Zero-carbon park inter-industry information fusion operation and maintenance system

Abstract

The invention discloses a zero-carbon park inter-industry information fusion operation and maintenance system, particularly relates to the technical field of carbon emission dynamic management and control, and is used for solving the problem of carbon accounting distortion caused by state deviation of an existing static carbon emission model and a physical system. The method comprises the following steps: establishing an initial carbon emission model containing energy efficiency reference parameters in a design stage, adopting multi-working-condition dynamic calibration in a construction stage to obtain an operation boundary characteristic spectrum and synchronously acquiring a thermodynamic trace, and identifying topological fragile nodes based on Lyapunov exponential stability analysis and optical flow method thermal gradient analysis; quantizing a carbon emission cascade conduction effect by combining an equipment energy topology network, generating a dynamic carbon emission model correction instruction to update an energy efficiency parameter, and finally executing double-instruction collaborative management and control in an operation and maintenance stage; a complete-cycle carbon data flow closed loop is realized, and through multi-source information fusion and conduction chain modeling, the effect of continuously reducing the carbon emission intensity of the park is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic carbon emission control, and more specifically, to an information fusion operation and maintenance system for industries in a zero-carbon park. Background Art

[0002] The construction and operation and maintenance of a zero-carbon park involve the coordination of multiple stages such as design, construction, and operation. Currently, the industry generally adopts the EPC (Engineering Procurement Construction) model to integrate the entire industrial chain and realizes carbon emission control through technologies such as BIM modeling and Internet of Things monitoring. Existing technologies usually establish a static carbon emission model in the design stage (such as fixed equipment energy efficiency parameters), install equipment according to the drawings in the construction stage, and conduct carbon accounting based on a preset model in the operation and maintenance stage, forming a one-way data flow.

[0003] Existing technologies have the defect of broken carbon data flow in the whole EPC cycle: the static carbon model in the design stage cannot dynamically absorb changes in equipment parameters in the construction link (such as energy efficiency upgrade or downgrade), resulting in a continuous deviation between the carbon accounting benchmark in the operation and maintenance stage and the true state of the physical system, leading to systematic carbon emission distortion, and the error accumulates and expands with equipment aging or transformation, seriously restricting the accurate achievement of the zero-carbon goal. Summary of the Invention

[0004] In order to overcome the above defects of the existing technology, the present invention provides an information fusion operation and maintenance system for industries in a zero-carbon park to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: An information fusion operation and maintenance system for industries in a zero-carbon park, comprising: An initial modeling module, used to establish an initial carbon emission model in the design stage, and the initial carbon emission model includes the energy efficiency benchmark parameters of park equipment; A dual-source acquisition module, used to obtain the operation boundary characteristic spectrum of park equipment through multi-condition dynamic calibration in the construction stage, and synchronously collect the infrared thermal imaging time series of the equipment heat dissipation system to generate a thermodynamic trace; A topology identification module, used to analyze the energy efficiency stability of the operation boundary characteristic spectrum based on the Lyapunov exponent, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace to identify the topological vulnerable nodes of park equipment in the energy network; A cascade evaluation module, used to evaluate the carbon emission cascade conduction effect of parameter changes of topological vulnerable nodes on associated equipment when topological vulnerable nodes are identified; A model update module, used to generate a dynamic carbon emission model correction instruction according to the carbon emission cascade conduction effect, update the energy efficiency benchmark parameters and the associated equipment parameters of topological vulnerable nodes, and form a dynamic carbon emission model; A carbon emission control module, used to perform park carbon emission control operations based on the dynamic carbon emission model in the operation and maintenance stage.

[0006] Further, an initial carbon emission model is established in the design stage. The initial carbon emission model includes the energy efficiency benchmark parameters of the park equipment, including: Construct an equipment energy topology network based on the design drawings of the park energy system; Obtain the rated energy efficiency parameters of the park equipment from the equipment supplier database; Bind the rated energy efficiency parameters to the corresponding nodes of the equipment energy topology network; Calculate the energy efficiency conduction coefficient of associated equipment through the cold, heat, and power coupling relationship of the park; Combine the equipment energy topology network, rated energy efficiency parameters, and energy efficiency conduction coefficient to generate an initial carbon emission model; Among them, the energy efficiency benchmark parameters include rated energy efficiency parameters and energy efficiency conduction coefficients.

[0007] Further, in the construction stage, the operating boundary characteristic spectrum of the park equipment is obtained through multi-condition dynamic calibration, and the infrared thermal imaging time series of the equipment cooling system is synchronously collected to generate a thermodynamic trace, including: Apply a stepped load change to the park equipment during the construction commissioning stage; Collect the dynamic response of the equipment energy efficiency parameters at different load rates through a frequency converter and a power sensor; Draw a curve of the energy efficiency parameters changing with the load rate according to the dynamic response of the energy efficiency parameters; Extract the energy efficiency inflection point interval in the change curve as the operating boundary characteristic spectrum; At the same time, use an infrared thermal imager to continuously photograph the surface temperature field of the equipment cooling system; Perform pixel-level thermal radiation intensity analysis on the time series images of the surface temperature field of the equipment cooling system; Generate a thermodynamic trace based on the migration trajectory of the thermal radiation intensity between consecutive frames.

[0008] Further, the operating boundary characteristic spectrum and the thermodynamic trace timestamp are synchronized and aligned.

[0009] Further, based on the Lyapunov exponent, analyze the energy efficiency stability of the operating boundary characteristic spectrum, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace to identify the topological vulnerable nodes of the park equipment in the energy network, including: Calculate the Lyapunov exponent based on the curve of the energy efficiency parameters changing with the load rate of the operating boundary characteristic spectrum; When the Lyapunov exponent is greater than the preset stability threshold, it is marked as the energy efficiency mutation interval; Perform optical flow method vector field calculation on the thermodynamic trace to obtain the thermal gradient vector field; Extract the spatial region where the heat flow intensity in the thermal gradient vector field exceeds the heat interaction threshold; Locate the coincidence point of the park equipment corresponding to the energy efficiency mutation interval and the equipment in the spatial area where the heat flow intensity exceeds the heat interaction threshold in the device energy topology network; Identify the device nodes that simultaneously meet the conditions of the energy efficiency mutation interval and the heat flow intensity exceeding the heat interaction threshold as topological vulnerable nodes.

[0010] Further, when topological vulnerable nodes are identified, evaluate the carbon emission cascading conduction effect of the parameter change of the topological vulnerable nodes on the associated devices, including: Determine the directly associated devices of the topological vulnerable nodes based on the device energy topology network; Calculate the primary carbon emission impact value of the parameter change of the topological vulnerable nodes on the directly associated devices according to the energy efficiency conduction coefficient in the initial carbon emission model; Trace the indirectly associated devices through the energy transfer path in the device energy topology network; Calculate the secondary carbon emission impact value of the indirectly associated devices by combining the energy efficiency conduction coefficient chain calculation; Accumulate the primary carbon emission impact value and the secondary carbon emission impact value to generate a quantitative value of the carbon emission cascading conduction effect.

[0011] Further, generate a dynamic carbon emission model correction instruction according to the carbon emission cascading conduction effect, update the energy efficiency benchmark parameters and the associated device parameters of the topological vulnerable nodes, and form a dynamic carbon emission model, including: Extract the energy efficiency parameter deviation amount of the topological vulnerable nodes from the quantitative value of the carbon emission cascading conduction effect; Calculate the parameter correction coefficient of the topological vulnerable nodes in the device energy topology network according to the energy efficiency parameter deviation amount; Locate the directly associated devices of the topological vulnerable nodes in the device energy topology network; Calculate the parameter compensation amount of the directly associated devices through the energy efficiency conduction coefficient; Convert the energy efficiency parameter deviation amount and the parameter compensation amount into a dynamic carbon emission model correction instruction; Update the energy efficiency benchmark parameters of the initial carbon emission model according to the dynamic carbon emission model correction instruction; Synchronously update the energy efficiency conduction coefficients of the topological vulnerable nodes and the directly associated devices; Combine the updated energy efficiency benchmark parameters and the energy efficiency conduction coefficients to generate a dynamic carbon emission model.

[0012] Further, perform park carbon emission control operations based on the dynamic carbon emission model during the operation and maintenance stage, including: Real-time collect the operation load rate and energy consumption data of park devices through Internet of Things sensors; Input the operation load rate into the dynamic carbon emission model to obtain the theoretical energy efficiency benchmark parameters under the current working conditions; Calculate the carbon emission deviation value between the actual energy consumption data and the theoretical energy efficiency benchmark parameters; Generate an equipment energy efficiency optimization instruction when the carbon emission deviation value exceeds the preset tolerance threshold; Determine the compensation regulation amount of the associated equipment according to the energy efficiency conduction coefficient in the dynamic carbon emission model; Send the energy efficiency optimization instruction to the topological vulnerable node device and send the compensation regulation instruction to the associated equipment.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. Achieve the continuous evolution of the carbon emission model through a multi-stage dynamic closed-loop mechanism; the initial carbon emission model in the design stage is dynamically calibrated under multiple working conditions in the construction stage to correct the energy efficiency benchmark parameters in real time, integrate multi-dimensional data of the operation boundary characteristic spectrum and the thermodynamic trace, accurately capture the deterioration of equipment energy efficiency stability and abnormal heat transfer, and break through the identification blind area of traditional single-dimensional monitoring for hidden topological vulnerable nodes; quantify the carbon emission conduction effect based on the equipment energy topology network, construct a computable model of carbon flow association between multi-hop devices, and provide a scientific basis for dynamic correction; the dual-instruction collaborative control strategy in the operation and maintenance stage realizes a qualitative leap from single-point regulation to system collaborative optimization through the parallel execution of the energy efficiency optimization instruction and the compensation regulation instruction.

[0014] 2. Establish a full-cycle carbon data flow closed-loop: the dynamic calibration data in the construction stage reversely corrects the design model, and the real-time carbon emission data in the operation and maintenance stage drives the model update forward, forming a cross-stage self-optimization mechanism; the introduction of the equipment energy topology network converts discrete equipment into a computable node network, enabling the mathematical modeling of the carbon emission conduction path; the multi-source information fusion technology based on the Lyapunov exponent and the optical flow method locates vulnerable nodes from the dual dimensions of energy efficiency stability and thermodynamic characteristics, greatly improving the identification reliability; generate parameter compensation instructions through the conduction effect quantization value to ensure the collaborative optimization of associated equipment; finally, realize the dynamic matching of the theoretical energy efficiency benchmark parameters and the actual working conditions at the operation and maintenance end. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic structural diagram of an information fusion operation and maintenance system among industries in a zero-carbon park according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment: Figure 1The structure diagram of an inter - industry information fusion operation and maintenance system for a zero - carbon park in the present invention is given. An inter - industry information fusion operation and maintenance system for a zero - carbon park includes: An initial modeling module, which is used to establish an initial carbon emission model in the design stage. The initial carbon emission model includes the energy efficiency benchmark parameters of park equipment; A dual - source acquisition module, which is used to obtain the operation boundary characteristic spectrum of park equipment through multi - condition dynamic calibration in the construction stage, and synchronously collect the infrared thermal imaging time series of the equipment cooling system to generate a thermodynamic trace; A topology identification module, which is used to analyze the energy efficiency stability of the operation boundary characteristic spectrum based on the Lyapunov exponent, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace to identify the topological vulnerable nodes of park equipment in the energy network; A cascading evaluation module, which is used to evaluate the carbon emission cascading conduction effect of parameter changes of topological vulnerable nodes on associated equipment when topological vulnerable nodes are identified; A model update module, which is used to generate a dynamic carbon emission model correction instruction according to the carbon emission cascading conduction effect, update the energy efficiency benchmark parameters and the parameters of associated equipment of topological vulnerable nodes, and form a dynamic carbon emission model; A carbon emission control module, which is used to perform park carbon emission control operations based on the dynamic carbon emission model in the operation and maintenance stage.

[0018] The initial modeling module, which is used to establish an initial carbon emission model in the design stage. The initial carbon emission model includes the energy efficiency benchmark parameters of park equipment. The specific implementation is as follows: The specific process of constructing the equipment energy topology network based on the design drawings of the park energy system is as follows: The equipment connection relationship in the design drawings is converted into a graph data structure, where park equipment such as refrigeration units, boilers, and photovoltaic inverters are used as nodes, and energy transmission media such as steam pipelines, power lines, and cooling water pipelines are used as connection edges. Each node records the unique identification code and spatial position coordinates of the equipment. The weight of each edge is determined according to the energy type. For example, the weight of a power line is the voltage level multiplied by the line length, and the weight of a thermal pipeline is the flow rate multiplied by the pipe diameter. The equipment energy topology network is stored in the form of an adjacency matrix, and the matrix element value represents the physical connection strength between equipment. The matrix element value between un - directly - connected equipment is set to zero.

[0019] The rated energy efficiency parameters of park equipment are obtained from the equipment supplier database in the following way: Access the cloud parameter library provided by the equipment manufacturer through the application programming interface, and use the equipment model as the retrieval keyword to obtain the performance parameters under rated conditions. For chillers, obtain the nominal value of the coefficient of performance (COP), for example, the COP value is 3.8; for gas boilers, obtain the rated value of the thermal efficiency η, for example, the thermal efficiency value is 92%; for photovoltaic panel arrays, obtain the power generation efficiency per unit area under standard test conditions, for example, the power generation efficiency value is 21.5%. These rated energy efficiency parameters are uniformly measured in kilojoules per kilowatt-hour and stored as a structured data table. Each record in the table contains four fields: equipment number, parameter name, parameter value, and measurement unit.

[0020] The specific operations for binding the rated energy efficiency parameters to the corresponding nodes in the equipment energy topology network include: Traverse each node in the equipment energy topology network and perform an exact match query in the equipment parameter table according to the node equipment number. When the chiller node with the equipment number CH-01 is located, its COP value, for example 3.8, is written into the energy efficiency attribute field of this node; when the gas boiler node with the equipment number GB-05 is located, its thermal efficiency value, for example 92%, is converted to the decimal number 0.92 and then stored in the corresponding field. The binding process is implemented using the hash table data structure, with the equipment number as the key and the energy efficiency parameter value as the value, ensuring fast data access with a constant time complexity.

[0021] The specific method for calculating the energy efficiency conduction coefficient of associated equipment through the park's cold, heat, and power coupling relationship is as follows: First, identify the conversion nodes in the equipment energy topology network that are simultaneously connected to multiple energy types, such as the waste heat boiler that is simultaneously connected to the steam pipe network and the generator set. Establish the equipment energy balance equation according to the law of conservation of energy. For the waste heat boiler node numbered WHB-12, its heat input comes from the exhaust heat of the gas turbine, and the outputs are the generated electricity and steam heat. The calculation formula for the energy efficiency conduction coefficient k is the sum of the output energies divided by the input energy, specifically, k is equal to (the generated electricity + steam heat) divided by the exhaust heat of the gas turbine. Among them, the generated electricity is measured in megawatt-hours, the heat is measured in gigajoules, and the dimensional consistency is achieved through the energy unit conversion coefficient of 3.6 gigajoules / megawatt-hour. The finally generated relational data table contains three columns of data: starting equipment number, ending equipment number, and energy efficiency conduction coefficient value.

[0022] The specific steps for generating the initial carbon emission model from the combined device energy topology network, rated energy efficiency parameters, and energy efficiency conduction coefficient include: performing multi-source data fusion on the device energy topology network in adjacency matrix format, the rated energy efficiency parameters stored in a hash table, and the energy efficiency conduction coefficient table in tabular form. The fusion operation uses the device number as the primary key, and through database connection operations, the energy efficiency parameters are attached to the node attributes of the device energy topology network, while the energy efficiency conduction coefficient table is used as an additional attribute of the connecting edge. The generated initial carbon emission model is stored in a graph database, where the node attributes include device type and rated energy efficiency value, and the edge attributes include energy efficiency conduction coefficient values. For example, the energy efficiency conduction coefficient field of the edge between the chiller node with device number NODE_001 and the cooling tower node with device number NODE_005 is assigned a value of 0.85.

[0023] The specific implementation of the energy efficiency benchmark parameters including the rated energy efficiency parameters and the energy efficiency conduction coefficient is as follows: The rated energy efficiency parameters are stored as static attributes in the node attribute set of the graph database. For example, the photovoltaic node with device number PV_Array_03 has an attribute item of "rated power generation efficiency", and its value is, for example, 21.5%. The energy efficiency conduction coefficient is stored as a dynamic attribute in the edge attributes. For example, the edge from the combined heat and power unit with device number CHP_Unit_07 to the heat storage tank with device number Heat_Storage_09 has an attribute item of "energy efficiency conduction coefficient", and its value is, for example, 0.93. These two types of parameters together constitute the carbon emission calculation benchmark and are stored in a hierarchical data format in the model file. The node parameter block and the edge parameter block are precisely associated and mapped through the device number and the connection identifier.

[0024] The dual-source acquisition module is used to obtain the operation boundary characteristic spectrum of the park devices through multi-condition dynamic calibration during the construction stage, and synchronously collect the infrared thermal imaging time series of the device cooling system to generate a thermodynamic trace. The specific implementation is as follows: The specific implementation process of applying a stepped load change to the park devices during the construction and commissioning stage is as follows: A stepped control signal is sent to the device drive motor through a programmable logic controller to make the device load rate change gradually according to a predetermined gradient. For example, the load rate of the chiller starts from 30% and increases step by step to 100% in steps of 10%, and each load step maintains a stable operation time of 5 minutes. The load change instruction is sent to the frequency converter through an industrial communication protocol, and the frequency converter adjusts the motor speed according to the received instruction value to achieve an accurate step change in the device output power.

[0025] The specific method for using a frequency converter and power sensor to collect the dynamic response of equipment energy efficiency parameters at different load rates is as follows: A three-phase power sensor is installed at the equipment power input to measure voltage, current, and power factor parameters in real time at a sampling frequency of 10 times per second. The frequency converter simultaneously records motor speed and output torque data. The raw measurement values are received and real-time energy efficiency calculations are performed based on the equipment type. For refrigeration equipment, the real-time coefficient of performance (COP) is calculated by the ratio of cooling capacity to input power. The cooling capacity is calculated by multiplying the refrigerant flow rate by the enthalpy difference. For water pump equipment, the efficiency value is calculated by the ratio of flow rate to input power. Dynamic response data is stored in a time series format, with each data point containing three pieces of information: the load rate value, a millisecond timestamp, and the energy efficiency parameter value.

[0026] The steps for plotting a curve of energy efficiency parameters changing with load rate based on the dynamic response of energy efficiency parameters include: extracting data corresponding to load rate and energy efficiency parameters from a time-series database, locating discrete points in a two-dimensional coordinate system with the load rate percentage as the horizontal axis and the energy efficiency parameter as the vertical axis. A curve fitting algorithm is used to process the discrete points, such as using the least squares method to fit a cubic polynomial function to generate a continuous and smooth curve. Abnormal data is filtered during the curve plotting process. When the energy efficiency parameter values of adjacent data points change by more than 20%, a data verification mechanism is triggered to confirm the actual operating conditions through the associated equipment operating status signals.

[0027] The specific operation of extracting the energy efficiency inflection point interval in the change curve as the operating boundary characteristic spectrum is as follows: calculate the first-order derivative function of the change curve and identify the load interval where the absolute value change rate of the derivative exceeds the set threshold. For example, when the derivative value change rate of the refrigeration unit COP curve suddenly increases from 0.05 / 1% to 0.15 / 1% at a load rate of 60%, the load range of 55% to 65% is determined to be the energy efficiency inflection point interval. Each inflection point interval records three characteristic parameters: the starting load rate value, the ending load rate value, and the maximum energy efficiency change rate, for example, a starting load rate of 55%, an ending load rate of 65%, and a maximum energy efficiency change rate of 0.15. The characteristic parameter set of multiple inflection point intervals constitutes the operating boundary characteristic spectrum, which is stored as a structured data table.

[0028] The technology used to continuously capture the surface temperature field of the equipment's cooling system using an infrared thermal imager involves placing fiducial markers on the surface of the equipment's heat sink, and using an infrared thermal imager mounted on a fixed bracket to capture thermal images at a rate of 5 frames per second. The thermal imager operates in the long-wave infrared range, with a temperature resolution of 0.1 degrees Celsius and a spatial resolution of 640 by 480 pixels. Each frame is timestamped with millisecond-level accuracy, and the time base is maintained consistent with the power acquisition system through a clock synchronization protocol. The acquisition process fully covers the entire step load variation process, generating a thermal imaging dataset with strict chronological order.

[0029] The specific process of pixel-level thermal radiation intensity analysis for the temporal image of the surface temperature field of the device's cooling system is as follows: Convert the thermal imaging sequence into a temperature matrix sequence, where each matrix element value represents the temperature measurement value of the corresponding pixel point. For each pixel coordinate, analyze the temperature change characteristics between consecutive frames. For example, the temperature change value between two adjacent frames at pixel coordinate row number 120 and column number 80 is 3.2 degrees Celsius. Convert the temperature value into a thermal radiation intensity value according to the physical law of thermal radiation. The conversion relationship is that the thermal radiation intensity is proportional to the fourth power of the absolute temperature. The analysis results generate a spatio-temporal change dataset of thermal radiation intensity.

[0030] The implementation method of generating a thermodynamic trace based on the migration trajectory of thermal radiation intensity between consecutive frames is as follows: In the thermal radiation intensity field of adjacent time frames, identify the movement trajectory of a specific high-temperature area through an image feature tracking algorithm. For the high-temperature spot on the heat sink, calculate the displacement vector of its centroid pixel coordinate between consecutive frames. For example, if the coordinate of the high-temperature area at time t1 is row 100 and column 200, and the coordinate at time t2 is row 102 and column 198, then the displacement vector is +2 pixels in the row direction and -2 pixels in the column direction. The displacement vectors between all consecutive frames are connected in chronological order to form the migration trajectory path of thermal radiation intensity. Each thermodynamic trace contains information such as the starting coordinate of the trajectory, the sequence of displacement vectors, and the total time length, and is stored in a structured data format.

[0031] The guarantee mechanism for synchronously aligning the operation boundary characteristic spectrum with the time stamp of the thermodynamic trace is as follows: Deploy a time synchronization server in the data acquisition network to provide a unified time reference for power sensors, frequency converters, and infrared thermal imagers. All data records are appended with Coordinated Universal Time (UTC) time stamps, and the time synchronization error is controlled within 10 milliseconds. In the data processing stage, through the time stamp matching algorithm, spatially align the operation boundary characteristic data and the thermodynamic trace data at the same time point. For example, establish an associated mapping relationship between the energy efficiency inflection point characteristic parameters at 14:35:20.500 and the thermal radiation migration trajectory at the corresponding time.

[0032] A topology recognition module, which is used to analyze the energy efficiency stability of the operation boundary characteristic spectrum based on the Lyapunov exponent, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace, and identify the topology vulnerable nodes of the park equipment in the energy network. The specific implementation is as follows: The specific implementation process of calculating the Lyapunov exponent based on the curve of the energy efficiency parameter varying with the load rate of the operating boundary characteristic spectrum is as follows: Obtain the energy efficiency parameter-load rate data sequence stored in the operating boundary characteristic spectrum, and construct a dynamic system model with the load rate as the time-equivalent variable. For the energy efficiency parameter sequence of the refrigeration unit in the load rate range of 40% to 90%, the time-delay embedding method is used to reconstruct the phase space, the embedding dimension parameter is set to 3, and the time-delay parameter is determined according to the position of the first zero-crossing point of the sequence autocorrelation function. Calculate the exponential divergence rate of adjacent phase space orbits, and the Lyapunov exponent λ is solved through the orbital evolution relationship. Specifically, when the distance between adjacent orbits evolves from the initial value d0 to the distance dt at time t, the relationship dt equals d0 multiplied by the exponential of λ times t of the natural constant e is satisfied. The calculation results of the Lyapunov exponent for each load interval are stored as numerical feature vectors.

[0033] The determination logic for marking as the energy efficiency mutation interval when the Lyapunov exponent is greater than the preset stability threshold is as follows: The preset stability threshold is set according to the equipment type classification. For example, the threshold for refrigeration equipment is set to 0.25, and the threshold for heating equipment is set to 0.30. Traverse all load intervals. If the calculation result of the Lyapunov exponent in a certain interval exceeds the threshold corresponding to the equipment type, for example, the λ value of the refrigeration unit in the load rate interval of 55% to 65% is 0.28, then mark this interval as the energy efficiency mutation interval. The marking result records three attributes: the starting load rate value, the ending load rate value, and the maximum Lyapunov exponent value, forming a structured mutation interval characteristic data table.

[0034] The specific method for obtaining the thermal gradient vector field by calculating the optical flow method vector field of the thermodynamic trace is as follows: Read the displacement vector sequence in the thermodynamic trace dataset, and establish a two-dimensional coordinate system on the equipment heat dissipation surface. Decompose the displacement vector of each pixel point into the horizontal component u and the vertical component v, and solve the motion vector field between consecutive frames through the optical flow algorithm. In the vector field calculation, the smoothing coefficient parameter is set to 0.2, the number of iterations is set to 100 times, and the convergence condition is set to that the change amount between adjacent iterations is less than 0.01 pixel per frame. The finally output thermal gradient vector field is a matrix data structure, and each element of the matrix contains the vector amplitude and the direction angle. For example, the vector amplitude at coordinate row 150 column 200 is 3.5 pixels per frame, and the direction angle is 120 degrees.

[0035] The operation of extracting the spatial regions in the thermal gradient vector field where the heat flux intensity exceeds the thermal interaction threshold includes: calculating the heat flux intensity values at each position point in the thermal gradient vector field, where the heat flux intensity is equal to the vector amplitude multiplied by the material's thermal conductivity. The thermal interaction threshold is set according to the material characteristics of the device's heat sink. For example, the threshold for an aluminum alloy heat sink is set to 200 watts per square meter. Scan the entire vector field matrix data and mark all the pixel points whose heat flux intensity values exceed the threshold corresponding to the material. For example, when the heat flux intensity value at coordinate row 120 column 80 is 250 watts per square meter, it is marked. Perform regional clustering on the continuously adjacent marked points to form a set of regions with excessive heat flux intensity, and record the boundary coordinate data and the maximum heat flux intensity value for each region.

[0036] The implementation steps for locating the coincidence points of the park devices corresponding to the energy efficiency mutation intervals and the devices in the spatial regions where the heat flux intensity exceeds the thermal interaction threshold in the device energy topology network are as follows: Establish a device spatial coordinate mapping relationship table, which contains three items of data: the unique device number, the central coordinate value in the thermal imaging map, and the node number in the device energy topology network. For example, the device number CH-01 has coordinates row 300 column 400 in the thermal imaging map and is node N15 in the topology network. Traverse the device list corresponding to all energy efficiency mutation intervals, and locate the projection area range in the thermal imaging map through the coordinate mapping relationship table. At the same time, obtain the spatial position information of the regions with excessive heat flux intensity, and calculate the spatial intersection area between the device projection area and the excessive area. When the overlapping area ratio between the device projection area in the thermal imaging map and the area with excessive heat flux intensity exceeds the 50% threshold, it is determined as a device coincidence point, and record the device number and the coordinate set of the overlapping area.

[0037] The specific process of identifying the device nodes that simultaneously meet the energy efficiency mutation interval conditions and the heat flux intensity exceeding the thermal interaction threshold as topological vulnerable nodes is as follows: Screen the device nodes corresponding to the set of device coincidence points and verify whether they exist in the characteristic data table of the energy efficiency mutation intervals. For example, if the device number P-07 appears in both the device coincidence point list and the list of mutated devices in the load rate range of 60% to 70%, then it is determined that this device node meets the dual conditions. The finally identified topological vulnerable nodes record the unique device number, the associated energy efficiency mutation interval range, the maximum heat flux intensity value, and the node number in the device energy topology network, forming a list of topological vulnerable nodes. This list is associated and stored with the network topology structure through the node numbers in the device energy topology network.

[0038] A cascaded evaluation module, which is used to evaluate the cascaded carbon emission conduction effect on the associated devices when topological vulnerable nodes are identified. The specific implementation is as follows: The specific implementation process of determining the directly associated devices of the topology vulnerable nodes based on the device energy topology network is as follows: The data processing system reads the device unique numbers in the list of topology vulnerable nodes and performs node positioning retrieval in the adjacency matrix data structure of the device energy topology network. Determine the directly connected device relationships according to the positions of the non-zero elements in the adjacency matrix. For example, when the topology vulnerable node number is N15, scan the column numbers where the non-zero elements are located in the 15th row of the adjacency matrix, and the device numbers N08 and N22 indicated by the corresponding column numbers are identified as directly associated devices. The determined set of directly associated devices records three attribute data items: the device unique number, the energy transmission type, and the connection weight value in the device energy topology network.

[0039] The specific method for calculating the primary carbon emission impact value of the parameter change of the topology vulnerable node on the directly associated devices according to the energy efficiency conduction coefficient in the initial carbon emission model is as follows: Extract the energy efficiency conduction coefficient values between the topology vulnerable node and each directly associated device from the initial carbon emission model database. Assume that the parameter change amount of the topology vulnerable node is expressed as a percentage, then the primary carbon emission impact value on a specific directly associated device is obtained through the following calculation relationship: The primary carbon emission impact value is equal to the parameter change amount multiplied by the energy efficiency conduction coefficient and then multiplied by the baseline carbon emission of the device. The baseline carbon emission is calculated by multiplying the device rated power value by the standard coal consumption coefficient and then by the annual operating time. For example, if the rated power of device N08 is 500 kilowatts, the standard coal consumption coefficient is 0.8 kilograms of carbon dioxide per kilowatt-hour, and the annual operating time is 2000 hours, then the baseline carbon emission is calculated as: 500×2000×0.8 = 800000, with the unit of kilograms of carbon dioxide. The calculation results of the primary carbon emission impact values of all directly associated devices form a list of primary impact data.

[0040] The operations of tracing indirectly associated devices through the energy transfer paths in the device energy topology network include: Starting from the topology vulnerable node, perform a breadth-first search algorithm in the device energy topology network. The search depth parameter is set to 3 layers, and each layer expansion follows the non-backtracking principle: Starting from the current node, traverse the unvisited device nodes along the connection edges. For example, starting the search from node N15, the devices directly reached in the first layer are N08 and N22; in the second layer, connect to N33 through N08 and connect to N41 through N22; in the third layer, connect to N57 through N33. During the search process, completely record the device path sequence and the path comprehensive weight value. The path comprehensive weight is equal to the product value of the energy efficiency conduction coefficients of each connection edge on the path. The finally generated data table of indirectly associated devices contains three fields: the device unique number, the path hop count value, and the path comprehensive weight value.

[0041] The implementation steps for calculating the secondary carbon emission impact value of indirectly associated devices by combining the chain calculation of energy efficiency conduction coefficients are as follows: For each indirectly associated device, extract the sequence of energy efficiency conduction coefficients on its device path. For example, the path sequence of device N57 is from N15 to N08 to N33 to N57, and the conduction coefficient chain consists of three coefficients: the coefficient of 0.85 from N15 to N08, the coefficient of 0.72 from N08 to N33, and the coefficient of 0.68 from N33 to N57. The product of the three coefficients is obtained as 0.85×0.72×0.68 = 0.416. The secondary carbon emission impact value is obtained through the following calculation relationship: The secondary carbon emission impact value is equal to the parameter change amount multiplied by the product of the conduction coefficient chain and the baseline carbon emission of the device itself. Among them, the parameter change amount is the parameter change amount of the topological vulnerable node, and the calculation method of the baseline carbon emission is the same as that of directly associated devices. The calculation results of the secondary carbon emission impact values of all indirectly associated devices form a secondary impact data list.

[0042] The process of accumulating the primary carbon emission impact value and the secondary carbon emission impact value to generate the quantified value of the carbon emission cascade conduction effect is as follows: Initialize the accumulator value to 0. First, traverse the primary impact data list and arithmetically accumulate the impact values of each directly associated device. For example, the impact value of device N08 is 12,000 kg of carbon dioxide, and the impact value of device N22 is 8,000 kg of carbon dioxide. Then the primary accumulation sum is calculated as: 12,000 + 8,000 = 20,000, with the unit of kg of carbon dioxide. Subsequently, traverse the secondary impact data list and arithmetically accumulate the impact values of each indirectly associated device. For example, the impact value of device N33 is 5,000 kg of carbon dioxide, the impact value of device N41 is 3,000 kg of carbon dioxide, and the impact value of device N57 is 2,000 kg of carbon dioxide. Then the secondary accumulation sum is calculated as: 5,000 + 3,000 + 2,000 = 10,000, with the unit of kg of carbon dioxide. Finally, the quantified value of the carbon emission cascade conduction effect is equal to the primary accumulation sum value plus the secondary accumulation sum value, calculated as: 20,000 + 10,000 = 30,000, with the unit of kg of carbon dioxide. This quantified value is output as the cascade effect evaluation result to downstream processing.

[0043] The model update module is used to generate dynamic carbon emission model correction instructions based on the carbon emission cascade conduction effect, update the energy efficiency baseline parameters and the associated device parameters of topological vulnerable nodes, and form a dynamic carbon emission model. The specific implementation is as follows: The specific implementation process of extracting the energy efficiency parameter deviation of topological vulnerable nodes from the quantified value of the carbon emission cascade conduction effect is as follows: Receive the evaluation result of the carbon emission cascade conduction effect, which includes quantified value data and its associated topological vulnerable node information. Determine the deviation through comparative analysis of the quantified value with a preset reference value, where the preset reference value is set in the initial carbon emission model according to the equipment type. For example, for the topological vulnerable node of the refrigeration unit, when the quantified value is 30,000 kilograms of carbon dioxide and the reference value is 25,000 kilograms of carbon dioxide, the energy efficiency parameter deviation is calculated as: (30,000 - 25,000) / 25,000 = 0.2, that is, a 20% deviation. The deviation is calculated in relative percentage form to ensure comparability between different types of equipment. The extracted energy efficiency parameter deviation records three attributes: equipment number, deviation value, and calculation timestamp.

[0044] The operation of calculating the parameter correction factor of the topological vulnerable node in the equipment energy topology network based on the energy efficiency parameter deviation includes: establishing a correction factor mapping function that considers three factors: equipment type, operating years, and current load rate. For example, for refrigeration equipment with an operating life exceeding 5 years, the correction factor is equal to the energy efficiency parameter deviation multiplied by the age decay factor, and the age decay factor is determined according to the equipment aging curve, with a typical value of 0.9; for equipment with a load rate below 40%, the correction factor is equal to the energy efficiency parameter deviation multiplied by the load compensation factor, and the load compensation factor is determined according to the equipment load characteristics, with a typical value of 1.2. The final value range of the parameter correction factor is limited between -50% and +50%, and the boundary value is taken when it exceeds the range. The calculated parameter correction factor is bound to the topological vulnerable node.

[0045] The implementation steps for locating the directly associated equipment of the topological vulnerable node in the equipment energy topology network are as follows: Retrieve the row data corresponding to the topological vulnerable node in the adjacency matrix of the equipment energy topology network. Scan the column positions where the non-zero elements in this row are located, and each non-zero element corresponds to a directly associated equipment node. For example, for the topological vulnerable node N15 in the 15th row of the adjacency matrix, the equipment nodes N08 and N22 corresponding to column positions 8 and 22 are identified as directly associated equipment. The location result forms a list of directly associated equipment, including equipment number, connection type, and connection weight value.

[0046] The specific method for calculating the parameter compensation amount of directly associated equipment through the energy efficiency conduction coefficient is as follows: Extract the energy efficiency conduction coefficient values between the topological vulnerable node and each directly associated equipment from the initial carbon emission model database. The calculation relationship of the parameter compensation amount is: the parameter compensation amount is equal to the parameter correction factor of the topological vulnerable node multiplied by the energy efficiency conduction coefficient and then multiplied by the equipment sensitivity factor. The equipment sensitivity factor is set according to the equipment type. For example, the typical value of the equipment sensitivity factor for variable frequency equipment is 0.8, and the typical value of the equipment sensitivity factor for fixed frequency equipment is 1.0. The calculation results of the parameter compensation amounts of all directly associated equipment form a compensation amount data table.

[0047] The process of converting the energy efficiency parameter deviation and the parameter compensation amount into a dynamic carbon emission model correction instruction is as follows: Create a structured instruction object, which includes an instruction type field, a target device field, and a parameter value field. Generate a parameter update instruction for the topology vulnerable node, set the instruction type to the node parameter correction type, and fill the parameter value field with the energy efficiency parameter deviation value. Generate a parameter compensation instruction for each directly associated device, set the instruction type to the associated device compensation type, and fill the parameter value field with the corresponding parameter compensation amount value. For example, the compensation instruction for device N08 is: the instruction type is the associated device compensation type, the target device is N08, and the parameter value is -0.15. All instructions form an instruction set.

[0048] The operations for updating the energy efficiency benchmark parameters of the initial carbon emission model according to the dynamic carbon emission model correction instruction include: loading the initial carbon emission model database and performing the update in the order of the instruction set. For the parameter update instruction, directly modify the energy efficiency benchmark parameter value of the target device. For the parameter compensation instruction, add the compensation value to the current energy efficiency benchmark parameter of the target device. For example, when the original energy efficiency benchmark parameter of device N08 is 3.5 and it receives a compensation value of -0.15, the new energy efficiency benchmark parameter = 3.5 + (-0.15) = 3.35. Set a version control mechanism during the update process, and generate a new model version identifier for each update.

[0049] The implementation steps for synchronously updating the energy efficiency conduction coefficients of the topology vulnerable node and the directly associated devices are as follows: Analyze the change in the energy transfer relationship between devices after the parameter change and recalculate the energy efficiency conduction coefficient. For the connection between the topology vulnerable node and the directly associated device, the new energy efficiency conduction coefficient is equal to the original coefficient multiplied by the correction factor, and the correction factor = 1 + the parameter change rate. For example, if the original coefficient is 0.85 and the parameter change rate is -0.1, then the correction factor = 1 + (-0.1) = 0.9, and the new coefficient = 0.85 × 0.9 = 0.765. The updated energy efficiency conduction coefficient overwrites the corresponding value in the initial carbon emission model.

[0050] The process of generating a dynamic carbon emission model by combining the updated energy efficiency benchmark parameters and the energy efficiency conduction coefficients is as follows: Create a new model data structure, which includes a device energy efficiency parameter table and a device - to - device conduction relationship table. Extract the energy efficiency benchmark parameters of all devices from the updated initial carbon emission model and fill them into the device energy efficiency parameter table. Extract the energy efficiency conduction coefficients of all device connection relationships from the updated conduction coefficient data and fill them into the conduction relationship table. For example, the device energy efficiency parameter table records that the energy efficiency benchmark parameter of device N15 is 2.8, and the conduction relationship table records that the energy efficiency conduction coefficient from the starting device N15 to the ending device N08 is 0.765. The finally generated dynamic carbon emission model is stored in the model database with a timestamp and a version number attached.

[0051] The carbon emission control module is used to perform carbon emission control operations for the park based on the dynamic carbon emission model during the operation and maintenance stage. The specific implementation is as follows: The specific implementation process of collecting the operation load rate and energy consumption data of park equipment in real time through Internet of Things sensors is as follows: Current sensors, voltage sensors and power metering devices are deployed at each key equipment node in the park. These sensors collect electrical parameters at a sampling frequency of once per second through the Modbus-RTU protocol. The operation load rate is obtained by calculating the real-time power divided by the rated power of the equipment. For example, if the rated power of a chiller is 500 kW and the real-time collected power is 350 kWh, the operation load rate is 70%. The energy consumption data is calculated by integrating power over time, and the cumulative energy consumption value is calculated in a 15-minute cycle, with the unit of kWh. The collected data is preprocessed by the edge computing gateway and then uploaded to the central data processing system.

[0052] The operation of inputting the operation load rate into the dynamic carbon emission model to obtain the theoretical energy efficiency benchmark parameters under the current working condition includes: The central data processing system receives the real-time operation load rate data and retrieves the corresponding record in the equipment energy efficiency parameter table of the dynamic carbon emission model according to the equipment number. Call the working condition matching algorithm, which matches the preset energy efficiency curve according to the load rate interval. For example, when the load rate is in the range of 60% to 80%, a quadratic polynomial fitting curve is used to calculate the theoretical value. The specific implementation is: Input the load rate value into the fitting function, and output the theoretical energy efficiency benchmark parameters. For example, when the load rate is 70%, the theoretical energy efficiency benchmark parameter is 3.2 kg CO₂ / kWh.

[0053] The specific method for calculating the carbon emission deviation value between the actual energy consumption data and the theoretical energy efficiency benchmark parameters is as follows: Convert the actual energy consumption value within the cycle into carbon emissions, and the conversion relationship is the actual energy consumption value multiplied by the theoretical energy efficiency benchmark parameter. For example, if the actual energy consumption within a 15-minute cycle is 87.5 kWh and the theoretical energy efficiency benchmark parameter is 3.2 kg CO₂ / kWh, the theoretical carbon emissions are 87.5×3.2 = 280 kg CO₂. The actual carbon emissions during the same period are measured by a flue gas analyzer as 315 kg CO₂, and the carbon emission deviation value is calculated as: (315 - 280) / 280×100% = 12.5%. The calculation process is executed every 30 minutes.

[0054] The implementation steps for generating an equipment energy efficiency optimization instruction when the carbon emission deviation value exceeds the preset tolerance threshold are as follows: The preset tolerance threshold is set according to the equipment type. For example, the threshold for refrigeration equipment is set to 10%. Real-time judge the relationship between the carbon emission deviation value and the threshold. When it is monitored that the deviation value reaches 12.5% and exceeds the 10% threshold, generate an energy efficiency optimization instruction containing the equipment number, deviation amount, and timestamp. The instruction content specifies the adjustment direction of the energy efficiency parameter. For example, a positive deviation generates a downward adjustment instruction, and a negative deviation generates an upward adjustment instruction. The instruction is encapsulated and transmitted to the control center in JSON format.

[0055] The process of determining the compensation regulation amount of associated devices according to the energy efficiency conduction coefficient in the dynamic carbon emission model is as follows: Retrieve the associated device records of the target device in the conduction relationship table of the dynamic carbon emission model, and extract the corresponding energy efficiency conduction coefficient values. The calculation relationship of the compensation regulation amount is: the compensation regulation amount is equal to the carbon emission deviation value multiplied by the energy efficiency conduction coefficient and then multiplied by the regulation factor, where the regulation factor is set as a floating-point number between 0.6 and 1.2 according to the device response characteristics, and the default value is 0.8. For example, if the conduction coefficient between the target device N15 and the associated device N08 is 0.75 and the carbon emission deviation value is 12.5%, then the compensation regulation amount = 12.5% × 0.75 × 0.8 = 7.5%. The calculation results form a compensation instruction dataset.

[0056] The final operations of sending energy efficiency optimization instructions to the topology-vulnerable node devices and compensation regulation instructions to the associated devices include: The control center parses the energy efficiency optimization instructions and sends parameter adjustment commands to the topology-vulnerable node device controllers through the industrial Ethernet. For example, it sends an instruction to device N15 to lower the energy efficiency parameter by 3%. Synchronously, it sends compensation regulation instructions to the associated devices through the OPC UA protocol. For example, it sends an instruction to device N08 to compensate 7.5% of the parameters. After all instructions are executed, the system records the operation logs and updates the device status database.

[0057] This system realizes precise control of a zero-carbon park through the closed-loop data flow of six modules. The initial modeling module constructs an initial carbon emission model containing the basic energy efficiency parameters of the devices during the design phase, providing a basic framework for the system. The dual-source acquisition module synchronously obtains the operation boundary characteristic spectrum and the thermodynamic trace during the construction phase to form a multi-dimensional dynamic database. The topology identification module combines the Lyapunov exponent stability analysis and the optical flow method for thermal gradient analysis to locate the topology-vulnerable nodes from the dimension of the energy network. The cascade evaluation module calculates the carbon emission cascade effect of parameter changes based on the device energy topology network and quantifies the conduction influence range. The model update module generates dynamic correction instructions according to the evaluation results and updates the basic energy efficiency parameters and the compensation amount of the associated devices in real time. The carbon emission control module executes the dual-instruction collaborative control based on the dynamic model during the operation and maintenance phase to realize the parallel issuance of energy efficiency optimization instructions and compensation regulation instructions. Each module is seamlessly connected through the data bus.

[0058] The solution of this embodiment proposes a carbon emission cascade effect evaluation and dynamic control mechanism: by establishing a coupling model of the device energy topology network and the energy efficiency conduction coefficient, the computable expression of the carbon emission conduction path is realized. In the cascade effect quantification link, the breadth-first search algorithm is used to calculate the indirect influence in combination with the chain product of the conduction coefficients, so as to solve the quantification problem of the carbon emission correlation between multi-hop devices. The dynamic model update mechanism generates compensation instructions based on the parameter deviation of the topological vulnerable nodes, realizing the leap from single-point regulation to system collaborative optimization. The dual-instruction collaborative control strategy (the energy efficiency optimization instruction and the compensation regulation instruction are executed in parallel) in the operation and maintenance stage forms a closed-loop feedback; the reduction of the carbon emission intensity of the park is realized in the industrial scenario measurement. By integrating multi-source heterogeneous data of the device operation boundary feature spectrum and the heat dissipation thermodynamics trace, a cross-industry device energy topology network is established, realizing the dynamic coupling modeling and collaborative control of the carbon emission conduction chain between devices, and accurately driving the overall carbon emission optimization of the park.

[0059] All the calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and threshold selections in the calculations are set by those skilled in the art according to the actual situation.

[0060] It should be noted that the present invention can be deployed on the device itself to achieve embedded applications, or can also be run on a PC or other terminals with a user interface, so as to meet various hardware environments and usage requirements.

[0061] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center in a wireless or wired manner; the wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; the wireless transmission includes infrared rays, microwaves, etc. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0062] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0063] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0064] The modules described as separate components may or may not be physically separated. The components displayed as modules may or may not be physical modules. They can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0065] In addition, in each embodiment of the present application, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0066] If the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0067] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0068] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An information fusion operation and maintenance system among industries in a zero-carbon park, characterized in that Including: An initial modeling module, which is used to establish an initial carbon emission model in the design stage. The initial carbon emission model includes the energy efficiency benchmark parameters of the park equipment; A dual-source acquisition module, which is used to obtain the operation boundary characteristic spectrum of the park equipment through multi-condition dynamic calibration in the construction stage, and synchronously collect the infrared thermal imaging time series of the equipment cooling system to generate a thermodynamic trace; A topology identification module, which is used to analyze the energy efficiency stability of the operation boundary characteristic spectrum based on the Lyapunov exponent, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace to identify the topological vulnerable nodes of the park equipment in the energy network; A cascaded evaluation module, which is used to evaluate the carbon emission cascading conduction effect of the parameter change of the topological vulnerable node on the associated equipment when the topological vulnerable node is identified; A model update module, which is used to generate a dynamic carbon emission model correction instruction according to the carbon emission cascading conduction effect, update the energy efficiency benchmark parameters and the associated equipment parameters of the topological vulnerable node, and form a dynamic carbon emission model; A carbon emission control module, which is used to perform park carbon emission control operations based on the dynamic carbon emission model in the operation and maintenance stage.

2. The zero-carbon park inter-industry information fusion operation and maintenance system according to claim 1, characterized in that, An initial carbon emission model is established in the design stage. The initial carbon emission model includes the energy efficiency benchmark parameters of the park equipment, including: Construct an equipment energy topology network based on the design drawings of the park energy system; Obtain the rated energy efficiency parameters of the park equipment from the equipment supplier database; Bind the rated energy efficiency parameters to the corresponding nodes of the equipment energy topology network; Calculate the energy efficiency conduction coefficient of the associated equipment through the park's cold, heat, and power coupling relationship; Combine the equipment energy topology network, rated energy efficiency parameters, and energy efficiency conduction coefficient to generate an initial carbon emission model; Among them, the energy efficiency benchmark parameters include rated energy efficiency parameters and energy efficiency conduction coefficients.

3. The zero-carbon park inter-industry information fusion operation and maintenance system according to claim 2, wherein In the construction stage, obtain the operation boundary characteristic spectrum of the park equipment through multi-condition dynamic calibration, and synchronously collect the infrared thermal imaging time series of the equipment cooling system to generate a thermodynamic trace, including: Apply stepped load changes to the park equipment during the construction commissioning stage; Collect the dynamic response of the equipment energy efficiency parameters at different load rates through the frequency converter and power sensor; Draw the change curve of the energy efficiency parameter with the load rate according to the dynamic response of the energy efficiency parameter; Extract the energy efficiency inflection point interval in the change curve as the operation boundary characteristic spectrum; At the same time, use an infrared thermal imager to continuously photograph the surface temperature field of the equipment cooling system; Perform pixel-level thermal radiation intensity analysis on the time series images of the surface temperature field of the equipment cooling system; Generate a thermodynamic trace based on the migration trajectory of the thermal radiation intensity between consecutive frames.

4. The zero-carbon park inter-industry information fusion operation and maintenance system according to claim 3, characterized in that, The operation boundary characteristic spectrum and the thermodynamic trace timestamp are synchronized and aligned.

5. The information fusion operation and maintenance system among industries in a zero-carbon park according to claim 3, characterized in that, Analyze the energy efficiency stability of the operation boundary characteristic spectrum based on the Lyapunov exponent, and combine the optical flow method to analyze the thermal gradient vector field of the thermodynamic trace to identify the topological vulnerable nodes of the park equipment in the energy network, including: Calculate the Lyapunov exponent based on the change curve of the energy efficiency parameter with the load rate of the operation boundary characteristic spectrum; When the Lyapunov exponent is greater than the preset stability threshold, it is marked as the energy efficiency mutation interval; Perform optical flow method vector field calculation on the thermodynamic trace to obtain the thermal gradient vector field; Extract the spatial region where the heat flow intensity in the thermal gradient vector field exceeds the heat interaction threshold; Locate the coincidence point of the park equipment corresponding to the energy efficiency mutation interval and the equipment in the spatial area where the heat flow intensity exceeds the heat interaction threshold in the device energy topology network; Identify the device nodes that simultaneously meet the conditions of the energy efficiency mutation interval and the heat flow intensity exceeding the heat interaction threshold as topology vulnerable nodes.

6. The zero-carbon park inter-industry information fusion operation and maintenance system according to claim 5, characterized in that, When topology vulnerable nodes are identified, evaluate the carbon emission cascading conduction effect of the parameter change of the topology vulnerable nodes on the associated devices, including: Determine the directly associated devices of the topology vulnerable nodes based on the device energy topology network; Calculate the primary carbon emission impact value of the parameter change of the topology vulnerable nodes on the directly associated devices according to the energy efficiency conduction coefficient in the initial carbon emission model; Trace the indirectly associated devices through the energy transfer path in the device energy topology network; Calculate the secondary carbon emission impact value of the indirectly associated devices by combining the energy efficiency conduction coefficient chain calculation; Accumulate the primary carbon emission impact value and the secondary carbon emission impact value to generate a quantitative value of the carbon emission cascading conduction effect.

7. An inter-industry information fusion operation and maintenance system for a zero-carbon park according to claim 6, characterized in that, Generate a dynamic carbon emission model correction instruction according to the carbon emission cascading conduction effect, update the energy efficiency benchmark parameter and the associated device parameter of the topology vulnerable node, and form a dynamic carbon emission model, including: Extract the energy efficiency parameter deviation amount of the topology vulnerable node from the quantitative value of the carbon emission cascading conduction effect; Calculate the parameter correction coefficient of the topology vulnerable node in the device energy topology network according to the energy efficiency parameter deviation amount; Locate the directly associated devices of the topology vulnerable node in the device energy topology network; Calculate the parameter compensation amount of the directly associated devices through the energy efficiency conduction coefficient; Convert the energy efficiency parameter deviation amount and the parameter compensation amount into a dynamic carbon emission model correction instruction; Update the energy efficiency benchmark parameter of the initial carbon emission model according to the dynamic carbon emission model correction instruction; Synchronously update the energy efficiency conduction coefficient of the topology vulnerable node and the directly associated devices; Combine the updated energy efficiency benchmark parameter and the energy efficiency conduction coefficient to generate a dynamic carbon emission model.

8. The zero-carbon park inter-industry information fusion operation and maintenance system according to claim 7, characterized in that, Execute the park carbon emission control operation based on the dynamic carbon emission model during the operation and maintenance stage, including: Real-time collect the operation load rate and energy consumption data of the park devices through the Internet of Things sensors; Input the operation load rate into the dynamic carbon emission model to obtain the theoretical energy efficiency benchmark parameter under the current working condition; Calculate the carbon emission deviation value between the actual energy consumption data and the theoretical energy efficiency benchmark parameter; Generate a device energy efficiency optimization instruction when the carbon emission deviation value exceeds the preset tolerance threshold; Determine the compensation regulation amount of the associated devices according to the energy efficiency conduction coefficient in the dynamic carbon emission model; Send the energy efficiency optimization instruction to the topology vulnerable node device and send the compensation regulation instruction to the associated devices.

Images