Fault detection rate distribution method considering multi-factor integration

Abstract

The invention discloses a fault detection rate allocation method considering multi-factor integration, belongs to the technical field of testability design, and is suitable for power systems of objects such as new energy automobiles and hybrid electric vehicles. The method comprises the following steps: considering the characteristics of a power system, taking the fault occurrence frequency, the electronization degree, the state monitoring coverage degree and the simple judgment fault proportion as fault detection rate distribution influence factors, and carrying out data preparation and normalization processing on the influence factors; determining the weight of each influence factor through multi-expert scoring; performing weighted summation on the influence factor data of each distribution object to obtain a comprehensive distribution coefficient of each distribution object; and completing fault detection rate index distribution according to the comprehensive distribution coefficient and the distribution model. According to the method, the problem of failure rate information loss in the system is effectively solved, the feasibility of engineering implementation is remarkably improved, the failure detection requirement of each distribution object in the system can be reflected more accurately, and good adaptability and stability are achieved.

Description

Technical Field

[0001] This invention relates to a fault detection rate allocation method that considers multiple factors, belonging to the field of testable design technology, and is applicable to the power systems of new energy vehicles, hybrid vehicles, and other similar applications. Background Technology

[0002] Testability is a design characteristic of a product that allows for timely and accurate determination of its state (operable, inoperable, or degree of performance degradation) and isolation of internal faults. Similar to characteristics such as reliability, maintainability, supportability, and safety, testability is one of the general quality characteristics of equipment. Testability allocation is a crucial step in testability design. Its main task is to decompose the quantitative testability requirements specified by the system, such as the failure detection rate, layer by layer to each subsystem and component to ensure that the overall design achieves the expected failure detection effect. Currently, commonly used allocation methods include equal-value allocation, failure rate allocation, allocation considering both failure rate and cost, and comprehensive weighted allocation.

[0003] In the practical application of test design for power systems, the applicability of traditional allocation methods faces challenges. On the one hand, power systems generally lack subsystem-level failure rate information, failing to meet the reliance of traditional methods on basic data, thus hindering the effective progress of allocation work. On the other hand, existing methods allocate based on general dimensions such as failure rate, failure impact, maintenance, and cost, failing to reflect the system characteristics of power systems in terms of electronics, condition monitoring coverage, and the ease of fault identification, and making it difficult to truly reflect the relative roles and priorities of different subsystems in fault detection. These problems make the direct application of traditional methods in power system fault detection rate allocation lacking in engineering feasibility and accuracy, urgently requiring innovation in fault detection rate allocation methods for such complex systems. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a fault detection rate allocation method that considers multiple factors. Based on the determination of testability requirements and allocation objects, the method considers the characteristics of the power system to set influencing factors for fault detection rate allocation and prepares data. It conducts independent scoring by multiple experts to determine the weight of each influencing factor and to determine the comprehensive allocation coefficient. Finally, it completes the allocation of fault detection rate indicators based on the allocation model.

[0005] This invention provides a fault detection rate allocation method that considers multiple factors. For a power system to which fault detection rate allocation is to be performed, the product hierarchy of the power system is determined, and fault detection rate allocation is performed hierarchically from top to bottom. For each product level, this invention takes it as the current allocation object, and then performs the following steps:

[0006] Step 1: Organize and acquire data on factors influencing the fault detection rate allocation of assigned objects;

[0007] The influencing factors include: frequency of failure, degree of electronic integration, coverage of condition monitoring, and proportion of easily identifiable failures. The frequency of failure is determined based on the number of failures that occur in similar products of the allocated object during field use. The degree of electronic integration refers to the proportion of electronic circuit-related components in the allocated object. The coverage of condition monitoring refers to the proportion of each functional item and status parameter of the allocated object that is effectively covered by existing condition monitoring methods. The proportion of easily identifiable failures refers to the proportion of the number of failures in the allocated object that can be determined by only a single parameter.

[0008] Step 2: Normalize the data of influencing factors for each allocation object;

[0009] Step 3: Determine the weight of each influencing factor through expert evaluation;

[0010] Step 4: For each allocation object, use the weights calculated in Step 3 to perform a weighted summation of the normalized influencing factor values ​​obtained in Step 2 to obtain the comprehensive allocation coefficient for each allocation object;

[0011] Step 5: Calculate the fault detection rate allocation value of the allocated object based on the comprehensive allocation coefficient of the allocated object.

[0012] In step 5, let the comprehensive allocation coefficient of the i-th allocation object be... Then calculate the fault detection rate allocation value for the assigned object. as follows:

[0013] ;

[0014] in, This is the required value for the system fault detection rate. The total number of objects currently allocated.

[0015] The advantages and positive effects of this invention are as follows: Compared to traditional methods that rely on complete failure rate data, this invention effectively addresses the problem of missing failure rate information in the system. By introducing more representative influencing factors such as failure frequency, level of electronicization, condition monitoring coverage, and simple fault identification ratio, the allocation process remains operable even in the absence of traditional basic data, significantly improving the feasibility of engineering implementation. Simultaneously, through quantitative evaluation and weighting of multiple influencing factors, this invention can more accurately reflect the fault detection needs of each allocation object in the system, thus providing a basis for the rational allocation of subsequent testing resources. Furthermore, this invention makes special modifications to the low detection rate target in the allocation model, giving the allocation method good adaptability and stability. Attached Figure Description

[0016] Figure 1This is an overall flowchart of the fault detection rate allocation method that considers multiple factors in an embodiment of the present invention. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0018] The fault detection rate allocation method of this invention, which considers multiple factors, first takes into account the characteristics of the power system and uses the frequency of fault occurrence, the degree of electronicization, the coverage of condition monitoring, and the proportion of simple-to-identify faults as influencing factors for fault detection rate allocation. Data preparation and normalization of these influencing factors are then completed to eliminate the influence of dimensions. Next, independent scoring by multiple experts is conducted to comprehensively determine the weight of each influencing factor. Based on the normalized values ​​and weights of the influencing factor data, a comprehensive allocation coefficient is determined for each object to be allocated. Finally, the fault detection rate index is allocated according to the comprehensive allocation coefficient and the allocation model.

[0019] like Figure 1 As shown in the figure, the fault detection rate allocation method considering multiple factors implemented in this embodiment of the invention includes five steps, which are described in detail below.

[0020] Step 1: For the system where fault detection rate allocation is to be carried out, compile data on influencing factors.

[0021] Because power systems are characterized by electromechanical-hydraulic hybridity, high functional integration, and tight system coupling, the main influencing factors to be considered when setting the power system fault detection rate allocation in this invention during test allocation include: fault occurrence frequency, degree of electronicization, condition monitoring coverage, and the proportion of simple-to-identify faults.

[0022] A typical power system is divided into multiple levels, such as system, subsystem, component, and part level. Testability allocation is a process of decomposing indicators from the whole to the parts, from top to bottom. The indicator allocation process is the same across products at each level. This embodiment of the invention uses the allocation of fault detection rate from the power system to its subsystems as an example to illustrate the fault detection rate of the power system.

[0023] Step 1.1: Determine the frequency of failures of the objects to be assigned.

[0024] The failure frequency of the object to be assigned refers to the number of failures that occur in similar products of that object during field use. In this embodiment of the invention, the failure frequency is statistically analyzed in the form of Table 1. The statistical items in Table 1 include the assigned object, the default number of failures, the statistical number of failures, and the total number of failures.

[0025] Table 1. Fault Occurrence Frequency Statistics

[0026]

[0027] The specific statistical method is as follows:

[0028] (1) Fill in the name of the allocation object in the "allocation object" project; in this embodiment of the invention, the allocation objects are the subsystems under the current power system;

[0029] (2) Determine the default number of faults for the assigned object. Enter 1 in the "Default Fault Count" field;

[0030] (3) Count the number of field failures of the assigned objects The number of failures should be entered in the "Statistical Fault Count" field. In this embodiment of the invention, the number of failures in the field of each subsystem under the current power system is collected and filled in the table.

[0031] (4) Total number of failures The sum of the default number of faults and the statistical number of faults, i.e. Then fill in the results in "Total number of faults".

[0032] Step 1.2: Determine the degree of electronic access of the objects to be allocated.

[0033] The degree of electronic integration of the objects to be allocated refers to the proportion of electronic circuit-related components in the objects to be allocated. The degree of electronic integration is statistically presented in Table 2, including the objects to be allocated, the number of electronic modules, the number of non-electrical components, and the degree of electronic integration. Electronic modules refer to integrated electronic components or assemblies with specific functions. They undertake functions such as electrical signal processing, control, and communication in the system. Common electronic modules include: control modules, storage modules, sensor modules, communication modules, and power management modules. Non-electrical components refer to components that do not directly participate in electrical signal processing and mainly serve mechanical structural or physical functions such as support, connection, transmission, and sealing. Common non-electrical components include: mechanical transmission mechanisms, cooling and heat dissipation components, hydraulic valve groups, and pneumatic actuators.

[0034] Table 2. Statistics on the Level of Electronicization

[0035]

[0036] The specific statistical method used in Table 2 is as follows:

[0037] (1) Enter the name of the assigned object in the "Assigned Object" column;

[0038] (2) Count the number of electronic modules contained in each allocation object. Enter the number of electronic modules in the "Number of Electronic Modules" column;

[0039] (3) Count the number of non-electrical components contained in each allocation object. Enter the number of non-electrical components in the "Non-electrical Components Quantity" column;

[0040] (3) Calculate the degree of electronicization Enter the calculation results in the "Level of Electronicization" column.

[0041] Step 1.3: Determine the status monitoring coverage of the objects to be assigned.

[0042] The status monitoring coverage of the object to be allocated refers to the proportion of functional items and status parameters of the object that can be effectively covered by existing status monitoring methods. Status monitoring coverage is statistically presented in Table 3, including the allocated object, number of functional items, number of status parameters, number of monitorable functional items, number of monitorable status parameters, and status monitoring coverage. A functional item refers to a clearly identifiable functional unit in the system that possesses a specific function, typically manifested as a specific functional behavior, such as the voltage regulation function of a power module or the signal transmission function in a communication system. Status parameters refer to physical or logical quantities closely related to the operating status of the allocated object, such as temperature, voltage, current, frequency, flow rate, operating time, and switch status. Monitorable functional items refer to functional items of the allocated object that have been configured with status monitoring methods and can sense their operating status in real time or periodically. Monitorable status parameters refer to status parameters that can be measured through sensors, monitoring circuits, or software logic.

[0043] Table 3 Status Monitoring Coverage Statistics

[0044]

[0045] The specific statistical methods used in Table 3 are as follows:

[0046] (1) Enter the name of the assigned object in the "Assigned Object" column;

[0047] (2) Count the number of main functional items contained in each allocation object. Enter the number of function items in the "Number of function items" column;

[0048] (3) Count the number of state parameters that can be used to determine the state of the assigned object. Enter the number of status parameters in the "Number of Status Parameters" column;

[0049] (4) The number of statistical function items that can be covered by existing status monitoring methods. Enter the number of monitorable functions in the "Number of Monitorable Functions" column;

[0050] (5) Count the number of monitorable state parameters. Enter the number of monitorable status parameters in the "Number of Monitorable Status Parameters" column;

[0051] (6) Calculate the coverage of status monitoring Enter the calculation results in the "Status Monitoring Coverage" column.

[0052] Step 1.4: Determine the percentage of simple faults in the objects to be assigned.

[0053] The simple-determined fault percentage of the objects to be assigned refers to the percentage of faults in the objects to be assigned whose occurrence can be determined using only a single parameter. The simple-determined fault percentage is statistically presented in Table 4, which includes the objects to be assigned, the number of faults that can be determined with a single parameter, the number of faults that can be determined by a combination of multiple parameters, and the simple-determined fault percentage.

[0054] Table 4. Statistics on the Percentage of Simple Fault Diagnosis

[0055]

[0056] The specific method is as follows:

[0057] (1) Enter the name of the assigned object in the "Assigned Object" column;

[0058] (2) Count the number of faults that can be determined by a single monitoring parameter. Enter the number of faults that can be determined by a single parameter in the column;

[0059] (3) The number of faults that require comprehensive analysis of two or more monitoring parameters to determine. Enter the number of faults determined by multi-parameter comprehensive judgment in the column;

[0060] (4) Calculate the proportion of simple faults Enter the calculation results in the "Simple Fault Ratio" column.

[0061] Step 2: Normalize the data of influencing factors for each allocation object.

[0062] To eliminate the influence of dimensions and achieve a unified and comprehensive analysis of all influencing factors, the data on these factors need to be normalized. The normalized value of the i-th allocation object under the j-th influencing factor is calculated using the following formula:

[0063]

[0064] in, This represents the i-th allocation object. The number of currently allocated objects corresponds to the number of subsystems under the power system in this embodiment of the invention; Let j represent the j-th influencing factor, which corresponds to: frequency of fault occurrence, degree of electronicization, coverage of condition monitoring, and proportion of simple-to-identify faults, respectively. The original value of the k-th assigned object under the j-th factor.

[0065] The normalized results of each allocation object under various influencing factors can be calculated using the above formula, and the obtained normalized data are summarized and recorded in Table 5.

[0066] Table 5. Testability Requirements Allocation Table

[0067]

[0068] Step 3: Determine the weight of influencing factors through expert evaluation.

[0069] Step 3.1: Experts assess the importance of influencing factors.

[0070] Based on the relative importance of each influencing factor in the allocation of system failure detection rate, a pairwise comparison method is used to evaluate all factors, thereby constructing a judgment matrix. Specifically, experts need to evaluate any two influencing factors... and The relative importance of each is compared and a scale value is assigned. , indicating factors Relative factors The importance of each factor. To obtain a quantifiable judgment matrix for comparison among the factors, a commonly used scale of 1 to 9 is employed. The specific meanings of the scales are shown in Table 6. Table 6 Meaning of the Scales

[0071]

[0072] The resulting judgment matrix has the following properties:

[0073]

[0074]

[0075]

[0076] Each expert, based on their professional knowledge and practical experience, and referring to Table 6, conducted pairwise comparisons of influencing factors such as failure frequency, level of electronicization, coverage of condition monitoring, and proportion of simple-to-identify failures, constructing a judgment matrix and calculating the influencing factors using the geometric mean method. Importance The specific calculation formula is as follows:

[0077]

[0078] Table 7 Judgment Matrix

[0079]

[0080] Step 3.2: Determine the weighting coefficients of the influencing factors.

[0081] Table 8 summarizes the importance of all expert assessment factors, and the average factor importance is calculated. :

[0082]

[0083] in, This represents the opinion of the k-th expert on the influencing factors. The assessment results for importance, where m is the total number of experts.

[0084] The calculated average importance of each factor is normalized to obtain the relative weight of each factor. The specific calculation formula is as follows:

[0085]

[0086] Received This represents the final weight of each influencing factor, which will serve as the basis for determining the overall allocation coefficient.

[0087] Table 8 Weights of Influencing Factors

[0088]

[0089] Step 4: Determine the multi-factor comprehensive allocation coefficient for the objects to be allocated.

[0090] To obtain the comprehensive allocation coefficient for each allocation object, a weighted sum is needed based on the weights of each influencing factor, building upon the aforementioned normalization process. The comprehensive allocation coefficient for the i-th allocation object... The calculation formula is as follows:

[0091]

[0092] in: It is the normalized value of the failure frequency of the i-th allocated object; It is the normalized value of the electronic degree of the i-th allocation object; It is the normalized value of the state monitoring coverage of the i-th allocation object; It is the normalized value of the simple judgment failure rate of the i-th allocation object; These are the weight values ​​of the four influencing factors.

[0093] Based on the above formula, calculate the comprehensive allocation coefficient for each allocation object and fill the results into the test requirement allocation table shown in Table 5.

[0094] Step 5: Determine the fault detection rate allocation value for the objects to be assigned.

[0095] Based on the comprehensive allocation coefficient of the allocation recipients The following allocation model is used to calculate the Fault Detection Rate (FDR) allocation value for the i-th allocation object. The calculation formula is as follows:

[0096]

[0097] In the formula, It is the FDR allocation value of the i-th allocation object; This is the required value for the system's FDR; This represents the total number of objects currently awaiting allocation.

[0098] Unlike traditional allocation methods that rely primarily on failure rates for calculation, the method of this invention uses a comprehensive allocation coefficient, which is more in line with the characteristics of power systems and provides better accuracy in detecting system failures. Cases with values ​​less than 0.5 were also considered, and the allocation model was modified to give the method of the present invention good adaptability and stability.

[0099] The FDR allocation value for each object is calculated according to the above formula, and the allocation value is filled into the second to last column of Table 5. In this embodiment of the invention, the allocation value is rounded to two decimal places, and the adjusted allocation value is filled into the last column of Table 5.

[0100] The following example, using specific data from a subsystem of a power system, illustrates the use of the multi-factor comprehensive testability allocation method of this invention.

[0101] Step 1: Compile data on factors influencing the allocation of object fault detection rates.

[0102] The power system can be divided into engine subsystem, transmission subsystem, auxiliary subsystem and electric subsystem in terms of system composition.

[0103] Step 1.1: Determine the frequency of failures of the objects to be assigned. Statistical analysis of the number of failures occurring in similar products across various subsystems of the power system during field use is performed; specific data is shown below.

[0104] Table 9. Statistics on the frequency of power system failures

[0105]

[0106] Step 1.2: Determine the degree of electronic access of the objects to be allocated.

[0107] The degree of electronic integration of each subsystem of the power system was statistically analyzed, and the specific data are shown below.

[0108] Table 10 Statistical data on the degree of digitalization

[0109]

[0110] Step 1.3: Determine the status monitoring coverage of the objects to be assigned.

[0111] The coverage of condition monitoring for each subsystem of the power system was statistically analyzed, and the specific data are shown below.

[0112] Table 11 Status monitoring coverage statistics

[0113]

[0114] Step 1.4: Determine the percentage of simple faults in the objects to be assigned.

[0115] The percentage of simple faults in each subsystem of the power system is statistically analyzed, and the specific data is shown below.

[0116] Table 12 Statistical data on the percentage of simple fault identification

[0117]

[0118] Step 2: Normalize the data of influencing factors for each subsystem.

[0119] The data on influencing factors in Tables 9-12 were normalized according to Equation (1), and the results are shown in Table 13 below.

[0120] Table 13 Testability Requirements Allocation Table

[0121]

[0122] Step 3: Determine the weight of the influencing factors for each subsystem through expert scoring.

[0123] Step 3.1: Construct the judgment matrix and calculate the importance of the factors.

[0124] In this embodiment of the invention, five experts in the field of power system-related fields evaluated the relative importance of factors such as the frequency of fault occurrence, the degree of electronicization, the coverage of condition monitoring, and the proportion of simple fault identification. The importance of the factors was calculated according to formula (5), and the results are shown in Tables 14-18.

[0125] Table 14 Judgment Matrix 1

[0126]

[0127] Table 15 Judgment Matrix 2

[0128]

[0129] Table 16 Judgment Matrix 3

[0130]

[0131] Table 17 Judgment Matrix 4

[0132]

[0133] Table 18 Judgment Matrix 5

[0134]

[0135] Step 3.2: Calculate the weights of influencing factors.

[0136] The evaluation results of 5 experts were summarized, and the weights of the influencing factors were calculated according to Equations (6) and (7). The results are shown below.

[0137] Table 19 Weights of Influencing Factors

[0138]

[0139] Step 4: Determine the overall allocation coefficient for each subsystem.

[0140] Based on the normalized data, the comprehensive allocation coefficient is calculated using equation (8). The results are shown below.

[0141] Table 20 Test Requirements Allocation Table

[0142]

[0143] Step 5: Calculate the fault detection rate allocation value for each subsystem.

[0144] Required power system fault detection rate =0.80, use formula (9) to calculate the fault detection rate allocation value of each subsystem of the power system, and fill the allocation value of the allocation object to two decimal places into the table, as shown below.

[0145] Table 21 Testability Requirements Allocation Table

[0146]

[0147] Calculated from Table 21 The adjustment value serves as the FDR requirement value for the corresponding subsystem, and the fault detection rate can continue to be allocated to the next layer of allocation objects in each subsystem according to the method of this invention.

[0148] As can be seen from the above embodiments, the method of the present invention can achieve fault detection rate allocation for the power system, ultimately ensuring that the overall design of the power system achieves the expected fault detection effect. The method of the present invention, combined with the characteristics of the power system, introduces more representative influencing factors, making the allocation process operable even in the absence of traditional basic data, significantly improving the feasibility of engineering implementation.

[0149] Except for the technical features described in the specification, all other technologies are known to those skilled in the art. Descriptions of well-known components and technologies are omitted in this invention to avoid redundancy and unnecessary limitation. The embodiments described above do not represent all embodiments consistent with this application. Various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this invention are still within the protection scope of this invention.

Claims

1. A fault detection rate allocation method considering multiple factors, which determines the product level of the power system to be allocated fault detection rates, and performs hierarchical fault detection rate allocation from top to bottom; characterized in that, Using the current level of products as the allocation object, the method performs the following steps: Step 1: Organize and acquire data on factors influencing the fault detection rate allocation of assigned objects; The influencing factors include: frequency of failure, degree of electronic integration, coverage of condition monitoring, and proportion of easily identifiable failures. The frequency of failure is determined based on the number of failures that occur in similar products of the allocated object during field use. The degree of electronic integration refers to the proportion of electronic circuit-related components in the allocated object. The coverage of condition monitoring refers to the proportion of each functional item and status parameter of the allocated object that is effectively covered by existing condition monitoring methods. The proportion of easily identifiable failures refers to the proportion of the number of failures in the allocated object that can be determined by only a single parameter. Step 2: Normalize the data of influencing factors for each allocation object; Step 3: Determine the weight of each influencing factor through expert evaluation; Step 4: For each allocation object, use the weights calculated in Step 3 to perform a weighted summation of the normalized influencing factor values ​​obtained in Step 2 to obtain the comprehensive allocation coefficient for each allocation object; Step 5: Calculate the fault detection rate allocation value of the allocated object based on the comprehensive allocation coefficient of the allocated object.

2. The method according to claim 1, characterized in that, Step 1 includes: (1) Count the frequency of failures for each allocation object, including: setting a default number of failures for each allocation object. The value is 1, indicating the number of malfunctions that occurred in similar products during field use. Calculate the frequency of failures of this object. The default number of faults With the number of statistical failures sum; (2) Determine the degree of digitalization of each allocation object, including: for each allocation object, count the number of electronic modules in that object. Number of non-electrical components Calculate the degree of electronicization ; (3) Determine the status monitoring coverage for each assigned object, including: for each assigned object, count the number of functional items. and the number of state parameters Statistical analysis of the number of functional items covered by existing status monitoring methods Count the number of state parameters that can be measured by sensors, monitoring circuits, or software logic. Calculate the coverage of status monitoring ; (4) Calculate the proportion of simple-discrimination faults for each allocation object, including: for each allocation object, count the number of faults that can be judged by a single parameter. And the number of faults that require comprehensive analysis based on two or more parameters to determine. Calculate the percentage of simple discriminant faults for this object. .

3. The method according to claim 1 or 2, characterized in that, In step 2, the data of the j-th influencing factor of the i-th allocation object is... Normalization yields as follows: ； in, The number of objects currently allocated. ; Data for the j-th influencing factor of the k-th allocation object; These correspond to four influencing factors: frequency of failures, level of electronicization, coverage of condition monitoring, and proportion of simple fault identification.

4. The method according to claim 1, characterized in that, Step 3 includes: First, obtain the judgment matrix for each expert's assessment of the influencing factors. The matrix contains the judgment matrix for the first influencing factor. Line number The elements in the column represent influencing factors. Relative to influencing factors Importance scale, with scale value set as follows: , The diagonal elements of the matrix are 1, and the element values ​​of the matrix are... For each expert's evaluation judgment matrix, the importance of each influencing factor is calculated using the geometric mean method. Importance ; Then, the importance of each influencing factor is obtained from the judgment matrix evaluated by m experts. The average of the m importance values ​​of each influencing factor is calculated, and the average importance values ​​of the four influencing factors are normalized to obtain the weight of each influencing factor.

5. The method according to claim 1 or 4, characterized in that, In step 5, let the comprehensive allocation coefficient of the i-th allocation object be... Then calculate the fault detection rate allocation value for the assigned object. as follows: ； in, This is the required value for the system fault detection rate. The total number of objects currently allocated.

Images