An emergency diesel engine starting performance evaluation method based on grey clustering

Abstract

This invention relates to a method for evaluating the starting performance of emergency diesel engines based on gray clustering. The method includes: establishing an evaluation index system for the starting performance of emergency diesel engines and determining the set of evaluation indices; determining the weight of each evaluation index based on the entropy weight method; dividing the starting performance of the emergency diesel engine into multiple gray classes, each corresponding to a performance level; establishing a whitening weight function for each evaluation index with respect to each gray class; collecting operational data from monthly tests of the emergency diesel engine to be evaluated, obtaining the observed values ​​of each evaluation index, and substituting these values ​​into the corresponding whitening weight function to calculate the whitening weight function value of each index with respect to each gray class; calculating the clustering coefficient of the evaluated object with respect to each gray class based on the whitening weight function value and the index weight; comparing the clustering coefficients of each gray class, assigning the starting performance of the emergency diesel engine to the gray class with the largest clustering coefficient, and completing the performance level evaluation.

Description

Technical Field

[0001] This invention relates to the field of nuclear power emergency diesel engine technology, and in particular to an emergency diesel engine start-up performance evaluation method based on grey clustering. Background Technology

[0002] Nuclear power plant emergency diesel generator sets are emergency equipment used in nuclear power plants. Their function is to provide reliable and independent emergency power to the nuclear power plant's safety system in extreme accident scenarios where both the main power source (external power grid) and backup power sources fail, ensuring safe reactor shutdown and protecting nuclear safety. According to nuclear safety guidelines for emergency diesel generators, they must be able to start and reach rated speed and be ready for loading within 10 seconds in an emergency, placing high demands on their starting performance. Therefore, monthly tests are necessary to regularly evaluate the starting performance of the emergency diesel generators and ensure their rapid start-up capability.

[0003] Currently, power plants typically assess and analyze the emergency diesel engine's starting performance during periodic testing by monitoring thermal parameters through the electronic control system and vibration parameters collected by offline vibration meters, combined with the auditory and tactile feedback from on-site personnel and the experience of experts, based on thresholds provided by the manufacturer. If the monitored data exceeds a fixed threshold, an alarm is issued, indicating that the starting performance fails to meet requirements.

[0004] However, due to the complex structure and numerous components of emergency diesel engines, current monitoring methods and assessment techniques are insufficient to achieve satisfactory monitoring and assessment results. On the one hand, they neglect the gradual deterioration of system startup performance, failing to accurately reflect the entire process from normal operation to failure. On the other hand, these methods isolate various parameters, ignoring the correlation between parameters and the importance of different indicators, thus failing to comprehensively evaluate the startup performance of emergency diesel engine sets. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an emergency diesel engine starting performance evaluation method based on gray clustering, in order to address the above-mentioned defects.

[0006] The technical solution adopted by this invention to solve its technical problem is: a method for evaluating the starting performance of an emergency diesel engine based on gray clustering, comprising the following steps: S1. Establish an evaluation index system for the starting performance of emergency diesel engines and determine the set of evaluation indicators; S2. Determine the weights of each evaluation index based on the entropy weight method; S3. Divide the starting performance of emergency diesel engines into multiple gray categories, with each gray category corresponding to a performance level; S4. Establish a whitening weight function for each evaluation index and for each gray class; the whitening weight function is used to describe the degree of membership of the index observation value to the gray class; S5. Collect the operating data of the emergency diesel engine to be evaluated during the monthly test, obtain the observed values ​​of each evaluation index, and substitute the observed values ​​into the corresponding whitening weight function to calculate the whitening weight function value of each index with respect to each gray class. S6. Based on the whitening weight function value and index weight, calculate the clustering coefficient of the object to be evaluated with respect to each gray class; S7. Compare the clustering coefficients of each gray class, assign the emergency diesel engine starting performance to the gray class with the largest clustering coefficient, and complete the performance level evaluation.

[0007] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, step S1 further includes: Based on the ideal relationship between the index values ​​and startup performance, each index is classified as positive, negative, or intermediate.

[0008] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, step S2 includes: Determine the corresponding normalization formula based on the type of each evaluation indicator, and then perform normalization processing. The information entropy of each evaluation indicator is calculated based on the normalized evaluation indicator data; The weight of each evaluation indicator is calculated based on its information entropy.

[0009] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, the step of determining the corresponding normalization formula according to the type of each evaluation index includes: For positive indicators, normalization is performed using Formula 1: (Formula 1) For negative indicators, normalization is performed using Formula 2: (Formula 2) For intermediate indicators, normalization is performed using Formula 3: (Formula 3) In the formula, i represents the i-th evaluation index. This represents the normalized value of the j-th data point for the i-th evaluation index. This represents the specific value of the j-th data point for the i-th evaluation indicator. This represents the sample data for the i-th evaluation index. This represents the optimal value for the moderate evaluation index i.

[0010] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, the step of calculating the information entropy of each evaluation index based on the normalized evaluation index data includes: The information entropy of the i-th evaluation index is calculated using the following formula: In the formula, This represents the information entropy of the i-th evaluation index. This represents the elements within the normalized data matrix, where n represents the number of indicators.

[0011] Furthermore, in the emergency diesel engine starting performance evaluation method based on grey clustering described in this invention, the step of calculating the corresponding weight based on the information entropy of each evaluation index includes: The weight of the i-th evaluation indicator is calculated using the following formula: In the formula, This represents the weight of the i-th evaluation indicator. ,and .

[0012] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, in step S5, the whitening weight function includes: For positive indicators, an upper limit measure whitening weight function is used; For negative indicators, a lower limit measure whitening weight function is used; For intermediate indicators, a moderate whitening weight function is adopted.

[0013] Furthermore, in the emergency diesel engine starting performance evaluation method based on gray clustering described in this invention, the gray clusters correspond to three performance levels: excellent, good, and poor. Specifically, an excellent emergency diesel engine starting performance level corresponds to a clustering coefficient value in the range of [0.8, 1]; a good starting performance level corresponds to a clustering coefficient value in the range of [0.4, 0.8]; and a poor starting performance level corresponds to a clustering coefficient value in the range of [0, 0.4].

[0014] Furthermore, in the emergency diesel engine starting performance evaluation method based on grey clustering described in this invention, the evaluation index set includes: At least one evaluation index is determined for each type of key characteristic; the key characteristics include work capacity, speed characteristics, and start-up time; wherein, work capacity includes evaluation indices: charge work and ignition work; speed characteristics include evaluation indices: speed increase that reflects the average speed change during the start-up process of the emergency diesel engine and speed fluctuation that reflects the overall working state during the start-up process of the emergency diesel engine; start-up time includes evaluation indices: time taken for the speed to reach the ignition speed from 0 and time taken to complete the start-up.

[0015] The gray clustering-based emergency diesel engine starting performance evaluation method of this invention has the following beneficial effects: This invention comprehensively considers the interaction and correlation of multiple evaluation indicators, rather than treating a single parameter in isolation. It reflects the differentiated contribution of different indicators to starting performance through a weight allocation mechanism, and achieves a holistic and comprehensive quantitative evaluation of emergency diesel engine starting performance based on the gray clustering algorithm. This invention can accurately identify intermediate transitional states of performance degradation, avoiding the abrupt defects of binary judgment of normal / fault conditions, and providing a basis for preventative maintenance decisions. It can also provide a scientific and reliable quantitative basis for subsequent condition-based maintenance and life prediction of emergency diesel engines. Attached Figure Description

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart illustrating the emergency diesel engine start-up performance evaluation method based on gray clustering provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a multi-level evaluation system for the starting performance of emergency diesel engines according to some embodiments of the present invention; Figure 3 These are impact signal diagrams corresponding to the air-pumping work and ignition work in some embodiments of the present invention; Figure 4 This is a speed diagram of the starting process of a certain type of emergency diesel engine according to some embodiments of the present invention; Figure 5 This is a schematic diagram of the instantaneous speed waveform of the emergency diesel engine within a single cycle in some embodiments of the present invention; Figure 6 This is a schematic diagram of a typical whitening weight function in some embodiments of the present invention; Figure 7 This is a schematic diagram of the moderate measure whitening weight function of some embodiments of the present invention; Figure 8 This is a schematic diagram of the upper limit measure whitening weight function of some embodiments of the present invention; Figure 9 This is a schematic diagram of the lower limit measure whitening weight function of some embodiments of the present invention. Detailed Implementation

[0017] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the invention are now described in detail with reference to the accompanying drawings. In the following description, specific details such as particular system structures and techniques are set forth for illustrative purposes rather than for limiting the scope of the invention, in order to provide a thorough understanding of the embodiments. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted to avoid unnecessary detail that could obscure the description of the invention.

[0018] In a preferred embodiment, reference Figure 1 The emergency diesel engine starting performance evaluation method based on gray clustering in this embodiment includes the following steps: S1. Establish an evaluation index system for the starting performance of emergency diesel engines and determine the set of evaluation indicators. This scheme mainly extracts evaluation indicators from three key characteristics of emergency diesel engines: work capacity, speed characteristics, and starting time, forming an evaluation system for the starting performance of emergency diesel engines. It can be understood that the evaluation index set in this embodiment includes: determining at least one evaluation indicator from each type of key characteristic. Key characteristics include work capacity, speed characteristics, and starting time. Work capacity includes evaluation indicators: charge work and ignition work. Speed ​​characteristics include evaluation indicators: the speed increase reflecting the average speed change during the emergency diesel engine starting process, and the speed fluctuation reflecting the overall working state during the emergency diesel engine starting process. Starting time includes evaluation indicators: the time taken for the speed to reach the ignition speed from 0, and the time taken to complete the start-up.

[0019] like Figure 2 As shown, the emergency diesel engine starting performance evaluation index system of this method can reflect the multi-level comprehensive evaluation method for analyzing the starting performance of emergency diesel generator sets. Since the starting performance of emergency diesel generators is affected by a variety of factors, it can be decomposed into key characteristics. Each key characteristic contains state parameters or evaluation indicators at the next level. This hierarchical structure provides the conditions for constructing a hierarchical analysis model. To make the evaluation results more accurate and truly reflect the starting performance of emergency diesel engines, the evaluation indicators are evaluated from the bottom up, and then each key characteristic is comprehensively evaluated from the bottom up, iterating until the evaluation result of the entire unit's starting performance is obtained. This embodiment determines the overall unit's starting performance (excellent, medium, or poor) from the bottom up by using each key starting performance characteristic and corresponding evaluation index, enabling timely early warning of unit anomalies and reducing the probability of accidents. For example, this scheme divides the evaluation of starting performance into three key characteristics: work capacity, speed characteristics, and starting time. Each key characteristic has its own evaluation index.

[0020] It is understandable that the angular domain signal of a diesel engine exhibits a periodic pattern, and its specific phase contains special physical meaning, which can reflect the true state of an emergency diesel engine. Therefore, this method starts from the angular domain signal to analyze evaluation indicators that can reflect the key characteristics of the emergency diesel engine's starting performance. The calculation methods for each key characteristic evaluation indicator are described in further detail below.

[0021] 1) Work capacity: The work capacity of the generator set during startup can be divided into two stages. The work capacity of each stage can well reflect the starting performance of the diesel engine. Figure 3 The diagrams show the impact signals corresponding to the inflator work and ignition work in some embodiments.

[0022] Injection power: The first stage is the injection stage. After the control system detects the unit start command, high-pressure starting air from the diesel generator rushes into the cylinder to drive the crankshaft and perform work, bringing the diesel engine speed to 60 rpm. Then, the diesel engine crankshaft is driven by both high-pressure air and the ignition of the cylinder. The quality of the injection power performance determines whether the diesel engine can start successfully within the required time and has the capacity to carry a load. Injection impact characteristics are extracted using cylinder head vibration signals as an indicator to evaluate the unit's injection power performance.

[0023] Root mean square value of cylinder head vibration during the power stroke: in, Let i be the vibration signal value at the i-th sampling point. The total number of sampling points for work done by the air pump.

[0024] Peak cylinder head vibration index during the air-fuel combustion period: in, This represents the maximum vibration signal value within 50° before and after the diesel engine exhaust valve opening phase. The root mean square value of cylinder head vibration during the combustion power period.

[0025] Ignition and Power: The second stage is the acceleration stage. When the speed reaches 350 rpm (ignition speed) or after 6 seconds of starting, the main start air valve will close. At this time, the diesel engine crankshaft is only driven by the cylinder ignition and power.

[0026] Root mean square value of cylinder head vibration during ignition: In the formula, The vibration signal value at the i-th sampling point during the ignition and power generation period is given. This represents the total number of sampling points during the fire-starting and work period.

[0027] Peak cylinder head vibration index during ignition and power stroke: In the formula, This represents the maximum vibration signal value within 50° before and after the top dead center phase of the diesel engine. The root mean square value of cylinder head vibration during ignition and power stroke.

[0028] 2) Rotational speed characteristics: such as Figure 4 As shown, during the startup process of a nuclear power plant emergency diesel engine, the engine speed continuously increases from zero until it reaches the rated speed. Its instantaneous speed can reflect the startup performance of the emergency diesel engine to a certain extent.

[0029] Average speed variation pattern: During the startup process of the nuclear power emergency diesel engine, the instantaneous speed waveform generally shows an upward trend, reflecting the rapid speed increase throughout the cycle. This information is called the speed increase, which is calculated from the average instantaneous speed increase rate. Specifically, the formula for calculating the speed increase is as follows: In the formula, To start and complete the corresponding speed, Time taken for startup to complete.

[0030] Overall operating status during startup: Figure 5 This is a waveform diagram of the instantaneous speed of the emergency diesel engine within a single working cycle after startup. As can be seen from the graph, the instantaneous speed exhibits a periodic fluctuation trend within a certain range in each cycle, reflecting the overall operating state of the unit during startup. This can be used as an evaluation indicator of startup performance, and this information is called the speed fluctuation, which can be obtained by the difference between the mean values ​​of the upper and lower envelopes of the instantaneous speed. Specifically, the formula for calculating the speed fluctuation is as follows: in, Let be the upper envelope of the instantaneous rotational speed within the i-th period. Let be the lower envelope of the instantaneous rotational speed within the i-th period.

[0031] 3) Startup time: Since emergency startup has strict time requirements, the startup cycle time collected by the detection system can be used as a measure of startup performance.

[0032] Time to reach 350 rpm (ignition speed): According to the starting requirements of emergency diesel engines, if the engine speed does not reach 350 rpm within 6 seconds of starting, the start is considered a failure. Therefore, the time it takes for the engine speed to reach 350 rpm (ignition speed) during the start-up process of an emergency diesel engine is an important indicator for evaluating starting performance.

[0033] Start-up completion time: According to the emergency diesel engine start-up requirements, the diesel engine must complete the emergency start within 10 seconds. If it cannot complete the start-up within the specified time, it is considered a failure. Therefore, the start-up completion time of the emergency diesel engine is used as an indicator to evaluate the start-up performance.

[0034] In some embodiments, step S1 can further classify the indicators into positive, negative, or intermediate types based on the ideal relationship between the indicator values ​​and startup performance. A positive indicator is one whose larger value indicates that the evaluated object is closer to the optimal target and has better performance. A negative indicator is one whose smaller value indicates that the evaluated object is closer to the optimal target and has better performance. An intermediate indicator is one whose value is closer to a specific ideal value (standard value) and has better performance; values ​​that are too large or too small are considered undesirable.

[0035] S2. Based on the entropy weight method, determine the weights of each evaluation indicator. It can be understood that for each evaluation indicator under each key characteristic, the weight reflects the relative importance of different evaluation indicators in the evaluation process. Therefore, whether the weight allocation is reasonable will directly affect the final evaluation result. This patent uses the entropy weight method to objectively assign weights to the indicators of each key characteristic in the startup performance evaluation model. This method can take into account the degree of variation of each evaluation indicator in the actual operation process and has the advantage of lower data requirements.

[0036] In some embodiments, step S2 includes: determining the corresponding normalization formula according to the type of each evaluation index, and performing normalization processing. The information entropy of each evaluation index is calculated based on the normalized evaluation index data. The corresponding weight is calculated based on the information entropy of each evaluation index. It should be noted that the order of step S2, i.e., determining the weights of each evaluation index, and the step of constructing the whitening weight function, can be interchanged. In this scheme, the execution / implementation order of certain steps can be adjusted according to specific actual needs, and is not limited here; all such adjustments fall within the protection scope of this embodiment.

[0037] It is understandable that this scheme treats the disorder of uncertain information in the emergency diesel generator set starting performance evaluation system as entropy. For each key characteristic of emergency diesel engine starting performance, the index and sample size are n and m respectively. Then the original data matrix X is: Entropy is a variable that ranges from 0 to 1. However, the various evaluation indicators (including vibration acceleration, rotational speed, and start-up time) have different dimensions and do not have standardized maximum and minimum values. Therefore, the raw data of each evaluation indicator of the emergency diesel generator set's start-up characteristics should be preprocessed and normalized before calculation.

[0038] Specifically, suppose there are n evaluation indicators for a certain key characteristic. ,in 'm' represents the sample size for a specific evaluation indicator, i.e., the amount of data. Based on the differences in how evaluation indicators reflect the starting performance of emergency diesel engines, each indicator is divided into three types: positive, negative, and intermediate. Assume that the normalized values ​​for each parameter are... The normalization formulas for the three types of data are as follows: For positive indicators, normalization is performed using Formula 1: (Formula 1) For negative indicators, normalization is performed using Formula 2: (Formula 2) For intermediate indicators, normalization is performed using Formula 3: (Formula 3) In the formula, i represents the i-th evaluation index. This represents the normalized value of the j-th data point for the i-th evaluation index. This represents the specific value of the j-th data point for the i-th evaluation indicator. This represents the sample data for the i-th evaluation index. This represents the optimal value for the moderate evaluation index i.

[0039] After normalization, the matrix is ​​obtained The information entropy of the i-th evaluation index is calculated using the following formula: In the formula, This represents the information entropy of the i-th evaluation index. This represents the elements within the normalized data matrix, where n represents the number of indicators.

[0040] The weight of the i-th evaluation indicator is calculated using the following formula: In the formula, This represents the weight of the i-th evaluation indicator. ,and .

[0041] This embodiment can determine the weight percentage of each evaluation index in the key characteristic assessment of the corresponding diesel generator set based on the above steps. .

[0042] S3. Divide the starting performance of the emergency diesel engine into multiple gray classes, each corresponding to a performance level. For example, a gray class can correspond to three performance levels: excellent, good, and poor. The excellent starting performance level corresponds to a clustering coefficient in the range [0.8, 1]. The good starting performance level corresponds to a clustering coefficient in the range [0.4, 0.8]. The poor starting performance level corresponds to a clustering coefficient in the range [0, 0.4].

[0043] S4. Establish a whitening weight function for each evaluation index and for each gray class. The whitening weight function is used to describe the degree to which the index observation belongs to the gray class.

[0044] S5. Collect the operating data of the emergency diesel engine to be evaluated during the monthly test period, obtain the observed values ​​of each evaluation index, and substitute the observed values ​​into the corresponding whitening weight function to calculate the whitening weight function value of each index with respect to each gray class. It can be understood that, considering the characteristics of emergency diesel engines with small sample size and scarce information, this embodiment can effectively complete the evaluation of the overall engine starting performance through multi-dimensional and multi-parameter fusion analysis using only the operating data monitored during the monthly test period.

[0045] S6. Calculate the clustering coefficients of the object to be evaluated for each gray class based on the whitening weight function value and the index weight.

[0046] S7. Compare the clustering coefficients of each gray class, assign the emergency diesel engine starting performance to the gray class with the largest clustering coefficient, and complete the performance level assessment.

[0047] It is understandable that the grey clustering method can define each grey class of diesel engine starting performance by studying the basic theory of emergency diesel engines, determine the whitening weight function of each grey class by analyzing the existing information in the system, and finally import the known information into the function to calculate the calculation result of each grey class, thereby solving the uncertainty problem and completing the identification of the starting performance state.

[0048] Specifically, in step S5, the whitening weight function includes: for positive indicators, an upper limit whitening weight function is used; for negative indicators, a lower limit whitening weight function is used; and for intermediate indicators, a moderate whitening weight function or a typical whitening weight function is used.

[0049] A typical whitening weight function has j indices and k gray classes. , , , Four inflection points of the function, the function expression is denoted as The diagram is as follows Figure 6 As shown.

[0050] The moderate whitening weight function has j indicators and k gray levels. , , Three inflection points of the function. The function expression is denoted as... The diagram is as follows Figure 7 As shown.

[0051] The upper limit measure whitening weight function has j indices and k gray classes. , Two inflection points of a function. The function expression is denoted as... The diagram is as follows Figure 8 As shown.

[0052] The lower limit measure whitening weight function has j indices and k gray classes. , Two inflection points of a function. The function expression is denoted as... The diagram is as follows Figure 9 As shown.

[0053] The ultimate goal of grey clustering is to calculate the grey clustering coefficients after processing with a whitening weight function, in order to determine the state level to which the emergency diesel engine's starting performance belongs. If... Let be the normalized value of a cluster object i (e.g., cluster object i is an emergency diesel engine) with respect to evaluation index j, and It is the weight of the evaluation index j, and satisfies Then it is called Let be the clustering coefficient of cluster object i in the startup performance state belonging to the k-th state level, and call it . Let be the clustering coefficient vector. And we have: Then the startup performance state of cluster object i belongs to the corresponding state level. .

[0054] This application proposes a method for extracting evaluation indicators for the starting performance of emergency diesel generators. It comprehensively considers the interaction and correlation of multiple evaluation indicators, rather than treating a single parameter in isolation. A weighting mechanism reflects the differentiated contribution of different indicators to starting performance, thereby achieving a holistic and comprehensive quantitative evaluation of emergency diesel generator starting performance. This method can accurately identify intermediate transitional states of performance degradation, avoiding the abrupt changes inherent in binary judgments of normal / fault conditions, and providing a basis for preventative maintenance decisions. Based on a grey clustering algorithm, the evaluation can intuitively display the unit's starting performance status level, reducing the workload and training costs for on-site inspection personnel. When starting performance is poor, it alerts on-site technicians to conduct troubleshooting, preventing emergency diesel generator malfunctions, avoiding external power supply failures caused by complete emergency diesel generator failures, preventing significant economic losses from reactor shutdowns, and eliminating the harm that nuclear power plant accidents can cause to society.

[0055] In another preferred embodiment, the emergency diesel engine monitoring and fault diagnosis system of this embodiment integrates the gray clustering-based emergency diesel engine start-up performance evaluation method of the above embodiment. Specifically, the gray clustering-based emergency diesel engine start-up performance evaluation method of the above embodiment can be integrated into a cloud platform through an algorithm. During monthly testing, emergency diesel engine operating data is collected, start-up performance evaluation indicators are extracted through computing modules deployed at the edge, and transmitted to the cloud computing module. The start-up performance status level is then determined through the start-up performance evaluation algorithm.

[0056] The emergency diesel engine monitoring and fault diagnosis system in this embodiment adopts the gray clustering-based emergency diesel engine starting performance evaluation method described in the previous embodiment. This method comprehensively considers the interaction and correlation of multiple evaluation indicators, reflects the differentiated contribution of different indicators to starting performance through a weight allocation mechanism, and achieves a holistic and comprehensive quantitative evaluation of emergency diesel engine starting performance based on the gray clustering algorithm. The gray clustering algorithm accurately identifies intermediate transitional states of performance degradation, avoiding the abrupt defects of binary judgment between normal and fault states, and providing a basis for preventative maintenance decisions. It also provides a scientific and reliable quantitative basis for subsequent emergency diesel engine condition-based maintenance and life prediction.

[0057] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0058] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0059] It is understood that the above embodiments only illustrate preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. Therefore, all equivalent transformations and modifications made with respect to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A method for evaluating the starting performance of emergency diesel engines based on grey clustering, characterized in that, Includes the following steps: S1. Establish an evaluation index system for the starting performance of emergency diesel engines and determine the set of evaluation indicators; S2. Determine the weights of each evaluation index based on the entropy weight method; S3. Divide the starting performance of emergency diesel engines into multiple gray categories, with each gray category corresponding to a performance level; S4. Establish a whitening weight function for each evaluation index and for each gray class; the whitening weight function is used to describe the degree of membership of the index observation value to the gray class; S5. Collect the operating data of the emergency diesel engine to be evaluated during the monthly test, obtain the observed values ​​of each evaluation index, and substitute the observed values ​​into the corresponding whitening weight function to calculate the whitening weight function value of each index with respect to each gray class. S6. Based on the whitening weight function value and index weight, calculate the clustering coefficient of the object to be evaluated with respect to each gray class; S7. Compare the clustering coefficients of each gray class, assign the emergency diesel engine starting performance to the gray class with the largest clustering coefficient, and complete the performance level evaluation.

2. The method for evaluating the starting performance of emergency diesel engines based on gray clustering according to claim 1, characterized in that, Step S1 also includes: Based on the ideal relationship between the index values ​​and startup performance, each index is classified as positive, negative, or intermediate.

3. The emergency diesel engine starting performance evaluation method based on gray clustering according to claim 2, characterized in that, Step S2 includes: Determine the corresponding normalization formula based on the type of each evaluation indicator, and then perform normalization processing. The information entropy of each evaluation indicator is calculated based on the normalized evaluation indicator data; The weight of each evaluation indicator is calculated based on its information entropy.

4. The emergency diesel engine starting performance evaluation method based on grey clustering according to claim 3, characterized in that, The step of determining the corresponding normalization formula based on the type of each evaluation indicator includes: For positive indicators, normalization is performed using Formula 1: (Official 1) For negative indicators, normalization is performed using Formula 2: (Official 2) For intermediate indicators, normalization is performed using Formula 3: (Official 3) In the formula, i represents the i-th evaluation index. This represents the normalized value of the j-th data point for the i-th evaluation index. This represents the specific value of the j-th data point for the i-th evaluation indicator. This represents the sample data for the i-th evaluation indicator. This represents the optimal value for the moderate evaluation index i.

5. The method for evaluating the starting performance of emergency diesel engines based on gray clustering according to claim 4, characterized in that, The information entropy of each evaluation indicator calculated based on the normalized evaluation indicator data in the step described above includes: The information entropy of the i-th evaluation index is calculated using the following formula: In the formula, This represents the information entropy of the i-th evaluation index. This represents the elements within the normalized data matrix, where n represents the number of indicators.

6. The method for evaluating the starting performance of emergency diesel engines based on gray clustering according to claim 5, characterized in that, The step of calculating the corresponding weight based on the information entropy of each evaluation indicator includes: The weight of the i-th evaluation indicator is calculated using the following formula: In the formula, This represents the weight of the i-th evaluation indicator. ,and .

7. The method for evaluating the starting performance of emergency diesel engines based on gray clustering according to claim 2, characterized in that, In step S5, the whitening weight function includes: For positive indicators, an upper limit measure whitening weight function is used; For negative indicators, a lower limit measure whitening weight function is used; For intermediate indicators, a moderate whitening weight function is adopted.

8. The grey clustering based emergency diesel engine starting performance evaluation method according to claim 1, characterized in that, The gray class corresponds to three performance levels: excellent, good, and poor. Among them, the emergency diesel engine starting performance level is excellent, with a clustering coefficient value range of [0.8, 1]; the starting performance level is good, with a clustering coefficient value range of [0.4, 0.8]; and the starting performance level is poor, with a clustering coefficient value range of [0, 0.4].

9. The grey clustering based emergency diesel engine starting performance evaluation method according to claim 1, characterized in that, The evaluation index set includes: At least one evaluation index is determined for each type of key characteristic; the key characteristics include work capacity, speed characteristics, and start-up time; wherein, work capacity includes evaluation indices: charge work and ignition work; speed characteristics include evaluation indices: speed increase that reflects the average speed change during the start-up process of the emergency diesel engine and speed fluctuation that reflects the overall working state during the start-up process of the emergency diesel engine; start-up time includes evaluation indices: time taken for the speed to reach the ignition speed from 0 and time taken to complete the start-up.

Images