Ground source heat pump system operation control method and control device based on building load
Through the operation control method of ground source heat pump system based on building load, the neural network predicts heat exchange ability and combines the non-dominant sorting optimization algorithm, the problems of long operation time and low prediction accuracy of the ground source heat pump system are solved, and the system is efficient , stable operation and cost reduction.
Patent Information
- Application Number
- CN202411318131.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-12-03
AI Technical Summary
The operation method of the ground source heat pump system has the problem that it takes a long time and the prediction accuracy needs to be improved. Traditional empirical prediction methods cannot adapt to environmental changes. The model prediction method is complex in calculations and low data timeliness, making it difficult to provide accurate predictions in complex environments.
The operation control method of ground source heat pump system based on building load is adopted. By obtaining ground source end parameters and equipment end parameters, input them into the pre-trained neural network model to predict heat exchange capacity, and the number of buried pipes is optimized using non-dominant sorting optimization algorithm. and spacing to dynamically adjust the system operation.
It realizes rapid and accurate prediction of heat exchange capacity and system optimization, reduces operating costs, improves the operating efficiency and stability of the system, and is suitable for changes in different regions and buildings.
Smart Images

Figure CN119063326A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of building energy conservation and HVAC, and relates to an operation control method and a control device of a ground source heat pump system based on building load. Background Art
[0002] The ground source heat pump system is a new type of energy system that is efficient and environmentally friendly and uses the constant temperature of the underground to heat and cool buildings.
[0003] However, despite the increasing use of geothermal heat pump systems, their operation strategies are relatively traditional. Usually, when formulating the operation strategy of geothermal heat pump systems, factors such as the energy demand of the building, underground hydrogeological conditions, and weather changes need to be considered. After comprehensively considering the above factors, the control system adjusts the circulating water temperature, compressor operating frequency and other parameters to meet the heating and cooling needs.
[0004] When the energy demand of the building system is determined and the exploration of underground hydrogeological conditions is completed, the operation of the ground source heat pump system is mainly determined by the environment in which the building is located. At present, most operators will formulate system operation strategies in advance based on weather forecasts and existing engineering experience, mainly including the flow rate of the circulating fluid in the pipe, the number of buried pipes, etc. At the same time, with the advancement of technology and the development of intelligent control systems, more and more ground source heat pump systems have begun to adopt advanced model prediction methods, that is, using building characteristics, underground hydrogeological conditions, climate data, etc. as input parameters, combined with mathematical models and computer algorithms, through simulation and prediction of system operation, optimize control parameters and strategies, so as to achieve more accurate operation strategies and reduce dependence on empirical methods.
[0005] However, both the traditional empirical method and the model prediction method developed in recent years have certain limitations:
[0006] The empirical prediction method usually makes hypothetical inferences about the system's operating strategy based on previous test data and engineering experience, and therefore cannot fully take into account changes in the current environment and the actual needs of the system; at the same time, the empirical prediction method is difficult to adapt to changes in different regions, different buildings and different seasons, and the operating strategies it formulates are not flexible enough; on the other hand, since the empirical prediction method mainly relies on historical data and experience judgment, its prediction accuracy may be limited, and it cannot accurately predict the performance of the system under different conditions.
[0007] The model prediction method needs to consider multiple factors when establishing an accurate model, including building characteristics, groundwater hydrogeological conditions, climate change, etc. Therefore, the model is highly complex and requires a large amount of data and computing resources to support it, which will also lead to problems such as long calculation time and low data timeliness. In addition, although the model prediction method can provide more accurate prediction results, there is still the possibility of prediction errors, especially when facing complex and changeable environmental conditions, the prediction accuracy of the model may be affected. Summary of the invention
[0008] The technical purpose of the present application is to provide a ground source heat pump system operation control method and control system based on building load in order to address the technical problems that the current operation method of the ground source heat pump system is time-consuming and the prediction accuracy needs to be improved.
[0009] In order to achieve the above technical objectives, this application adopts the following technical solutions.
[0010] In a first aspect, an embodiment of the present application provides a method for controlling the operation of a ground source heat pump system based on a building load, comprising:
[0011] Obtain ground source parameters, equipment parameters and building loads at target time scale;
[0012] Inputting the ground source end parameters and the equipment end parameters into the pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the control conditions corresponding to the ground source end parameters and the equipment end parameters;
[0013] According to the building load and the heat exchange capacity, the minimum number of underground pipes that meets the building load is obtained;
[0014] Based on the minimum number of buried pipes, the operating cost and the spacing between buried pipe heat exchangers are optimized using a non-dominated sorting optimization algorithm, and the number of buried pipe heat exchangers and the spacing between buried pipe heat exchangers are ultimately determined.
[0015] In a second aspect, the embodiment of the present application provides a ground source heat pump system operation control device based on building load, including:
[0016] The first parameter input module is used to obtain the ground source end parameters;
[0017] The second parameter input module is used to obtain device-side parameters;
[0018] The third parameter input module obtains the building load at the target time scale;
[0019] A heat exchange capacity determination module is used to input the ground source end parameters and the device end parameters into a pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the control conditions corresponding to the ground source end parameters and the device end parameters;
[0020] A minimum number of underground pipes determining module, used to obtain the minimum number of underground pipes that meets the building load according to the building load and the heat exchange capacity;
[0021] A parameter optimization module is used to optimize the operation cost and the spacing between ground heat exchangers based on the minimum number of ground heat exchangers by using a non-dominated sorting optimization algorithm, and finally determine the number of ground heat exchangers and the spacing between ground heat exchangers;
[0022] The result output module is used to output the number of ground heat exchangers and the spacing between the ground heat exchangers.
[0023] Compared with the prior art, the ground source heat pump system operation control method based on building load provided in the embodiment of the present application has the following advantages: through the pre-trained neural network model, the heat exchange capacity of a single underground pipe heat exchanger under the pre-selected ground source end parameters can be accurately predicted with high accuracy and reliability; the method can obtain the building load in real time, and dynamically adjust the number and spacing of the underground pipe heat exchangers according to the load changes to ensure that the system always operates in the optimal state. By calculating the minimum number of underground pipes required to meet the building load, over-allocation and waste of resources are avoided; by using the non-dominated sorting optimization algorithm, not only the operating cost is considered, but also the optimization of the spacing of the underground pipe heat exchangers is taken into account, so as to achieve the reduction of operating costs while ensuring system performance. The entire control process is automated, which reduces the possibility of human intervention and wrong decisions, and improves the operating efficiency and stability of the system.
[0024] Compared with the prior art, the ground source heat pump system operation control device based on building load provided in the embodiment of the present application has the same beneficial technical effects as the ground source heat pump system operation control method based on building load provided in the embodiment of the present application compared with the above technology, which will not be elaborated. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The drawings described herein are for explanation purposes only and are not intended to limit the scope of the present application in any way. In addition, the shapes and proportional dimensions of the components in the drawings are only for illustration purposes and are used to help understand the present application. They do not specifically limit the shapes and proportional dimensions of the components of the present application. Under the guidance of the present application, those skilled in the art can select various possible shapes and proportional dimensions to implement the present application according to specific circumstances. In the drawings:
[0026] Figure 1 A schematic flow chart of a method for controlling the operation of a ground source heat pump system based on building loads provided in an embodiment of the present application;
[0027] Figure 2 A schematic diagram of the structure of a ground source heat pump system operation control device based on building load provided in an embodiment of the present application;
[0028] Figure 3 An overall schematic diagram of the input and output windows of the ground source heat pump system operation control device based on building load provided in an embodiment of the present application;
[0029] Figure 4 The MATLAB source code file included in the embodiments of the present application;
[0030] Figure 5 The specific distribution of the optimized buried pipes is shown in Sheet 1 in the file coordinates.xls in the embodiment of this application. DETAILED DESCRIPTION
[0031] In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.
[0032] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features.
[0033] See also Figure 1 The embodiment of the present application provides an operation control method for a ground source heat pump system based on a building load, including: obtaining ground source end parameters, equipment end parameters and a building load at a target time scale; inputting the ground source end parameters and the equipment end parameters into a pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the control conditions corresponding to the ground source end parameters and the equipment end parameters; according to the building load and the heat exchange capacity at the target time scale, obtaining the minimum number of buried pipes that meet the building load; based on the minimum number of buried pipes, using a non-dominated sorting optimization algorithm, optimizing the operating cost and the spacing between buried pipe heat exchangers, and finally determining the number of buried pipe heat exchangers and the spacing between buried pipe heat exchangers.
[0034] Before starting to input the ground source parameters, it is necessary to ensure that all necessary measurement tools and data are ready, including but not limited to geological exploration reports, ground temperature monitoring data, soil thermal physical parameters, etc.
[0035] In some embodiments, the ground source parameters include ground temperature, comprehensive thermal conductivity, buried pipe depth, buried pipe spacing, number of buried pipe rows, number of buried pipe columns, fluid inlet temperature and fluid flow rate.
[0036] The average ground temperature in the current season or operating period can be accurately input based on historical monitoring data or ground temperature information provided by professional institutions; the comprehensive thermal conductivity of the soil in the area where the ground source heat pump system is located can be input based on geological exploration reports or experimental test data; the burial depth of the buried pipes, the spacing between the buried pipes, and the number of rows and columns of the buried pipes can be input one by one according to the design plan of the heat pump system; the temperature and flow rate of the heat exchange fluid (water or antifreeze) when it enters the buried pipe heat exchanger can be input based on the operating requirements of the heat pump unit or actual test data.
[0037] In some embodiments, the plurality of ground source parameters are a set of random values within a range of the parameters, and the energy efficiency ratio of the heat pump unit is a random value within a range of the parameters.
[0038] In some embodiments, the equipment-side parameter includes the energy efficiency ratio of the heat pump unit. The energy efficiency ratio of the heat pump unit can be input according to the specific model data or actual test results provided by the manufacturer of the heat pump unit.
[0039] According to the specific needs of building operation, combined with the actual situation, the building load value is entered in the system in sequence according to the target time scale. When there are the same loads, they also need to be entered separately. As an example, the target time scale can be the hourly load of the building end hour by hour, such as referring to the hourly load value of a single building on a certain day in the historical operation cycle.
[0040] After completing the input of all the above parameters, the data will be processed and analyzed according to the built-in neural network and optimization algorithm. Subsequently, the number of ground heat exchangers involved in heat exchange and their specific distribution locations will be given. These results will help operators adjust the operation strategy of the heat pump system according to actual conditions to achieve the purpose of energy saving and consumption reduction.
[0041] In some embodiments, the neural network model includes an input layer with 9 nodes, a hidden layer with 10 nodes (or 15 nodes) and an output layer with 1 node, and the training process is completed by calling the ground source heat pump system operation database.
[0042] The neural network model is used to obtain the heat exchange capacity of a single buried pipe heat exchanger under the corresponding control conditions of the ground source end parameters and the equipment end parameters through the mapping relationship according to the ground source end parameters and the equipment end parameters. The mapping relationship can be expressed as:
[0043] C GHE =F(T soil , k soil , l GHE ,s GHE , row GHE ,col GHE , T in , v f, COP) (1);
[0044] Among them, C GHE is the heat transfer capacity, T soil is ground temperature, k soil is the comprehensive thermal conductivity, l GHE is the buried depth of underground pipe, s GHE is the buried pipe spacing, row GHE is the number of buried pipe rows, col GHE is the number of buried pipes, T in is the fluid inlet temperature, v f is the fluid flow rate, and COP is the energy efficiency ratio of the heat pump unit.
[0045] The algorithm used in some embodiments for training the neural network model is the Levenberg-Marquardt (LM) algorithm.
[0046] The LM algorithm is an optimization algorithm for nonlinear least squares problems. Its implementation steps can be briefly summarized as: initializing parameters, calculating the objective function and Jacobian matrix, calculating the parameter update amount, updating the parameter vector and judging the termination condition.
[0047] During the parameter initialization process, the initial parameter vector x is given 0 , set the number of iterations k, and set the adjustment factor λ and the threshold ε for controlling convergence.
[0048] In the process of calculating the objective function, according to the current parameter vector x k , calculate the value of the objective function F(x k ) and the Jacobian matrix J k The Jacobian matrix contains the partial derivatives of the objective function F(x) with respect to each parameter. In the process of calculating the parameter update amount, the Gauss-Newton algorithm and the gradient descent algorithm theory are combined to calculate the update amount Δx of the parameter vector. k When calculating the parameter update amount, the update value is equal to the sum of the current parameter vector and the calculated update amount. When determining the termination condition, when ||Δx k When ||<ε or the number of iterations reaches a given value, the iteration stops. The important calculation formulas in the process are shown in formulas (2) and (3):
[0049] Δx k =(J T J+λI) -1 J T (F(x k )) (2);
[0050] x k+1 =x k +Δx k (3);
[0051] In the formula, J and J T are the Jacobian matrix and its transposed matrix respectively, and I is the identity matrix.
[0052] In some embodiments, the advantages of the Newton iteration method and the gradient descent algorithm can be combined in the LM algorithm to achieve a better balance between convergence speed and calculation accuracy.
[0053] In some embodiments, the ratio of the building load to the heat exchange capacity on a time scale is determined as the minimum number of underground pipes, expressed as:
[0054]
[0055] Where n is the minimum number of underground pipes required to meet the building load at the target time scale (e.g. hourly), Q load is the building load at the target time scale (which can be hourly building load); C GHE It is the heat exchange capacity of a single buried pipe heat exchanger under the corresponding control conditions of the ground source end parameters and the equipment end parameters. The number of buried pipe heat exchangers obtained by subsequent optimization shall not be less than this value.
[0056] In specific implementation, the minimum number of buried pipes that meet the building load can be used as a parameter for subsequent optimization of operating costs and geothermal balance. The optimization goal is the economy of the ground source heat pump system and the heat exchange efficiency under long-term operation: when a small number of buried pipes are enabled, the system operation and maintenance costs are reduced and the economy is improved; when a small number of buried pipes are enabled, the heat exchange capacity of a single buried pipe increases, and the nearby ground temperature changes more significantly, which is not conducive to maintaining the efficiency of the entire system during long-term operation. There is a contradictory relationship between the two optimization goals, so they can be handled using the optimization algorithm.
[0057] In the embodiment, the decision variables required by the optimization algorithm include the minimum number of underground pipes to meet the building load obtained by the neural network, the cost and power of the ground source heat pump system, the outlet temperature of the underground heat exchanger, the power consumption of the heat pump unit, and the energy loss. The optimization algorithm (such as the NSGA-II algorithm) can comprehensively consider the optimization problem of multiple objectives. The optimization target parameters in the embodiment are:
[0058] 1. Operating expenses:
[0059] The operating cost is used to evaluate the economic efficiency of the ground source heat pump system and is determined by the number of underground pipes involved in heat exchange. The more underground pipes involved in heat exchange, the higher the operating cost. Among them, the operating cost is mainly composed of three parts, including the operation and maintenance cost of the underground pipes, the operation and maintenance cost of the heat pump heat exchange, and the operation cost of the water pump, which can be expressed by the mathematical formula as follows:
[0060] OC=OC GHE +OCheat_pump +OC water_pump (5);
[0061] In the formula, OC GHE , OC heat_pump , OC water_pump They are the costs of underground pipes, heat pump units and water pump operation and maintenance, which can be calculated as follows:
[0062] OC GHE =N GHE ×SC GHE
[0063] OC heat_pump =Cap heat_pump ×SC heat_pump +M re ×SC re ×(1+(T op -1)×ALR)+∑(1+2%)×(HEC+CEC)×SC ele (6);
[0064] OC water_pump =(m heat_pump +m water_pump )×SC water_pump +∑(1+2%)×(EC heat_pump +EC water_pump )×SC ele (7);
[0065] Where N GHE is the number of ground heat exchangers involved in operation, SC GHE The cost of operating a single ground heat exchanger, Cap heat_pump is the rated heat exchange power of the heat pump unit, SC heat_pump M is the cost of the heat pump unit at rated power. re is the mass of coolant in the heat pump unit, SC re is the coolant cost per unit mass, T op is the service life, ALR is the loss rate of the coolant during the service life of the heat pump unit, HEC and CEC are the heating capacity and cooling capacity of the heat pump unit respectively. The calculation method is shown in formula (8) and formula (9), SC ele is the electricity price, m heat_pump and m water_pump are the fluid mass flow rates in the heat pump and water pump, SC water_pump EC is the cost of processing unit mass of fluid. heat_pump and EC water_pumpare the electricity consumption of the heat pump and the water pump, respectively. The calculation methods are shown in formulas (10) and (11); P% (optional 2%) is the energy loss rate during operation, which can represent the energy loss rate of most ground source heat pump systems. For some special systems, this value can also be flexibly adjusted according to actual needs.
[0066] HEC=1.63×T out +2.37×50+15.33 (8)
[0067] CEC=2.48×T out +1.45×7+35.88 (9);
[0068] EC heat_pump =m heat_pump ×H heat_pump ×g / 1000(10);
[0069] EC water_pump =m water_pump ×H water_pump ×g / 1000(11);
[0070] Where, T out is the outlet temperature of the ground heat exchanger, H heat_pump and H water_pump are the water head heights of the heat pump unit and the water pump respectively, and g is the acceleration of gravity.
[0071] 2. Spacing of buried pipes.
[0072] The buried pipe spacing is used to evaluate the long-term operating efficiency of the ground source heat pump system and is determined by the distribution of the buried pipes. The larger the buried pipe spacing, the lower the local ground temperature rise, and the smaller the impact on the long-term operation of the ground source heat pump system. The calculation method is shown in formula (12):
[0073]
[0074] Where D is the sum of the distances between the underground pipes, x i and i are the horizontal and vertical coordinates of a single buried pipe, respectively, and n is the minimum number of buried pipes required to meet the building load of the target time scale.
[0075] In the specific implementation, the optimization algorithm used can be the non-dominated sorting algorithm NSGA-II (Non-dominated Sorting Genetic Algorithm II), which is a multi-objective genetic algorithm with an elite strategy. The non-dominated sorting algorithm can search the parameter space and generate a series of ground source heat pump system operating parameter combinations with good fitness, and then select high-quality individuals based on the results of the non-dominated sorting algorithm. The population is updated using crossover and mutation operations to further optimize the parameter combination.
[0076] In NSGA-II, the density of individuals in the population is measured by calculating and comparing the crowding degree, and it is used as a comparison standard between individuals in the population. This allows individuals to be evenly expanded to the entire Pareto domain, ensuring the diversity of the population. At the same time, each individual also has a dominant sorting attribute, so that the best individual that meets the requirements can be selected.
[0077] Crowding is an important concept used by NSGA-II to maintain population diversity. It measures the density of individuals around a given point in the population, thereby helping the algorithm avoid falling into local optimality during the search process and ensuring that the solution is evenly distributed in the target space. The calculation method can be regarded as the distance between individuals and can be expressed by the following mathematical formula.
[0078]
[0079] Where, d b and d w They represent the distance between a single individual function value and the best and worst function values respectively, and Objk represents the function value of the kth objective function.
[0080] The dominance relationship depends on the size relationship between the corresponding function values of the individuals. Based on the concept of Pareto optimality, it can be defined as: if individual x 1 All objective function values F k (x 1 ) are no better than individual x 2 difference, and at least one objective function value F i (x 1 ) than individual x 2 Better, then we call individual x 1 Dominant individual x 2 . Or individuals are affected by individuals x 1 Dominate. It can be expressed as:
[0081]
[0082] Where k, i represents the number of the objective function.
[0083] Crossover is one of the main methods for generating new individuals in genetic algorithms. It usually involves selecting two parent individuals and generating two new offspring individuals by exchanging some of their genes. The crossover operation can combine the excellent characteristics of different individuals, thus potentially producing a solution with better performance. This patent uses the real number coding crossover method, which means that the genes on the chromosome are considered to be real numbers. The process is expressed mathematically as follows:
[0084] C 1 =αP 1 +γ1-α)P 2 (17);
[0085] C 2 =αP 2 +(1-α)P 1 (18);
[0086] In the formula, C 1 , C 2 They represent the offspring individuals produced by crossover, P 1 , P 2 They represent the parent individuals used for the crossover operation, and α is a constant between 0 and 1, which is used to control the contribution of the two parent individuals in generating offspring individuals.
[0087] Mutation is also one of the main methods for generating offspring individuals in genetic algorithms. It is manifested by adding or subtracting a small random number based on the existing gene value. Since the probability of mutation in nature is low, the mutation probability P is specified when implementing the mutation operation process. mu The process can be expressed mathematically as follows:
[0088]
[0089] N mu =N pop *P mu (20);
[0090] In the formula, is the gene value in the variant individual The value after mutation, the change is Δ mu , N mu and N pop are the number of mutant individuals and the population size respectively.
[0091] The specific distribution position of the buried pipe heat exchanger, in order to balance the heat exchange distribution, the buried pipe heat exchangers in different positions can be opened in different operating periods to make each pipeline run intermittently, thereby reducing the heat exchange of a single pipeline, making the local ground temperature changes more uniform and the variation range lower during long-term operation.
[0092] The ground source heat pump system operation control method based on building load provided in the embodiment can quickly obtain the heat exchange capacity of the buried pipe heat exchanger by calling the ground source heat pump system operation database to train the neural network model, which greatly saves time cost compared with the traditional simulation method and has considerable reliability; comprehensively considers the economy of system operation and the heat exchange efficiency during long-term use, and optimizes the operation strategy of the ground source heat pump system with this as the objective function, thereby saving system operating costs while evenly distributing the positions of the buried pipe heat exchangers involved in heat exchange; at the same time, since the non-dominated sorting algorithm itself does not rely on a specific model or assumption, and has no restrictions on the search range of the parameter space, it is suitable for various types of ground source heat pump systems, has strong versatility and adaptability, and is conducive to popularization and use.
[0093] Based on the same inventive concept as the above ground source heat pump system operation control method based on building load, the embodiment of the present application also provides a ground source heat pump system operation control device based on building load, such as Figure 2 As shown, it includes a first parameter input module, a second parameter input module, a third parameter input module, a heat exchange capacity determination module, a minimum number of buried pipes determination module, a parameter optimization module and a result output module.
[0094] Among them, the first parameter input module is used to obtain the ground source parameters; the second parameter input module is used to obtain the equipment end parameters; the third parameter input module obtains the building load at the target time scale; the heat exchange capacity determination module is used to input the ground source parameters and the equipment end parameters into the pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the corresponding control conditions of the ground source parameters and the equipment end parameters; the minimum number of buried pipes determination module is used to obtain the minimum number of buried pipes that meet the building load based on the building load and the heat exchange capacity; the parameter optimization module is used to optimize the operating cost and the spacing between buried pipe heat exchangers based on the minimum number of buried pipes using a non-dominated sorting optimization algorithm, and finally determine the number of buried pipe heat exchangers and the spacing between buried pipe heat exchangers; the result output module is used to output the number of buried pipe heat exchangers and the spacing between buried pipe heat exchangers.
[0095] Figure 3 The overall schematic diagram of the input and output window of the ground source heat pump system operation control device based on building load provided in the embodiment of the present application is as follows: Figure 3 In the illustrated embodiment, the first parameter input module, the second parameter input module and the third parameter input module correspond to area A, area B and area C in the external receiving parameter input window respectively.
[0096] Enter the following parameter values according to the prompts through the ground source parameter input window (area A): ground temperature, comprehensive thermal conductivity, buried pipe depth, buried pipe spacing, number of buried pipe rows, number of buried pipe columns, fluid inlet temperature and fluid flow rate.
[0097] Enter the following parameter values according to the prompts through the device-side parameter input window (area B): energy efficiency ratio of the heat pump unit.
[0098] Enter the following parameter values according to the prompts through the building parameter input window (area C): The building's hourly cooling / heating load, even if the load is the same in multiple time periods, also needs to be entered separately.
[0099] like Figure 3 As shown in area D, the results of the output optimization result window are generated and displayed as follows: After completing the input of all the above parameters, the system will process and analyze the data according to the built-in neural network and optimization algorithm. In the non-dominated sorting optimization algorithm, calculation parameters including population size, number of iterations, crossover probability, and mutation probability are set. The algorithm parameters can be changed according to the specific requirements of the user. In addition to providing optimization strategies for daily use, they are also suitable for research activities with certain accuracy requirements. Subsequently, the system will give the number of buried pipe heat exchangers involved in heat exchange and their specific distribution locations. These results will help operators adjust the operation strategy of the heat pump system according to actual conditions to achieve the purpose of energy saving and consumption reduction.
[0100] Considering the different operating conditions and operating conditions of ground source heat pump systems in different regions: the operating cost function in the optimization objective can be adjusted according to the specific costs of different ground source heat pump systems, which is universal. The ground pipe heat exchanger spacing parameter in the optimization objective is derived from the user's input value in the ground source end parameter input window, so the specific operating location distribution method of ground source heat pump systems of different sizes can be provided without repeated changes.
[0101] Figure 4 is the MATLAB source code file included in the embodiments of the present application; Figure 4 As shown, the file BPnetwork.m builds and trains the artificial neural network;
[0102] The file BPpredict.m is the main program, which includes calling the neural network, printing the input and output panes, receiving parameter input, implementing the NSGA-II algorithm, and generating an Excel table of the specific location of the ground pipe heat exchanger;
[0103] The file CalcCrowdingDistance.m is the crowding calculation function for implementing the NSGA-II algorithm;
[0104] The file Crossover.m is the crossover function for implementing the NSGA-II algorithm;
[0105] The file data.mat stores the database data used for neural network training and verification;
[0106] The file database.xlsx is an Excel table that stores database data;
[0107] The file datapro.m is used for importing and processing data;
[0108] The file Dominates.m is the dominance relationship judgment function for implementing the NSGA-II algorithm;
[0109] The file GHEdis.m is the objective function considering the distribution of ground heat exchangers in the optimization algorithm;
[0110] The file inputps.mat is the file for normalization processing in the neural network;
[0111] The file MOP.m is the test function of the optimization process;
[0112] The file MOP4.m is the objective function of the optimization process;
[0113] The file Mutate.m is the mutation function for implementing the NSGA-II algorithm;
[0114] The file net.mat is used to store the trained neural network file;
[0115] The file NonDominatedSorting.m is the non-dominated sorting function for implementing the NSGA-II algorithm;
[0116] The file outputps.mat is the file for denormalization in the neural network;
[0117] The file SortPopulation.m is the selection function for implementing the NSGA-II algorithm;
[0118] The file coordinates.xls is an Excel table that stores the specific distribution of buried pipes.
[0119] Figure 5 The specific distribution of the optimized buried pipes is shown in Sheet 1 in the file coordinates.xls in the embodiment of this application. Figure 5 In the figure, column A is the X coordinate of the buried pipe, and column B is the Y coordinate of the buried pipe.
[0120] This application implements the heat exchange capacity of the vertical buried heat exchanger quickly through the neural network according to the input of multiple input ground source end parameters, equipment end parameters and building loads at the target time scale, and then calculates the minimum number of buried pipes required. Then enter the optimization algorithm part to complete the planning of the operation time and distribution position of the buried pipes, and obtain the number of buried pipe heat exchangers involved in the heat exchange and their specific distribution positions. This method can quickly obtain the system's operating strategy while taking into account economic efficiency, and at the same time reasonably allocates the spatial position of the heat exchanger, which is suitable for long-term use of ground source heat pump systems.
[0121] It should be noted that in the present application, the emphasis of the schemes described in various embodiments is different, but the various embodiments are also somewhat interrelated. When understanding the scheme of the present application, the various embodiments can refer to each other.
[0122] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily make various changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A ground source heat pump system operation control method based on building load, characterized in that: include: Obtain ground source parameters, equipment parameters and building loads at target time scale; Inputting the ground source end parameters and the device end parameters into a pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the control conditions corresponding to the ground source end parameters and the device end parameters; According to the building load and the heat exchange capacity, the minimum number of underground pipes that meets the building load is obtained; Based on the minimum number of buried pipes, the operating cost and the spacing between buried pipe heat exchangers are optimized using a non-dominated sorting optimization algorithm, and the number of buried pipe heat exchangers and the spacing between buried pipe heat exchangers are ultimately determined.
2. The ground source heat pump system operation control method according to claim 1, characterized in that: The ground source end parameters include ground temperature, comprehensive thermal conductivity, buried pipe depth, buried pipe spacing, number of buried pipe rows, number of buried pipe columns, fluid inlet temperature and fluid flow rate.
3. The ground source heat pump system operation control method according to claim 1, characterized in that: The equipment-side parameters include the energy efficiency ratio of the heat pump unit.
4. The ground source heat pump system operation control method based on building load according to claim 1, characterized in that: The ground source heat pump system operation control method also includes: using the Levenberg-Marquardt algorithm to train the neural network model.
5. The ground source heat pump system operation control method based on building load according to claim 1, characterized in that: The functional expression of the operating cost is: OC=OC GHE +OC heat_pump +OC water_pump ; Among them, OC is the operating cost, OC GHE The cost of underground pipe operation and maintenance, OC heat_pump The cost of heat pump unit operation and maintenance, OC water_pump The cost of operating and maintaining the water pump.
6. The method for controlling the operation of a ground source heat pump system based on building load according to claim 5, characterized in that: The calculation expression of the cost of operation and maintenance of the buried pipe is as follows: OC GHE =N GHE ×SC GHE ; Among them, N GHE is the number of ground heat exchangers involved in operation, SC GHE Cost of operating a single BHE.
7. The method for controlling the operation of a ground source heat pump system based on building load according to claim 5, characterized in that: The calculation expression of the operation and maintenance cost of the heat pump unit is as follows: OC heat_pump =Cap heat_pump ×SC heat_pump +M re ×SC re ×(1+(T op -1)×ALR)+∑(1+P%)×(HEC+CEC)×SC ele ; Among them, Cap heat_pump is the rated heat exchange power of the heat pump unit, SC heat_pump M is the cost of the heat pump unit at rated power. re is the mass of coolant in the heat pump unit, SC re is the coolant cost per unit mass, T op is the service life of the heat pump unit, ALR is the loss rate of the coolant during the service life of the heat pump unit, HEC and CEC are the heating and cooling capacities of the heat pump unit respectively, SC ele is the electricity price, and P% is the energy loss rate during operation.
8. The method for controlling the operation of a ground source heat pump system based on building load according to claim 5, characterized in that: The calculation expression of the cost of water pump operation and maintenance is as follows: OC water_pump =(m heat_pump +m water_pump )×SC water_pump +∑(1+P%)×(EC heat_pump +EC water_pump )×SC ele ; Among them, m heat_pump is the fluid mass flow rate in the heat pump, m water_pump is the fluid mass flow rate in the pump, SC water_pump EC is the cost of processing unit mass of fluid. heat_pump is the electrical energy consumption of the heat pump, EC water_pump is the power consumption of the water pump, P% is the energy loss rate during operation, SC ele For electricity price.
9. The method for controlling the operation of a ground source heat pump system based on building load according to claim 1, characterized in that: The ground source heat pump system operation control method further includes: determining the ratio of the building load to the heat exchange capacity at the target time scale as the minimum number of the buried pipes.
10. A ground source heat pump system operation control device based on building load, characterized in that: include: The first parameter input module is used to obtain the ground source end parameters; The second parameter input module is used to obtain device-side parameters; The third parameter input module obtains the building load at the target time scale; A heat exchange capacity determination module is used to input the ground source end parameters and the device end parameters into a pre-trained neural network model to obtain the heat exchange capacity of a single buried pipe heat exchanger under the control conditions corresponding to the ground source end parameters and the device end parameters; A minimum number of underground pipes determining module, used to obtain the minimum number of underground pipes that meets the building load according to the building load and the heat exchange capacity; A parameter optimization module is used to optimize the operation cost and the spacing between ground heat exchangers based on the minimum number of ground heat exchangers by using a non-dominated sorting optimization algorithm, and finally determine the number of ground heat exchangers and the spacing between ground heat exchangers; The result output module is used to output the number of ground heat exchangers and the spacing between the ground heat exchangers.
Citation Information
Cited By
Intelligent management and control method and system for heat pump energy system
CN119802789A
