A near-zero energy consumption building energy supply method and system based on multi-objective optimization

Abstract

The present invention discloses an energy supply method and system for nearly zero - energy buildings based on multi - objective optimization, which relates to the technical field of building energy efficiency and renewable energy applications. The method includes collecting energy consumption data in a building and constructing a multi - objective evaluation system according to the energy consumption data; optimizing the weights of the multi - objective evaluation system to obtain the optimal weight coefficients; combining the optimal weight coefficients with a solar photovoltaic system and a ground - source heat pump system to establish a renewable energy supply strategy model; and based on the renewable energy supply strategy model, outputting energy supply scheduling instructions for each time period of the building to adjust the operating parameters of the solar photovoltaic system and the ground - source heat pump system. The present invention ensures the balance between energy efficiency and comfort by real - time adjusting the equipment operation mode, avoiding the problems of common energy efficiency waste or insufficient comfort in traditional methods. Through intelligent scheduling and optimization, it can effectively reduce the energy consumption of buildings and improve the sustainability of their energy utilization.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy efficiency and renewable energy application, and in particular to a near-zero energy building energy supply method and system based on multi-objective optimization. Background Art

[0002] With the intensification of the global energy crisis and increasingly stringent environmental protection requirements, the optimization and innovation of energy utilization in the construction industry has become an important topic of concern in modern society. In particular, the concept of zero energy buildings (ZEBs) has gradually emerged and has become one of the directions of sustainable building design. Zero energy buildings achieve zero net energy exchange with external energy networks by optimizing the energy demand of buildings, improving energy efficiency and combining the supply of renewable energy. In order to achieve this goal, renewable energy technologies such as solar photovoltaic systems and ground source heat pump systems are increasingly used in the construction field. These technologies reduce the energy consumption of buildings to a certain extent by directly utilizing natural resources such as solar energy and geothermal energy, and realize a self-sufficient energy supply model for buildings. In addition, as an important part of building energy optimization, energy management systems (EMS) are gradually moving towards efficient and refined management with the help of intelligent sensing technology, data acquisition technology and control algorithms.

[0003] However, the existing energy supply methods for zero-energy buildings still have some shortcomings in terms of energy collection, storage and scheduling. Most current zero-energy buildings rely mainly on a single energy source when designing energy management systems, or simply match the use time of renewable energy with the load demand of the building, and fail to fully consider the synergy between different energy systems. In addition, the optimization algorithms in the existing technologies are usually guided by a single goal and fail to comprehensively consider the multi-dimensional needs of the building, such as comfort, energy efficiency, environmental impact, etc., resulting in uneven energy efficiency performance of the system under different operating modes. When faced with different climatic conditions, building structures and usage methods, the lack of flexible scheduling strategies can easily lead to insufficient or wasteful energy supply. Therefore, how to establish a multi-objective optimization energy supply strategy and accurately schedule different energy systems to improve the energy self-sufficiency and comfort of buildings is still a technical problem that needs to be solved in this field. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention proposes a near-zero energy building energy supply method and system based on multi-objective optimization.

[0005] Therefore, the present invention provides a near-zero energy building energy supply method based on multi-objective optimization, which can solve the problems mentioned in the background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a near-zero energy consumption building energy supply method based on multi-objective optimization, which includes:

[0008] Collect energy consumption data in the building and construct a multi-objective evaluation system according to the energy consumption data;

[0009] Optimize the weights of the multi-objective evaluation system to obtain the optimal weight coefficients;

[0010] Combine the optimal weight coefficients with a solar photovoltaic system and a ground source heat pump system to establish a renewable energy supply strategy model;

[0011] Based on the renewable energy supply strategy model, output energy supply scheduling instructions for each time period of the building, and adjust the operating parameters of the solar photovoltaic system and the ground source heat pump system.

[0012] As a preferred solution of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the energy consumption data includes temperature sensor data, humidity sensor data, light sensor data, and electrical load data;

[0013] The temperature sensor data is collected according to temperature sensors arranged in each functional area, corridor, staircase, and the building facade of the building;

[0014] The humidity sensor data is collected according to humidity sensors arranged in each functional area of the building and the building facade;

[0015] The light sensor data is collected according to light sensors arranged in the indoor and outdoor window areas and the roof area of the building;

[0016] The electrical load data is collected according to the building's electrical system.

[0017] As a preferred solution of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the multi-objective evaluation system includes indoor comfort index, energy consumption index, and daylighting index;

[0018] The indoor comfort index includes temperature and humidity deviation, and the indoor comfort index is calculated according to the temperature and humidity deviation;

[0019] The temperature and humidity deviation is calculated by comparing the temperature sensor data with a preset temperature comfort range and comparing the humidity sensor data with a preset relative humidity comfort range;

[0020] Standardize the electricity consumption load data according to the building's usable area to obtain the electricity consumption per unit area, and compare it with the reference value specified in the building energy efficiency design standard to generate the energy consumption index;

[0021] Calculate the indoor daylighting factor based on the data from the light sensors, and compare the daylighting factor with the minimum required value specified in the building daylighting standard to generate the daylighting index; where the daylighting factor is the percentage of the indoor natural daylight illuminance to the total outdoor horizontal natural illuminance at the same moment.

[0022] As a preferred embodiment of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the calculation of the temperature and humidity deviation value is shown in the following formula:

[0023]

[0024] where, T is the measured temperature value, T set is the set temperature value, H is the measured humidity value, H set is the set humidity value, ΔT max and ΔH max are the allowable maximum temperature and humidity deviations respectively, and α and β are weight coefficients;

[0025] The calculation of the indoor comfort index is shown in the following formula:

[0026]

[0027] where, P i is the electricity power at time period i, t i is the duration of time period i, A is the building area, E0 is the reference energy consumption value, λ is the temperature correction coefficient, and T0 is the outdoor design temperature;

[0028] The calculation of the daylighting index is shown in the following formula:

[0029]

[0030] where, E in is the indoor natural daylight illuminance, E out is the total outdoor horizontal natural illuminance, θ is the solar altitude angle, d is the distance from the measuring point to the window, d0 is the reference distance, and μ and ρ are correction coefficients.

[0031] As a preferred embodiment of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the optimal weight coefficient is obtained by optimizing the objective function of the multi-objective evaluation system according to the multi-objective particle swarm optimization algorithm;

[0032] The multi-objective particle swarm optimization algorithm includes particle velocity update, particle position update, and fitness function;

[0033] The update of the particle velocity is expressed by the following formula:

[0034] V id (t + 1)= ω·V id (t)+ c1·r1·[P id (t)- X id (t)]+ c2·r2·[P gd (t)- X id (t)]+ c3·r3·(1 - ∑w i )

[0035] Wherein, V id is the velocity of the i-th particle in the d-th dimension, X id is the position of the i-th particle in the d-th dimension, P id (t) is the individual optimal position, P gd is the global optimal position, ω is the inertia weight, c1, c2, c3 are learning factors, and r1, r2, r3 are random numbers between [0, 1];

[0036] The update of the particle position is expressed by the following formula:

[0037]

[0038] Wherein, σ is the adaptive adjustment coefficient, and F(t) is the objective function value of the current iteration.

[0039] As a preferred embodiment of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the fitness function is shown in the following formula:

[0040]

[0041] Wherein, η is the gradient penalty factor, θ is the weight balance factor, F is the objective function of the multi-objective evaluation system, and w1, w2, w3 are the weight coefficients in the objective function of the multi-objective evaluation system;

[0042] During the iteration of the multi-objective particle swarm optimization algorithm, the weight combination obtained in each iteration is evaluated. When the change in the objective function value of consecutive preset iterations is less than the first preset threshold and the maximum particle velocity is less than the second preset threshold, the weight combination corresponding to the current global optimal particle is output as the optimal weight coefficient.

[0043] As a preferred embodiment of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the construction of the renewable energy supply strategy model includes:

[0044] Collect the operation data of the solar photovoltaic system and the ground source heat pump system, and construct a device performance characteristic model;

[0045] Construct the renewable energy supply strategy model according to the optimal parameters and the device performance characteristic model;

[0046] The renewable energy supply strategy model further includes optimizing the renewable energy supply strategy model according to a neural network model and a loss function, and outputting an optimal supply strategy;

[0047] The device performance characteristic model includes the power generation efficiency of the solar photovoltaic system and the coefficient of performance of the ground source heat pump system.

[0048] As a preferred embodiment of the near-zero energy consumption building energy supply method based on multi-objective optimization according to the present invention, wherein: the objective function of the renewable energy supply strategy model is represented by the following formula:

[0049] S(t) = arg min{L + μ1[P PV (t) - P PV,max (t)] + + μ2[Q HP (t) - Q HP,max (t)] +

[0050] + μ3∑δ(t)}

[0051] Wherein, P PV (t) and Q HP (t) are the solar photovoltaic power generation power at the current moment and the actual energy supply load of the ground source heat pump at the current moment respectively, P PV,max (t) and Q HP,max (t) are the maximum power generation power limit of the solar photovoltaic system and the maximum energy supply load limit of the ground source heat pump system respectively, δ(t) is the device start-stop penalty term, μ1, μ2, μ3 are weight coefficients, S(t) is the optimization objective function value of the supply strategy, and L is the loss function.

[0052] In a second aspect, an embodiment of the present invention provides a near-zero energy consumption building energy supply system based on multi-objective optimization, which includes:

[0053] A data acquisition module, configured to acquire the energy consumption data in a building and construct a multi-objective evaluation system according to the energy consumption data;

[0054] A parameter optimization module, configured to optimize the weights of the multi-objective evaluation system to obtain optimal weight coefficients;

[0055] A model construction module, configured to combine the optimal weight coefficients with a solar photovoltaic system and a ground source heat pump system to establish a renewable energy supply strategy model;

[0056] A control module, configured to output energy supply scheduling instructions for each time period of a building based on the renewable energy supply strategy model, and adjust the operating parameters of a solar photovoltaic system and a ground source heat pump system.

[0057] The beneficial effects of the present invention are as follows: by collecting temperature, humidity, lighting, and electricity load data inside and outside the building, and combining the particle swarm optimization algorithm to optimize the weights of three objectives: indoor comfort, energy consumption, and natural lighting, a renewable energy supply strategy model based on a multi-objective evaluation system is established. Through in-depth analysis of the real-time operation data of the solar photovoltaic system and the ground source heat pump system, combined with the optimal weight coefficients, the model can accurately calculate the energy supply demand for each time period and output corresponding scheduling instructions. This scheduling strategy can optimize the operating parameters of the solar photovoltaic and ground source heat pump systems on the premise of ensuring indoor temperature and humidity comfort and natural lighting, and maximize the energy utilization efficiency. Especially in the face of dynamically changing meteorological conditions, electricity loads, and system operating states, this solution ensures the balance of energy efficiency and comfort by adjusting the device operating mode in real time, avoiding common problems of energy efficiency waste or insufficient comfort in traditional methods. Finally, through intelligent scheduling and optimization, this method can effectively reduce the energy consumption of the building and improve the sustainability of its energy utilization. Description of the Drawings

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0059] Figure 1 It is a flowchart of an energy supply method for a nearly zero-energy consumption building based on multi-objective optimization. Detailed Embodiments

[0060] To make the above objects, features, and advantages of the present invention more obvious and understandable, the detailed embodiments of the present invention will be described below in conjunction with the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0061] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0062] Secondly, the so-called "one embodiment" or "embodiment" herein refers to specific features, structures or characteristics that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0063] The present invention is described in detail in conjunction with schematic diagrams. When detailing the embodiments of the present invention, for the sake of convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions of length, width and depth should be included.

[0064] At the same time, in the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper, lower, inner and outer" is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0065] Unless otherwise clearly defined and limited in the present invention, the terms "installed, connected, connected" shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection or an integral connection; it may also be a mechanical connection, an electrical connection or a direct connection, or may be indirectly connected through an intermediate medium, or may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0066] Embodiment 1

[0067] Referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a near-zero energy consumption building energy supply method based on multi-objective optimization, including:

[0068] S1: Collect the temperature sensor data, humidity sensor data, light sensor data and electricity load data in the building, and store the temperature sensor data, the humidity sensor data, the light sensor data and the electricity load data in a database;

[0069] Temperature sensors, humidity sensors, and light sensors are deployed in the interior and exterior areas of the building. Among them, the temperature sensors are deployed in each functional area, corridor, stairwell, and the exterior facade of the building indoors. The humidity sensors are deployed in each functional area and the exterior facade of the building indoors. The light sensors are deployed in the window areas and roof areas inside and outside the building. The indoor and outdoor temperature data is collected by the temperature sensors, the indoor and outdoor relative humidity data is collected by the humidity sensors, and the indoor and outdoor light intensity data is collected by the light sensors. At the same time, the real-time electricity load data collected by the building's electricity consumption system is recorded. The temperature sensor data, humidity sensor data, light sensor data, and electricity load data are transmitted to the data processing unit through a wired communication network or a wireless communication network, and the temperature sensor data, humidity sensor data, light sensor data, and electricity load data received by the data processing unit are stored in the database according to the time series.

[0070] S2: Establish an indoor comfort index based on the temperature sensor data and the humidity sensor data, establish an energy consumption index based on the electricity load data, establish a natural lighting index based on the light sensor data, and construct a multi-objective evaluation system.

[0071] Retrieve the temperature sensor data and the humidity sensor data from the database, compare the temperature sensor data with the preset temperature comfort range of 18°C - 26°C, compare the humidity sensor data with the preset relative humidity comfort range of 40% - 70%, and calculate the temperature and humidity deviation value. The temperature and humidity deviation can be expressed by the following formula:

[0072]

[0073] Among them, T is the measured temperature value, T set is the set temperature value (18°C - 26°C), H is the measured humidity value, H set is the set humidity value (40% - 70%), ΔT max and ΔH max are the allowable maximum temperature and humidity deviations respectively, and α and β are weight coefficients and α + β = 1.

[0074] Substitute the temperature and humidity deviation value into the preset PMV-PPD human comfort model to obtain the indoor comfort index, as shown in the following formula:

[0075]

[0076] Among them, M is the human metabolic rate, A Du is the human body surface area, and γ is the correction coefficient.

[0077] Retrieve the electricity load data from the database, standardize the electricity load data according to the building's usable area to obtain the electricity consumption per unit area, and compare it with the benchmark value specified in the building energy efficiency design standard to generate an energy consumption index, as shown in the following formula:

[0078]

[0079] where P i is the electricity power at time period i, t i is the duration of time period i, A is the building area, E0 is the benchmark energy consumption value, λ is the temperature correction coefficient, and T0 is the outdoor design temperature.

[0080] Retrieve the light sensor data from the database, calculate the indoor daylighting factor, where the daylighting factor is the percentage of the indoor natural daylight illumination to the outdoor horizontal plane natural total illumination at the same moment, and compare the daylighting factor with the minimum requirement value of 2% specified in the building daylighting standard to generate a natural daylighting index, as shown in the following formula:

[0081]

[0082] where E in is the indoor natural daylight illumination, E out is the outdoor horizontal plane natural total illumination, θ is the solar altitude angle, d is the distance from the measuring point to the window, d0 is the reference distance, and μ and ρ are correction coefficients.

[0083] Combine the indoor comfort index, the energy consumption index, and the natural daylighting index through linear weighting to construct the objective function of the multi-objective evaluation system.

[0084] F = w1·I c + w2·(1 - I e ) + w3·I l

[0085] where w1, w2, and w3 are weight coefficients and satisfy w1 + w2 + w3 = 1.

[0086] S3: Optimize the weights of the indoor comfort index, the energy consumption index, and the natural daylighting index through the particle swarm algorithm to obtain the optimal weight coefficients;

[0087] Based on the indoor comfort index, the energy consumption index, and the natural daylighting index, optimize the weight coefficients using an improved multi-objective particle swarm optimization algorithm. The improved multi-objective particle swarm optimization algorithm sets the initial population size to 50, the maximum number of iterations to 200, the learning factors to 2.05 and 2.05 respectively, and the range of the inertia weight to be from 0.4 to 0.9 and linearly decreasing with the number of iterations.

[0088] During the iteration process of the particle swarm optimization algorithm, each particle contains weight information in three dimensions, and the sum of the weights in each dimension is 1. The velocity update of the particle adopts an adaptive inertia weight strategy;

[0089] More specifically, each particle is represented as a weight vector P(w1, w2, w3), and the velocity update formula of the particle is:

[0090] V id (t + 1) = ω·V id (t) + c1·r1·[P id (t) - X id (t)] + c2·r2·[P gd (t) - X id (t)] + c3·r3·(1 - ∑w i )

[0091] where, V id is the velocity of the i-th particle in the d-th dimension, X id is the position of the i-th particle in the d-th dimension, P id (t) is the individual optimal position, P gd is the global optimal position, ω is the inertia weight, c1, c2, c3 are learning factors, and r1, r2, r3 are random numbers in the range of [0, 1].

[0092] The position update of the particle introduces a contraction factor, and at the same time, a weight balance penalty term is added to the fitness function;

[0093] Then the particle position update is shown as follows:

[0094]

[0095] where, σ is the adaptive adjustment coefficient, and F(t) is the objective function value of the current iteration.

[0096] Its fitness function is shown as follows:

[0097]

[0098] where, η is the gradient penalty factor, and θ is the weight balance factor;

[0099] Evaluate the weight combination obtained in each iteration. When the change in the objective function value for 20 consecutive iterations is less than the preset threshold of 0.001 and the maximum particle velocity is less than the preset threshold of 0.0001, output the weight combination corresponding to the current global optimal particle as the optimal weight coefficient, that is:

[0100] |F(t) - F(t - 1)| ≤ ε and max(|Vid |) ≤ δ

[0101] The final output is the combination of weight coefficients (w1*, w2*, w3*) that satisfies the optimal fitness function as the optimal weight coefficients.

[0102] S4: Combine the optimal weight coefficients with the power generation of the solar photovoltaic system and the cooling capacity of the ground source heat pump system to establish a renewable energy supply strategy model;

[0103] Collect operation parameters such as battery module temperature, solar radiation intensity, power generation power, component surface dust coverage rate, and component aging coefficient from the solar photovoltaic system in real time; collect operation parameters such as soil temperature field distribution, system inlet and outlet water temperature, cooling / heating power, compressor frequency, and heat exchanger efficiency from the ground source heat pump system in real time; establish an equipment performance characteristic model considering multiple influencing factors based on these parameters. Among them, the photovoltaic system efficiency model considers the coupled effects of temperature effect, radiation intensity effect, incident angle effect, dust attenuation effect, and aging effect, and the ground source heat pump performance coefficient model considers the combined effects of temperature difference effect, soil temperature effect, part-load characteristics, and heat exchanger efficiency.

[0104] Furthermore, establish an energy supply strategy model based on the optimal weight coefficients (w1*, w2*, w3*). First, establish equipment performance characteristic equations, including the power generation efficiency of the solar photovoltaic system and the performance coefficient of the ground source heat pump system.

[0105] Among them, the power generation efficiency of the solar photovoltaic system can be expressed by the following formula:

[0106]

[0107] Among them, η0 is the standard condition efficiency, β is the temperature coefficient, T c is the component temperature, T ref is the reference temperature of 25 °C, G is the radiation intensity, G0 is the standard radiation intensity of 1000 W / m 2 , θ is the incident angle, κ and ψ are correction coefficients;

[0108] The calculation of the performance coefficient of the ground source heat pump system is shown in the following formula:

[0109]

[0110] Among them, COP r is the rated performance coefficient, ΔT is the temperature difference between the inlet and outlet water, ΔT r is the rated temperature difference, T s is the soil temperature, T sr is the soil reference temperature, Q is the actual load, Q r is the rated load, β1, α2, and ξ are correction coefficients.

[0111] Furthermore, a deep neural network prediction model is used to predict the predicted values of the device operation parameters for the next 48 time periods, including the power generation power of the photovoltaic system, the cooling / heating power of the ground source heat pump, the total system efficiency, etc. It adopts a 5-layer deep neural network structure. The input layer includes meteorological forecast data (temperature, humidity, radiation, etc.), load prediction data (cooling load, heating load, electrical load), device status data (efficiency, COP, etc.), and optimal weight coefficients. Each of the three hidden layers is set with 64, 32, and 16 neurons respectively, uses the ReLU activation function, and a Dropout layer is added between each layer to prevent overfitting. The output of the neuron in the l-th layer of this neural network is:

[0112]

[0113] Among them, W (l) is the weight matrix, b (l) is the bias vector, f is the ReLU activation function, φ (l) is the weight adjustment coefficient between layers, h (l) is the output of the l-th layer of the neural network, h (l-1) is the output of the (l - 1)-th layer of the neural network, I c is the comfort index, I e is the energy consumption index, I l is the daylighting index.

[0114] Furthermore, a loss function for the energy supply strategy model is designed. This loss function consists of four parts: (1) the prediction error term, which measures the deviation between the predicted value and the actual value; (2) the temporal smoothing term, which ensures the time continuity of the prediction result; (3) the device constraint term, which guarantees that the prediction result meets the device operation limitations; (4) the multi-objective balance term, which coordinates the three objectives of indoor comfort, energy consumption efficiency, and natural daylighting according to the optimal weight coefficients. The balance between the terms is achieved through adaptive weight coefficients, and the weight coefficients are dynamically adjusted during the training process. As shown in the following formula:

[0115]

[0116] Among them, y t is the collected system output value, is the output value predicted by the model, λ1, λ2, and λ3 are the balance coefficients, I i,ref is the reference value of the i-th index, L is the loss function of the energy supply strategy model, η PV,t is the solar photovoltaic efficiency at the current moment, η PV,t-1 is the solar photovoltaic efficiency at the previous moment, COP t is the performance coefficient of the ground source heat pump at the current moment, COP t-1 is the performance coefficient of the ground source heat pump at the previous moment, I i,tis the actual value of the i-th index at the current moment.

[0117] A 48-hour rolling time domain is adopted, and the prediction model is updated every 1 hour to receive the latest meteorological data, load data, and equipment operation data in real time. In each prediction cycle, first, the predicted values of the equipment operation parameters are obtained based on the deep neural network, and then considering the actual operation restrictions such as equipment start-stop constraints, ramp constraints, and capacity constraints, a mixed-integer programming method is used to solve the optimal operation strategy. To improve the optimization efficiency, the 48-hour optimization cycle is divided into 6 sub-cycles of 8 hours each, and a segmented optimization method is adopted. The finally obtained functional strategy model can be shown by the following formula:

[0118] S(t) = arg min{L + μ1[P PV (t) - P PV,max (t)] + + μ2[Q HP (t) - Q HP,max (t)] +

[0119] + μ3∑δ(t)}

[0120] where P PV (t) and Q HP (t) are the solar photovoltaic power generation power at the current moment and the actual energy supply load of the ground source heat pump at the current moment respectively, P PV,max (t) and Q HP,max (t) are the maximum power generation power limit of the solar photovoltaic system and the maximum energy supply load limit of the ground source heat pump system respectively, δ(t) is the equipment start-stop penalty term, μ1, μ2, μ3 are weight coefficients, S(t) is the optimization objective function value of the energy supply strategy, and L is the loss function.

[0121] S5: Based on the renewable energy supply strategy model, output the energy supply scheduling instructions for each period of the building, and adjust the operation parameters of the solar photovoltaic system and the ground source heat pump system.

[0122] Based on the renewable energy supply strategy model, first, the building energy supply demand and equipment operation parameters in future periods are predicted through a deep neural network model, including the power generation power of the solar photovoltaic system and the cooling / heating power of the ground source heat pump. Combining with the actual operation data, meteorological forecast data and load prediction data, the operation status of the equipment is adjusted in real time. Through the optimized energy supply strategy model, considering the multi-objective balance of indoor comfort, energy efficiency and natural lighting, in each period, the model calculates the optimal power generation efficiency of the solar photovoltaic system and the performance coefficient of the ground source heat pump, and then determines its optimal operation parameters. The energy supply scheduling instructions output by the model will, according to the actual factors such as the maximum power limit of the current system, equipment start-stop constraints, temperature difference, load, etc., adjust the working status of the solar photovoltaic modules (such as adjusting the module temperature, radiation intensity, etc.) and the operation mode of the ground source heat pump system (such as adjusting the water temperature difference between inlet and outlet, load regulation, etc.) in real time, ensuring that each system meets the building's energy demand with the optimal efficiency in each period, achieving an overall balance between efficient energy utilization and building comfort.

[0123] Furthermore, this embodiment also provides a near-zero energy consumption building energy supply system based on multi-objective optimization, including:

[0124] A data acquisition module, which is used to collect the energy consumption data in the building and construct a multi-objective evaluation system according to the energy consumption data;

[0125] A parameter optimization module, which is used to optimize the weights of the multi-objective evaluation system to obtain the optimal weight coefficients;

[0126] A model construction module, which is used to combine the optimal weight coefficients with the solar photovoltaic system and the ground source heat pump system to establish a renewable energy supply strategy model;

[0127] A control module, which is used to output the energy supply scheduling instructions for each period of the building based on the renewable energy supply strategy model and adjust the operation parameters of the solar photovoltaic system and the ground source heat pump system.

[0128] This embodiment also provides a computer device, which is applicable to the case of the near-zero energy consumption building energy supply method based on multi-objective optimization, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the near-zero energy consumption building energy supply method based on multi-objective optimization as proposed in the above embodiment.

[0129] The computer device may be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball, or touchpad provided on the outer shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0130] This embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method for supplying energy to a nearly zero-energy consumption building based on multi-objective optimization as proposed in the above embodiment.

[0131] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. Technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0132] Embodiment 2

[0133] This is the second embodiment of the present invention. This embodiment provides a method for supplying energy to a nearly zero-energy consumption building based on multi-objective optimization. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0134] In this embodiment, a 25-story intelligent office building located in a provincial capital city was selected as the experimental object, with a total building area of approximately 45,000 square meters. Before the experiment began, a comprehensive sensor layout work was first carried out on the building. PT100 platinum resistance temperature sensors were used for temperature sensors, with a measurement accuracy of ±0.1°C. A total of 375 measurement points were arranged in various functional areas inside the building, and 48 measurement points were arranged on the exterior facade; high-precision capacitive humidity sensors were selected for relative humidity sensors, with a measurement range of 0-100%RH and an accuracy of ±2%RH. 280 measurement points were arranged indoors, and 32 measurement points were arranged on the exterior facade; silicon photocell illuminometers were used for light sensors, with a measurement range of 0-200,000 lux and an accuracy of ±3%. 156 measurement points were arranged in the indoor and outdoor window areas, and 24 measurement points were arranged in the roof area. All sensors were connected to the data collector through the RS485 bus, using the Modbus-RTU communication protocol, and the sampling frequency was 1 minute / time. At the same time, a power quality analyzer was installed in the building's power distribution room to monitor the electricity load data of each area in real time.

[0135] To verify the innovation and practicality of the present invention, the experiment was carried out in three stages: In the first stage, a traditional building control scheme was adopted, relying only on simple thermostats and timing control strategies; in the second stage, the basic scheme of the present invention was applied, introducing a multi-objective evaluation system and particle swarm optimization algorithm; in the third stage, on the basis of the second stage, the parameter configuration of the energy supply strategy model was further optimized. The experiment lasted for three months, covering different weather conditions and building usage situations. During the data collection process, special attention was paid to the system operation status under different scenarios such as weekdays and weekends, sunny days and cloudy days. The solar photovoltaic system used polycrystalline photovoltaic modules with a total installed capacity of 500 kWp, and the ground source heat pump system included 200 vertical buried tube heat exchangers with a drilling depth of 120 meters and a total system cooling capacity of 2000 kW.

[0136] During the implementation of the experiment, the training of the deep neural network prediction model used the historical data of the first two months as the training set and the data of the last month as the test set. The Adam optimizer was used during the model training process, the initial value of the learning rate was set to 0.001, and a learning rate decay strategy was used, decaying to 0.9 times the original every 50 epochs. To prevent overfitting, an early stopping strategy was adopted during the training process, and the training was stopped when the validation set loss did not improve for 10 consecutive epochs. The prediction accuracy of the model reached more than 90% on the test set, meeting the requirements of engineering applications. The following experimental data as shown in Table 1 were finally obtained:

[0137] Table 1 Experimental data of multi-objective optimization of intelligent buildings

[0138]

[0139]

[0140] Through the analysis of experimental data, it can be clearly seen that the solution of the present invention has significant advantages compared with the traditional control solution, and the optimized solution further improves the system performance. The specific analysis is as follows:

[0141] Under the traditional control solution, the indoor temperature is generally on the high side (26.8°C - 27.5°C), and the relative humidity is on the low side (35% - 38%), at the edge of the human comfort range. While the solution of the present invention precisely controls the indoor temperature between 23.5°C and 24.2°C, and maintains the relative humidity in the ideal range of 52% - 58%. The optimized solution further stabilizes the temperature within the best range of 23.1°C - 23.4°C. This shows that the present invention realizes more precise indoor environment regulation through a multi-objective evaluation system and a deep learning prediction model.

[0142] Under similar outdoor illuminance conditions, the solution of the present invention significantly improves the indoor daylighting effect. Taking a sunny weekday as an example, the indoor illuminance of the traditional solution is only 320 lux, while that of the present invention reaches 460 lux, and the optimized solution is further increased to 485 lux, with an increase of 51.6%. Even under cloudy conditions, the solution of the present invention can maintain a good indoor illuminance level, which benefits from the precise control of the natural daylighting index and the deep neural network prediction model introduced in the present invention.

[0143] The energy consumption data per unit area shows that the energy consumption on weekdays of the traditional solution is 0.185 kWh / m 2 , and that of the solution of the present invention is reduced to 0.142 kWh / m 2 . The energy-saving effect reaches 23.2%. The optimized solution further reduces the energy consumption to 0.135 kWh / m 2 . The overall energy saving compared with the traditional solution is 27%. At the same time, the solution of the present invention significantly improves the operating efficiency of the renewable energy system: the photovoltaic power generation efficiency is increased from 15.2% to 17.8% (under sunny conditions), and the highest reaches 18.4% after optimization, with an increase of 21.1%; the heat pump COP value is increased from 3.2 to 4.1, and the highest reaches 4.5 after optimization, with an increase of 40.6%. These improvements benefit from the real-time optimization of the system operating parameters by the improved particle swarm algorithm adopted in the present invention, and the precise regulation of the equipment operating state by the energy supply strategy model.

[0144] Data shows that the solution of the present invention can maintain a stable control effect under different weather conditions and building usage patterns. Taking the indoor temperature as an example, the temperature fluctuation ranges are 0.7°C and 0.4°C under sunny and cloudy conditions respectively, while the temperature fluctuation of the traditional solution reaches 1.2°C. This shows that the present invention successfully realizes the adaptive optimization of building environment control through a multi-objective evaluation system and a deep learning prediction model.

[0145] In summary, the experimental data fully demonstrate the significant advantages of the present invention in improving the quality of the indoor building environment and enhancing energy utilization efficiency. Through the innovative multi-objective evaluation system, the improved particle swarm optimization algorithm, and the prediction model based on deep learning, the present invention achieves precision, intelligence, and energy conservation in building environment control, and has important practical value and promotional significance.

[0146] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A near-zero energy consumption building energy supply method based on multi-objective optimization, characterized in that: Including: Collecting the energy consumption data in the building and constructing a multi-objective evaluation system according to the energy consumption data; Optimizing the weights of the multi-objective evaluation system to obtain the optimal weight coefficients; Combining the optimal weight coefficients with a solar photovoltaic system and a ground source heat pump system to establish a renewable energy supply strategy model; Based on the renewable energy supply strategy model, outputting the energy supply scheduling instructions for each time period of the building and adjusting the operating parameters of the solar photovoltaic system and the ground source heat pump system; The energy consumption data includes temperature sensor data, humidity sensor data, light sensor data and electrical load data; The multi-objective evaluation system includes indoor comfort indicators, energy consumption indicators and daylighting indicators; The optimal weight coefficients are obtained by optimizing the objective function of the multi-objective evaluation system according to the multi-objective particle swarm optimization algorithm; The multi-objective particle swarm optimization algorithm includes particle velocity update, particle position update and fitness function; During the iteration of the multi-objective particle swarm optimization algorithm, the weight combinations obtained in each iteration are evaluated. When the change in the objective function value of consecutive preset iterations is less than the first preset threshold and the maximum particle velocity is less than the second preset threshold, the weight combination corresponding to the current global optimal particle is output as the optimal weight coefficient; The construction of the renewable energy supply strategy model includes: Collecting the operation data of the solar photovoltaic system and the ground source heat pump system and constructing an equipment performance characteristic model; Constructing the renewable energy supply strategy model according to the optimal weight coefficients and the equipment performance characteristic model; The renewable energy supply strategy model also includes optimizing the renewable energy supply strategy model according to a neural network model and a loss function and outputting an optimal energy supply strategy; The equipment performance characteristic model includes the power generation efficiency of the solar photovoltaic system and the performance coefficient of the ground source heat pump system; The objective function of the renewable energy supply strategy model is represented by the following formula: ， Among them, and are the solar photovoltaic power generation power at the current moment and the actual energy supply load of the ground source heat pump at the current moment respectively, and are the maximum power generation power limit of the solar photovoltaic system and the maximum energy supply load limit of the ground source heat pump system respectively, is the equipment start-stop penalty term, , , are the weight coefficients, is the optimized objective function value of the energy supply strategy, is the loss function.

2. The energy supply method for a nearly zero-energy consumption building based on multi-objective optimization according to claim 1, characterized in that: The temperature sensor data is collected by temperature sensors arranged in each functional area, corridor, staircase and the building facade of the building; The humidity sensor data is collected by humidity sensors arranged in each functional area of the building and the building facade; The light sensor data is collected by light sensors arranged in the indoor and outdoor window areas and the roof area of the building; The electrical load data is collected by the building electrical system.

3. The energy supply method for a nearly zero-energy consumption building based on multi-objective optimization according to claim 1, characterized in that: The indoor comfort indicator includes the temperature and humidity deviation, and the indoor comfort indicator is calculated according to the temperature and humidity deviation; The temperature and humidity deviation is calculated by comparing the temperature sensor data with the preset temperature comfort range and comparing the humidity sensor data with the preset relative humidity comfort range to obtain the temperature and humidity deviation value; Standardize the electricity consumption load data according to the building's usable area to obtain the electricity consumption per unit area, and compare it with the baseline value specified in the building energy efficiency design standard to generate the energy consumption index; Calculate the indoor daylighting factor based on the daylight sensor data, and compare the daylighting factor with the minimum requirement value specified in the building daylighting standard to generate the daylighting index; wherein the daylighting factor is the percentage of the indoor natural daylight illuminance to the total outdoor horizontal natural illuminance at the same moment.

4. The near-zero energy consumption building energy supply method based on multi-objective optimization according to claim 3, characterized in that: The calculation of the temperature and humidity deviation value is shown in the following formula: ， Among them, is the measured temperature value, is the set temperature value, is the measured humidity value, is the set humidity value, and are the allowable maximum temperature and humidity deviations respectively, and are the weight coefficients; The calculation of the indoor comfort index is shown in the following formula: ， Among them, is the power consumption at time period i, is the duration of time period i, is the building area, is the reference energy consumption value, is the temperature correction coefficient, is the outdoor design temperature; The calculation of the daylighting index is shown in the following formula: ， Among them, is the indoor natural daylight illumination, is the total outdoor horizontal natural illumination, is the solar altitude angle, is the distance from the measurement point to the window, is the reference distance, and are correction factors.

5. The near-zero energy consumption building energy supply method based on multi-objective optimization according to claim 4, characterized in that: The update of the particle velocity is represented by the following formula: ， Among them, is the velocity of the -th particle in dimensions, is the position of the -th particle in d dimensions, is the individual optimal position, is the global optimal position, is the inertia weight, , , are the learning factors, , , are random numbers in [0, 1]; The update of the particle position is represented by the following formula: ， Among them, is the adaptive adjustment coefficient, is the objective function value of the current iteration.

6. The near-zero energy consumption building energy supply method based on multi-objective optimization according to claim 5, characterized in that: The fitness function is shown in the following formula: ， Among them, is the gradient penalty factor, is the weight balance factor, is the objective function of the multi-objective evaluation system, , , are the weight coefficients in the objective function of the multi-objective evaluation system.

7. A near-zero energy consumption building energy supply system based on multi-objective optimization, based on the near-zero energy consumption building energy supply method based on multi-objective optimization according to any one of claims 1 to 6, characterized in that: Including: A data acquisition module for collecting energy consumption data in a building and constructing a multi-objective evaluation system based on the energy consumption data; A parameter optimization module for optimizing the weights of the multi-objective evaluation system to obtain the optimal weight coefficients; A model construction module for combining the optimal weight coefficients with a solar photovoltaic system and a ground source heat pump system to establish a renewable energy supply strategy model; A control module for outputting the energy supply scheduling instructions for each time period of the building based on the renewable energy supply strategy model and adjusting the operating parameters of the solar photovoltaic system and the ground source heat pump system.

Images