Comprehensive energy efficiency evaluation method and system for smart park

Abstract

The invention provides a comprehensive energy efficiency evaluation method and system for a smart park, and the method comprises the steps: constructing a nonlinear metabolism matrix reflecting the energy incidence relation between equipment, and solving the optimal metabolic flux distribution through mixed integer nonlinear programming; defining an interaction rule base with a metabolic network, dynamically triggering rules according to real-time sensor data, and adjusting real-time metabolic flux; calculating the quantized value of each current energy efficiency index according to the real-time metabolic flux, quantifying the uncertainty through a fuzzy entropy theory, dynamically distributing multi-target weights, and calculating a comprehensive energy efficiency score; and calculating the quantized value of each energy efficiency index in the optimal state according to the optimal metabolic flux distribution, comparing the fuzzy entropy, determining a to-be-optimized energy efficiency index, and identifying high risks and potential risks to formulate an optimization strategy. According to the method, various equipment and energy flow conditions in the park are considered, the energy efficiency level of the system is accurately reflected, dynamic adjustment and evaluation are carried out according to real-time data, uncertainty factors of the energy system are identified and solved, and efficient and stable operation of the system is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart park energy efficiency evaluation, and particularly relates to a comprehensive energy efficiency evaluation method and system for a smart park. BACKGROUND

[0002] A smart park covers multiple fields such as industrial production, commercial office, and residential life. The energy system of the smart park is complex and diverse, involving multiple energy forms such as electricity, heat, and gas, and there is a complex energy interaction relationship between devices. In the operation process of the smart park, efficient use and reasonable allocation of energy are crucial. On the one hand, energy cost is an important part of park operation cost, and reducing energy consumption and waste can effectively improve the economic benefit of the park; on the other hand, reducing energy consumption and pollutant emissions is of great significance to achieving environmental protection and sustainable development goals.

[0003] However, the current traditional energy efficiency evaluation method often focuses on using a single energy efficiency indicator, such as unit energy consumption per output value or unit area energy consumption, to measure the energy utilization efficiency of the park. These indicators are usually obtained through simple statistical calculations and only consider the relationship between energy consumption and a certain specific output. Using a single indicator for evaluation cannot fully reflect the complexity and diversity of the park's energy system, and ignores the mutual influence and correlation between different devices and different energy forms. For example, focusing only on unit energy consumption per output value may not be able to discover energy waste problems existing in the operation of certain devices, because the energy consumption of these devices may be hidden by the overall output value.

[0004] Some current energy efficiency evaluation methods may use a simple linear model to describe the park's energy system, by establishing a linear relationship between energy consumption and various influencing factors to predict and evaluate energy efficiency levels. These models are usually based on empirical formulas or simple statistical analysis, assuming a linear relationship between energy consumption and influencing factors. However, the energy system of a smart park is a complex nonlinear system, and there is often a complex nonlinear relationship between energy consumption and various influencing factors. A simple linear model cannot accurately describe this nonlinear relationship, resulting in large errors in prediction and evaluation results. Moreover, the park's energy system is affected by various uncertain factors such as device failure, energy supply fluctuations, and environmental changes. A simple linear model cannot effectively handle these uncertain factors and cannot accurately evaluate the energy efficiency level of the system under uncertain conditions. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a comprehensive energy efficiency evaluation method and system for a smart park to solve the above-mentioned problems.

[0006] According to the comprehensive energy efficiency evaluation method for a smart park provided by the present application, the method comprises:

[0007] constructing a nonlinear metabolic matrix reflecting the energy correlation between devices, and solving the optimal metabolic flux distribution through mixed integer nonlinear programming;

[0008] defining an interaction rule library with the metabolic network, dynamically triggering the rules and adjusting the real-time metabolic flux of each device according to the real-time sensor data of each device;

[0009] According to the real-time metabolic flux of each device, the quantitative value of each energy efficiency index in the current state is calculated, and the uncertainty of the energy efficiency index in the current state is quantified through fuzzy entropy theory, and the multi-objective weight is dynamically allocated, and the comprehensive energy efficiency score is calculated;

[0010] According to the optimal metabolic flux distribution, the quantitative value of each energy efficiency index in the optimal state is calculated, the fuzzy entropy of the energy efficiency index in the current state and the optimal state is compared, the energy efficiency index to be optimized is determined, and the high risk and potential risk are identified from the energy efficiency index to be optimized, and the targeted optimization strategy is formulated.

[0011] Further, the construction of the nonlinear metabolic matrix reflecting the energy correlation between devices, and solving the optimal metabolic flux distribution through mixed integer nonlinear programming, comprises:

[0012] According to the energy flow relationship of the park devices, the device is defined as a metabolic enzyme, and the energy flow is defined as a metabolite, and a nonlinear metabolic matrix S = [S ij ] n×n , n is the total number of park devices, S ij = f ij (v i , v j ), f ij is a nonlinear function, which represents the energy correlation between the i-th device and the j-th device;

[0013] The device energy flow rate vector is defined as the metabolic flux vector v = [v1, v2, …, v n ] T , v i is the metabolic flux of the i-th device, which represents the energy flow rate of the i-th device, v j is the metabolic flux of the j-th device, which represents the energy flow rate of the j-th device;

[0014] Construct a nonlinear metabolic flux balance equation: S(v)·v = 0, v ≥ 0, where 0 is a zero vector, representing the overall balance of park energy input and output;

[0015] Introducing dynamic constraints, including energy supply and demand fluctuation constraints and device start-stop frequency constraints:

[0016] The optimal metabolic flux distribution is obtained by solving the mixed integer nonlinear programming problem by taking the energy efficiency objective function as the optimization objective and combining dynamic constraints Reflects the energy flow rate of each device under the optimal energy efficiency, wherein, is the metabolic flux of the i-th device under the optimal metabolic flux distribution.

[0017] Further, the optimal metabolic flux distribution is obtained by solving the mixed integer nonlinear programming problem by taking the energy efficiency objective function as the optimization objective and combining dynamic constraints Comprise:

[0018] The energy efficiency objective function F(v) is constructed based on the energy flow cost and the energy flow difference between devices, and is:

[0019] Wherein, c i is the energy flow cost coefficient of the i-th device, r ij is the penalty coefficient of the energy flow difference between the i-th device and the j-th device.

[0020] The minimum energy efficiency objective function F(v) is taken as the optimization objective of the mixed integer nonlinear programming problem, and the optimization objective formula is:

[0021]

[0022] s.t.S(v)·v=0,v≥0;

[0023] The optimal metabolic flux distribution is obtained by solving the optimization objective of the mixed integer nonlinear programming problem

[0024] Further, the definition and the interaction rule library of the metabolic network, according to the real-time sensor data of each device, dynamically triggers the rules and adjusts the real-time metabolic flux of each device, comprising:

[0025] A rule library containing basic rules, preference rules and exception rules is constructed.

[0026] For each device, the rule library is matched in real time according to its sensor data, and if the rule is triggered, the real-time metabolic flux is adjusted, and the adjustment formula is: i，new v i，old =v i，rule +Δv i，preference +Δv i，fault +Δv i，old , wherein v i，rule is the metabolic flux of the i-th device before adjustment, Δv i，preference is the flux change of the i-th device triggered by the basic rule, Δv faultTrigger flux change of the i-th device for abnormal rules;

[0027] The adjusted real-time metabolic flux of each device is fed back to the nonlinear metabolic matrix S, and the energy efficiency objective function is recalculated.

[0028] Further, the quantitative value of each energy efficiency indicator in the current state is calculated according to the real-time metabolic flux of each device, and the formula is:

[0029] Wherein, x a represents the quantitative value of the a-th energy efficiency indicator in the current state, reflecting the energy efficiency level of the system in the current actual running state, and the energy efficiency indicators include total energy consumption, carbon emission and energy cost, v i is the metabolic flux of the i-th device, e i，a is the unit energy flow coefficient of the a-th energy efficiency indicator of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n is the total number of park devices.

[0030] Further, the uncertainty of the energy efficiency indicator in the current state is quantified by fuzzy entropy theory, the multi-objective weight is dynamically allocated, and the comprehensive energy efficiency score is calculated, including:

[0031] For each energy efficiency indicator in the current state, its fuzzy entropy is calculated, and the formula is:

[0032]

[0033] Wherein, H a is the fuzzy entropy of the a-th energy efficiency indicator in the current state, indicating the uncertainty degree of the a-th energy efficiency indicator in the current state, μ a，b is the membership degree of the quantitative value of the a-th energy efficiency indicator in the current state under the energy efficiency state b, and n is the number of membership function division levels of energy efficiency state;

[0034] According to the fuzzy entropy, the dynamic weight is allocated, and the formula is:

[0035] Wherein, w a reflects the importance adjustment of the a-th energy efficiency indicator in the current state under uncertainty, A is the total number of energy efficiency indicators, and λ is the temperature coefficient, which is used to adjust the sensitivity of weight allocation;

[0036] According to the dynamic weight energy efficiency indicator, the weighted sum is obtained, and the formula of the comprehensive energy efficiency score Score is:

[0037] Wherein, Norm(x a ) is the normalized value of the quantitative value of the a-th energy efficiency indicator in the current state, x ax a is the quantitative value of the a th energy efficiency indicator in the current state, min(x a ) is the minimum value of all quantitative values of the a th energy efficiency indicator in the actual operation, and max(x a ) is the maximum value of all quantitative values of the a th energy efficiency indicator in the actual operation.

[0038] Further, the quantitative value of each energy efficiency indicator in the optimal state is calculated according to the optimal metabolic flux distribution, and the formula is:

[0039] wherein, x a represents the quantitative value of the a th energy efficiency indicator in the optimal state, reflecting the ideal energy efficiency level in the ideal state of the system, and the energy efficiency indicators include total energy consumption, carbon emission and energy cost, e i represents the metabolic flux of the i th device under the optimal metabolic flux distribution, e i，a e a i represents the unit energy flow coefficient of the a th energy efficiency indicator of the i th device, including unit energy consumption, unit carbon emission and unit cost, and n represents the total number of devices in the park.

[0040] Further, the fuzzy entropy of the energy efficiency indicators in the current state and the optimal state is compared to determine the energy efficiency indicators to be optimized, and high-risk and potential risk are identified from the energy efficiency indicators to be optimized, and targeted optimization strategies are developed, including:

[0041] For each energy efficiency indicator in the optimal state, the fuzzy entropy thereof is calculated, and the formula is:

[0042]

[0043] wherein, H a represents the fuzzy entropy of the a th energy efficiency indicator in the optimal state, indicating the uncertainty degree of the a th energy efficiency indicator in the optimal state, μ b a represents the membership degree of the quantitative value of the a th energy efficiency indicator in the optimal state under the energy efficiency state b, and n represents the number of division levels of the membership function to the energy efficiency state;

[0044] the difference between the fuzzy entropy of the energy efficiency indicators in the current state and the optimal state is calculated and compared with a preset threshold ΔH;

[0045] if the a th energy efficiency indicator is determined as the energy efficiency indicator to be optimized, indicating that there is a significant difference between the current state and the optimal state in terms of the a th energy efficiency indicator, which needs to be focused on and optimized;

[0046] if the a th energy efficiency indicator is determined as the high-risk energy efficiency indicator, indicating that the uncertainty of the a th energy efficiency indicator in the current state is significantly higher than that in the optimal state, and factors affecting the uncertainty are analyzed, and corresponding optimization measures are developed.

[0047] If The a-th energy efficiency index is determined as a potential risk energy efficiency index, and attention and preventive measures are made in advance.

[0048] The application also provides a comprehensive energy efficiency evaluation system for a smart park, which is used to realize the comprehensive energy efficiency evaluation method for the smart park.

[0049] Metabolic flux module: used to construct a nonlinear metabolic matrix reflecting the energy correlation between devices, and to solve the optimal metabolic flux distribution through mixed integer nonlinear programming;

[0050] Real-time metabolic flux module: used to define an interaction rule library of the metabolic network, dynamically trigger the rules and adjust the real-time metabolic flux of each device according to the real-time sensor data of each device;

[0051] Comprehensive energy efficiency evaluation module: used to calculate the quantitative values of each energy efficiency index in the current state according to the real-time metabolic flux of each device, to quantify the uncertainty of the energy efficiency index in the current state through fuzzy entropy theory, to dynamically allocate multi-objective weights, and to calculate the comprehensive energy efficiency score;

[0052] Energy efficiency optimization module: used to calculate the quantitative values of each energy efficiency index in the optimal state according to the optimal metabolic flux distribution, to compare the fuzzy entropy of the energy efficiency index in the current state and the optimal state, to determine the energy efficiency index to be optimized, to identify high risk and potential risk from the energy efficiency index to be optimized, and to develop a targeted optimization strategy.

[0053] In summary, the comprehensive energy efficiency evaluation method for the smart park of the present application firstly constructs a nonlinear metabolic matrix reflecting the energy correlation relationship between devices to accurately capture the complex energy interaction characteristics between park devices, and uses mixed integer nonlinear programming to solve the optimal metabolic flux distribution, realizing dynamic optimization and balance of system energy flow, thereby effectively improving energy utilization efficiency and reducing energy waste; then, the interaction rule library of the metabolic network is defined, and the real-time metabolic flux of each device is dynamically triggered based on real-time sensor data, so that the metabolic flux can accurately reflect the real-time changes of the park energy demand and the abnormal fluctuations of the current device operation state, ensuring the stability and reliability of energy supply; then, the quantitative value of each energy efficiency index is calculated according to the real-time metabolic flux, reflecting the energy efficiency level of the system under the current actual operation state, and the uncertainty of each energy efficiency index under the current state is quantified by fuzzy entropy theory to reflect the uncertainty degree of the energy efficiency index under the current state, and the dynamic weight of the energy efficiency index under the current state is distributed according to the fuzzy entropy, that is, the weight of the energy efficiency index is automatically adjusted according to the uncertainty degree, so that the energy efficiency index with low uncertainty plays a greater role in the comprehensive evaluation, thereby improving the accuracy and reliability of the energy efficiency evaluation. Finally, the quantitative value of each energy efficiency index under the optimal state is calculated according to the optimal metabolic flux distribution, which is used to reflect the energy efficiency level of the park energy system under the ideal operation state, and the fuzzy entropy of each energy efficiency index under the optimal state is calculated by using the fuzzy entropy theory, which quantifies the uncertainty of each energy efficiency index under the optimal state, provides a comparison benchmark, and compares the energy efficiency index fuzzy entropy under the current state and the optimal state to quantify the difference degree of the current state and the optimal state in terms of energy efficiency index uncertainty, determine the to-be-optimized index, and identify high-risk and potential risks from the to-be-optimized index, and perform targeted optimization to solve the uncertainty factors of the park energy system and improve the energy efficiency level, thereby ensuring the efficient and stable operation of the park energy system.

[0054] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter in the description of embodiments of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0055] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the description of the embodiments, taken in conjunction with the following drawings in which:

[0056] Figure 1 A flow chart of the comprehensive energy efficiency evaluation method for the smart park of the present application embodiment one;

[0057] Figure 2 A system block diagram of the comprehensive energy efficiency evaluation system for the smart park of the present application embodiment two. DETAILED DESCRIPTION

[0058] For the purpose of promoting an understanding of the principles of the application, reference will now be made to the embodiments illustrated in the drawings. There is shown in the drawings several embodiments of the application. However, it should be understood that the application can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. It is therefore an object of the present application to provide a method and system for providing a user with a virtual environment.

[0059] It is to be understood that where an element is referred to as being "on" another element, it can be directly on the other element or intervening elements can also be present. Where an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements can also be present. As used herein the terms "vertical", "horizontal", "left", "right" and the like are merely used for illustration and do not imply any particular orientation.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises", "comprising", "including" and "having" are inclusive and therefore specify the presence of stated features but do not preclude the presence or addition of one or more other features. The term "and / or" includes any and all combinations of one or more of the associated listed items.

[0061] Embodiment one

[0062] Referring to Figure 1 , the application provides a comprehensive energy efficiency evaluation method for a smart park, which comprises steps S101-S104:

[0063] S101, a nonlinear metabolic matrix reflecting the energy correlation relationship between devices is constructed, and the optimal metabolic flux distribution is solved by mixed integer nonlinear programming.

[0064] Further optionally, the construction of the nonlinear metabolic matrix reflecting the energy correlation relationship between devices and the solving of the optimal metabolic flux distribution by mixed integer nonlinear programming comprises:

[0065] According to the energy flow relationship of the park devices, the devices are defined as metabolic enzymes, and the energy flow is defined as metabolites, and a nonlinear metabolic matrix S=[S ij ] n×n is constructed, where n is the total number of park devices, S ij =f ij (v i , v j ), f ij is a nonlinear function for representing the energy correlation relationship between the i-th device and the j-th device;

[0066] The device energy flow rate vector is defined as the metabolic flux vector v=[v1, v2, …, v n ]T , v i is the metabolic flux of the ith device, representing the energy flow rate of the ith device;

[0067] A nonlinear metabolic flux balance equation is constructed: S(v) · v = 0, v ≥ 0, where 0 is a zero vector, representing the overall balance of energy input and output in the park;

[0068] Dynamic constraints are introduced, including energy supply and demand fluctuation constraints and device start-stop frequency limit constraints:

[0069] Taking the energy efficiency objective function as the optimization objective, combined with dynamic constraints, the optimal metabolic flux distribution is obtained by mixed integer nonlinear programming Reflecting the energy flow rate of each device under the optimal energy efficiency.

[0070] Understandably, in a biological metabolic system, metabolic enzymes catalyze metabolites to undergo chemical reactions, realizing the conversion and transmission of matter and energy. Similarly, in the energy system of a park, devices such as metabolic enzymes bear the functions of energy conversion, transmission and utilization; energy flow is similar to metabolites, which are transmitted and converted between different devices. For example, in an industrial park, a boiler device is like a kind of metabolic enzyme, which can convert fuel such as coal (energy input, similar to metabolic material) into heat energy (energy output, similar to metabolic product), and then the heat energy is transmitted to other devices that need heat energy, such as a dryer, which further utilizes the heat energy to dry the material, completing the further conversion and utilization of energy.

[0071] This embodiment defines the devices in the park as metabolic enzymes and the energy flow as metabolites, and constructs a nonlinear metabolic matrix S = [S ij ] n×n , where S ij = f ij (v i , v j ), f ij as a nonlinear function, used to describe the complex energy correlation between devices (i.e. between the ith device and the jth device). This nonlinear relationship can reflect actual situations such as efficiency decay and load dependence. For example, there are two devices A and B in the park, and device A transmits electrical energy to device B. When the output power of device A (i.e. v i ) is low, the transmission efficiency is relatively high due to relatively fixed line loss and other factors; but as the output power of device A increases, the line loss increases in a nonlinear manner, resulting in a gradual decrease in the increase in the actual power transmitted to device B (related to S ij ), which is a kind of efficiency decay nonlinear relationship, which can be described by a suitable nonlinear function f ij .

[0072] Specifically, when constructing the nonlinear function f ij , the inter-device energy flow rate can be taken as the independent variable, and the index reflecting the inter-device energy correlation can be taken as the dependent variable to draw a scatter plot. The type of nonlinear relationship can be preliminarily judged according to the scatter distribution trend, and the possible form of f ij is assumed. Then, the parameters of the nonlinear function are estimated by minimizing the sum of squared errors function through the least squares method, and the least squares problem is solved by using an iterative algorithm (such as gradient descent method), and finally the specific form of f ij (v i , v j ) is determined, which provides an accurate function basis for constructing the nonlinear metabolic matrix.

[0073] In the smart park, representative equipment combinations are selected, such as production equipment (numerical control machine tools, industrial robots, etc.), energy conversion equipment (air compressors, boilers, etc.), auxiliary equipment (ventilation and air conditioning systems, lighting, etc.), etc. Through energy monitoring sensors and load sensors, device energy input and output rates and load state information are collected at certain time intervals (such as 15 minutes), and park energy consumption, carbon emissions, cost, and other energy efficiency index data are recorded. After that, the collected data is preprocessed (remove outliers and missing values) and normalized (eliminate dimension effects) to ensure data quality.

[0074] The device energy flow rate vector is defined as the metabolic flux vector v = [v1, v2, …, v n ] T , where v i is the metabolic flux of the i-th device, representing the energy flow rate (including input or output rate) of the i-th device, which is used to intuitively reflect the activity and energy transfer efficiency of the device in the energy system. v i > 0 indicates output, and v i < 0 indicates input, which clearly quantifies the energy flow.

[0075] The nonlinear metabolic flux balance equation S(v) · v = 0, v ≥ 0 is constructed, where 0 is the zero vector. This equation ensures the overall balance between energy input and output in the park, which is an important constraint for stable operation of the energy system.

[0076] At the same time, energy supply and demand fluctuation constraints (to ensure that the metabolic flux meets the balance within the prediction error range) and device start-stop frequency constraints (to limit the number of device start-stop times and avoid frequent start-stop damage to devices and increase costs) are introduced to make the optimization process more in line with the actual park energy system operation requirements.

[0077] Then, taking the energy efficiency objective function as the optimization objective, combining with the dynamic constraints, the optimal metabolic flux distribution is obtained by solving the mixed integer nonlinear programming The optimal metabolic flux distribution is the energy flow rate of each device under optimal energy efficiency (i.e. ideal energy efficiency state). The optimal metabolic flux distribution obtained by optimization can realize the dynamic balance and energy efficiency maximization of the park's energy system, and provide optimization guidance for subsequent energy efficiency management.

[0078] Optionally, the optimal metabolic flux distribution is obtained by using a mixed-integer nonlinear programming approach, with the energy efficiency objective function as the optimization target and incorporating dynamic constraints. include:

[0079] Based on energy flow costs and differences in energy flow between equipment, the energy efficiency objective function F(v) is constructed as follows:

[0080] Among them, c i Let r be the energy flow cost coefficient for the i-th device. ij The penalty coefficient for the difference in energy flow between the i-th device and the j-th device;

[0081] Minimizing the energy efficiency objective function F(v) is taken as the objective of optimizing the mixed-integer nonlinear programming problem. The optimization objective formula is:

[0082]

[0083] stS(v)·v=0,v≥0;

[0084] The optimal metabolic flux distribution is obtained by solving the optimization objective of the mixed-integer nonlinear programming problem.

[0085] Understandably, constructing an energy efficiency objective function It consists of two parts. Part One This reflects the economics of energy flow in equipment. By considering the energy flow cost coefficient for each piece of equipment, the optimization process can balance the relationship between energy flow volume and cost, avoiding unnecessary energy consumption and cost increases. Part Two This serves as a penalty term for differences in energy flow between devices, used to balance energy flow between them. This is achieved by introducing a penalty coefficient r. ij When the energy flow difference between devices is too large, the value of the penalty term will increase, thereby promoting the energy flow between devices to tend to be balanced during the optimization process, avoiding excessive or insufficient energy flow for some devices, and ensuring the stable operation of the park's energy system.

[0086] Minimizing the energy efficiency objective function F(v) is taken as the objective of optimizing the mixed-integer nonlinear programming problem, and the optimization formula is: Meanwhile, the nonlinear metabolic flux balance equation S(v)·v=0 and dynamic constraints (such as energy supply and demand fluctuation constraints and device start-stop number limit constraints) need to be met. These constraint conditions ensure that the metabolic flux distribution obtained by optimization solving not only considers the energy efficiency target, but also meets the actual operation requirements of the park energy system, including energy supply and demand balance, device operation stability and the like.

[0087] By solving the optimization target of the mixed integer nonlinear programming problem, the optimal metabolic flux distribution The optimal metabolic flux distribution is the energy flow rate of each device under the optimal energy efficiency (i.e. the ideal energy efficiency state). The optimal metabolic flux distribution obtained by optimization solving can realize dynamic balance and energy efficiency maximization of the park energy system, that is, under the premise of meeting the energy supply and demand balance and the device operation stability, by optimizing the energy flow among devices, the energy flow cost can be reduced, the energy flow difference among devices can be balanced, and thus the energy efficiency level of the entire park energy system can be improved, thereby providing a scientific basis and optimization direction for energy management and decision-making of the smart park.

[0088] Further optionally, the dynamic constraints are introduced, including:

[0089] The energy supply and demand fluctuation constraint is introduced to ensure that the metabolic flux meets the balance within the prediction error range. The energy supply and demand fluctuation constraint formula is:

[0090] wherein d j is the energy demand of the jth user, m is the total number of users, and ∈ is the allowed energy supply and demand fluctuation proportion;

[0091] The constraint start-stop number is introduced, and the formula is:

[0092] wherein z i (t) is an integer variable, representing the start-stop state of the ith device at time slot t, z i (t)∈{0,1}, N max is the upper limit of the device start-stop number, M is a large enough constant, and T1 is the total number of time slots in the evaluation period.

[0093] Understandably, by introducing the energy supply and demand fluctuation constraint, it is ensured that the metabolic flux meets the balance within the prediction error range. In the actual smart park energy system, the energy demand has certain uncertainty. By introducing this constraint, a certain fluctuation range (controlled by ∈) is allowed between the actual energy supply and demand and the predicted value, so that the metabolic flux distribution obtained by optimization solving is more in line with the actual situation, avoiding the problem that the actual operation cannot be realized due to excessive pursuit of accurate balance, and ensuring the stability and reliability of the park energy supply.

[0094] The equipment start-stop frequency constraint limits the number of times the equipment can be started and stopped within the evaluation period. Frequent start-stop operations not only damage the equipment itself and shorten its lifespan, but also increase energy consumption and operating costs. This is addressed by setting an upper limit N for the number of equipment start-stop operations. max , and combined with v i (t)≤M·z i The condition (t) (ensuring that energy flows only when the equipment is turned on) allows the optimization process to consider both energy efficiency targets and the operational stability and cost factors of the equipment.

[0095] The formula for constraining the number of equipment start-stop cycles is as follows: Among them, z i (t) is an integer variable representing the start / stop state of the i-th device in time slot t, z i (t)∈{0,1}(z i (t) = 1 indicates that the device is turned on, z i (t) = 0 indicates that the device is off); N max The upper limit for the number of device start-ups and shutdowns; M is a sufficiently large constant; T1 is the total number of time slots within the evaluation period, within the preset evaluation period [t]. start , t end Within this period, the time interval is divided according to a certain time interval Δt to obtain the result. t start To evaluate the start time t of the time period end The end time of the evaluation period is Δt, which is a preset time interval (for example, Δt can be set to 1 hour, 15 minutes, etc., and the specific value is determined according to the evaluation accuracy requirements).

[0096] S102 defines the interaction rule base with the metabolic network, which dynamically triggers rules and adjusts the real-time metabolic flux of each device based on real-time sensor data from each device.

[0097] Further optionally, the defined interaction rule base with the metabolic network dynamically triggers rules and adjusts the real-time metabolic flux of each device based on real-time sensor data from each device, including:

[0098] Build a rule base that includes basic rules, preference rules, and exception rules;

[0099] For each device, a rule base is matched in real time based on its sensor data (such as electricity price, temperature, and user feedback). If a rule is triggered, the real-time metabolic flux is adjusted using the following formula: v i，new =v i，old +Δv i，rule +Δv i，preference +Δv i，fault , where v i，oldAdjusted metabolic flux of the ith device, Δv i，rule Flux change of the ith device triggered by the base rule, Δv i，preference Flux change of the ith device triggered by the preference rule, Δv fault Flux change of the ith device triggered by the exception rule, Δv

[0100] The adjusted metabolic flux of each device is fed back to the nonlinear metabolic matrix S, and the energy efficiency objective function is recalculated.

[0101] It is understood that a rule library containing base rules (such as electricity price / temperature triggers), preference rules (such as user comfort priority), and exception rules (such as device failure) is constructed. By setting rules, various situations that may be encountered during the operation of the park energy system are comprehensively covered, providing a wealth of basis for dynamically adjusting metabolic flux. For example:

[0102] Base rule example: IF P > P threshold THEN v 照明 ← 0.7·v 照明 When the real-time electricity price P exceeds the set threshold P threshold , the metabolic flux v 照明 of the lighting device is adjusted to 0.7 times the original value. This rule takes into account the electricity price factor and reduces the energy consumption of the lighting device when the electricity price is high, which helps to reduce energy costs.

[0103] Preference rule example: IF U user < U min THEN v 空调 ← max(v 空调 , 0.8·v 空调，max ). When the user feedback U user is lower than the set minimum value U min , the metabolic flux v 空调 of the air conditioning device is adjusted to the larger value of the current value and 0.8·v 空调，max (the upper limit of the metabolic flux of the air conditioning device is assumed to be v 空调，max ). This rule prioritizes user comfort, and when user feedback indicates insufficient comfort, it appropriately increases the energy supply of the air conditioning device to improve user satisfaction.

[0104] Exception rule example: IF Deviceifails THEN v backup，i ← 1.0·v backup，i When the ith device fails, the metabolic flux v backup，i of the standby device of the ith device is adjusted to 1 times the original value (i.e., full load operation). This rule ensures that standby devices can be put into use in a timely manner when a device fails, ensuring the normal operation of the park energy system.

[0105] For each device, the flux change triggered by different types of rules is superimposed by adjusting the formula v i，new = v i，old + Δv i，rule + Δv i，preference + Δv i，fault , to obtain the adjusted metabolic flux v new , realizing the dynamic adjustment of metabolic flux. Wherein v i，old is the metabolic flux of the i-th device before adjustment, Δv i，rule is the flux change of the i-th device triggered by the basic rule, Δv i，preference is the flux change of the i-th device triggered by the preference rule, Δv fault is the flux change of the i-th device triggered by the exception rule.

[0106] The adjusted actual metabolic flux of each device is fed back to the nonlinear metabolic matrix S, ensuring that the real-time adjustment of metabolic flux can accurately reflect the current running state of the system in time, and the energy efficiency objective function is recalculated, so that the system can optimize the energy efficiency according to the new metabolic flux distribution, to ensure that the park energy system is always in or close to the optimal operating state.

[0107] S103, according to the actual metabolic flux of each device, calculate the quantitative value of each energy efficiency index in the current state, and quantify the uncertainty of energy efficiency index in the current state through fuzzy entropy theory, dynamically allocate multi-objective weight, calculate the comprehensive energy efficiency score.

[0108] Further optionally, the formula for calculating the quantitative value of each energy efficiency index in the current state according to the actual metabolic flux of each device is:

[0109] Wherein, x a represents the quantitative value of the a-th energy efficiency index in the current state, reflecting the energy efficiency level of the system in the current actual running state, and the energy efficiency index includes total energy consumption, carbon emission and energy cost, v i is the metabolic flux of the i-th device, e i，a is the unit energy flow coefficient of the a-th energy efficiency index of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n is the total number of devices in the park.

[0110] Understandably, the formula By multiplying the absolute value of the actual metabolic flux of each device with the unit energy flow coefficient of the corresponding energy efficiency index, and adding up the results of all devices, the quantitative value of each energy efficiency index in the current state is obtained. Comprehensive consideration of the contribution of all devices in the park to the energy efficiency index can accurately reflect the actual running state of the park energy system under the actual metabolic flux distribution.

[0111] The energy efficiency indicators include total energy consumption, carbon emissions, and energy cost, which are important aspects of measuring the operation efficiency and environmental impact of the park energy system. Among them, the total energy consumption reflects the overall energy consumption of the park energy system, which is the basic indicator for evaluating energy utilization efficiency; carbon emissions reflect the impact of the park energy system on the environment, in line with the current energy-saving and emission-reducing development trend; and energy cost is directly related to the operating cost of the park and has an important impact on the economic benefits of the park. By calculating the quantitative values of these energy efficiency indicators, the performance of the park energy system can be comprehensively evaluated.

[0112] Further, the uncertainty of the energy efficiency indicators in the current state is quantified by the fuzzy entropy theory, the multi-objective weights are dynamically allocated, and the comprehensive energy efficiency score is calculated, including:

[0113] For each energy efficiency indicator in the current state, its fuzzy entropy is calculated, and the formula is:

[0114]

[0115] where H a is the fuzzy entropy of the a-th energy efficiency indicator in the current state, indicating the uncertainty degree of the a-th energy efficiency indicator in the current state, μ a，b is the membership degree of the quantitative value of the a-th energy efficiency indicator in the current state under the energy efficiency state b, and n is the number of division levels of the membership function to the energy efficiency state.

[0116] The dynamic weight is allocated according to the fuzzy entropy, and the formula is:

[0117] where w a reflects the importance adjustment of the a-th energy efficiency indicator in the current state under uncertainty, and the fuzzy entropy H a is greater (i.e. the uncertainty is higher), the weight w a is smaller, so as to reduce the influence of the energy efficiency indicator with high uncertainty on the overall evaluation, A is the total number of energy efficiency indicators, and λ is the temperature coefficient for adjusting the sensitivity of weight allocation.

[0118] The energy efficiency indicators are weighted and summed according to the dynamic weight w a to obtain the comprehensive energy efficiency score Score, and the formula is:

[0119] where Norm(x a ) is the normalized value of the a-th energy efficiency indicator in the current state, x a is the quantitative value of the a-th energy efficiency indicator in the current state, min(x a ) is the minimum value of all quantitative values of the a-th energy efficiency indicator in actual operation, and max(x a) is the maximum value of all quantitative values of the a-th energy efficiency indicator in actual operation.

[0120] It can be understood that, since the smart park energy system is a complex system, the uncertainty of the energy efficiency indicator is affected by multiple factors. The present embodiment uses the fuzzy entropy theory formula to quantify the uncertainty of each energy efficiency indicator under the current state. Fuzzy entropy is an indicator that measures the uncertainty of a fuzzy set, which can well adapt to the complexity and uncertainty of the park energy system. Here, the membership degree μ a，b of the quantitative value of the energy efficiency indicator under the current operation under different energy efficiency states is calculated, and then the fuzzy entropy H a of the energy efficiency indicator under the current state is obtained, to reflect the degree of uncertainty of the energy efficiency indicator under the current state.

[0121] Then, according to the fuzzy entropy H a of the energy efficiency indicator under the current state, a dynamic weight is allocated, and the formula is where the temperature coefficient λ is used to adjust the sensitivity of weight allocation, and the fuzzy entropy H a is greater (i.e. the uncertainty is higher), the weight w a is smaller, to reduce the influence of the energy efficiency indicator with high uncertainty on the overall evaluation. Dynamic weight allocation can automatically adjust the weight of the energy efficiency indicator according to the degree of uncertainty, so that the energy efficiency indicator with low uncertainty plays a greater role in the comprehensive evaluation, thereby improving the accuracy and reliability of the evaluation.

[0122] S104, according to the optimal metabolic flux distribution, the quantitative value of each energy efficiency indicator under the optimal state is calculated, the fuzzy entropy of the energy efficiency indicator under the current state and the optimal state is compared, the energy efficiency indicator to be optimized is determined, and the high risk and potential risk are identified from the energy efficiency indicator to be optimized, and the targeted optimization strategy is formulated.

[0123] Further optionally, according to the optimal metabolic flux distribution, the quantitative value of each energy efficiency indicator under the optimal state is calculated, and the formula is:

[0124] wherein, represents the quantitative value of the a-th energy efficiency indicator under the optimal state, reflecting the ideal energy efficiency level under the ideal state of the system, and the energy efficiency indicator includes total energy consumption, carbon emission and energy cost, is the metabolic flux of the i-th device under the optimal metabolic flux distribution, e i，a is the unit energy flow coefficient of the a-th energy efficiency indicator of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n is the total number of park devices.

[0125] It can be understood that, the formula The quantitative value of each energy efficiency indicator in the optimal state is obtained by multiplying the absolute value of the metabolic flux of each device under the optimal metabolic flux distribution with the corresponding unit energy flow coefficient, and accumulating the results of all devices to comprehensively consider the contribution of all devices in the park to the energy efficiency indicators, for reflecting the ideal running state of the park energy system under the optimal metabolic flux distribution.

[0126] Since the metabolic flux directly determines the contribution of each device to each energy efficiency indicator x a . The greater the metabolic flux of the device, the greater its influence on the energy efficiency indicators. For example, if the metabolic flux of a device increases, the values of the indicators such as energy consumption, carbon emissions and cost x a will also increase accordingly. Therefore, by optimizing each metabolic flux, the optimal metabolic flux distribution v * (such as minimizing total energy consumption or carbon emissions) is solved to optimize each energy efficiency indicator x a . For example, during the optimization of the metabolic flux distribution, the metabolic flux of some devices can be reduced by adjusting the running state of the devices, so as to achieve the purpose of reducing the total energy consumption and carbon emissions, and reducing the energy cost.

[0127] Further optionally, the fuzzy entropy of the comparative energy efficiency indicator under the current state and the optimal state is determined to determine the energy efficiency indicator to be optimized, and high-risk and potential risk are identified from the energy efficiency indicator to be optimized, and a targeted optimization strategy is formulated, including:

[0128] For each energy efficiency indicator in the optimal state, the fuzzy entropy is calculated, and the formula is:

[0129]

[0130] Wherein, is the fuzzy entropy of the a-th energy efficiency indicator in the optimal state, indicating the uncertainty degree of the a-th energy efficiency indicator in the optimal state, is the membership degree of the quantitative value of the a-th energy efficiency indicator in the optimal state under the energy efficiency state b, and n is the number of division levels of the membership function to the energy efficiency state;

[0131] The difference between the fuzzy entropy of the energy efficiency indicator under the current state and the optimal state is calculated and compared with a preset threshold ΔH;

[0132] If , the a-th energy efficiency indicator is determined as the energy efficiency indicator to be optimized, indicating that there is a significant difference between the current state and the optimal state in the a-th energy efficiency indicator, which needs to be focused on and optimized;

[0133] If If the fuzzy entropy of the energy efficiency indicator a in the current state H a is greater than the fuzzy entropy of the energy efficiency indicator a in the optimal state H , it indicates that the uncertainty of the energy efficiency indicator a in the current state is significantly higher than that in the optimal state, and the energy efficiency indicator a is determined as a high-risk energy efficiency indicator. The factors affecting the uncertainty of the energy efficiency indicator a are analyzed, and corresponding optimization measures are developed.

[0134] If the fuzzy entropy of the energy efficiency indicator a in the current state H a is less than the fuzzy entropy of the energy efficiency indicator a in the optimal state H , it indicates that the uncertainty of the energy efficiency indicator a in the current state is significantly lower than that in the optimal state, and the energy efficiency indicator a is determined as a potential risk energy efficiency indicator, which needs to be paid attention to and preventive measures are developed in advance. It can be understood that the fuzzy entropy of each energy efficiency indicator in the optimal state is calculated using the fuzzy entropy theory formula , where represents the fuzzy entropy of the energy efficiency indicator a in the optimal state, reflecting the degree of uncertainty, is the membership degree of the quantitative value of the energy efficiency indicator a in the optimal state under the energy efficiency state b, and n is the number of division levels of the membership function for the energy efficiency state. By calculating the fuzzy entropy in the optimal state, the uncertainty of each energy efficiency indicator in this state can be quantified scientifically, providing a benchmark for subsequent comparison with the current state.

[0135]

[0136] The difference between the fuzzy entropy of the energy efficiency indicator in the current state and the optimal state is calculated , and compared with the preset threshold ΔH to determine the energy efficiency indicator to be optimized. By calculating the difference in fuzzy entropy, the difference in the degree of uncertainty of the energy efficiency indicators between the current state and the optimal state can be quantified, providing an objective basis for determining the indicator to be optimized. The preset threshold ΔH can be adjusted according to actual needs and evaluation standards.

[0137] If the fuzzy entropy of the energy efficiency indicator a in the current state H a is greater than the fuzzy entropy of the energy efficiency indicator a in the optimal state H , it indicates that the uncertainty of the energy efficiency indicator a in the current state is significantly higher than that in the optimal state, and the energy efficiency indicator a is determined as a high-risk energy efficiency indicator. The factors affecting the uncertainty of the energy efficiency indicator a are analyzed, and corresponding optimization measures are developed, such as maintenance of abnormal equipment, optimization of energy supply strategy, etc. By analyzing and optimizing the high-risk energy efficiency indicator, the uncertainty of the energy system in the park can be quickly reduced, and the energy efficiency level can be improved.

[0138] If the fuzzy entropy of the energy efficiency indicator a in the current state H a is less than the fuzzy entropy of the energy efficiency indicator a in the optimal state H , it indicates that the uncertainty of the energy efficiency indicator a in the current state is significantly lower than that in the optimal state, and the energy efficiency indicator a is determined as a potential risk energy efficiency indicator, which needs to be paid attention to and preventive measures are developed in advance.The energy efficiency indicator is determined as a potential risk energy efficiency indicator, although the current state uncertainty is low, but the potential risk factors with high optimal state fuzzy entropy still need to be concerned. For the energy efficiency indicator, preventive measures can be made in advance, such as establishing a device operation early warning mechanism, reserving emergency energy, etc., to promote the energy system of the park to approach the optimal state. By identifying the potential risk energy efficiency indicator, the potential risk can be identified in advance, and preventive measures can be taken to help avoid potential problems and ensure the efficient and stable operation of the energy system of the park.

[0139] In summary, the comprehensive energy efficiency evaluation method for the smart park of the present application firstly constructs a nonlinear metabolic matrix reflecting the energy correlation relationship between devices to accurately capture the complex energy interaction characteristics between devices in the park, and uses mixed integer nonlinear programming to solve the optimal metabolic flux distribution, realizing dynamic optimization and balance of system energy flow, thereby effectively improving energy utilization efficiency and reducing energy waste; secondly, define the interaction rule library of the metabolic network, and dynamically trigger the rules based on real-time sensor data to adjust the real-time metabolic flux of each device, so that the metabolic flux can accurately reflect the real-time changes of the park's energy demand and the abnormal fluctuations of the current device operation state, ensuring the stability and reliability of energy supply; then, calculate the quantitative value of each energy efficiency indicator according to the real-time metabolic flux, reflecting the energy efficiency level of the system under the current actual operating state, and use fuzzy entropy theory to quantify the uncertainty of each energy efficiency indicator under the current state, to reflect the uncertainty degree of the energy efficiency indicator under the current state, and according to the fuzzy entropy of the energy efficiency indicator under the current state, a dynamic weight is assigned, that is, the weight of the energy efficiency indicator is automatically adjusted according to its uncertainty degree, so that the energy efficiency indicator with low uncertainty plays a greater role in the comprehensive evaluation, thereby improving the accuracy and reliability of the energy efficiency evaluation. Finally, according to the optimal metabolic flux distribution, the quantitative value of each energy efficiency indicator under the optimal state is calculated, which is used to reflect the energy efficiency level of the park energy system under the ideal operating state, and then the fuzzy entropy of each energy efficiency indicator under the optimal state is calculated using the fuzzy entropy theory, to quantify the uncertainty of each energy efficiency indicator under the optimal state, to provide a comparison benchmark, and by comparing the fuzzy entropy of the energy efficiency indicator under the current state, the difference degree of the current state and the optimal state in terms of energy efficiency indicator uncertainty is quantified, to determine the to-be-optimized indicators, and identify high-risk and potential risks from the to-be-optimized indicators, for targeted optimization, to solve the uncertainty factors of the park energy system and improve the energy efficiency level, thereby ensuring the efficient and stable operation of the park energy system.

[0140] Embodiment two

[0141] Please refer to Figure 2 The present application provides a comprehensive energy efficiency evaluation system for a smart park, which comprises:

[0142] Metabolic flux module: used to construct a nonlinear metabolic matrix reflecting the energy correlation between devices, and solve the optimal metabolic flux distribution through mixed integer nonlinear programming;

[0143] Real-time metabolic flux module: used to define a rule library for interaction with the metabolic network, dynamically trigger rules and adjust the real-time metabolic flux of each device according to real-time sensor data of each device;

[0144] Comprehensive energy efficiency evaluation module: used to calculate the quantitative value of each energy efficiency indicator in the current state according to the real-time metabolic flux of each device, and quantize the uncertainty of the energy efficiency indicator in the current state through fuzzy entropy theory, dynamically allocate multi-objective weights, and calculate the comprehensive energy efficiency score;

[0145] Energy efficiency optimization module: used to calculate the quantitative value of each energy efficiency indicator in the optimal state according to the optimal metabolic flux distribution, compare the fuzzy entropy of the energy efficiency indicator in the current state and the optimal state, determine the energy efficiency indicator to be optimized, and identify high-risk and potential risks from the energy efficiency indicator to be optimized, and develop targeted optimization strategies.

[0146] Further optionally, the metabolic flux module is further used for:

[0147] According to the energy flow relationship of the park devices, the devices are defined as metabolic enzymes, and the energy flow is defined as metabolites, and a nonlinear metabolic matrix S = [S ij ] n×n is constructed, where n is the total number of park devices, S ij = f ij (v i , v j ), f ij is a nonlinear function, used to represent the energy correlation between the i-th device and the j-th device;

[0148] The device energy flow rate vector is defined as the metabolic flux vector v = [v1, v2, …, v n ] T , v i is the metabolic flux of the i-th device, representing the energy flow rate of the i-th device, v j is the metabolic flux of the j-th device, representing the energy flow rate of the j-th device;

[0149] A nonlinear metabolic flux balance equation S(v)·v = 0, v ≥ 0 is constructed, where 0 is a zero vector, representing the overall balance of park energy input and output;

[0150] Dynamic constraints are introduced, including energy supply and demand fluctuation constraints and device start-stop frequency limit constraints:

[0151] Using energy efficiency as the optimization objective and incorporating dynamic constraints, the optimal metabolic flux distribution is obtained through mixed-integer nonlinear programming. Reflecting the energy flow rate of each device under optimal energy efficiency, among which, Let be the metabolic flux of the i-th device under the optimal metabolic flux distribution.

[0152] Further optionally, the metabolic flux module is also used for:

[0153] Based on energy flow costs and differences in energy flow between equipment, the energy efficiency objective function F(v) is constructed as follows:

[0154] Among them, c i Let r be the energy flow cost coefficient for the i-th device. ij The penalty coefficient for the difference in energy flow between the i-th device and the j-th device;

[0155] Minimizing the energy efficiency objective function F(v) is taken as the objective of optimizing the mixed-integer nonlinear programming problem. The optimization objective formula is:

[0156]

[0157] stS(v)·v=0,v≥0;

[0158] The optimal metabolic flux distribution is obtained by solving the optimization objective of the mixed-integer nonlinear programming problem.

[0159] Further optionally, the real-time metabolic flux module is also used for:

[0160] Build a rule base that includes basic rules, preference rules, and exception rules;

[0161] For each device, a rule base is matched in real time based on its sensor data. If a rule is triggered, the real-time metabolic flux is adjusted using the following formula: v i，new =v i，old +Δv i，rule +Δv i，preference +Δv i，fault , where v i，old Let Δv be the metabolic flux of the i-th device before adjustment. i，rule For the i-th device, the flux change that triggers the basic rule is Δv i，preference For the i-th device, the flux change that triggers the preference rule is Δv fault The flux change that triggers the abnormal rule for the i-th device;

[0162] The adjusted real-time metabolic flux of each device is fed back to the nonlinear metabolic matrix S, and the energy efficiency objective function is recalculated.

[0163] Further optionally, in the comprehensive energy efficiency evaluation module, the formula for calculating the quantitative value of each energy efficiency indicator under the current state according to the actual metabolic flux of each device is:

[0164] wherein x a represents the quantitative value of the a-th energy efficiency indicator under the current state, reflecting the energy efficiency level of the system under the current actual operation state, the energy efficiency indicators including total energy consumption, carbon emission and energy cost, v i is the metabolic flux of the i-th device, e i，a is the unit energy flow coefficient of the a-th energy efficiency indicator of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n is the total number of park devices.

[0165] Further optionally, the comprehensive energy efficiency evaluation module is further used for:

[0166] For each energy efficiency indicator under the current state, the fuzzy entropy thereof is calculated, and the formula is:

[0167]

[0168] wherein H a is the fuzzy entropy of the a-th energy efficiency indicator under the current state, representing the uncertainty degree of the a-th energy efficiency indicator under the current state, μ a，b is the membership degree of the quantitative value of the a-th energy efficiency indicator under the current state to the energy efficiency state b, and n is the number of division levels of the membership function to the energy efficiency state.

[0169] According to the fuzzy entropy, a dynamic weight is distributed, and the formula is:

[0170] wherein w a reflects the importance adjustment of the a-th energy efficiency indicator under the current state in the uncertainty, A is the total number of energy efficiency indicators, and λ is a temperature coefficient for adjusting the sensitivity of weight distribution.

[0171] According to the dynamic weight energy efficiency indicator, a weighted sum is performed to obtain a comprehensive energy efficiency score Score, and the formula is:

[0172] wherein Norm(x a ) is the normalized value of the quantitative value of the a-th energy efficiency indicator under the current state, x a is the quantitative value of the a-th energy efficiency indicator under the current state, min(x a ) is the minimum value of all quantitative values of the a-th energy efficiency indicator in the actual operation, and max(x a ) is the maximum value of all quantitative values of the a-th energy efficiency indicator in the actual operation.

[0173] Further optionally, in the energy efficiency optimization module, the formula for calculating the quantitative value of each energy efficiency indicator under the optimal state according to the optimal metabolic flux distribution is:

[0174] wherein, represents the quantitative value of the a-th energy efficiency indicator under the optimal state, reflecting the ideal energy efficiency level under the ideal state of the system, and the energy efficiency indicators include total energy consumption, carbon emission, and energy cost, is the metabolic flux of the i-th device under the optimal metabolic flux distribution, e i，a is the unit energy flow coefficient of the a-th energy efficiency indicator of the i-th device, including unit energy consumption, unit carbon emission, and unit cost, and n is the total number of devices in the park.

[0175] Further optionally, the energy efficiency optimization module is further used for:

[0176] For each energy efficiency indicator under the optimal state, the fuzzy entropy thereof is calculated, and the formula is:

[0177]

[0178] wherein, is the fuzzy entropy of the a-th energy efficiency indicator under the optimal state, indicating the uncertainty degree of the a-th energy efficiency indicator under the optimal state, is the membership degree of the quantitative value of the a-th energy efficiency indicator under the optimal state under the energy efficiency state b, and n is the number of division levels of the membership function to the energy efficiency state;

[0179] the difference between the fuzzy entropy of the energy efficiency indicator under the current state and the optimal state is calculated and compared with a preset threshold ΔH;

[0180] If the a-th energy efficiency indicator is determined as a to-be-optimized energy efficiency indicator, indicating that there is a significant difference between the current state and the optimal state in the a-th energy efficiency indicator, which needs to be focused on and optimized;

[0181] If the a-th energy efficiency indicator is determined as a high-risk energy efficiency indicator, indicating that the uncertainty of the a-th energy efficiency indicator under the current state is significantly higher than that under the optimal state, and the factors affecting the uncertainty are analyzed, and corresponding optimization measures are developed;

[0182] If the a-th energy efficiency indicator is determined as a potential risk energy efficiency indicator, which is paid attention to and preventive measures are developed in advance.

[0183] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A comprehensive energy efficiency assessment method for smart parks, characterized in that, The method includes: A nonlinear metabolic matrix reflecting the energy correlation between devices is constructed, and the optimal metabolic flux distribution is solved by mixed integer nonlinear programming. Define an interaction rule base with the metabolic network, dynamically trigger rules and adjust the real-time metabolic flux of each device based on real-time sensor data from each device; Based on the real-time metabolic flux of each device, the quantitative values ​​of each energy efficiency index under the current state are calculated, and the uncertainty of the energy efficiency index under the current state is quantified by fuzzy entropy theory. Multi-objective weights are dynamically allocated, and the comprehensive energy efficiency score is calculated. Based on the optimal metabolic flux distribution, the quantitative values ​​of each energy efficiency index under the optimal state are calculated. The fuzzy entropy of the energy efficiency index under the current state and the optimal state is compared to determine the energy efficiency index to be optimized. High-risk and potential risks are identified from the energy efficiency index to be optimized, and targeted optimization strategies are formulated.

2. The comprehensive energy efficiency assessment method for smart parks according to claim 1, characterized in that, The construction of a nonlinear metabolic matrix reflecting the energy correlation between devices, and the solution of the optimal metabolic flux distribution through mixed-integer nonlinear programming, includes: Based on the energy flow relationships of equipment in the park, the equipment is defined as metabolic enzymes, and the energy flow is defined as metabolites. A nonlinear metabolic matrix S = [S ij ] n×n Where n is the total number of devices in the park, and S ij =f ij (v i v j ), f ij It is a nonlinear function used to represent the energy correlation between the i-th device and the j-th device; The energy flow rate vector of the equipment is defined as the metabolic flux vector v = [v1, v2, ..., v]. n ] T v i Let v be the metabolic flux of the i-th device, and v represent the energy flow rate of the i-th device. j Let be the metabolic flux of the j-th device, and let represent the energy flow rate of the j-th device. A nonlinear metabolic flux balance equation is constructed: S(v)·v=0,v≥0,where 0 is the zero vector, representing the overall balance between energy input and output in the park; Introducing dynamic constraints, including constraints on energy supply and demand fluctuations and constraints on the number of equipment start-ups and shutdowns: Using energy efficiency as the objective function and incorporating dynamic constraints, the optimal metabolic flux distribution is obtained through mixed-integer nonlinear programming. Reflecting the energy flow rate of each device under optimal energy efficiency, among which, Let be the metabolic flux of the i-th device under the optimal metabolic flux distribution.

3. The comprehensive energy efficiency assessment method for smart parks according to claim 2, characterized in that, The optimal metabolic flux distribution is obtained by using an energy efficiency objective function as the optimization objective and incorporating dynamic constraints through mixed-integer nonlinear programming. include: Based on energy flow costs and differences in energy flow between equipment, the energy efficiency objective function F(v) is constructed as follows: Among them, c i Let r be the energy flow cost coefficient for the i-th device. ij The penalty coefficient for the difference in energy flow between the i-th device and the j-th device; Minimizing the energy efficiency objective function F(v) is taken as the objective of optimizing the mixed-integer nonlinear programming problem. The optimization objective formula is: stS(v)·v=0,v≥0; The optimal metabolic flux distribution is obtained by solving the optimization objective of the mixed-integer nonlinear programming problem.

4. The comprehensive energy efficiency assessment method for smart parks according to claim 2, characterized in that, The defined interaction rule base with the metabolic network dynamically triggers rules and adjusts the real-time metabolic flux of each device based on real-time sensor data from each device, including: Build a rule base that includes basic rules, preference rules, and exception rules; For each device, a rule base is matched in real time based on its sensor data. If a rule is triggered, the real-time metabolic flux is adjusted using the following formula: v i，new =v i，old +Δv i，rule +Δv i，preference +Δv i，fault , where v i，old Let Δv be the metabolic flux of the i-th device before adjustment. i，rule For the i-th device, the flux change that triggers the basic rule is Δv i，preference For the i-th device, the flux change that triggers the preference rule is Δv fault The flux change that triggers the abnormal rule for the i-th device; The adjusted real-time metabolic flux of each device is fed back to the nonlinear metabolic matrix S, and the energy efficiency objective function is recalculated.

5. The comprehensive energy efficiency assessment method for smart parks according to claim 1, characterized in that, The quantitative values ​​of each energy efficiency index are calculated based on the real-time metabolic flux of each device under the current state, using the following formula: Where, x a This represents the quantified value of the *a*-th energy efficiency indicator under the current state, reflecting the system's energy efficiency level under its current actual operating conditions. Energy efficiency indicators include total energy consumption, carbon emissions, and energy costs. i Let e ​​be the metabolic flux of the i-th device. i，a Let be the unit energy flow coefficient of the a-th energy efficiency index of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n be the total number of devices in the park.

6. The comprehensive energy efficiency assessment method for smart parks according to claim 5, characterized in that, The process of quantifying the uncertainty of energy efficiency indicators under the current state using fuzzy entropy theory, dynamically allocating multi-objective weights, and calculating a comprehensive energy efficiency score includes: For each energy efficiency index under the current state, its fuzzy entropy is calculated using the following formula: m a，b ∈[0,1], Among them, H a Let μ be the fuzzy entropy of the a-th energy efficiency index in the current state, representing the degree of uncertainty of the a-th energy efficiency index in the current state. a，b Let n be the membership degree of the quantified value of the a-th energy efficiency index under the current state in energy efficiency state b, and n be the number of levels of energy efficiency state division by the membership function. Dynamic weights are assigned based on fuzzy entropy, using the following formula: Among them, w a This reflects the adjustment of the importance of the a-th energy efficiency indicator under uncertainty in the current state, where A is the total number of energy efficiency indicators and λ is the temperature coefficient used to adjust the sensitivity of weight allocation. The overall energy efficiency score (Score) is obtained by weighting and summing the dynamic weighted energy efficiency indicators. The formula is as follows: Among them, Norm(x a Let x be the normalized value of the quantified value of the a-th energy efficiency index under the current state. a Let x be the quantified value of the a-th energy efficiency index under the current state, min(x) a ) represents the minimum value among all quantified values ​​of the a-th energy efficiency index in actual operation, max(x) a ) represents the maximum value among all quantified values ​​of the a-th energy efficiency indicator in actual operation.

7. The comprehensive energy efficiency assessment method for smart parks according to claim 2, characterized in that, The quantitative values ​​of each energy efficiency index under optimal conditions are calculated based on the optimal metabolic flux distribution, using the following formula: in, This represents the quantified value of the a-th energy efficiency index under optimal conditions, reflecting the ideal energy efficiency level of the system under ideal conditions. Energy efficiency indicators include total energy consumption, carbon emissions, and energy costs. Let e ​​be the metabolic flux of the i-th device under the optimal metabolic flux distribution. i，a Let be the unit energy flow coefficient of the a-th energy efficiency index of the i-th device, including unit energy consumption, unit carbon emission and unit cost, and n be the total number of devices in the park.

8. The comprehensive energy efficiency assessment method for smart parks according to claim 7, characterized in that, The comparison of the fuzzy entropy of energy efficiency indicators in the current state and the optimal state determines the energy efficiency indicators to be optimized, and identifies high-risk and potential risks from the energy efficiency indicators to be optimized, and formulates targeted optimization strategies, including: For each energy efficiency index under optimal conditions, its fuzzy entropy is calculated using the following formula: in, Let be the fuzzy entropy of the a-th energy efficiency index under optimal conditions, representing the degree of uncertainty of the a-th energy efficiency index under optimal conditions. Let n be the membership degree of the quantified value of the a-th energy efficiency index under the optimal state in energy efficiency state b, and n be the number of levels of the membership function for energy efficiency state. Calculate the difference in fuzzy entropy between the energy efficiency index in the current state and the optimal state. Compare with the preset threshold ΔH; like The a-th energy efficiency index is then defined as the energy efficiency index to be optimized, indicating that there is a significant difference between the current state and the optimal state in terms of the a-th energy efficiency index, which needs to be focused on and optimized. like The a-th energy efficiency indicator is then identified as a high-risk energy efficiency indicator, indicating that the uncertainty of the a-th energy efficiency indicator under the current state is significantly higher than that under the optimal state. The factors affecting its uncertainty are analyzed, and corresponding optimization measures are formulated. like The a-th energy efficiency indicator will be identified as a potential risk energy efficiency indicator, and attention will be paid to it and preventive measures will be developed in advance.

9. A comprehensive energy efficiency assessment system for smart parks, used to implement the comprehensive energy efficiency assessment method for smart parks as described in any one of claims 1 to 8, characterized in that, The system includes: Metabolic flux module: used to construct a nonlinear metabolic matrix that reflects the energy correlation between devices, and to solve for the optimal metabolic flux distribution through mixed integer nonlinear programming; Real-time metabolic flux module: used to define the rule base for interaction with the metabolic network, dynamically trigger rules and adjust the real-time metabolic flux of each device based on real-time sensor data from each device; The comprehensive energy efficiency assessment module is used to calculate the quantitative values ​​of each energy efficiency index under the current state based on the real-time metabolic flux of each device, and to quantify the uncertainty of the energy efficiency index under the current state through fuzzy entropy theory, dynamically allocate multi-objective weights, and calculate the comprehensive energy efficiency score. Energy efficiency optimization module: It is used to calculate the quantitative values ​​of each energy efficiency index under the optimal state based on the optimal metabolic flux distribution, compare the fuzzy entropy of the energy efficiency index under the current state and the optimal state, determine the energy efficiency index to be optimized, identify high-risk and potential risks from the energy efficiency index to be optimized, and formulate targeted optimization strategies.

Images