A power-on device salt spray performance test method

Abstract

The present application relates to the field of electronic equipment reliability testing, and specifically relates to a power-on equipment salt fog performance test method, comprising: obtaining internal heat generation data of the power-on equipment under a plurality of different operating powers; determining the surface temperature difference change distribution under each operating power based on the same; determining the condensation rate distribution and identifying the corrosion critical area through numerical simulation; extracting dynamic change characteristics representing the correlation between power change and condensation rate; adjusting the operating power parameters to build an optimized temperature difference change model; using the model to quantitatively evaluate the influence of corrosive atmosphere on the tolerance of the device; and finally generating a salt fog performance test scheme containing regulation requirements. The present application can realize the dynamic correlation between internal heat generation and corrosion medium condensation rate, accurately locate the salt fog preferential condensation site, and effectively improve the scientificity, prediction accuracy and reliability of the test results.

Description

Technical Field

[0001] This invention relates to the field of electronic equipment reliability testing, and specifically to a method for testing the salt spray performance of powered equipment. Background Technology

[0002] Electrical equipment faces corrosion threats in salt spray environments. Existing salt spray performance testing methods primarily focus on the corrosion resistance of the equipment's external materials, neglecting the impact of internal changes in the equipment's condition during operation on salt spray erosion. In particular, the internal heat generation of equipment varies under different workloads, resulting in internal and external temperature differences. These temperature differences significantly alter the condensation and dew formation rates of salt spray on the equipment surface, leading to accelerated localized corrosion. This dynamic process is not adequately reflected in existing testing methods.

[0003] Therefore, accurately grasping the impact of equipment operating power, internal heat generation, and surface temperature difference changes on salt spray condensation rate has become a key issue in improving the accuracy of salt spray performance testing and truly reflecting the equipment's tolerance to multiple operating conditions. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned shortcomings and provide a method for testing the salt spray performance of electrical equipment.

[0005] To achieve the above objectives, the specific solution of the present invention is as follows: A method for testing the salt spray performance of electrical equipment includes the following specific steps: Step 1: Obtain the internal heat generation data of the powered equipment under multiple different operating power levels. The internal heat generation data includes the heating power and temperature rise data. Step 2: Based on the internal heat generation data, determine the surface temperature difference distribution of the powered equipment at various operating power levels. The temperature difference distribution reflects the surface temperature gradient difference caused by internal heat generation. Step 3: Based on the surface temperature difference distribution under each operating power, the condensation rate distribution of the corrosive medium on the surface of the powered equipment is determined by numerical simulation, and the critical corrosion area is identified based on the condensation rate distribution. The critical corrosion area corresponds to the salt spray preferential condensation site caused by the temperature difference. Step 4: Based on the critical corrosion regions identified under multiple different operating power levels and their corresponding condensation rate distributions, extract dynamic change features characterizing the relationship between power changes and condensation rate changes; Step 5: Compare the dynamic change characteristics with the preset threshold. If the dynamic change characteristics exceed the preset threshold, adjust the operating power parameters of the powered equipment, and re-acquire the internal heat generation data and the corresponding temperature difference change distribution based on the adjusted operating power, so as to iteratively optimize the temperature difference change model that characterizes the dynamic relationship between internal heat generation and the condensation rate of the corrosive medium. Step 6: Using the optimized temperature difference change model, quantitatively evaluate the impact of corrosive atmosphere on the tolerance of the powered equipment, and generate tolerance assessment reports for different operating power levels. Step 7: Based on the tolerance assessment report, simulate the temperature difference evolution and condensation rate changes under various workload scenarios, determine the regulation requirements for the internal heating state of the powered equipment, and generate a salt spray performance test plan that includes the regulation requirements.

[0006] Compared with the prior art, the present invention has the following beneficial effects: This invention overcomes the shortcomings of traditional testing methods that neglect the impact of equipment operating heat effects on the corrosion process by acquiring internal heat generation data of electrical equipment under multiple different operating power levels and determining the surface temperature difference distribution based on this data. Numerical simulation is used to determine the condensation rate distribution and identify key corrosion areas, achieving precise location of areas where salt spray preferentially condenses due to temperature differences, effectively improving the targeting of the test. Dynamic variation characteristics between power changes and condensation rates are extracted to construct a temperature difference variation model. This model is iteratively optimized by comparing dynamic characteristics with thresholds, quantitatively reflecting the nonlinear dynamic correlation between internal heat generation and the condensation rate of the corrosive medium. Especially under conditions of drastic power fluctuations, the model's prediction accuracy is adaptively improved, enhancing its prediction accuracy under multiple operating conditions. The generated endurance assessment report comprehensively considers the coupling effect of workload and temperature difference evolution, realistically reflecting the environmental endurance of electrical equipment in complex service scenarios. The final salt spray performance testing scheme, which includes control requirements, achieves coordinated control of the electrical equipment's heating state and the external corrosive environment, improving the scientific validity and reliability of the test results and providing important data support for the design optimization and life prediction of electrical equipment. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating the overall technical solution architecture of the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 2 This is a schematic diagram of the core principle framework for determining critical corrosion areas based on internal thermal effects in the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 3 This is a flowchart illustrating the process of obtaining internal heat generation data and determining the surface temperature difference distribution based on a three-dimensional heat conduction model in the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 4 This is a flowchart illustrating the process of determining the condensation rate distribution and identifying key corrosion regions through numerical simulation of gas-liquid two-phase flow in the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 5This is a flowchart illustrating the logic of extracting dynamic change features characterizing the relationship between power change and condensation rate in the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 6 This is a schematic diagram of the multi-level interactive logic of the temperature difference change model constructed and optimized based on a deep neural network in the salt spray performance testing method for electrical equipment according to an embodiment of this application. Figure 7 This is a flowchart illustrating the process of introducing a cumulative corrosion damage parameter to quantitatively assess the tolerance of electrical equipment and generate an assessment report in the salt spray performance testing method for electrical equipment according to embodiments of this application. Figure 8 This is a flowchart illustrating the generation of a salt spray performance test scheme based on control requirements and load scenario simulation in the salt spray performance test method for powered equipment according to an embodiment of this application. Detailed Implementation

[0008] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, but this is not to limit the scope of the invention to this.

[0009] Example 1

[0010] like Figures 1 to 8 As shown in this embodiment, a method for testing the salt spray performance of electrical equipment may specifically include the following steps: Step 1: Acquire internal heat generation data of the powered equipment under multiple different operating power levels. This data includes heating power and temperature rise. Specifically, this process is first achieved through multiple temperature sensors deployed on the surface of key heat-generating components inside the powered equipment. Key heat-generating components include, but are not limited to, power conversion modules, microprocessor units, inductor components, and high-power-density integrated circuits. High-precision platinum resistance temperature sensors or thermistors are selected, with a measurement accuracy better than ±0.1 degrees Celsius, and a sampling frequency set to 10 to 100 times per second to ensure the capture of transient thermal fluctuation characteristics. The temperature sensors convert analog electrical signals into digital signals through a data acquisition unit.

[0011] In practice, acquiring temperature rise data is not simply a matter of reading the temperature. Temperature rise data is obtained by calculating the difference between the real-time temperature signal and a preset environmental reference temperature, provided by an environmental monitoring point placed inside the salt spray test chamber away from the heat-generating area of ​​the powered equipment. For temperature rise data extraction, the system smooths the original temperature curve and uses a moving average filtering algorithm to remove random fluctuations caused by sensor thermal noise.

[0012] Determining the heating power involves real-time monitoring of electrical parameters. Specifically, the heating power is obtained by real-time monitoring of the input current and input voltage of the powered equipment and calculating their product to obtain the instantaneous total input power. Subsequently, the effective power output by the powered equipment under the current operating state and the known losses, including switching losses and line resistance losses, which were measured in advance through no-load experiments, are subtracted from the total input power. This calculation process is executed in real time within the logic controller.

[0013] Furthermore, acquiring internal heat generation data also involves in-depth monitoring of the cooling system's status. The cooling system includes a cooling fan controlled by a pulse width modulation signal and internal circulating liquid cooling piping. The heat generation data needs to be corrected for the cooling system's heat dissipation contribution. The correction process involves measuring the temperature difference between the cooling medium at the inlet and outlet, and combining this with the cooling medium's flow rate, specific heat capacity, and flow velocity to calculate the amount of heat removed by the cooling system. The temperature rise data is obtained by smoothing the temperature change curves at each measurement point, extracting not only the steady-state value after reaching equilibrium, but also the transient rate of change at the moment of power switching.

[0014] The acquisition of internal heat generation data also includes filtering sensor signals under electromagnetic interference conditions. Since the powered equipment generates high-frequency switching noise during power switching, the system employs an adaptive filter that dynamically adjusts the filtering parameters based on the power electronic switching frequency to remove common-mode interference. Simultaneously, the calculation of heat generation power also considers the impact of ambient temperature on component resistance; by introducing the temperature coefficient formula for metal materials, temperature correction is applied to the real-time calculated ohmic losses.

[0015] Step 2: Based on the internal heat generation data, determine the surface temperature difference distribution of the powered equipment at various operating power levels. The temperature difference distribution reflects the surface temperature gradient differences caused by internal heat generation. For example... Figure 2 and Figure 3 As shown, the process includes establishing a high-precision three-dimensional heat conduction model of the powered device. This model not only includes the external geometric contours of the device but also meticulously describes the internal circuit board layout, the spatial location of heat-generating components, and the physical properties of the heat-conducting interfaces. During the model parameter setting phase, the system assigns corresponding physical properties to different components, including the material's thermal conductivity, specific heat capacity, and mass density.

[0016] The temperature difference distribution was calculated using the finite element method. Specifically, the overall three-dimensional structure of the power supply equipment was divided into tiny computational units consisting of millions of nodes. Within each computational unit, based on the internal heat generation data obtained in step 1, the heat generation power was applied as an internal heat source boundary condition to the corresponding heating element location. Simultaneously, the model incorporated the convective and radiative heat transfer coefficients between the power supply equipment's casing and the external salt spray environment.

[0017] In step 2, the determination of the surface temperature difference distribution also considers the heat sink effect of the mounting bracket of the electrical equipment. The heat sink effect refers to the fact that the mounting bracket, as a large thermal mass, conducts heat through mechanical contact points, resulting in a significant difference in temperature distribution between the bottom of the electrical equipment and the top or other suspended parts. When constructing the model, the contact surface between the mounting bracket and the electrical equipment is set as a contact thermal resistance boundary condition, that is, the heat flux density and contact thermal resistance of this interface are clearly defined. The temperature difference distribution is ultimately presented as the difference between the temperature of each sampling point on the surface of the electrical equipment casing and the salt spray environment temperature, forming a temperature gradient field as the spatial coordinates change.

[0018] To improve the accuracy of the simulation, infrared thermal imaging data was used for auxiliary verification in determining the surface temperature difference distribution. During the pre-experiment phase before formal testing, infrared thermal imagers were used to acquire full-field surface temperature distribution maps of the powered equipment under no-load, half-load, and full-load conditions. By comparing the geometric similarity and numerical deviation between the infrared thermal images and the temperature cloud maps generated by finite element analysis, the local heat transfer coefficients in the model were adjusted in reverse to ensure that the simulated surface temperature difference distribution remained highly consistent with the physical reality. The analysis of the temperature difference distribution covered both the transient startup phase and the steady-state operation phase. In the transient phase, the focus was on analyzing the temperature rise slope as heat was conducted from the inside to the outside during the instant of equipment startup, while in the steady-state phase, the distribution of the fixed thermal gradient formed under long-term power operation was analyzed.

[0019] Step 3: Based on the surface temperature difference distribution under various operating power levels, determine the condensation rate distribution of the corrosive medium on the surface of the powered equipment through numerical simulation, and identify critical corrosion areas based on the condensation rate distribution. These critical corrosion areas correspond to the preferential condensation sites of salt spray caused by temperature differences. Figure 4 As shown, the numerical simulation involves complex computational fluid dynamics simulations to analyze the trajectory, collision mechanism, and phase transition process of salt spray particles on the surface of electrical equipment. The numerical simulation is based on a two-phase flow model of gas and liquid phases.

[0020] The gas-phase model describes the mixed flow of air, water vapor, and sodium chloride aerosol within the salt spray test chamber, using the Navier-Stokes equations. The liquid-phase model tracks the discrete trajectory of each salt spray particle using the Lagrangian method, considering the effects of gravity, buoyancy, aerodynamic drag, and thermophoretic forces on particle motion. The core basis for determining the condensation rate distribution is that when the local temperature on the surface of the electrical equipment is lower than the dew point temperature under the current ambient humidity, water vapor and salt spray particles in the gas phase will condense on that surface.

[0021] Specifically, the numerical simulation calculates the mass of liquid film condensed per unit time for each surface computational cell. The calculation of the condensation rate distribution also incorporates a surface wettability parameter, which assesses the droplet formation and spreading characteristics on the surface through the contact angle. The critical corrosion region identification process involves sorting the condensation rates of all surface computational cells according to their numerical values ​​and defining the regions with condensation rates in a preset high-order range as critical corrosion regions. These regions are typically located in heat sink gaps, structural corners, connector interfaces, and areas with the greatest temperature difference due to internal heating. The preset high-order range comprises the top 10% of regions in terms of condensation rate values.

[0022] The dynamics of salt spray particles in the numerical simulation include Brownian motion and turbulent diffusion effects. For tiny salt spray particles with a diameter less than 1 micrometer, Brownian motion leads to random deposition behavior within the boundary layer, and the system obtains a stable condensation rate through statistical averaging. For larger particles with a diameter greater than 5 micrometers, the focus is on the capture efficiency on the windward side due to inertial impaction. Furthermore, when identifying critical corrosion areas, the protective performance of the coating on the surface of the electrical equipment must be considered. If an anti-corrosion coating exists, the coating's thermal resistance and thickness parameters are incorporated into the numerical simulation. The simulation determines the likelihood of microcracks forming in the coating under thermal stress, and, combined with the condensation rate distribution, assesses the risk of corrosive media penetrating the coating and entering the substrate.

[0023] Step 4: Based on the identified critical corrosion regions and their corresponding condensation rate distributions under multiple different operating power conditions, extract dynamic change characteristics that characterize the relationship between power changes and condensation rate changes. For example... Figure 5 As shown, this step uses data mining techniques to establish a mathematical correlation between input power and output corrosion risk. Specifically, the extraction process includes establishing a response function with power parameters as the independent variable and condensation rate as the dependent variable.

[0024] Dynamic characteristics include the derivative relationship between the condensation rate and power, i.e., the slope characteristic, which characterizes the sensitivity of the corrosion medium deposition rate to changes when the operating power fluctuates by a unit increment. In addition, dynamic characteristics also include the hysteresis effect during power switching. Due to the thermal inertia of the powered equipment, temperature rise typically lags behind power changes, resulting in an asymmetry in the condensation process during the power rise and fall phases.

[0025] The extraction process employed multivariate regression analysis to eliminate random numerical fluctuations and noise generated during the simulation, obtaining characteristic parameters reflecting underlying physical laws. Time series analysis was also used to extract dynamic characteristics. For each sampling moment, the cross-correlation coefficient between the power change rate and the condensation rate change rate was calculated. These dynamic characteristics reflect the instantaneous peak value of corrosion risk during the system's transition from one thermal equilibrium state to another.

[0026] Feature extraction also includes identifying resonance points, i.e., at specific power fluctuation frequencies, where the oscillation period of surface temperature synchronizes with the phase transition period of salt spray deposition, leading to a frequency characteristic that accelerates the deposition of corrosive media. Furthermore, dynamic characteristics include the thermal response time constant, which is the time required for the powered equipment to reach a new equilibrium in surface temperature distribution after a power change. This time constant directly determines the dwell time in each stage of the subsequent testing scheme.

[0027] Step 5: Compare the dynamic change characteristics with a preset threshold. When the dynamic change characteristics exceed the preset threshold, adjust the operating power parameters of the powered equipment, and re-acquire the internal heat generation data and corresponding temperature difference distribution based on the adjusted operating power to iteratively optimize the temperature difference change model characterizing the dynamic relationship between internal heat generation and the condensation rate of the corrosive medium. Specifically, based on the dynamic change characteristics extracted in Step 4, the internal heat generation data obtained in Step 1, and the surface temperature difference distribution determined in Step 2, a temperature difference change model characterizing the dynamic relationship between internal heat generation and the condensation rate of the corrosive medium is constructed. Specifically, multiple regression analysis or machine learning algorithms can be used, with operating power, ambient humidity, and salt spray concentration as inputs, and the predicted condensation rate of each key corrosion area as output, to establish an initial temperature difference change model. Compare the dynamic change characteristics extracted in Step 4 with the preset threshold. When the dynamic change characteristics exceed the preset threshold, the operating power parameters of the powered equipment are adjusted, and steps 1 to 2 are re-executed based on the adjusted operating power to obtain new internal heat generation data and the corresponding temperature difference change distribution. The new data is fed back into the temperature difference change model constructed in step 5 to iteratively optimize the temperature difference change model until the dynamic change characteristics meet the preset threshold requirements or reach the preset number of iterations.

[0028] The preset threshold is a slope safety threshold for the condensation rate as a function of power. The slope safety threshold is determined based on the design life of the powered equipment and the maximum allowable corrosion rate.

[0029] When the dynamic characteristics show that the condensation rate increases non-linearly with increasing power, i.e., the slope exceeds the preset safety threshold, the system will decrease or increase the operating power by a preset step size to observe the feedback response of the condensation rate. The preset step size is 5%-10% of the rated operating power.

[0030] During the iterative optimization process, the temperature difference change model can be deeply trained using machine learning algorithms on historical experimental data and the simulation data obtained in steps 1 to 4. Specifically, a deep neural network containing multiple hidden layers is established.

[0031] This neural network uses internal heat generation, ambient humidity, salt spray concentration, and initial ambient temperature as input vectors, and the predicted condensation rate at each key location as the output vector. Through continuous updates of neuron weights, the model can achieve real-time prediction of corrosion risk under dynamic operating conditions. The optimized temperature difference change model is validated through residual analysis, calculating the root mean square error (RMS) between the model's predicted values ​​and the actual observed temperature or condensation rate. If the RMS error exceeds a preset allowable range, the model is reconstructed by increasing the sampling point density and adjusting the number of network layers. The preset allowable range is the permissible prediction error range for the temperature difference change model, set based on the allowable accuracy deviation requirements of engineering testing, and used to determine whether the model meets the prediction accuracy requirements.

[0032] The model also incorporates a material aging factor, considering the changes in surface emissivity and condensation characteristics caused by the formation of surface oxide layers and salt accumulation over time, thus achieving accurate simulation of the entire test lifecycle. Furthermore, the optimized temperature difference model can calculate the relative humidity distribution within the internal cavity of the electrical equipment. Internal heating causes an increase in the air temperature inside the cavity, increasing its saturated moisture content and altering the condensation conditions on the cavity walls. The model uses a fluid-thermal coupling algorithm to simultaneously solve for the convection flow and heat exchange of the gas within the cavity, predicting the condensation risk on the surfaces of sensitive internal components.

[0033] Step 6: Using a temperature difference change model, quantitatively assess the impact of corrosive atmospheres on the withstand capability of powered equipment, and generate withstand capability assessment reports for different operating power levels. For example... Figure 7 As shown, a cumulative corrosion damage parameter was introduced during the quantitative evaluation process. The cumulative corrosion damage parameter is defined as the integral sum of the product of the condensation rate and the concentration of the corrosive medium per unit time over the entire test period.

[0034] The endurance assessment report details the estimated time required for the powered equipment to experience functional failure or for the insulation resistance to drop to a preset critical value under different workload percentages. The report employs a multi-dimensional health index evaluation system. This system integrates indicators such as thermal resistance change rate, leakage current growth rate, and mechanical structure integrity. A weighted sum of these indicators yields the overall endurance score of the powered equipment.

[0035] The tolerance assessment report also includes a failure mode and effect analysis (FMEA). For each identified critical corrosion area, the system analyzes specific failure modes that corrosion products may cause, such as short circuits, open circuits, or increased contact resistance. The quantitative assessment also includes monitoring the signal integrity of the powered equipment, mapping the signal transmission error rate to cumulative corrosion damage parameters, and determining the impact curve of corrosion on the electrical performance degradation of the powered equipment.

[0036] The endurance assessment report is highly traceable. Each conclusion in the report corresponds to specific simulation data and original sampling records. By clicking on virtual links in the report, technicians can access the corresponding temperature difference evolution curves and dynamic videos simulating condensation distribution. Furthermore, the assessment report provides targeted improvement suggestions, such as adding insulation design or strengthening local sealing levels in areas with significant temperature differences, providing a basis for product design iterations.

[0037] Step 7: Based on the tolerance assessment report, simulate the temperature difference evolution and condensation rate changes under various workload scenarios to determine the control requirements for the internal heating state of the powered equipment, and generate a salt spray performance test plan that includes the control requirements. Figure 8 As shown, the process of generating the scheme includes accurately determining the temperature cycling curve, humidity control program, and physical intensity of salt spray in the test chamber.

[0038] The control requirements involve programmed control commands for the operating power of the powered equipment. The salt spray performance test scheme divides the entire test process into several test phases with specific objectives. In the first phase, the powered equipment is in a low-power standby state to simulate the severe condensation effect caused by the lack of internal heat generation in a high-humidity environment. In the second phase, the powered equipment switches to a full-load operation state, and the thermal evaporation process of the condensate and the residual salt distribution after evaporation are observed through intense internal heating. In the third phase, complex cyclic load switching is performed to simulate the coupling effect of thermal stress and chemical corrosion in actual service conditions.

[0039] The salt spray performance testing scheme also includes dynamic adjustment logic for the salt spray concentration. When the powered equipment is in a high-heat-generating state, the scheme automatically increases the salt spray density to offset the loss of corrosive media caused by surface heat evaporation, ensuring the continuity of corrosion stress. The control requirements also specify the calibration cycle of each sensor node during the test, as well as the recovery procedure after abnormal test interruption.

[0040] The final solution is output as a control algorithm script, which can be directly read and executed by the central controller of the salt spray testing system. The automated process for generating the solution includes protocol matching with the hardware interface of the testing equipment, and the solution generator automatically converting the control command format according to the model of the target test chamber. The solution also specifies regular maintenance and cleaning cycles, determining the timing for replacing the brine solution and cleaning the nozzles based on the simulated salt accumulation rate. Furthermore, the solution includes automatic backup of test data and remote monitoring configuration, enabling unattended operation of the testing process.

[0041] Example 2

[0042] In another specific embodiment, for powered devices with high-frequency power switching characteristics, a more refined time analysis is employed in the process of acquiring internal heat generation data. Specifically, the calculation of heat generation power not only considers the fundamental power but also extracts the higher harmonic components of current and voltage through fast Fourier transform, calculating the additional eddy current losses caused by harmonics. Temperature rise data is acquired through miniature thin-film thermocouples embedded in the middle layer of multilayer circuit boards to capture the deep thermal response inside the package.

[0043] In step 2, when determining the surface temperature difference distribution, the three-dimensional heat conduction model further refined the nonlinear characteristics of the contact thermal resistance. The contact thermal resistance decreases with increasing surface pressure, and the model adjusts the heat conduction efficiency of each contact surface based on bolt preload data. Simultaneously, the influence of ambient wind speed on forced convection heat transfer on the surface of the electrical equipment was considered. By setting fluid velocity field boundary conditions in the finite element analysis, the weakening or enhancing effect of wind speed on the surface temperature gradient was simulated.

[0044] In step 3, the numerical simulation process incorporates chemical equilibrium calculations of the electrolyte solution particles. The corrosive medium is considered as a dynamic equilibrium system containing sodium ions, chloride ions, and water molecules, taking into account the inducing effect of surface electrochemical potential on condensate film formation. The calculation of the condensation rate distribution also involves the correction of saturated vapor pressure by surfactants, considering the inhibitory effect of increased solution concentration due to water evaporation on the condensation rate. The identification of critical corrosion regions relies not only on the absolute value of the condensation rate but also on the characteristics of the surface geometry, using curvature analysis to identify pits and narrow gap structures that are prone to liquid film accumulation.

[0045] In step 4, the extraction of dynamic change features incorporates multi-scale analysis based on wavelet transform. By decomposing the time-series signals of power and condensation rate into different frequency bands, the different contributions of high-frequency operational fluctuations and long-period thermal trends to corrosion risk are identified. The dynamic change features also include thermal strain rate features, i.e., the rate of thermal expansion and contraction of the structure caused by temperature differences, used to assess the contribution of physical stress to the peeling of the protective coating.

[0046] In step 5, the optimized temperature difference change model is trained using an adaptive weight allocation mechanism. When the system detects a drastic change in ambient humidity, the model automatically increases the weight of the humidity feature term to improve prediction sensitivity. The deep neural network employs a recurrent neural network architecture, which can use historical temperature rise data to predict future condensation trends, enhancing the model's ability to handle dynamic operating conditions.

[0047] In step 6, when quantifying the resilience assessment, the resilience assessment report incorporates a reliability analysis model based on the Weibull distribution. Through statistical analysis of a large number of simulated samples, the mean time to failure and reliability function of the powered equipment under specific temperature stress are derived. The assessment report also includes stress derating recommendations for critical components, guiding hardware engineers to reduce the risk of localized corrosion by adjusting the layout.

[0048] In step 7, the generated salt spray performance test scheme includes a "stress-accelerated" test mode for extreme operating conditions. Without altering the chemical corrosion mechanism, the thermal equilibrium establishment time is artificially shortened by adjusting the power switching frequency, thereby accelerating the test progress. The scheme also integrates emergency shutdown logic. When the temperature of a critical node in the powered equipment exceeds a preset safety threshold or electrical short-circuit symptoms occur, the test scheme will automatically switch to safety protection mode, cutting off operating power and activating high-pressure ventilation cooling.

[0049] Example 3

[0050] In another specific embodiment, the method of the present invention is integrated into a remote cloud monitoring and testing system. The data acquired in step 1 is uploaded to a cloud server in real time via an industrial IoT gateway. Sensor signals are stored in binary compressed format after adaptive filtering. The calculation of heat generation power is performed on a cloud computing cluster, utilizing a distributed computing architecture to process large-scale real-time electrical parameter streams.

[0051] In step 2, the calculation of the three-dimensional heat conduction model utilizes a GPU acceleration unit in the cloud, enabling the calculation of millions of degrees of freedom in the thermal field within minutes. During the determination of the temperature difference distribution, the system automatically retrieves a database of the thermal characteristics of similar historical products and rapidly converges the model parameters using transfer learning methods.

[0052] In step 3, multiphysics coupling simulation was introduced into the numerical simulation process. In addition to the gas-liquid two-phase flow, the charge transfer field during the corrosion process was also solved simultaneously. The identification of critical corrosion regions combined with artificial intelligence image recognition technology, which automatically marked potential corrosion initiation points by extracting features from the simulated cloud map.

[0053] In step 4, a causal inference algorithm was used to extract the dynamic change features. The system automatically determines whether the power change is a direct cause of the change in condensation rate or an effect produced through intermediate variables, thereby more accurately identifying key control variables. The dynamic change features also include an environmental adaptability envelope, which defines the temperature difference safety boundary for the equipment to operate stably for a long time under different loads.

[0054] In step 5, the optimized temperature difference change model exhibits self-evolution capabilities. As test data accumulates, the model automatically adjusts the hyperparameters of the deep neural network to adapt to power-on equipment with different materials and structures. The model also integrates corrosion kinetic equations, using condensation rate, chloride ion concentration, and surface electrochemical potential as inputs to accumulate the corrosion depth at each point in three-dimensional space in real time.

[0055] In step 6, during the generation of the tolerance assessment report, the system automatically compares industry standards with enterprise standards to determine the compliance status of the powered equipment. The report also includes a virtual reality demonstration system, allowing users to view animations of humidity evolution inside the powered equipment's cavities at different power levels through an interactive interface.

[0056] In step 7, the generated salt spray performance test plan is highly programmable. Users can customize temperature cycles and salt spray patterns according to specific service environment conditions (such as tropical coastal areas or temperate industrial areas). The plan generator can automatically calculate the optimal test sequence to obtain the most comprehensive tolerance data with minimal testing time cost. In addition, the plan includes an interface with the laboratory management system to automatically schedule test tasks and allocate resources.

[0057] In any of the above embodiments, the acquisition of internal heat generation data involves real-time correction of ambient humidity. Since changes in humidity affect the heat capacity and thermal conductivity of air, the system dynamically corrects the temperature rise data processing algorithm based on humidity sensor readings. In determining the surface temperature difference distribution, the finite element analysis employs nonlinear material parameters, considering the nonlinear change in the thermal conductivity of metallic materials with increasing temperature, ensuring simulation accuracy over a wide temperature range.

[0058] In the numerical simulation of condensation rate distribution, a capillary condensation effect model was introduced for capillary structures on the surface of electrical equipment (such as the tiny gaps in heat sinks). In these areas, condensation will occur even if the ambient humidity is not 100%, due to the decrease in saturated vapor pressure caused by the concave liquid surface. The identification process for critical corrosion areas fully considers the droplet drainage path driven by gravity, identifying structural parts that are prone to forming dead zones and stagnant zones. These parts, because moisture does not evaporate easily, often become the areas with the most severe corrosion.

[0059] Extracting dynamic change features also includes identifying the temperature fluctuation amplitude after the system reaches steady state. This amplitude reflects the coupled oscillation relationship between the internal control logic of the powered equipment (such as the fan speed regulation algorithm) and the heat generation power. The optimized temperature difference change model also integrates a thermal infrared emissivity correction module, considering that the thermal radiation performance of the object changes with surface salt deposition, thus affecting the subsequent temperature field simulation results.

[0060] The tolerance assessment report utilized a knowledge graph-based failure analysis expert system. When a specific corrosion damage mode was identified, the system automatically referenced a historical failure case library, providing possible root cause analyses. The salt spray performance test plan also included compensation suggestions for the uniformity of the test environment. Based on the placement of the powered equipment within the test chamber, the angle and pressure of the spray nozzles were adjusted to ensure that the surface condensation environment met the simulated boundary conditions.

[0061] During the test execution, the system monitors the electrical performance indicators of the powered equipment in real time, such as the leakage current and insulation resistance. When the rate of change of these indicators exceeds the preset dynamic alarm threshold, the system automatically triggers an enhanced sampling mode to increase data recording density and provide detailed data support for capturing transient failures. The dynamic alarm threshold is defined as a leakage current hourly change rate exceeding 5%. After the test, the system automatically generates a complete closed-loop evaluation logic, feeding the measured data back into the optimized temperature difference change model to achieve continuous model learning and accuracy improvement.

[0062] In a specific application scenario, this method was used for salt spray performance testing of a certain type of powered-on communication base station controller. First, internal heat generation data of the controller was acquired at three typical operating power levels: 10 watts, 50 watts, and 100 watts. Internal sensors monitored a temperature rise of 45 degrees Celsius in the processor under a full load of 100 watts. Next, a finite element thermal model of the controller was established, simulating a maximum temperature gradient of 5 degrees Celsius per centimeter on the controller's outer casing surface at an ambient temperature of 25 degrees Celsius.

[0063] CFD numerical simulations revealed that at the air inlet grid at the bottom of the controller, due to alternating heating and cooling and stagnant airflow, the salt spray condensation rate reached 0.5 milligrams per square centimeter per hour. This area was identified as a critical area for primary corrosion. Subsequently, dynamic change characteristics were extracted, revealing a brief peak in the condensation rate when the power rapidly switched from 50 watts to 100 watts, with a thermal response hysteresis of approximately 10 minutes.

[0064] Based on this, an initial deep neural network model was constructed, and the extracted dynamic change features were used to iteratively optimize the model, reducing its root mean square error in dynamic load prediction to within 5%. Evaluation using this model showed that in a coastal high-salt-spray environment, if the controller is subjected to fluctuating loads of 50% to 80% for an extended period, the cumulative corrosion damage at the edge of its sealing ring will reach a preset critical value after 500 hours, potentially leading to damage to the internal circuitry. Finally, based on this evaluation result, a 10-day cyclic salt spray test plan was generated. This plan specifically included a power switching requirement every 4 hours to accurately simulate the accelerating effect of temperature stress on the corrosion process during actual operation. After execution, the plan successfully reproduced the localized corrosion failure mode observed in field service in the laboratory, verifying the scientific validity and engineering practical value of the test method.

[0065] The above description is only a preferred embodiment of the present invention. Therefore, all equivalent changes or modifications made to the structure, features and principles described in the claims of this patent application are included within the protection scope of this patent application.

Claims

1. A method for testing the salt spray performance of electrical equipment, characterized in that, Includes the following steps: Acquire internal heat generation data of powered equipment under multiple different operating power levels; Based on internal heat generation data, the distribution of surface temperature difference changes of the powered equipment under various operating power levels was determined. Based on the surface temperature difference distribution under various operating power levels, the condensation rate distribution of the corrosive medium on the surface of the power supply equipment is determined by numerical simulation, and the key corrosion areas are identified based on the condensation rate distribution. Based on the critical corrosion regions identified under multiple different operating power levels and their corresponding condensation rate distributions, dynamic change characteristics characterizing the relationship between power variation and condensation rate variation were extracted. The dynamic change characteristics are compared with the preset threshold. If the dynamic change characteristics exceed the preset threshold, the operating power parameters of the powered equipment are adjusted. Based on the adjusted operating power, the internal heat generation data and the corresponding temperature difference change distribution are re-acquired to iteratively optimize the temperature difference change model that characterizes the dynamic relationship between the internal heat generation and the condensation rate of the corrosive medium. Using the optimized temperature difference change model, the impact of corrosive atmosphere on the tolerance of electrical equipment is quantitatively evaluated, and tolerance assessment reports for different operating power are generated. Based on the tolerance assessment report, the temperature difference evolution and condensation rate changes under various workload scenarios are simulated to determine the regulation requirements for the internal heating state of the powered equipment, and a salt spray performance test plan including the regulation requirements is generated.

2. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for obtaining internal heat generation data is as follows: Real-time temperature signals are obtained by multiple temperature sensors deployed on the surface of key heating elements inside the power supply equipment. The key heating elements include a power conversion module, a processor unit, and an inductor assembly. The real-time temperature signal is acquired by the data acquisition unit at a preset sampling frequency, and the difference between the real-time temperature signal and the preset ambient reference temperature is calculated to obtain temperature rise data. The method for determining the heating power is as follows: By real-time monitoring of the input current and input voltage of the powered equipment, the product of the input current and the input voltage is calculated to obtain the total input power. The output effective power and the known loss power are then subtracted from the total input power to obtain the preliminary heating power. Monitor the status of the cooling system of the powered equipment. The cooling system includes a cooling fan and a cooling circulation pipe. Measure the temperature difference between the inlet and outlet of the cooling medium, and calculate the amount of heat carried away by the cooling system based on the flow rate, flow volume, and specific heat capacity of the cooling medium. Use the heat value to correct the initial heating power to obtain the internal heat generation data. An adaptive filter is used to remove common-mode interference from the real-time temperature signal, and the heating power is corrected in real time based on the temperature coefficient of resistance of the key heating element.

3. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for determining the surface temperature difference distribution is as follows: A three-dimensional heat conduction model of the power supply equipment is established, which describes the material thermal conductivity, specific heat capacity and density of the power supply equipment. The power supply equipment is divided into multiple computing units. Within each computing unit, internal heat source boundary conditions are set based on the internal heat generation data. At the same time, the convective heat transfer coefficient and radiative heat transfer coefficient between the power supply equipment shell and the external environment are combined, and the surface temperature difference distribution is calculated by finite element analysis. The three-dimensional heat conduction model includes a description of the heat sink effect of the mounting bracket of the power supply equipment, sets the contact surface between the mounting bracket and the power supply equipment as a contact thermal resistance boundary condition, and specifies the heat flux density and contact thermal resistance of the contact surface. The surface temperature difference distribution is verified using infrared thermal imaging data. By comparing the geometric similarity between the infrared thermal image and the simulated cloud image, the local heat transfer coefficient in the three-dimensional heat conduction model is adjusted. The determination of the surface temperature difference distribution covers the transient start-up phase and the steady-state operation phase of the powered equipment. The temperature rise slope of heat conduction is extracted during the transient start-up phase, and the fixed thermal gradient distribution is extracted during the steady-state operation phase.

4. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for numerical simulation and identification of critical corrosion regions is as follows: Computational fluid dynamics was used to simulate and analyze the motion trajectory and phase transition process of salt spray particles on the surface of the electrical equipment. A two-phase flow model is established, consisting of a gas phase and a liquid phase. The gas phase model describes the mixed flow of air and water vapor, while the liquid phase model tracks the discrete motion of the salt spray particles using the Lagrange method, taking into account the effects of gravity, buoyancy, aerodynamic drag, and thermophoretic force on the motion of the salt spray particles. The basis for determining the condensation rate distribution is that the local temperature of the surface computing unit is lower than the dew point temperature under the corresponding humidity, and the mass of liquid film condensed by the surface computing unit per unit time is calculated. Furthermore, a surface wettability parameter is introduced to describe the effect of the contact angle on droplet formation, and the condensation rates of each surface calculation unit are sorted, with the region where the condensation rate is in the preset high range defined as the critical corrosion region. When determining the critical corrosion area, the thermal resistance and coating thickness parameters of the coating on the surface of the electrical equipment are introduced to simulate the possibility of microcracks in the coating under temperature stress and assess the risk of the corrosive medium penetrating the coating and entering the substrate.

5. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for extracting dynamic change features is as follows: Establish a response function with power parameter as independent variable and condensation rate as dependent variable; The dynamic change characteristics include the derivative relationship between the condensation rate and the power parameter, which is used to characterize the sensitivity of power fluctuations to the deposition rate of the corrosive medium; the dynamic change characteristics also include the hysteresis effect characteristics during different power switching processes, which are used to reflect the asymmetry of the condensation process caused by temperature rise lag. Multiple regression analysis was used to obtain characteristic parameters reflecting physical laws, and time series analysis was used to calculate the cross-correlation coefficient between the power change rate and the condensation rate change rate. Identify the resonance point where the power fluctuation frequency of the powered device is synchronized with the surface temperature oscillation period, and the thermal response time constant required for the powered device to reach a new equilibrium in surface temperature distribution from a power change. The thermal response time constant is used to determine the duration of each stage in the salt spray performance test scheme.

6. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for iteratively optimizing the temperature difference change model is as follows: An iterative optimization algorithm is used to adjust the operating power parameter by a preset step size when the condensation rate shown by the dynamic change characteristics exhibits a non-linear and rapid increase trend with the increase of the power parameter. A deep neural network with multiple hidden layers was established by training historical experimental data using machine learning algorithms. The deep neural network takes internal heat generation, ambient humidity and salt spray concentration as input vectors and the predicted condensation rate at each key location as output vector. The optimized temperature difference change model was verified by residual analysis. The root mean square error between the model prediction and the actual observation point was calculated. When the root mean square error exceeded the preset range, the model was reconstructed by increasing the sampling point density and adjusting the neuron weight parameters. A material aging factor is introduced into the optimized temperature difference change model. The material aging factor reflects the change in thermal emissivity and condensation characteristics caused by the formation of the surface oxide layer as the test time increases. The gas convection and heat exchange in the internal cavity of the power supply equipment are solved simultaneously by using a fluid-thermal coupling algorithm to predict the condensation risk on the surface of sensitive components inside the power supply equipment.

7. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for quantitative assessment and generating tolerance assessment reports is as follows: A cumulative corrosion damage parameter is introduced, which is equal to the time integral of the product of the condensation rate and the concentration of the corrosive medium per unit time over the entire test cycle. The tolerance assessment report records the estimated time required for the powered equipment to experience functional failure or for the insulation resistance to drop to a preset critical value under different workload percentages. A multi-dimensional health index evaluation system is established, which integrates thermal resistance change rate, leakage current growth rate and mechanical structure integrity index. The comprehensive endurance score of the power supply equipment is obtained by weighted summation of the above indicators. The failure modes of each of the aforementioned critical corrosion regions were analyzed, including short circuits, open circuits, and increased contact resistance caused by corrosion products. Monitor the signal integrity of the powered equipment, correlate and map the signal transmission error rate with the cumulative corrosion damage parameter, and determine the electrical performance degradation curve; The tolerance assessment report includes a temperature difference evolution curve and a condensation distribution simulation video, and provides improvement suggestions for thermal insulation design or enhanced sealing for areas where the surface temperature difference distribution is significant.

8. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, The method for generating a salt spray performance test scheme is as follows: Determine the temperature cycling curve, humidity control program, and salt spray intensity in the test chamber; The control requirements involve programmed control commands for the operating power of the powered equipment; The testing process is divided into several testing phases: In the first phase, the powered equipment is controlled to be in a low-power standby state to simulate the condensation effect under high humidity; in the second phase, the powered equipment is controlled to switch to full-load operation, and the evaporation of condensate and salt residue are observed through internal heating; in the third phase, cyclic load switching is performed to simulate the coupling effect of thermal stress and chemical corrosion in actual service conditions. The salt spray intensity and spray density are dynamically adjusted based on the amount of corrosive medium lost from the surface of the electrical equipment due to thermal evaporation. The salt spray performance test scheme is output in the form of a control algorithm script, and the hardware interface protocol is automatically matched according to the model of the target test chamber.

9. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, In determining the condensation rate distribution, the chemical equilibrium calculation of the electrolyte solution particles in the corrosive medium is introduced, and the inducing effect of surface electrochemical potential on the formation of condensate film is considered. The inhibitory effect of the solution concentration of the condensate film on the condensation rate as the water evaporates is considered, and the calculation is performed in conjunction with the correction factor of the surfactant on the saturated vapor pressure. For the tiny gaps in the heat sink on the surface of the above-mentioned electrical equipment, a capillary condensation effect model is introduced to calculate the amount of condensation caused by the decrease in saturated vapor pressure due to the concave liquid surface. By combining the analysis of the droplet discharge path under gravity, the structural parts on the surface of the power supply equipment that are prone to forming dead zones and stagnation zones are identified.

10. The method for testing the salt spray performance of electrical equipment according to claim 1, characterized in that, During the execution of the salt spray performance test plan, the leakage current and insulation resistance values ​​of the powered equipment are monitored in real time. When the rate of change of the leakage current exceeds the preset dynamic alarm threshold, the enhanced sampling mode is automatically triggered to increase the data recording density. The salt spray performance test scheme also includes emergency shutdown logic. When the temperature of the critical node of the powered equipment exceeds the preset safety threshold or an electrical short circuit occurs, the operating power is automatically cut off and the ventilation and cooling mode is activated. Based on the simulated salt accumulation rate, the time cycle for replacing the brine solution and cleaning the nozzle in the salt spray performance test scheme is determined, and the test data is automatically backed up.

Images