Method and system for predicting service life of explosion-proof luminaire based on thermal resistance matrix characteristics

Abstract

This invention discloses a method and system for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix features. It relates to the technical field of lifespan prediction, including: deploying temperature sensing nodes in the explosion-proof lighting fixture and simultaneously acquiring time-series temperature data from multiple locations; acquiring structural heat transfer relationships and establishing an initial thermal resistance matrix; extracting features from the time-series temperature data from multiple locations and establishing a feature set; calculating the thermal resistance change increment coefficient of each heat transfer path and iteratively updating the time-varying thermal resistance matrix; performing matrix feature decomposition after topology consistency constraint analysis; and outputting the remaining lifespan prediction result. This invention solves the technical problems in existing technologies, such as the inability to dynamically characterize the time-varying degradation characteristics of the thermal resistance network between multiple thermal units in explosion-proof lighting fixtures and the difficulty in extracting early thermal degradation fingerprint features from multi-location temperature signals for accurate lifespan prediction. It achieves the technical effect of improving the accuracy of remaining lifespan prediction for explosion-proof lighting fixtures and increasing sensitivity to thermal path degradation.

Description

Technical Field

[0001] This invention relates to the field of life prediction technology, specifically to a method and system for predicting the life of explosion-proof lighting fixtures based on thermal resistance matrix characteristics. Background Technology

[0002] The safety and reliability of explosion-proof lights are directly related to production safety. Due to long-term exposure to harsh working conditions such as high temperature, high humidity, dust, vibration, and corrosive gases, the thermal aging and heat accumulation effects of the internal functional thermal units (such as LED chips, driver power supplies, heat sinks, and sealing structures) of explosion-proof lights will significantly accelerate the decline in their insulation performance, material deterioration, and structural failure, ultimately leading to the loss of the explosion-proof performance of the lights or premature scrapping. Accurately predicting the remaining service life of explosion-proof lights is extremely important for preventing safety accidents and optimizing maintenance strategies. Traditional life assessment of explosion-proof lighting fixtures mainly relies on standardized accelerated aging tests or empirical models based on a single temperature threshold. For example, life is estimated by monitoring the highest surface temperature of the enclosure or the cumulative operating time of the power module, combined with the Arrhenius equation. However, this approach has significant limitations. First, it ignores the complex thermal coupling and differences in heat transfer paths between multiple heat sources inside the fixture, resulting in insensitivity to localized heat concentration or thermal channel degradation. Second, it cannot reflect the time-varying characteristics of thermal resistance as it evolves over time; degradation processes such as poor thermal contact, aging of thermally conductive materials, or micro-deformation of the structure can alter the topology of the thermal resistance network. Third, it lacks in-depth analysis of the dynamic characteristics such as phase, response speed, and differences between temperature signals at multiple locations, making it difficult to detect subtle signs of thermal degradation early. Furthermore, IoT-based sensing monitoring methods cannot dynamically characterize the evolution of the thermal resistance matrix and extract degradation fingerprint features for life prediction in explosion-proof lighting systems with multiple functional thermal units and strong thermal interactions between units.

[0003] Therefore, current technologies have limitations in dynamically characterizing the time-varying degradation characteristics of the thermal resistance network between multiple thermal units in explosion-proof lighting fixtures, and in extracting early thermal degradation fingerprint features from temperature signals at multiple locations for accurate lifetime prediction. Summary of the Invention

[0004] This application provides a method and system for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix characteristics. This solves the technical problems in the prior art, such as the inability to dynamically characterize the time-varying degradation characteristics of the thermal resistance network between multiple thermal units of explosion-proof lighting fixtures and the difficulty in extracting early thermal degradation fingerprint features from temperature signals at multiple locations for accurate lifespan prediction. This achieves the technical effect of improving the accuracy of predicting the remaining lifespan of explosion-proof lighting fixtures and increasing their sensitivity to thermal path degradation.

[0005] This application provides a method for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix features. The method includes: deploying temperature sensing nodes on multiple functional thermal units of the explosion-proof lighting fixture, wherein the temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation; acquiring the structural heat transfer relationship between the functional thermal units and establishing an initial thermal resistance matrix; extracting features from the multi-location temperature time-series data to establish a feature set, wherein the feature set includes temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features; calculating the thermal resistance change increment coefficient of each heat transfer path using the feature set, and iteratively updating the initial thermal resistance matrix according to the calculation results to establish a time-varying thermal resistance matrix; performing topological consistency constraint analysis on the time-varying thermal resistance matrix, and performing matrix feature decomposition processing to extract thermal resistance degradation fingerprint vectors including coupling enhancement features, main heat path bottleneck features, and thermal network stability spectrum features; predicting the lifespan of the explosion-proof lighting fixture based on the thermal resistance degradation fingerprint vectors, and outputting the remaining lifespan prediction result.

[0006] In a possible implementation, the lifespan prediction of explosion-proof lighting fixtures based on thermal resistance degradation fingerprint vectors includes: performing time series alignment processing on thermal resistance degradation fingerprint vectors acquired at consecutive time points within a thermal inertia time window, extracting the degradation increment and degradation rate between adjacent time points, and constructing a dynamic change sequence characterizing the continuous evolution of thermal resistance degradation state over time; performing trend consistency analysis and sliding cumulative fusion processing on the dynamic change sequence to suppress the influence of instantaneous thermal disturbances on degradation characteristics, and then constructing a thermal resistance degradation trajectory characterizing the stable evolution trend of thermal resistance degradation; and using the thermal resistance degradation trajectory to predict the lifespan of explosion-proof lighting fixtures.

[0007] In a possible implementation, the lifespan prediction of explosion-proof lighting fixtures using the thermal resistance degradation trajectory includes: calling a dataset of thermal resistance characteristic distribution of the explosion-proof lighting fixture under a preset failure state; constructing a thermal instability threshold spatial boundary model based on the calling result, wherein the thermal instability threshold spatial boundary model is used to characterize the critical region where the explosion-proof lighting fixture transitions from a normal conductive state to a thermal instability state; projecting and mapping the thermal resistance degradation trajectory in the feature space to establish a spatial correspondence with the thermal instability threshold spatial boundary model, and establishing the position evolution path of the trajectory in the thermal instability space; calculating the minimum distance and time decay rate between the position evolution path and the thermal instability boundary; constructing a lifespan mapping function using the decay gradient of the minimum distance; and outputting the remaining lifespan prediction result using the lifespan mapping function.

[0008] In possible implementations, the construction of the thermal inertia time window includes: calculating the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient between adjacent time points using multi-location temperature time series data; performing joint evolution analysis on the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient to identify the starting time point when thermal resistance evolution enters the quasi-steady-state input interval; using the starting time point as the starting point of the thermal inertia time window and the moment when the thermal resistance degradation fingerprint vector continuously deviates from the quasi-steady-state baseline characteristics as the ending node of the thermal inertia time window, thus constructing the thermal inertia time window.

[0009] In a possible implementation, the structural heat transfer relationship between functional thermal units is obtained, and an initial thermal resistance matrix is ​​established. This includes: obtaining the spatial connection structure characteristics, material thermal conductivity parameters, and contact surface characteristics between adjacent thermal units of the functional thermal units; establishing the heat transfer topology relationship between the functional thermal units; identifying the heat flow conduction path between each functional thermal unit using the heat transfer topology relationship; calculating the basic thermal resistance value of the heat path based on the heat transfer distance, contact area, and thermal conductivity of the conduction path; and constructing the initial thermal resistance matrix using the basic thermal resistance value of the heat path.

[0010] In possible implementations, feature extraction is performed on multi-location temperature time-series data to establish a feature set. This includes: performing time synchronization processing on the location temperature time-series data corresponding to each functional thermal unit, and establishing corresponding temperature response sequences between adjacent functional thermal units according to the heat transfer topology; calculating the degree of temperature rise offset between nodes using the temperature change at the same time in the corresponding temperature response sequences, and establishing a temperature rise difference feature characterizing the strength of heat transfer; calculating the response time offset of heat transfer from upstream thermal units to downstream thermal units based on the time difference of adjacent functional thermal units reaching the preset temperature rise ratio node, and establishing a thermal response hysteresis feature; and performing time-series correlation analysis on the temperature fluctuation sequences of adjacent functional thermal units to extract the phase offset relationship between the main periods of temperature change, and establishing a phase delay feature between nodes.

[0011] In possible implementations, topological consistency constraint analysis is performed on the time-varying thermal resistance matrix, including: based on the heat transfer topological relationship between functional thermal units, using heat flow conservation constraints, heat path connectivity constraints, and boundary heat dissipation continuity constraints to perform constraint analysis on the time-varying thermal resistance matrix, and establishing a modified thermal resistance matrix that satisfies the physical constraints of the thermal network.

[0012] In possible implementations, matrix feature decomposition processing is performed, including: performing thermal path contribution decomposition processing on the modified thermal resistance matrix, identifying dominant thermal paths and heat flow-limited paths based on the proportion of heat transfer in the overall thermal network; using dominant thermal paths and heat flow-limited paths to establish coupling enhancement features characterizing the degree of heat coupling enhancement and main path bottleneck features characterizing the degree of local heat transfer blockage; performing spectral feature decomposition processing on the modified thermal resistance matrix, extracting the matrix eigenvalue distribution, spectral radius variation trend, and main eigenvector evolution relationship to establish thermal network stability spectrum features characterizing changes in the stable state of the thermal network; and performing time-series correlation mapping on coupling enhancement features, main path bottleneck features, and thermal network stability spectrum features to establish a thermal resistance degradation fingerprint vector.

[0013] In a possible implementation, linear interpolation is performed on the multi-location temperature time series data for data cleaning and outlier removal, and feature extraction is performed based on the processed multi-location temperature time series data.

[0014] This application also provides a life prediction system for explosion-proof lighting fixtures based on thermal resistance matrix features. The system includes: a temperature data acquisition module for deploying temperature sensing nodes on multiple functional thermal units of the explosion-proof lighting fixture, wherein the temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation; an initial thermal resistance matrix establishment module for acquiring the structural heat transfer relationship between functional thermal units and establishing an initial thermal resistance matrix; and a feature extraction module for extracting features from the multi-location temperature time-series data and establishing a feature set, wherein the feature set includes temperature rise difference features, thermal response hysteresis features, and other features. The system includes: a point-to-point phase delay feature module; a time-varying thermal resistance matrix establishment module, used to calculate the thermal resistance change increment coefficient of each heat transfer path using the feature set, and iteratively update the initial thermal resistance matrix based on the calculation results to establish a time-varying thermal resistance matrix; a feature decomposition processing module, used to perform matrix feature decomposition processing on the time-varying thermal resistance matrix after performing topological consistency constraint analysis, and extracting thermal resistance degradation fingerprint vectors including coupling enhancement features, main heat path bottleneck features, and thermal network stability spectrum features; and a lifetime prediction result output module, used to predict the lifetime of explosion-proof lighting fixtures based on the thermal resistance degradation fingerprint vectors and output the remaining lifetime prediction results.

[0015] This application proposes a method and system for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix features. The method involves deploying temperature sensing nodes within the explosion-proof lighting fixture and simultaneously acquiring time-series temperature data from multiple locations. It then obtains the structural heat transfer relationships between functional thermal units and establishes an initial thermal resistance matrix. Features are extracted from the time-series temperature data at multiple locations to establish a feature set. The incremental coefficient of thermal resistance change for each heat transfer path is calculated, and the time-varying thermal resistance matrix is ​​iteratively updated. After performing topology consistency constraint analysis, matrix eigenvalue decomposition is executed. Finally, the remaining lifespan prediction result is output. This method solves the technical problems in existing technologies, such as the inability to dynamically characterize the time-varying degradation characteristics of the thermal resistance network between multiple thermal units in explosion-proof lighting fixtures and the difficulty in extracting early thermal degradation fingerprint features from multi-location temperature signals for accurate lifespan prediction. It achieves the technical effect of improving the accuracy of remaining lifespan prediction for explosion-proof lighting fixtures and increasing sensitivity to thermal path degradation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic flowchart of the explosion-proof lighting fixture life prediction method based on thermal resistance matrix characteristics provided in the embodiments of this application.

[0018] Figure 2 This is a schematic diagram of the explosion-proof lighting fixture life prediction system based on thermal resistance matrix characteristics provided in an embodiment of this application.

[0019] Figure labeling: Temperature data acquisition module 10, initial thermal resistance matrix establishment module 20, feature extraction module 30, time-varying thermal resistance matrix establishment module 40, feature decomposition processing module 50, lifetime prediction result output module 60. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structures, features and effects of the present invention.

[0021] This application provides a method for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix characteristics, such as... Figure 1 As shown, the method includes: Step S100: Temperature sensing nodes are deployed on multiple functional thermal units of the explosion-proof lighting fixture. The temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation.

[0022] Preferably, the functional thermal unit inside the explosion-proof lighting fixture refers to an independent functional component that generates heat or participates in heat transfer during the operation of the lighting fixture, including but not limited to LED light source modules, driver power supply modules, thermally conductive substrates, heat sinks, sealed housings, and wiring cavities. The functional thermal units are spatially interconnected, and heat is transferred sequentially along the contact interface between the units. Temperature sensing nodes are arranged on the surface or inside each functional thermal unit, which may be thermocouples, platinum resistance temperature detectors, or semiconductor temperature sensors. Each node has a unique identification number and can output the temperature value at the corresponding location. Thermal contact between the node and the functional thermal unit is achieved through thermally conductive adhesive, mechanical pressing, or welding. Then, all temperature sensing nodes use the same clock reference source to synchronously acquire the temperature of the functional thermal unit during continuous operation. This refers to the entire working time period from the start of the explosion-proof lighting fixture to the stop of power failure, including the initial temperature rise stage, the stable stage of reaching thermal equilibrium, and the possible temperature fluctuation stage. The temperature output values ​​of all temperature sensing nodes at this moment are arranged in chronological order to determine the multi-location temperature time sequence data.

[0023] Step S200: Obtain the structural heat transfer relationship between functional thermal units and establish an initial thermal resistance matrix.

[0024] Step S200 further includes: acquiring the spatial connection structure characteristics of the functional thermal units, the thermal conductivity parameters of the materials, and the contact surface characteristics between adjacent thermal units; establishing the heat transfer topology relationship between the functional thermal units; identifying the heat flow conduction path between each functional thermal unit using the heat transfer topology relationship; calculating the basic thermal resistance value of the heat path based on the heat transfer distance, contact area, and thermal conductivity of the conduction path; and constructing an initial thermal resistance matrix using the basic thermal resistance value of the heat path.

[0025] Preferably, the spatial connection structure characteristics, material thermal conductivity parameters, and contact surface characteristics between adjacent thermal units of the explosion-proof lighting fixture are obtained. The spatial connection structure characteristics refer to the relative positions and contact relationships of each functional thermal unit in physical space, including topological connections such as the contact between the LED light source module and the thermally conductive substrate, the contact between the thermally conductive substrate and the heat sink, and the contact between the heat sink and the sealed housing; the spatial arrangement order, i.e., the order in which heat is transferred from the heat source outwards through the units; and the geometric dimensions of each functional thermal unit. Material thermal conductivity parameters refer to the inherent thermophysical properties of the materials constituting each functional thermal unit, including thermal conductivity, specific heat capacity, and material density. Different functional thermal... Different materials are used for each unit. For example, the LED light source module substrate is made of aluminum substrate, the heat sink is made of die-cast aluminum, and the sealing shell is made of cast steel or stainless steel. The contact surface characteristics between adjacent thermal units refer to the geometric and physical properties of two directly contacting functional thermal units at the interface, including contact area, contact interface type, contact pressure, surface roughness, and contact thermal resistance coefficient. Then, the heat transfer topology between each functional thermal unit is abstractly represented using graph theory. Each functional thermal unit corresponds to a topological node, and the directed edge represents the direction of heat flow transfer, from the high-temperature unit to the low-temperature unit. A directed edge is established when two units are in physical contact and heat flow can be transferred through the contact surface.

[0026] Preferably, the heat-generating functional thermal unit (such as an LED light source module or a driver power supply module) is used as the starting point for heat flow. The heat flow is identified by traversing the directed edges in the heat transfer topology, identifying the conduction paths between each functional thermal unit. This represents the complete sequence of heat transfer from the initial heat source through several intermediate functional thermal units to the ambient air. Each conduction path includes the functional thermal units traversed and the contact interface between each pair of adjacent units. For each pair of adjacent units in each conduction path, the heat transfer distance refers to the physical distance traveled in the normal direction of the contact interface when heat is transferred from one unit to an adjacent unit. For example, for internal heat conduction within a unit, it refers to the average thickness of the heat traveled within that unit; for the contact interface, it is the thickness of the thermally conductive medium layer, such as the thickness of thermally conductive grease. The contact area is the actual interface area where heat transfer occurs between two adjacent functional thermal units, and the thermal conductivity is selected based on the type of medium in the heat transfer segment. For a segment between adjacent units or within a unit, the heat transfer distance L, contact area A, and thermal conductivity are considered. According to the formula Calculate the basic thermal resistance R of the thermal path. For a complete conduction path, its total basic thermal resistance is the sum of the thermal resistances of each segment on the path.

[0027] Preferably, an initial thermal resistance matrix is ​​constructed using the basic thermal resistance values ​​of the thermal path. Each row and column of the matrix corresponds to a functional thermal unit. The main diagonal elements represent the internal thermal resistance of the i-th functional thermal unit, taking the thermal resistance value along the main heat transfer direction. The off-diagonal elements represent the thermal resistance value along the direct heat transfer path from unit i to unit j. Assuming the functional thermal units are LED (L), substrate (S), heat sink (H), and housing (C), the initial thermal resistance matrix takes the form:

[0028] in, This represents the thermal resistance from which heat flows from the LED to the substrate, calculated based on the heat transfer distance between the LED and the substrate, the contact area, and the interfacial thermal conductivity. This indicates the thermal resistance from which heat flows from the substrate to the heat sink. This indicates the thermal resistance of heat flowing from the radiator to the casing, and the reverse path is as follows: , , Depending on the actual situation, the value may be assigned to infinity.

[0029] Step S300: Extract features from the multi-location temperature time series data and establish a feature set, which includes temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features.

[0030] Step S300 further includes performing data cleaning and outlier removal processing on the multi-location temperature time series data by linear interpolation, and performing feature extraction based on the processed multi-location temperature time series data.

[0031] Preferably, the multi-location temperature time-series data collected by each temperature sensing node is traversed and checked, and linear interpolation is performed for data cleaning, i.e., detection and correction to remove invalid or erroneous data points introduced by sensor failure, transmission interference, data acquisition system jitter, etc. Linear interpolation is used to fill in missing or lost data points in the temperature time-series data. Assuming at time point... The temperature collected is At the point in time The temperature collected is And the midpoint of time ( < < If temperature data is missing, calculate using the following formula. Temperature value at time: Time point and All temperature data are valid. Then, abnormal data points with temperature values ​​that significantly deviate from the normal range of the node under the same operating conditions are identified. These include values ​​that exceed the sensor's range, temperature abrupt changes that cannot be physically realized between adjacent sampling times, and values ​​that do not conform to the physical laws of heat transfer with the temperature values ​​of other nodes at the same time. Abnormal temperature data points are then marked as invalid, deleted, and refilled according to linear interpolation to obtain complete, valid, and continuous multi-location temperature time series data after processing.

[0032] Furthermore, step S300 also includes: performing time synchronization processing on the location temperature time series data corresponding to each functional thermal unit, establishing corresponding temperature response sequences between adjacent functional thermal units according to the heat transfer topology; calculating the degree of temperature rise offset between nodes using the temperature change at the same time in the corresponding temperature response sequences, and establishing temperature rise difference characteristics characterizing the strength of heat transfer; calculating the response time offset of heat transfer from upstream thermal units to downstream thermal units based on the time difference of adjacent functional thermal units reaching the preset temperature rise ratio node, and establishing thermal response hysteresis characteristics; performing time-series correlation analysis on the temperature fluctuation sequences of adjacent functional thermal units, extracting the phase offset relationship between the main periods of temperature change, and establishing phase delay characteristics between nodes.

[0033] Preferably, the location temperature time series data corresponding to each functional thermal unit are time-synchronized, that is, the temperature data of all nodes are resampled according to a unified time base to ensure that all nodes have a corresponding temperature value at each sampling time. Then, all adjacent unit pairs with direct physical contact and heat flowing from one side to the other are identified from the heat transfer topology. For each pair of adjacent units, the temperature data of adjacent functional thermal units are extracted from the time-synchronized data and arranged in chronological order to form a corresponding temperature response sequence. The temperature change of the corresponding temperature response sequence at the same time is calculated, that is, the difference between the upstream unit temperature and the downstream unit temperature at the same time. The degree of temperature rise offset between nodes is calculated to characterize the efficiency and intensity of heat transfer from upstream to downstream. This includes statistically calculating all temperature rise offsets within a continuous time window to obtain the average temperature rise offset, the maximum temperature rise offset, and the standard deviation of the temperature rise offset. Then, a temperature rise difference feature is established to characterize the strength of heat conduction on the current heat transfer path. The smaller the temperature rise difference feature value, the smaller the temperature difference between upstream and downstream, and the smoother the heat transfer. The larger the temperature rise difference feature value, the larger the temperature difference between upstream and downstream, and the more obstructed the heat transfer.

[0034] Preferably, the preset temperature rise ratio node refers to the moment when the temperature of the upstream unit rises to a certain percentage relative to its initial temperature. The preset temperature rise ratio is typically set to 50%. Let the upstream unit's initial temperature at startup be... Rise to the final steady-state temperature Then, the time when the upstream unit temperature meets the following conditions This refers to the preset temperature rise ratio node. Where p is a preset ratio, ranging from 0 to 1, and the response time offset is the time difference between the moment when the downstream unit reaches the same preset temperature rise ratio and the moment when the upstream unit reaches the same ratio. A response time offset greater than 0 indicates that the downstream unit lags behind the upstream unit. The larger the absolute value of the response time offset, the slower the heat transfer response. Then, the response time offsets of multiple preset temperature rise ratio nodes are combined to determine the thermal response lag characteristics.

[0035] Preferably, the original temperature sequence is subjected to high-pass filtering or detrending processing to obtain a temperature fluctuation sequence, which is a time series that retains the periodic fluctuation component after removing the long-term trend and steady-state components from the original temperature time series data. The similarity between the two temperature fluctuation sequences at different time offsets is calculated using a cross-correlation function to determine the time delay of one sequence relative to the other. The main period of temperature change refers to the period length corresponding to the frequency component with the highest energy in the temperature fluctuation sequence. It is obtained by performing a Fourier transform on the temperature fluctuation sequence, calculating the power spectral density, and identifying the frequency corresponding to the power peak. The phase offset relationship refers to the phase delay of the downstream temperature fluctuation relative to the upstream temperature fluctuation at the frequency corresponding to the main period. The offset of the cross-correlation function with the maximum value is determined as the phase delay, and finally, the phase delay feature between nodes is established. Among them, a phase delay of 0° indicates that the upstream and downstream temperature fluctuations are synchronized, a phase delay of 90° indicates that the downstream fluctuation lags behind the upstream by 1 / 4 of a cycle, and a phase delay of 180° indicates that the downstream fluctuation lags behind the upstream by 1 / 2 of a cycle. The larger the phase delay, the greater the proportion of the total cycle required for heat to be transferred from the upstream to the downstream. Finally, the extracted temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features are integrated to obtain a feature set.

[0036] Step S400: Calculate the thermal resistance change increment coefficient of each heat transfer path using the feature set, and iteratively update the initial thermal resistance matrix based on the calculation results to establish a time-varying thermal resistance matrix.

[0037] Preferably, the thermal resistance change increment coefficient is used to characterize the degree of thermal resistance change of a heat transfer path per unit time, reflecting the degradation rate of thermal conductivity on the path. For each heat transfer path, the thermal resistance change increment coefficient is calculated by fusing the feature values ​​of all adjacent unit pairs contained in the path. Specifically, the temperature rise difference feature, thermal response hysteresis feature, and inter-node phase delay feature of the i-th adjacent unit pair on the path at the current moment are obtained. The feature change of the i-th adjacent unit pair relative to the initial state is calculated, and then the changes of the three features are weighted and fused to obtain the thermal resistance change increment coefficient of the adjacent unit pair. The weight coefficients are set according to the sensitivity of each feature to thermal resistance change and adjusted according to the actual working conditions of the explosion-proof lighting fixture. Features with higher sensitivity should be given greater weight. For example, the weight coefficients of temperature rise difference feature, thermal response hysteresis feature, and inter-node phase delay feature are 0.5 respectively. For heat transfer paths containing multiple adjacent unit pairs, the thermal resistance change increment coefficient is the sum or weighted average of the coefficients of each segment. Then, the initial thermal resistance matrix is ​​iteratively updated according to the thermal resistance change increment coefficient. That is, the elements in the thermal resistance matrix are successively corrected at discrete time steps (e.g., every 1 minute, every 10 minutes, or every 1 hour). The original thermal resistance value is multiplied by (1 + thermal resistance change increment coefficient) to obtain the updated thermal resistance value, which reflects the true state of thermal resistance at the current moment. If the thermal resistance change increment coefficient is greater than 0, it indicates that the thermal resistance increases and the heat transfer performance degrades. If the thermal resistance change increment coefficient is less than 0, it indicates that the thermal resistance decreases and the heat transfer performance improves. Then, a time-varying thermal resistance matrix is ​​updated and established to record the dynamic evolution trajectory of thermal resistance on each heat transfer path inside the explosion-proof lighting fixture during the entire operation process from start-up to the current moment. Each element changes with time, reflecting the degradation process of the thermal resistance network.

[0038] Step S500: After performing topological consistency constraint analysis on the time-varying thermal resistance matrix, perform matrix eigenvalue decomposition to extract thermal resistance degradation fingerprint vectors, including coupling enhancement features, main thermal path bottleneck features, and thermal network stability spectrum features.

[0039] Step S500 further includes performing constraint analysis of the time-varying thermal resistance matrix based on the heat transfer topology relationship between functional thermal units, using heat flow conservation constraints, heat path connectivity constraints, and boundary heat dissipation continuity constraints, to establish a modified thermal resistance matrix that satisfies the physical constraints of the thermal network.

[0040] Preferably, the time-varying thermal resistance matrix is ​​corrected element-by-element by applying heat flow conservation constraints, thermal path connectivity constraints, and boundary heat dissipation continuity constraints to obtain a corrected thermal resistance matrix that satisfies the physical constraints of the thermal network. This matrix is ​​numerically close to the original time-varying thermal resistance matrix but strictly satisfies the physical laws of the thermal network. Specifically, the heat flow conservation constraint checks whether the heat flow entering each non-heat source unit is equal to the heat flow leaving the unit. If they are not equal, the values ​​of the relevant thermal resistance matrix elements are adjusted to balance the heat flow. The thermal path connectivity constraint checks whether there is at least one complete path with a finite positive thermal resistance value between each heat source unit and each heat dissipation boundary unit. If a path is interrupted, the abnormal thermal resistance value that caused the interruption is replaced with a reasonable value to restore path connectivity. The boundary heat dissipation continuity constraint checks whether the heat transferred from the inside of the shell to the outside surface is equal to the heat lost from the shell surface to the ambient air. If they are not equal, the correct shell-air thermal resistance value is deduced from the actual temperature measurement value and the corresponding element in the original matrix is ​​replaced.

[0041] Furthermore, step S500 also includes performing thermal path contribution decomposition processing on the modified thermal resistance matrix, identifying the dominant thermal path and the heat flow-limited path based on the proportion of heat transfer in the overall thermal network according to different heat transfer paths; using the dominant thermal path and the heat flow-limited path to establish coupling enhancement features characterizing the degree of heat coupling enhancement and main path bottleneck features characterizing the degree of local heat transfer blockage; performing spectral feature decomposition processing on the modified thermal resistance matrix to extract the matrix eigenvalue distribution, spectral radius change trend, and main eigenvector evolution relationship, and establishing thermal network stability spectrum features characterizing the change of the thermal network's stable state; and performing time-series correlation mapping on the coupling enhancement features, main path bottleneck features, and thermal network stability spectrum features to establish a thermal resistance degradation fingerprint vector.

[0042] Preferably, performing thermal path contribution decomposition processing on the modified thermal resistance matrix refers to allocating the total heat flow in the overall thermal network according to each heat transfer path, calculating the proportion of each path in the total heat flow. Specifically, the ambient temperature is set as the reference temperature, the linear equation of the thermal network is solved, and then the heat flow conduction proportion of each heat transfer path is calculated and the paths are sorted from largest to smallest proportion. Then, the heat transfer paths that undertake the main heat flow transmission task in the overall thermal network are identified as dominant heat paths, such as the heat flow conduction proportion exceeding a preset threshold (e.g., 20%), the top K paths in terms of proportion (e.g., K=3), and the top few paths with a cumulative proportion exceeding 80%. Then, the heat transfer paths with a heat flow conduction proportion significantly lower than the design expectation or with an abnormally high thermal resistance value are identified as heat flow restricted paths, such as the current heat flow conduction proportion decreasing by more than a preset threshold (e.g., 50%) compared to the initial state, the current thermal resistance value increasing by more than a preset threshold (e.g., 100%) compared to the initial thermal resistance value, and anomalies such as excessive temperature rise difference, excessive response lag, or excessive phase delay in a certain segment of the path.

[0043] Preferably, the ratio of the sum of the heat transfer ratios of all dominant heat paths at the current moment to the sum of the heat transfer ratios of all dominant heat paths at the initial moment is calculated to determine the coupling enhancement feature. This feature is used to characterize the degree of heat redistribution in the heat network due to path degradation. That is, heat that was originally dispersed on multiple paths is forced to concentrate on the remaining dominant heat paths for transmission. A ratio of 1 indicates that the heat distribution state is consistent with the initial state, and a ratio greater than 1 indicates that heat is concentrated on the dominant heat paths. The larger the value, the stronger the coupling of the heat network and the more severe the degradation. Then, the segment with the maximum thermal resistance on the dominant heat path at the current moment is identified, and the thermal resistance growth factor of this segment relative to the initial state is calculated. The value of the most severe bottleneck among all dominant heat paths is taken as the bottleneck feature of the main path. This feature is used to characterize the degree of heat transfer capacity reduction caused by local blockage on the dominant heat path. The larger the value, the more severe the blockage.

[0044] Preferably, the modified thermal resistance matrix undergoes spectral feature decomposition, representing it as eigenvalues ​​and eigenvectors. Then, the eigenvalue distribution is extracted, including the maximum eigenvalue, mean eigenvalue, eigenvalue variance, and eigenvalue entropy; the spectral radius trend, i.e., the spectral radius values ​​at different times arranged in chronological order, where the spectral radius refers to the largest absolute value among all eigenvalues ​​of the thermal resistance matrix; and the principal eigenvector evolution relationship, i.e., the temporal relationship of the principal eigenvector components at different times, including the rate of change and concentration of the principal eigenvectors. Then, the eigenvalue distribution, spectral radius trend, and principal eigenvector evolution relationship are combined to determine the thermal network stability spectral features, used to characterize the stable state of the thermal network. Smaller and more gradual changes in each indicator indicate a stable thermal network, while increased eigenvalues, accelerated growth of the spectral radius, and significant changes in the principal eigenvectors indicate that the thermal network is developing towards instability. Finally, the coupling enhancement features, main path bottleneck features, and thermal network stability spectral features are aligned on the time axis and fused into a thermal resistance degradation fingerprint vector, used to uniquely characterize the degradation state of the thermal network at that moment.

[0045] Step S600: Predict the lifespan of the explosion-proof lighting fixture based on the thermal resistance degradation fingerprint vector, and output the remaining lifespan prediction result.

[0046] Step S600 further includes performing time series alignment processing on the thermal resistance degradation fingerprint vectors acquired at consecutive time points within the thermal inertia time window, extracting the degradation increment change and degradation rate between adjacent time points, and constructing a dynamic change sequence characterizing the continuous evolution of the thermal resistance degradation state over time; performing trend consistency analysis and sliding cumulative fusion processing on the dynamic change sequence to suppress the influence of instantaneous thermal disturbances on the degradation characteristics, and then constructing a thermal resistance degradation trajectory characterizing the stable evolution trend of thermal resistance degradation; and using the thermal resistance degradation trajectory to predict the lifespan of explosion-proof lighting fixtures.

[0047] Preferably, the thermal inertia time window refers to the time interval during which the thermal resistance degradation fingerprint vector enters the quasi-stable evolution stage after the explosion-proof lighting fixture reaches thermal equilibrium from startup. The thermal resistance degradation fingerprint vector within the thermal inertia time window is time-series aligned, arranged in chronological order, and the time interval between adjacent moments is equal. Then, it is checked whether there is any missing data at each moment. If there is any missing data, linear interpolation or forward padding is used to fill in the missing data. Then, the degradation increment change between adjacent moments is extracted, that is, the change difference of each component of the thermal resistance degradation fingerprint vector between two adjacent moments, as well as the degradation change rate, and the magnitude of the degradation increment change per unit time, that is, the degradation increment change divided by the time interval. The degradation increment change and degradation change rate at each moment are organized in chronological order to form a dynamic change sequence, which represents the continuous evolution of the thermal resistance degradation state over time. Next, the direction of change of each component in the dynamic change sequence over time is checked to identify a stable degradation trend and eliminate random fluctuations and instantaneous disturbances. Then, a sliding time window is set, for example, containing 10 sampling moments, and the degradation increment of the dynamic change sequence is accumulated, summed, and weighted averaged to suppress instantaneous disturbances, highlight the long-term evolution trend, construct the thermal resistance degradation trajectory, characterize the stable evolution trend of thermal resistance degradation, and finally use the thermal resistance degradation trajectory as the input for life prediction to predict the life of explosion-proof lighting fixtures.

[0048] Furthermore, step S600 also includes: calculating the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient between adjacent time points using multi-location temperature time series data; performing joint evolution analysis on the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient to identify the starting time point when thermal resistance evolution enters the quasi-steady-state input interval; using the starting time point as the starting point of the thermal inertia time window and the moment when the thermal resistance degradation fingerprint vector continuously deviates from the quasi-steady-state baseline characteristics as the ending node of the thermal inertia time window to construct the thermal inertia time window.

[0049] Preferably, the consistency coefficient of thermal response change between adjacent time points is calculated using multi-location temperature time series data. This coefficient characterizes whether the temperature change direction of all functional thermal units is consistent between adjacent time points, i.e., whether the units heat up or cool down simultaneously. Specifically, the temperature change of each temperature sensing node between adjacent time points is calculated, the sign of the change direction of each node is determined, the proportion of nodes in the same direction is calculated, and the consistency coefficient of thermal response change is determined. Then, the thermal diffusion hysteresis stability coefficient is calculated to characterize whether the time delay of heat transfer from upstream unit to downstream unit tends to stabilize, i.e., whether the heat transfer process has entered a quasi-steady state. Specifically, a pair of adjacent functional thermal units is selected, the response relationship of upstream and downstream temperature changes between adjacent time points is calculated, the thermal diffusion hysteresis response coefficient of a single pair of units is determined, and the average hysteresis response coefficient of all adjacent unit pairs is calculated and its reciprocal standard deviation is used as the thermal diffusion hysteresis stability coefficient. Then, the changes in the thermal response consistency coefficient and the thermal diffusion hysteresis stability coefficient over time are comprehensively analyzed to determine whether the thermal network has entered the quasi-steady-state evolution stage from the transient response stage at the initial stage of startup. The starting time point of the thermal resistance evolution entering the quasi-steady propagation interval is identified. The quasi-steady propagation interval refers to the stage in which the explosion-proof lighting thermal network transitions from the unstable state at the initial stage of startup to a relatively stable state. Next, the Euclidean distance between the thermal resistance degradation fingerprint vector of the current thermal network and the baseline is calculated as the deviation of the thermal resistance degradation fingerprint vector from the quasi-steady-state baseline characteristics. The first moment that meets the continuous deviation condition is taken as the termination node of the thermal inertia time window. Finally, the thermal inertia time window is determined to ensure that degradation trajectory extraction and lifetime prediction are only performed after the thermal network enters the stable evolution stage.

[0050] Furthermore, step S600 also includes: calling the thermal resistance characteristic distribution dataset of the explosion-proof lighting fixture under a preset failure state; constructing a thermal instability threshold spatial boundary model based on the calling result, wherein the thermal instability threshold spatial boundary model is used to characterize the critical region where the explosion-proof lighting fixture transitions from a normal conduction state to a thermal instability state; projecting and mapping the thermal resistance degradation trajectory in the feature space to establish a spatial correspondence with the thermal instability threshold spatial boundary model, and establishing the position evolution path of the trajectory in the thermal instability space; calculating the minimum distance and time decay rate between the position evolution path and the thermal instability boundary; constructing a lifetime mapping function using the decay gradient of the minimum distance; and outputting the remaining service life prediction result using the lifetime mapping function.

[0051] Preferably, the preset failure state refers to the state in which the explosion-proof lighting fixture loses its normal explosion-proof performance or lighting function. This includes thermal resistance degradation causing the surface temperature of the housing to exceed the maximum allowable temperature specified by the explosion-proof standard, complete blockage of the thermal path, inability to effectively dissipate heat, triggering over-temperature protection or thermal fuse, and the thermal network stability spectrum characteristics reaching a preset critical threshold. The thermal resistance feature distribution dataset is pre-established by conducting accelerated aging tests on explosion-proof lighting fixtures of the same model, continuously monitoring until the lighting fixture fails, recording the thermal resistance degradation fingerprint vector at the time of failure, or collecting historical data from actual failure cases on site and extracting the thermal resistance features before failure. The thermal resistance feature distribution dataset is then called to construct a thermal instability threshold space boundary model, which is used to characterize the critical region where the explosion-proof lighting fixture transitions from a normal conduction state to a thermal instability state, in order to distinguish between the normal conduction state and the thermal instability state. Specifically, each sample in the thermal resistance feature distribution dataset is regarded as a point in the feature space, and the convex hull of these points is calculated, which is the smallest convex polyhedron containing all failure points. The boundary of the convex hull is the thermal instability threshold space boundary.

[0052] Preferably, each point in the thermal resistance degradation trajectory is placed in the same feature space as the thermal instability threshold space boundary model, ensuring that the coordinate system of the trajectory point is consistent with the coordinate system of the boundary model. If the feature dimensions are different, dimensionality reduction and alignment are performed through principal component analysis or linear transformation. The Euclidean distance from each trajectory point to the center of the sphere is calculated to determine the spatial correspondence with the thermal instability threshold space boundary model, that is, the positional relationship of each trajectory point relative to the thermal instability boundary, such as inside, on the boundary, or outside. Then, the trajectory curve traversed by the trajectory point as it moves in the feature space over time is calculated to determine the positional evolution path of the trajectory in the thermal instability space. Then, the minimum distance between the trajectory point and the thermal instability boundary is calculated using the positional evolution path, that is, the shortest Euclidean distance from the trajectory point to the thermal instability threshold space boundary at the current moment, as well as the time decay rate, that is, how fast the minimum distance changes over time. The decay gradient refers to the decay rate of the minimum distance over time, reflecting the speed at which the explosion-proof lighting fixture approaches the failure state. The decay gradient of the minimum distance is used to construct a lifetime mapping function to map the current degradation state to the remaining lifetime. For example, the current degradation trajectory is matched with the degradation trajectory of historical failure samples to determine the most similar historical trajectory, and its remaining lifetime is used as the current prediction value. Finally, the prediction result of the remaining lifetime is output. When the predicted remaining lifetime is lower than the preset threshold, an early warning signal is issued.

[0053] In the above text, refer to Figure 1 This paper describes in detail a method for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix characteristics according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a life prediction system for explosion-proof lighting fixtures based on thermal resistance matrix characteristics, according to an embodiment of the present invention.

[0054] The explosion-proof lighting fixture life prediction system based on thermal resistance matrix characteristics according to embodiments of the present invention addresses the technical problems in the prior art, namely, the inability to dynamically characterize the time-varying degradation characteristics of the thermal resistance network between multiple thermal units of explosion-proof lighting fixtures, and the difficulty in extracting early thermal degradation fingerprint features from multi-location temperature signals for accurate life prediction. This system achieves the technical effects of improving the accuracy of remaining life prediction for explosion-proof lighting fixtures and increasing their sensitivity to thermal path degradation. Figure 2 As shown, the explosion-proof lighting fixture life prediction system based on thermal resistance matrix features includes: a temperature data acquisition module 10, an initial thermal resistance matrix establishment module 20, a feature extraction module 30, a time-varying thermal resistance matrix establishment module 40, a feature decomposition processing module 50, and a life prediction result output module 60.

[0055] The temperature data acquisition module 10 is used to deploy temperature sensing nodes on multiple functional thermal units of the explosion-proof lighting fixture. The temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation. The initial thermal resistance matrix establishment module 20 is used to acquire the structural heat transfer relationship between functional thermal units and establish an initial thermal resistance matrix. The feature extraction module 30 is used to extract features from the multi-location temperature time-series data and establish a feature set, which includes temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features. The time-varying thermal resistance matrix establishment module 40 is used to calculate the thermal resistance change increment coefficient of each heat transfer path using the feature set, and iteratively update the initial thermal resistance matrix according to the calculation results to establish a time-varying thermal resistance matrix. The feature decomposition processing module 50 is used to perform matrix feature decomposition processing on the time-varying thermal resistance matrix after performing topological consistency constraint analysis, and extract thermal resistance degradation fingerprint vectors including coupling enhancement features, main heat path bottleneck features, and thermal network stability spectrum features. The lifetime prediction result output module 60 is used to predict the lifetime of the explosion-proof lighting fixture based on the thermal resistance degradation fingerprint vectors and output the remaining lifetime prediction result.

[0056] The specific configuration of the lifetime prediction result output module 60 will be described in detail below. The lifetime prediction result output module 60 further includes: performing time series alignment processing on the thermal resistance degradation fingerprint vectors acquired at consecutive time points within a thermal inertia time window; extracting the degradation increment and degradation rate between adjacent time points; constructing a dynamic change sequence characterizing the continuous evolution of the thermal resistance degradation state over time; performing trend consistency analysis and sliding cumulative fusion processing on the dynamic change sequence to suppress the influence of instantaneous thermal disturbances on the degradation characteristics; and constructing a thermal resistance degradation trajectory characterizing the stable evolution trend of thermal resistance degradation; and using the thermal resistance degradation trajectory to predict the lifetime of explosion-proof lighting fixtures.

[0057] The specific configuration of the lifetime prediction result output module 60 will be described in detail below. The lifetime prediction result output module 60 further includes: calling a thermal resistance characteristic distribution dataset of the explosion-proof lighting fixture under a preset failure state; constructing a thermal instability threshold spatial boundary model based on the calling result, wherein the thermal instability threshold spatial boundary model is used to characterize the critical region where the explosion-proof lighting fixture transitions from a normal conductive state to a thermal instability state; projecting and mapping the thermal resistance degradation trajectory in the feature space to establish a spatial correspondence with the thermal instability threshold spatial boundary model, and establishing the position evolution path of the trajectory in the thermal instability space; calculating the minimum distance and time decay rate between the position evolution path and the thermal instability boundary; constructing a lifetime mapping function using the decay gradient of the minimum distance; and outputting the remaining lifetime prediction result using the lifetime mapping function.

[0058] The specific configuration of the lifetime prediction result output module 60 will be described in detail below. The lifetime prediction result output module 60 further includes: calculating the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient between adjacent time points using multi-location temperature time series data; performing joint evolution analysis on the thermal response change consistency coefficient and thermal diffusion hysteresis stability coefficient to identify the starting time point when thermal resistance evolution enters the quasi-steady-state input interval; and constructing the thermal inertia time window by using the starting time point as the starting point of the thermal inertia time window and the moment when the thermal resistance degradation fingerprint vector continuously deviates from the quasi-steady-state baseline characteristics as the ending node of the thermal inertia time window.

[0059] The specific configuration of the initial thermal resistance matrix establishment module 20 will be described in detail below. The initial thermal resistance matrix establishment module 20 further includes: acquiring the spatial connection structure characteristics of functional thermal units, material thermal conductivity parameters, and contact surface characteristics between adjacent thermal units; establishing the heat transfer topology between functional thermal units; identifying the heat flow conduction path between each functional thermal unit using the heat transfer topology; calculating the basic thermal resistance value of the heat path based on the heat transfer distance, contact area, and thermal conductivity of the conduction path; and constructing the initial thermal resistance matrix using the basic thermal resistance value of the heat path.

[0060] The specific configuration of the feature extraction module 30 will be described in detail below. The feature extraction module 30 further includes: performing time synchronization processing on the location temperature time series data corresponding to each functional thermal unit, establishing corresponding temperature response sequences between adjacent functional thermal units according to the heat transfer topology; calculating the degree of temperature rise offset between nodes using the temperature change at the same time in the corresponding temperature response sequences, and establishing temperature rise difference features characterizing the strength of heat transfer; calculating the response time offset of heat transfer from upstream thermal units to downstream thermal units based on the time difference between adjacent functional thermal units reaching the preset temperature rise ratio node, and establishing thermal response hysteresis features; performing time-series correlation analysis on the temperature fluctuation sequences of adjacent functional thermal units, extracting the phase shift relationship between the main periods of temperature change, and establishing phase delay features between nodes.

[0061] The specific configuration of the feature decomposition processing module 50 will be described in detail below. The feature decomposition processing module 50 further includes: performing constraint analysis of the time-varying thermal resistance matrix based on the heat transfer topology relationship between functional thermal units, using heat flow conservation constraints, heat path connectivity constraints, and boundary heat dissipation continuity constraints, to establish a modified thermal resistance matrix that satisfies the physical constraints of the thermal network.

[0062] The specific configuration of the feature decomposition processing module 50 will be described in detail below. The feature decomposition processing module 50 further includes: performing thermal path contribution decomposition processing on the modified thermal resistance matrix; identifying dominant thermal paths and heat flow-limited paths based on the proportion of heat transfer in the overall thermal network according to different heat transfer paths; establishing coupling enhancement features characterizing the degree of heat coupling enhancement and main path bottleneck features characterizing the degree of local heat transfer blockage using the dominant thermal paths and heat flow-limited paths; performing spectral feature decomposition processing on the modified thermal resistance matrix; extracting the matrix eigenvalue distribution, spectral radius variation trend, and main eigenvector evolution relationship to establish thermal network stability spectrum features characterizing changes in the stable state of the thermal network; and performing time-series correlation mapping on the coupling enhancement features, main path bottleneck features, and thermal network stability spectrum features to establish a thermal resistance degradation fingerprint vector.

[0063] The specific configuration of the feature extraction module 30 will be described in detail below. The feature extraction module 30 further includes: performing data cleaning and outlier removal processing on the multi-location temperature time series data by linear interpolation, and performing feature extraction based on the processed multi-location temperature time series data.

[0064] The explosion-proof lighting fixture life prediction system based on thermal resistance matrix characteristics provided in this embodiment of the invention can execute the explosion-proof lighting fixture life prediction method based on thermal resistance matrix characteristics provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0065] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for predicting the lifetime of an explosion-proof luminaire based on thermal resistance matrix characteristics, characterized by, The method includes: Temperature sensing nodes are deployed on multiple functional thermal units of the explosion-proof lighting fixture. The temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation. Obtain the structural heat transfer relationship between functional thermal units and establish the initial thermal resistance matrix; Feature extraction is performed on multi-location temperature time-series data to establish a feature set, which includes temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features. The thermal resistance change increment coefficient of each heat transfer path is calculated using the feature set. The initial thermal resistance matrix is ​​iteratively updated based on the calculation results to establish a time-varying thermal resistance matrix. After performing topological consistency constraint analysis on the time-varying thermal resistance matrix, matrix eigenvalue decomposition is performed to extract thermal resistance degradation fingerprint vectors, including coupling enhancement features, main thermal path bottleneck features, and thermal network stability spectrum features. The lifespan of explosion-proof lighting fixtures is predicted based on the thermal resistance degradation fingerprint vector, and the remaining lifespan prediction result is output.

2. The method for predicting the lifetime of an explosion-proof luminaire based on thermal resistance matrix characteristics according to claim 1, characterized in that, Lifespan prediction of explosion-proof lighting fixtures based on thermal resistance degradation fingerprint vectors includes: Within the thermal inertia time window, the thermal resistance degradation fingerprint vectors acquired at consecutive time points are aligned with the time series. The degradation increment and degradation rate between adjacent time points are extracted to construct a dynamic change sequence characterizing the continuous evolution of the thermal resistance degradation state over time. Trend consistency analysis and sliding cumulative fusion processing are performed on the dynamic change sequence to suppress the influence of instantaneous thermal disturbance on the degradation characteristics, and then a thermal resistance degradation trajectory characterizing the stable evolution trend of thermal resistance degradation is constructed. The lifespan of explosion-proof lighting fixtures is predicted using the aforementioned thermal resistance degradation trajectory.

3. The method of predicting the lifetime of an explosion-proof luminaire based on thermal resistance matrix characteristics according to claim 2, wherein, Predicting the lifespan of explosion-proof lighting fixtures using the aforementioned thermal resistance degradation trajectory includes: Access the thermal resistance characteristic distribution dataset of explosion-proof lighting fixtures under preset failure states; Based on the call results, a thermal instability threshold spatial boundary model is constructed. The thermal instability threshold spatial boundary model is used to characterize the critical region where the explosion-proof lighting fixture transitions from a normal conductive state to a thermal instability state. The thermal resistance degradation trajectory is projected and mapped in the feature space to establish a spatial correspondence with the thermal instability threshold spatial boundary model, and the position evolution path of the trajectory in the thermal instability space is established. The minimum distance and time decay rate between the location evolution path and the thermal instability boundary are calculated. A lifetime mapping function is constructed using the attenuation gradient of the minimum distance, and the remaining lifetime prediction result is output using the lifetime mapping function.

4. The method of predicting the lifetime of an explosion-proof luminaire based on thermal resistance matrix characteristics according to claim 2, wherein, The construction of the thermal inertia time window includes: The consistency coefficient of thermal response change and the thermal diffusion hysteresis stability coefficient between adjacent time points are calculated using multi-location temperature time series data. A joint evolution analysis of the thermal response change consistency coefficient and the thermal diffusion hysteresis stability coefficient was conducted to identify the starting time point when the thermal resistance evolution entered the quasi-stable propagation interval. The thermal inertia time window is constructed by taking the starting point as the starting point of the thermal inertia time window and the moment when the thermal resistance degradation fingerprint vector continuously deviates from the quasi-steady-state baseline characteristics as the ending point of the thermal inertia time window.

5. The method of predicting the life of an explosion-proof luminaire based on thermal resistance matrix characteristics according to claim 1, wherein, Obtain the structural heat transfer relationship between functional thermal units and establish the initial thermal resistance matrix, including: Acquire the spatial connection structure characteristics, material thermal conductivity parameters, and contact surface characteristics between adjacent thermal units of the functional thermal units, and establish the heat transfer topology between the functional thermal units. The heat transfer topology is used to identify the heat flow conduction path between each functional thermal unit, and the basic thermal resistance of the heat path is calculated based on the heat transfer distance, contact area, and thermal conductivity of the conduction path. An initial thermal resistance matrix is ​​constructed using the basic thermal resistance values ​​of the aforementioned thermal path.

6. The method of predicting the life of an explosion-proof luminaire based on thermal resistance matrix characteristics according to claim 1, wherein, Feature extraction was performed on multi-location temperature time-series data to establish a feature set, including: The location temperature time series data corresponding to each functional thermal unit are processed for time synchronization, and the corresponding temperature response sequence between adjacent functional thermal units is established according to the heat transfer topology. The temperature rise offset between nodes is calculated by using the temperature change at the same time in the corresponding temperature response sequence, and a temperature rise difference feature characterizing the change in the strength of heat transfer is established. The response time offset of heat transfer from upstream to downstream thermal units is calculated based on the time difference between adjacent functional thermal units reaching the preset temperature rise ratio node, and thermal response hysteresis characteristics are established. Temporal correlation analysis was performed on the temperature fluctuation sequences of adjacent functional thermal units to extract the phase shift relationship between the main periods of temperature change and to establish the phase delay characteristics between nodes.

7. The method for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix characteristics as described in claim 1, characterized in that, Topological consistency constraint analysis of the time-varying thermal resistance matrix is ​​performed, including: Based on the heat transfer topology between functional thermal units, constraint analysis of the time-varying thermal resistance matrix is ​​performed using heat flow conservation constraints, heat path connectivity constraints, and boundary heat dissipation continuity constraints to establish a modified thermal resistance matrix that satisfies the physical constraints of the thermal network.

8. The method for predicting the lifespan of explosion-proof lighting fixtures based on thermal resistance matrix characteristics as described in claim 7, characterized in that, Perform matrix eigenvalue decomposition, including: The thermal resistance matrix is ​​decomposed by thermal path contribution, and the dominant thermal path and the heat flow-limited path are identified according to the proportion of heat flow conduction of different heat transfer paths in the overall thermal network. By utilizing the dominant thermal path and the heat flow-limited path, we can establish coupling enhancement features that characterize the degree of heat coupling enhancement and main path bottleneck features that characterize the degree of local heat transfer blockage. Perform spectral eigenvalue decomposition on the modified thermal resistance matrix to extract the matrix eigenvalue distribution, spectral radius variation trend, and principal eigenvector evolution relationship, and establish thermal network stability spectral features characterizing the changes in the steady state of the thermal network; A thermal resistance degradation fingerprint vector is established by temporal correlation mapping of coupling enhancement features, main path bottleneck features, and thermal network stability spectrum features.

9. The method of predicting the life of an explosion-proof luminaire based on thermal resistance matrix characteristics as claimed in claim 1, wherein, The multi-location temperature time series data is subjected to linear interpolation for data cleaning and outlier removal, and feature extraction is performed based on the processed multi-location temperature time series data.

10. A system for predicting the lifetime of an explosion-proof luminaire based on thermal resistance matrix characteristics, characterized by, The system is used to implement the explosion-proof lighting fixture life prediction method based on thermal resistance matrix characteristics as described in any one of claims 1 to 9, and the system comprises: A temperature data acquisition module is used to deploy temperature sensing nodes on multiple functional thermal units of the explosion-proof lighting fixture. The temperature sensing nodes are used to synchronously acquire multi-location temperature time-series data of the functional thermal units during continuous operation. The initial thermal resistance matrix establishment module is used to obtain the structural heat transfer relationship between functional thermal units and establish the initial thermal resistance matrix. The feature extraction module is used to extract features from multi-location temperature time-series data and establish a feature set, which includes temperature rise difference features, thermal response hysteresis features, and inter-node phase delay features. The time-varying thermal resistance matrix establishment module is used to calculate the thermal resistance change increment coefficient of each heat transfer path using the feature set, and to iteratively update the initial thermal resistance matrix based on the calculation results to establish the time-varying thermal resistance matrix. The eigenvalue decomposition module is used to perform matrix eigenvalue decomposition after topological consistency constraint analysis of the time-varying thermal resistance matrix, and extract thermal resistance degradation fingerprint vectors including coupling enhancement features, main thermal path bottleneck features and thermal network stability spectrum features. The life prediction result output module is used to predict the life of explosion-proof lighting fixtures based on the thermal resistance degradation fingerprint vector and output the remaining life prediction result.

Images