Power distribution cabinet comprehensive management method and system based on multiple protection and energy efficiency optimization

Abstract

This invention belongs to the field of power distribution cabinet technology and provides a comprehensive management method and system for power distribution cabinets based on multiple protections and energy efficiency optimization. In terms of safety status, this invention constructs a multi-layered progressive collaborative architecture including structural safety protection, electrical safety protection, fire safety protection, environmental safety protection, and intrusion safety protection layers. Regarding energy efficiency status, it selects electrical components at the source, performs dynamic power quality management to eliminate additional power losses during the process, and dynamically adjusts strategies at the control stage. Then, it constructs a unified state vector including safety status, energy efficiency status, module topology status, and health status. Based on the unified state vector, it determines the safety risk margin, loss gradient, and lifetime loss rate, and performs collaborative control with safety constraints as the priority condition. This achieves collaborative control that considers safety protection, energy efficiency optimization, modular design, and operation and maintenance management.

Description

Technical Field

[0001] This invention belongs to the field of power distribution cabinet technology, and in particular relates to a comprehensive management method and system for power distribution cabinets based on multiple protections and energy efficiency optimization. Background Technology

[0002] As a key node device in the power system, the AC low-voltage integrated switchgear undertakes the core functions of power distribution, control and protection. With the continuous expansion of the power system scale and the continuous improvement of power supply quality requirements, the AC low-voltage integrated switchgear is becoming increasingly important.

[0003] With the continuous expansion of power system scale and the increasing demands for power quality, traditional AC low-voltage integrated switchgear has revealed numerous shortcomings in terms of safety, energy efficiency, and maintainability. It fails to achieve coordinated control that considers safety protection, energy efficiency optimization, modular design, and operation and maintenance management. Specifically, safety protection lacks systematic integration and collaborative mechanisms, resulting in the overall protective effect of various protective measures failing to meet expectations. Energy efficiency optimization methods are scattered and unsystematic, lacking a comprehensive technical solution encompassing the selection of high-efficiency components, power quality management, and optimized control. This hinders coordinated control that considers safety protection, energy efficiency optimization, modular design, and operation and maintenance management, impacting the safety, energy efficiency, and maintainability of AC low-voltage integrated switchgear under complex operating conditions. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a comprehensive management method and system for distribution cabinets based on multiple protections and energy efficiency optimization. Regarding safety, this invention constructs a multi-layered, progressive collaborative architecture comprising structural safety protection, electrical safety protection, fire safety protection, environmental safety protection, and intrusion safety protection layers. Regarding energy efficiency, it selects electrical components at the source, performs dynamic power quality management to eliminate additional power losses during operation, and dynamically adjusts strategies at the control stage. This solves the problems of a lack of systematic integration and collaborative mechanisms in safety protection and fragmented and unsystematic energy efficiency optimization methods. This paper addresses the problem by providing the prerequisites for collaborative control that considers safety protection, energy efficiency optimization, modular design, and operation and maintenance management. Then, a unified state vector is constructed, including safety state, energy efficiency state, module topology state, and health state. Based on this unified state vector, safety risk margin, loss gradient, and lifetime loss rate are determined. Based on these parameters, collaborative control is implemented with safety constraints as the priority condition. This achieves collaborative control that considers safety protection, energy efficiency optimization, modular design, and operation and maintenance management, ensuring the safety, energy efficiency, and maintainability of the AC low-voltage integrated switchgear under complex operating conditions.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: In a first aspect, the present invention provides a comprehensive management method for power distribution cabinets based on multiple protections and energy efficiency optimization, comprising: Acquire the safety status, energy efficiency status, module topology status, and health status of the AC low-voltage integrated switchgear; construct a unified state vector including safety status, energy efficiency status, module topology status, and health status; in terms of safety status, construct a multi-layered progressive collaborative architecture including structural safety protection layer, electrical safety protection layer, fire safety protection layer, environmental safety protection layer, and intrusion safety protection layer; in terms of energy efficiency status, select electrical components at the source, perform dynamic power quality management to eliminate additional power losses in the middle stage, and dynamically adjust strategies in the control stage; The safety risk margin, loss gradient, and lifetime loss rate are determined based on the unified state vector. Based on safety risk margin, loss gradient, and lifetime loss rate, coordinated control is carried out with safety constraints as the priority.

[0006] Furthermore, the unified state vector is: ; in, Indicates safety features related to temperature, smoke, access control, vibration, insulation, and leakage current; It represents the energy efficiency characteristics of active power, reactive power, power factor, harmonics, voltage deviation, and line loss; This indicates the module type, rated parameters, interface status, and communication address characteristics. It represents contact resistance, insulation resistance, number of switching operations, hot spot temperature, partial discharge count, and health life characteristics.

[0007] Furthermore, the objective function and security constraints of the cooperative control are as follows: ; in, The control variables include static var generator output, active power filter compensation current, capacitor switching, temperature control setpoint, load derating, and maintenance actions. , and These are energy consumption targets, lifespan loss targets, and operating cost targets; The security risk value after fusion. To allow for a risk ceiling.

[0008] Furthermore, the security risk value after fusion is: ; Among them, N, P, A, and E represent normal, early warning, alarm, and emergency states, respectively; , , , These are the scalarized coefficients for each risk state.

[0009] Furthermore, the security risk assessment will output fuzzy reasoning. With Bayesian network output Perform DS evidence fusion; assuming the identification framework is normal (N), warning (P), alarm (A), and emergency (E), the fusion quality function is: ; First, determine the overall cost of each candidate action, then select the action with the lowest overall cost: (34) in, Indicates the total cost of the action. Indicates the cost of the action. Indicates taking an action Penalties for exceeding limits afterward This indicates the additional penalty imposed by the action.

[0010] Furthermore, the structural safety protection layer includes: a double-sealed structure between the cabinet door and the cabinet body; a multi-layer sealed cable connector for the cable inlet; a labyrinth-style waterproof ventilation structure for the ventilation and heat dissipation vents; vibration-damping rubber supports installed at the bottom of the cabinet body; and the determination of the thickness of the outer shell includes: Determine the bending stiffness of cabinet doors or side panels : ; For plate regions supported on four sides or with stiffened ribs at the boundary, the maximum equivalent stress and maximum deflection According to boundary coefficient and Sure: ; Strength constraints and stiffness constraints The lower limit of plate thickness is obtained: ; equivalent width of impact The design verification of local plastic energy dissipation and energy absorption capacity was carried out. and the lower limit of impact constraint plate thickness for: ; For coastal salt spray, chemical dust, or high humidity environments, by superimposing the full-life corrosion thinning, negative tolerance of the sheet metal, and welding heat deformation allowance, the lower limit of corrosion and manufacturing constraints is obtained: ; Select the minimum thickness that satisfies all constraints from the standard plate thickness sequence: ; in, For thickness The bending stiffness of the lower part; and These are the long and short sides of the equivalent plate region, respectively; ν is the elastic modulus; ν is Poisson's ratio; This is the equivalent surface load; , These are the boundary factors for stress and deflection, respectively. Yield strength; and These are allowable stress and allowable deflection, respectively. The equivalent width for impact effect; To design impact energy; The effective thickness that needs to be retained even at the end of its service life; Annual corrosion thinning rate; Design life; and These are the negative tolerance of the sheet metal and the welding heat deformation allowance, respectively. To meet the lower limit of the process for installing sealing flanges, hinges, and locking components; For procureable standard plate thickness series; To finalize the selection of the plate thickness.

[0011] Furthermore, when the strategy is dynamically adjusted, the objective function is composed of energy consumption, power quality, lifespan loss, and operating cost: ; in, For energy consumption items, For power quality, This is a lifespan depletion item. The cost is the operating cost item; the optimization constraints include harmonic constraint indicators and lifetime loss rate constraint indicators.

[0012] Furthermore, regarding the module topology status, a compatibility score is performed: ; When compatibility rating Power is prohibited when the current is below the preset threshold; otherwise, protection settings are generated based on the module's rated current, backplane bus capacity, and contact temperature rated curve. ; The overload protection inverse time curve is as follows: ; in, Rate the module compatibility; For the first Class compatibility input; This is the individual score after mapping the input; For single-item weights; This is the module's rated current. This refers to the capacity of the backplane busbar; For temperature Lower contact allowable current; , and This is the reduction factor; To protect the set value; For current The tripping time is specified below; A, B, and p are parameters of the inverse time curve.

[0013] Furthermore, the prediction of the health status includes: acquiring current, voltage, hotspot temperature, humidity, total harmonic distortion of current, contact resistance, insulation resistance, number of switching actions, vibration intensity, and partial discharge count to construct an input vector; robustly normalizing outliers using median absolute deviation; and determining the hybrid remaining lifetime. ; ; ; ; in, This is insulation damage; This is the failure threshold; , , and , representing the prediction error and uncertainty of the physical channel and the data channel, respectively; c is the context vector; , For weighting; constant; The bias term for the data-driven remaining lifetime output layer; Transpose of the data-driven remaining lifetime output layer weight vector.

[0014] Secondly, the present invention also provides a comprehensive management system for power distribution cabinets based on multiple protections and energy efficiency optimization, comprising: The data processing module is configured to: acquire the safety status, energy efficiency status, module topology status, and health status of the AC low-voltage integrated distribution cabinet; construct a unified state vector including safety status, energy efficiency status, module topology status, and health status; specifically, regarding safety status, construct a multi-layered progressive collaborative architecture including structural safety protection layer, electrical safety protection layer, fire safety protection layer, environmental safety protection layer, and intrusion safety protection layer; regarding energy efficiency status, select electrical components at the source, perform dynamic power quality management to eliminate additional power losses in the intermediate stage, and dynamically adjust strategies in the control stage. The constraint determination module is configured to determine the safety risk margin, loss gradient, and lifetime loss rate based on the unified state vector. The collaborative control module is configured to perform collaborative control based on safety risk margin, loss gradient, and lifetime loss rate, prioritizing safety constraints.

[0015] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in the first aspect.

[0016] Fourthly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in the first aspect.

[0017] Fifthly, the present invention also provides a computer program product, the computer program product comprising a computer program, which, when executed by a processor, implements the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in the first aspect.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: In terms of safety status, this invention constructs a multi-layered, progressive collaborative architecture including structural safety protection, electrical safety protection, fire safety protection, environmental safety protection, and intrusion safety protection layers. Regarding energy efficiency, it selects electrical components at the source, performs dynamic power quality management to eliminate additional power losses midway, and dynamically adjusts strategies at the control stage. This addresses the lack of systematic integration and collaborative linkage mechanisms in safety protection and the fragmented and unsystematic nature of energy efficiency optimization methods, providing a prerequisite for collaborative control considering safety protection, energy efficiency optimization, modular design, and operation and maintenance management. Then, it constructs a unified state vector including safety status, energy efficiency status, module topology status, and health status. Based on this unified state vector, it determines the safety risk margin, loss gradient, and lifetime loss rate. Based on the safety risk margin, loss gradient, and lifetime loss rate, it performs collaborative control with safety constraints as the priority condition. This achieves collaborative control considering safety protection, energy efficiency optimization, modular design, and operation and maintenance management, ensuring the safety, energy efficiency, and maintainability of the AC low-voltage integrated distribution cabinet under complex operating conditions. Attached Figure Description

[0019] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0020] Figure 1 This is a system architecture diagram of Embodiment 1 of the present invention; Figure 2 This is a flowchart of the security risk assessment algorithm in Embodiment 1 of the present invention; Figure 3 This is a flowchart of the multi-objective energy efficiency optimization algorithm of Embodiment 1 of the present invention; Figure 4 This is a flowchart of the modular intelligent recognition configuration of Embodiment 1 of the present invention; Figure 5 This is a flowchart of the life prediction and maintenance management process of Embodiment 1 of the present invention; Figure 6 This is a system economic benefit analysis diagram of Embodiment 1 of the present invention; Figure 7 This is a flowchart illustrating the preferred method for multiple constraints on shell plate thickness in Embodiment 1 of the present invention. Figure 8 This is a structural diagram of the physical degradation-data-driven hybrid lifetime prediction model of Embodiment 1 of the present invention; Figure 9 This is a flowchart of the risk-energy efficiency-lifetime coordinated control of Embodiment 1 of the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0023] Example 1: As a key node device in the power system, the AC low-voltage integrated switchgear (JP) cabinet undertakes the core functions of power distribution, control, and protection. With the continuous expansion of the power system and the increasing requirements for power supply quality, traditional JP cabinets have revealed many shortcomings in terms of safety, energy efficiency, and maintainability.

[0024] First, there are significant systemic deficiencies in safety protection. In traditional JP cabinets, electrical safety, fire safety, environmental safety, and intrusion safety are often designed and deployed independently, lacking a coordinated mechanism. This fragmented protection model can lead to problems in actual operation: for example, when electrical overheating causes a fire, only independent smoke detectors trigger an alarm after the fire starts, without coordinated prevention of electrical temperature anomalies and fire protection systems; similarly, when high humidity leads to a decrease in electrical insulation, dehumidification systems and leakage protection systems respond independently, lacking a coordinated protection strategy. Furthermore, traditional JP cabinets mostly have protection ratings of IP30 to IP40, meaning they can only prevent the entry of solid foreign objects larger than 2.5 mm in diameter and the intrusion of vertically dripping water droplets. Their protection capabilities are severely insufficient when facing real-world environmental challenges such as rain splashes, dust accumulation, or coastal salt spray intrusion. Safety monitoring is also incomplete, often focusing only on a few electrical parameters while neglecting equally important safety boundary conditions such as thermal state, mechanical vibration, and gas concentration, leaving many potential safety hazards in monitoring blind spots.

[0025] Secondly, low energy efficiency is another prominent problem with traditional JP switchgear. The overall energy efficiency of traditional JP switchgear is typically only 88% to 92%, which means that about 8% to 12% of electrical energy is lost as heat in the transformer, switching devices, and connecting wires. The main reasons for low energy efficiency include: First, due to the inherent hysteresis loss and eddy current loss characteristics of the core material, traditional silicon steel transformers have high no-load losses and continue to consume electrical energy even under zero load conditions; Second, inductive loads in the system (motors, transformers themselves) reduce the power factor. Typically, the power factor of an uncompensated JP switchgear is only 0.75 to 0.85. Low power factor leads to increased line current, increased line losses, and limited transformer output; Third, nonlinear loads such as frequency converters, switching power supplies, and rectifiers generate a large amount of harmonic current. The total harmonic current distortion rate of a typical JP switchgear can reach 20% to 30%. Harmonics not only increase line losses and transformer core losses, but may also cause malfunctions of protection devices and burnout of capacitors. Under light load conditions with a load rate below 30%, the energy efficiency of traditional JP switchgear deteriorates further because the no-load loss of the transformer core does not change with the load, while the load loss of devices and lines is proportional to the square of the load rate.

[0026] Furthermore, insufficient intelligent operation and maintenance management is a significant drawback of traditional JP cabinets. Installation and maintenance heavily rely on specialized technicians. A typical troubleshooting process requires technicians to bring various testing instruments to the site and spend several hours meticulously checking each item to determine the cause of the fault, with a mean time to repair (MTTR) typically ranging from 4 to 8 hours. Regarding fault diagnosis, due to the lack of accumulated and analyzed historical operating data, technicians often rely solely on experience and on-site observation, making it difficult to guarantee the accuracy and efficiency of fault diagnosis. Lifespan prediction relies almost entirely on the manufacturer's stated lifespan and periodic visual inspections, failing to provide dynamic assessments based on the actual operating status and degradation trends of the equipment. Operating data utilization is low. Although some newer JP cabinets are equipped with basic power monitoring functions, this data is usually only used to generate monthly reports and is not systematically used to optimize operating strategies, predict faults, and develop maintenance plans.

[0027] Existing technologies collectively reflect the shortcomings in the three aspects mentioned above. For example, the modular JP cabinet solution achieves physical modularity at the component level, allowing different functional components to be replaced independently. However, at the electrical connection level, this solution still relies on manual wiring; each module replacement requires technicians to rewire and manually set parameters, failing to achieve true plug-and-play functionality. It focuses on electric shock protection and simple temperature monitoring, with insufficient coverage for fire protection, environmental protection, and intrusion protection. The energy efficiency optimization JP cabinet solution compensates for power factor by adding a static var generator, but this solution only focuses on optimizing reactive power and fails to systematically optimize from multiple dimensions such as electrical component selection, comprehensive power quality management, and automated control.

[0028] Based on the above analysis, the existing JP cabinet technology has the following technical problems that need to be solved: the lack of systematic integration and collaborative linkage mechanism for security protection makes the overall protection effect of various protection measures fail to reach the expected level; energy efficiency optimization methods are scattered and unsystematic, lacking a comprehensive technical solution from the selection of high-efficiency components and power quality management to optimization control; modular design remains at the mechanical level and has not been extended to the electrical connection and intelligent configuration level; operation and maintenance management lacks predictive and proactive technical means, and cannot fully realize the value of operation data.

[0029] To address at least one of the aforementioned issues, this embodiment provides a comprehensive management method for distribution cabinets based on multiple protections and energy efficiency optimization. At the physical and security level, a comprehensive security system is constructed through five progressively layered protection measures. At the energy efficiency optimization level, significant improvements in energy efficiency are achieved through three technical aspects: the selection of high-efficiency electrical components, dynamic power quality management, and intelligent optimization control algorithms. At the deployment and maintenance level, plug-and-play installation and automatic configuration of modules are achieved through standardized electrical interfaces and intelligent identification technology. At the operation and maintenance level, predictive intelligent operation and maintenance is achieved through four technical aspects: status monitoring, fault diagnosis, lifespan prediction, and maintenance management.

[0030] This embodiment unifies four subsystems—safety protection, energy efficiency optimization, modular design, and operation and maintenance management—into a single state variable and constraint optimization framework. The collaborative control center, based on safety status, energy efficiency status, module topology status, and health status, forms a unified state vector. It first calculates the safety risk margin, loss gradient, and lifetime attrition rate, and then, under the condition of prioritizing safety constraints, outputs control actions such as compensation, derating, isolation, heat dissipation, and maintenance work orders. The unified state vector... Defined as Equation (1): (1) in, Indicates safety features such as temperature, smoke, access control, vibration, insulation, and leakage current; It represents energy efficiency characteristics such as active power, reactive power, power factor, harmonics, voltage deviation, and line loss; This indicates module characteristics such as module type, rated parameters, interface status, and communication address; This represents lifetime characteristics such as contact resistance, insulation resistance, number of switching actions, hot spot temperature, partial discharge count, and health status. The objective function and safety constraints of the coordinated control are determined by equation (2); the controller selects the action with the lowest J value from the candidate actions that meet the risk constraints. (2) in, These include control variables such as Static Var Generator (SVG) output, Active Power Filter (APF) compensation current, capacitor switching, temperature control setpoint, load derating, and maintenance actions; , and These are energy consumption targets, lifespan loss targets, and operating cost targets; The security risk value after fusion. To allow for a risk ceiling, this embodiment implements a closed-loop control method that prioritizes safety constraints, from unified state modeling to coordinated optimization of energy efficiency and lifespan.

[0031] Specifically, the security risk margin is calculated as the difference between the allowable risk ceiling and the current fusion risk value, i.e. .in, The results are obtained by fusing evidence from the fuzzy inference results and the Bayesian network results using equations (32) and (33) as described later. This represents the upper limit of the permissible risk in equation (2). When A value greater than or equal to 0 indicates that the current risk has not exceeded the allowable limit; the larger the value, the more sufficient the safety margin. When the value is less than 0, it indicates that the current safety risk has exceeded the limit. The collaborative control center prohibits energy efficiency actions that conflict with safety protection and increases the priority of isolation, derating, heat dissipation, or maintenance actions.

[0032] The loss gradient is used to evaluate the marginal impact of changes in each candidate control variable on the total system loss. The total system loss is expressed by equation (9). Based on this, for the j-th continuous control variable Finite difference calculation ,in This is the minimum allowable adjustment step size for this control variable. For the first A unit vector; for discrete control actions such as capacitor switching and load derating, calculate the loss difference before and after the candidate action is implemented. .when or When the value is less than 0, it indicates that the corresponding adjustment can reduce system losses, and the larger the absolute value, the more obvious the marginal energy saving benefit; when it is greater than 0, it indicates that the action will introduce additional losses.

[0033] Lifetime attrition rate is used to characterize the rate at which the lifespan of critical equipment is consumed per unit time. Preferably, it is the cumulative insulation damage calculated using equation (23). Based on this, in adjacent control cycles Internal computation ,in The failure threshold is used as the threshold. When multiple critical devices such as contacts, insulation, and transformers exist simultaneously, the normalized damage increment of each critical device is calculated separately, and the maximum value among them is used as the current life loss rate. Figure 8 The mixed remaining lifetime prediction results shown are used to verify the lifetime loss trend; when the mixed remaining lifetime decreases continuously and the rate of decrease exceeds the preset threshold, the weight of the lifetime loss target is increased, or load derating, enhanced heat dissipation and maintenance actions are triggered.

[0034] This embodiment constrains four functional categories—safety status, energy efficiency status, module topology status, and health status—to the same state vector, the same risk assessment result, and the same action cost function. When high temperature, increased harmonics, increased contact resistance, and insufficient module thermal rating occur simultaneously, the safety risk margin, loss gradient, and lifespan attrition rate are first calculated. Then, a decision is made regarding whether to reduce the load, improve heat dissipation, adjust SVG / APF output, postpone low-priority energy-saving actions, or generate a maintenance work order. The method in this embodiment includes the following steps: S1. Multiple security protections (multiple security protection subsystem design): The architecture employs a five-layered, progressively collaborative protection system: the first layer is the structural safety protection layer, providing a physical barrier; the second layer is the electrical safety protection layer, providing core electrical protection; the third layer is the fire safety protection layer, providing thermal disaster response capabilities; the fourth layer is the environmental safety protection layer, providing external environmental adaptability; and the fifth layer is the intrusion safety protection layer, providing proactive defense against human-caused damage.

[0035] The structural safety protection layer achieves an IP65 protection rating through material selection, sealing design, and mechanical structure design. This means that under specified test conditions, dust-tight protection is formed, preventing dust from entering to the extent that it affects the safe operation of the equipment; it can withstand low-pressure water spray from any direction, simulating severe weather conditions such as heavy rain. Key technologies for achieving the IP65 protection rating include: a double-seal structure between the cabinet door and the cabinet body (an outer labyrinth mechanical seal and an inner silicone rubber compression seal); a multi-layer sealed cable connector at the cable inlet, consisting of an elastic rubber sealing ring, a metal compression ring, and a waterproof nut; and a labyrinth waterproof ventilation structure at the ventilation and heat dissipation vents, ensuring an airflow rate of no less than 60% while utilizing a multi-turn ventilation path to ensure that water droplets are repeatedly impacted and trapped on the channel walls under inertial force, thus preventing water droplets from entering the cabinet. The seismic design meets the requirements of a seismic fortification intensity of 8 degrees. By installing vibration-damping rubber supports (a multi-layer alternating vulcanized structure of natural rubber and metal plates) at the bottom of the cabinet, a ground acceleration of 0.3g is attenuated to no more than 0.1g when transmitted to the equipment inside the cabinet. The outer shell is made of 304 stainless steel sheet with a thickness of not less than 2.0 mm and has an impact energy resistance of not less than 20 joules.

[0036] The outer shell is made of 304 stainless steel sheet with a thickness of not less than 2.0 mm; specifically, "not less than 2.0 mm" is the minimum feasible thickness determined by strength, stiffness, impact energy absorption, corrosion thinning, manufacturing tolerances, and a standard sheet thickness sequence. The largest unreinforced plate area in the cabinet door or side panel is equivalent to the long side. short side A rectangular thin plate with a modulus of elasticity of Poisson's ratio is Bearing equivalent surface load When, its bending stiffness As in equation (3): (3) For plate regions supported on four sides or with stiffened ribs at the boundary, the maximum equivalent stress and maximum deflection Can be based on boundary coefficient and The estimation is performed as shown in equation (4): (4) Strength constraints and stiffness constraints The lower limit of the plate thickness is obtained as shown in equation (5): (5) For the requirement of impact energy not less than 20 joules, the equivalent width of the impact action is used. The design verification of local plastic energy dissipation and energy absorption capacity was carried out. and the lower limit of impact constraint plate thickness As shown in equation (6): (6) For coastal salt spray, chemical dust, or high humidity environments, it is also necessary to add the full-life corrosion thinning amount, plate negative tolerance, and welding heat deformation allowance to obtain the lower limit of corrosion and manufacturing constraints, as shown in equation (7): (7) Finally, the minimum thickness that satisfies all constraints is selected from the standard plate thickness sequence, as shown in equation (8): (8) In equations (3) to (8), For thickness The bending stiffness of the lower part; and These are the long and short sides of the equivalent plate region, respectively; ν is the elastic modulus of 304 stainless steel; ν is Poisson's ratio. This is the equivalent surface load; , These are the boundary factors for stress and deflection, respectively. Yield strength; and These are allowable stress and allowable deflection, respectively. The equivalent width for impact effect; To design impact energy; The effective thickness that needs to be retained even at the end of its service life; Annual corrosion thinning rate; Design life; and These are the negative tolerance of the sheet metal and the welding heat deformation allowance, respectively. To meet the lower limit of the process for installing sealing flanges, hinges, and locking components; For procureable standard plate thickness series; To finalize the selection of the plate thickness.

[0037] In ordinary indoor or general industrial environments, based on the maximum unreinforced plate area size of the cabinet, the equivalent pressure of IP65 spraying, the impact energy of 20 Joules, and the lifespan boundary of 20 to 25 years, a plate thickness of 1.5 mm cannot simultaneously satisfy the impact energy absorption and welding deformation margin. 2.0 mm is the minimum feasible solution in the standard plate thickness sequence. In salt spray, strong corrosion, or high impact scenarios, the corrosion thinning amount and impact term in equation (7) increase, and the thickness shifts to 2.5 mm or higher. Therefore, not less than 2.0 mm is the minimum feasible boundary, and using 2.5 mm is a scenario-specific and stricter choice for salt spray and dust environments.

[0038] The insulation resistance of the main circuit of the electrical safety protection layer is not less than 100 megohms, allowing sufficient margin for insulation degradation due to aging. The power frequency withstand voltage level is 3000 volts / 1 minute, and the impulse withstand voltage level is 8000 volts. The grounding system adopts an independent grounding copper busbar design with a cross-section of 50 mm × 5 mm. It is reliably connected to the building's grounding grid through hot-dip galvanized flat steel, and the measured grounding resistance does not exceed 0.5 ohms. In the TN-S grounding system, the protective grounding conductor (PE) and the neutral conductor (N) are strictly separated from the power supply point to ensure that the fault current has an independent low-impedance return path. The leakage current protection adopts a three-level setting: the first level is the main incoming line protection, with an operating current set at 300 mA and an operating time not exceeding 0.3 seconds, mainly used for fire protection; the second level is the branch line protection, with an operating current set at 100 mA and an operating time not exceeding 0.2 seconds, used for equipment protection; the third level is the end-point protection, with an operating current set at 30 mA and an operating time not exceeding 0.1 seconds, used for personal electric shock protection. This hierarchical protection scheme enables selective protection, meaning that a fault at the end is only triggered by the end protection system, without affecting other circuits at the upper or same level, thus minimizing the impact of the fault.

[0039] The fire safety protection layer constitutes a multi-layered active defense system through three aspects: fire-resistant materials, fire detection, and automatic fire suppression. Regarding fire-resistant materials, the cabinet interlayer is filled with 50mm thick ceramic fiberboard. This ceramic fiberboard, manufactured from aluminum silicate fiber using a wet-forming process, exhibits extremely low thermal conductivity at flame temperatures of 1000℃ to 1200℃, utilizing its high-temperature stability to construct an effective thermal barrier. Intumescent fire-resistant sealant is filled between the ceramic fiberboards; under high fire temperatures, the sealant expands to form a dense carbonized layer, preventing the spread of flames and hot smoke. The fire-resistant partition design divides the cabinet interior into several independent fire-resistant units, each separated by a fire-resistant partition extending from the inner surface of the cabinet door to the outer surface of the back panel, forming a complete fire-resistant partition surface. Fire detection employs a dual-detector configuration logic: a fire event is confirmed and the fire suppression system is triggered only when both detectors alarm simultaneously. This dual-detection logic effectively reduces the probability of false alarms from a single detector. The automatic fire suppression system employs thermal aerosol extinguishing technology. The extinguishing agent is a mixture of strontium and potassium salts, which generates a large number of micron-sized solid particles during combustion. The extinguishing mechanism of these particles is as follows: they capture active free radicals in the combustion chain reaction through chemical reactions, interrupting the chain propagation of combustion; simultaneously, the large specific surface area of ​​the solid particles generates physical heat absorption and isolation effects. The extinguishing agent dosage is designed at 100 grams per cubic meter of cabinet volume, with an extinguishing concentration of 100 grams per cubic meter. The total flooding arrangement ensures that all electrical clearances within the cabinet are covered by the extinguishing medium.

[0040] The environmental safety protection layer primarily addresses four environmental factors: temperature, humidity, corrosion, and dust. For temperature control, a dual-purpose cooling and heating solution using semiconductor cooling chips and PTC heaters is employed. The semiconductor cooling chips cool when a forward current is applied and heat when a reverse current is applied. Multiple cooling chips are arranged in parallel, with a total cooling power of 600 watts, capable of maintaining the cabinet temperature below 30°C at an ambient temperature of 50°C. The heat generated at the hot end of the cooling chips is conducted to the aluminum heat sink outside the cabinet via heat pipes. In cooling mode, the low temperature at the cold end creates controlled cold air circulation within the cabinet through a built-in air circulation fan. For humidity control, the principle of cold-end condensation dehumidification using cooling chips is employed: when the cold end temperature is controlled below the dew point temperature, water vapor in the air condenses into liquid water at the cold end, and the condensate is drained out of the cabinet through a drain pipe. The advantages of cooling chip dehumidification are: the entire process has no moving mechanical parts, a simple structure, requires no maintenance, has a long lifespan, and does not use compressors or refrigerants, meeting environmental protection requirements. When the ambient temperature is below 5℃ and the cooling coil cannot effectively dehumidify, the system switches to PTC heater dehumidification mode: heating is used to reduce relative humidity, and increased external ventilation is used to expel moisture from the cabinet. For corrosion protection, the aforementioned epoxy-polyurethane coating system is used, coupled with online corrosion rate monitoring, providing real-time corrosion status feedback based on electrochemical sensors. The dustproof design employs a replaceable pleated filter with an F9 filtration rating, achieving a filtration efficiency of no less than 95% for particles with a diameter of 0.4 microns. Differential pressure sensors are installed before and after the filter; a filter replacement reminder is triggered when the pressure difference exceeds 200 Pascals. The balance between the filter's dustproof and heat dissipation / ventilation functions is achieved by rationally determining the filter area: after calculating the required ventilation volume, a filter with a sufficiently large area is selected to ensure that the initial pressure drop of the filter does not exceed 100 Pascals, and that the maximum pressure drop during the filter's lifespan does not exceed the fan's rated air pressure capacity.

[0041] The intrusion security protection layer achieves physical security through three stages: multi-factor authentication, video surveillance, and intrusion detection. Access control authentication employs a three-layer progressive authentication scheme: the first layer is IC card swipe authentication, providing basic access; the second layer is a 6-digit numeric password verification, activated after successful IC card authentication, with a 30-minute lockout after three consecutive incorrect entries; the third layer is capacitive fingerprint sensor verification. The progressive logic of the three-layer authentication is designed so that routine operations by maintenance personnel only require two layers of authentication (IC card and password, approximately 15 seconds), while critical operations involving changes to system settings and parameter configurations require all three layers of authentication (approximately 30 seconds, including fingerprint verification time). Video surveillance uses a 2-megapixel CMOS camera, supporting wide dynamic range (WDR) technology to adapt to scenarios with drastic changes in internal and external lighting when the cabinet door is open. The camera has a built-in passive infrared (PIR) motion detection sensor; recording is triggered when a change in human infrared signal is detected around the cabinet, and it enters standby mode when no one is present to save storage and communication resources. For intrusion detection, a sensor network with multiple detection technologies is configured: microswitches detect unauthorized opening of cabinet doors; ball-bearing tilt sensors detect abnormal tilting of the cabinet; and piezoelectric accelerometers detect destructive behaviors such as drilling and impact. When any intrusion sensor is triggered, the system immediately activates an audible and visual alarm, simultaneously sending alarm information including alarm type, time, and camera-captured image to the monitoring center via the network, and sending SMS notifications to pre-set management personnel's mobile phones.

[0042] S2. Intelligent Energy Efficiency Optimization (Design of Intelligent Energy Efficiency Optimization Subsystem): Achieving a systematic improvement in energy efficiency is achieved through three technological steps: at the source, high-efficiency electrical components are selected to fundamentally reduce energy loss; in the middle stage, dynamic power quality management is used to eliminate additional losses caused by low power factor and harmonic pollution; and in the control stage, intelligent optimization algorithms are used to dynamically adjust the operating strategy.

[0043] High-efficiency electrical components comprise three parts: amorphous alloy transformers, intelligent circuit breakers, and low-loss busbar systems. The energy-saving mechanism of amorphous alloy transformers lies in the unique properties of their core material: amorphous alloy strips are formed by directly cooling and solidifying molten iron-based alloys at an ultra-high cooling rate. This ultra-fast solidification process prevents crystal formation, maintaining the material in a disordered amorphous structure with atomic arrangement. In this amorphous structure, the lack of magnetic domain walls and grain boundaries, as seen in crystals, significantly reduces hysteresis losses during magnetization. Simultaneously, the thickness of the amorphous strip (25 micrometers) is only about one-tenth that of traditional grain-oriented silicon steel sheets. Eddy current losses in an alternating magnetic field are proportional to the square of the thickness, thus the thin strip further significantly reduces eddy current losses. These two reductions in losses result in a 60% to 70% reduction in no-load losses for amorphous alloy transformers compared to traditional grain-oriented silicon steel transformers of the same capacity. Taking a 2000 kVA transformer as an example, the no-load loss of a traditional silicon steel transformer is about 3450 watts, while the no-load loss of an amorphous alloy transformer is about 1200 watts, a reduction of about 65%.

[0044] Intelligent circuit breakers integrate an electronic control unit (ECU) on top of the mechanical protection functions of traditional circuit breakers. This ECU incorporates current transformers, voltage transformers, and temperature sensors, enabling real-time measurement of the current, voltage, and terminal temperatures passing through the circuit breaker. The ECU transmits the measured data to the central control system via Modbus TCP or Profibus DP communication protocols, achieving remote online monitoring of the circuit breaker's status. Based on the measured current and temperature data, the ECU employs a physics-based thermal model to estimate the real-time temperature of the circuit breaker contacts. This thermal model is based on the balance equation between Joule heat generation and heat dissipation; the contact temperature recursion and contact resistance degradation model are given in the health management model section below. When the estimated contact temperature approaches or exceeds the manufacturer's specified long-term allowable operating temperature, the system issues an early warning, prompting maintenance to be arranged before equipment damage occurs. Furthermore, the circuit breaker's ECU integrates an arc fault detection algorithm: by analyzing the high-frequency component characteristics of the current waveform and distinguishing it from the normal load current waveform, once arc characteristics are identified, the circuit is tripped and disconnected before the arc causes a fire.

[0045] The energy-saving mechanism of the low-loss busbar system is as follows: First, T2 industrial pure copper is used as the conductor material, and its conductivity is about 101% of the international standard conductivity of annealed copper, which is higher than that of ordinary electrical copper. Second, an irregular cross-sectional geometry optimized by finite element analysis of current density distribution is adopted. Compared with a rectangular cross-section of the same copper cross-sectional area, although the skin effect of current tending to be distributed on the conductor surface is not significant under power frequency conditions, the local effect at the conductor connection can be suppressed by reasonable rounded transition, thereby further reducing the equivalent resistance. Third, the surface of the copper busbar is silver-plated, which utilizes the extremely high conductivity of silver to reduce contact resistance, and at the same time utilizes the excellent anti-oxidation properties of silver to maintain a long-term stable contact resistance.

[0046] Dynamic power quality management comprehensively employs three complementary technologies: Static Var Generator (SVG), Active Power Filter (APF), and Smart Capacitor Bank. The selection and division of labor among these three technologies are as follows: SVG primarily handles rapid compensation of dynamic reactive power, with a response time of less than 5 milliseconds, suitable for situations where the power factor changes rapidly with the load; APF primarily handles harmonic current compensation, its high-frequency switching action generating a compensation current of the same magnitude but opposite phase to the load harmonic current, reducing the total harmonic distortion rate on the grid side to below the standard limit; Smart Capacitor Bank handles steady-state compensation of fundamental reactive power. Due to its lower cost compared to SVG, it shares the compensation capacity of SVG in steady-state conditions where rapid response is not required, thus reducing the overall system cost.

[0047] The operating principle of an SVG (Static Var Generator) is based on voltage source inverter technology. After its DC bus capacitor is charged to a stable operating voltage, the SVG uses PWM to control the on / off state of its IGBT switching transistors, generating a sinusoidal voltage with adjustable amplitude and phase at the AC output. When the SVG's output voltage amplitude is higher than the grid voltage, the SVG provides capacitive reactive power to the grid (equivalent to the function of a capacitor); when the output voltage amplitude is lower than the grid voltage, the SVG absorbs inductive reactive power from the grid (equivalent to the function of an inductor). By adjusting the output voltage amplitude in real time, the SVG can achieve continuous reactive power control from rated capacitive capacity to rated inductive capacity. The SVG capacity is selected as 800 kVvar, the response time is within 5 milliseconds, and the reactive power compensation accuracy is ±1% of the rated reactive power.

[0048] The operating principle of an Active Power Filter (APF) is based on harmonic current detection and reverse injection technology. A current sensor measures the current waveform of the nonlinear load in real time. A digital signal processor (DSP) uses algorithms such as Fast Fourier Transform (FFT) or Synchronous Reference Coordinate Transform (SRC) to separate the fundamental and harmonic current components from the total current. The controller generates a compensation current command with the same amplitude but opposite phase to the harmonic components. The IGBT inverter tracks this command and injects compensation current into the grid, thus canceling out the harmonic components in the total current observed from the grid side, leaving only the fundamental component. The APF's tracking speed is required to be no more than 100 microseconds, its harmonic compensation range covers the 2nd to 50th harmonics, its harmonic suppression rate is no less than 95%, and its rated compensation current is 600 amperes.

[0049] The intelligent capacitor bank adopts a hybrid compensation configuration with unequal capacity groups: the first group consists of four small-capacity capacitors, each with a capacity of 50,000 kV, and the second group consists of four large-capacity capacitors, each with a capacity of 100 kV, for a total installed capacity of 600 kV. The number of groups is determined based on matching the reactive power compensation capacity gradient of the SVG (Static Var Generator): the SVG's fast response and stepless adjustment cover continuous reactive power demand from 0 to 800 kV, while the capacitor bank provides stepped baseline reactive power compensation. Switching control uses thyristor solid-state switches instead of traditional mechanical contactors. The key technology for thyristor switching is that the switch is activated when the voltage across the thyristor crosses zero, avoiding inrush current surges caused by the difference between the residual voltage of the capacitor and the grid voltage. To this end, the controller continuously monitors the voltage across the thyristor and the grid voltage, triggering the thyristor gate signal when the voltage difference crosses zero. This zero-crossing switching technology reduces the inrush current, which is tens of times the rated current when using traditional contactors, to less than 1.5 times the rated current, protecting the capacitors from overcurrent damage and eliminating the impact of inrush current on the power quality of the grid. The switching life of capacitors has been increased from approximately 100,000 cycles for contactors to over 1 million cycles for thyristors.

[0050] Intelligent energy efficiency management operates at both the monitoring and control levels. At the monitoring level, a power quality analyzer with Class A harmonic measurement capabilities is employed. This instrument, based on high-speed digital signal processing technology, synchronously samples voltage and current at a rate of 256 sampling points per power frequency cycle (approximately 12.8 kHz). Through a fast Fourier transform algorithm, the analyzer can provide amplitude and phase information for up to 50 harmonics, as well as power quality indicators such as total harmonic distortion (THD), harmonic distortion rates, interharmonic content, and flicker. The harmonic analysis accuracy meets the requirements of Class A measurement instruments. At the control level, a coordinated control algorithm based on multi-objective optimization is used. The algorithm's four-dimensional objective function is: the first dimension is minimizing the total system power consumption; the second dimension is minimizing the degree of power quality degradation (characterized by a combination of THD and power factor offset); the third dimension is minimizing the rate of equipment lifespan loss (modeled based on the accelerating effects of temperature, voltage, and current on insulation lifespan); and the fourth dimension is minimizing operating economic costs (including electricity purchase costs, reactive power penalty costs, and maintenance costs). Constraints include permissible voltage deviation (±10% of rated voltage), permissible current overload (e.g., not exceeding 110% of rated current), harmonic limits, and upper limit of transformer insulation life loss rate. An improved NSGA-Ⅲ multi-objective genetic algorithm is used for optimization. This algorithm uses a predefined reference direction vector to ensure a uniformly distributed Pareto front solution set in the four-dimensional objective space. The population size is 120 individuals, and the maximum number of generations is 600. The decision method for selecting the final solution from the final Pareto front is as follows: after normalizing the four objective function values ​​to the [0, 1] interval, a weighted sum method is used. The weight coefficients are set according to user preferences (default settings are energy consumption weight 0.35, power quality weight 0.25, equipment life weight 0.20, and operating cost weight 0.20). The solution with the smallest weighted sum is selected as the final control strategy.

[0051] The improvement in energy efficiency optimization does not lie in setting up SVG, APF, or capacitor banks separately, but in calculating the loss gradient based on the real-time measured load rate, harmonic content, bus temperature, and compensation device status, and treating reactive power compensation, harmonic mitigation, and temperature control power as the same optimization variable. The total system loss is calculated according to equation (9): (9) in, For no-load loss, For rated load loss, For load rate, For the first Subharmonic equivalent resistance This includes the losses of the fan, refrigeration system, control power supply, and compensation devices. The harmonic equivalent resistance and system efficiency are calculated according to equation (10): (10) Target power factor The required reactive power compensation is calculated according to formula (11): (11) The controller performs capacity decomposition between SVG continuous compensation and discrete capacitor switching, selecting the capacitor bank switching state according to equation (12): (12) The compensation current of the APF is injected in reverse from the harmonic detection result, as shown in equation (13): (13) Within each control cycle, the collaborative control center uses energy consumption, power quality, lifespan loss, and operating costs as objective functions, as shown in equation (14): (14) In equation (14), For energy consumption items, For power quality, This is a lifespan depletion item. This is the operating cost item. The optimization constraints are as shown in equation (15), where As a harmonic constraint index, The upper limit of harmonic constraints, It serves as a constraint indicator for lifespan loss rate. Maximum lifespan attenuation rate: (15) In equations (9) to (15), and These are system losses and output active power, respectively. and These are the no-load loss and the rated load loss, respectively. Transformer load factor; and These are the fundamental current and Subharmonic current; and Temperature Lower fundamental frequency and Subharmonic equivalent resistance; For skin effect and proximity effect correction coefficients; The amount of reactive power compensation required to achieve the target power factor; The active power of the load; and These are the current power factor and the target power factor, respectively. Output reactive power to SVG; For the first Capacitor bank switching status; For the first Capacitor group capacity; For APF compensation current; The fundamental angular frequency; for Subharmonic phase; to For multi-objective weights.

[0052] The aforementioned model eliminates the independent control of compensation equipment capacity, switching actions, and temperature control power. Instead, it enables collaborative decision-making based on loss gradients and safe lifespan constraints. Compared to conventional solutions that only configure reactive power compensation devices, this model demonstrates when low-cost capacitors should handle steady-state reactive power, when SVG should handle rapid fluctuations, and when APF should prioritize harmonic mitigation, thus forming a reproducible control process.

[0053] S3, Modular Intelligent Design (Modular Intelligent Design Subsystem Design): The JP cabinet's functional modules can be plug-and-play installed and automatically configured through standardized electrical interfaces, intelligent identification technology, and automatic configuration algorithms.

[0054] Standardized electrical interface design is the core technology for achieving true modularity. In traditional JP cabinets, similar modules from different manufacturers (such as circuit breaker modules and contactor modules) have different installation dimensions, wiring methods, and electrical parameters. Replacing modules in the field requires professional technicians to rewire, verify parameters, and perform functional tests. This invention solves this problem by defining a unified electrical interface standard on the intelligent connection backplane. The rated parameters of the electrical interface are 100 amps for rated current and 600 volts for rated voltage, covering the electrical requirements of most functional modules in low-voltage JP cabinets. The connection method uses a silver alloy spring-loaded finger plug-in structure, replacing the traditional bolt crimping method. The plug-in structure is designed with an insertion force of 50 to 150 Newtons and a withdrawal holding force of 200 to 400 Newtons. The interface's anti-misinsertion design employs a dual anti-misinsertion method using mechanical and electrical coding: mechanical coding uses 12 different arrangements of positioning pins and holes around the interface to ensure that only modules with the corresponding codes can be inserted into the interface; electrical coding uses the level combination of the coding pins during connection to secondary confirm the module type, serving as a redundant verification method for mechanical coding. The key safety design for supporting hot-swapping is as follows: the connector contacts are designed with unequal lengths, with the longest contact being the ground contact, which connects first and disconnects last, ensuring that grounding protection remains effective during hot-swapping of the module; the second longest contact is the signal contact, which completes signal handshake and data transmission within 200 milliseconds before the power contact is turned on; the shortest contact is the power contact, which is powered on by an electronic switch only after grounding and signal connection are confirmed and the controller verifies that the module insertion is legitimate.

[0055] The intelligent identification technology is based on an RFID radio frequency identification system. The selection of 13.56 MHz high-frequency RFID technology is primarily based on the following technical and economic trade-offs: While ultra-high frequency RFID (860 to 960 MHz) offers a long identification distance, it is susceptible to interference from metal reflections; low-frequency RFID (125 kHz), while possessing strong metal penetration capabilities, suffers from low data rates and small storage capacity, making it unsuitable for storing module parameter information; 13.56 MHz high-frequency RFID achieves a good balance between identification distance (5 to 10 cm), data rate (approximately 26 kbit / s), and anti-metal performance. The RFID tag is a passive tag (obtaining operating power from the radio frequency energy emitted by the reader), meeting the requirement of maintenance-free operation. It has a 2 kilobyte EEPROM storage capacity, storing the following information: the module's unique serial number, module type code, rated electrical parameters, manufacturer's name and manufacturing date, module calibration information (such as current transformer ratio and calibration coefficient), and module firmware version number. The reader antenna features a low-profile design and is embedded in the module slot on the back panel, forming near-field coupling with the RFID tag inserted into the bottom of the module.

[0056] The automatic configuration system's workflow includes the following steps: (1) Module insertion detection: The micro switch or capacitive sensor on the back panel detects that the module has been pushed into the slot; (2) RFID information reading: The reader excites the RFID tag at the bottom of the module with an RF carrier and reads all the module information stored in the tag; (3) Compatibility detection: The controller compares the read module parameters with the current system configuration and checks the compatibility conditions such as voltage level, current capacity, communication protocol version, and physical space. If any one of them is not met, it will issue a voice and light prompt and display the specific reason for incompatibility on the monitoring interface; (4) Resource and parameter allocation: The controller allocates resources according to the module type and the current system configuration. , assign a communication network address and priority to the module, calculate the protection parameter settings of the module (such as the operating current and time curve of overload protection), and configure the communication protocol stack parameters of the module; (5) Configuration verification: the controller sends a self-test command to the module, the module runs a self-test program to verify whether its sensor calibration, output drive circuit and protection function are normal; after the verification is passed, the controller sends a test command through the main communication bus to simulate normal working conditions and fault working conditions, and confirm that the module's measurement data and protection actions are consistent with expectations; (6) Configuration completion: the controller updates the system topology diagram, adds the new module to the system status monitoring and maintenance management database, and displays the installation completion information on the monitoring interface.

[0057] The focus of modular improvement has shifted from mechanical plug-in structures to a method of "identification-compatibility scoring-automatic generation of protection parameters-closed-loop verification". After the RFID reads the module parameters, the controller constructs compatibility features. The compatibility score is calculated according to formula (16), which includes six categories of characteristics: voltage level, current capacity, communication protocol version, thermal rating, physical coding, and firmware trust status. (16) when When the temperature is below a preset threshold, the module is prohibited from powering on and only allows reading and writing diagnostic information; when... When the threshold is met, the controller generates protection settings based on the module's rated current, backplane bus capacity, and contact temperature rated curve, as shown in equation (17): (17) The overload protection inverse time curve is automatically generated according to formula (18): (18) Equations (17) and (18) ensure that the protection parameters after module insertion do not depend on manual experience settings, avoiding misconfiguration due to differences in rated capacity, contact thermal state, or backplane capacity of similar modules. The automatic configuration verification program then confirms the configuration effectiveness through four steps: communication self-test, low-voltage power-on, short-time no-load measurement, and simulated fault response, thus forming an executable plug-and-play technical solution.

[0058] In equations (16) to (18), Rate the module compatibility; For the i-th type of compatibility input; This is the individual score after mapping the input; For single-item weights; This is the module's rated current. This refers to the capacity of the backplane busbar; This refers to the allowable current of the contact at temperature T. , , This is the reduction factor; To protect the set value; is the tripping time under current I; A, B, and p are parameters of the inverse time curve.

[0059] S4. Condition Prediction and Health Management (Design of Condition Prediction and Health Management Subsystem): Based on a four-layer architecture consisting of a monitoring layer, a diagnostic layer, a prediction layer, and a management layer, the system enables comprehensive perception of the status of JP cabinet equipment, accurate diagnosis of faults, scientific prediction of lifespan, and optimized maintenance decisions.

[0060] At the monitoring layer, electrical condition monitoring employs multiple sensor technologies: Non-contact batch temperature scanning of all electrical connection points within the field of view is performed using an infrared thermometer, while contact thermocouples directly measure the absolute temperature of key connections as a calibration benchmark for infrared temperature measurement; ultrasonic sensors detect acoustic signals in the air within a frequency range of 20 kHz to 100 kHz, which may originate from partial discharge in conductors or corona discharge in insulation (the movement of ions generated by the discharge produces sound waves at characteristic frequencies); current measurement is performed using Rogowski coils, which are coreless air coils whose output signal is proportional to the time-varying rate of change of the measured current, and is restored to a current waveform by an integrator. The linearity of Rogowski coils is unaffected by core saturation, making them suitable for a wide range of current measurements. Voltage measurement is performed using a combination of resistive and capacitive voltage dividers, with the frequency response of the dividers designed to cover the fundamental frequency of 50 Hz to the 50th harmonic frequency of 2500 Hz. Mechanical condition monitoring is achieved through triaxial MEMS accelerometers installed at key structural nodes of the cabinet and laser displacement sensors mounted on the transformer base plate. Stress monitoring utilizes fiber Bragg grating sensors. Their working principle is as follows: a periodic refractive index modulation structure is written into the core of a single-mode fiber. The center wavelength of the reflected light depends on the grating's period and the effective refractive index of the core. When the grating is subjected to axial strain, the grating period is stretched or compressed, causing the reflected wavelength to shift. By measuring this wavelength shift, the applied strain can be calculated. Fiber Bragg grating sensors are immune to electromagnetic interference; the sensor itself is part of the fiber optic cable, eliminating the need for electronic circuitry at the measurement point. Therefore, they are ideal for use in JP cabinets operating in strong electromagnetic environments. Environmental condition monitoring covers parameters such as temperature and humidity, condensation risk, gas concentration, and water quality. Condensation risk is assessed by comparing the dew point temperature (based on measurements from temperature and humidity sensors) with the temperature of the coldest surface inside the cabinet. When the dew point temperature is higher than the coldest surface temperature, condensation is at risk on that surface, triggering dehumidification or heating measures.

[0061] At the diagnostic layer, fault diagnosis employs a hybrid strategy combining physical model-based, knowledge rule-based, and data-driven approaches, as a single diagnostic method often cannot cover all fault types and operating conditions. The physical model-based approach uses physical models of the equipment (such as transformer thermal models, circuit breaker arc models, insulation aging models, etc.) to estimate internal parameters representing the equipment's health status from online measurement data. When estimated parameters (such as transformer winding resistance, circuit breaker contact resistance, dielectric loss factor of insulators, etc.) deviate from their baseline values ​​by more than a set threshold, a related fault is diagnosed. The knowledge rule-based approach uses a fault diagnosis rule base compiled by domain experts, expressed in "If..., Then..." format. Each rule includes the fault type, related symptoms, judgment threshold, and confidence assessment. The data-driven approach uses machine learning techniques to train the fault diagnosis model using historical monitoring data and corresponding fault labels. The advantage of this approach is its ability to discover complex fault patterns that are difficult to capture using physical models and expert rules; the disadvantage is the need for a sufficient quantity and quality of training data. The three methods are integrated through a weighted voting mechanism, with the weights dynamically adjusted based on the accuracy of each method in diagnosing the fault types it is responsible for over the past 12 months.

[0062] In the prediction layer, the remaining lifetime prediction adopts a hybrid prediction method of "parameter identification - physical degradation integral - time series learning - error adaptive fusion". Unlike schemes that only give the names of the physical degradation model and the data-driven model, this invention first unifies current, voltage, hot spot temperature, humidity, harmonics, contact resistance, insulation resistance, number of switching actions, vibration, and partial discharge count into a device health state vector; then, the physical degradation channel calculates the damage integral caused by contact aging, insulation aging, and thermal aging; at the same time, the data-driven channel learns the degradation sequence within the sliding time window and outputs the data-driven remaining lifetime and its uncertainty; finally, adaptive weighted fusion is performed based on the prediction error and uncertainty of the two types of models within the most recent verification window.

[0063] The historical training samples preferably include normal operation segments, early warning segments, maintenance confirmation segments, and pre-fault degradation segments. For newly commissioned equipment lacking real failure labels, initial training can be performed using manufacturer life labels and maintenance records of similar equipment, followed by updating the fusion weights using the verification window error after the equipment's operation. This training and updating mechanism ensures that the data-driven channel does not replace physical constraints, but rather supplements the degradation pattern recognition capability in complex operating conditions where the physical model extrapolation is insufficient.

[0064] The input vector for health status prediction is based on the sampling time. The construction is as shown in equation (19): (19) All sensor data were resampled at a uniform time granularity of 1 second to 1 minute. Short-term missing data were compensated by linear or spline methods. Missing data with more than 3 consecutive sampling windows were marked as low-confidence samples. Outliers were robustly normalized using the median absolute deviation, as shown in equation (20). (20) The physical degradation pathways include three types of models: contact temperature rise recursion, contact resistance growth, and insulation damage integral. The circuit breaker contact temperature is recursively derived from equation (21): (twenty one) The increase in contact resistance is driven by the combined effects of current heating, wear from switching action, and vibration loosening, as shown in equation (22): (twenty two) The insulation damage of transformers and insulating components is integrated according to the heat, voltage, humidity, and harmonic coupling, as shown in equation (23): (twenty three) When the damage score reaches the failure threshold At that time, the remaining lifetime of the physical channel output is defined according to equation (24): (twenty four) The data-driven channel adopts a gated cyclic unit model with a sliding time window length L. For the normalized state sequence within each window, the GRU gate calculation is as shown in equations (25) and (26): (25) (26) To highlight critical moments such as recent abnormal temperature rises, sudden drops in insulation resistance, or increases in partial discharge counts, attention weights are introduced, as shown in equation (27): (27) Data-driven remaining lifetime output is shown in equation (28): (28) During the training phase, actual maintenance records, pre-fault degradation segments, and manufacturer lifetime labels were used to form samples. The loss function consisted of absolute error and lifetime monotonicity penalty, as shown in equation (29). (29) The weights of the physical model and the data model are not fixed empirical values, but are determined by the prediction error e_p, e_d, and uncertainty in the most recent validation window. , The adaptive approach is given as in equation (30): (30) The final hybrid remaining lifetime is calculated according to equation (31): (31) When the device operates within the temperature, voltage, and humidity range well covered by the physical model, the physical channel receives a higher weight due to its extrapolation stability. Conversely, when the device exhibits complex harmonics, abnormal vibrations, frequent interruptions, or degradation patterns not covered by the physical model appear in the sensing sequence, the data channel's weight increases. This structure clearly defines the model structure, input data, preprocessing, training objectives, and fusion rules, avoiding insufficient disclosure of "physical degradation model and data-driven model" in the manual.

[0065] In equations (19) to (31), This is the health status input vector at the k-th sampling time. , , , , , , , , and These are current, voltage, hot spot temperature, humidity, total harmonic distortion of current, contact resistance, insulation resistance, number of switching operations, vibration intensity, and partial discharge count. The i-th dimension feature after robust normalization; and These are equivalent heat capacity and equivalent thermal resistance, respectively. Ambient temperature; For insulation damage, This is the failure threshold; , , and These are the GRU update gate, reset gate, hidden state, and candidate hidden state, respectively. c represents the attention weights; c represents the context vector. and These are actual lifespan labels and predicted lifespans, respectively. , , and These represent the prediction errors and uncertainties for the physical channel and the data channel, respectively. , For weighting; A positive constant is set to prevent the denominator from being zero when calculating the fusion weights; The bias term for the data-driven remaining lifetime output layer is used to characterize the lifetime baseline offset when the baseline value is taken in each context feature. This is the transpose of the weight vector of the data-driven remaining lifetime output layer, and its components are used to characterize the contribution direction and relative weight of the corresponding context features to the remaining lifetime prediction results.

[0066] At the management level, maintenance decision optimization considers the remaining lifespan forecast of each piece of equipment, its importance level (e.g., critical equipment, important equipment, general equipment), and system operating constraints (e.g., whether downtime is possible, availability of backup equipment, and maintenance resource availability) to optimize future maintenance plans. The optimization employs a mixed-integer linear programming model, with the objective function being to minimize the total maintenance cost (including the direct cost of preventative maintenance, the expected direct cost of post-failure repair, and the production loss cost of equipment downtime). The output of the optimization model is a maintenance plan for the next 12 months, including the start time of each maintenance task, the execution team, required spare parts, and maintenance downtime windows, and providing alternative suggestions for situations where the maintenance plan may be adjusted due to unforeseen events.

[0067] S5. Integrated and Collaborative Control: While ensuring safety, energy efficiency is optimized, and ease of operation and maintenance is also taken into account. This design principle is reflected in a strict priority system.

[0068] The priority system categorizes system decisions into three priority levels: The highest priority is safety incident handling. When any safety sensor triggers an alarm, the safety subsystem independently takes pre-set protective actions (such as tripping power, activating fire suppression, and locking access control), independent of other subsystems. Its execution does not depend on the decisions of the collaborative control center. The medium priority is energy efficiency optimization. The energy efficiency optimization algorithm continuously monitors the status output of the safety system during operation. When the safety system is in a "normal" or "warning" state, the energy efficiency optimization algorithm operates normally; when the safety system enters an "alarm" or "emergency" state, the energy efficiency optimization algorithm pauses optimization calculations, reverts to pre-set conservative operating parameters (safety-first parameter configuration), and waits for the safety status to return to normal before restarting optimization. The lowest priority is maintenance plan adjustment. Maintenance management proposes maintenance plan suggestions based on the equipment's operating status and maintenance needs, but these suggestions must be verified by both the safety system (confirming that the maintained equipment can be safely shut down) and the energy efficiency system (confirming that the economic losses from maintenance shutdowns are within acceptable limits) before implementation.

[0069] The state machine of the safety subsystem includes five states and state transition conditions. The normal state is the system's routine operating state, where all safety parameters are within normal ranges. When any safety parameter enters the warning zone (e.g., temperature reaches 55℃, humidity reaches 70%, vibration amplitude exceeds 1.0g, etc.) but does not reach the alarm zone, the system enters the warning state. At this time, the monitoring frequency increases from once every 10 seconds to once per second, and the warning information is pushed to the maintenance personnel's terminal. When any one or more safety parameters enter the alarm zone (e.g., temperature reaches 65℃, smoke detector alarm, vibration amplitude exceeds 2.0g, etc.), the system enters the alarm state, triggering audible and visual alarms and remote alarm information push notifications, and begins to execute corresponding protection strategies (e.g., increasing heat dissipation power, starting dehumidification equipment, preparing the fire extinguishing system, etc.). When the alarm state lasts for more than a set time window (e.g., temperature exceeds 65℃ for more than 5 minutes or two or more safety parameters are simultaneously in the alarm zone), the system enters the emergency state, automatically executing emergency protection actions (e.g., cutting off some non-critical loads, starting the fire extinguishing system, locking cabinet doors, etc.). Once all safety parameters return to normal ranges and stabilize for a certain period, the system recovers from the emergency state to the normal state. The time series of state transitions and parameter changes are all recorded in an immutable event log for post-incident analysis and system reliability improvement.

[0070] Security risk assessment will output fuzzy reasoning With Bayesian network output Perform DS evidence fusion. Let the identification framework be normal (N), warning (P), alarm (A), and emergency (E). The fusion quality function is calculated according to equation (32): (32) The scalar risk value after fusion is calculated according to formula (33): (33) If a conflict arises between safety and energy efficiency actions at the same time, for example, under high temperature conditions, the energy efficiency algorithm requires a reduction in fan power while the safety algorithm requires an increase in fan power, the control center first calculates the comprehensive cost of each candidate action according to equation (34), and then selects the action with the smallest comprehensive cost: (34) In equation (34), Indicates the total cost of the action. Indicates the cost of the action. Indicates taking an action Penalties for exceeding limits afterward This indicates the additional penalty introduced by the action. Because... Set to higher At least an order of magnitude, the system automatically sacrifices energy efficiency to meet safety constraints during alarm and emergency states; under normal and warning states, it continues to perform energy efficiency optimization without exceeding the risk limit.

[0071] In equations (32) to (34), and These are the quality functions given by fuzzy inference and Bayesian networks, respectively; N, P, A, and E represent normal, early warning, alarm, and emergency states, respectively; K is the evidence fusion conflict coefficient; , , , These are the scalarized coefficients for each risk state; , , , These are the overall cost of action a, the cost of the action, the penalty for exceeding the limit, and the penalty for additional losses. and These are the penalty weights for safety risks and energy efficiency losses, respectively.

[0072] In one specific embodiment, such as Figure 1 As shown, the integrated JP cabinet, when deployed in an engineering project, consists of a multi-source sensing and execution layer, four types of functional subsystems, a collaborative control center, and an external interface layer. The multi-source sensing and execution layer collects information on temperature, humidity, smoke, access control, vibration, voltage, current, harmonics, partial discharge, and module identification, and transmits this information synchronously to the safety protection subsystem, energy efficiency optimization subsystem, modular intelligent subsystem, and state prediction and health management subsystem. The collaborative control center forms a unified state vector according to equation (1). Then, according to formula (2), the control calculation with safety constraints as the priority is executed, and actions such as tripping, compensation, derating, isolation, heat dissipation and maintenance work orders are output.

[0073] like Figure 2 As shown, the specific execution order of the safety risk assessment is as follows: first, multi-source safety data is generated from structural, electrical, fire, environmental, and intrusion sensors; then, time alignment, outlier removal, feature extraction, and threshold grading are performed; then, the data enters the fuzzy inference path and the Bayesian network path respectively; finally, DS evidence fusion is performed using equations (32) and (33) to obtain the risk value. If the risk level is alarm or emergency, the collaborative control center will immediately disable energy efficiency actions that conflict with safety actions and switch to the audible and visual alarm, non-critical load cut-off, fire extinguishing preparation, or fire extinguishing activation process.

[0074] like Figure 3 As shown, the specific execution process of energy efficiency optimization is as follows: the energy efficiency subsystem reads... , , , , Given variables, first calculate the loss, reactive power compensation, and APF compensation current according to equations (9) to (13), and then construct a four-objective optimization problem according to equations (14) and (15). After the optimization algorithm outputs Pareto candidate solutions, the collaborative control center selects the combination of SVG output, capacitor bank switching, APF compensation current, and temperature control power according to the user-set weights and safety risk margin, so that energy consumption is reduced without exceeding the safety threshold or accelerating lifespan loss.

[0075] like Figure 4 As shown, the specific execution process of modular intelligent configuration is as follows: After the module is inserted, the backplane first completes the handshake through the grounding contact and the signal contact. The RFID reader reads the module serial number, type, rated parameters and firmware version. The controller calculates the compatibility score according to formula (16). After the compatibility is approved, the protection setting and inverse time trip curve are automatically generated according to Equations (17) and (18); if the communication self-test, low voltage power-on, no-load measurement and simulated fault response are all passed, the system will close the electronic switch and update the topology database.

[0076] like Figure 5 As shown, full lifecycle management is not a standalone maintenance reporting function, but rather integrates five stages—new equipment installation and acceptance, initial operation monitoring, stable operation, performance degradation, and end-of-life—into a single timeline. At each stage, the system performs short-term anomaly prediction, mid-term maintenance window prediction, and long-term replacement decision prediction, and... Figure 8 The mixed remaining useful life prediction results shown are input into the maintenance optimization model to generate corrective maintenance, preventive maintenance, or predictive maintenance tasks.

[0077] like Figure 6 As shown, the system's economic benefit analysis consists of four parts: investment cost, operating revenue, social benefits, and financial indicators. In practical implementation, the energy efficiency management server measures transformer losses, line losses, harmonic losses, maintenance costs, and downtime losses separately. It then summarizes the value of electricity cost savings, maintenance savings, reduced downtime losses, and extended equipment lifespan into the investment payback period, rate of return, and net present value, used to illustrate... Figure 3 Energy efficiency optimization and Figure 8 The contribution of life prediction to actual engineering benefits.

[0078] like Figure 7 As shown, the preferred process for the outer shell thickness corresponds to equations (3) to (8). During implementation, the designer first inputs the maximum unreinforced area dimensions a and b of the cabinet door or side panel, the equivalent spray or impact load q, and the corrosion rate. And the target lifespan L; the controller or design software calculates the lower limit of strength respectively. Stiffness lower limit Impact energy absorption lower limit Corrosion manufacturing lower limit and the lower limit of the process Subsequently, select a plate thickness not less than 10 mm from the standard plate thickness sequence S. minimum value Therefore, 2.0 mm is the minimum feasible thickness to meet multiple constraints under normal operating conditions, while 2.5 mm is the preferred implementation value under salt spray, strong corrosion, or high impact conditions.

[0079] like Figure 8 As shown, the hybrid lifetime prediction model consists of data preprocessing, a physical degradation channel, a data-driven channel, and an adaptive fusion channel. Multi-source monitoring data is first resampled, missing data compensated, and robustly normalized to form... The physical channel calculates the contact temperature rise, contact resistance increase, and insulation damage according to equations (21) to (24); the data-driven channel learns the degradation sequence and outputs it according to equations (25) to (29). The fusion channel outputs according to equations (30) and (31). This structure enables remaining lifetime prediction to have clearly defined inputs, processing procedures, model structure, and output rules.

[0080] like Figure 9 As shown, the risk-energy efficiency-lifespan collaborative control process is used to handle conflicts between safety-oriented and energy-saving actions. The collaborative control center first reads... Figure 1 The unified state vector, then by Figure 2 To obtain a risk level, by Figure 3 Obtain energy efficiency candidate control quantities from Figure 8 Obtain the lifetime attrition rate; by Figure 8 The mixed remaining lifetime prediction results are obtained, and the lifetime loss rate is calculated by combining the normalized damage increment in the aforementioned adjacent control cycles; when a conflict occurs between candidate actions, the comprehensive cost is calculated according to equation (34). And prioritize those that can reduce The system operates in a manner that optimizes energy efficiency under normal conditions, increases monitoring frequency and limits aggressive energy-saving actions under warning conditions, and directly implements safety protection measures under alarm and emergency conditions.

[0081] Taking a power distribution center in a large industrial park as an example, this paper demonstrates the specific engineering application of the integrated JP cabinet system of this invention in a power distribution center of an industrial park. This power distribution center serves an area of ​​approximately 10 square kilometers, connecting more than 50 industrial enterprises (including approximately 15 continuous production chemical enterprises, approximately 10 steel processing enterprises with impact loads, and approximately 25 precision electronics manufacturing enterprises), with a total connected load capacity of 200 MVA and an annual electricity consumption of approximately 1.2 billion kWh. The technical requirements of the project are: the power supply reliability protocol of the power distribution center requires no less than 99.99% (i.e., annual cumulative power outage time not exceeding 52.56 minutes), a comprehensive energy efficiency of no less than 95%, and a power factor of no less than 0.95.

[0082] The system configuration and implementation instructions are as follows, and its system deployment structure corresponds to... Figure 1 The plate thickness optimization process corresponds to Figure 7 Health management closed loop corresponds to Figure 8 The process of resolving the conflict between safety and energy efficiency corresponds to Figure 9 : In terms of safety protection, this power distribution center is equipped with a complete five-layer protection scheme. For structural protection, the cabinet panels are made of 304 stainless steel with a thickness of 2.5 mm (based on an assessment of the salt spray and dust corrosion rates in this industrial area, the 2.5 mm panel will not thin by more than 0.5 mm due to corrosion over its expected 25-year service life, providing a sufficient safety margin). The surface coating process is as follows: pretreatment involves sandblasting to Sa2½ grade, with a roughness Ra of 6.3 to 12.5 microns, providing an anchor-type surface for mechanical paint adhesion; primer is applied using high-pressure airless spraying of epoxy zinc-rich paint, with a dry film thickness of 80 microns; intermediate coat is applied using high-pressure airless spraying of epoxy micaceous iron oxide paint, with a dry film thickness of 120 microns; topcoat is applied using high-pressure airless spraying of aliphatic polyurethane paint, with a dry film thickness of 80 microns, and the color selected is RAL 7035 light gray; the total dry film thickness is 280 microns. After painting, the cross-cut adhesion test was conducted according to ISO 2409 standard, and the test result reached level 0 (the cut edge was completely smooth and there was no peeling); the pull-off adhesion test was conducted according to ISO 4624 standard, and the adhesion strength was not less than 8 MPa.

[0083] The grounding system for electrical protection adopts the TN-S standard. The neutral conductor (N) and protective earth (PE) are separated at the neutral point on the low-voltage side of the distribution transformer. The PE conductor uses a 50mm x 5mm copper busbar that runs through the entire cabinet and is welded to the building's steel reinforcement grounding grid using hot-dip galvanized 40mm x 4mm flat steel. The weld length is twice the width of the flat steel (80mm), and the weld joint is treated with heavy-duty anti-corrosion. The grounding resistance, measured using the three-electrode method, is 0.38 ohms, meeting the design target value of 0.5 ohms. The selective coordination of the three levels of leakage current protection is achieved through the coordinated design of the current-time characteristic curves: the Joule integral withstand margin of the upper-level 300mA / 0.3s protection is set to be more than twice the Joule integral of the lower-level 100mA / 0.2s and 30mA / 0.1s protection, ensuring that the upper-level protection does not malfunction when the lower-level protection operates.

[0084] The fire protection system divides a 2000mm high cabinet into three independent fire-resistant isolation zones at 750mm and 1500mm along the height. The fire-resistant partitions have a sandwich structure: a 50mm thick ceramic fiberboard in the center and 2mm thick hot-dip galvanized steel plates on both sides. The gaps between the partitions and the inner walls of the cabinet are filled with intumescent fire-resistant sealant. Aerosol extinguishing devices are installed independently in each fire-resistant unit. The extinguishing agent tank capacity is calculated at 100g / m³ per unit volume. The nozzles feature a 360-degree omnidirectional spray design to ensure uniform diffusion of extinguishing agent particles throughout the unit.

[0085] In terms of energy efficiency optimization, according to Figure 3 The multi-objective energy efficiency optimization process shown has the following core configuration for the primary system design: two SCBH15-2000 / 10 type amorphous alloy transformers (rated 2000 kVA, voltage ratio 10 / 0.4 kV, connection group Dyn11, impedance voltage 6%). Under normal conditions, each transformer operates at approximately 50% load (total load approximately 2000 kVA). In the event of a single transformer failure, the other transformer can operate with the full load (load rate approximately 100%), complying with the N-1 power supply safety standard. The transformers use amorphous alloy cores, with a no-load loss of 1200 W, which is approximately 48% lower than the 2300 W no-load loss of a Class 1 energy-efficient silicon steel transformer of the same capacity, and approximately 65% ​​lower than the 3450 W no-load loss of a Class 3 energy-efficient silicon steel transformer. The load loss (at a reference temperature of 75℃) is 15500 W, which is 11% lower than the load loss limit of 17500 W for a 2000 kVA transformer specified in the national standard GB 20052-2020.

[0086] The power quality management system is configured with the following capacity: SVG capacity of 800 kVAR (based on previous measured data showing that the reactive power demand fluctuation range of this distribution center is inductive from 300 kVAR to 700 kVAR, 800 kVAR is selected to cover the upper limit of the fluctuation plus a 10% margin), APF rated compensation current of 600 amps (based on the measured total value of harmonic current minus the 5th harmonic, approximately 520 amps, 600 amps is selected), and a total installed capacity of 600 kVAR capacitor banks (divided into unequal capacity groups of 50,000 VAR × 4 groups + 100 kVAR × 4 groups). The operational effectiveness of the power quality management system was compared using online records from the same monitoring point before and after commissioning: before the system was implemented, the total harmonic distortion rate of the 10 kV bus voltage was approximately 5.8% (exceeding the limit of less than 4% for the total harmonic distortion rate of the 10 kV bus voltage specified in GB / T 14549-93), which decreased to approximately 2.1% after the system was implemented; the power factor improved from an average of 0.82 before the system to an average of 0.98 after the system.

[0087] The energy efficiency monitoring system uses a power quality analyzer with Class A harmonic measurement capabilities as its core equipment. It is installed at 28 monitoring points on the low-voltage side of the transformer's main incoming cabinet, the main outgoing feeder cabinets, and the feeder cabinets of key equipment (such as feeders for large motors and frequency converters). Monitoring data is uploaded to the energy efficiency management server via Ethernet. The server runs energy efficiency analysis software, calculating the power consumption of each feeder at 15-minute intervals and the load rate and efficiency of each transformer at 1-hour intervals.

[0088] The operational effectiveness of this embodiment was verified through 12 months of continuous monitoring. The main operational performance indicators are as follows: system availability reached 99.998% (excluding 4 planned maintenance shutdowns totaling 8.2 hours within 12 months, no unplanned shutdowns occurred); the weighted average actual operating efficiency of the transformer within the load range of 35% to 85% was 99.12%, and the comprehensive energy efficiency (considering transformer losses, line losses, and harmonic additional losses) was approximately 96.5%, higher than the target value of 95%; the monthly average power factor was between 0.97 and 0.99, and the annual average was approximately 0.98. The power factor in all months was higher than the assessment standard of 0.95, and no power factor assessment penalties were incurred throughout the year. The response time and compensation accuracy of the SVG to load fluctuations were verified through waveform analysis: Under the test condition of direct starting of a 200 kW motor, the time required for the SVG to establish the compensation current output from detecting a sudden increase in reactive power was approximately 4.2 milliseconds. The lowest transient power factor after compensation was 0.96, and the steady-state power factor reached 0.99, with a compensation accuracy of approximately ±0.8% of the rated capacity. Three equipment anomaly warnings occurred within 12 months: one was a high temperature warning for the circuit breaker terminals (maximum temperature 78℃, below the 85℃ alarm threshold), which was found to be due to slightly loose connecting bolts; after tightening, the temperature dropped to 45℃. Two were air filter differential pressure warnings, both of which were resolved by timely replacement of the filter element. All three warnings were handled promptly and accurately, without escalating into actual faults.

[0089] The economic benefit analysis is based on the actual operating data of the above 12 months. In terms of investment costs: the purchase cost of two amorphous alloy transformers is about RMB 180,000 higher than that of silicon steel transformers of the same capacity; the total purchase and installation cost of SVG and APF equipment is about RMB 750,000; the hardware and software cost of the intelligent monitoring and control system is about RMB 320,000; and the enhanced cost of the safety protection system is about RMB 250,000. The total increase in investment is about RMB 1.5 million (compared to a power distribution system of the same scale that only meets the standard configuration requirements). In terms of operational benefits: Based on the improvement of the monthly average power factor from 0.82 to 0.98, approximately RMB 420,000 in power factor adjustment penalties were avoided throughout the year; based on the reduction in transformer losses (the annual power loss of amorphous alloy transformers is approximately 186,000 kWh less than that of silicon steel level 3 energy-efficient transformers of the same capacity), plus reactive power compensation and harmonic mitigation reducing additional line and transformer losses by approximately 65,000 kWh, the total annual power savings are approximately 251,000 kWh, which translates to approximately RMB 181,000 in electricity cost savings based on a weighted average electricity price of RMB 0.72 / kWh; based on three early warnings preventing events that could have developed into faults, it is conservatively estimated that approximately RMB 80,000 was saved in fault repair and downtime losses. The total annual operational benefits are approximately RMB 681,000. The investment payback period (static) is approximately 2.2 years (150 / 681,000 ≈ 2.2 years).

[0090] Taking the power distribution system of a commercial building as an example, this invention is demonstrated in the power distribution system of a large commercial complex. The commercial complex has a total building area of ​​approximately 200,000 square meters, including a shopping mall (approximately 120,000 square meters), Grade A office buildings (approximately 50,000 square meters), a five-star hotel (approximately 30,000 square meters), and supporting facilities. The complex's annual electricity consumption is approximately 25 million kilowatt-hours, with a peak load of approximately 18 megawatts, and a total transformer installed capacity of 25 megavolt-amperes.

[0091] The building's power distribution system faces unique challenges, including: diverse load types, with varying power quality requirements for different types of lighting, air conditioning, elevators, and commercial equipment; distinct temporal characteristics, exhibiting a typical daily load curve (rising from 10:00 AM, peaking between 2:00 and 3:00 PM, and declining after 8:00 PM) and seasonal variations (summer air conditioning loads in July and August are approximately 40% higher than in January and February); and the complex aims to achieve LEED Platinum certification for green buildings and WELL Building Standard certification, imposing stringent requirements on energy efficiency and environmental quality.

[0092] To meet the high standards for building energy efficiency, this embodiment follows... Figure 3 and Figure 6 The energy efficiency optimization and benefit assessment logic shown adopts a time-of-use energy efficiency optimization strategy. The technical logic of this strategy is based on the response mechanism of peak-valley electricity price signals: in the industrial and commercial electricity price structure of the city where the building is located, the electricity price during off-peak hours (23:00 to 7:00 the next day) is approximately 45% of the price during normal hours, and the electricity price during peak hours (9:00 to 11:30, 14:00 to 16:30, and 19:00 to 21:00) is approximately 180% of the price during normal hours. The optimization objective of the strategy can be expressed as: minimizing the total cost of electricity purchase through load shifting and energy storage charging and discharging scheduling, while meeting building comfort and operational constraints. The decision variables of the strategy include: the temperature setpoint of the air conditioning system and the pre-cooling / preheating start-up time (utilizing the thermal inertia of the building envelope), the operating time of transferable loads (such as some hot water preparation and laundry equipment), and the charging and discharging plan of the energy storage system (such as the energy storage battery pack configured in electric vehicle charging facilities).

[0093] The specific operational strategies are broken down as follows: Off-peak period strategy: During off-peak hours, the energy storage system charges at its maximum allowable charging power (approximately 1.5 MW) to bring the battery charge level to at least 90%; the air conditioning system pre-cools or preheats all areas of the building to a comfortable range of 20-24℃ (winter / summer), significantly reducing the operating power of the air conditioning load during peak hours. Normal period strategy: All equipment operates in economic mode, with air conditioning temperature settings maintained at 24-26℃, fresh air volume adjusted as needed based on indoor CO2 concentration (target concentration below 800ppm), lighting dimmed in stages according to natural light intensity and occupant activity, reactive power compensation and harmonic control equipment operating normally, and the power factor of all feeders maintained at at least 0.97. Peak period strategy: The energy storage system discharges at a power level not exceeding 1.2 MW, partially replacing grid power and reducing peak-hour electricity purchases; air conditioning temperature settings are moderately adjusted to 26℃ (summer) or 20℃ (winter), leaning towards energy saving within the human comfort range; all transferable loads are scheduled to operate during off-peak hours.

[0094] The load classification management scheme categorizes all loads in the power distribution system into two dimensions: power supply reliability and control flexibility. The power supply reliability dimension is divided into three levels: "uninterruptible," "allowable short-term interruption," and "allowable planned interruption." The control flexibility dimension is divided into three levels: "non-adjustable," "downward adjustable," and "bidirectionally adjustable." Loads are mapped into three categories: Critical non-adjustable loads (fire pumps, security systems, emergency lighting, elevator traction machines, etc.) must be supplied with power under all circumstances and do not participate in demand response or load regulation; their feeders are connected to a dual-circuit power supply system with uninterruptible power supply protection. Important adjustable loads (air conditioning units, chilled water pumps, cooling tower fans, and chilled water circulating pumps, etc.) have high power supply reliability requirements but are adjustable; they are the main participants in demand response, and their operating power can be adjusted upwards or downwards within a safe allowable range. General adjustable loads (charging piles, landscape lighting, some hot water preparation equipment, etc.) have a high tolerance for power supply reliability and are demand response resources with high flexibility adjustment potential.

[0095] The building energy efficiency management system is built upon the hardware network architecture of the building automation system. The data acquisition layer obtains real-time data through sensing devices such as 28 multi-functional power quality meters, temperature sensors distributed in approximately 800 rooms, and CO2 concentration sensors distributed in major public areas. The data transmission layer aggregates data from each floor and area to the building management server via the BACnet / IP building communication protocol. The analysis and decision-making layer runs energy efficiency analysis and optimization software. Core functions include: calculating the energy efficiency baseline (using a regression model based on historical data, with outdoor temperature and building occupancy rate as independent variables and building energy consumption as the dependent variable, establishing an hourly energy consumption baseline model); real-time energy consumption deviation analysis (decomposing the deviation between current energy consumption and the baseline, and the causes of the deviation); and generating and pushing optimization strategies (generating suggested operating strategies for the next 24 hours based on electricity price periods, weather forecasts, and building occupancy reservation information).

[0096] The operational effectiveness of this embodiment was evaluated using complete data from 12 months after commissioning. The main energy efficiency results are as follows: Annual total electricity consumption before the upgrade was 25.43 million kWh, and after the upgrade it was 19.26 million kWh, saving 6.17 million kWh, a saving rate of 24.3%. Total electricity costs (including basic electricity costs, energy consumption costs, and power factor adjustment costs) decreased from RMB 18.68 million / year before the upgrade to RMB 13.23 million / year, saving RMB 5.45 million, a saving rate of 29.2%. The annual average power factor increased from 0.83 to 0.98, and the annual power factor adjustment cost changed from a penalty of RMB 420,000 to a reward of approximately RMB 80,000 (the power company provides a reward for power factors exceeding 0.95), resulting in an annualized net benefit of approximately RMB 500,000. Regarding power quality, the 10 kV bus voltage qualification rate increased from 99.2% to 99.9%, and the total harmonic distortion rate decreased from approximately 6% to approximately 2.5%, both within the standard limits. Regarding the comfort of the built environment, according to statistics on temperature complaints received by the property management, the average monthly complaints were about 35 before the renovation, and about 8 after the renovation, a reduction of 77%. After the renovation, the annual average concentration of indoor PM2.5 was 28 micrograms per cubic meter, which is about 55% lower than the annual average outdoor value, and better than the WELL certification requirement of an annual average indoor PM2.5 concentration of less than 35 micrograms per cubic meter.

[0097] The economic benefit analysis is based on 12 months of operational data. The total incremental investment in the power distribution system upgrade is approximately RMB 4.8 million, with transformer upgrades (replacing four original silicon steel transformers with amorphous alloy transformers), power quality improvement equipment purchase and installation, and energy efficiency management system hardware and software procurement, deployment, and commissioning costs each accounting for approximately one-quarter. The annual incremental operational revenue is approximately RMB 5.87 million, including RMB 5.45 million in electricity cost savings and RMB 420,000 in power factor cost improvement benefits. The static payback period for the incremental investment, converted from the incremental investment to the annual incremental operational revenue, is approximately 9.8 months, meaning the entire incremental investment can be recovered in less than one year. Considering that the building's entire lifespan exceeds 30 years, and the expected lifespan of the power distribution system equipment is approximately 20 to 25 years, it will continuously generate net revenue exceeding the book depreciation cost throughout the equipment's lifespan. The certification report specifically mentions the contribution of power distribution system energy efficiency optimization and power quality improvement measures to the building's overall energy efficiency score.

[0098] Taking a smart grid distribution substation as an example, this paper demonstrates the application of the integrated JP cabinet system of this invention in a distribution substation of a smart grid demonstration project in a city. This distribution substation is a typical new urban area, serving an area of ​​approximately 2.5 square kilometers, with connected users including approximately 3,800 residential households, approximately 200 small and medium-sized businesses, and approximately 30 light industrial enterprises. The substation is equipped with a 1.2 MW rooftop distributed photovoltaic power generation system, a 1.0 MW / 2.5 MWh lithium iron phosphate battery energy storage system, and 20 electric vehicle charging piles with a total power of approximately 1.4 MW (covering both AC slow charging and DC fast charging). The total peak load of the substation is approximately 4.5 MW, and the total transformer capacity is 6.3 MVA.

[0099] The core technical challenges faced by the demonstration project are: the high penetration rate of distributed photovoltaic power generation causes significant power flow reversal and voltage over-limit risks (the photovoltaic output may reach more than 150% of the load during the midday period on sunny days, and a large amount of reverse power transmission causes the voltage in the distribution area to rise, and the voltage of some end users may exceed 1.07 times the nominal voltage); the short-term drastic fluctuations in photovoltaic output (due to cloud movement and shading, the photovoltaic output may drop from 80% to 20% of the installed capacity within 1 minute, and vice versa), causing frequent voltage fluctuations in the distribution area; the strong randomness and peak superposition effect of electric vehicle charging load (the concentrated charging peak during the evening rush hour) makes the original distribution transformers in the distribution area face serious short-term overload risks.

[0100] To address the aforementioned challenges, this embodiment... Figure 9Based on the collaborative control center shown, the system deeply integrates the coordinated control functions of power generation, grid, load, and energy storage. The design logic of the technical solution is as follows: using the energy storage system as the core buffer, leveraging its rapid charging and discharging response capability (response time less than 20 milliseconds) to smooth out short-term fluctuations in photovoltaic output; using the orderly scheduling of electric vehicle charging and the flexible control of distribution area loads as auxiliary means, transferring some of the transferable electricity demand from peak system periods to periods with sufficient photovoltaic output or low system load; and coordinating the reactive power and voltage control strategies of distribution areas through information interaction with the upper-level distribution automation system.

[0101] The algorithm layer of the coordinated control of power generation, grid, load, and energy storage is built on a model predictive control framework. The predictive model includes three sub-models: a photovoltaic ultra-short-term output prediction model, which predicts the photovoltaic output of the distribution area within the next 0 to 30 minutes based on sky cloud images taken by an all-sky imager, through image processing and cloud tracking algorithms, with a time resolution of 1 minute and a prediction error of approximately 5% to 8% of the installed capacity; a load ultra-short-term prediction model, which predicts the power demand of various loads within the next 0 to 1 hour based on the time-series decomposition of historical loads (decomposed into periodic, trend, and random components) and online updates, with a time resolution of 5 minutes; and an energy storage state of charge prediction model, which predicts the state of charge, available charging capacity, and available discharging capacity of the energy storage system at various points in the future based on the equivalent circuit model of the energy storage system and the simulation of the charging and discharging plan.

[0102] The objective function of MPC is a multi-objective weighted form, where each term represents the total electricity cost of power exchange between the distribution area and the upstream grid, the penalty term for node voltage deviation within the distribution area, the battery capacity degradation caused by the charge-discharge cycle of the energy storage system, and the penalty term for forced reduction in photovoltaic power output due to inability to absorb power. The four weight coefficients can be flexibly configured according to the dispatch strategy. Optimization variables include the energy storage charge-discharge power every 15 minutes over the next 24 hours, the charging power of dispatchable electric vehicle charging piles, the power adjustment amount of adjustable loads, and the compensation capacity of reactive power compensation devices. The optimization solution is executed on a rolling basis every 15 minutes, issuing dispatch instructions only for the first 15 minutes of the future to each executing device, and then updating again in the next cycle to address the cumulative impact of prediction errors.

[0103] The corresponding MPC objective function is shown in equation (35): (35) In equation (35), F is the rolling optimization objective of the source-grid-load-storage model predictive control; The difference between the cost of purchasing electricity and the revenue from selling electricity generated from the exchange of power between the distribution area and the upper-level power grid; This is a penalty for node voltage deviation from the nominal value; This refers to the energy storage cycle life loss item; This is a penalty for abandoning light; to The corresponding weights are set according to voltage quality, light curtailment constraints, and battery life priority in each rolling optimization cycle.

[0104] The specific implementation of orderly charging scheduling for electric vehicles in this embodiment is as follows: The charging pile management system provides electric vehicle users with a mobile application, allowing users to set their vehicle pickup time and desired charging level. Based on the total charging demand and available charging time window of all connected vehicles, the system dynamically allocates charging power with the constraints of preventing transformer overload and maximizing local photovoltaic (PV) consumption, while aiming to minimize total charging costs. When the PV output in the distribution area is sufficient, a large number of charging vehicles are allocated higher charging power; when the load in the distribution area is at its peak and PV output is insufficient, the overall charging power is reduced, but each vehicle is still guaranteed to be fully charged before the set pickup time. According to statistics from the six-month operation of this demonstration project, through orderly charging scheduling, the peak charging load decreased from approximately 1.2 MW during disordered charging to approximately 0.8 MW (a reduction of approximately 33%), preventing transformer overload caused by the superposition of charging peaks; the local PV consumption rate (the proportion of PV power generation immediately consumed within the distribution area to the total PV power generation) increased from approximately 78% during disordered charging to approximately 92%.

[0105] The key implementation points of distributed energy management in this embodiment are as follows: In addition to maximum power point tracking (MPPT) functionality, the photovoltaic power control inverter is also equipped with reactive power-voltage droop control based on grid voltage and power rate of change limiting functionality. The reactive power-voltage droop control works by automatically absorbing inductive reactive power (the amount absorbed is determined based on the droop coefficient) when the inverter detects that the grid connection voltage is higher than a set reference value (e.g., 1.03 pu), using the voltage drop caused by the absorbed reactive power to suppress further voltage increases. The power rate of change limiting function limits the rate of increase and decrease of photovoltaic output power to no more than 10% / minute of the installed capacity, buying time for the energy storage system to respond and preventing the power grid from being impacted by sudden power drops caused by cloud cover. In this embodiment, the energy storage system is configured with a multi-scenario composite operation strategy: during the daytime when photovoltaic output is abundant, the energy storage system operates in a "consumption priority" mode, charging and storing the excess photovoltaic power (the surplus after deducting immediate load and charging load); during the evening peak load period, the energy storage system discharges in a "peak shaving" mode, replacing part of the grid's power purchase, while reserving approximately 30% of the target state of charge after discharge to provide short-term support in case of possible grid emergencies at night. Based on six months of operational data, the energy storage system completes an average of approximately 1.2 equivalent full charge-discharge cycles per day. According to the cycle life curve of lithium iron phosphate batteries (approximately 6500 cycles at 80% depth of discharge), the estimated calendar life of the energy storage system is approximately 12 to 15 years.

[0106] The operational data and performance evaluation of this embodiment in the demonstration project over six months are as follows: Power supply reliability in the transformer area: No unplanned power outages occurred within six months, achieving a 100% power supply reliability rate. Voltage quality: Based on the 10-minute statistical values ​​continuously recorded by the power quality monitoring terminal, the voltage qualification rate (the proportion of time with voltage values ​​between 0.93 pu and 1.07 pu) of all monitoring nodes in the transformer area increased from approximately 97.5% before the upgrade to over 99.8%; voltage flicker value... The daily 95% probability maximum value of (short-term flicker severity) decreased from approximately 1.1 to approximately 0.6, far below the limit of 1.0. Photovoltaic grid integration rate and carbon emissions: The on-site grid integration rate of photovoltaic power generation reached approximately 92% (including electricity transferred and integrated through energy storage), a significant improvement compared to 78% under the disordered mode; the six-month curtailment loss decreased from approximately 180,000 kWh to approximately 45,000 kWh; equivalent to a six-month reduction in carbon emissions of approximately 1,400 tons.

[0107] Economic Benefit Analysis: The incremental investment in the distribution system of the transformer substation is approximately RMB 3.65 million, mainly including RMB 2.4 million for the purchase and installation of the energy storage system, approximately RMB 600,000 for the intelligent upgrade of charging facilities and the orderly charging management system, approximately RMB 400,000 for the construction of the source-grid-load-storage coordinated control system and communication system, and approximately RMB 250,000 for the multi-functional integrated upgrade of the JP cabinet. Based on six months of operating data, the annual revenue is extrapolated as follows: the economic loss from curtailment due to increased photovoltaic grid integration is approximately RMB 270,000 per year; the net revenue from energy storage operation (discharge revenue minus charging costs minus lifespan depreciation) based on peak shaving and valley filling and peak-valley electricity price arbitrage is approximately RMB 420,000 per year; the savings in grid connection fees due to increased photovoltaic grid integration from orderly charging is approximately RMB 150,000 per year; the power factor tariff bonus obtained by maintaining a power factor above 0.98 for a long period is approximately RMB 80,000 per year; and the investment in transformer capacity expansion projects delayed to avoid transformer overload is approximately RMB 250,000 per year when converted to an annuitable annual return. The total incremental annual revenue is approximately RMB 1.17 million. The static payback period for incremental investment is approximately 3.1 years (365 / 117). From a technology promotion perspective, this demonstration project verifies that the integrated JP cabinet system of this invention has good technical feasibility and replicability in urban distribution substations with a high proportion of distributed energy. Its technical solution and operational data provide a valuable reference for the planning and construction of similar urban new area distribution networks in China.

[0108] This invention integrates traditionally fragmented security measures into a collaborative and comprehensive protection system through five progressively layered protection levels and intelligent linkages between these levels. Compared to traditional JP cabinets with protection levels of only IP30 to IP40 and independent protective measures, this system's IP65 protection level provides dust and low-pressure water spray protection, fire resistance reaches a fire separation level with a fire resistance limit of not less than 2 hours, and seismic performance meets the earthquake protection requirements for areas with an intensity of 8 degrees. This integrated security system significantly improves the safety of JP cabinets in various application scenarios, including harsh industrial environments, high-salt-fog coastal environments, and densely populated public buildings.

[0109] Through the combined effects of three technical aspects—the selection of high-efficiency electrical components (reducing no-load losses of amorphous alloy transformers by approximately 60% to 70%), comprehensive dynamic power quality management (improving the power factor to above 0.96 and harmonic mitigation rate to no less than 95%), and intelligent optimization control algorithms—a comprehensive improvement in system performance and efficiency is achieved. Under the load rate, harmonic content, and power factor conditions listed in the examples, compared to a reference JP switchgear system using traditional silicon steel transformers and only performing basic reactive power compensation, the overall system energy efficiency can be increased from approximately 88% to 92% to approximately 96% to 98%; the actual energy saving is related to the load curve, compensation capacity, and operating strategy.

[0110] By combining standardized electrical interfaces, RFID smart identification, and automatic configuration technology, a truly plug-and-play modular installation method has been achieved in the JP cabinet field. Compared to traditional modular installations that require on-site wiring and manual parameter settings by technicians, the average replacement time for a single functional module has been reduced from 60 minutes to less than 5 minutes. The automatic configuration system eliminates the risk of errors caused by manual parameter settings.

[0111] By employing condition monitoring, fault diagnosis based on hybrid methods, life prediction combining physical and data-driven models, and optimized maintenance decisions, a shift from traditional planned maintenance to predictive maintenance based on the actual health status of the equipment has been achieved. Under the implemented operating conditions and maintenance strategies, the mean time between failures (MTBF) can be increased from approximately 30,000 hours to the order of 50,000 hours, extending the overall lifespan of the equipment; the specific improvement depends on the load spectrum, environmental stress, and maintenance execution.

[0112] Given load, electricity price, maintenance costs, and equipment configuration, the static payback period for incremental investment (compared to a distribution system that only meets standard configuration requirements) can be between 1 and 3 years. Annual electricity cost savings and carbon emission reductions per unit vary with capacity, operating hours, local electricity price, and grid emission factor.

[0113] This integrated JP cabinet system can be widely used in power plants, substations, and distribution rooms in the power industry; steel, chemical, and manufacturing industries in the industrial sector; commercial buildings and residential areas in the construction sector; transportation hubs and hospitals in the infrastructure sector; distributed photovoltaic and energy storage distribution substations in the new energy sector; and power distribution facilities in special environments such as coastal areas and high altitudes. The system can meet voltage level requirements from 0.4 kV to 35 kV, is suitable for various power distribution equipment with capacities from 100 kVA to 5000 kVA, and can operate normally under ambient temperatures from -20℃ to 60℃ and relative humidity from 10% to 95%. The system's communication capabilities support mainstream communication protocols such as Modbus, Profibus, CAN, Ethernet, Wi-Fi, and mobile communication. The localization rate of key components exceeds 85%. The estimated annual market demand is 150,000 to 250,000 units, with a market size of approximately 3 billion to 5 billion yuan.

[0114] Example 2: This embodiment provides a comprehensive distribution cabinet management system based on multiple protections and energy efficiency optimization, including: The data processing module is configured to: acquire the safety status, energy efficiency status, module topology status, and health status of the AC low-voltage integrated distribution cabinet; construct a unified state vector including safety status, energy efficiency status, module topology status, and health status; specifically, regarding safety status, construct a multi-layered progressive collaborative architecture including structural safety protection layer, electrical safety protection layer, fire safety protection layer, environmental safety protection layer, and intrusion safety protection layer; regarding energy efficiency status, select electrical components at the source, perform dynamic power quality management to eliminate additional power losses in the intermediate stage, and dynamically adjust strategies in the control stage. The constraint determination module is configured to determine the safety risk margin, loss gradient, and lifetime loss rate based on the unified state vector. The collaborative control module is configured to perform collaborative control based on safety risk margin, loss gradient, and lifetime loss rate, prioritizing safety constraints.

[0115] The working method of the system is the same as that of the integrated management method of distribution cabinet based on multiple protection and energy efficiency optimization in Embodiment 1, and will not be repeated here.

[0116] Example 3: This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in Embodiment 1.

[0117] Example 4: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, it implements the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in Embodiment 1.

[0118] Example 5: This embodiment provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization described in Embodiment 1.

[0119] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A power distribution cabinet integrated management method based on multiple protection and energy efficiency optimization, characterized in that, include: Acquire the safety status, energy efficiency status, module topology status, and health status of the AC low-voltage integrated switchgear; construct a unified state vector including safety status, energy efficiency status, module topology status, and health status; in terms of safety status, construct a multi-layered progressive collaborative architecture including structural safety protection layer, electrical safety protection layer, fire safety protection layer, environmental safety protection layer, and intrusion safety protection layer; in terms of energy efficiency status, select electrical components at the source, perform dynamic power quality management to eliminate additional power losses in the middle stage, and dynamically adjust strategies in the control stage; The safety risk margin, loss gradient, and lifetime loss rate are determined based on the unified state vector. Based on safety risk margin, loss gradient, and lifetime loss rate, coordinated control is carried out with safety constraints as the priority.

2. The power distribution cabinet integrated management method based on multiple protection and energy efficiency optimization according to claim 1, characterized in that, The unified state vector is: ； in, Indicates safety features related to temperature, smoke, access control, vibration, insulation, and leakage current; It represents the energy efficiency characteristics of active power, reactive power, power factor, harmonics, voltage deviation, and line loss; This indicates the module type, rated parameters, interface status, and communication address characteristics. It represents contact resistance, insulation resistance, number of switching operations, hot spot temperature, partial discharge count, and health life characteristics.

3. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 1, characterized in that, The objective function and security constraints of the coordinated control are: ； in, The control variables include static var generator output, active power filter compensation current, capacitor switching, temperature control setpoint, load derating, and maintenance actions. , and These are energy consumption targets, lifespan loss targets, and operating cost targets; The security risk value after fusion. To allow for a risk ceiling.

4. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 3, characterized in that, The security risk value after fusion is: ； Among them, N, P, A, and E represent normal, early warning, alarm, and emergency states, respectively; , , , These are the scalarized coefficients for each risk state.

5. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 4, characterized in that, Security risk assessment will output fuzzy reasoning With Bayesian network output Perform DS evidence fusion; assuming the identification framework is normal (N), warning (P), alarm (A), and emergency (E), the fusion quality function is: ； First, determine the overall cost of each candidate action, then select the action with the lowest overall cost: ； in, Indicates the total cost of the action. Indicates the cost of the action. Indicates taking action Penalties for exceeding limits afterward This indicates the additional penalty imposed by the action.

6. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 1, characterized in that, The structural safety protection layer includes: a double-sealed structure between the cabinet door and the cabinet body; a multi-layer sealed cable connector for the cable inlet; a labyrinth-style waterproof ventilation structure for the ventilation and heat dissipation vents; vibration-damping rubber supports installed at the bottom of the cabinet body; and the determination of the thickness of the outer shell includes: Determine the bending stiffness of cabinet doors or side panels : ； For plate regions with four-sided support or boundaries containing reinforcing ribs, the maximum equivalent stress and maximum deflection According to boundary coefficient and Sure: ； Strength constraints and stiffness constraints The lower limit of plate thickness is obtained: ； equivalent width of impact The design verification of local plastic energy dissipation and energy absorption capacity was carried out. and the lower limit of impact constraint plate thickness for: ； For coastal salt spray, chemical dust, or high humidity environments, by superimposing the full-life corrosion thinning, negative tolerance of the sheet metal, and welding heat deformation allowance, the lower limit of corrosion and manufacturing constraints is obtained: ； Select the minimum thickness that satisfies all constraints from the standard plate thickness sequence: ； in, For thickness The bending stiffness of the lower part; and These are the long and short sides of the equivalent plate region, respectively; ν is the elastic modulus; ν is Poisson's ratio; This is the equivalent surface load; , These are the boundary factors for stress and deflection, respectively. Yield strength; and These are allowable stress and allowable deflection, respectively. The equivalent width for impact effect; To design impact energy; The effective thickness that needs to be retained even at the end of its service life; Annual corrosion thinning rate; Design life; and These are the negative tolerance of the sheet metal and the welding heat deformation allowance, respectively. To meet the lower limit of the process for installing sealing flanges, hinges, and locking components; For procureable standard plate thickness series; To finalize the selection of the plate thickness.

7. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 1, characterized in that, When the strategy is dynamically adjusted, the objective function is composed of energy consumption, power quality, lifespan loss, and operating cost: ； in, For energy consumption items, For power quality, This is a lifespan depletion item. The cost is the operating cost item; the optimization constraints include harmonic constraint indicators and lifetime loss rate constraint indicators.

8. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 1, characterized in that, Regarding the module topology status, a compatibility score is performed: ； When compatibility rating Power is prohibited when the current is below the preset threshold; otherwise, protection settings are generated based on the module's rated current, backplane bus capacity, and contact temperature rated curve. ； The overload protection inverse time curve is as follows: ； in, Rate the module's compatibility. For the first Class compatibility input; This is the individual score after mapping the input; For single-item weights; This is the module's rated current. This refers to the capacity of the backplane busbar; For temperature Lower contact allowable current; , and This is the reduction factor; To protect the set value; For current The tripping time is specified below; A, B, and p are parameters of the inverse time curve.

9. The integrated management method for distribution cabinets based on multiple protections and energy efficiency optimization as described in claim 1, characterized in that, The prediction of the health status includes: acquiring current, voltage, hotspot temperature, humidity, total harmonic distortion of current, contact resistance, insulation resistance, number of switching actions, vibration intensity, and partial discharge count to construct an input vector; robustly normalizing outliers using median absolute deviation; and determining the hybrid remaining lifetime. ； ； ； ； in, This is insulation damage; This is the failure threshold; , , and , representing the prediction error and uncertainty of the physical channel and the data channel, respectively; c is the context vector; , For fusion weights; constant; The bias term for the data-driven remaining lifetime output layer; Transpose of the data-driven remaining lifetime output layer weight vector.

10. A comprehensive management system for distribution cabinets based on multiple protections and energy efficiency optimization, characterized in that: include: The data processing module is configured to: acquire the safety status, energy efficiency status, module topology status, and health status of the AC low-voltage integrated distribution cabinet; construct a unified state vector including safety status, energy efficiency status, module topology status, and health status; specifically, regarding safety status, construct a multi-layered progressive collaborative architecture including structural safety protection layer, electrical safety protection layer, fire safety protection layer, environmental safety protection layer, and intrusion safety protection layer; regarding energy efficiency status, select electrical components at the source, perform dynamic power quality management to eliminate additional power losses in the intermediate stage, and dynamically adjust strategies in the control stage. The constraint determination module is configured to determine the safety risk margin, loss gradient, and lifetime loss rate based on the unified state vector. The collaborative control module is configured to perform collaborative control based on safety risk margin, loss gradient, and lifetime loss rate, prioritizing safety constraints.

Images