A method and system for monitoring the temperature of a ring network enclosure based on data fusion

Abstract

This invention belongs to the field of power distribution network operation monitoring technology, specifically relating to a method and system for monitoring ring main unit temperature based on data fusion. The method includes: collecting historical load current sequences and actual temperatures for each phase in the ring main unit and writing them into a time-aligned buffer; using a thermal time constant to accumulate and calculate the predicted temperature of the target phase by decaying over time based on the historical load current sequence; converting the abnormal temperature rise (actual temperature exceeding predicted temperature) into redundant heat energy, and using this to inversely deduce the resistivity degradation of the target phase based on the historical load current sequence; determining the heat source correlation between the target phase and adjacent phases based on the temporal fluctuation correlation between the redundant heat sequence and the historical heat generation characteristic sequences of each phase; and determining the contact degradation state of the target phase based on the resistivity degradation of the target phase. This invention can compensate for heat conduction hysteresis effects and eliminate spatial thermal radiation interference, achieving accurate identification of the true degradation state.

Description

Technical Field

[0001] This invention relates to the field of power distribution network operation monitoring technology. More specifically, this invention relates to a method and system for monitoring the temperature of ring main units based on data fusion. Background Technology

[0002] In the operation monitoring of distribution network ring main units, measuring the contact resistance at the joints is a core means of preventing insulation breakdown. For high-voltage insulation safety considerations, temperature sensors cannot directly contact live conductors and are usually placed on the outer surface of the insulating bushing. Existing systems indirectly assess whether the internal contact resistance exceeds the standard by synchronously collecting the temperature of the outer surface of the insulating bushing and directly registering and comparing it with the current line load current.

[0003] However, the apparent temperature relied upon by existing data synchronization and comparison methods is highly susceptible to interference from both internal and external physical factors, leading to severe distortion of the underlying evaluation logic. On the one hand, media such as epoxy resin have heat capacity and thermal damping characteristics, resulting in a long lag time for internal heat conduction to the outside. When the load increases sharply, high temperatures are already generated internally while the surface temperature has not risen, causing the system to draw a normal but erroneous conclusion. On the other hand, when the load drops sharply, the delayed heat reaches the surface, and the system compares low current with high temperature rise, thus triggering false alarms.

[0004] On the other hand, due to the extremely small internal space of the ring main unit and the extremely close physical distance between the three-phase connectors, when a conductor of a certain phase experiences severe contact deterioration and emits a huge amount of heat, the intense thermal radiation and convection in the closed air chamber will force the surface temperature of the adjacent healthy phase to rise abnormally passively. The existing independent single-phase monitoring cannot identify and eliminate the spatial heat interference transmitted from the adjacent phase, and will mistakenly attribute it to the internal resistance heating of the healthy phase itself, thus triggering multiple false alarms, making it impossible for on-site maintenance personnel to accurately locate the real source of the defect. Summary of the Invention

[0005] To address the technical problem that existing data comparison methods are unable to accurately identify the true source of degradation due to thermal conduction hysteresis and spatial thermal interference, this invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for monitoring the temperature of a ring network enclosure based on data fusion, comprising: The historical load current sequence and actual temperature of each phase in the continuously collected ring network box are written into the time-aligned buffer, and the ambient temperature at the current moment is obtained. Using the thermal time constant of the target phase insulating bushing, the historical load current sequence of the target phase in the time-aligned buffer is accumulated and calculated to decay over time, so as to reconstruct the cumulative contribution of historical electrical input to the current thermal state on the time axis, and combine it with the ambient temperature to obtain the predicted temperature of the target phase at the current moment, so as to compensate for the thermal conduction hysteresis of the insulating bushing. The abnormal temperature rise of the target phase is calculated when the actual temperature is higher than the predicted temperature. Based on the principle of energy conservation, the abnormal temperature rise is converted into redundant heat energy accumulated inside by using the equivalent heat capacity. Combined with the historical load current sequence in the time-aligned buffer, the resistance degradation variable corresponding to the internal conductor resistance degradation of the target phase is derived in reverse. Within the association window, construct the redundant heat sequence of the target phase, the intrinsic heat sequence of the target phase, and the intrusive heat sequence of the adjacent phase; calculate the heat source correlation between the redundant heat sequence and the intrinsic heat sequence of the target phase, and the heat source correlation between the redundant heat sequence and each intrusive heat sequence of the adjacent phase. The contact degradation state of the target phase is determined based on the resistivity of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases.

[0007] Preferably, the predicted temperature satisfies the expression: ; In the formula, This indicates the predicted temperature of the outer surface of the target phase insulating bushing at the current moment; This indicates the ambient temperature measured at the current moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates counting backwards from the current time. The root mean square current of the target phase at each time step; Indicates the heat conversion coefficient of the target phase; Represents the natural constant; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing.

[0008] Preferably, the reverse derivation of the resistance degradation due to the deterioration of the conductor resistance inside the target phase includes: When the difference between the actual and predicted temperatures of the target phase is greater than 0, the resistivity inferior variable satisfies the expression: ; In the formula, The resistivity variation of the target phase; Indicates the actual temperature of the target phase; Indicates the predicted temperature of the target phase; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates going back 1 / 2 from the current time. The root mean square current of the target phase line at each time step; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing; Represents the natural constant; This represents the equivalent heat capacity of the target phase.

[0009] Preferably, it further includes: If the difference between the actual temperature and the predicted temperature of the target phase is less than or equal to 0, the resistivity of the target phase is determined to be 0.

[0010] Preferably, the construction of the redundant thermal sequence of the target phase, the intrinsic thermal sequence of the target phase, and the intrusive thermal sequence of the adjacent phase includes: Extract redundant heat accumulated inside the target phase at each sampling time. This constitutes a redundant heat sequence; Extract the heat integral term of the target phase at the corresponding time. This constitutes the intrinsic heating sequence of the target phase; the nearest neighboring phase in physical space is obtained. If there are multiple nearest neighboring phases with equal distances, the heating integral term of each nearest neighboring phase at the corresponding time is obtained, and the corresponding adjacent phase intrusion heating sequence is constituted. in, Indicates the target phase in the first... The excess heat accumulated internally at a given moment Indicates equivalent heat capacity, Indicates the target phase in the first... The actual temperature at that moment. Indicates the target phase in the first... The predicted temperature at that moment. Indicates the target phase in the first... The heat integral term at each moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; Indicates the target phase in the first... Counting back from the first moment The line root mean square current at each time step. Represents the natural constant; Indicates the time step; This represents the thermal time constant of the insulating bushing.

[0011] Preferably, the correlation degree of the target phase heat source satisfies the expression: ; In the formula, Indicates the correlation degree of the target phase heat source; Indicates the length of the associated window; Indicates the time sliding sequence number within the associated window; Indicates the target phase in the first... The excess heat accumulated internally at a given moment; Represents the arithmetic mean of redundant heat sequences; Indicates the target phase in the first... The intrinsic heating sequence values ​​of the target phase at each moment; This represents the arithmetic mean of the intrinsic heating sequence of the target phase.

[0012] Preferably, determining the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases includes: If the resistance of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the heat source correlation of the target phase is greater than the heat source correlation of the adjacent phase, and the heat source correlation of the target phase is greater than the preset correlation judgment threshold, while the heat source correlation of the adjacent phase is less than or equal to the correlation judgment threshold, it is determined that the target phase has experienced contact oxidation and loosening, and an alarm command indicating a fault is output.

[0013] Preferably, determining the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases further includes: If the resistance variation of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the heat source correlation between the target phase and any nearest adjacent phase is greater than the heat source correlation of the target phase, and the heat source correlation of the adjacent phase is greater than the preset correlation judgment threshold, while the heat source correlation of the target phase is less than or equal to the correlation judgment threshold, it is determined that the temperature rise of the target phase is a heating phenomenon induced by the high temperature radiation of the adjacent phase, and the alarm command for the target phase is discarded.

[0014] Preferably, determining the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases further includes: In response to the target phase's resistance degradation exceeding a preset tolerance limit within a preset tolerance time window, and the correlation between the target phase's heat source and that of adjacent phases exceeding a correlation determination threshold, while the absolute value of the difference between the target phase's heat source correlation and that of adjacent phases is less than or equal to a preset small tolerance threshold, the actual temperature difference between the target phase and adjacent phases is calculated. When the absolute value of the actual temperature difference exceeds a preset interphase temperature difference threshold, the phase with the higher actual temperature is used as the source of degradation, and an alarm is output. When the absolute value of the actual temperature difference is less than or equal to the preset interphase temperature difference threshold, the target phase and adjacent phases are determined to be in a uniform degradation state, and a warning command for both the target phase and adjacent phases is simultaneously output. The interphase temperature difference threshold is set based on the nominal measurement accuracy of the deployed temperature sensor.

[0015] Secondly, the present invention provides a ring network box temperature monitoring system based on data fusion, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned ring network box temperature monitoring method based on data fusion is implemented.

[0016] By adopting the above technical solution, a computer program for a ring network box temperature monitoring method based on data fusion is generated and stored in a memory so that it can be loaded and executed by a processor. This allows for the creation of a terminal device based on the memory and processor, making it convenient to use.

[0017] The beneficial effects of this invention are as follows: By performing a cumulative calculation on the historical load current sequence with decay over time, this invention reconstructs the thermal contribution of historical electrical input to the current state, effectively compensating for the time lag effect of heat transfer in the insulating bushing, and eliminating erroneous monitoring results caused by the mismatch between surface temperature and bottom heating during severe load fluctuations; This invention utilizes equivalent heat capacity to convert abnormal temperature rises where the actual temperature exceeds the predicted temperature into redundant heat energy accumulated internally, thereby deriving the resistivity degradation variable reflecting the degree of internal conductor degradation. At the same time, it extracts the temporal fluctuation characteristics of the redundant heat sequence and the historical heating characteristic sequence of each phase, and by comparing the relative magnitude of the correlation between the target phase and the heat source of adjacent phases, it eliminates spatial thermal radiation interference emitted from adjacent phases in the narrow air chamber. Thus, under the condition of overcoming the dual obstacles of heat conduction lag and spatial physical crosstalk, it restores the true electrical heating law and realizes the identification and early warning of the true contact degradation state inside the ring network box. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for monitoring the temperature of a ring network box based on data fusion, as described in this invention. Detailed Implementation

[0019] 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, not all, of the embodiments of the present invention. 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.

[0020] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0021] This invention discloses a method for monitoring the temperature of a ring network enclosure based on data fusion, referring to... Figure 1 This includes steps S1-S4: S1: The historical load current sequence and actual temperature of each phase in the continuously collected ring network box are continuously written into the time-aligned buffer, and the ambient temperature at the current moment is obtained synchronously.

[0022] Specifically, a continuous first-in-first-out (FIFO) storage space is allocated within the monitoring terminal of the ring main unit as a time alignment buffer. Wide-range current transformers are independently deployed on each of the three-phase incoming and outgoing circuits of the ring main unit. At the same time, a patch-type digital temperature sensor is independently deployed on the outer surface of the insulating bushing of each phase, and an ambient temperature sensor is deployed at the air inlet at the bottom of the unit.

[0023] Obtain the insulating bushings of each phase in the ring main unit. The specific process is as follows: The basic thermal conduction hysteresis time is obtained by consulting the manufacturer's material property table of the insulating sleeve and used directly as... This parameter, set based on a material property table, utilizes objective material physical properties as the basis for subsequent attenuation calculations, providing an accurate physical transfer benchmark. In other embodiments, implementers can use actual heat conduction hysteresis measurement data during on-site power outages and cooling. Corrected settings.

[0024] Since the temperature change of the insulating medium is an extremely slow low-frequency physical process, the time step for data acquisition needs to be set appropriately. Get the time step The method is to obtain the minimum value of the thermal time constant of all phases as a global benchmark. Based on the low-frequency physical response characteristics of thermal conduction in insulating media, the time step is... Set as global benchmark of Time step The reason for setting this specific ratio is that it meets the high-fidelity capture requirements of the discrete sampling theorem for first-order inertial temperature rise curves. This avoids the accumulation of invalid data and storage overflow caused by blindly using extremely short sampling intervals, while reducing the system's computational load while ensuring the accuracy of heat change capture. In other embodiments, the implementer can adjust the time step according to the actual storage capacity of the hardware microprocessor. Adjust the settings.

[0025] By time step The real-time load current of each phase line is continuously collected, and then converted into the root mean square current of each phase line using the root mean square algorithm. Simultaneously, it independently records the actual temperature physically acquired by each phase's surface-mount digital temperature sensor, as well as the ambient temperature acquired by the ambient temperature sensor. The root mean square current of each phase of the line Actual temperature vs. ambient temperature Write data to the time-aligned cache sequentially.

[0026] Determine the cache depth of the time-aligned cache Specifically: the global benchmark Multiply by 5 to get the total duration, then divide the total duration by the time step. Get the cache depth of the corresponding phase The specific value varies depending on the time step in this embodiment. Set as global benchmark of ,therefore This invention will increase the cache depth. The time span of the data is set to cover 5 global benchmarks. The reason for the total duration is that, under the natural logarithmic decay law, after 5 constant periods, the contribution rate of the earliest load current to the heat at the current moment has decreased to less than 1%. This setting allows for the preservation of a complete historical heat generation data in a thermodynamic sense. In other embodiments, implementers can adjust the cache depth based on a comprehensive consideration of calculation accuracy and storage space on-site. Flexible expandable settings.

[0027] S2: The thermal time constant of the target phase insulating bushing is used to perform a cumulative calculation on the historical load current sequence of the target phase in the time alignment buffer area to obtain the predicted temperature of the target phase at the current moment, so as to compensate for the thermal conduction hysteresis of the target phase insulating bushing.

[0028] Specifically, the temperature of the outer surface of the insulating bushing is not determined by the instantaneous current, but is the cumulative result of the continuous heat conduction and heat dissipation of all the current heating in the insulating medium over a period of time. Based on the first law of classical thermodynamics, the thermal balance relationship of the insulating layer in the lumped parameter heat capacity model satisfies: the heating power of the conductor minus the heat dissipation power of the insulating layer to the environment equals the rate of change of the internal energy of the insulating layer.

[0029] Taking any one phase in each phase of the ring main unit as the target phase, according to Joule's law, the heating power of the conductor in the target phase satisfies the following relationship:

[0030] In the formula, Indicates the heating power of the target phase; This represents the root-mean-square current of the target phase in the continuous-time domain. Represents the initial healthy conductor resistance of the target phase. Heat generation power. Compared with the root mean square current of the line in the continuous time domain The square of and the initial healthy conductor resistance The magnitudes are positively correlated, and this relationship can be used to quantify the heat generation rate of the electrical source of the equipment under non-deterioration conditions.

[0031] According to Newton's law of cooling, the heat dissipation power from the insulation layer to the environment satisfies the following relationship:

[0032] In the formula, This indicates the heat dissipation power of the target phase; This represents the convective heat transfer coefficient of the outer surface of the target phase insulating sleeve; This indicates the effective heat dissipation area of ​​the outer surface of the target phase insulating sleeve; Indicates the temperature of the target phase insulation layer; This represents the ambient temperature. It should be noted that, based on the lumped parameter model, it is assumed that the internal thermal resistance of the insulating medium is extremely small and the temperature distribution is uniform; the target phase insulation layer temperature here... In an engineering physics sense, this is equivalent to the outer surface temperature of the target phase insulating sleeve. Heat dissipation power. Temperature of the insulation layer and ambient temperature The difference is positively correlated, and this relationship can be used to define the rate of heat loss to the external space.

[0033] The rate of change of internal energy of the target phase insulation layer satisfies the following relationship:

[0034] In the formula, Represents the rate of change of internal energy of the target phase; Represents the equivalent heat capacity of the target phase; This represents the rate of change of the temperature of the target phase insulation layer over time. (Internal energy change rate) Rate of change of temperature over time The magnitudes are positively correlated, and this relationship can be used to characterize the heat storage dynamics of the insulating medium itself.

[0035] Substituting the heat generation power of the target phase conductor, the heat dissipation power of the target phase insulation layer to the environment, and the rate of change of internal energy of the target phase insulation layer into the thermal equilibrium relationship, we obtain the basic expression of the first-order linear ordinary differential equation:

[0036] In the formula, This represents the root-mean-square current of the target phase in the continuous-time domain. Represents the initial healthy conductor resistance of the target phase; This represents the convective heat transfer coefficient of the outer surface of the target phase insulating sleeve; This indicates the effective heat dissipation area of ​​the outer surface of the target phase insulating sleeve; Indicates the temperature of the target phase insulation layer; Indicates ambient temperature; Represents the equivalent heat capacity of the target phase; This represents the rate of change of the target phase insulation layer temperature over time. Among these, the heating term and the root-mean-square current of the line in the continuous-time domain are also included. The square of the terms is positively correlated with the magnitude of the heat dissipation term and the temperature difference, and the rate of change of temperature over time is also positively correlated. The magnitude of the equation depends on the net difference between the heating term and the heat dissipation term. This equation allows us to establish the underlying physical mapping between the conductor current heating and the dynamic evolution of the insulation layer temperature.

[0037] Temperature rise Due to ambient temperature If considered constant over a short period, then the rate of change of temperature with time Equal to the rate of change of temperature over time , will raise the temperature Substitute the above fundamental expression for the first-order linear ordinary differential equation and divide by the term. Simplifying, we obtain the expression for the differential equation containing the time constant:

[0038] In the formula, Indicates temperature rise; Represents the equivalent heat capacity of the target phase; This represents the convective heat transfer coefficient of the outer surface of the target phase insulating sleeve; This indicates the effective heat dissipation area of ​​the outer surface of the target phase insulating sleeve; This represents the rate of change of temperature rise over time; Represents the initial healthy conductor resistance of the target phase; This represents the root-mean-square current of the target phase in the continuous time domain. This equation simplifies complex multi-element heat transfer networks into a standard first-order inertial element form that is easier to solve mathematically.

[0039] definition and the basic heat conversion coefficient Under the boundary condition of an initial temperature rise of 0, the differential equation containing a time constant is solved using the integral factor method to obtain the analytical solution function of the temperature rise in the continuous time domain:

[0040] In the formula, Indicates the target phase in time The continuous temperature rise value at the location; This represents the thermal time constant of the target phase insulating bushing; Indicates the variable that is integrated over historical time. The square of the root mean square current of the target phase in the continuous time domain; Indicates the heat conversion coefficient of the target phase; Represents the natural constant; Indicates the current time; This represents the time integral differential term. Among them, the continuous temperature rise value... The magnitude of the square value of the root mean square current of the line in the continuous time domain at each historical moment is positively correlated with the magnitude of the physical contribution, and its physical contribution decreases exponentially with the increase of the time difference. The analytical solution function of temperature rise can realize the continuous integral presentation of historical electrical input with respect to the current thermal state.

[0041] To adapt to the discrete sampling characteristics of digital monitoring terminals, the analytical solution function for temperature rise in the continuous time domain is discretized at equal intervals along the time axis. Combined with the configuration of a time-aligned buffer, the current time... Integral variables over historical time The physical difference is replaced by the reduced step product. The integration operation is then transformed into an operation for a finite cache depth. Discrete summation and accumulation; the discretization process makes the time integral differential term... Convert to discrete time steps After integration and demodulation, the predicted temperature of the outer surface of the insulating sleeve at the current moment is finally derived. Satisfying the expression:

[0042] In the formula, This indicates the predicted temperature of the outer surface of the target phase insulating bushing at the current moment; This indicates the ambient temperature measured at the current moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates counting backwards from the current time. The root mean square current of the target phase at each time step; Indicates the heat conversion coefficient of the target phase; Represents the natural constant; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing. Among them, the predicted temperature... With the root mean square current of the line The magnitudes are positively correlated, and the root mean square current of the line is... The larger the value, the more Joule heat is generated, leading to a higher predicted temperature. The higher the temperature; and The magnitudes are negatively correlated. The larger the diameter, the greater the internal heat dissipation resistance of the insulating sleeve and the slower the heat transfer, resulting in a higher predicted temperature. The lower the value. This invention introduces a proportional term. It can restore the physical integral accumulation effect in discrete states and compensate for the time delay effect of heat conduction at the algorithm level.

[0043] When the device is initially powered on and in a healthy state, the system calculates the root mean square current of the target phase in the buffer by time alignment. The sliding variance is used to lock the initial calibration interval: the duration for which the sliding variance is continuously less than or equal to a preset steady-state variance threshold is greater than or equal to the cache depth. When the time span of the data is considered, it is determined that the power supply branch has entered a quasi-steady-state range where the load fluctuation is within the allowable error range of the sensor. The steady-state variance threshold is set as follows: when the ring main unit is unloaded, the background noise variance of the target phase current transformer output signal is extracted, multiplied by a preset reliability factor, and used as the steady-state variance threshold. This invention sets the steady-state variance threshold based on the unloaded background noise variance, which can effectively eliminate the interference caused by the inherent measurement fluctuations of the sensor itself on the steady-state determination, ensuring that the system identifies the true stable physical load state. In other embodiments, implementers can directly read the static noise parameters according to the accuracy grade specification of the current sensor to set the steady-state variance threshold. In this embodiment, based on the Raida criterion in statistics, the reliability factor is set to 3, which can ensure the system's sensitivity to capturing true load fluctuations while preventing the steady-state calibration range from being erroneously interrupted due to occasional inherent measurement noise spikes.

[0044] Within the quasi-steady-state region, the heating power of the conductor inside the target phase and the convective heat dissipation power of the insulating bushing surface of the target phase reach a rigid physical equilibrium. The current amplitude and actual surface temperature at this point are extracted, and combined with the target phase insulating bushing obtained in the preceding steps... By substituting the steady-state boundary conditions into the heat transfer differential equation and performing inverse arithmetic solutions, the heat conversion coefficient of the target phase reflecting the health baseline can be spontaneously calibrated. .

[0045] During the transition period after the device is initially powered on and before the first quasi-steady-state interval capture is completed, the system uses a factory-preset conservative value of the same model as a transition parameter to maintain the baseline temperature estimation. This conservative value is derived by inversely calculating the rated current and rated temperature rise from the device's nameplate using Newton's law of cooling steady-state formula. The transition period begins when the device is first powered on and continues until the system first captures the time period that meets the quasi-steady-state interval determination condition. The quasi-steady-state interval determination condition is the root mean square current of the target phase within the time alignment buffer. The sliding variance, the duration for which it is continuously less than or equal to a preset steady-state variance threshold is greater than or equal to the cache depth. The time span of the data is as follows: After the initial calibration is completed, the parameters are seamlessly overwritten and stored as the reference thermal conversion coefficient for the current equipment lifecycle; when the equipment reaches the regular maintenance cycle, the quasi-steady-state range is redefined and the reference thermal conversion coefficient is calibrated to update the reference thermal conversion coefficient.

[0046] S3: The portion of the target phase temperature that exceeds the predicted temperature is identified as an abnormal temperature rise. The abnormal temperature rise is converted into redundant heat energy accumulated internally using the equivalent heat capacity. Combined with the historical load current sequence of the target phase in the time-aligned buffer, the resistance variation of the conductor inside the target phase is derived in reverse.

[0047] This invention uses historical current sequences to establish a predicted temperature of the target phase at the current moment. Because of the predicted temperature The time lag effect of thermal conduction in the insulation layer has been compensated for, ensuring that historical heat generation data and current temperature data are compared on the same timeline. The actual temperature of the target phase at the current moment is then used as the benchmark. With predicted temperature By performing difference calculations and extracting the difference in heat, the change in contact resistance is calculated in reverse. Specifically: Based on the thermodynamic law of conservation of energy, the difference between the actual temperature and the predicted temperature is the abnormal temperature rise. The redundant heat energy corresponding to the abnormal temperature rise accumulating inside the insulation layer satisfies the following relationship:

[0048] In the formula, This represents redundant heat accumulated inside the target phase; Represents the equivalent heat capacity of the target phase; Indicates the actual temperature of the target phase; This represents the predicted temperature of the target phase. This relationship allows us to convert observable temperature surface deviations into specific redundant thermal energy indices.

[0049] At the same time, the contact resistance increases due to oxidation and deterioration at the contact points of the conductors inside the target phase; this increased resistance is known as the resistance degradation. The total residual Joule heat generated during the historical buffer cycle, after dissipating through the insulation layer, satisfies the discrete energy integral relationship:

[0050] In the formula, This represents the total residual Joule heat of the target phase; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates going back 1 / 2 from the current time. The root mean square current of the target phase line at each time step; The resistivity variation of the target phase; Represents the natural constant; Indicates the time step; This represents the thermal time constant of the target phase insulating bushing. This relationship allows for the quantification of the total dissipated energy accumulated over historical time by abnormal resistance.

[0051] Because the physical oxidation damage to the internal conductors of the target phase is a slow, gradual process within the time span corresponding to the time alignment buffer, the physical damage morphology of the internal conductors is basically stable, and the resistance is significantly reduced. The value of can be considered a constant term throughout the entire accumulation period, thus allowing for the reduction of resistance variation. Extracted to the outside of the discrete accumulation symbol. According to the principle of thermal equilibrium, the total residual Joule heat generated by the deterioration of resistance... This must equal the redundant heat accumulated inside the insulating component. That is, satisfying the equation Combining the two equations, we obtain the thermoelectric energy balance mapping equation:

[0052] In the formula, The resistivity variation of the target phase; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates going back 1 / 2 from the current time. The root mean square current of the target phase line at each time step; Represents the natural constant; Indicates the time step; This represents the thermal time constant of the target phase insulating bushing; Represents the equivalent heat capacity of the target phase; Indicates the actual temperature of the target phase; This represents the predicted temperature of the target phase. The terms on the left-hand side of the equation are related to the resistivity variation. The magnitudes are positively correlated, and the terms on the right side of the equation are positively correlated with the magnitude of the temperature difference. This equation establishes an energy conservation equation between macroscopic heat accumulation and the work done by microscopic resistance degradation.

[0053] In performing resistance inferior variable Before the stripping calculation, the system calculates the heat integral term in real time. If the integral term for heat generation is equal to 0, the ring main unit is determined to be in a long-term no-load or power outage state. The system automatically bypasses the current calculation cycle and suspends the warning logic to avoid an overflow error where the denominator of the microprocessor is 0. After determining that the integral term for heat generation is greater than 0, the above thermoelectric energy balance mapping equation is rearranged to finally obtain the resistance variable. The expression is:

[0054] In the formula, The resistance degradation variable of the target phase reflects the difference in resistance degradation between the current actual internal contact resistance and the initial healthy state. Indicates the actual temperature of the target phase; Indicates the predicted temperature of the target phase; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates going back 1 / 2 from the current time. The root mean square current of the target phase line at each time step; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing; Represents the natural constant; This represents the equivalent heat capacity of the target phase. Among these, the resistivity is a significant factor. Compared with actual temperature and predicted temperature The difference in magnitude is positively correlated with the actual temperature. Temperatures exceeding predictions The more [amount of heat], the more severe the abnormal heat dissipation inside, leading to a decrease in the calculated resistance. The larger the resistance, the worse the variable. With equivalent heat capacity The magnitudes are positively correlated with the equivalent heat capacity. The larger the value, the more redundant heat the insulation absorbs when the temperature rises by the same amount, and the greater the corresponding resistance variation of the internal heat source. The larger the denominator. Through this operation, the denominator... It possesses the dimension of current squared times time, combined with the equivalent heat capacity in the molecule. The product of the product and the temperature difference, represented by the joule heat dimension, results in a significant decrease in the final calculated resistivity. Its dimension is ohms, which can successfully convert the apparent temperature, which is subject to severe interference from both the environment and the load, into an increase in the internal resistance in ohms, thereby improving the accuracy of identifying the underlying health status.

[0055] Obtain the equivalent heat capacity of the target phase The specific method is as follows: the specific heat capacity constant of epoxy resin in the equipment material parameters is multiplied by its material density, and then multiplied by the actual encapsulation volume of the insulating sleeve of the target phase at that location. This value is then pre-fixed in the internal read-only memory of the monitoring terminal. The reason for setting this physical parameter as a constant value is that the heat consumption rate during the damping transfer process can be accurately quantified using the objective physical material, density, and geometric volume of the insulating component, ensuring the rigor of the reverse calculation of redundant heat into the resistance ohm value. In other embodiments, the implementer can calculate the equivalent heat capacity based on the degree of material aging and shedding of the insulating sleeve after years of operation. Regular updates and dynamic adjustments to the settings.

[0056] It should be noted that if the large current generated in the past does not currently affect the actual temperature... Temperatures exceeding predictions That is, the actual temperature With predicted temperature The difference between When the resistance is less than or equal to 0, the system determines that the resistance is inferior. A value of 0 indicates that the heat is in the normal heat storage and damping process of the insulation layer; if the actual temperature is 0... With predicted temperature The difference between If the value is greater than 0, it proves that the internal structure is emitting redundant abnormal heat beyond what the historical current should produce. This redundant abnormal heat is transmitted through the aforementioned resistance variation. The expression is converted back to the increase in the physical resistance in ohms.

[0057] S4: Based on the temporal fluctuation correlation between the redundant heat sequence and the historical heat generation characteristic sequence of each phase, calculate the heat source correlation degree of the target phase and the heat source correlation degree of the adjacent phase respectively. Based on the resistance degradation variable, the heat source correlation degree of the target phase and the heat source correlation degree of the adjacent phase, determine the contact degradation state of the target phase.

[0058] It should be noted that due to the extremely confined space inside the ring main unit and the extremely compact physical arrangement of the three-phase cable joints, when severe degradation of adjacent phases generates extreme high temperatures, the intense spatial thermal radiation and thermal convection within the enclosed air chamber will force the actual temperature of the target phase, which is in a normal state, to rise abnormally. This spatial heat transferred from adjacent phases to the target phase will cause the calculated resistance degradation of the target phase to show a false over-limit phenomenon. However, based on the objective physical characteristics of electrothermal conversion, the waveform fluctuation evolution of redundant heat generated by the degradation within the target phase itself on the time axis will be highly synchronized with its own historical load heating accumulation process; and the waveform fluctuation of spatial heat radiated from adjacent phases will be highly consistent with the historical load heating accumulation process of adjacent phases. Based on this principle of thermoelectric waveform homogeneity and synchronization, this invention dynamically traces the true physical source of the redundant heat accumulated internally to eliminate spatial thermal interference.

[0059] Set the length of the association window for timing feature tracking of the target phase: ,in, The length of the associated window for tracking the temporal features of the target phase. This represents the thermal time constant of the target phase insulating bushing. Indicates the time step. This indicates the rounding up sign. A short correlation window cannot reflect the waveform fluctuations over time, while a length covering two thermal time constant periods can encompass the main transient response processes of the thermodynamic system to step load changes, thus ensuring sufficient time-series feature comparison. Therefore, this invention uses an expression... Length of the associated window The length is set to cover two thermal time constant periods. In other embodiments, the implementer can determine the length of the correlation window based on the actual load fluctuation frequency of the grid node. The proportional scaling setting.

[0060] Within the correlation window of the time-series feature tracking of the target phase, the redundant heat accumulated inside the target phase at each sampling time is extracted. Construct a redundant calorimetric sequence and simultaneously extract the calorimetric integral term of the target phase at the corresponding time. This constitutes the intrinsic thermal sequence of the target phase, in which, Indicates the target phase in the first... The excess heat accumulated internally at a given moment Indicates equivalent heat capacity, Indicates the target phase in the first... The actual temperature at that moment. Indicates the target phase in the first... The predicted temperature at that moment. Indicates the target phase in the first... The intrinsic heating sequence value of the target phase at time n, that is, the target phase at time n. The heat integral term at each moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; Indicates the target phase in the first... Counting back from the first moment The line root mean square current at each time step. Represents the natural constant; Indicates the time step; This represents the thermal time constant of the insulating bushing.

[0061] Similarly, within the associated window of the time-series feature tracking of the target phase, the nearest neighboring phase in physical space is obtained. If there are multiple nearest neighboring phases with equal distances, the heat integral term of each nearest neighboring phase at the corresponding time is obtained, and the corresponding neighboring phase intrusion heat sequence is formed.

[0062] The numerical variance of the redundant heat series and the numerical variance of the intrinsic heat series of the target phase are calculated. If either the numerical variance of the redundant heat series or the numerical variance of the intrinsic heat series of the target phase is zero, it indicates that the physical quantity is in an absolutely stationary state during that time period. The system automatically correlates the heat sources of the target phase. Assigning a value of 0 avoids triggering a division-by-zero crash error at the processor level; conversely, in response to the numerical variance of the redundant heat sequence being greater than 0 and the numerical variance of the intrinsic heat sequence of the target phase being greater than 0, the heat source correlation degree of the target phase is calculated:

[0063] In the formula, Indicates the correlation degree of the target phase heat source; Indicates the length of the associated window for tracking the temporal features of the target phase; Indicates the time sliding sequence number within the associated window; Indicates the target phase in the first... The excess heat accumulated internally at a given moment; Represents the arithmetic mean of redundant heat sequences; Indicates the target phase in the first... The intrinsic heating sequence values ​​of the target phase at each moment; This represents the arithmetic mean of the intrinsic heating sequence of the target phase. The closer the time axes of the transient peaks and troughs of the redundant heating sequence and the intrinsic heating sequence of the target phase match, the more the redundant heating depends on the work done by the target phase's own current. The calculated heat source correlation of the target phase is... The closer the value is to 1, the less independent the two are, indicating a low correlation between the target and the heat source. Approaching 0 or a negative value.

[0064] Similarly, using the same computational structure as the target phase heat source correlation expression, the intrinsic heat generation sequence of the target phase is replaced with the adjacent phase intrusion heat generation sequence, thereby calculating the adjacent phase heat source correlation between the redundant heat generation sequence and the adjacent phase intrusion heat generation sequence. .

[0065] The maximum allowable contact resistance margin from the equipment's factory parameters is extracted as the upper tolerance limit, and a tolerance time window for observing state stability is set. This tolerance time window contains multiple consecutive time steps, and the specific method for setting the number of time steps is as follows: obtain the thermal time constant of the target phase. The thermal time constant Multiply by a preset multiplier. Divide by the time step The time is then rounded up to determine the number of discrete time steps covered by the tolerance time window, with a preset multiplier factor. A tolerance time window greater than or equal to 1 ensures that the time span is sufficient to smoothly filter out occasional transient interference peaks in the power grid, while also preventing delays in alarm response to actual contact degradation faults due to excessively long tolerance time windows. In practical engineering applications, to ensure effective filtering of transient overshoots caused by thermal inertia, a preset multiplier is used. The preferred setting is 1.5 to 2, so that the time span of the tolerance time window covers 1.5 to 2 thermal time constant periods.

[0066] The resistivity degradation of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the calculated heat source correlation of the target phase is also considered. Greater than the correlation of adjacent phase heat sources And the correlation of the target phase heat source The correlation is greater than the preset correlation threshold, and the correlation between adjacent phase heat sources is also greater. If the value is less than or equal to the preset correlation judgment threshold, it proves that the excess redundant heat is generated by the load current flowing through the target phase itself. The system determines that the target phase has experienced real contact oxidation and loosening, and outputs an alarm command to the power distribution dispatch master station indicating that the target phase has a serious fault. If the resistance variation of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the heat source correlation between the target phase and any nearest neighboring phase is greater than the heat source correlation of the target phase, and the heat source correlation of the neighboring phase is greater than the preset correlation judgment threshold, while the heat source correlation of the target phase is less than or equal to the correlation judgment threshold, it is determined that the temperature rise of the target phase is a heating phenomenon induced by the high temperature radiation of the neighboring phase, and the alarm command for the target phase is discarded. Furthermore, considering that under normal three-phase balanced power grid operation conditions, the load current fluctuation curves of each phase are often highly synchronized, if the resistivity degradation of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the calculated heat source correlation of the target phase is also considered... Correlation with adjacent phase heat sources All are greater than the preset correlation judgment threshold, and the correlation degree of the target phase heat source is greater than the preset correlation judgment threshold. Correlation with adjacent phase heat sources If the absolute value of the temperature difference is less than or equal to a preset small tolerance threshold, it indicates that the three-phase load fluctuations are highly synchronized. At this time, the system extracts the actual temperature of the target phase and compares it directly with the actual temperature of the adjacent phase, calculates the actual temperature difference between the target phase and the adjacent phase, and when the absolute value of the actual temperature difference is greater than the preset interphase temperature difference threshold, the phase with the higher actual temperature is used as the source of degradation and an alarm is output; when the absolute value of the actual temperature difference is less than or equal to the preset interphase temperature difference threshold, it is determined that the target phase and the adjacent phase are in a uniform degradation state, and warning commands for the target phase and the adjacent phase are output simultaneously. The interphase temperature difference threshold is set according to the nominal measurement accuracy of the deployed temperature sensor, thereby fully ensuring the reliability of identification under extreme three-phase balance conditions.

[0067] In this embodiment, the correlation determination threshold is set to 0.75 in the monitoring system's software configuration file because 0.75 statistically represents a strong linear correlation boundary, sufficient to filter out occasional electromagnetic interference. In other embodiments, the implementer can fine-tune the correlation determination threshold according to the tolerance for false alarm rate. In this embodiment, the small tolerance threshold is set to 0.05 in the monitoring system's software configuration file because this value can effectively define the reasonable calculation error range when the three-phase load currents are highly synchronized. In other embodiments, the implementer can adjust the small tolerance threshold according to the sampling accuracy of the current transformer.

[0068] This invention also discloses a ring network box temperature monitoring system based on data fusion, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a ring network box temperature monitoring method based on data fusion according to the present invention.

[0069] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0070] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise expressly and specifically defined.

[0071] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A ring main unit temperature monitoring method based on data fusion, characterized in that, include: The historical load current sequence and actual temperature of each phase in the continuously collected ring network box are written into the time-aligned buffer, and the ambient temperature at the current moment is obtained. Using the thermal time constant of the target phase insulating bushing, the historical load current sequence of the target phase in the time-aligned buffer is accumulated and calculated to decay over time, so as to reconstruct the cumulative contribution of historical electrical input to the current thermal state on the time axis, and combine it with the ambient temperature to obtain the predicted temperature of the target phase at the current moment, so as to compensate for the thermal conduction hysteresis of the insulating bushing. The abnormal temperature rise of the target phase is calculated when the actual temperature is higher than the predicted temperature. Based on the principle of energy conservation, the abnormal temperature rise is converted into redundant heat energy accumulated inside by using the equivalent heat capacity. Combined with the historical load current sequence in the time-aligned buffer, the resistance degradation variable corresponding to the internal conductor resistance degradation of the target phase is derived in reverse. Within the association window, construct the redundant heat sequence of the target phase, the intrinsic heat sequence of the target phase, and the intrusive heat sequence of the adjacent phase; calculate the heat source correlation between the redundant heat sequence and the intrinsic heat sequence of the target phase, and the heat source correlation between the redundant heat sequence and each intrusive heat sequence of the adjacent phase. The contact degradation state of the target phase is determined based on the resistivity of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases.

2. The ring main unit temperature monitoring method based on data fusion according to claim 1, characterized in that, The predicted temperature satisfies the expression: ； In the formula, This indicates the predicted temperature of the outer surface of the target phase insulating bushing at the current moment; This indicates the ambient temperature measured at the current moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates counting backwards from the current time. The root mean square current of the target phase at each time step; Indicates the heat conversion coefficient of the target phase; Represents the natural constant; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing.

3. The method for monitoring the temperature of a ring network enclosure based on data fusion according to claim 1, characterized in that, The reverse derivation yields the resistance degradation due to the deterioration of the conductor resistance inside the target phase, including: When the difference between the actual and predicted temperatures of the target phase is greater than 0, the resistivity inferior variable satisfies the expression: ； In the formula, The resistivity variation of the target phase; Indicates the actual temperature of the target phase; Indicates the predicted temperature of the target phase; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; This indicates going back 1 / 2 from the current time. The root mean square current of the target phase line at each time step; Indicates the time step between two consecutive data collections; This represents the thermal time constant of the target phase insulating bushing; Represents the natural constant; This represents the equivalent heat capacity of the target phase.

4. The method for monitoring the temperature of a ring network box based on data fusion according to claim 3, characterized in that, Also includes: If the difference between the actual temperature and the predicted temperature of the target phase is less than or equal to 0, the resistivity of the target phase is determined to be 0.

5. The method for monitoring the temperature of a ring network box based on data fusion according to claim 1, characterized in that, The construction of the redundant thermal sequence of the target phase, the intrinsic thermal sequence of the target phase, and the intrusive thermal sequence of the adjacent phase includes: Extract redundant heat accumulated inside the target phase at each sampling time. This constitutes a redundant heat sequence; Extract the heat integral term of the target phase at the corresponding time. This constitutes the intrinsic heating sequence of the target phase; the nearest neighboring phase in physical space is obtained. If there are multiple nearest neighboring phases with equal distances, the heating integral term of each nearest neighboring phase at the corresponding time is obtained, and the corresponding adjacent phase intrusion heating sequence is constituted. in, Indicates the target phase in the first... The excess heat accumulated internally at a given moment Indicates equivalent heat capacity, Indicates the target phase in the first... The actual temperature at that moment. Indicates the target phase in the first... The predicted temperature at that moment. Indicates the target phase in the first... The heat integral term at each moment; Indicates the cache depth of the time-aligned cache; Indicates the sequence number of the backward time step; Indicates the target phase in the first... Counting back from the first moment The line root mean square current at each time step. Represents the natural constant; Indicates the time step; This represents the thermal time constant of the insulating bushing.

6. The method for monitoring the temperature of a ring network box based on data fusion according to claim 1, characterized in that, The correlation degree of the target phase heat source satisfies the expression: ； In the formula, Indicates the correlation degree of the target phase heat source; Indicates the length of the associated window; Indicates the time sliding sequence number within the associated window; Indicates the target phase in the first... The excess heat accumulated internally at a given moment; Represents the arithmetic mean of the redundant heat sequence; Indicates the target phase in the first... The intrinsic heating sequence values ​​of the target phase at each moment; This represents the arithmetic mean of the intrinsic heating sequence of the target phase.

7. The method for monitoring the temperature of a ring network box based on data fusion according to claim 1, characterized in that, The determination of the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases includes: If the resistance of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the heat source correlation of the target phase is greater than the heat source correlation of the adjacent phase, and the heat source correlation of the target phase is greater than the preset correlation judgment threshold, while the heat source correlation of the adjacent phase is less than or equal to the correlation judgment threshold, it is determined that the target phase has experienced contact oxidation and loosening, and an alarm command indicating a fault is output.

8. The method for monitoring the temperature of a ring network box based on data fusion according to claim 1, characterized in that, The step of determining the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases further includes: If the resistance variation of the target phase is greater than the preset tolerance limit within the preset tolerance time window, and the heat source correlation between the target phase and any nearest adjacent phase is greater than the heat source correlation of the target phase, and the heat source correlation of the adjacent phase is greater than the preset correlation judgment threshold, while the heat source correlation of the target phase is less than or equal to the correlation judgment threshold, it is determined that the temperature rise of the target phase is a heating phenomenon induced by the high temperature radiation of the adjacent phase, and the alarm command for the target phase is discarded.

9. The method for monitoring the temperature of a ring network box based on data fusion according to claim 1, characterized in that, The step of determining the contact degradation state of the target phase based on the resistivity degradation of the target phase, the heat source correlation of the target phase, and the heat source correlation of adjacent phases further includes: In response to the target phase's resistance degradation exceeding a preset tolerance limit within a preset tolerance time window, and the correlation between the target phase's heat source and that of adjacent phases exceeding a correlation determination threshold, while the absolute value of the difference between the target phase's heat source correlation and that of adjacent phases is less than or equal to a preset small tolerance threshold, the actual temperature difference between the target phase and adjacent phases is calculated. When the absolute value of the actual temperature difference exceeds a preset interphase temperature difference threshold, the phase with the higher actual temperature is used as the source of degradation, and an alarm is output. When the absolute value of the actual temperature difference is less than or equal to the preset interphase temperature difference threshold, the target phase and adjacent phases are determined to be in a uniform degradation state, and a warning command for both the target phase and adjacent phases is simultaneously output. The interphase temperature difference threshold is set based on the nominal measurement accuracy of the deployed temperature sensor.

10. A ring network box temperature monitoring system based on data fusion, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a ring network box temperature monitoring method based on data fusion according to any one of claims 1-9.

Images