshold expression to calculate the alarm maintenance threshold function with the lowest cost.

[0121] Secondly, the process of solving the opportunity maintenance threshold function is:

[0122] Assume that the minimum maintenance cost after a failure is C3, which includes downtime loss, fixed maintenance cost and equipment minor repair cost. Minimum maintenance means restoring the component to normal operation through maintenance, but the failure rate inside the component has the same trend as before the maintenance, and the cost of preventive overhaul is C2; since the failure rate of the component after the minimum maintenance has the same trend as before the maintenance, the average length of the component's opportunity maintenance cycle is T and remains constant; the cost of maintenance within the cycle is C u It also consists of two parts, namely the cost of maintenance implementation and C s and decision misjudgment costs C q ;

[0123] The cost C required for maintenance implementation within the period T s for:

[0124]

[0125] Assume that the subsequent cost caused by fault misjudgment is C6, the loss caused by fault omission is C7, the probability of misjudgment is P3, and the probability of omission is P4, then the cost of decision misjudgment is C q for:

[0126] C q =P3·T·C6+P4·T·C7

[0127] C u =C s +C q

[0128] The cost of preventive minor repairs per unit time is k2:

[0129]

[0130] Derivative the above formula with respect to T and set it to zero to solve the optimal opportunity maintenance time T o , find the current value at the corresponding moment is the optimal minor repair state value Z o ; T o and Z o Substituting it into the calculation formula of the threshold function, the opportunity maintenance threshold function can be solved.

[0131] After the first combined maintenance, considering the maintenance effect ε n T n is not 0; then the failure function expression in the second maintenance cycle is:

[0132]

[0133] Where T1 is the time required for the first maintenance, ε1 can determine the maintenance cost according to the first maintenance method, and ε1 can be regarded as a function of the running time t. Then the maintenance threshold function of the second cycle is:

[0134]

[0135] Similarly, the above method is used to solve and obtain the maintenance threshold function of the second cycle;

[0136] Similarly, after each component maintenance, the correlation coefficient of the improvement factor ε is determined according to the maintenance method, and it is iterated on the failure function of the previous maintenance cycle. Then, the maintenance threshold function is updated using the method that minimizes the maintenance cost per unit time to ensure the effectiveness of the threshold function setting.

[0137] S4. Considering the emergency fault situation, when the current monitoring value suddenly increases and the status monitoring value exceeds the replacement threshold, in order to ensure the safe operation of the converter valve, the component is replaced, that is, the status is restored to be as good as new, ∑ε n T nBack off to 0, the repair threshold function iterates again.

[0138] In addition, considering that the number of maintenance times of a component should not be too many, when the number of maintenance times of the component exceeds the set value N, the maintenance measure of replacement is also taken, and the maintenance threshold function is also updated accordingly.

[0139] S5. Compare the calculated maintenance threshold function in each cycle with the condition monitoring value, and finally formulate a suitable combined maintenance strategy.

[0140] Therefore, the present invention adopts the above-mentioned method for adjusting the maintenance plan of the converter valve and optical CT based on the dynamic alarm function. As the components are maintained, the maintenance threshold function is adjusted according to different maintenance effects and maintenance times to make it correspond to the real-time status of the components, thereby ensuring the effectiveness of the maintenance strategy.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for adjusting the maintenance plan of a converter valve and optical CT based on a dynamic alarm function, characterized in that: The following steps are involved: S1. Establish a state assessment model for the converter valve bridge arm and optical CT; S2, formulate state opportunity maintenance strategy; Establish the alarm maintenance function M of the converter valve bridge arm and optical CT respectively c k , opportunity maintenance function M o k and the replacement threshold M(max), in a converter valve, including six converter valve bridge arms and two optical CTs, that is, these eight components are grouped together to formulate a combined maintenance strategy; S3, solving the alarm maintenance threshold function and the opportunity maintenance threshold function; S4. When the current monitoring value suddenly increases and the status monitoring value exceeds the replacement threshold, the component is replaced; S5. Compare the calculated maintenance threshold function in each cycle with the condition monitoring value, and finally formulate a suitable combined maintenance strategy.

2. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 1, characterized in that: In step S1, the specific process of establishing the state assessment model for the converter valve bridge arm and the optical CT is as follows: Considering the impact of maintenance effect on the real-time health status of components, the component status assessment model is established as the Weibull proportional strength model. This model considers equipment life, maintenance activities and status information while calculating the failure rate. Its expression is as follows: Among them, β is the shape parameter, η is the life parameter, γ is the p-dimensional vector of the regression coefficient of Z arranged in the corresponding order, Z represents the real-time monitoring value of the covariate, and the covariate in the status evaluation of the converter valve and optical CT is the current value at each location, T n is the time required for maintenance, ε n is the improvement factor, with a value range of [0,1]. If the minimum maintenance is adopted, ε n =1, if the replacement method is adopted, then ε n =0, β, η, γ are solved by maximum likelihood estimation based on historical fault data and current monitoring values; ε n The calculation formula is as follows: Among them, u is the cost adjustment coefficient, C jx The cost of maintenance, C gh is the cost required for replacement, v is the time coefficient, t is consistent with t in the Weibull proportional strength model, is the component operation time, n is the number of maintenance times, and w is the maintenance coefficient.

3. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 2, characterized in that: In step S2, when the status monitoring value of one component exceeds the alarm maintenance function, the real-time status of other components is monitored, and the components whose real-time status exceeds the alarm maintenance threshold function are overhauled, and the components whose real-time status exceeds the opportunity maintenance threshold function are minor repaired; Considering the sudden serious failure of components, a replacement threshold is set. When the component status characterization value exceeds the replacement threshold, replacement maintenance measures are taken. Considering that the number of repairs of a component should not be too many, when the number of repairs of the component exceeds the set value, replacement maintenance measures are also taken.

4. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 3 is characterized in that: In step S3, the goal of solving the alarm maintenance threshold function is: The cost-effectiveness index of preventive overhaul is the largest within a maintenance cycle unit time. The variable is the overhaul time. The corresponding optimal preventive overhaul time and the current value under the corresponding time are obtained and substituted into the alarm maintenance function. The goal of solving the opportunity maintenance threshold function is: The maintenance cost of preventive minor repairs per unit time in a maintenance cycle is the lowest. The variable is the minor repair time. The corresponding optimal preventive minor repair time and the current value under the corresponding time are obtained. Substituting them into the solution is the opportunity maintenance function. The cost-effectiveness index is the ratio of the difference between the economic benefit brought by maintenance and the cost required for maintenance to the maintenance cycle.

5. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 4, characterized in that: In step S3, the alarm maintenance threshold function M is obtained by rewriting the proportional strength model c (t) and the opportunity maintenance threshold function M o (t) is expressed as follows: Among them, β and η are known, ε n T n Solve according to the maintenance situation, so the unknown quantity in the maintenance threshold function is H c and H o , H c and H o The expressions for the two unknown quantities are: For any component i, in the two maintenance threshold expressions of the warning maintenance threshold and the opportunity maintenance threshold, the unknown quantities are the optimal overhaul time, the optimal minor repair time and the current value at the corresponding time. Substitute them into the expressions to find H c and H o , and then solve the warning maintenance threshold function and opportunity maintenance threshold function of component i.

6. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 5, characterized in that: In step S3, in the first maintenance cycle, since all components are not repaired, ε0T0=0, and the failure function expression in this cycle is: The process of solving the alarm maintenance threshold function is as follows: According to the relevant knowledge of reliability theory, let the reliability function of the component life X be R(x), the time interval from the last preventive maintenance moment to the next preventive maintenance moment is called the maintenance cycle, let the preventive maintenance cycle be T, E(Y) represents the average length of the state maintenance cycle, and we have: Maintenance costs within the cycle C c It consists of two parts: the cost of maintenance implementation and C m and decision misjudgment costs C n ; Cost of maintenance implementation C m : Let the failure repair cost be C1, which includes downtime loss, fixed repair cost and equipment failure repair cost; let the preventive overhaul cost be C2, which includes downtime loss, fixed repair cost and equipment overhaul cost, then: C m =C1N1+C2R(x) Where N is the predicted number of failures in one cycle, expressed as follows: Costs of misjudgment of decision making C n : Decision misjudgment includes two situations, namely, fault misjudgment and fault omission. Assume that the subsequent cost caused by fault misjudgment is C4, the loss caused by fault omission is C5, the probability of misjudgment is P1, and the probability of omission is P2: C n =P1·E(Y)·C4+P2·E(Y)·C5 C c =C m +C n Assume that the economic benefit generated by maintenance during the period is W c , which is the benefit brought by the reduced downtime. Assuming the loss caused by the shutdown of the converter valve system per unit time is Q1, we have: W c =(T-E(Y))·Q1 Assume the cost-effectiveness index is k1: The above equation is differentiated with respect to T and is set to zero to solve the optimal overhaul time T1. The current value at the corresponding time is the optimal overhaul state value Z. c , T1 and Z c Substitute it into the state maintenance threshold expression to calculate the alarm maintenance threshold function with the lowest cost.

7. The method for adjusting the maintenance plan of converter valve and optical CT based on dynamic alarm function according to claim 6, characterized in that: In step S3, the process of solving the opportunity maintenance threshold function is: Assume that the minimum maintenance cost after a failure is C3, which includes downtime loss, fixed maintenance cost and equipment minor repair cost. Minimum maintenance means restoring the component to normal operation through maintenance, but the failure rate inside the component has the same trend as before the maintenance. The cost of preventive overhaul is C2. Since the failure rate of the component after the minimum maintenance has the same trend as before the maintenance, the average length of the component's opportunity maintenance cycle is T and remains constant. The cost of maintenance within the cycle is C u It also consists of two parts, namely the cost of maintenance implementation and C s and decision misjudgment costs C q ; The cost C required for maintenance implementation within the period T s for: Assume that the subsequent cost caused by fault misjudgment is C6, the loss caused by fault omission is C7, the probability of misjudgment is P3, and the probability of omission is P4, then the cost of decision misjudgment is C q for: C q =P3·T·C6+P4·T·C7 C u =C s +C q The cost of preventive minor repairs per unit time is k2: Derivative the above formula with respect to T and set it to zero to solve the optimal opportunity maintenance time T o , find the current value at the corresponding moment is the optimal minor repair state value Z o ; T o and Z o Substitute it into the calculation formula of the threshold function to solve the opportunity maintenance threshold function; After the first combined maintenance, considering the maintenance effect ε n T n is not 0; then the failure function expression in the second maintenance cycle is: Where T1 is the time required for the first maintenance, ε1 determines the maintenance cost according to the first maintenance method, and ε1 is regarded as a function of the running time t. The maintenance threshold function of the second cycle is: The above method is used to solve and obtain the maintenance threshold function of the second cycle; After each component maintenance, the correlation coefficient of the improvement factor ε is determined according to the maintenance method, and it is iterated on the failure function of the previous maintenance cycle, and then the maintenance threshold function is updated using the method that minimizes the maintenance cost per unit time.

Images