Energy economic index early warning analysis method and system

Abstract

The invention relates to an energy economic index early warning analysis method and system, and relates to the technical field of comprehensive energy system operation monitoring and evaluation.The method comprises the steps that system real-time operation data and external economic parameters are collected, and an energy flow coupling model containing equipment nonlinear characteristics is established; based on the real-time state, the dynamic economic benchmark is solved by taking the operation cost minimization as the target; the economic drift degree of the actual operation cost relative to the dynamic economic reference is calculated, and the early warning level is judged according to the economic drift degree; when early warning is triggered, the total cost deviation is decoupled into equipment performance degradation deviation and scheduling strategy mismatch deviation through model reverse calculation, and specific factors causing early warning are positioned. According to the invention, through a dynamic reference and deviation attribution mechanism, self-adaptive adjustment of evaluation standards and accurate identification of cost abnormity causes are realized.

Description

Technical Field

[0001] This application relates to the field of integrated energy system operation monitoring and evaluation technology, and in particular to an energy economic indicator early warning analysis method and system. Background Technology

[0002] Integrated energy systems can significantly improve energy efficiency by coupling and complementing multiple energy forms such as electricity, heat, cooling, and gas. In the actual operation of the system, economy is one of the core indicators for measuring its operating level. In order to ensure that the system is in the economic operating range, operation and maintenance personnel usually need to monitor and warn of economic indicators such as operating costs in real time.

[0003] Existing methods for monitoring the economic efficiency of energy systems mostly employ fixed threshold methods or historical data comparison methods. Fixed threshold methods rely primarily on theoretical parameters or empirical values ​​established during the system design phase to set alarm upper limits. However, the actual operating boundary conditions of a comprehensive energy system, such as energy prices, ambient temperature, and load demand, are highly time-varying. Fixed thresholds struggle to adapt to these dynamic changes, easily generating false alarms during peak load periods or when energy prices rise, or missing energy efficiency anomalies during low load periods. While historical data comparison methods incorporate year-on-year or month-on-month analysis, they neglect the differences in external environmental parameters between current and historical operating conditions. This results in a lack of objectivity in the evaluation benchmark, failing to accurately reflect the system's current true energy efficiency level.

[0004] Furthermore, existing monitoring methods typically focus on the statistics of total macro costs, lacking the ability to deeply analyze the causes of cost anomalies. The operating cost of an integrated energy system is jointly determined by the physical conversion efficiency of the underlying equipment and the scheduling strategy of the top-level energy management system. When an abnormal increase in operating costs is detected, existing technologies often struggle to distinguish whether it is due to a decline in physical efficiency caused by equipment aging and failure, or due to energy allocation mismatch caused by improper scheduling strategy settings. This ambiguity in the source of economic deviation makes it difficult for operation and maintenance personnel to quickly locate the root cause of the problem and take timely and targeted maintenance or optimization measures, thus hindering the improvement of the system's refined management level. Summary of the Invention

[0005] The purpose of this application is to provide an energy economic indicator early warning analysis method and system, which aims to solve the technical problems of existing energy system economic monitoring methods relying on fixed thresholds or historical data, which cannot adjust the evaluation benchmark according to real-time operating conditions and external price changes, resulting in insufficient early warning accuracy, and difficulty in distinguishing whether the detected cost anomalies are due to a decline in equipment physical performance or improper scheduling strategy configuration, leading to ambiguous fault location.

[0006] Firstly, the energy economic indicator early warning analysis method provided in this application adopts the following technical solution:

[0007] Includes the following steps:

[0008] Collect real-time operating data and external economic parameters of the integrated energy system, and perform cleaning and time alignment processing to construct the system's real-time state vector and economic parameter vector;

[0009] Based on the variable operating condition characteristics of energy conversion equipment, an energy flow coupling model including the influence of load rate and ambient temperature is established; based on real-time operating data and economic parameter vectors, the real-time actual operating cost index of the system is calculated.

[0010] Based on the system's real-time state vector, with the goal of minimizing operating costs, the dynamic economic benchmark under the current operating conditions is solved using an energy flow coupling model.

[0011] Calculate the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark, and determine the warning level based on the economic drift.

[0012] When the judgment result is a warning or alarm level, a deviation attribution analysis based on deviation component decomposition is performed to decouple the total cost deviation into equipment performance degradation deviation and scheduling strategy inefficiency deviation, and to locate the specific factors that cause the warning.

[0013] The construction of the system's real-time state vector and economic parameter vector includes:

[0014] Acquire physical layer operation data, which includes various energy load demand data and environmental parameter data, and encapsulate the physical layer operation data into a system real-time state vector;

[0015] Acquire economic layer parameter data, which includes real-time energy price information and equipment operation and maintenance cost coefficients, and encapsulate the economic layer parameter data into an economic parameter vector.

[0016] The sampling frequency of the physical layer operation data and the time section of the economic layer parameter data are unified and synchronized.

[0017] For gas turbine units and refrigeration equipment, their energy conversion efficiency is defined as a bivariate nonlinear function of load rate and ambient temperature;

[0018] For gas-fired boilers, their thermal efficiency is defined as a univariate nonlinear function of load rate;

[0019] Based on the law of conservation of energy, the input and output energy ports of each device are associated with the binary nonlinear function and the univariate nonlinear function to construct a set of mathematical constraints describing the energy flow and conversion of the system.

[0020] The real-time actual operating cost indicators of the computing system include:

[0021] Calculate the cost of external energy procurement based on real-time energy exchange volume measurements and real-time energy prices;

[0022] Calculate the equipment maintenance loss cost based on the actual output power measurement value of each device and the unit output maintenance cost coefficient.

[0023] The external energy procurement cost is added to the equipment operation and maintenance loss cost to obtain the real-time total operating cost.

[0024] The method of using an energy flow coupling model to solve for the dynamic economic benchmark under the current operating conditions includes:

[0025] Construct system power balance constraint equations covering three energy flows: electricity, heat, and cold, and set upper and lower limits for the output of each device and limits for the ramp rate.

[0026] Construct an optimization model with the objective of minimizing total operating cost, where equipment output is the decision variable to be solved;

[0027] The optimal solution vector is searched within the feasible region that satisfies all constraints using a mathematical optimization solver, and the objective function value corresponding to the optimal solution vector is used as the dynamic economic benchmark at the current moment.

[0028] The calculation of the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark, and the determination of the early warning level based on the economic drift, includes:

[0029] Calculate the difference between the real-time actual operating cost index and the dynamic economic benchmark, and divide the difference by the dynamic economic benchmark to obtain the economic drift degree;

[0030] The economic drift is compared with the preset early warning threshold and alarm threshold.

[0031] When the economic drift is less than or equal to the warning threshold, it is considered to be in a normal state.

[0032] When the economic drift is greater than the warning threshold but less than or equal to the alarm threshold, it is determined to be a warning level.

[0033] When the economic drift exceeds the alarm threshold, it is determined to be an alarm level.

[0034] The decoupling of total cost deviation into equipment performance degradation deviation and scheduling strategy inefficiency deviation includes:

[0035] The difference between the real-time actual operating cost index and the dynamic economic benchmark is calculated as the total cost deviation;

[0036] By using the energy flow coupling model, the actual output power of the equipment and the current environmental parameters are substituted into the efficiency function to calculate the theoretical input energy in reverse.

[0037] Based on the difference between the actual input energy consumption of the equipment and the theoretical input energy, the unit efficiency deviation cost is calculated and summarized to obtain the equipment performance degradation deviation; the equipment performance degradation deviation is obtained by subtracting the total cost deviation from the total cost deviation.

[0038] The specific factors that cause the warning based on the location include:

[0039] The weighting of computing equipment performance degradation deviation in total cost deviation, and the weighting of scheduling strategy inefficiency deviation in total cost deviation;

[0040] If the weight of the deviation in equipment performance degradation exceeds the preset threshold, the warning reason is determined to be abnormal equipment performance, and abnormal equipment is screened according to the size of the cost of the individual efficiency deviation.

[0041] If the weight ratio of the scheduling strategy failure deviation exceeds the preset threshold, the warning reason is determined to be scheduling strategy failure, and the actual load allocation vector is compared with the optimal load allocation vector obtained when solving the dynamic economic benchmark to locate the control node.

[0042] The warning threshold and alarm threshold are adaptive thresholds set based on the statistical distribution characteristics of the system's historical operating data, and the adaptive thresholds are dynamically adjusted according to the seasonal operating mode of the system.

[0043] Secondly, the energy economic indicator early warning and analysis system provided in this application adopts the following technical solution:

[0044] include:

[0045] The data acquisition module is used to collect real-time operating data and external economic parameters of the integrated energy system, and to construct the system's real-time state vector and economic parameter vector.

[0046] The model storage module is used to store energy flow coupling models that include the nonlinear efficiency characteristics of energy conversion devices;

[0047] The calculation engine module is used to calculate the real-time actual operating cost indicators of the system, solve the dynamic economic benchmark based on the real-time state vector of the system, calculate the economic drift degree and determine the warning level, and perform deviation attribution analysis based on deviation component decomposition.

[0048] The early warning interaction module is used to display the early warning level and the specific factors that led to the early warning, as determined by the deviation attribution analysis.

[0049] In summary, this application includes at least one of the following beneficial technical effects:

[0050] 1. This invention constructs a dynamic economic benchmark solution mechanism based on real-time operating condition optimization, and uses a nonlinear energy flow coupling model to calculate the theoretical minimum operating cost under the current boundary conditions. Compared with the traditional fixed threshold or historical year-on-year method, this mechanism enables the evaluation criteria to be adaptively adjusted with external load demand and energy price fluctuations, effectively eliminating the interference of natural cost growth caused by rising energy unit prices or rigid load increases, and ensuring the objectivity and accuracy of economic assessment.

[0051] 2. This invention proposes an economic drift index and its adaptive early warning judgment logic, using the percentage deviation of actual operating costs from a dynamic economic benchmark as the core evaluation criterion. This dimensionless relative value evaluation method shields the impact of system scale differences and monetary inflation. Even when external energy prices fluctuate drastically, as long as the internal energy dispatch of the system remains efficient, the index value can remain stable, thus avoiding false alarms that are easily caused by a single absolute cost index and significantly improving the robustness of the early warning system.

[0052] 3. This invention establishes a deviation attribution analysis mechanism based on deviation component decomposition. By reverse-calculating the theoretical energy consumption of the equipment, the total cost deviation is decoupled into equipment performance degradation deviation at the physical level and scheduling strategy inefficiency deviation at the control level. This mechanism can accurately distinguish whether the root cause of increased operating costs is reduced efficiency of hardware equipment or improper configuration of energy management system strategies. It overcomes the shortcomings of traditional methods that can only monitor cost anomalies but cannot locate the cause, providing maintenance personnel with a clear direction for troubleshooting. Attached Figure Description

[0053] Figure 1 This is a flowchart of the method of the present invention;

[0054] Figure 2 This is a system architecture diagram of the present invention;

[0055] Figure labeling: 10, Data acquisition module; 20, Model storage module; 30, Calculation engine module; 40, Early warning interaction module. Detailed Implementation

[0056] The following is in conjunction with the appendix Figure 1 - Appendix Figure 2 This application will be described in further detail below.

[0057] A method for early warning analysis of energy economic indicators, referring to Figure 1 This includes the following steps:

[0058] S100: Collects real-time operating data and external economic parameters of the integrated energy system, performs time alignment processing, and constructs the system's real-time state vector;

[0059] S200. Establish an energy flow coupling model that includes the nonlinear efficiency characteristics of the energy conversion equipment. The energy flow coupling model describes the functional relationship between the output power of the equipment and the input energy consumption under different load rates and environmental parameters.

[0060] S300. Calculate the real-time actual operating cost index of the system based on the collected real-time operating data.

[0061] S400: Based on the real-time state vector of the system, with the goal of minimizing operating costs, the theoretical minimum operating cost under the current operating conditions is solved as a dynamic economic benchmark, under the premise of meeting load demand and equipment operation constraints.

[0062] S500 calculates the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark, and compares the economic drift with the preset threshold range to determine the warning level.

[0063] S600 When the judgment result is a warning level, the deviation between the actual operating cost and the dynamic economic benchmark is decomposed into equipment efficiency deviation component and scheduling strategy deviation component to determine the factors that lead to the increase in economic drift.

[0064] An energy economic indicator early warning and analysis system, referring to Figure 2 ,include:

[0065] The system includes a data acquisition module 10, a model storage module 20, a computing engine module 30, and an early warning interaction module 40.

[0066] The data acquisition module 10 connects to the underlying sensor network and external data service interface, periodically acquiring equipment operating parameters, environmental parameters and energy price parameters of the integrated energy system, and outputting the system state vector after time synchronization.

[0067] The model storage module 20 stores the input-output coupling relationship data and nonlinear efficiency characteristic curve data of each energy conversion device.

[0068] The computing engine module 30 is connected to the data acquisition module 10 and the model storage module 20 respectively. The computing engine module 30 receives the system state vector output by the data acquisition module 10, calculates the real-time actual operating cost index and dynamic economic benchmark based on the equipment model data in the model storage module 20, and calculates the economic drift degree based on the two.

[0069] The early warning interaction module 40 is connected to the calculation engine module 30. The early warning interaction module 40 receives economic drift data and cost deviation decomposition data, compares the economic drift with the built-in early warning threshold, draws a comparison chart containing the actual cost curve and the dynamic baseline on the display interface, and outputs an early warning signal containing the source tracing results when the economic drift exceeds the threshold.

[0070] In step S100, real-time operating data and external economic parameters of the integrated energy system are collected and time-aligned to construct the system's real-time state vector. This step may specifically include the following processing procedures:

[0071] The system acquires physical layer operational data and communicates with the integrated energy system's monitoring and data acquisition system or underlying metering devices via a data acquisition interface to obtain the current time. The system acquires various energy load data and environmental parameter data. Specifically, the energy load data includes: electricity load demand data, including active and reactive power; heat load demand data, calculated from supply water temperature, return water temperature, and flow rate, or directly read from the heat meter; cooling load demand data and gas load demand data; and environmental parameter data, specifically including: outdoor dry-bulb temperature, relative humidity, and solar radiation intensity of the area where the system is located. Acquiring the above physical layer operating data by reading sensor or controller values ​​using industrial communication protocols is a well-known technique to those skilled in the art and will not be elaborated here.

[0072] The system obtains economic layer parameter data through the application programming interface of an external energy trading platform or a preset local database to retrieve the current time. The energy price information includes: the grid purchase price, which is the peak, flat, or valley price for the current period when the time-of-use pricing mechanism is in place, and the real-time clearing price when the spot market mechanism is in place; the natural gas price, which is the unit price of gas per unit volume or unit calorific value; and the tap water price. In addition, it is necessary to read the pre-set equipment economic parameters in the database, including the depreciation coefficient and unit power operation and maintenance cost coefficient of each piece of equipment.

[0073] The system cleans and aligns the collected multi-source data over time. Given that the sampling frequencies of physical layer operational data and economic layer parameter data may differ, the system uses a preset calculation cycle as a benchmark. High-frequency data is downsampled or averaged, while low-frequency or missing data is linearly interpolated or preserved to ensure all data is aligned to the same timeframe. At the same time, abnormal noise data that clearly exceeds physical limits is removed.

[0074] Construct the system's real-time state vector and economic parameter vector, and encapsulate the processed data into mathematical vectors as input boundary conditions for subsequent model calculations.

[0075] Define time System real-time state vector It represents the physical boundary conditions that the system must satisfy at that moment, and the system's real-time state vector. Includes electrical load demand Heat load demand Cooling load demand Ambient temperature and light radiation intensity In practical implementation, the system's real-time state vector Represented as:

[0076] ;

[0077] Among them, superscript Indicates vector transpose; ambient temperature and light radiation intensity As key state variables affecting equipment conversion efficiency and photovoltaic output, they are included in the vector.

[0078] Define time Economic parameter vector It represents the external economic environment in which the system operates at that moment, and the economic parameter vector. Includes real-time grid purchase price Natural gas unit price and water price In practical implementation, the economic parameter vector Represented as:

[0079] ;

[0080] Through the above steps, the system completes the transformation from heterogeneous raw data to standardized mathematical vectors, providing a unified data foundation for subsequent cost calculation and benchmark solution based on the energy flow coupling model.

[0081] In step S200, an energy flow coupling model incorporating the nonlinear efficiency characteristics of the energy conversion devices is established. This model describes the conversion relationship between input and output energy for each device within the system. Unlike traditional fixed-efficiency models, the model in this embodiment considers the nonlinear effects of load rate and ambient temperature on the device conversion efficiency to improve the accuracy of describing the system's operating characteristics under varying conditions. This step specifically includes establishing corresponding mathematical models for different types of devices:

[0082] A variable operating condition coupled model of the gas turbine unit is established. As a typical combined heat and power (CHP) unit, the gas turbine unit's input is natural gas, and its output is electrical energy and high-temperature flue gas waste heat. The power generation efficiency of the gas turbine is significantly affected by the load rate and ambient temperature. A time-varying coupled model of the gas turbine unit is defined. gas consumption With output power and the recovery of waste heat power The relationship between the gas turbine's power generation efficiency and the power generation efficiency of the gas turbine. Represented as about load rate and ambient temperature A binary function, load factor Defined as current output electrical power With rated power The ratio of the two variables, which can be obtained by polynomial fitting of the performance curves provided by the equipment manufacturer, is based on the above power generation efficiency and the gas consumption of the gas turbine. The calculation is as follows:

[0083] ;

[0084] in, The lower heating value of natural gas, and the waste heat recovery efficiency of the gas turbine. Primarily affected by load factor, expressed as load factor. The function of recovering waste heat power The calculation is as follows:

[0085] ;

[0086] The above formula establishes a nonlinear coupling relationship between the electrical, thermal, and gas components of a gas turbine.

[0087] A variable operating condition efficiency model for a gas-fired boiler is established. This boiler is used to supplement heat load gaps, with natural gas as input and heat energy as output. The thermal efficiency of the gas-fired boiler fluctuates with changes in load rate, typically decreasing under partial load. The thermal efficiency of the gas-fired boiler is defined as follows. Regarding its load rate The function where load rate Current output thermal power The ratio of gas consumption to rated thermal power in a gas-fired boiler. With output thermal power The relationship is calculated as follows:

[0088] ;

[0089] This model reflects the actual energy consumption characteristics of the boiler at different output levels.

[0090] A variable-condition performance model for refrigeration equipment is established, which mainly includes electric chillers and absorption chillers. Electric chillers consume electrical energy for cooling, while absorption chillers consume thermal energy for cooling. For electric chillers, the coefficient of performance (COP) is... Load factor And cooling water temperature is affected by ambient temperature The combined effect of these factors establishes the input power of the electric chiller. With output cooling capacity The relationship is as follows:

[0091] ;

[0092] For absorption chillers, their coefficient of performance (COP) Mainly affected by load rate Impact, establishing the input heat power of the absorption chiller With output cooling capacity The relationship is as follows:

[0093] ;

[0094] The above-mentioned refrigeration equipment model can reflect the dynamic changes in the equipment's energy efficiency ratio when the temperature changes or the cooling load fluctuates.

[0095] The independent models of the above-mentioned devices are integrated into the overall energy flow coupling model of the system. By using the law of conservation of energy, the input and output ports of each device are physically connected to form a set of mathematical constraints describing the energy flow and conversion of the entire integrated energy system. This set covers the entire process conversion logic from primary energy input (electricity and gas) to secondary energy output (electricity, heat, and cold). It can reverse-calculate the required source-end energy consumption based on any given device output state.

[0096] In step S300, based on the collected real-time operating data, the real-time actual operating cost index of the system is calculated. This step aims to quantify the real economic consumption generated by the system maintaining operation at the current moment, serving as the comparison object for subsequent comparison with theoretical benchmarks. The calculation process specifically includes the following steps:

[0097] To calculate the real-time external energy procurement cost, the system calculates the electricity procurement cost and the natural gas procurement cost based on the real-time energy price obtained in step S100 and the actual energy exchange volume read from the metering device. For the electricity procurement cost, the system reads the real-time interactive power metering value at the connection point with the external power grid. When the metering value is positive, it means that the system is purchasing electricity from the power grid, and the cost is the product of the real-time electricity price and the purchased power. When the metering value is negative, it means that the system is returning electricity to the power grid, and the cost is negative, i.e., the revenue from selling electricity. For the natural gas procurement cost, the system reads the real-time readings of the flow meters at the inlet of the gas turbine and gas boiler, sums them up to obtain the total gas consumption, and multiplies it by the current natural gas unit price. The calculation of external energy procurement cost is not limited to electricity and gas. If the system purchases heat or cooling from the municipal pipeline network, it is also included in the calculation according to the same logic.

[0098] To calculate real-time equipment operation and maintenance costs, the system's operating costs include not only fuel costs but also wear, depreciation, and maintenance expenses incurred during equipment operation. This embodiment employs a variable operation and maintenance cost calculation method based on equipment output. For each key energy conversion device in the system, such as a gas turbine, boiler, or refrigeration unit, a unit output operation and maintenance cost coefficient is pre-set. This coefficient represents the equipment lifespan loss and maintenance amortization amount corresponding to each unit of energy generated by the equipment, such as 1 kilowatt-hour of electrical or thermal energy. The system reads the actual output power of each device in real time, multiplies it by the corresponding operation and maintenance cost coefficient, and sums up the operation and maintenance costs of all operating equipment to obtain the total operation and maintenance cost at the current moment.

[0099] The total real-time operating cost of the system is calculated by summing the external energy procurement cost obtained in step 1 and the equipment operation and maintenance loss cost obtained in step 2, thus obtaining the total real-time operating cost of the system at the current time segment tt. The calculation logic can be expressed mathematically as follows:

[0100] ;

[0101] in: The real-time electricity price at time tt; For a moment The actual power metering value interacting with the power grid; For a moment The unit price of natural gas; and They are time points Actual gas consumption of gas turbines and gas boilers; This refers to the set of all devices currently in operation within the system. Represents the first element in the set. One device; For the first Unit output maintenance cost coefficient of each device; For the first Each device at time The actual output power measurement value.

[0102] Through the above calculations, the system transforms physical-level flow and power data into unified economic-level monetary data, providing a quantitative basis for subsequent economic assessments.

[0103] In step S400, based on the system's real-time state vector and with the goal of minimizing operating costs, the theoretical minimum operating cost under the current operating conditions is calculated as a dynamic economic benchmark, while satisfying load demand and equipment operating constraints. This step constructs a virtual optimal operating scenario using mathematical programming methods. This scenario shares the current external boundary conditions of load, weather, and price, but assumes that the theoretical optimum has been achieved in the equipment scheduling strategy. This process specifically includes the following steps:

[0104] To ensure the calculated dynamic baseline is physically feasible, the system power balance constraint equations must be constructed. This constraint requires that the total energy supply of the system equal the total load demand collected in step S100. The constraint equations cover the balance relationships of the three energy flows: electricity, heat, and cooling.

[0105] For power balance, the system is constrained to purchase power from the grid. Gas turbine power generation and photovoltaic power generation The sum minus the power consumption of the electric chiller and power consumption of system auxiliary equipment After that, it equals the user's electricity load demand. ,Right now:

[0106] ;

[0107] For thermal balance, constrain the waste heat recovery power of the gas turbine. Heat output power of gas boiler The sum, minus the heat power consumed by the absorption chiller. After that, it equals the user's heat load demand. ,Right now:

[0108] ;

[0109] For cold energy balance, the cooling power output of the electric chiller is constrained. Cooling power of absorption chiller The sum equals the user's cooling load demand. ,Right now:

[0110] ;

[0111] Among them, the waste heat power of the gas turbine Energy consumption of the refrigeration unit and The variables are associated with their respective decision variables through the nonlinear coupling function established in step S200.

[0112] To construct safety constraints for equipment operation, and to prevent the solved theoretical operating conditions from exceeding the physical capacity of the equipment, upper and lower limits of output and ramp rate constraints are set for each piece of equipment. These constraints apply to any controllable equipment in the system. Its power Must meet:

[0113] ;

[0114] in, and These are the minimum technical output and rated capacity of the equipment, respectively. Additionally, to account for the dynamic inertia of equipment adjustments, the magnitude of power changes between adjacent time points is constrained.

[0115] ;

[0116] in, and These are the maximum downhill and uphill speeds allowed by the equipment, respectively.

[0117] Construct an objective function that minimizes the operating cost. This is based on the system's current operating cost. Total operating costs Minimization is the optimization objective. The objective function has the same structure as the actual cost calculation formula in the steps, but the equipment output variable is no longer a measured value, but an optimization decision variable to be solved. The objective function is expressed as:

[0118] ;

[0119] in, and Based on the nonlinear efficiency function in step S200, the decision variables... and Export.

[0120] Solving the optimization problem to obtain a dynamic economic benchmark involves summarizing the mathematical models constructed from the steps into a nonlinear programming or mixed-integer linear programming problem. Using a pre-built mathematical optimization solver, the search is conducted within the feasible region that satisfies all physical and safety constraints to find a solution that satisfies the objective function. The optimal solution vector that reaches the minimum value is the objective function value corresponding to the optimal solution, which is the dynamic economic benchmark at the current moment. This benchmark value represents the theoretically lowest operating cost that should occur under the current external environmental load, electricity price, and temperature, assuming that the system scheduling strategy is perfectly matched and the equipment is in the best efficiency coordination state. This benchmark value fluctuates in real time with changes in the external environment, providing an accurate dynamic reference system for subsequent economic evaluation.

[0121] In step S500, the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark is calculated, and the warning level is determined based on the economic drift. This step aims to accurately identify the problem of declining operating efficiency within the system by comparing the relative deviation between the actual value and the theoretical optimal value, eliminating cost increases caused by general increases in energy prices or rigid increases in load. This process specifically includes the following steps:

[0122] The economic drift index is calculated, defined as the percentage deviation of actual operating costs from a dynamic economic benchmark. This is achieved using the real-time total actual operating cost calculated step by step. The dynamic economic benchmark obtained by solving the steps Calculate the time using the following formula Economic drift :

[0123] ;

[0124] Among them, due to It is the theoretical minimum value obtained through optimization algorithms, and under normal circumstances... ,therefore As a non-negative value, the physical meaning of this indicator lies in quantifying the gap between the current operating state of the system and the perfect economic state. The larger the value, the greater the potential for energy-saving optimization of the system, or the more unreasonable the current operating strategy.

[0125] To differentiate management of different levels of economic deviation, an adaptive tiered early warning threshold is set. The system presets two levels of threshold parameters: early warning threshold... and alarm threshold ,and These two threshold levels can be set based on the statistical distribution characteristics of the system's historical operating data. For example, the 95th percentile of the historical drift data can be used as the warning threshold. In a preferred embodiment, the threshold can be dynamically adjusted according to the system's operating mode, such as summer cooling mode or winter heating mode, to adapt to the differences in control accuracy under different seasons.

[0126] The system will calculate the economic drift in real time based on the execution of the early warning logic. The system is compared with the above thresholds to generate corresponding system operating status labels. :

[0127] ;

[0128] When the judgment result is (Normal) indicates that although the current operating cost of the system is higher than the theoretical minimum, the deviation is within a reasonable range of random fluctuations or measurement errors, and no manual intervention is required. When the judgment result is... When a warning is issued, it indicates that the system is deviating from its optimal operating range, resulting in unnecessary energy waste. The system generates a warning signal, suggesting that maintenance personnel monitor equipment parameters or fine-tune the control strategy. When the judgment result is... When an alarm is triggered, it indicates that the system has engaged in serious uneconomical operating behavior, such as a high-efficiency unit failing and shutting down, causing a low-efficiency unit to operate in lieu of it, or a logical error in the energy management strategy, such as purchasing large amounts of electricity during peak electricity price periods. In this case, the system triggers a high-level audible and visual alarm and automatically links to the subsequent source analysis module.

[0129] Based on the above logic, this method implements a relative value early warning mechanism, which can detect even sudden increases in external electricity prices. and At the same time, even with a significant increase, as long as the internal scheduling of the system remains efficient, its ratio This will maintain stability, thus avoiding false alarms that are prone to occur with traditional fixed threshold methods.

[0130] In step S600, when the determination result is a warning or alarm level, a two-way tracing mechanism based on deviation component decomposition is executed. This step aims to decouple the overall economic deviation into physical-level equipment performance deviation and control-level scheduling strategy deviation, thereby providing specific fault location guidance for operation and maintenance personnel. This process is specifically implemented through the following steps:

[0131] Calculate the absolute value of the total cost deviation based on the obtained real-time actual total operating cost. and the obtained dynamic economic benchmark Calculate the total cost deviation at the current moment. :

[0132] ;

[0133] This deviation value represents the amount of unnecessary costs incurred by the system per unit of time, such as per hour.

[0134] The equipment efficiency deviation component is calculated, reflecting the increased cost due to the equipment's operating efficiency being lower than the theoretical model efficiency, for each piece of equipment in the system that is in operation. To obtain its actual output power and actual input energy consumption The energy flow coupling model corresponding to the device established in step S200 is called to calculate the current actual output power. And the theoretically required amount of input energy under current environmental parameters. .

[0135] Calculate the first Unit efficiency deviation cost of the equipment :

[0136] ;

[0137] in, The current unit price of the energy consumed by the equipment, such as natural gas or electricity.

[0138] The total equipment efficiency deviation component of the system is obtained by summing the individual efficiency deviation costs of all operating equipment. :

[0139] ;

[0140] If a certain device A significantly positive value indicates that the energy efficiency ratio or thermal efficiency of the equipment under the current operating conditions is significantly lower than the model baseline, suggesting that there may be physical faults such as heat exchanger scaling, burner carbon buildup, or mechanical wear.

[0141] The scheduling strategy deviation component is calculated, reflecting the cost increase caused by the deviation of the energy allocation strategy (i.e., the output setpoint of each device) from the global optimal solution. The scheduling strategy deviation component is obtained by subtracting the device efficiency deviation component from the total cost deviation. :

[0142] ;

[0143] The physical meaning of this component is: assuming all equipment is in a healthy theoretical efficiency state, but due to the scheduling system selecting an uneconomical combination of equipment, such as incorrectly starting the gas turbine during periods of low electricity prices, or not fully utilizing the heat pump during periods of high gas prices, the actual operating point deviates from the optimal operating point planned in step S400, resulting in opportunity cost.

[0144] Perform two-way attribution analysis, calculate the weight of each deviation component in the total deviation, determine the main causes of the warning based on the weight, and calculate the efficiency deviation weight. and scheduling deviation weight :

[0145] ;

[0146] The system executes the following decision logic:

[0147] like If the value exceeds a preset weight threshold, the warning type is determined to be equipment performance abnormality, and the system further analyzes each device. Sort the data and prioritize those with the largest contribution from the bias. Each device is identified as a suspected fault device, and a comparison chart of its actual efficiency curve and theoretical curve is output.

[0148] like If the load exceeds the preset weight threshold, the warning type is determined to be a scheduling strategy failure. The system compares the actual load allocation vector with the optimal load allocation vector solved in step S400, finds the control node with the largest output deviation, such as the power grid purchase or the charging and discharging of the energy storage device, and prompts the dispatcher to check the control logic or constraint parameter settings.

[0149] Through the above decomposition and tracing process, this invention can distinguish between two fundamentally different causes of economic decline: equipment failure and incorrect strategy, thereby achieving precise positioning from macroeconomic indicators to specific micro-level causes.

Claims

1. A method for early warning analysis of energy economic indicators, characterized in that, Includes the following steps: Collect real-time operating data and external economic parameters of the integrated energy system, and perform cleaning and time alignment processing to construct the system's real-time state vector and economic parameter vector; Based on the variable operating condition characteristics of energy conversion equipment, an energy flow coupling model including the influence of load rate and ambient temperature is established. Calculate the real-time actual operating cost index of the system based on real-time operating data and economic parameter vectors; Based on the system's real-time state vector, with the goal of minimizing operating costs, the dynamic economic benchmark under the current operating conditions is solved using an energy flow coupling model. Calculate the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark, and determine the warning level based on the economic drift. When the judgment result is a warning or alarm level, a deviation attribution analysis based on deviation component decomposition is performed to decouple the total cost deviation into equipment performance degradation deviation and scheduling strategy inefficiency deviation, and to locate the specific factors that cause the warning.

2. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The construction of the system's real-time state vector and economic parameter vector includes: Acquire physical layer operation data, which includes various energy load demand data and environmental parameter data, and encapsulate the physical layer operation data into a system real-time state vector; Acquire economic layer parameter data, which includes real-time energy price information and equipment operation and maintenance cost coefficients, and encapsulate the economic layer parameter data into an economic parameter vector. The sampling frequency of the physical layer operation data and the time section of the economic layer parameter data are unified and synchronized.

3. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The establishment of the energy flow coupling model, which incorporates the influence of load rate and ambient temperature, includes: For gas turbine units and refrigeration equipment, their energy conversion efficiency is defined as a bivariate nonlinear function of load rate and ambient temperature; For gas-fired boilers, their thermal efficiency is defined as a univariate nonlinear function of load rate; Based on the law of conservation of energy, the input and output energy ports of each device are associated with the binary nonlinear function and the univariate nonlinear function to construct a set of mathematical constraints describing the energy flow and conversion of the system.

4. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The real-time actual operating cost indicators of the computing system include: Calculate the cost of external energy procurement based on real-time energy exchange volume measurements and real-time energy prices; Calculate the equipment maintenance loss cost based on the actual output power measurement value of each device and the unit output maintenance cost coefficient. The external energy procurement cost is added to the equipment operation and maintenance loss cost to obtain the real-time total operating cost.

5. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The method of using an energy flow coupling model to solve for the dynamic economic benchmark under the current operating conditions includes: Construct system power balance constraint equations covering three energy flows: electricity, heat, and cold, and set upper and lower limits for the output of each device and limits for the ramp rate. Construct an optimization model with the objective of minimizing total operating cost, where equipment output is the decision variable to be solved; The optimal solution vector is searched within the feasible region that satisfies all constraints using a mathematical optimization solver, and the objective function value corresponding to the optimal solution vector is used as the dynamic economic benchmark at the current moment.

6. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The calculation of the economic drift of the real-time actual operating cost index relative to the dynamic economic benchmark, and the determination of the early warning level based on the economic drift, includes: Calculate the difference between the real-time actual operating cost index and the dynamic economic benchmark, and divide the difference by the dynamic economic benchmark to obtain the economic drift degree; The economic drift is compared with the preset early warning threshold and alarm threshold. When the economic drift is less than or equal to the warning threshold, it is considered to be in a normal state. When the economic drift is greater than the warning threshold but less than or equal to the alarm threshold, it is determined to be a warning level. When the economic drift exceeds the alarm threshold, it is determined to be an alarm level.

7. The energy economic indicator early warning analysis method according to claim 1, characterized in that, The decoupling of total cost deviation into equipment performance degradation deviation and scheduling strategy inefficiency deviation includes: The difference between the real-time actual operating cost index and the dynamic economic benchmark is calculated as the total cost deviation; By using the energy flow coupling model, the actual output power of the equipment and the current environmental parameters are substituted into the efficiency function to calculate the theoretical input energy in reverse. Based on the difference between the actual input energy consumption of the equipment and the theoretical input energy, the unit efficiency deviation cost is calculated and summarized to obtain the equipment performance degradation deviation; the equipment performance degradation deviation is obtained by subtracting the total cost deviation from the total cost deviation.

8. The energy economic indicator early warning analysis method according to claim 7, characterized in that, The specific factors that cause the warning based on the location include: The weighting of computing equipment performance degradation deviation in total cost deviation, and the weighting of scheduling strategy inefficiency deviation in total cost deviation; If the weight of the deviation in equipment performance degradation exceeds the preset threshold, the warning reason is determined to be abnormal equipment performance, and abnormal equipment is screened according to the size of the cost of the individual efficiency deviation. If the weight ratio of the scheduling strategy failure deviation exceeds the preset threshold, the warning reason is determined to be scheduling strategy failure, and the actual load allocation vector is compared with the optimal load allocation vector obtained when solving the dynamic economic benchmark to locate the control node.

9. The energy economic indicator early warning analysis method according to claim 6, characterized in that, The warning threshold and alarm threshold are adaptive thresholds set based on the statistical distribution characteristics of the system's historical operating data, and the adaptive thresholds are dynamically adjusted according to the seasonal operating mode of the system.

10. An energy economic indicator early warning analysis system, comprising an energy economic indicator early warning analysis method according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to collect real-time operating data and external economic parameters of the integrated energy system, and to construct the system's real-time state vector and economic parameter vector. The model storage module is used to store energy flow coupling models that include the nonlinear efficiency characteristics of energy conversion devices; The calculation engine module is used to calculate the real-time actual operating cost indicators of the system, solve the dynamic economic benchmark based on the real-time state vector of the system, calculate the economic drift degree and determine the warning level, and perform deviation attribution analysis based on deviation component decomposition. The early warning interaction module is used to display the early warning level and the specific factors that led to the early warning, as determined by the deviation attribution analysis.

Images