Method for evaluating working reliability of generator set of ship power station

Abstract

The invention relates to a method for evaluating the working reliability of a ship power station generator set, and the method comprises the steps: firstly constructing evaluation indexes according to the working characteristics of the ship power station generator set, wherein the evaluation indexes comprise a steady-state speed regulation rate, a transient speed regulation rate, a transient voltage change rate, stabilization time, an active distribution error, a reactive distribution error and a harmonic distortion rate; constructing a generator set evaluation matrix by using the evaluation indexes in cooperation with index weights and state data of multiple tests, then constructing a positive ideal matrix by using factory inspection report data of the ship power station generator set, and constructing a negative ideal matrix by using historical worst state data of the generator set stored in a database; and calculating the similarity between the evaluation matrix and the positive and negative ideal matrix by using a projection algorithm specially designed for the ship power station generator set, further calculating a reliability coefficient according to the positive and negative similarity, and finally judging the working reliability of the ship power station generator set according to the value range of the coefficient.

Description

Technical Field

[0001] The invention belongs to the technical field of ship power station reliability assessment, and in particular relates to a method for assessing the working reliability of a ship power station generator. Background Art

[0002] As a core component of the ship's power system, the working reliability of the ship's power station generator set is directly related to the stability of the ship's power supply, navigation safety and mission execution capabilities. Especially in ocean voyages, complex sea conditions or emergency conditions, key indicators such as the dynamic response performance, power distribution accuracy and power quality of the generator set need to meet strict technical requirements. Once a generator set fails or its performance deteriorates, it may cause the entire ship's power system to be paralyzed, endangering personnel, property, and environmental safety. For example, the direct cause of the Dali container collision with the Francis Scott Key Bridge on March 26, 2024 was the multiple power outages of the Dali ship. Therefore, a scientific and accurate assessment of the working reliability of the ship's power station generator set is an important technical basis for the design, maintenance and fault warning of the ship's power system.

[0003] At present, the reliability assessment methods of ship power station generator sets mainly include traditional expert assessment methods, empirical methods, and statistical analysis methods based on operating data. Although these methods can reflect the reliability of generator sets to a certain extent, they have the following limitations: First, the assessment dimension is single, data utilization is insufficient, and the traditional assessment process mostly relies on single test data or static thresholds. There is a lack of effective fusion of historical operating data (such as historical worst state data) and comparative analysis of factory inspection benchmarks, which leads to the assessment results being greatly disturbed by accidental factors and insufficient robustness; second, the algorithm has poor adaptability. The existing projection algorithms are mostly designed for general scenarios, and are not optimized for the high dynamic and multi-disturbance working characteristics of ship power station generator sets. It is difficult to accurately quantify the similarity between the assessment matrix and the ideal matrix, which affects the calculation accuracy of the reliability coefficient.

[0004] In addition, the operating environment of ship generator sets is complex and changeable (such as sudden load changes, mechanical vibrations, temperature and humidity fluctuations, etc.), and their performance degradation often presents nonlinear characteristics. Traditional methods are difficult to capture early performance degradation trends in a timely manner, resulting in delayed fault warning. Therefore, a multi-dimensional, data-driven, and algorithmically adaptable reliability assessment method is urgently needed to improve the state monitoring capability and maintenance decision-making efficiency of ship power station generator sets, and provide technical support for the safe and stable operation of ship power systems.

[0005] The present invention proposes a method for evaluating the working reliability of a generator set in a ship power station, aiming to comprehensively evaluate the working reliability of the generator set by building an accurate quantitative model and adopting an innovative projection algorithm, comprehensively considering multi-dimensional factors such as the steady-state and transient performance, power distribution error and harmonic distortion, and providing a more scientific, objective and accurate evaluation method. Summary of the invention

[0006] The present invention is a method for evaluating the reliability of a generator set in a ship power station, which is used for evaluating the reliability of a ship power station and specifically comprises the following steps:

[0007] Step 1: Construct evaluation indicators and evaluation matrix

[0008] According to the working characteristics of the ship power station generator set, an evaluation index set Z = {z j |j∈N}, where N={1,2,…,7}, and z1=steady-state rate regulation, z2=transient rate regulation, z3=transient voltage change rate, z4=stabilization time, z5=active power distribution error, z6=reactive power distribution error, z7=harmonic distortion rate, construct two subsets P and I of Z, satisfying P∪I=Z and If z j ∈P, the index is expressed using exact numbers (j=1,5,6,7). If z j ∈I, the index is expressed using interval numbers (j = 2, 3, 4), and the weight vectors corresponding to the above indicators are w = (w1, w2, …, w j ), satisfying 0≤w j ≤1(j∈N) and Among them, w1 = 0.1, w2 = 0.2, w3 = 0.2, w4 = 0.2, w5 = 0.1, w6 = 0.1, w7 = 0.1, all the generator sets to be evaluated constitute a set A = {a k |k∈L}, where L={1,2,…,l}, and the test conditions constitute a set G={g i |i∈M}, where M={1,2,…,m}, for the kth unit a in set A, k Perform m tests under different working conditions to obtain the original matrix X k The definition is as follows:

[0009]

[0010] In formula (1), the matrix elements represents the data of the jth indicator of the kth unit during the i-th operating condition test (i∈M, j∈N, k∈L). If, z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, Is the upper bound of the interval number, and then use the original matrix X k Combine the indicator weight vector w to construct the evaluation matrix Y k as follows:

[0011]

[0012] In formula (2), if z j ∈P, then If z j ∈I, then

[0013] Step 2: Construct positive and negative ideal matrices

[0014] The factory reference matrix Y is obtained by using the factory inspection report data of the ship power station generator set according to the calculation method of (1) and (2). * as follows:

[0015]

[0016] In formula (3), the matrix elements The calculation process is the same as step 1, except that the data of the original matrix comes from the factory inspection report of the generator set. If z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number. Next, construct the positive ideal matrix Y + as follows:

[0017]

[0018] In formula (4), the matrix elements Represents traversal of all The optimal value obtained after , since the elements of the evaluation index set are all cost-type, that is, the smaller the value, the better the evaluation result. j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number;

[0019] Negative ideal matrix Y - It is constructed by iterative method, that is, the negative ideal matrix Y - The initial value of Y +The same is stored in the database. Each test value is subsequently compared with the corresponding historical record stored in the database. If the test value is greater than the historical record value (indicating that the current test result is poor), the current test value replaces the original historical record and constructs the negative ideal matrix Y - as follows:

[0020]

[0021] In formula (5), the matrix elements Represents traversing all current evaluation matrices Y k and the worst value obtained after comparing the historical records stored in the database, if z j ∈P, then is an exact number, where Is the historical record stored in the database, if Replace and refresh the database storage record; if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number, and Is the historical record stored in the database, if or Then replace and refresh the corresponding database storage record;

[0022] Step 3: Calculate the similarity between the evaluation matrix and the positive and negative ideal matrices

[0023] Design of projection algorithm for ship power station generator set to calculate evaluation matrix Y k With the positive ideal matrix Y + The similarity of p and the evaluation matrix Y k With the negative ideal matrix Y - The similarity of n as follows:

[0024]

[0025] In equations (6) and (7),

[0026] Step 4: Calculate the reliability coefficient

[0027] The reliability coefficient ρ is calculated by the following formula (8):

[0028]

[0029] Step 5: Conduct reliability assessment of ship power station generator sets based on the quantitative characterization of reliability coefficients

[0030] The reliability of the ship power station generator set is divided into three levels: when 0.8≤ρ≤1, the system is in the reliability level I state, the working performance of the generator set is in the stable operating range, the operating parameters meet the requirements of the classification society, and the normal working conditions can be maintained without active intervention; when 0.6≤ρ<0.8, the system is downgraded to the reliability level II state, and the working performance of the generator set shows a degradation trend. Preventive maintenance strategies need to be implemented, including but not limited to power station system parameter inspection and calibration, relay protection function verification, electromechanical coordination and linkage, electrical insulation and temperature rise detection, etc.; when 0≤ρ<0.6, the system enters the reliability level III critical failure state, and it is necessary to immediately perform graded unloading and start the emergency generator, and further carry out shutdown maintenance, focusing on the voltage regulator (AVR), speed governor, excitation system and power management system (PMS) diagnostic tests, and thus build a multi-level early warning mechanism to avoid the risk of power outage accidents on the entire ship;

[0031] The beneficial effects of the present invention are:

[0032] (1) By constructing a multi-dimensional evaluation index system covering steady-state speed regulation, transient speed regulation, transient voltage change rate, stabilization time, active power distribution error, reactive power distribution error and harmonic distortion rate, it is possible to comprehensively evaluate the performance of ship power station generator sets under different working conditions and reflect the working status and reliability of the generator sets more comprehensively and accurately;

[0033] (2) The factory inspection report data of the ship power station generator set is combined with the historical worst state data to construct a positive and negative ideal matrix, and the similarity is calculated using a projection algorithm designed specifically for the generator set, which further improves the accuracy of the reliability assessment.

[0034] (3) This method is applicable to all types of ship power station generator sets and has strong versatility and promotion value. It can provide an important reference for the design, manufacture, operation and maintenance of ship power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention relates to a flow chart of a method for evaluating the reliability of a generator set in a ship power station; DETAILED DESCRIPTION

[0036] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings:

[0037] Step 1: Construct evaluation indicators and evaluation matrix

[0038] According to the working characteristics of the ship power station generator set, an evaluation index set Z = {z j|j∈N}, where N={1,2,…,7}, and z1=steady-state rate regulation, z2=transient rate regulation, z3=transient voltage change rate, z4=stabilization time, z5=active power distribution error, z6=reactive power distribution error, z7=harmonic distortion rate, construct two subsets P and I of Z, satisfying P∪I=Z and If z j ∈P, the index is expressed using exact numbers (j=1,5,6,7). If z j ∈I, the index is expressed using interval numbers (j = 2, 3, 4), and the weight vectors corresponding to the above indicators are w = (w1, w2, …, w j ), satisfying 0≤w j ≤1(j∈N) and Among them, w1 = 0.1, w2 = 0.2, w3 = 0.2, w4 = 0.2, w5 = 0.1, w6 = 0.1, w7 = 0.1, all the generator sets to be evaluated constitute a set A = {a k |k∈L}, where L={1,2,…,l}, and the test conditions constitute a set G={g i |i∈M}, where M={1,2,…,m}, for the kth unit a in set A, k Perform m tests under different working conditions to obtain the original matrix X k The definition is as follows:

[0039]

[0040] In formula (1), the matrix elements represents the data of the jth indicator of the kth unit during the i-th operating condition test (i∈M, j∈N, k∈L). If, z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, Is the upper bound of the interval number, and then use the original matrix X k Combine the indicator weight vector w to construct the evaluation matrix Y k as follows:

[0041]

[0042] In formula (2), if z j ∈P, then If z j ∈I, then

[0043] Step 2: Construct positive and negative ideal matrices

[0044] The factory reference matrix Y is obtained by using the factory inspection report data of the ship power station generator set according to the calculation method of (1) and (2). * as follows:

[0045]

[0046] In formula (3), the matrix elements The calculation process is the same as step 1, except that the data of the original matrix comes from the factory inspection report of the generator set. If z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number. Next, construct the positive ideal matrix Y + as follows:

[0047]

[0048] In formula (4), the matrix elements Represents traversal of all The optimal value obtained after , since the elements of the evaluation index set are all cost-type, that is, the smaller the value, the better the evaluation result. j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number;

[0049] Negative ideal matrix Y - It is constructed by iterative method, that is, the negative ideal matrix Y - The initial value of Y + The same is stored in the database. Each test value is subsequently compared with the corresponding historical record stored in the database. If the test value is greater than the historical record value (indicating that the current test result is poor), the current test value replaces the original historical record and constructs the negative ideal matrix Y - as follows:

[0050]

[0051] In formula (5), the matrix elements Represents traversing all current evaluation matrices Y k and the worst value obtained after comparing the historical records stored in the database, if z j ∈P, then is an exact number, where Is the historical record stored in the database, if Replace and refresh the database storage record; if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number, and Is the historical record stored in the database, if or Then replace and refresh the corresponding database storage record;

[0052] Step 3: Calculate the similarity between the evaluation matrix and the positive and negative ideal matrices

[0053] Design of projection algorithm for ship power station generator set to calculate evaluation matrix Y k With the positive ideal matrix Y + The similarity of p and the evaluation matrix Y k With the negative ideal matrix Y - The similarity of n as follows:

[0054]

[0055] In equations (6) and (7),

[0056] Step 4: Calculate the reliability coefficient

[0057] The reliability coefficient ρ is calculated by the following formula (8):

[0058]

[0059] Step 5: Conduct reliability assessment of ship power station generator sets based on the quantitative characterization of reliability coefficients

[0060] The reliability of ship power station generator sets is divided into three levels: when 0.8≤ρ≤1, the system is in reliability level I state, the working performance of the generator set is in a stable operating range, the operating parameters meet the requirements of the classification society, and the normal operating conditions can be maintained without active intervention; when 0.6≤ρ<0.8, the system is downgraded to reliability level II state, and the working performance of the generator set shows a degradation trend. It is necessary to implement preventive maintenance strategies, including but not limited to power station system parameter inspection and calibration, relay protection function verification, electromechanical coordination and linkage, electrical insulation and temperature rise detection, etc.; when 0≤ρ<0.6, the system enters reliability level III critical failure state, and it is necessary to immediately perform graded unloading and start the emergency generator, and further carry out shutdown maintenance, focusing on the diagnosis and testing of the voltage regulator (AVR), speed governor, excitation system and power management system (PMS). Based on this, a multi-level early warning mechanism is constructed to avoid the risk of power outage accidents on the entire ship.

Claims

1. A method for evaluating the reliability of a generator set in a ship power station, characterized in that: The following steps are involved: Step 1: Construct evaluation indicators and evaluation matrix According to the working characteristics of the ship power station generator set, an evaluation index set Z = {z j |j∈N}, where N={1,2,…,7}, and z1=steady-state rate regulation, z2=transient rate regulation, z3=transient voltage change rate, z4=stabilization time, z5=active power distribution error, z6=reactive power distribution error, z7=harmonic distortion rate, construct two subsets P and I of Z, satisfying P∪I=Z and If z j ∈P, the index is expressed using exact numbers (j=1,5,6,7). If z j ∈I, the index is expressed using interval numbers (j = 2, 3, 4), and the weight vectors corresponding to the above indicators are w = (w1, w2, …, w j ), satisfying 0≤w j ≤1(j∈N) and Among them, w1 = 0.1, w2 = 0.2, w3 = 0.2, w4 = 0.2, w5 = 0.1, w6 = 0.1, w7 = 0.1, all the generator sets to be evaluated constitute a set A = {a k |k∈L}, where L={1,2,…,l}, and the test conditions constitute a set G={g i |i∈M}, where M={1,2,…,m}, for the kth unit a in set A, k Perform m tests under different working conditions to obtain the original matrix X k The definition is as follows: In formula (1), the matrix elements represents the data of the jth indicator of the kth unit during the i-th operating condition test (i∈M, j∈N, k∈L). If, z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, Is the upper bound of the interval number, and then use the original matrix X k Combine the indicator weight vector w to construct the evaluation matrix Y k as follows: In formula (2), if z j ∈P, then If z j ∈I, then Step 2: Construct positive and negative ideal matrices The factory reference matrix Y is obtained by using the factory inspection report data of the ship power station generator set according to the calculation method of (1) and (2). * as follows: In formula (3), the matrix elements The calculation process is the same as step 1, except that the data of the original matrix comes from the factory inspection report of the generator set. If z j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number. Next, construct the positive ideal matrix Y + as follows: In formula (4), the matrix elements Represents traversal of all The optimal value obtained after , since the elements of the evaluation index set are all cost-type, that is, the smaller the value, the better the evaluation result. j ∈P, then is an exact number, if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number; Negative ideal matrix Y - It is constructed by iterative method, that is, the negative ideal matrix Y - The initial value of Y + The same is stored in the database. Each test value is subsequently compared with the corresponding historical record stored in the database. If the test value is greater than the historical record value (indicating that the current test result is poor), the current test value replaces the original historical record and constructs the negative ideal matrix Y - as follows: In formula (5), the matrix elements Represents traversing all current evaluation matrices Y k and the worst value obtained after comparing the historical records stored in the database, if z j ∈P, then is an exact number, where Is the historical record stored in the database, if Replace and refresh the database storage record; if z j ∈I, then is an interval number, where is the lower bound of the interval number, is the upper bound of the interval number, and Is the historical record stored in the database, if or Then replace and refresh the corresponding database storage record; Step 3: Calculate the similarity between the evaluation matrix and the positive and negative ideal matrices Design of projection algorithm for ship power station generator set to calculate evaluation matrix Y k With the positive ideal matrix Y + The similarity of p and the evaluation matrix Y k With the negative ideal matrix Y - The similarity of n as follows: In equations (6) and (7), Step 4: Calculate the reliability coefficient The reliability coefficient ρ is calculated by the following formula (8): Step 5: Conduct reliability assessment of ship power station generator sets based on the quantitative characterization of reliability coefficients The reliability of ship power station generator sets is divided into three levels: when 0.8≤ρ≤1, the system is in reliability level I state, the working performance of the generator set is in a stable operating range, the operating parameters meet the requirements of classification society regulations, and the normal operating conditions can be maintained without active intervention; when 0.6≤ρ<0.8, the system is downgraded to reliability level II state, and the working performance of the generator set shows a degradation trend. It is necessary to implement preventive maintenance strategies, including but not limited to power station system parameter inspection and calibration, relay protection function verification, electromechanical coordination and linkage, electrical insulation and temperature rise detection, etc.; when 0≤ρ<0.6, the system enters reliability level III critical failure state, and it is necessary to immediately perform graded unloading and start the emergency generator, and further carry out shutdown maintenance, focusing on the diagnosis and testing of the voltage regulator (AVR), speed governor, excitation system and power management system (PMS). Based on this, a multi-level early warning mechanism is constructed to avoid the risk of power outage accidents on the entire ship.

Images