A method and system for selecting temperature-sensitive characteristics of relay protection devices
By constructing a thermal analysis model of the relay protection device and using correlation coefficients and random forest algorithms to determine temperature-sensitive characteristics, the problem of unclear temperature-sensitive characteristics of the device is solved, and the thermal management capability of the device is improved.
Patent Information
- Application Number
- CN202410898078.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2044-07-05
AI Technical Summary
The lack of quantitative thermal simulation analysis methods in existing technologies leads to unclear temperature sensitivity characteristics of relay protection devices, making it impossible to detect overheating problems in a timely manner and affecting the thermal reliability of the devices.
A thermal analysis model for relay protection devices is established. By using correlation coefficient and random forest algorithms, a set of temperature-sensitive features is determined, a temperature sensitivity matrix is constructed, and the key factors affecting the device temperature are accurately identified.
This study enabled quantitative analysis of the temperature distribution characteristics of relay protection devices, grasped the temperature sensitivity of core components, and improved the thermal management level of the devices.
Smart Images

Figure CN118861611B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of relay protection technology, and more specifically, to a method and system for selecting the temperature-sensitive characteristics of a relay protection device. Background Technology
[0002] Relay protection devices are among the key equipment in the power grid dispatch and communication field, serving as the first line of defense for ensuring the safe and stable operation of the power grid. Overheating of components severely impacts the reliability of relay protection devices. During the device development phase, manufacturers primarily rely on design experience and high / low temperature testing of the entire device to ensure its thermal reliability, lacking quantitative thermal simulation analysis methods, and the temperature sensitivity characteristics of the devices remain unclear. Current practices do not meet the high reliability requirements of protection devices. With the widespread application of domestically produced chips in protection devices, the relatively higher power consumption of domestically produced chips compared to imported chips due to differences in manufacturing processes and power consumption leads to overheating issues, exacerbating the need for thermal simulation analysis of relay protection devices.
[0003] Currently, relay protection device manufacturers possess years of experience in research and development, testing, and operation. During the device development process, they leverage past design experience, allow for parameter margins, and conduct subsequent high and low temperature tests to ensure the thermal reliability of the relay protection equipment. However, this approach suffers from limitations in accuracy, making it difficult to detect overheating issues promptly and to identify the key factors influencing the device's temperature.
[0004] Therefore, a technology is needed to enable the selection of temperature-sensitive characteristics in relay protection devices. Summary of the Invention
[0005] The present invention provides a method and system for selecting temperature-sensitive characteristics of relay protection devices, in order to solve the problem of how to select temperature-sensitive characteristics of relay protection devices.
[0006] To address the above problems, the present invention provides a method for selecting the temperature-sensitive characteristics of a relay protection device, the method comprising:
[0007] Based on the structural characteristics of relay protection devices, a thermal analysis model of the relay protection device is established; the thermal analysis model includes the determined key simulation parameters;
[0008] Based on the thermal analysis model, the temperature distribution characteristics of the relay protection device are determined;
[0009] Based on the temperature distribution characteristics, a first temperature-sensitive feature set M is established using the correlation coefficient algorithm; a second temperature-sensitive feature set N is established using the random forest algorithm.
[0010] Based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, the temperature sensitivity matrix Ω of the relay protection device is determined.
[0011] Preferably, the thermal analysis model includes a housing thermal analysis model and a board thermal analysis model;
[0012] The enclosure analysis model includes: an outer shell, flow channel baffles, and a back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffles are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area.
[0013] The board thermal analysis model includes: CPU board, power supply board, and analog signal acquisition board; the heat-generating and heat-conducting components on the CPU board, power supply board, and analog signal acquisition board are determined according to the dimensions in the actual product.
[0014] Preferably, the key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature.
[0015] Preferably, the step of establishing a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm includes:
[0016] Calculate the correlation coefficient between random variables X and Y based on the Pearson correlation coefficient:
[0017]
[0018] Where X and Y are the temperature characteristic variable and the component temperature variable, respectively, cov(X,Y) is the covariance of X and Y, and σ X ,σ Y These are the standard deviations of X and Y, respectively.
[0019] The correlation coefficient between random variables X and Y is calculated using the number of rank transpositions based on Kendall's rank correlation coefficient.
[0020]
[0021] Where n is the number of samples, and ∑ is the total number of transpositions;
[0022] The correlation coefficient between random variables X and Y is calculated based on the number of ranks between them, according to the Spearman correlation coefficient.
[0023]
[0024] Where n is the number of samples, i is the index of the variable element, and r i and s iThey represent x respectively i and y i rank, x i and y i Let be the i-th elements of random variables X and Y, respectively. and x i and y i The average rank;
[0025] When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factor all exceed a preset threshold, the influencing factor will be regarded as a temperature-sensitive feature.
[0026] Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
[0027] Preferably, the first temperature-sensitive feature set M includes: power consumption of electronic devices, power consumption of circuit boards, ambient temperature, altitude, and opening ratio of the casing.
[0028] Preferably, the step of establishing the second temperature-sensitive feature set N using the random forest algorithm includes:
[0029] The information entropy of dataset Q is:
[0030]
[0031] Where c represents the number of data categories, p i This represents the proportion of the i-th category of data in the total data;
[0032] After feature A is applied, the dataset Q is divided into n subsets Q. j For j = 1, 2, ..., n, the information entropy of dataset Q is expressed as:
[0033]
[0034] Where j represents the j-th category, p j H(Q) represents the proportion of the j-th category of data in the total data; j ) represents the information entropy of the j-th category sample;
[0035] Information gain is:
[0036] G(Q,A)=H(Q)-H A (Q) (6)
[0037] Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features.
[0038] Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
[0039] Preferably, the second temperature-sensitive feature set N includes: device power consumption, board power consumption, altitude, shell opening ratio, and ambient temperature.
[0040] Preferably, determining the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N includes:
[0041] Ω=M∩N={device power consumption, board power consumption, altitude, shell opening ratio, ambient temperature}.
[0042] Based on another aspect of the present invention, the present invention provides a selection system for the temperature-sensitive characteristics of a relay protection device, the system comprising:
[0043] An initial unit is used to establish a thermal analysis model of the relay protection device based on its structural characteristics; the thermal analysis model includes determined key simulation parameters.
[0044] The determining unit is used to determine the temperature distribution characteristics of the relay protection device based on the thermal analysis model.
[0045] A unit is established to establish a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm, and to establish a second temperature-sensitive feature set N using a random forest algorithm.
[0046] The result unit is used to determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N.
[0047] Preferably, the thermal analysis model includes a housing thermal analysis model and a board thermal analysis model;
[0048] The enclosure analysis model includes: an outer shell, flow channel baffles, and a back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffles are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area.
[0049] The board thermal analysis model includes: CPU board, power supply board, and analog signal acquisition board; the heat-generating and heat-conducting components on the CPU board, power supply board, and analog signal acquisition board are determined according to the dimensions in the actual product.
[0050] Preferably, the key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature.
[0051] Preferably, the establishing unit is used to establish a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm, and is further used to:
[0052] Calculate the correlation coefficient between random variables X and Y based on the Pearson correlation coefficient:
[0053]
[0054] Where X and Y are the temperature characteristic variable and the component temperature variable, respectively, cov(X,Y) is the covariance of X and Y, and σ X ,σ Y These are the standard deviations of X and Y, respectively.
[0055] The correlation coefficient between random variables X and Y is calculated using the number of rank transpositions based on Kendall's rank correlation coefficient.
[0056]
[0057] Where n is the number of samples, and ∑ is the total number of transpositions;
[0058] The Spearman correlation coefficient is used to calculate the correlation coefficient between random variables X and Y based on the order of their ranks.
[0059]
[0060] Where n is the number of samples, i is the index of the variable element, and r i and s i They represent x respectively i and y i rank, x i and y i Let X and Y be the i-th elements of random variables X and Y, respectively. and x i and y i The average rank;
[0061] When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factor all exceed a preset threshold, the influencing factor will be regarded as a temperature-sensitive feature.
[0062] Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
[0063] Preferably, the first temperature-sensitive feature set M includes: power consumption of electronic devices, power consumption of circuit boards, ambient temperature, altitude, and opening ratio of the casing.
[0064] Preferably, the establishing unit is used to establish a second temperature-sensitive feature set N using a random forest algorithm, and is also used to:
[0065] The information entropy of dataset Q is:
[0066]
[0067] Where c represents the number of data categories, p i This represents the proportion of the i-th category of data in the total data;
[0068] After feature A is applied, the dataset Q is divided into n subsets Q. j For j = 1, 2, ..., n, the information entropy of dataset Q is expressed as:
[0069]
[0070] Where j represents the j-th category, p j H(X) represents the proportion of the j-th category of data in the total data; j ) represents the information entropy of the j-th category sample;
[0071] Information gain is:
[0072] G(Q,A)=H(Q)-H A (Q) (6)
[0073] Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features.
[0074] Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
[0075] Preferably, the second temperature-sensitive feature set N includes: device power consumption, board power consumption, altitude, shell opening ratio, and ambient temperature.
[0076] Preferably, the result unit is used to determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, and is also used to:
[0077] Ω=M∩N={device power consumption, board power consumption, altitude, shell opening ratio, ambient temperature}.
[0078] According to another aspect of the present invention, the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing a method for selecting temperature-sensitive features of a relay protection device.
[0079] According to another aspect of the present invention, the present invention provides an electronic device, characterized in that the electronic device comprises: a processor and a memory; wherein,
[0080] The memory is used to store the processor-executable instructions;
[0081] The processor is configured to read the executable instructions from the memory and execute the instructions to implement a method for selecting temperature-sensitive features of a relay protection device.
[0082] This invention provides a method and system for selecting temperature-sensitive characteristics of relay protection devices. The method includes: establishing a thermal analysis model of the relay protection device based on its structural characteristics; setting key simulation parameters for the thermal analysis model; determining the temperature distribution characteristics of the relay protection device based on the thermal analysis model with the set key simulation parameters; establishing a set of temperature-sensitive characteristics M using a correlation coefficient algorithm based on the temperature distribution characteristics; establishing a set of temperature-sensitive characteristics N using a random forest algorithm; and determining the temperature sensitivity matrix Ω of the relay protection device based on the temperature-sensitive characteristic sets M and N. This invention provides a method and system for selecting temperature-sensitive characteristics of relay protection devices. By using a thermal analysis model to understand the temperature-sensitive characteristics affecting the core components of the device, and by analyzing these sensitive characteristics, it achieves accurate determination of the temperature-sensitive characteristics affecting the device, which is of great significance for improving the thermal management level of relay protection devices. Attached Figure Description
[0083] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures:
[0084] Figure 1 This is a flowchart of a method for selecting temperature-sensitive features of a relay protection device according to a preferred embodiment of the present invention.
[0085] Figure 2 This is a flowchart of a method for selecting temperature-sensitive features of a relay protection device according to a preferred embodiment of the present invention.
[0086] Figure 3 This is a schematic diagram of a box analysis model according to a preferred embodiment of the present invention;
[0087] Figure 4 This is a schematic diagram of a thermal analysis model for a circuit board according to a preferred embodiment of the present invention;
[0088] Figure 5 This is a schematic diagram illustrating the correlation between the temperature and characteristic parameters of the core components according to a preferred embodiment of the present invention.
[0089] Figure 6A schematic diagram illustrating the importance of temperature-sensitive characteristics of core components according to a preferred embodiment of the present invention; and
[0090] Figure 7 This is a structural diagram of a system for selecting the temperature-sensitive characteristics of a relay protection device according to a preferred embodiment of the present invention. Detailed Implementation
[0091] Exemplary embodiments of the invention will now be described with reference to the accompanying drawings. However, the invention may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the invention and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the invention. In the drawings, the same units / elements are referred to by the same reference numerals.
[0092] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.
[0093] Figure 1 This is a flowchart of a method for selecting temperature-sensitive features of a relay protection device according to a preferred embodiment of the present invention.
[0094] Currently, existing technologies lack methods for thermal simulation modeling and temperature-sensitive feature selection in the field of relay protection equipment, failing to fundamentally understand and solve potential overheating problems in relay protection devices. This invention proposes a device-level thermal simulation modeling method applicable to different types of protection equipment by conducting research on modeling and thermal simulation methods for typical relay protection equipment. It constructs thermal simulation models of typical protection devices and obtains the internal and external excitation factors affecting the temperature field distribution of the device through simulation analysis. Then, based on a large amount of simulation data, it proposes a method for selecting temperature-sensitive features of core components of relay protection devices. This enables quantitative analysis of the temperature distribution characteristics of relay protection devices and helps to understand the temperature-sensitive features affecting the core components, which is of great significance for improving the thermal management level of relay protection devices.
[0095] This invention addresses the challenge of lacking quantitative analysis methods for the thermal characteristics of relay protection devices and methods for selecting temperature-sensitive features. It proposes a method for selecting temperature-sensitive features of relay protection devices. This method enables quantitative analysis of the temperature distribution characteristics of relay protection devices and helps to understand the sensitive features affecting the temperature of core components, which is of great significance for improving the thermal management level of the devices.
[0096] like Figure 1As shown, the present invention provides a method for selecting the temperature-sensitive characteristics of a relay protection device, the method comprising:
[0097] Step 101: Based on the structural characteristics of the relay protection device, establish a thermal analysis model of the relay protection device; the thermal analysis model includes the determined key simulation parameters;
[0098] Preferably, the thermal analysis model includes a box thermal analysis model and a board thermal analysis model;
[0099] The enclosure analysis model includes: the outer shell, the flow channel baffle, and the back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffle are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area.
[0100] The thermal analysis model of the circuit boards includes: CPU boards, power supply boards, and analog signal acquisition boards; the heat-generating and heat-conducting components on the CPU boards, power supply boards, and analog signal acquisition boards are determined according to the dimensions in the actual products.
[0101] Preferably, the key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature.
[0102] This invention conducts an analysis of the temperature distribution characteristics of key components in relay protection devices. Based on the constructed thermal simulation model of the protection device, it analyzes the temperature distribution characteristics of core components under the influence of internal and external excitation factors.
[0103] The device thermal simulation modeling method of the present invention includes:
[0104] (1) Box modeling method
[0105] The relay protection device primarily dissipates heat externally through its enclosure. The enclosure modeling follows the principle of maximizing the preservation of the original heat dissipation path, removing components that do not affect heat dissipation, and simplifying features that do not impact heat dissipation. Based on the original channel design, the enclosure is divided into three parts: the outer shell, the flow channel baffle, and the backplate assembly mounting area. The dimensions and specifications of each opening in the flow channel baffle are retained from the actual product, while curved edges at the openings that do not impede airflow are reduced or simplified. The backplate assembly mounting area retains the original internal framework and preserves the airflow vortex area.
[0106] (2) Board Modeling Method
[0107] When modeling the PCB, prioritize retaining high-power components and components that obstruct gas flow. Remove low-power, heat-generating components that do not affect gas flow, and include their power consumption in the PCB designation. When modeling components, retain the original outline dimensions of heat-generating and heat-conducting parts. Repairing components that obstruct gas flow should not affect the flow path. The PCBs involved include CPU boards, power supply boards, analog input / output boards, etc.
[0108] (3) Set key parameters for thermal simulation
[0109] Set boundary conditions and basic solution parameters, including ambient temperature, component power consumption, altitude, and turbulence model. Simultaneously, parameter settings for heat-conducting devices are required, the most important of which is thermal conductivity, including the thermal conductivity of the device chassis, panel, heat sink, and PCB.
[0110] Step 102: Based on the thermal analysis model, determine the temperature distribution characteristics of the relay protection device;
[0111] This invention calculates the temperature distribution characteristics of core components:
[0112] Determine the factors R that affect the temperature distribution of the device. i (i = 1, 2, 3, ..., n), including the ambient temperature of the thermal conduction environment, the power consumption of components, the power consumption of the board, etc. Then, based on the ambient temperature, altitude, and power consumption of the relay protection device as specified in the standard, the temperature distribution characteristics of the core components of the device under different operating conditions are calculated.
[0113] Step 103: Based on the temperature distribution characteristics, establish the first temperature-sensitive feature set M using the correlation coefficient algorithm; establish the second temperature-sensitive feature set N using the random forest algorithm;
[0114] Preferably, based on temperature distribution characteristics, a first temperature-sensitive feature set M is established using a correlation coefficient algorithm, including:
[0115] Calculate the correlation coefficient between random variables X and Y based on the Pearson correlation coefficient:
[0116]
[0117] Where X and Y are the temperature characteristic variable and the component temperature variable, respectively, cov(X,Y) is the covariance of X and Y, and σ X ,σ Y These are the standard deviations of X and Y, respectively.
[0118] The correlation coefficient between random variables X and Y is calculated using the number of rank transpositions based on Kendall's rank correlation coefficient.
[0119]
[0120] Where n is the number of samples, and ∑ is the total number of transpositions;
[0121] The correlation coefficient between random variables X and Y is calculated based on the number of ranks between them, according to the Spearman correlation coefficient.
[0122]
[0123] Where n is the number of samples, i is the index of the variable element, and r i and s i They represent x respectively i and y i rank, x i and y i Let be the i-th elements of random variables X and Y, respectively. and x i and y i The average rank;
[0124] When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factors all exceed the preset threshold, the influencing factors will be regarded as temperature-sensitive features.
[0125] Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
[0126] Preferably, the first temperature-sensitive feature set M includes: power consumption of electronic devices, power consumption of circuit boards, ambient temperature, altitude, and opening ratio of the casing.
[0127] Preferably, a second temperature-sensitive feature set N is established using the random forest algorithm, including:
[0128] The information entropy of dataset Q is:
[0129]
[0130] Where c represents the number of data categories, p i This represents the proportion of the i-th category of data in the total data;
[0131] After feature A is applied, the dataset Q is divided into n subsets Q. j For j = 1, 2, ..., n, the information entropy of dataset Q is expressed as:
[0132]
[0133] Where j represents the j-th category, p j H(Q) represents the proportion of the j-th category of data in the total data; j ) represents the information entropy of the j-th category sample;
[0134] Information gain is:
[0135] G(Q,A)=H(Q)-H A (Q) (6)
[0136] Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features.
[0137] Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
[0138] Preferably, the second set of temperature-sensitive features N includes: device power consumption, board power consumption, altitude, shell opening ratio, and ambient temperature.
[0139] Based on the aforementioned temperature field distribution data of the device under different operating conditions, this invention selects the temperature-sensitive features of the core components of the device using the correlation coefficient method and the random forest method.
[0140] In the component temperature-sensitive feature selection method based on correlation coefficient of the present invention, the correlation coefficient is used to describe the correlation between variables. The present invention intends to use Pearson correlation coefficient, Kendall rank correlation coefficient and Spearman correlation coefficient to evaluate the correlation between the temperature of core components and thermal simulation parameters, and select temperature-sensitive features.
[0141] The correlation threshold is set to 0.1. Based on the correlation coefficient method described above, the correlation between the core component temperature and the thermal simulation parameters is calculated. If the correlation is greater than 0.1, the thermal simulation parameter is considered to have a strong correlation with the component temperature, and this parameter is selected. Finally, the first set of temperature-sensitive features, M, is obtained.
[0142] In the component temperature-sensitive feature selection method based on random forest in this invention, the random forest method is used for feature selection, the contribution of thermal simulation feature parameters to the temperature of core components is evaluated, and temperature-sensitive features are selected.
[0143] A contribution threshold of 0.1 was set, and the contribution of thermal simulation parameters to the temperature of core components was calculated using the random forest selection method. Parameters with a contribution greater than 0.1 were considered to have a significant impact on component temperature and were therefore selected, resulting in the second set of temperature-sensitive features, N. Figure 2 As shown.
[0144] Step 105: Determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N.
[0145] Preferably, the temperature sensitivity matrix Ω of the relay protection device is determined based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, including:
[0146] Ω=M∩N={device power consumption, board power consumption, altitude, shell opening ratio, ambient temperature}.
[0147] This invention comprehensively considers the correlation coefficient and the selection results obtained from random forests to determine the temperature sensitivity matrix Ω:
[0148] Ω=M∩N.
[0149] Currently, manufacturers mainly rely on design experience and high and low temperature testing of the entire device to ensure its thermal reliability, lacking quantitative thermal simulation analysis methods, and the temperature sensitivity characteristics of the device are still unclear. This invention, based on the true physical structure of a typical relay protection device, proposes a refined modeling method for device thermal simulation, constructs a finite element simulation model of the device, analyzes the mapping relationship between internal and external excitation factors such as ambient temperature, altitude, and component power consumption and the device temperature, and quantitatively analyzes the temperature distribution characteristics of key components under different operating conditions.
[0150] This invention, based on extensive simulation data obtained from thermal simulation models, employs a combination of correlation coefficient and random forest methods to select factors influencing temperature distribution in relay protection devices, constructing temperature-sensitive characteristics for key components. This provides data support for improving the thermal management level of the devices.
[0151] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.
[0152] S1: Conduct temperature distribution characteristic analysis of key components of relay protection devices. Based on the constructed thermal simulation model of the protection device, analyze the temperature distribution characteristics of core components under the action of internal and external excitation factors.
[0153] S11: Thermal Simulation Modeling Method for Devices
[0154] (1) Box modeling method
[0155] The relay protection device primarily dissipates heat externally through its enclosure. The enclosure modeling follows the principle of maximizing the preservation of the original heat dissipation path, removing components that do not affect heat dissipation, and simplifying features that do not impact heat dissipation. Based on the original channel design, the enclosure is divided into three parts: the outer shell, the flow channel baffles, and the backplate assembly mounting area. The dimensions and specifications of each opening in the flow channel baffles are retained from the actual product, while curved edges at the openings that do not impede airflow are reduced and simplified. The backplate assembly mounting area retains the original internal framework and the airflow vortex region. The enclosure thermal simulation analysis model constructed based on the proposed method is as follows: Figure 3 As shown.
[0156] (2) Board Modeling Method
[0157] When modeling the PCB, prioritize retaining high-power components and components that obstruct gas flow. Remove low-power heat-generating components that do not affect gas flow, and include their power consumption in the PCB designation. When modeling components, retain the original outline dimensions of heat-generating and heat-conducting parts. Repairing components that obstruct gas flow should not affect the flow path. Figure 4 As shown. The boards involved include the CPU board, power supply board, and analog signal acquisition board.
[0158] (3) Set key parameters for thermal simulation
[0159] Set boundary conditions and basic solution parameters, including ambient temperature, component power consumption, altitude, and turbulence model. Simultaneously, parameter settings for heat conduction devices are required, the most important of which is thermal conductivity, including the thermal conductivity of the device chassis, panel, heat sink, and PCB, as shown in Tables 1 and 2.
[0160] Table 1 Thermal conductivity of main thermally conductive materials
[0161]
[0162]
[0163] Table 2 Power consumption and thermal conductivity of each board
[0164]
[0165] S12: Calculate the temperature distribution characteristics of core components
[0166] Determine the factors R that affect the temperature distribution of the device. i (i=1,2,3,...,n), including thermal conductivity ambient temperature, component power consumption, board power consumption, etc., as shown in Table 3. Then, based on the ambient temperature, altitude, and power consumption of the protection device as specified in the standard, the temperature distribution characteristics of the core components of the device under different operating conditions are calculated.
[0167] Table 3. Composition of Feature Vectors
[0168]
[0169] S2: Based on the temperature field distribution data of the device under different operating conditions obtained above, the temperature-sensitive characteristics of the core components of the device are selected based on the correlation coefficient method and the random forest method.
[0170] S21: A method for selecting temperature-sensitive characteristics of components based on correlation coefficients
[0171] Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient were used to evaluate the correlation between the temperature of core components and thermal simulation parameters, and temperature-sensitive features were selected.
[0172] The Pearson correlation coefficient is used to measure the correlation between two sets of data.
[0173]
[0174] Where cov(X,Y) is the covariance of X and Y, and σ X ,σ Y ρ represents the standard deviations of X and Y, respectively. X,Y The closer the value is to 1, the stronger the correlation between the two sets of data.
[0175] Kendall's rank correlation coefficient is derived by calculating the correlation coefficient between random variables X and Y using the number of rank transpositions. The formula is:
[0176]
[0177] Where n is the number of samples, and ∑ is the total number of transpositions. K Between [-1, 1], when r K When r = 1, it indicates that the two random variables have consistent rank correlation; when r = 1, it indicates that the two random variables have consistent rank correlation. K When r = -1, it indicates that the two random variables have completely opposite rank correlations; when r K When = 0, it means that the two random variables are independent.
[0178] Spearman correlation coefficient is used to analyze linear correlation between two variables based on their ranks. The Spearman correlation coefficient (SRCC) between two random variables X and Y is defined as follows:
[0179]
[0180] Where n is the number of samples, r i and s i They represent x respectively i and y i The rank of r. s Between [-1, 1], when one of two random variables is monotonically increasing as the other variable increases, r... s =1; conversely, when one variable monotonically decreases as another variable decreases, r s =-1.
[0181] The correlation threshold was set at 0.1. Based on the correlation coefficient method described above, the correlation between the core component temperature and the thermal simulation parameter was calculated. A correlation greater than 0.1 indicates a strong correlation between the thermal simulation parameter and the component temperature; therefore, this thermal simulation parameter is selected. Figure 5 As shown.
[0182] Based on the correlation results, influencing factors with correlation scores greater than 0.1 were selected, and it was considered that they had a high correlation with temperature. For core devices such as CPU, FPGA, and MOSFET, the temperature-sensitive feature set M = {power consumption of electronic devices, power consumption of circuit boards, ambient temperature, altitude, and opening ratio of the casing}.
[0183] S22: A Random Forest-Based Method for Selecting Temperature-Sensitive Features of Components
[0184] The random forest method was used to select features that affect the temperature of the core components of the device.
[0185] Decision trees are constructed based on a heuristic strategy that considers impurity. The impurity of a dataset can be represented by its entropy value; the higher the information entropy, the lower the data purity. The information entropy of dataset Q is:
[0186]
[0187] In the formula, c represents the number of data categories, and p i This represents the proportion of the i-th category of data in the total data.
[0188] After the effect of feature A (such as ambient temperature), the sample Q is divided into n subsets Q. j Let j = 1, 2, ..., n, and let the information entropy of sample Q be expressed as follows:
[0189]
[0190] In the formula, j represents the j-th category.
[0191] Information gain can be obtained in the following ways:
[0192] G(Q,A)=H(Q)-H A (Q) (6)
[0193] The greater the information gain, the better the classification effect using feature A.
[0194] The contribution threshold for temperature characteristic parameters is set to 0.1. Based on the random forest selection method, the contribution of thermal simulation parameters to the temperature of core components is calculated. A thermal simulation parameter with a contribution greater than 0.1 is considered to have a significant impact on the component temperature and is therefore selected. Figure 6 As shown.
[0195] This invention evaluates the contribution of different features in the feature vector to regression prediction based on random forest. Through... Figure 6 It can be seen that the second temperature-sensitive feature set N = {device power consumption, board power consumption, altitude, shell opening ratio, ambient temperature}, among which the influence of ambient temperature is the most prominent.
[0196] S3: Taking into account both the correlation coefficient and the selection results obtained from the random forest, determine the temperature sensitivity matrix Ω.
[0197] Ω=M∩N={device power consumption, board power consumption, altitude, enclosure aperture ratio, ambient temperature}
[0198] This invention proposes a device-level thermal simulation modeling method applicable to different types of protection equipment, constructs a thermal simulation model of a typical protection device, and obtains the internal and external excitation factors affecting the temperature field distribution of the device and the temperature distribution characteristics of its core components through simulation analysis.
[0199] This invention proposes a method for selecting temperature-sensitive features of core components in relay protection devices. It uses a combination of correlation coefficient method and random forest method to select the factors affecting the temperature distribution of relay protection devices and constructs the temperature-sensitive features of key components.
[0200] like Figure 7 As shown, the present invention provides a selection system for the temperature-sensitive characteristics of a relay protection device, the system comprising:
[0201] The initial unit 701 is used to establish a thermal analysis model of the relay protection device based on its structural characteristics; the thermal analysis model includes determined key simulation parameters; preferably, the thermal analysis model includes a housing thermal analysis model and a board thermal analysis model.
[0202] The enclosure analysis model includes: the outer shell, the flow channel baffle, and the back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffle are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area.
[0203] The thermal analysis model of the circuit boards includes: CPU boards, power supply boards, and analog signal acquisition boards; the heat-generating and heat-conducting components on the CPU boards, power supply boards, and analog signal acquisition boards are determined according to the dimensions in the actual products.
[0204] Preferably, the key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature.
[0205] Unit 702 is used to determine the temperature distribution characteristics of the relay protection device based on the thermal analysis model;
[0206] Unit 703 is established to establish a first temperature-sensitive feature set M based on temperature distribution characteristics using a correlation coefficient algorithm; and to establish a second temperature-sensitive feature set N using a random forest algorithm.
[0207] Result unit 704 is used to determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature sensitive feature set M and the second temperature sensitive feature set N.
[0208] Preferably, the establishing unit 703 is used to establish a first temperature-sensitive feature set M based on temperature distribution characteristics and through a correlation coefficient algorithm, and is also used for:
[0209] Calculate the correlation coefficient between random variables X and Y based on the Pearson correlation coefficient:
[0210]
[0211] Where X and Y are the temperature characteristic variable and the component temperature variable, respectively, cov(X,Y) is the covariance of X and Y, and σ X ,σ Y These are the standard deviations of X and Y, respectively.
[0212] The correlation coefficient between random variables X and Y is calculated using the number of rank transpositions based on Kendall's rank correlation coefficient.
[0213]
[0214] Where n is the number of samples, and ∑ is the total number of transpositions;
[0215] The Spearman correlation coefficient is used to calculate the correlation coefficient between random variables X and Y based on the order of their ranks.
[0216]
[0217] Where n is the number of samples, i is the index of the variable element, and r i and s i They represent x respectively i and y i rank, x i and y i Let be the i-th elements of random variables X and Y, respectively. and x i and y i The average rank;
[0218] When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factors all exceed the preset threshold, the influencing factors will be regarded as temperature-sensitive features.
[0219] Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
[0220] Preferably, the first temperature-sensitive feature set M includes: power consumption of electronic devices, power consumption of circuit boards, ambient temperature, altitude, and opening ratio of the casing.
[0221] Preferably, the establishment unit 703 is used to establish a second temperature-sensitive feature set N using a random forest algorithm, and is also used to:
[0222] The information entropy of dataset Q is:
[0223]
[0224] Where c represents the number of data categories, p i This represents the proportion of the i-th category of data in the total data;
[0225] After feature A is applied, the dataset Q is divided into n subsets Q. j For j = 1, 2, ..., n, the information entropy of dataset Q is expressed as:
[0226]
[0227] Where j represents the j-th category, p j H(Q) represents the proportion of the j-th category of data in the total data; j ) represents the information entropy of the j-th category sample;
[0228] Information gain is:
[0229] G(Q,A)=H(Q)-H A (Q) (6)
[0230] Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features.
[0231] Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
[0232] Preferably, the second set of temperature-sensitive features N includes: device power consumption, board power consumption, altitude, shell opening ratio, and ambient temperature.
[0233] Preferably, the result unit 704 is used to determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, and is also used to:
[0234] Ω=M∩N={device power consumption, board power consumption, altitude, shell opening ratio, ambient temperature}.
[0235] The present invention provides a computer-readable storage medium storing a computer program for executing a method for selecting temperature-sensitive features of a relay protection device.
[0236] This invention provides an electronic device, which includes: a processor and a memory; wherein,
[0237] Memory, used to store processor-executable instructions;
[0238] A processor for reading executable instructions from memory and executing the instructions to implement a method for selecting temperature-sensitive features of a relay protection device.
[0239] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.
[0240] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0241] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0242] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0243] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.
[0244] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
[0245] The invention has been described with reference to a few embodiments. However, as will be known to those skilled in the art, and as defined in the appended claims, other embodiments besides those disclosed above fall equivalently within the scope of the invention.
[0246] Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the art, unless otherwise expressly defined herein. All references to “a / the / the [device, component, etc.]” are openly interpreted as at least one instance of said device, component, etc., unless otherwise expressly stated. The steps of any method disclosed herein need not be performed in the exact order disclosed unless explicitly stated otherwise.
Claims
1. A method for selecting the temperature-sensitive characteristics of a relay protection device, the method comprising: Based on the structural characteristics of relay protection devices, a thermal analysis model of relay protection devices is established. The thermal analysis model includes the determined key simulation parameters; The thermal analysis model includes a housing thermal analysis model and a board thermal analysis model; The enclosure analysis model includes: an outer shell, flow channel baffles, and a back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffles are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area. The board thermal analysis model includes: CPU board, power supply board, and analog signal acquisition board; the heat-generating and heat-conducting components on the CPU board, power supply board, and analog signal acquisition board are determined according to the dimensions in the actual product; The key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature. Based on the thermal analysis model, the temperature distribution characteristics of the relay protection device are determined; Based on the aforementioned temperature distribution characteristics, a first temperature-sensitive feature set M is established using a correlation coefficient algorithm; a second temperature-sensitive feature set N is established using a random forest algorithm. The first temperature-sensitive feature set M includes: electronic device power consumption, board power consumption, ambient temperature, altitude, and casing opening ratio. The second temperature-sensitive feature set N includes: electronic device power consumption, board power consumption, altitude, casing opening ratio, and ambient temperature. Based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, the temperature-sensitive matrix Ω of the relay protection device is determined, including: Ω = M∩N = {Power consumption of electronic devices, power consumption of circuit boards, altitude, opening ratio of the casing, ambient temperature}.
2. The method according to claim 1, wherein establishing a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm comprises: Random variables calculated based on Pearson correlation coefficient X , Y Correlation coefficient between them: (1) in, X and Y These are temperature characteristic variables and component temperature variables, respectively. cov ( X , Y )yes X and Y covariance, They are X and Y Standard deviation; Calculating random variables based on the number of rank transpositions using Kendall's rank correlation coefficient. X , Y Correlation coefficient between them: (2) in, n Let be the number of samples, and ∑ be the total number of transpositions; Based on Spearman correlation coefficient and random variables X , Y Calculating the number of rank between random variables X , Y Correlation coefficient between them: (3) in, n For the number of samples, i For the number of the variable element, r i and s i They represent x i and y i rank, x i and y i They are random variables X , Y The i One element, and They are respectively x i and y i The average rank; When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factor all exceed a preset threshold, the influencing factor will be regarded as a temperature-sensitive feature. Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
3. The method according to claim 1, wherein establishing the second temperature-sensitive feature set N using the random forest algorithm comprises: Dataset Q The information entropy is: (4) Where c represents the number of data categories. p i Indicates the first i The proportion of each category of data in the total data; In features A After the effect, the dataset Q Classified as n Subdatasets Q j , j =1, 2, ..., n Dataset Q Information entropy is represented as: (5) in, j Indicates the first j Categories Indicates the first j The proportion of each category of data in the total data; H ( Q j ) indicates the first j Information entropy of each category of samples; Information gain is: (6) Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features. Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
4. A selection system for the temperature-sensitive characteristics of a relay protection device, the system comprising: The initial unit is used to establish a thermal analysis model of the relay protection device based on its structural characteristics. The thermal analysis model includes the determined key simulation parameters; The thermal analysis model includes a housing thermal analysis model and a board thermal analysis model; The enclosure analysis model includes: an outer shell, flow channel baffles, and a back panel assembly mounting area. The dimensions and specifications of each opening in the flow channel baffles are determined based on the dimensions and specifications in the actual product. The back panel assembly mounting area retains the original internal skeleton and the airflow vortex area. The board thermal analysis model includes: CPU board, power supply board, and analog signal acquisition board; the heat-generating and heat-conducting components on the CPU board, power supply board, and analog signal acquisition board are determined according to the dimensions in the actual product; The key simulation parameters include: thermal conductivity, component power consumption, and ambient temperature. A determination unit is used in the thermal analysis model to determine the temperature distribution characteristics of the relay protection device; A unit is established to establish a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm; and to establish a second temperature-sensitive feature set N using a random forest algorithm. The first temperature-sensitive feature set M includes: electronic device power consumption, board power consumption, ambient temperature, altitude, and casing opening ratio; the second temperature-sensitive feature set N includes: electronic device power consumption, board power consumption, altitude, casing opening ratio, and ambient temperature. The result unit is used to determine the temperature sensitivity matrix Ω of the relay protection device based on the first temperature-sensitive feature set M and the second temperature-sensitive feature set N, and is also used to: Ω = M∩N = {Power consumption of electronic devices, power consumption of circuit boards, altitude, opening ratio of the casing, ambient temperature}.
5. The system according to claim 4, wherein the establishing unit is configured to establish a first temperature-sensitive feature set M based on the temperature distribution characteristics using a correlation coefficient algorithm, and is further configured to: Random variables calculated based on Pearson correlation coefficient X , Y Correlation coefficient between them: (1) in, X and Y These are temperature characteristic variables and component temperature variables, respectively. cov ( X , Y )yes X and Y covariance, They are X and Y Standard deviation; Calculating random variables based on the number of rank transpositions using Kendall's rank correlation coefficient. X , Y Correlation coefficient between them: (2) in, n Let be the number of samples, and ∑ be the total number of transpositions; The Spearman correlation coefficient is used to calculate random variables based on the number of ranks between two variables. X , Y Correlation coefficient between them: (3) in, n For the number of samples, i For the number of the variable element, r i and s i They represent x i and y i rank, x i and y i They are random variables. X , Y The i One element, and They are respectively x i and y i Mean rank; When the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman correlation coefficient of the influencing factor all exceed a preset threshold, the influencing factor will be regarded as a temperature-sensitive feature. Based on the determined temperature-sensitive characteristics, a first temperature-sensitive characteristic set M is established.
6. The system according to claim 4, wherein the establishing unit is configured to establish a second temperature-sensitive feature set N using a random forest algorithm, and is further configured to: Dataset Q The information entropy is: (4) in, c represents the number of data categories. p i Indicates the first i The proportion of each category of data in the total data; In features A After the effect, the dataset Q Classified as n Subdatasets Q j , j =1, 2, ..., n Dataset Q Information entropy is represented as: (5) in, j Indicates the first j Categories Indicates the first j The proportion of each category of data in the total data; H ( Q j ) indicates the first j Information entropy of each category of samples; Information gain is: (6) Factors whose information gain exceeds a preset threshold are considered temperature-sensitive features. Based on the determined temperature-sensitive features, a second set of temperature-sensitive features N is established.
7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program for performing the method according to any one of claims 1-3.
8. An electronic device, characterized in that, The electronic device includes: a processor and a memory; wherein... The memory is used to store the processor-executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method of any one of claims 1-3.
Citation Information
Patent Citations
Method and device for acquiring temperature distribution of relay
CN115238601A
Enhanced switchgear monitoring and diagnostics in a protection relay
US20220246378A1
