A method and system for adaptive environmental control of intelligent electrical cabinets

By using an intelligent electrical cabinet adaptive environmental control method, the changes in dew point temperature difference are identified by utilizing environmental state parameters and thermal and humidity characteristics. The risk of condensation is dynamically assessed and the optimal control mode is selected, which solves the problems of control lag and energy waste in the existing technology and achieves safe and reliable operation of the equipment and energy-saving effect.

CN123114166APending Publication Date: 2026-11-17扬州长泉电器设备有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610949859.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-29
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing intelligent electrical cabinets rely on fixed thresholds and simple logic judgments for adaptive environmental control, which causes control actions to lag behind condensation formation and fail to accurately detect potential cold spot temperatures, resulting in energy waste or condensation risks.

Method used

By acquiring environmental status parameters and load current inside and outside the electrical cabinet, calculating dew point temperature and temperature difference, identifying thermal and humidity characteristic parameters, generating dew point temperature difference change curves, and selecting combined control modes of heaters, fans, and air conditioners based on Bayesian formulas and physical information neural networks, dynamic condensation risk assessment and proactive intervention are achieved.

Benefits of technology

It enables early identification and proactive protection against condensation risks, avoids energy waste, and ensures safe and reliable equipment operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123114166A_ABST
    Figure CN123114166A_ABST
Patent Text Reader

Abstract

This invention provides an intelligent electrical cabinet adaptive environmental control method and system, relating to the field of electrical cabinet adaptive control technology. The method includes: acquiring environmental state parameters inside and outside the electrical cabinet and load current during the current sampling period to calculate heating power and dew point temperature, and determining the current dew point temperature difference; based on the environmental state parameters inside and outside the electrical cabinet, heating power, and the current dew point temperature difference, identifying the electrical cabinet's thermal and humidity characteristic parameters and power distribution coefficient, generating a dew point temperature difference change curve, and extracting the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero during future sampling periods; comparing the minimum dew point temperature difference with a preset safety threshold, determining the condensation risk level based on the comparison result and the condensation arrival time, and selecting a combined control mode for the heater, fan, and air conditioner inside the electrical cabinet based on the condensation risk level. This invention effectively prevents condensation from harming equipment safety and avoids unnecessary energy consumption due to excessive control.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of adaptive control of electrical cabinets, and in particular to an intelligent adaptive environmental control method and system for electrical cabinets. Background Technology

[0002] In power systems, electrical cabinets are core equipment for power distribution and control. The internal environment of the cabinet directly affects the operational reliability and service life of electrical components. When the temperature and humidity inside the cabinet reach the condensation critical point, water droplets easily condense on the metal cabinet walls and the surface of electrical insulation components, leading to a decrease in insulation performance, corrosion of metal components, and even surface discharge and short circuit accidents. Existing intelligent electrical cabinet adaptive environmental control methods usually calculate the dew point temperature based on the measurement values ​​of the air temperature and humidity sensors inside the cabinet, compare it with the cabinet wall temperature to obtain the current dew point temperature difference, and then compare it with a preset fixed threshold. When it is lower than the threshold, the heater or air conditioner and other actuators are controlled.

[0003] The method has the following problems in practical applications: First, the existing adaptive control method mainly relies on fixed thresholds or simple logical judgments to make passive responses, which often causes the control action to lag behind the actual formation process of condensation; Second, since condensation first occurs on the surface of the metal structural parts with the lowest temperature, the existing adaptive control method only uses sensors to measure the temperature and humidity of the air at a certain point in the cabinet to calculate the dew point. It cannot sense the actual temperature of all potential cold points. In engineering, the safety margin can only be artificially increased. If the margin is too large, it will lead to energy waste, and if the margin is insufficient, there will still be a risk of condensation. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent electrical cabinet adaptive environment control method and system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An adaptive environmental control method for intelligent electrical cabinets includes the following steps:

[0007] Obtain the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period to calculate the heating power and dew point temperature, and determine the current dew point temperature difference;

[0008] Based on the environmental condition parameters inside and outside the electrical cabinet, the heating power and the current dew point temperature difference, the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet are identified, the dew point temperature difference change curve is generated, and the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero under the future sampling period are extracted.

[0009] The minimum dew point temperature difference is compared with the preset safety threshold. The condensation risk level is determined based on the comparison result and the condensation arrival time. The combined control mode of the heater, fan and air conditioner in the electrical cabinet is selected according to the condensation risk level.

[0010] Furthermore, the environmental parameters inside and outside the electrical cabinet include the cabinet temperature, cabinet humidity, cabinet inner wall temperature, cabinet outer ambient temperature, and cabinet outer ambient humidity. The thermal and humidity characteristic parameters include thermal resistance, heat capacity, moisture time constant, and dehumidification efficiency.

[0011] Furthermore, the calculation of heating power and dew point temperature, and the determination of the current dew point temperature difference, includes the following steps:

[0012] During the current sampling period, the temperature and humidity at multiple heights inside the electrical cabinet, the temperature at multiple locations on the inner side of the cabinet wall, the ambient temperature outside the cabinet, the ambient humidity outside the cabinet, and the main circuit current are collected simultaneously, and the heating power is calculated based on the main circuit current.

[0013] The temperature and humidity at each height measuring point inside the cabinet are treated as independent operating conditions for that measuring point, and the dew point temperature of each measuring point is calculated separately.

[0014] The difference between each collected temperature inside the cabinet wall and each dew point temperature is calculated, and the minimum difference is taken as the current dew point temperature difference.

[0015] Furthermore, the process of identifying the thermal and humidity characteristics of the electrical cabinet and generating a dew point temperature difference change curve includes the following steps:

[0016] Multiple consecutive historical sampling periods are extracted from adjacent historical periods. The current dew point temperature difference of each historical sampling period is obtained. The temperature and humidity of the measuring point at a specific location on the inner side of the cabinet wall that generates the current dew point temperature difference, the temperature and humidity of the measuring point at a specific height inside the cabinet, the ambient temperature, the ambient humidity and the heating power outside the cabinet are obtained, and arranged according to the chronological relationship of the historical sampling periods to form a local state sequence.

[0017] According to the law of conservation of energy, a thermal balance relationship is established for the local area that generates the current dew point temperature difference. The thermal balance relationship includes the thermal resistance and heat capacity of the local area, as well as the power distribution coefficient that characterizes the total heat generation power distributed to the local area.

[0018] According to the law of conservation of mass, a moisture balance relationship is established for the local area, which includes the moisture time constant and dehumidification efficiency.

[0019] By taking the local state sequence as input and combining the thermal equilibrium relationship and the moisture equilibrium relationship, thermal and moisture characteristic parameters and power distribution coefficients are identified.

[0020] The temperature of a specific measuring point on the inner side of the cabinet wall, the temperature and humidity of a measuring point at a specific height inside the cabinet, the ambient temperature outside the cabinet, and the heating power are obtained at the current dew point temperature difference during the current sampling period. Combined with the heat balance relationship and the humidity balance relationship, the predicted values ​​of the cabinet wall temperature and the humidity inside the cabinet are calculated for each sampling period in the future.

[0021] Based on the predicted values ​​of cabinet wall temperature and cabinet humidity, the predicted values ​​of dew point temperature for each future sampling period are calculated, and the difference between the predicted values ​​of cabinet wall temperature and dew point temperature is calculated point by point to form a dew point temperature difference change curve.

[0022] Furthermore, the preset of the security threshold includes the following steps:

[0023] The system acquires environmental state parameters, heating power, and corresponding current dew point temperature difference for multiple historical sampling periods, and calibrates the safety status or condensation risk status of each historical sampling period.

[0024] The first probability density distribution is constructed by using kernel density estimation for the current dew point temperature difference of historical sampling periods that are in a safe state, and the second probability density distribution is constructed by using kernel density estimation for the current dew point temperature difference of historical sampling periods that are in a condensation risk state.

[0025] Based on the environmental state parameters and heat generation power during the current sampling period, determine the prior probability that the current sampling period belongs to a condensation risk state.

[0026] Based on the first probability density distribution, the second probability density distribution, and the prior probability, the posterior probability of the current sampling period actually belonging to the condensation risk state under the current dew point temperature difference is calculated using Bayes' theorem.

[0027] The additional estimated energy consumption of heaters, fans, and air conditioners caused by misjudging a safe state as a condensation risk level is taken as the false alarm cost, and the condensation loss caused by underestimating a condensation risk state as a safe state is taken as the underreport cost. Based on the false alarm cost, the underreport cost, and the posterior probability, the current dew point temperature difference judgment boundary that minimizes the expected total cost is determined, and this judgment boundary is taken as the safety threshold.

[0028] Furthermore, the preset security threshold also includes the following steps:

[0029] Based on the heat capacity, heating power of the electrical cabinet and the variation range of environmental state parameters in adjacent historical sampling periods, the maximum allowable rate of change of the current dew point temperature difference in adjacent historical sampling periods is determined.

[0030] If the rate of change of the current dew point temperature difference in a certain historical sampling period relative to the previous historical sampling period exceeds the maximum allowable rate of change, the current dew point temperature difference of that historical sampling period is removed.

[0031] Based on historical sampling data after removing outliers, the safety status or condensation risk status of each historical sampling period is identified.

[0032] Furthermore, the process of determining the safety status or condensation risk status of each historical sampling period includes the following steps:

[0033] Compare the temperature at a specific location on the inside of the cabinet wall that generates the current dew point temperature difference with the dew point temperature at a specific height inside the cabinet.

[0034] In response to the existence of multiple consecutive historical sampling periods, where the temperature at a specific location on the inner side of the cabinet wall is lower than the dew point temperature at a specific height inside the cabinet during each period, and the number of these multiple consecutive historical sampling periods reaches the preset number of condensation risk confirmation periods, these multiple consecutive historical sampling periods are defined as a condensation risk state, and the remaining historical sampling periods are defined as a safe state.

[0035] Furthermore, the step of selecting the combined control mode of the heater, fan, and air conditioner in the electrical cabinet based on the condensation risk level includes the following steps:

[0036] When the minimum dew point temperature difference is greater than the safety threshold, the condensation risk level is determined to be a safe level, and a combined control mode is selected to turn off the heater and air conditioner and make the fan run at the preset minimum speed.

[0037] When the minimum dew point temperature difference is less than or equal to the safety threshold and the condensation arrival time is greater than zero, the condensation risk level is determined to be the warning level. Multiple candidate combination control parameters are generated from heater power, fan speed and air conditioner start / stop status. The condensation arrival time is used as the time constraint to determine the recovery time and estimated energy consumption required for each candidate combination control parameter to restore the dew point temperature difference to above the safety threshold. Candidate combination control parameters with a recovery time less than the condensation arrival time are screened out. Then, the candidate combination control parameter with the lowest estimated energy consumption is selected from the screening results as the target combination control mode.

[0038] When the minimum dew point temperature difference is less than the safety threshold and the condensation arrival time is less than or equal to zero, the condensation risk level is determined to be emergency level. The candidate combination control parameters with the lowest estimated energy consumption are selected from the candidate combination control parameters with the shortest recovery time as the target combination control mode.

[0039] Furthermore, based on the analysis of the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet during the current sampling period using a physical information neural network, the recovery time and estimated energy consumption required to restore the dew point temperature difference to above the safe threshold are determined. The physical information neural network includes an input layer for receiving the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet; a hidden layer for extracting features from the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet and combining them with thermal balance and humidity balance relationships for physical constraints; and an output layer for outputting the recovery time and estimated energy consumption.

[0040] The present invention also provides an intelligent electrical cabinet adaptive environmental control system, comprising:

[0041] The parameter acquisition and dew point temperature difference determination module is used to acquire the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period, so as to calculate the heating power and dew point temperature, and determine the current dew point temperature difference.

[0042] The thermal and humidity characteristic identification and prediction module is used to identify the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet based on the environmental state parameters, the heating power and the current dew point temperature difference, generate the dew point temperature difference change curve, and extract the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero in the future sampling period.

[0043] The combined control module is used to compare the minimum dew point temperature difference with the preset safety threshold, determine the condensation risk level based on the comparison result and the condensation arrival time, and select the combined control mode of the heater, fan and air conditioner in the electrical cabinet according to the condensation risk level.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] By identifying the thermal and humidity characteristics and power distribution coefficients of the electrical cabinet in real time, a dew point temperature difference prediction curve is generated. The minimum dew point temperature difference and condensation arrival time are obtained in advance and compared with the safety threshold that fluctuates in real time with environmental state parameters, heating power and actual risk costs. The condensation risk is dynamically divided into different levels, and the optimal combination control mode is selected among heaters, fans and air conditioners accordingly. It can adapt to the thermal and humidity characteristics of the cabinet and load changes, and actively intervene before condensation occurs. It can effectively prevent condensation from harming equipment safety and avoid unnecessary energy consumption caused by over-control. It realizes intelligent environmental control that balances protection reliability and operation economy. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0047] Figure 1 This is a schematic diagram of the adaptive environmental control method for intelligent electrical cabinets according to the present invention;

[0048] Figure 2 This is a schematic diagram illustrating the process for determining the safety or condensation risk status of each historical sampling period according to the present invention.

[0049] Figure 3 This is a schematic diagram of the intelligent electrical cabinet adaptive environment control system of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The following is combined Figures 1-3 This invention describes an intelligent electrical cabinet adaptive environmental control method and system.

[0052] Example 1

[0053] Please see Figure 1 and Figure 2 This invention provides an adaptive environmental control method for intelligent electrical cabinets, comprising the following steps:

[0054] Step S100: Obtain the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period, so as to calculate the heating power and dew point temperature, and determine the current dew point temperature difference.

[0055] The environmental parameters inside and outside the electrical cabinet include the cabinet temperature, cabinet humidity, cabinet inner wall temperature, cabinet outer ambient temperature, and cabinet outer ambient humidity.

[0056] The calculation of heating power and dew point temperature, and the determination of the current dew point temperature difference, includes the following steps:

[0057] Step S110: During the current sampling period, simultaneously collect the temperature and humidity at multiple heights inside the electrical cabinet, the temperature at multiple locations on the inner side of the cabinet wall, the ambient temperature outside the cabinet, the ambient humidity outside the cabinet, and the main circuit current, and calculate the heating power based on the main circuit current.

[0058] Inside the electrical cabinet, multiple integrated temperature and humidity sensors are installed vertically in layers to simultaneously collect the temperature and humidity inside the cabinet. Thermocouples are attached to multiple representative locations on the inner surface of the cabinet wall to collect the temperature inside the cabinet wall. Outside the cabinet, an ambient temperature and humidity sensor is installed to collect the ambient temperature and humidity outside the cabinet. These sensors are connected to the data acquisition module via signal cables, and the controller reads and stores them as environmental status parameters inside and outside the electrical cabinet.

[0059] At the start of each sampling period, the controller sequentially reads the measured values ​​from all integrated temperature and humidity sensors inside the cabinet, thermocouples on the inner wall of the cabinet, ambient temperature and humidity sensors outside the cabinet, and the main circuit current transformer via the internal bus. After reading, the controller calls a heat power calculation function, which solves the problem of calculating heat power from current. The heat power calculation function used is as follows:

[0060]

[0061] in, The current sampling period represents the total heat generation power of the main circuit of the electrical cabinet, I represents the current current collected by the current transformer of the main circuit, and R represents the equivalent resistance value of the main circuit.

[0062] Step S120: Treat the temperature and humidity at each height measuring point inside the cabinet as an independent operating condition for that measuring point, and calculate the dew point temperature for each measuring point.

[0063] The controller retrieves the data packets for the current sampling period from the memory buffer, iterates through the temperature and humidity values ​​corresponding to each height measurement point inside the cabinet, and for each measurement point, the controller calls a dew point temperature calculation function to calculate the dew point temperature. This dew point temperature calculation function is as follows:

[0064]

[0065] in, The saturated vapor pressure parameter is related to temperature. For Magnus's empirical coefficient, = ; For Magnus's empirical coefficient, ; The temperature inside the cabinet is measured at a certain height. Dew point temperature, The humidity inside the cabinet was collected at this height measurement point.

[0066] By calculating the dew point temperature at each height measurement point as an independent operating condition, the difference in dew point temperature in the vertical direction inside the cabinet caused by temperature stratification and uneven humidity distribution is accurately characterized, providing a spatially discernible basis for multi-location condensation risk assessment.

[0067] Step S130: Calculate the difference between each collected temperature inside the cabinet and each dew point temperature, and take the minimum difference as the current dew point temperature difference.

[0068] The controller extracts the cabinet wall inner temperature array and the dew point temperature array calculated in the previous step from the current data packet. The controller executes a nested loop algorithm, with the outer loop iterating through each cabinet wall inner temperature value and the inner loop iterating through each dew point temperature value, calculating the difference between the two. The controller temporarily stores the difference calculated each time in a temporary temperature difference array. After the iteration is complete, the controller calls a minimum value function to retrieve the smallest difference value from all elements of the temperature difference array and defines it as the current dew point temperature difference in the current sampling period.

[0069] By calculating the difference point by point globally and taking the minimum value, the system can automatically locate the most unfavorable wall surface and the most humid air combination inside the electrical cabinet and quantify the risk of condensation, providing the most conservative and reliable risk assessment benchmark for control decisions.

[0070] Step S200: Based on the environmental state parameters inside and outside the electrical cabinet, the heating power and the current dew point temperature difference, identify the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet, generate the dew point temperature difference change curve, and extract the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero during the future sampling period.

[0071] The thermal and humidity characteristic parameters include thermal resistance, heat capacity, moisture time constant, and dehumidification efficiency.

[0072] The process of identifying the thermal and humidity characteristics of the electrical cabinet and generating a dew point temperature difference curve includes the following steps:

[0073] Step S210: Extract multiple consecutive historical sampling periods from the adjacent historical time periods, obtain the current dew point temperature difference of each historical sampling period, extract the temperature of a specific measuring point on the inner side of the cabinet wall that generates the current dew point temperature difference, the temperature and humidity of a measuring point at a specific height inside the cabinet, and obtain the ambient temperature, ambient humidity and heating power outside the cabinet, and arrange them according to the chronological relationship of the historical sampling periods to form a local state sequence.

[0074] Once the current dew point temperature difference is determined, the controller first extracts data from its historical database, using the current sampling period as a reference, and then extracts data from multiple consecutive historical sampling periods, including the current period. For each extracted historical sampling period, the controller reads the stored current dew point temperature difference and extracts the temperature of a specific measuring point on the inner side of the cabinet wall that generated the current dew point temperature difference, as well as the temperature and humidity of a measuring point at a specific height inside the cabinet, based on the associated measuring point identifier recorded when the current dew point temperature difference was generated. At the same time, the external ambient temperature, external ambient humidity, and heating power of each historical sampling period are also extracted.

[0075] The controller arranges these data into a two-dimensional data table according to the time sequence of historical sampling periods. Each row represents a historical sampling period, and each column represents the temperature of a specific measuring point on the inside of the cabinet wall, the temperature of a measuring point at a specific height inside the cabinet, the humidity of a measuring point at a specific height inside the cabinet, the ambient temperature outside the cabinet, the ambient humidity outside the cabinet, the heating power, and the current dew point temperature difference. This data table is the local state sequence for the most dangerous local area.

[0076] By associating the current dew point temperature difference with the specific physical location that generates that temperature difference and constructing a time series, it is possible to extract the state evolution data of the weakest locality from the global state.

[0077] Step S220: According to the law of conservation of energy, a thermal balance relationship is established for the local area that generates the current dew point temperature difference. The thermal balance relationship includes the thermal resistance, thermal capacity, and power distribution coefficient representing the total heat generation power distributed to the local area.

[0078] The thermal equilibrium relationship is expressed as follows:

[0079]

[0080] in, For the heat capacity of a local area; This represents the derivative of the temperature at a specific location on the inside of the cabinet wall with respect to time, i.e., the rate of change of the cabinet wall temperature. This represents the power distribution factor, i.e., the total heat generation power. The proportion allocated to this local area; This indicates the temperature at a specific height point inside the cabinet; This indicates the temperature measured at a specific location on the inside of the cabinet wall; This indicates the thermal resistance between the air inside the cabinet and the cabinet wall in a localized area; Indicates the ambient temperature outside the cabinet; This indicates the thermal resistance of a local area through the cabinet wall to the external environment.

[0081] It should also be noted that, due to the uneven distribution of heating elements inside the electrical cabinet, only a portion of the heat will directly affect the cabinet wall temperature in the most dangerous area. Therefore, the power distribution coefficient α is between 0 and 1.

[0082] Step S230: According to the law of conservation of mass, establish a moisture balance relationship for the local area, which includes the moisture time constant and dehumidification efficiency.

[0083] The moisture balance relationship is expressed as follows:

[0084]

[0085] in, This indicates the absolute humidity measured at a specific height inside the cabinet. This indicates the temperature at a specific height point inside the cabinet; This indicates the temperature measured at a specific location on the inside of the cabinet wall; Indicates the humidity at a specific height point inside the cabinet; This indicates the absolute saturation humidity at a specific measuring point on the inside of the cabinet wall; Indicates the moisture time constant; This indicates the absolute humidity change rate at a specific height point inside the cabinet. This indicates the rate of moisture generation inside the cabinet, which can be used as a constant or estimated based on historical data. This indicates the dehumidification efficiency.

[0086] Step S240: Using the local state sequence as input, and combining the thermal balance relationship and the moisture balance relationship, the thermal and moisture characteristic parameters and power distribution coefficient are identified.

[0087] The controller takes the local state sequence data table as input and calls a least-squares parameter identification algorithm to minimize the sum of squared residuals between the predicted cabinet wall temperature and humidity for the next sampling period based on the thermal and humidity balance relationships and the actual measured values ​​recorded in the local state sequence. The parameter vector to be identified represents the thermal resistance between the air inside the cabinet and the cabinet wall in the local area. Thermal resistance of localized areas through the cabinet walls to the external environment Power distribution coefficient Heat capacity of local areas Dehumidification efficiency Moisture time constant .

[0088] Step S250: Obtain the temperature of a specific measuring point on the inner side of the cabinet wall, the temperature and humidity of a measuring point at a specific height inside the cabinet, the ambient temperature outside the cabinet, and the heating power corresponding to the current dew point temperature difference during the current sampling period. Combine the heat balance relationship and the humidity balance relationship to calculate the predicted values ​​of the cabinet wall temperature and the humidity inside the cabinet for each future sampling period.

[0089] The controller will display the measured values ​​during the current sampling period (i.e., the temperature at a specific measuring point on the inside of the cabinet wall). Temperature at a specific height measuring point inside the cabinet Humidity at a specific height inside the cabinet External ambient temperature of the cabinet Using the total heat generation power P of the main circuit of the electrical cabinet as the initial state, a numerical differential equation solver is invoked to calculate the thermal resistance between the air inside the local area and the cabinet wall identified in step S240. Thermal resistance of localized areas through the cabinet walls to the external environment Power distribution coefficient Heat capacity of local areas Dehumidification efficiency Moisture time constant As a constant input, the solver iteratively solves the thermal balance and humidity balance relationships with the sampling period as the step size. In each iteration, the derivative term is calculated using the value of the current step, and then the predicted values ​​of the cabinet wall temperature and the cabinet humidity for the next sampling period are calculated. This process is repeated until the preset future sampling period length is covered, generating a sequence of predicted values ​​of the cabinet wall temperature and the cabinet humidity for each future sampling period.

[0090] Step S260: Based on the predicted values ​​of cabinet wall temperature and cabinet humidity, calculate the predicted values ​​of dew point temperature for each future sampling period, and calculate the difference between the predicted values ​​of cabinet wall temperature and dew point temperature point by point to form a dew point temperature difference change curve.

[0091] For each future sampling period, the predicted values ​​of cabinet wall temperature and cabinet humidity are used as inputs. The dew point temperature calculation function is called, using the predicted cabinet wall temperature as the temperature parameter in the function input, to calculate the predicted dew point temperature for that future sampling period. Then, the predicted cabinet wall temperature for that future sampling period is subtracted from the newly calculated predicted dew point temperature to obtain the predicted dew point temperature difference for that sampling period. All the predicted dew point temperature difference data points are connected in chronological order of the future sampling periods to form a curve, which is the dew point temperature difference change curve.

[0092] Mark the intersection of this curve with the time axis as the condensation arrival time, and find the minimum value within the entire prediction interval, marking it as the minimum dew point temperature difference.

[0093] Step S300: Compare the minimum dew point temperature difference with the preset safety threshold, determine the condensation risk level based on the comparison result and the condensation arrival time, and select the combined control mode of the heater, fan and air conditioner in the electrical cabinet according to the condensation risk level.

[0094] The preset security threshold includes the following steps:

[0095] Step S310: Obtain environmental state parameters, heating power and corresponding current dew point temperature difference for multiple historical sampling periods, and mark the safety status or condensation risk status of each historical sampling period.

[0096] Determining the safety or condensation risk status for each historical sampling period includes the following steps:

[0097] Step S311: Compare the temperature of the measuring point at a specific location on the inner side of the cabinet wall that generates the current dew point temperature difference with the dew point temperature of the measuring point at a specific height inside the cabinet.

[0098] First, based on the minimum dew point temperature difference correlation information stored in the record, the specific location of the measuring point on the inner side of the cabinet wall and the measuring point at a specific height inside the cabinet that generated the current dew point temperature difference are located. Then, the controller extracts the temperature and humidity of the measuring point at the specific location on the inner side of the cabinet wall and the measuring point at the specific height inside the cabinet from the record, calls the dew point temperature calculation function, and calculates the dew point temperature at that location during the sampling period using the temperature and humidity of the measuring point at the specific height inside the cabinet. Finally, the temperature of the measuring point at the specific location on the inner side of the cabinet wall is compared with the calculated dew point temperature.

[0099] Step S312: In response to the existence of multiple consecutive historical sampling periods, the temperature of the measuring point at a specific location on the inner side of the cabinet wall in each period is lower than the dew point temperature of the measuring point at a specific height inside the cabinet, and the number of these multiple consecutive historical sampling periods reaches the preset number of condensation risk confirmation periods, these multiple consecutive historical sampling periods are defined as condensation risk status, and the remaining historical sampling periods are defined as safe status.

[0100] The preset number of condensation risk confirmation periods is set based on the safety requirements and operation and maintenance experience of electrical cabinet equipment. For example, it is set to 5 consecutive periods to avoid misjudging instantaneous fluctuations as risks. By introducing the preset number of condensation risk confirmation periods, isolated and brief moments when the temperature is below the dew point are filtered out, ensuring that only event sequences that form stable condensation conditions are identified as risks, thereby generating historical labels that are closer to the real situation and have less noise.

[0101] Step S320: Construct a first probability density distribution for the current dew point temperature difference of historical sampling periods that are in a safe state using the kernel density estimation method, and construct a second probability density distribution for the current dew point temperature difference of historical sampling periods that are in a condensation risk state using the kernel density estimation method.

[0102] The mathematical expression for the probability density distribution is as follows:

[0103]

[0104] in, Denotes the first probability density distribution function; This represents the total number of samples during the historical sampling period that falls under a safe state. The bandwidth parameter represents the first probability density distribution; This represents the Gaussian kernel function that represents a safe state. This represents the second probability density distribution function; This represents the total number of samples from historical sampling periods that fall under the condensation risk status; The bandwidth parameter represents the second probability density distribution; This represents the Gaussian kernel function that indicates a state of condensation risk.

[0105] The mathematical expression for the Gaussian kernel function is as follows:

[0106]

[0107] in, Represents the standardized residual. This is the current dew point temperature difference calculated from real-time measurements during the current sampling period. This indicates the current dew point temperature difference when the j-th historical sampling period prior to the current sampling period was either in a safe state or a state at risk of condensation. The bandwidth parameter represents either the first probability density distribution or the second probability density distribution. , This represents the standard deviation of the current dew point temperature difference sample. This represents the total number of samples from historical sampling periods that fall under either a safe or condensation-risk condition.

[0108] Step S330: Based on the environmental state parameters and heating power of the current sampling period, determine the prior probability that the current sampling period belongs to the condensation risk state.

[0109] The mathematical expression for the prior probability that the current sampling period belongs to a condensation risk state is as follows:

[0110]

[0111] in, This represents the prior probability that the current sampling period falls under the condensation risk state; This indicates the ambient temperature outside the cabinet during the current sampling period; This indicates the ambient humidity outside the cabinet during the current sampling period. This indicates the heating power during the current sampling period; This represents the conditional probability of observing this set of operating condition characteristics under condensation risk conditions; This represents the basic probability of condensation risk status obtained from historical data statistics, which is the number of condensation risk status periods in the historical sampling period divided by the total number of periods; This represents the marginal probability of observing the characteristics of this set of operating conditions.

[0112] Based on the feature conditional independence assumption, conditional probability It can be decomposed into the product of three independent conditional probabilities, i.e. These three conditional probabilities are obtained by kernel density estimation of the corresponding features of condensation risk state samples in historical data.

[0113] By determining dynamic prior probabilities based on operating condition characteristics, the initial risk perception can be adaptively adjusted according to changes in environment and load, providing a more reasonable prior basis for subsequent integration of real-time temperature difference observations.

[0114] Step S340: Based on the first probability density distribution, the second probability density distribution, and the prior probability, use Bayes' theorem to calculate the posterior probability that the current sampling period actually belongs to the condensation risk state under the current dew point temperature difference.

[0115] Obtain the current dew point temperature difference value calculated during the current sampling period. In the first probability density distribution function Second probability density distribution function The above calculation is in The likelihood value is obtained by taking the value at the given location. and Then, the Bayesian formula is used to calculate the posterior probability that the current sampling period is actually in a condensation risk state under the current dew point temperature difference.

[0116] The expression for the posterior probability is as follows:

[0117]

[0118] in, This indicates that the current dew point temperature difference is... Under the given conditions, the posterior probability of actually belonging to a condensation risk state. Indicates the current dew point temperature difference The likelihood value under the second probability density distribution. Indicates the current dew point temperature difference The likelihood value under the first probability density distribution This represents the prior probability.

[0119] Step S350: The additional estimated energy consumption of heaters, fans and air conditioners caused by misjudging the safe state as a condensation risk level is taken as the false alarm cost, and the condensation loss caused by misjudging the condensation risk state as a safe level is taken as the false alarm cost. Based on the false alarm cost, the false alarm cost and the posterior probability, the current dew point temperature difference judgment boundary that minimizes the expected total cost is determined, and this judgment boundary is taken as the safety threshold.

[0120] The mathematical expression for the expected total cost is as follows:

[0121]

[0122] in, This represents the expected total cost; This indicates the cost of underreporting; Indicates the temperature difference judgment boundary for candidates; This indicates that historical data is used to statistically determine when the actual risk state is a condensation risk state, and the temperature difference boundary is set to... The probability of missed detection at that time; Indicates the cost of false alarms; This indicates that historical data is used to statistically determine when the actual risk state is a safe state and the temperature difference boundary is set to... The probability of false alarms at that time; This indicates that the current dew point temperature difference is... The posterior probability that the actual state is a condensation risk under the given conditions.

[0123] The preset security threshold also includes the following steps:

[0124] Step S360: Based on the heat capacity, heating power, and environmental state parameter variation range of the electrical cabinet in adjacent historical sampling periods, determine the maximum allowable rate of change of the current dew point temperature difference in adjacent historical sampling periods.

[0125] The mathematical expression for the maximum permissible rate of change is as follows:

[0126]

[0127] in, This represents the absolute value of the maximum permissible rate of change of the current dew point temperature difference over adjacent historical sampling periods; Indicates the heat capacity of a local area; This represents the maximum possible change in heating power between adjacent time periods, derived from historical data. This represents the maximum possible change in temperature inside the cabinet between adjacent time periods, derived from historical data. This represents the maximum possible change in ambient temperature outside the cabinet between adjacent time periods, derived from historical data.

[0128] Step S370: In response to the fact that the rate of change of the current dew point temperature difference of a certain historical sampling period relative to the previous historical sampling period exceeds the maximum allowable rate of change, the current dew point temperature difference of that historical sampling period is removed.

[0129] Because outliers in the current dew point temperature difference may occur due to sensor malfunctions, cabinet door openings, or sudden disturbances during historical sampling periods, directly using the environmental state parameters and heating power of historical sampling periods and their corresponding current dew point temperature differences to determine the safety or condensation risk status of each historical sampling period may cause subsequent judgment boundaries to deviate from the true physical laws. Therefore, it is necessary to remove the abnormal current dew point temperature difference data of the historical sampling period.

[0130] Step S380: Based on the historical sampling period data after removing outliers, determine the safety status or condensation risk status of each historical sampling period.

[0131] By eliminating outliers based on thermal inertia constraints, we ensure that the historical samples of the constructed probability density distribution reflect the normal operating conditions of the electrical cabinet, thereby improving the conformity of the safety threshold to physical laws and avoiding extreme outlier sampling from pushing the threshold to overly conservative or overly aggressive extremes.

[0132] The selection of the combined control mode for the heater, fan, and air conditioner in the electrical cabinet based on the condensation risk level includes the following steps:

[0133] Step S391: When the minimum dew point temperature difference is greater than the safety threshold, the condensation risk level is determined to be a safe level, and a combined control mode is selected that turns off the heater and air conditioner and makes the fan run at the preset minimum speed.

[0134] Under the safety level, the ventilation strategy of maintaining the lowest power consumption is actively selected instead of completely shutting down. This saves energy and avoids the risk of local hot spots or humidity accumulation that may be caused by the complete stillness of the air inside the cabinet.

[0135] Step S392: When the minimum dew point temperature difference is less than or equal to the safety threshold and the condensation arrival time is greater than zero, the condensation risk level is determined to be a warning level. Multiple candidate combination control parameters are generated from heater power, fan speed and air conditioner start / stop status. The condensation arrival time is used as a time constraint to determine the recovery time and estimated energy consumption required for each candidate combination control parameter to restore the dew point temperature difference to above the safety threshold. Candidate combination control parameters with a recovery time less than the condensation arrival time are selected. Then, the candidate combination control parameter with the lowest estimated energy consumption is selected from the selection results as the target combination control mode.

[0136] Specifically, based on the analysis of the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet during the current sampling period using a physical information neural network, the recovery time and estimated energy consumption required to restore the dew point temperature difference to above the safe threshold are determined. The physical information neural network includes an input layer for receiving the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet; a hidden layer for extracting features from the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet and combining them with thermal balance and humidity balance relationships for physical constraints; and an output layer for outputting the recovery time and estimated energy consumption.

[0137] The number of nodes in the input layer is the same as the total number of environmental state parameters, thermal and humidity characteristic parameters, power distribution coefficients, and candidate combination control parameters inside and outside the electrical cabinet. These are used to receive the cabinet temperature, cabinet humidity, cabinet inner wall temperature, cabinet outer ambient temperature, cabinet outer ambient humidity, heating power, thermal resistance, heat capacity, power distribution coefficient, humidity time constant, dehumidification efficiency, heater power, fan speed, and air conditioner start / stop status. The hidden layer can be set to 6 to 8 layers, and the number of nodes in each layer can be set according to the actual situation, such as 64, 128, or 256 nodes. The ReLU function is used to improve the nonlinear expression capability of the model. The output layer has two nodes, which are used to output the recovery time and estimated energy consumption.

[0138] The process of training a physical information neural network is as follows:

[0139] Multiple internal and external environmental conditions and multiple candidate combinations of control parameters are preset, and several different operating scenarios are constructed by permutation and combination.

[0140] The environmental condition working conditions are constructed by discretely sampling the cabinet internal temperature, cabinet internal humidity, cabinet inner wall temperature, cabinet external ambient temperature, and cabinet external ambient humidity within their respective typical variation ranges; discretely sampling the heat generation power within the electrical cabinet from light load to full load range; and discretely sampling the thermal and humidity characteristic parameters and power distribution coefficient within the distribution range of the identified parameters, combining them to form a working condition sample set covering various possible states.

[0141] The candidate combination control parameters are constructed by discretely sampling the heater power from zero to rated power, discretely sampling the fan speed from minimum speed to maximum speed, and taking two states for the air conditioner start-stop state: on and off.

[0142] Each set of operating conditions in the operating condition sample set is arranged and combined with candidate combination control parameters to form several operating scenarios. For each operating scenario, a numerical simulation model is established using the thermal balance relationship and moisture balance relationship of the electrical cabinet. Through numerical iteration, the recovery time required for the dew point temperature difference to recover to above the safe threshold from the initial state under this operating scenario is calculated, as well as the total estimated energy consumption of the heater, fan and air conditioner during the recovery process. The input parameters of each operating scenario, the corresponding recovery time and the estimated energy consumption are collected as a sample to form a sample set.

[0143] As one implementation method, the sampling range for cabinet internal temperature is 0℃-60℃, with a sampling interval of 5 degrees Celsius; the sampling range for cabinet internal humidity is 20%-95%, with a sampling interval of 5%; the sampling range for cabinet inner wall temperature is 0℃-50℃, with a sampling interval of 5℃; the sampling range for cabinet external ambient temperature is -20℃-50℃, with a sampling interval of 5℃; the sampling range for cabinet external ambient humidity is 10%-100%, with a sampling interval of 10%; the sampling range for heating power is 0 watts to the rated heating power of the electrical cabinet, with a sampling interval of 50 watts; the sampling range for heater power is 0 watts to the rated power of the heater, with a sampling interval of 100 watts; and the sampling range for fan speed is from the lowest fan speed to the highest fan speed, with a sampling interval of 10% of the speed setting.

[0144] It should be noted that the numerical simulation model established using the thermal balance relationship and the moisture balance relationship can be constructed based on the thermal resistance, heat capacity, power distribution coefficient, moisture time constant and dehumidification efficiency identified in step S240. This numerical simulation model can simulate the dynamic change process of cabinet wall temperature and cabinet humidity under the control of different heater power, fan speed and air conditioner start and stop states, and then calculate the trajectory of dew point temperature difference changing with time. The recovery time and estimated energy consumption are extracted from the trajectory as sample labels. This is existing technology and will not be elaborated here.

[0145] The sample set is divided into training, testing, and validation sets in a ratio of 70:15:15. Batch size and training period are set, with the batch size set between 32 and 256 and the training period set between 50 and 100. The environmental state parameters, thermal and humidity characteristics, power distribution coefficients, and candidate combination control parameters inside and outside the electrical cabinet in the training set are used as inputs, and the corresponding recovery time and estimated energy consumption are used as output labels to train the physical information neural network.

[0146] The loss function of this physical information neural network consists of two parts: the first part is the mean square error between the predicted recovery time and the predicted energy consumption of the network output and the sample label value; the second part is the physical residual penalty term for the thermal balance relationship and the moisture balance relationship.

[0147] The physical residual penalty term is calculated by constructing residual expressions for the heat balance equation and the moisture balance equation using features extracted from the hidden layer. The mean square value of the residuals is added as an additional loss term to the total loss function. The total loss function is expressed as a weighted sum of the mean square error of the data and the physical residual penalty term. The weight coefficient of the physical residual penalty term can be set between 0.1 and 1.0. During training, the model parameters are updated through backpropagation to minimize the total loss function. Specifically, the model parameters can be updated using the Adam optimization algorithm. During training, the model hyperparameters, such as the learning rate, batch size, number of hidden layers, and weight coefficient of the physical residual penalty term, are adjusted using the validation set to optimize model performance. The initial value of the learning rate can be set between 0.001 and 0.01.

[0148] After the predetermined training period is reached, the test set is input into the physical information neural network for performance testing. If the relative error is less than the preset threshold, the training is considered complete. Otherwise, the model hyperparameters are adjusted and retraining is performed. The preset threshold can be set between 3% and 8% to ensure that the trained model has a good simulation fitting ability. The specific value can be set by the staff according to the actual situation and is not restricted here.

[0149] It should also be noted that by embedding a physical residual penalty term in the loss function, the physical information neural network must comply with the physical laws described by the thermal equilibrium relationship and the moisture equilibrium relationship while learning the data mapping relationship.

[0150] After training, the physical information neural network is deployed into the inference module of the controller. During each sampling period, the controller inputs the current environmental state parameters inside and outside the electrical cabinet, the identified thermal and humidity characteristics parameters and power distribution coefficient, as well as the candidate combination control parameters that need to be evaluated into the network. The network can then quickly output the recovery time and estimated energy consumption under the combination control parameters.

[0151] Step S393: When the minimum dew point temperature difference is less than the safety threshold and the condensation arrival time is less than or equal to zero, the condensation risk level is determined to be an emergency level. The candidate combination control parameters with the lowest estimated energy consumption are selected from the candidate combination control parameters with the shortest recovery time as the target combination control mode.

[0152] At the emergency level, by prioritizing recovery speed as the primary optimization objective and supplementing it with energy consumption comparisons, the system is ensured to respond in the fastest and most economical way when condensation has occurred or is in a critical state, thus minimizing the duration and extent of condensation hazards.

[0153] Example 2

[0154] The following describes an intelligent electrical cabinet adaptive environment control system provided by the present invention. The intelligent electrical cabinet adaptive environment control system and the intelligent electrical cabinet adaptive environment control method described below can be referred to in correspondence with each other.

[0155] Please see Figure 3 The present invention provides an intelligent electrical cabinet adaptive environmental control system, including a parameter acquisition and dew point temperature difference determination module 410, a thermal and humidity characteristic identification and prediction module 420, and a combined control module 430.

[0156] The parameter acquisition and dew point temperature difference determination module 410 is used to acquire the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period, so as to calculate the heating power and dew point temperature, and determine the current dew point temperature difference.

[0157] The thermal and humidity characteristic identification and prediction module 420 is used to identify the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet based on the environmental state parameters, the heating power and the current dew point temperature difference, generate the dew point temperature difference change curve, and extract the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero in the future sampling period.

[0158] The combined control module 430 is used to compare the minimum dew point temperature difference with the preset safety threshold, determine the condensation risk level based on the comparison result and the condensation arrival time, and select the combined control mode of the heater, fan and air conditioner in the electrical cabinet according to the condensation risk level.

[0159] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0160] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0161] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for adaptive environmental control of an intelligent electrical cabinet, characterized in that, Includes the following steps: Obtain the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period to calculate the heating power and dew point temperature, and determine the current dew point temperature difference; Based on the environmental condition parameters inside and outside the electrical cabinet, the heating power and the current dew point temperature difference, the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet are identified, the dew point temperature difference change curve is generated, and the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero are extracted in the future sampling period. The minimum dew point temperature difference is compared with the preset safety threshold. The condensation risk level is determined based on the comparison result and the condensation arrival time. The combined control mode of the heater, fan and air conditioner in the electrical cabinet is selected according to the condensation risk level.

2. The intelligent electrical cabinet adaptive environment control method according to claim 1, characterized in that, The environmental parameters inside and outside the electrical cabinet include the cabinet temperature, cabinet humidity, cabinet inner wall temperature, cabinet outer ambient temperature, and cabinet outer ambient humidity. The thermal and humidity characteristic parameters include thermal resistance, heat capacity, moisture time constant, and dehumidification efficiency.

3. The intelligent electrical cabinet adaptive environment control method according to claim 1, characterized in that, The calculation of heating power and dew point temperature, and the determination of the current dew point temperature difference, includes the following steps: During the current sampling period, the temperature and humidity at multiple heights inside the electrical cabinet, the temperature at multiple locations on the inner side of the cabinet wall, the ambient temperature outside the cabinet, the ambient humidity outside the cabinet, and the main circuit current are collected simultaneously, and the heating power is calculated based on the main circuit current. The temperature and humidity at each height measurement point inside the cabinet are treated as independent operating conditions for that measurement point, and the dew point temperature of each measurement point is calculated separately. The difference between each collected temperature inside the cabinet wall and each dew point temperature is calculated, and the minimum difference is taken as the current dew point temperature difference.

4. The intelligent electrical cabinet adaptive environment control method according to claim 1, characterized in that, The process of identifying the thermal and humidity characteristics of the electrical cabinet and generating a dew point temperature difference curve includes the following steps: Multiple consecutive historical sampling periods are extracted from adjacent historical periods. The current dew point temperature difference of each historical sampling period is obtained. The temperature and humidity of the measuring point at a specific location on the inner side of the cabinet wall that generates the current dew point temperature difference, the temperature and humidity of the measuring point at a specific height inside the cabinet, the ambient temperature, the ambient humidity and the heating power outside the cabinet are obtained, and arranged according to the chronological relationship of the historical sampling periods to form a local state sequence. According to the law of conservation of energy, a thermal balance relationship is established for the local area that generates the current dew point temperature difference. The thermal balance relationship includes the thermal resistance and heat capacity of the local area, as well as the power distribution coefficient that characterizes the total heat generation power distributed to the local area. According to the law of conservation of mass, a moisture balance relationship is established for the local area, which includes the moisture time constant and dehumidification efficiency. By taking the local state sequence as input and combining the thermal equilibrium relationship and the moisture equilibrium relationship, thermal and moisture characteristic parameters and power distribution coefficients are identified. The temperature of a specific measuring point on the inner side of the cabinet wall, the temperature and humidity of a measuring point at a specific height inside the cabinet, the ambient temperature outside the cabinet, and the heating power are obtained at the current dew point temperature difference during the current sampling period. Combined with the heat balance relationship and the humidity balance relationship, the predicted values ​​of the cabinet wall temperature and the humidity inside the cabinet are calculated for each sampling period in the future. Based on the predicted values ​​of cabinet wall temperature and cabinet humidity, the predicted values ​​of dew point temperature for each future sampling period are calculated, and the difference between the predicted values ​​of cabinet wall temperature and dew point temperature is calculated point by point to form a dew point temperature difference change curve.

5. The intelligent electrical cabinet adaptive environment control method according to claim 1, characterized in that, The preset security threshold includes the following steps: The system acquires environmental state parameters, heating power, and corresponding current dew point temperature difference for multiple historical sampling periods, and calibrates the safety status or condensation risk status of each historical sampling period. The first probability density distribution is constructed by using kernel density estimation for the current dew point temperature difference of historical sampling periods that are in a safe state, and the second probability density distribution is constructed by using kernel density estimation for the current dew point temperature difference of historical sampling periods that are in a condensation risk state. Based on the environmental state parameters and heat generation power during the current sampling period, determine the prior probability that the current sampling period belongs to a condensation risk state. Based on the first probability density distribution, the second probability density distribution, and the prior probability, the posterior probability of the current sampling period actually belonging to the condensation risk state under the current dew point temperature difference is calculated using Bayes' theorem. The additional estimated energy consumption of heaters, fans, and air conditioners caused by misjudging a safe state as a condensation risk level is taken as the false alarm cost, and the condensation loss caused by underestimating a condensation risk state as a safe state is taken as the underreport cost. Based on the false alarm cost, the underreport cost, and the posterior probability, the current dew point temperature difference judgment boundary that minimizes the expected total cost is determined, and this judgment boundary is taken as the safety threshold.

6. The intelligent electrical cabinet adaptive environment control method according to claim 5, characterized in that, The preset security threshold also includes the following steps: Based on the heat capacity, heating power of the electrical cabinet and the variation range of environmental state parameters in adjacent historical sampling periods, the maximum allowable rate of change of the current dew point temperature difference in adjacent historical sampling periods is determined. If the rate of change of the current dew point temperature difference in a certain historical sampling period relative to the previous historical sampling period exceeds the maximum allowable rate of change, the current dew point temperature difference of that historical sampling period is removed. Based on historical sampling data after removing outliers, the safety status or condensation risk status of each historical sampling period is identified.

7. The intelligent electrical cabinet adaptive environment control method according to claim 6, characterized in that, The determination of the safety status or condensation risk status of each historical sampling period includes the following steps: Compare the temperature at a specific location on the inside of the cabinet wall that generates the current dew point temperature difference with the dew point temperature at a specific height inside the cabinet. In response to the existence of multiple consecutive historical sampling periods, where the temperature at a specific location on the inner side of the cabinet wall is lower than the dew point temperature at a specific height inside the cabinet during each period, and the number of these multiple consecutive historical sampling periods reaches the preset number of condensation risk confirmation periods, these multiple consecutive historical sampling periods are defined as a condensation risk state, and the remaining historical sampling periods are defined as a safe state.

8. The intelligent electrical cabinet adaptive environment control method according to claim 1, characterized in that, The selection of the combined control mode for the heater, fan, and air conditioner in the electrical cabinet based on the condensation risk level includes the following steps: When the minimum dew point temperature difference is greater than the safety threshold, the condensation risk level is determined to be a safe level, and a combined control mode is selected to turn off the heater and air conditioner and make the fan run at the preset minimum speed. When the minimum dew point temperature difference is less than or equal to the safety threshold and the condensation arrival time is greater than zero, the condensation risk level is determined to be the warning level. Multiple candidate combination control parameters are generated from heater power, fan speed and air conditioner start / stop status. The condensation arrival time is used as the time constraint to determine the recovery time and estimated energy consumption required for each candidate combination control parameter to restore the dew point temperature difference to above the safety threshold. Candidate combination control parameters with a recovery time less than the condensation arrival time are screened out. Then, the candidate combination control parameter with the lowest estimated energy consumption is selected from the screening results as the target combination control mode. When the minimum dew point temperature difference is less than the safety threshold and the condensation arrival time is less than or equal to zero, the condensation risk level is determined to be emergency level. The candidate combination control parameters with the lowest estimated energy consumption are selected from the candidate combination control parameters with the shortest recovery time as the target combination control mode.

9. The intelligent electrical cabinet adaptive environment control method according to claim 8, characterized in that, The physical information neural network analyzes the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet during the current sampling period to determine the recovery time and estimated energy consumption required to restore the dew point temperature difference to above the safe threshold. The physical information neural network includes an input layer for receiving the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet; a hidden layer for extracting features from the environmental state parameters, thermal and humidity characteristics, power allocation coefficient, and candidate combination control parameters inside and outside the electrical cabinet and combining them with thermal balance and humidity balance relationships for physical constraints; and an output layer for outputting the recovery time and estimated energy consumption.

10. An intelligent electrical cabinet adaptive environment control system, used to execute the intelligent electrical cabinet adaptive environment control method according to any one of claims 1-9, characterized in that, include: The parameter acquisition and dew point temperature difference determination module is used to acquire the environmental status parameters inside and outside the electrical cabinet and the load current during the current sampling period, so as to calculate the heating power and dew point temperature, and determine the current dew point temperature difference. The thermal and humidity characteristic identification and prediction module is used to identify the thermal and humidity characteristic parameters and power distribution coefficient of the electrical cabinet based on the environmental state parameters, the heating power and the current dew point temperature difference, generate the dew point temperature difference change curve, and extract the minimum dew point temperature difference and the condensation arrival time when the dew point temperature difference drops to zero in the future sampling period. The combined control module is used to compare the minimum dew point temperature difference with the preset safety threshold, determine the condensation risk level based on the comparison result and the condensation arrival time, and select the combined control mode of the heater, fan and air conditioner in the electrical cabinet according to the condensation risk level.