Multi-time scale prediction based multi-level geothermal well coordinated scheduling method and system

Abstract

The present disclosure provides a multi-level geothermal well coordinated scheduling method and system based on multi-time scale prediction, relating to the technical field of data processing. The method comprises: based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, constructing a well group output model; based on historical load data and historical meteorological data, constructing a multi-scale load prediction model; using the multi-scale load prediction model to determine the load prediction result of the target period, the load prediction result including day-ahead load, day-time load and real-time load; using the well group output model, the load prediction result and the multi-objective optimization function to jointly calculate, obtaining the optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in the current scheduling period. The technical scheme in the present disclosure can dynamically coordinate the output ratio of shallow and deep geothermal wells according to the predicted load demand of different time scales, and realize efficient utilization of multi-level geothermal resources.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, in particular, to a multi-level geothermal well cooperative scheduling method and system based on multi-time scale prediction. BACKGROUND

[0002] As a stable, renewable and low-carbon energy, geothermal energy has been widely used in building heating, district energy stations and integrated energy systems. According to the development depth, geothermal wells can be divided into shallow and deep two categories, and there are significant differences in temperature level, heat storage capacity and adjustability between different levels of resources. In actual engineering, shallow geothermal wells have the advantages of fast response speed and suitability for daily fluctuating load, but the heat storage capacity is limited and temperature attenuation is easy to occur; deep geothermal wells have higher heat grade and stable output capacity, and are often used to bear the basic load or energy guarantee under extreme weather.

[0003] However, in the related geothermal well scheduling technology, shallow geothermal wells or deep geothermal wells are usually controlled separately, and the dynamic cooperation and complementary adjustment of multiple heat sources cannot be realized, which leads to resource redundancy, overloading of part of the resources, and reduces the operation efficiency and energy utilization rate of the overall energy system. Especially in the regional energy supply scene with large load fluctuations, one type of well group may be in a high-load operation state for a long time, leading to heat field attenuation and operation fatigue, while the other type of well group is in a redundant state and is not effectively scheduled, causing unbalanced resource utilization. Therefore, there is still room for improvement in the resource utilization efficiency and energy supply stability of the existing geothermal well scheduling technology.

[0004] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the embodiments of the present disclosure is to provide a multi-level geothermal well cooperative scheduling method based on multi-time scale prediction, a multi-level geothermal well cooperative scheduling system based on multi-time scale prediction, an electronic device and a computer readable storage medium, by constructing a well group output model and a multi-scale load prediction model, and combining a multi-objective optimization function to jointly calculate the scheduling parameters of the shallow geothermal well group and the deep geothermal well group, thereby at least to a certain extent improving the resource utilization efficiency and energy supply stability in the geothermal well scheduling process.

[0006] Other characteristics and advantages of the present disclosure will become apparent from the following detailed description, or will be learned by practice of the present disclosure.

[0007] According to a first aspect of the embodiments of the present disclosure, a multi-level geothermal well cooperative scheduling method based on multi-time scale prediction is provided, which comprises: constructing a well group output model based on operation parameters of a shallow geothermal well group and a deep geothermal well group, the well group output model being capable of representing unit heat output capacity, heat storage capacity and temperature attenuation characteristics of each well group; constructing a multi-scale load prediction model based on historical load data and historical meteorological data; determining a load prediction result of a target period by using the multi-scale load prediction model, the load prediction result comprising day-ahead load, daytime load and real-time load; and performing joint calculation by using the well group output model, the load prediction result and a multi-objective optimization function to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period.

[0008] In some example embodiments of the present disclosure, based on the foregoing scheme, the constructing a well group output model based on operation parameters of a shallow geothermal well group and a deep geothermal well group comprises: obtaining operation parameters of the shallow geothermal well group and the deep geothermal well group, the operation parameters comprising temperature field data, well output parameters, soil thermal conductivity, auxiliary equipment performance parameters and shallow geothermal well recharge parameters; constructing a temperature response function for representing temperature variation trend of each geothermal well in the operation process based on the temperature field data; constructing an output function for representing heat production capacity of each geothermal well in a unit time based on the well output parameters; constructing a ground temperature attenuation function for representing time attenuation of formation temperature of each geothermal well based on the soil thermal conductivity; constructing an auxiliary energy supply function for representing output characteristics of auxiliary equipment in a unit time based on the auxiliary equipment performance parameters; constructing a shallow heat storage function for representing heat storage capacity of the shallow geothermal well in a non-heating period based on the shallow geothermal well recharge parameters; and constructing the well group output model based on the temperature response function, the output function, the ground temperature attenuation function, the auxiliary energy supply function and the shallow heat storage function.

[0009] In some example embodiments of the present disclosure, based on the foregoing scheme, the constructing a multi-scale load prediction model based on historical load data and historical meteorological data comprises: performing feature extraction on the historical load data and the historical meteorological data to obtain structural index features and time series variation features; constructing a day-ahead prediction sub-model for outputting a first time granularity load based on the structural index features; constructing a daytime prediction sub-model for outputting a second time granularity load and a real-time prediction sub-model for outputting a third time granularity load based on the time series variation features; and wherein the time interval of the first time granularity is longer than the time interval of the second time granularity, and the time interval of the second time granularity is longer than the time interval of the third time granularity.

[0010] In some example embodiments of the present disclosure, based on the foregoing scheme, the feature extraction on the historical load data and the historical meteorological data obtains structural index features and time series change features, including: based on the historical load data and the historical meteorological data, extracting multi-type structure data, wherein the multi-type structure data includes any one or more of date type, month, energy supply area, historical peak load and average load index; performing discretization processing on the multi-type structure data to obtain the structural index features; and using the time series coupling relationship of the historical load data and the historical meteorological data to construct the time series change features.

[0011] In some example embodiments of the present disclosure, based on the foregoing scheme, the construction of the time series change features using the time series coupling relationship of the historical load data and the historical meteorological data includes: constructing a load change rate sequence based on the historical load data, constructing a temperature change sequence and a humidity change sequence based on the historical meteorological data; performing cross operation on the load change rate sequence and the temperature change sequence to obtain a load-temperature cross sequence; and combining the load change rate sequence and the humidity change sequence to obtain a load-humidity joint sequence; and constructing the time series change features based on the load change rate sequence, the temperature change sequence, the humidity change sequence, the load-temperature cross sequence and the load-humidity joint sequence.

[0012] In some example embodiments of the present disclosure, based on the foregoing scheme, the joint calculation using the well group output model, the load prediction result and the multi-objective optimization function to obtain the optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in the current scheduling period includes: constructing the multi-objective optimization function based on stratum heat balance, operation cost, carbon emission and power generation efficiency; obtaining a plurality of sets of candidate scheduling parameters in the current scheduling period based on the well group output model and the load prediction result; and using a target genetic algorithm to obtain the optimal scheduling parameters satisfying the multi-objective optimization function from the plurality of sets of candidate scheduling parameters.

[0013] In some example embodiments of the present disclosure, based on the foregoing scheme, the construction of the multi-objective optimization function based on stratum heat balance, operation cost, carbon emission and power generation efficiency includes: constructing a temperature change rate function reflecting the influence of heat extraction process on stratum temperature according to the geologic layer thermal conductivity of the geothermal well group, the temperature difference between the recharge fluid and the water outflow fluid, and the temperature gradient between the wellhead and the well bottom; determining a heat extraction upper limit value that makes the temperature change rate equal to a preset stratum temperature change rate threshold value according to the temperature change rate function; and constructing the multi-objective optimization function based on the heat extraction upper limit value, an operation cost function, a carbon emission function and a power generation efficiency function.

[0014] In some example embodiments of the present disclosure, based on the foregoing scheme, the target genetic algorithm is a non-dominated sorting multi-objective genetic algorithm, and the obtaining, by the target genetic algorithm, of the optimal scheduling parameters satisfying the multi-objective optimization function from the multiple sets of candidate scheduling parameters includes: constructing an initial scheduling population based on the sets of candidate scheduling parameters, each scheduling parameter individual in the initial scheduling population containing output configuration variables of the shallow geothermal well group and the deep geothermal well group; performing multi-objective fitness evaluation on the scheduling parameter individuals in the initial scheduling population by using the multi-objective optimization function, and dividing non-dominated levels according to the non-dominated sorting principle based on the evaluation results, while calculating corresponding congestion indicators in each level; selecting part of the scheduling parameter individuals as parent individuals based on the non-dominated levels and the congestion indicators, performing crossover operation and mutation operation, and generating a next generation of scheduling parameter population; and repeating the multi-objective fitness evaluation and non-dominated sorting operation on the next generation of scheduling parameter population, and selecting the optimal scheduling parameters from all current scheduling parameter individuals when a preset iteration termination condition is met.

[0015] In some example embodiments of the present disclosure, based on the foregoing scheme, the multi-level geothermal well cooperative scheduling method based on multi-time scale prediction further includes: collecting operation parameters of the shallow geothermal well group, the deep geothermal well group, the short-time heat storage device, and the air source heat pump, and determining the operation states of each heat source unit based on the operation parameters; in response to the operation state of any heat source unit being in failure, calling a backup output configuration strategy corresponding to the failure type, and adjusting the output proportions of each heat source unit in the scheduling parameters in combination with the current load prediction result and the backup output capacity; in a non-heating period, using the deep geothermal well group and / or building heat to replenish heat for the shallow geothermal well group; obtaining temperature distribution data of the shallow geothermal well, and updating the shallow geothermal well recharge parameters in the well group output model according to the temperature distribution data.

[0016] According to a second aspect of the embodiments of the present disclosure, a multi-level geothermal well cooperative scheduling system based on multi-time scale prediction is provided, which includes: an output model construction module configured to construct a well group output model based on operation parameters of a shallow geothermal well group and a deep geothermal well group, the well group output model being capable of characterizing unit heat output capacity, heat storage capacity, and temperature attenuation characteristics of each well group; a prediction model construction module configured to construct a multi-scale load prediction model based on historical load data and historical weather data; a load prediction module configured to determine a load prediction result of a target period by using the multi-scale load prediction model, the load prediction result including day-ahead load, day-time load, and real-time load; and an intelligent scheduling module configured to perform joint calculation by using the well group output model, the load prediction result, and a multi-objective optimization function to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period.

[0017] According to a third aspect of the embodiments of the present disclosure, an electronic device is provided, comprising: a processor; and a memory having computer readable instructions stored thereon, the computer readable instructions, when executed by the processor, implement the multi-level geothermal well coordinated scheduling method based on multi-time scale prediction as in the first aspect.

[0018] According to a fourth aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, having a computer program stored thereon, the computer program, when executed by a processor, implements the multi-level geothermal well coordinated scheduling method based on multi-time scale prediction as in the first aspect.

[0019] The technical solutions provided by the embodiments of the present disclosure can include the following beneficial effects:

[0020] The multi-level geothermal well coordinated scheduling method based on multi-time scale prediction in the embodiments of the present disclosure, by modeling the operation parameters of the shallow geothermal well group and the deep geothermal well group, constructing a well group output model capable of representing unit heat output capacity, heat storage capacity and temperature decay characteristics, introducing the hierarchical difference characteristics of geothermal resources in the scheduling parameter formation process, compared with the scheduling mode based on a single well type, can more accurately represent the heat output characteristics and operation boundaries of geothermal wells of different levels, thereby improving the matching and allocation accuracy of the scheduling strategy for multiple types of geothermal resources. On the one hand, by constructing a multi-scale load prediction model, taking historical load data and historical meteorological data as input, generating day-ahead, daytime and real-time granularity load prediction results, the scheduling strategy can cover long-term operation trends and short-term load fluctuations, thereby overcoming the problems of single time scale and slow response to sudden load changes in related scheduling methods. On the other hand, based on the well group output model and the load prediction results, through joint calculation with the multi-objective optimization function, the heat storage capacity and temperature decay trend of the geothermal well group can be considered while meeting the load demand, realizing dynamic allocation of heat output proportion, and significantly improving the configuration efficiency and operation balance ability of geothermal resources. On the other hand, through the introduction of the multi-objective optimization function, the generation of scheduling parameters is not limited to load matching targets, but can also realize energy supply optimization among indicators such as operation cost, carbon emissions and energy efficiency, effectively enhancing the stability and sustainability of the geothermal utilization system under different load scenarios and operation cycles.

[0021] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0022] The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, further serve to explain the principles of the present disclosure. It is apparent that the accompanying drawings described below are only some embodiments of the present disclosure, and other drawings can be obtained by those of ordinary skill in the art without creative effort based on these drawings.

[0023] Figure 1 A flowchart diagram of a multi-level geothermal well collaborative scheduling method based on multi-time scale prediction is schematically shown according to some embodiments of the present disclosure.

[0024] Figure 2 A flowchart diagram of obtaining optimal scheduling parameters is schematically shown according to some embodiments of the present disclosure.

[0025] Figure 3 A structural composition diagram of a multi-level geothermal well collaborative scheduling system based on multi-time scale prediction is schematically shown according to some embodiments of the present disclosure.

[0026] Figure 4 A block diagram of a multi-level geothermal well collaborative scheduling system based on multi-time scale prediction is schematically shown according to some embodiments of the present disclosure.

[0027] Figure 5 A structural diagram of a computer system of an electronic device is schematically shown according to some embodiments of the present disclosure.

[0028] Figure 6 A diagram of a computer readable storage medium is schematically shown according to some embodiments of the present disclosure.

[0029] In the drawings, the same or corresponding numbers represent the same or corresponding parts. DETAILED DESCRIPTION

[0030] The exemplary embodiments will be described in detail herein with reference to the accompanying drawings. The following description is with reference to the drawings, in which like numerals represent like elements throughout the several views. The following description of exemplary embodiments is not representative of all embodiments consistent with the present description. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present description as detailed in the appended claims.

[0031] The terminology used in this description is for the purpose of describing particular embodiments only and is not intended to be limiting of this description. As used in this description and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0032] It will be understood that, although the terms first, second, third, etc. can be used herein to describe various information, these terms are not intended to denote a temporal or chronological order. Rather, these terms are used only to distinguish different sets of information from one another. For example, a first information can also be termed a second information, and similarly, a second information can also be termed a first information, without departing from the scope of the present description. As used herein, the word "if' can be interpreted to mean "when" or "upon" or "in response to determining" depending on the context.

[0033] Example embodiments now will be described more fully hereinafter with reference to the accompanying drawings; however, these embodiments should not be construed as limiting the scope of the disclosure, but merely as illustrating different aspects of it. Accordingly, while example embodiments can be modified, and equivalents can be used, without departing from the scope of the disclosure.

[0034] Additionally, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the aspects of the disclosure can be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures, devices, and operations have not been shown or described in detail to avoid obscuring aspects of the disclosure.

[0035] Further, the drawings are diagrammatic and schematic representations of such information, but they are not necessarily drawn to scale. The embodiments depicted in the figures can advantageously be implemented in software, firmware, hardware, or various combinations thereof. It should be appreciated that the depicted block diagrams, as well as the method steps of the aspects of the disclosure are only meant to be exemplary. Numerous changes such as substitution, re-arrangement or omission can be made by one skilled in the art without departing from the spirit of the aspects of the disclosure.

[0036] In the present example embodiment, a multi-level geothermal well cooperative scheduling method based on multi-time scale prediction is first provided, which can be applied to heat exchange type geothermal wells, and can also be applied to water-heat type geothermal wells. Figure 1 A flowchart of a multi-level geothermal well cooperative scheduling method based on multi-time scale prediction according to some embodiments of the present disclosure is schematically shown. Referring toFigure 1 As shown, the multi-level geothermal well cooperative scheduling method based on multi-time scale prediction can include the following steps:

[0037] In step S110, based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, a well group output model is constructed, which can represent the unit heat output capacity, heat storage capacity and temperature attenuation characteristics of each well group.

[0038] In step S120, a multi-scale load prediction model is constructed based on historical load data and historical weather data.

[0039] In step S130, the multi-scale load prediction model is used to determine the load prediction result of the target period, which includes day-ahead load, daytime load and real-time load.

[0040] In step S140, the well group output model, the load prediction result and the multi-objective optimization function are used for joint calculation to obtain the optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in the current scheduling period.

[0041] According to the multi-level geothermal well cooperative scheduling method based on multi-time scale prediction in the present example embodiment, by modeling the operation parameters of the shallow geothermal well group and the deep geothermal well group, a well group output model is constructed which can represent the unit heat output capacity, heat storage capacity and temperature attenuation characteristics. The introduction of the hierarchical difference characteristics of geothermal resources in the scheduling parameter formation process can more accurately represent the heat output characteristics and operation boundaries of geothermal wells of different levels. On the one hand, by constructing a multi-scale load prediction model, historical load data and historical weather data are input to generate day-ahead, daytime and real-time granularity load prediction results, so that the scheduling strategy can cover long-term operation trends and short-term load fluctuations. On the other hand, based on the well group output model and the load prediction result, joint calculation is performed with the multi-objective optimization function, which can consider the heat storage capacity and temperature attenuation trend of the geothermal well group while meeting the load demand, realize dynamic allocation of heat output proportion, and significantly improve the configuration efficiency and operation balance ability of geothermal resources. On the other hand, the introduction of the multi-objective optimization function makes the generation of scheduling parameters not only limited to load matching targets, but also can realize energy supply optimization among operation cost, carbon emissions and energy efficiency indicators, effectively enhancing the stability and sustainability of geothermal utilization system under different load scenarios and operation periods.

[0042] In the following, the multi-level geothermal well cooperative scheduling method based on multi-time scale prediction in the present example embodiment will be further described.

[0043] In step S110, a well group output model is constructed based on the operation parameters of the shallow geothermal well group and the deep geothermal well group. The well group output model can represent the unit heat output capacity, heat storage capacity and temperature attenuation characteristics of each well group.

[0044] The shallow geothermal well group can represent a set of geothermal wells with a well depth less than a first preset threshold, which can be 200 meters, 300 meters or other suitable values. The deep geothermal well group can represent a set of high-temperature geothermal wells with a well depth greater than a second preset threshold, which can be 3000 meters, 3200 meters or other suitable values. Of course, in other embodiments of the present disclosure, when the well depth is greater than the first preset threshold, the corresponding geothermal well can be attributed to the deep geothermal well group. The well group output model can represent a mathematical function combination relationship established based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, and serve as one of the input bases for the scheduling parameter calculation. The unit heat output capacity can represent the effective heat that a single well group can output per unit time, which can be calculated by the flow parameter, heat exchange temperature difference and geothermal heat exchange coefficient of the well group, and reflects the instantaneous energy supply capacity of the geothermal resource. The heat storage capacity can represent the accumulation capacity of the underground medium corresponding to the well group for heat energy during the non-heating season operation, which is related to the thermal conductivity, specific heat capacity and chronological operation temperature change of the well area formation. The temperature attenuation characteristic can represent the change law that the well group water temperature gradually decreases with time under continuous operation conditions.

[0045] In addition, the operation parameters can represent various real-time and historical data sets collected during the operation of the shallow geothermal well group and the deep geothermal well group. For example, the operation parameters can include one or more of temperature field data, well output parameters, pipe network pressure data, equipment operation data, formation parameters and auxiliary equipment performance parameters. The temperature field data can include wellhead temperature, bottom hole temperature and geothermal gradient, the well output parameters can include single well flow, inlet and outlet temperature difference, inlet and outlet pressure and temperature, the pipe network pressure data can include pressure fluctuation value of the main pipe network and temperature, pressure and flow information of key nodes, the equipment operation data can include instantaneous flow, operating pressure, temperature, operating state, operating power, power consumption and speed of the equipment, the formation parameters can include the thermal conductivity of the soil or rock formation in the shallow layer region, and the auxiliary equipment performance parameters can include the performance coefficient of the heat pump system and the heat collection efficiency of the photothermal system. Of course, in other embodiments of the present disclosure, the operation parameters can also include other suitable geothermal well operation parameters.

[0046] In step S120, a multi-scale load prediction model is constructed based on historical load data and historical weather data.

[0047] The historical load data can represent a sequence of heating or cooling load data recorded by the geothermal well energy supply area in the past period of time, and can include instantaneous load, daily average load, peak load, and load change rate information at different time granularities. The historical meteorological data can represent a set of meteorological monitoring data in the corresponding time period of the historical load data, including temperature, humidity, wind speed, solar radiation, and other indicators, for characterizing the influence factors of external environment on load demand changes. The multi-scale load prediction model can represent a mathematical model capable of outputting multi-scale load prediction results for the target time period. The multi-scale load prediction model can be constructed using a hybrid neural network.

[0048] In step S130, a multi-scale load prediction model is used to determine a load prediction result for the target time period, and the load prediction result includes day-ahead load, day-time load, and real-time load.

[0049] The load prediction result can represent the data output result for the heating or cooling demand in the target time period generated by the multi-scale load prediction model. The result is divided into day-ahead load, day-time load, and real-time load according to different time granularities. The day-ahead load can represent the pre-forecast result generated based on historical load data and historical meteorological data. The prediction result has a time unit of a single day and a time granularity of 24 hours, and is used to form an initial plan data for the thermal load in the dispatch day. The day-time load can represent the medium-term prediction result generated by rolling update in the dispatch operation day. The prediction result has a preset day-time granularity as the update period, and the preset day-time granularity can be 2 hours, 3 hours, 4 hours, or other suitable lengths. The real-time load can represent real-time load data generated by short-term prediction based on historical data and current state in the geothermal system operation period. The real-time load data has a minute-level time granularity and can be updated rolling with a preset interval of 5 minutes, 10 minutes, 15 minutes, 30 minutes, etc.

[0050] In step S140, a well group output model, a load prediction result, and a multi-objective optimization function are used for joint calculation to obtain optimal scheduling parameters for the shallow geothermal well group and the deep geothermal well group in the current scheduling period.

[0051] The multi-objective optimization function can represent a set of optimization objectives for guiding the heat source output configuration based on the geothermal system operation constraints, which can include operation cost, carbon emission, energy utilization efficiency, and stratum heat balance constraint, and the like. The optimal scheduling parameter can represent a set of geothermal system operation configuration parameters calculated based on the well group output model, the heat load prediction result, and the multi-objective optimization function in a specific scheduling period. The optimal scheduling parameter can be used to determine the start-stop state of the shallow geothermal well group and the deep geothermal well group, the heat output distribution ratio, the heat charging-discharging strategy of the heat storage unit, and the enabling condition of the auxiliary energy unit, so as to realize the collaborative optimization among the operation cost, carbon emission, and energy efficiency under the premise of meeting the stratum heat balance, user load demand, and pipe network safety constraint.

[0052] In the following, the technical content in the above embodiments is described in detail.

[0053] In some embodiments, based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, the well group output model is constructed, including: obtaining the operation parameters of the shallow geothermal well group and the deep geothermal well group, the operation parameters including temperature field data, well output parameters, soil thermal conductivity, auxiliary equipment performance parameters, and shallow geothermal well recharge parameters; based on the temperature field data, a temperature response function is constructed for representing the temperature change trend of each geothermal well in the operation process; based on the well output parameters, an output function is constructed for representing the heat production capacity of each geothermal well per unit time; based on the soil thermal conductivity, a geotemperature decay function is constructed for representing the stratum temperature decay of each geothermal well with time; based on the auxiliary equipment performance parameters, an auxiliary energy supply function is constructed for representing the output characteristics of the auxiliary equipment per unit time; based on the shallow geothermal well recharge parameters, a shallow heat storage function is constructed for representing the heat storage capacity of the shallow geothermal well in the non-heating period; and based on the temperature response function, the output function, the geotemperature decay function, the auxiliary energy supply function, and the shallow heat storage function, the well group output model is constructed.

[0054] The temperature field data can represent data for representing the temperature distribution state of the geothermal well at different depth positions. The well output parameters can represent operation data for representing the heat exchange capacity of the geothermal well per unit time. The soil thermal conductivity can represent a geothermal property index for representing the heat conduction rate per unit length of the shallow or deep stratum. The auxiliary equipment performance parameters can represent data reflecting the performance characteristics of the auxiliary energy supply equipment in the cold and heat regulation and energy storage process. The shallow geothermal well recharge parameters can represent related data for representing the hydraulic and thermal characteristics in the shallow well recharge process, which can reflect the heat storage behavior and stratum heat recovery capacity of the shallow geothermal well during the non-heating season.

[0055] For example, the temperature response function can be constructed as:

[0056]

[0057] wherein, represents the effective temperature of the wellhead of the geothermal well numbered at time ; represents the initial temperature of the well; represents the temperature decay coefficient, which can be calculated according to the wellhead temperature, the bottom temperature and the geothermal gradient.

[0058] The output function can be constructed as:

[0059]

[0060] wherein, represents the heat power output of the geothermal well numbered at time ; represents the flow rate of the well at time ; represents the temperature difference between the inlet and outlet of the well; is the specific heat capacity constant.

[0061] The ground temperature decay function can be constructed as:

[0062]

[0063] wherein, represents the ground temperature decay factor of the geothermal well numbered at time ; represents the heat decay coefficient determined based on the ground thermal conductivity coefficient.

[0064] The shallow layer heat storage function can be constructed as:

[0065]

[0066] wherein, represents the heat storage capacity estimation of the shallow geothermal well at time ; represents the recharge flow rate; represents the temperature difference between the recharge water and the ground temperature; is the heat storage efficiency coefficient.

[0067] The auxiliary energy supply function can represent the peak shaving capacity of auxiliary equipment (such as air source heat pump), which can be constructed as:

[0068]

[0069] wherein, represents the heat storage capacity estimation of the shallow geothermal well numbered auxiliary equipment at time available heat; represents the energy efficiency coefficient of the equipment, such as the heating or cooling performance coefficient; represents the rated power of the equipment; represents the operating state coefficient, whose value is between 0 and 1.

[0070] Of course, in other embodiments of the present disclosure, the temperature response function, the output function, the ground temperature attenuation function, the auxiliary energy supply function, and the shallow layer heat storage function can also be adaptively set according to the actual application scenario.

[0071] In some embodiments, a multi-scale load prediction model is constructed based on historical load data and historical meteorological data, including the following technical steps: feature extraction is performed on the historical load data and the historical meteorological data to obtain structural index features and time series change features; a day-ahead prediction sub-model for outputting a first time granularity load is constructed based on the structural index features; a mid-day prediction sub-model for outputting a second time granularity load and a real-time prediction sub-model for outputting a third time granularity load are respectively constructed based on the time series change features; wherein the time interval of the first time granularity is longer than the time interval of the second time granularity, and the time interval of the second time granularity is longer than the time interval of the third time granularity.

[0072] The structural index features can represent periodic or categorical structural features in the load data and the meteorological data. The time series change features can represent continuity features constructed based on the numerical fluctuation trend in the historical load data and the historical meteorological data. The first time granularity can represent the time interval unit of the longest prediction period adopted in the load prediction process, which can be 24 hours. The second time granularity can represent the time interval unit of the medium prediction period adopted in the load prediction process, which can be 3 to 6 hours, for outputting the load prediction result in the mid-day stage. The third time granularity can represent the time interval unit of the shortest prediction period adopted in the load prediction process, which can be 10 minutes, 15 minutes, or 30 minutes, etc., for outputting the short-period load prediction result at the real-time level. The day-ahead prediction sub-model can represent a prediction model for outputting the first time granularity load prediction result, which can be constructed using gradient boosting decision trees, random forests, or other supervised learning models for static features. The mid-day prediction sub-model can represent a prediction model for outputting the second time granularity load prediction result constructed based on the time series change features, and the real-time prediction sub-model can represent a prediction model for outputting the third time granularity load prediction result constructed based on the time series change features. The mid-day prediction sub-model and the real-time prediction sub-model can be constructed using gated recurrent neural networks, long short-term memory networks, or other models suitable for time series data.

[0073] In the embodiment, different granularity prediction models are provided with targeted inputs, day-ahead prediction sub-models are constructed based on structural index features, the regularity and stability of single-day heat load can be effectively mined, and the planning and accuracy of the feedforward scheduling strategy can be improved. The day-middle prediction sub-model and the real-time prediction sub-model are respectively constructed based on the time sequence change feature, which further enhances the adaptability to dynamic fluctuations of the load.

[0074] In some embodiments, structural index features and time sequence change features are extracted from historical load data and historical meteorological data, including the following technical steps: based on the historical load data and the historical meteorological data, multi-type structural data is extracted, wherein the multi-type structural data includes any one or more of date type, month, energy supply area, historical peak load and historical average load; the multi-type structural data is discretized to obtain the structural index features; and the time sequence change features are constructed using the time sequence coupling relationship of the historical load data and the historical meteorological data.

[0075] The date type can represent the calendar attribute of the load data, including weekdays, weekends or holidays, and is used to represent the periodic change characteristics of energy consumption behavior. The month can represent the monthly time identifier of the load data, and is used to represent the change trend of the load in different seasons. The energy supply area can represent the geographic number of the geothermal well energy supply area, and is used to reflect the difference in load distribution between regions. The historical peak load can represent the maximum load value appearing in a given statistical period, and is used to represent extreme energy demand. The historical average load can represent the average load level measured in a specified period. Discretization processing can represent the process of converting original structural data into discrete input features through interval division or numerical mapping. The time sequence coupling relationship can represent the corresponding relationship between the historical load data and the historical meteorological data in the time dimension.

[0076] Specifically, when discretizing the multi-type structural data, the following technical steps can be used: the data fields belonging to the category type in the multi-type structural data are processed by one-hot encoding to obtain the corresponding category vector. Then, the data fields belonging to the numerical type in the multi-type structural data are discretely divided according to the preset numerical interval to generate the corresponding interval label vector, and the category vector and the interval label vector are spliced to obtain the structural index features. In the embodiment, by discretizing the multi-type structural data, heterogeneous information can be converted into unified structural index features, the convergence speed and generalization ability of the model in the training stage can be improved, and the prediction accuracy of the day-ahead prediction sub-model can be improved.

[0077] In some embodiments, a time series change feature is constructed by utilizing the time series coupling relationship between historical load data and historical meteorological data, including the following technical steps: constructing a load change rate sequence based on historical load data, constructing a temperature change sequence and a humidity change sequence based on historical meteorological data; performing cross operation on the load change rate sequence and the temperature change sequence to obtain a load-temperature cross sequence; combining the load change rate sequence and the humidity change sequence to obtain a load-humidity combined sequence; and constructing the time series change feature based on the load change rate sequence, the temperature change sequence, the humidity change sequence, the load-temperature cross sequence, and the load-humidity combined sequence.

[0078] The load change rate sequence can represent a time sequence formed by difference or ratio calculation of thermal load values at adjacent time points based on historical load data at a set time interval, and is used to represent the dynamic change trend of thermal load per unit time. The temperature change sequence can represent a temperature value sequence constructed in time sequence based on the external environment temperature values recorded in historical meteorological data. The humidity change sequence can represent a humidity value sequence constructed in time sequence based on the humidity values in historical meteorological data. The load-temperature cross sequence can represent a sequence generated by point-by-point cross operation of the load change rate sequence and the temperature change sequence, and can be used to represent the relationship between load and temperature. The load-humidity combined sequence can represent a sequence constructed by splicing and combining the load change rate sequence and the humidity change sequence in the feature dimension, and can be used to describe the synchronous change feature between load and humidity. In addition, the time intervals of the load change rate sequence, the temperature change sequence, the humidity change sequence, and the load-temperature cross sequence can be set according to the time granularity of the corresponding prediction sub-model. Time series change features of different time granularities are used to construct different prediction sub-models.

[0079] In this embodiment, the load-temperature cross sequence and the load-humidity combined sequence can more accurately reflect the correlation between load change and temperature and humidity conditions, especially in weather with high humidity and rapid temperature change. The model can identify the trend of load increase or decrease in advance, thereby improving the timeliness and matching degree of the prediction result.

[0080] In some embodiments, the day-ahead prediction sub-model can be constructed based on gradient boosting decision trees. Specifically, first, a training dataset is constructed based on structural indicator features, each sample in the training dataset containing structural indicator features and corresponding historical load values, the structural indicator features including one or more of date type, month number, energy supply area number, historical peak load, and historical average load. Then, based on the training dataset, a gradient boosting decision tree regression algorithm is used for modeling, and the first regression sub-tree is initialized to fit the initial residual of the sample. According to the preset iteration number or error convergence condition, a plurality of regression sub-trees are gradually constructed, and the output result of the overall model is updated in each iteration. The plurality of regression sub-trees are integrated to obtain a day-ahead prediction sub-model for representing the mapping relationship between different structural indicator features and day-ahead thermal load.

[0081] In some embodiments, the day-ahead prediction sub-model and the real-time prediction sub-model can be constructed using a long short-term memory network. Specifically, first, a training dataset is constructed based on time series variation features, the training dataset including a plurality of sample sequences obtained by sliding and intercepting at a fixed time step, the sample sequence consisting of a load change rate sequence, a temperature change sequence, a humidity change sequence, a load-temperature cross sequence, and a load-humidity joint sequence. Second, a neural network structure is constructed, including an input layer, a plurality of long short-term memory network unit layers, a fully connected layer, and an output layer, the input layer receiving the sample sequence as input, the long short-term memory network unit layer being used to extract time correlation features in the sample sequence, the fully connected layer being used to convert the extraction result, and the output layer being used to output a feature embedding vector. Then, the real load values within a historical time window are used as supervision labels, the difference between the predicted value and the actual value is calculated based on the mean square error loss function, and the gradient descent algorithm is used to iteratively update the weight parameters of the neural network structure. Finally, the foregoing training process is repeatedly executed until the loss function meets the preset convergence condition, and finally the day-ahead prediction sub-model for medium time granularity load modeling and the real-time prediction sub-model for short time granularity load modeling are obtained.

[0082] In some embodiments, when using a multi-scale load forecasting model to determine the load forecast results for a target time period, firstly, historical load data and historical meteorological data covering the current time point and its preceding preset duration are acquired. The historical load data characterizes the dynamic change trend of regional heat load, while the historical meteorological data includes ambient temperature, humidity, and other environmental data within the corresponding time period. Secondly, based on the historical load data and historical meteorological data, structural index features and temporal variation features are extracted. Then, the structural index features are input into the trained daytime forecasting sub-model to obtain the daytime load forecast results corresponding to the first time granularity. Next, the temporal variation features are input into the trained midday forecasting sub-model and real-time forecasting sub-model to obtain the midday load forecast results corresponding to the second time granularity and the real-time load forecast results corresponding to the third time granularity, respectively. Finally, considering the time granularity requirements of the current scheduling cycle, the daytime load forecast results, midday load forecast results, and real-time load forecast results are adapted and combined to generate the load forecast results for the target time period.

[0083] Furthermore, when adapting and combining the day-ahead load forecast results, mid-day load forecast results, and real-time load forecast results, the following steps can be taken: First, based on the target time granularity corresponding to the current scheduling cycle, determine the required forecast time range and divide the target time range into multiple forecast segments, with the duration of each forecast segment consistent with the target time granularity. Second, for forecast segments located after the current time point with an interval greater than a preset duration (e.g., 12 hours), select the forecast value of the corresponding time point in the day-ahead load forecast results as the initial forecast value for that segment. Then, for forecast segments located after the current time point with an interval within the short to medium range (e.g., 1 hour to 12 hours), select the forecast value of the corresponding time point in the mid-day load forecast results as the initial forecast value for that segment. Finally, for forecast segments close to the current time point (e.g., with a granularity less than 1 hour), select the forecast value of the corresponding time point in the real-time load forecast results as the initial forecast value for that segment. Finally, after obtaining the prediction results at each granularity, for prediction segments with time overlap, the output results of the prediction model with the smallest time granularity are retained first, and the boundary transition is processed using the sliding window weighted smoothing method to obtain continuous and consistent target period load prediction results.

[0084] In some embodiments, reference Figure 2 As shown, the optimal scheduling parameters for shallow and deep geothermal well groups within the current scheduling cycle are obtained through joint calculation using a well group output model, load prediction results, and a multi-objective optimization function. The specific technical steps include the following:

[0085] Step S210: Construct a multi-objective optimization function based on formation thermal balance, operating cost, carbon emissions, and efficiency.

[0086] wherein, the formation thermal balance can represent the ability of the formation temperature field to maintain a dynamic stable state during the continuous operation of the geothermal well group. The operation cost can represent the specific expenditure generated by the operation of various types of equipment in the geothermal system, including the geothermal well group, the heat pump system, the heat storage unit, etc. during the dispatching period. The carbon emission can represent the carbon dioxide emissions caused by the consumption of external power or auxiliary fuel by various energy conversion equipment during the dispatching period. The efficiency can represent the ratio between the output of effective heat and the total energy consumed per unit of time.

[0087] Step S220, based on the well group output model and the load prediction result, a plurality of sets of candidate scheduling parameters in the current dispatching period are obtained.

[0088] wherein, the scheduling parameter can represent a set of control variables used to regulate the operation state of the shallow geothermal well group and the deep geothermal well group in the current dispatching period, which includes but is not limited to: the start-stop state of each geothermal well; the target water outlet temperature of the geothermal well; the target heat supply flow of each well group; the participation ratio or working state of different energy equipment (such as heat pump, heat storage unit); the energy supply bearing ratio of the target energy supply load in the current dispatching period.

[0089] In the specific implementation process, first, based on the heat load prediction result corresponding to the current dispatching period, the target heat supply load value in the dispatching period is determined, which is used to represent the heat power demand that each heat source unit needs to meet together. Second, based on the well group output model, combined with the current operation parameters of the shallow geothermal well group and the deep geothermal well group, the adjustable output boundary of each well group in the current dispatching period is obtained, which includes the minimum output limit and the maximum output limit. Then, according to the current operation scene, including the season type, the load demand and the abnormal state, etc., the adjustable equipment set that meets the start-stop constraint condition is selected from the geothermal system, the short-time heat storage system and the air source heat pump system, forming a preliminary available equipment list. Finally, under the condition of meeting the target heat supply load value, a plurality of sets of candidate scheduling parameters containing different device output ratios are generated by using a plurality of combination modes, each set of candidate scheduling parameters can include the start-stop state, the unit output level and the target energy supply ratio of each device in the current dispatching period, etc.

[0090] Step S230, the optimal scheduling parameter meeting the multi-objective optimization function is obtained from the plurality of sets of candidate scheduling parameters using the target genetic algorithm.

[0091] The genetic algorithm can represent an optimization algorithm simulating natural selection and biological evolution mechanism. In the embodiments of the present disclosure, the genetic algorithm can be a non-dominated sorting based multi-objective genetic optimization algorithm, a multi-objective particle swarm optimization algorithm, and other suitable genetic algorithms. The optimal scheduling parameter can represent a set of scheduling parameters obtained by filtering from the candidate scheduling parameter set in the current scheduling period through the multi-objective optimization function.

[0092] In some embodiments, a multi-objective optimization function is constructed based on formation heat balance, operating cost, carbon emission and exergy efficiency, specifically including the following technical steps: constructing a temperature change rate function reflecting the influence of heat extraction process on formation temperature according to the thermal conductivity coefficient of the geological layer corresponding to the geothermal well group, the temperature difference between the recharge fluid and the water outlet fluid, and the temperature gradient between the wellhead and the well bottom; determining a heat extraction upper limit value that makes the temperature change rate equal to a preset rock layer temperature change rate threshold according to the temperature change rate function; constructing a multi-objective optimization function based on the heat extraction upper limit value, an operating cost function, a carbon emission function and an exergy efficiency function.

[0093] The heat extraction upper limit value can represent the maximum heat extraction amount of the geothermal well group in the current scheduling period under the constraint condition of formation heat balance. The operating cost function can represent a function expression reflecting the cost relationship of unit heat corresponding to each type of heat source equipment in the scheduling period. The carbon emission function can represent a function expression reflecting the total amount of carbon dioxide emissions generated in the operation process of different heat source equipment. The exergy efficiency function can represent a function expression for measuring the efficiency relationship between the unit input energy consumption and the effective output heat of each heat source in the multi-heat source collaborative energy supply system.

[0094] For example, the temperature change rate function can be represented as:

[0095]

[0096] wherein, represents the change rate of the formation temperature per unit time under the action of the geothermal well; represents the thermal conductivity coefficient of the formation; represents the temperature gradient between the well bottom temperature and the wellhead temperature; represents the density of the formation medium; represents the specific heat capacity of the formation medium; represents the thermal influence volume; represents the heat extraction disturbance coefficient; represents the temperature difference between the water outlet temperature and the recharge temperature.

[0097] According to the above temperature change rate function, the upper limit of the allowable annual temperature change rate of the rock layer is set as To avoid heat imbalance of the formation, the following heat balance constraint condition should be met:

[0098]

[0099] Under the premise of meeting the above constraints, the time length of the scheduling period is combined , and a heat extraction upper limit value function is constructed:

[0100]

[0101] wherein, represents the maximum allowed heat extraction amount of the corresponding well group in the current scheduling period.

[0102] In some embodiments, the target genetic algorithm is a non-dominated sorting genetic algorithm II (NSGA-II), and the optimal scheduling parameters that meet the multi-objective optimization function are obtained from the multiple sets of candidate scheduling parameter sets by using the target genetic algorithm. The specific technical steps include the following: an initial scheduling population is constructed based on the candidate scheduling parameter set, and each scheduling parameter individual in the initial scheduling population includes output configuration variables of the shallow geothermal well group and the deep geothermal well group; the scheduling parameter individuals in the initial scheduling population are subjected to multi-objective fitness evaluation by using the multi-objective optimization function, and based on the evaluation results, the non-dominated levels are divided according to the non-dominated sorting principle, and at the same time, the corresponding crowding degree indexes are calculated in each level; based on the non-dominated levels and the crowding degree indexes, part of the scheduling parameter individuals are selected as parent individuals, and cross operation and mutation operation are performed to generate the next generation of scheduling parameter population; the multi-objective fitness evaluation and non-dominated sorting operation are repeatedly performed on the next generation of scheduling parameter population, and under the premise of meeting the preset iteration termination condition, the optimal scheduling parameters are selected from all the current scheduling parameter individuals.

[0103] wherein, the non-dominated sorting genetic algorithm is used as the target genetic algorithm, and fixed weights do not need to be set for the operating cost, carbon emission, and 㶲 efficiency, etc. targets, which avoids the scheduling parameters from being biased towards a certain target, and has lower time complexity and good convergence, and is suitable for the application scenarios of geothermal energy supply systems with short scheduling periods and limited computing resources.

[0104] In some embodiments, the multi-time scale prediction based multi-level geothermal well coordinated scheduling method described above can further include the following technical steps: collecting the operation parameters of the shallow geothermal well group, the deep geothermal well group, the short-time heat storage device and the air source heat pump, and determining the operation state of each heat source unit based on the operation parameters; in response to the operation state of any heat source unit being faulty, calling the backup output configuration strategy corresponding to the fault type, and adjusting the output proportion of each heat source unit in the scheduling parameters in combination with the current load prediction result and the backup output capacity; in the non-heating period, using the deep geothermal well group and / or the building heat to replenish the heat of the shallow geothermal well group; obtaining the temperature distribution data of the shallow geothermal well, and updating the shallow geothermal well recharge parameters in the well group output model according to the temperature distribution data.

[0105] Among them, the backup output configuration strategy can represent the preset scheduling parameter adjustment rule for coping with specific heat source unit fault conditions, which allocates the target output proportion of the replacement heat source according to the operation capacity of the current available heat source unit. The backup output capacity can represent the schedulable output capacity of the heat source unit that is not in the running state but can be called in the fault response, and the backup heat source unit can be any one or more of the deep geothermal well group, the short-time heat storage device and the air source heat pump. The heat replenishment can represent the process of replenishing heat energy from the deep geothermal well group and / or the building heat load to the shallow geothermal well in the non-heating period. The building heat can represent the recoverable heat energy generated during the operation of the building due to equipment operation, personnel activity, solar radiation and the heat storage and heat transfer effect of the building envelope. The shallow geothermal well recharge parameters can represent a parameter set for describing the operation condition characteristics of the shallow geothermal well in the heat replenishment process, including but not limited to recharge temperature, recharge flow, recharge duration and recharge pressure, etc.

[0106] In the implementation process, the operation parameters of the shallow geothermal well group, the deep geothermal well group, the short-time heat storage device and the air source heat pump can be acquired first, and the operation parameters include the water outlet temperature, the water return temperature, the flow, the pressure, the operation power, the operation state identifier and the device alarm signal and the like. Based on the operation parameters, the operation state of each heat source unit is analyzed, and if the operation state identifier of any heat source unit is detected to be abnormal or there is a device alarm signal, it is determined that the heat source unit is in a fault state. In response to the failure of a certain heat source unit, a standby output configuration strategy matched with the fault type is called from a preset output configuration strategy library, and the standby output configuration strategy can include the activation sequence of other heat source units for replacing the current fault unit, the target load distribution ratio and the scheduling priority information. According to the load prediction result at the current time point and the output capacity of the available standby heat source unit, the standby output capacity of each standby heat source unit is determined, and the scheduling parameters are updated accordingly, specifically including: adjusting the target output ratio of the shallow geothermal well group, the deep geothermal well group, the short-time heat storage device and the air source heat pump to form a dynamic scheduling scheme matched with the current operation state. Exemplarily, if the deep geothermal well group fails, the air source heat pump is started preferentially, and the operation power of the shallow geothermal well group is increased; if the shallow geothermal well group fails, the circulating flow rate of the deep geothermal well group is increased, and the air source heat pump is called as a supplementary heat source; if the air source heat pump fails, the shallow and deep well groups are combined for heat extraction, and the short-time heat storage device is activated to release the heat storage capacity; if the short-time heat storage device fails, the output ratio of the shallow and deep geothermal wells is redistributed according to the current heat extraction capacity, and the air source heat pump capacity is fully mobilized to make up for the heat gap.

[0107] Further, when updating the shallow geothermal well recharge parameters in the well group output model according to the temperature distribution data, the following steps can be taken:

[0108] First, the heat received by the shallow geothermal well is calculated according to the water supply parameters of the heat source:

[0109]

[0110] wherein, represents the total heat received by the shallow well per unit time, represents the number of heat sources participating in the recharge, represents the specific heat capacity of water, represents the density of water, represents the flow of the th heat source, represents the water supply temperature of the th heat source, represents the initial temperature of the shallow well recharge.

[0111] Second, the temperature distribution data of the shallow geothermal well group is collected , for judging the degree of thermal field recovery, wherein, represents the geothermal distribution at the three-dimensional spatial position and time . The heat recovery rate of the shallow geothermal well is calculated in combination with the heat supplement and temperature rise:

[0112]

[0113] wherein, represents the heat recovery rate, represents the effective heat exchange area of the shallow well, represents the average temperature rise, and is calculated by difference based on represents the heat supplement time period.

[0114] In the third step, the heat recovery rate is used to update the shallow well heat storage capacity function:

[0115]

[0116] wherein, represents the total amount of the shallow heat storage at time , represents the starting time of heat supplement, represents the integral variable.

[0117] In addition, in other embodiments of the present disclosure, when multi-level geothermal well coordinated scheduling based on multi-time scale prediction is performed, different scheduling scene strategies can be dynamically matched in combination with the actual operation season, environmental temperature interval and load level, and the corresponding scheduling parameter set is generated under the guidance of the strategy.

[0118] Specifically, in the winter heating scheduling scene, in response to the operating condition that the external environmental temperature is between -15 and 5°C, the load prediction results at the day-ahead, daytime and real-time three time nodes are generated by the load prediction model, which are used as the input basis for constructing the device output model. Based on the device output model, the scheduling system can call the deep geothermal well and plate heat exchanger for joint heating under the condition of low load, call the shallow and deep geothermal wells for joint heating under the condition of medium load, further call the ground source heat pump system to provide energy support under the condition of high load, and collect parameters such as underground temperature field, pipe network pressure difference and user end temperature change rate with a 15-minute update cycle, and dynamically adjust the output proportion of various heat source units based on real-time feedback results.

[0119] In the summer cooling scheduling scene, in response to the environmental temperature of 25-40°C and the cooling load demand of 6-18 MW, the shallow well group is preferentially called to bear the basic cooling load, and the day and night heat storage and heat release functions of the short-time heat storage system and the elastic regulation capacity of the air source heat pump system are combined to complete the cooling load distribution. ​​

[0120] Further, in the seasonal scheduling stage, the shallow well group heat extraction intensity decline gradient can be set according to the operation history of the shallow well group and the formation heat balance state in the spring recovery period, the target recovery temperature of each well and the formation is calculated, and the deep well group is controlled to supply heat to the shallow well group, and the well group output model is updated; in the summer cooling period, the building cooling load transfer is taken as the carrier, the heat is supplemented to the formation through the shallow well, the underground heat reservoir is reconstructed, the deep well group is opened in the heat penetration operation mode, and the dynamic soil thermal property database is used to correct the heat conduction model in real time; in the autumn reserve period, combined with the seasonal transition load fluctuation characteristics, the heat storage adjustment mechanism of the phase change heat storage unit is triggered, and the well group preventive maintenance and deep to shallow heat supply strategy are combined to realize the system energy pre-charging adjustment. In addition, in the cross-season energy storage process, the shallow well heat storage unit can be replaced or supplemented by building an aquifer heat storage, so as to ensure the stability of the geothermal system heat field while enhancing the medium and long-term heat storage capacity of the system.

[0121] In the extreme weather emergency scheduling scenario, the scheduling system can start the pre-activation mode of the deep geothermal well, and through the injection of high-temperature fluid into the short-time heat storage system, the output temperature of the system is increased to the design interval, and at the same time, the impact heat extraction operation is applied to the deep well group, and the pulse release mechanism of the heat storage body is enabled to provide rapid heat compensation at a rate of 15% per 2 hours. The air source heat pump system acts as a backup heat source in this process, and its operating power is limited to within 20% of the total system load to achieve energy security redundancy control.

[0122] In some embodiments, the scheduling method in the above embodiments can be completed by a scheduling system as shown in Figure 3 Specifically, the collaborative scheduling system provided in this embodiment includes four functional closed loops of data acquisition, load prediction, intelligent scheduling and effectiveness evaluation, which supports the joint operation and optimal control of multi-level geothermal wells (including shallow geothermal well group and deep geothermal well group) in different operation scenarios by constructing a multi-time scale load prediction and multi-source heat energy optimization distribution mechanism.

[0123] In the data acquisition module, local historical meteorological data, historical operation data of each system and other load data, including personnel activities, equipment status, building characteristics, etc. are obtained in sequence. The above raw data is archived through the data storage module, and structured data results are formed through business analysis and data governance processes. The data service module provides data access capability to downstream modules based on unified standards, so that the load prediction and intelligent scheduling modules can access the required information based on consistent data interfaces.

[0124] The load prediction module utilizes structured data to construct a double-layer structure of feature extraction and algorithm modeling. The feature extraction part includes time feature, meteorological feature and lag feature extraction, which is used to describe the correlation pattern between thermal load and meteorology. The algorithm modeling part includes time series model, tree model and BP neural network, collectively referred to as load prediction model, which is used to construct load prediction mechanism of different time scales (day-ahead, mid-day and real-time). The load prediction model outputs the load prediction result based on the feature input, and transmits it as input to the intelligent scheduling module.

[0125] In the intelligent scheduling module, a model system for describing the operation mechanism of the geothermal well system and the auxiliary heat source system is constructed through the device model, the system model and the operation model, and then the multi-objective optimization control is driven by the intelligent scheduling strategy engine. After receiving the load prediction result, the intelligent scheduling strategy engine generates the optimal scheduling strategy according to the combination of the set cross-season energy storage strategy, the peak-valley cooperation strategy, the day-level energy storage strategy and the emergency linkage strategy. At the same time, the module integrates boundary conditions, optimization objectives and optimization algorithm construction components, completes the scheduling parameter space search and solution in the algorithm optimization unit, and sends the optimized strategy to the actual operation system for execution.

[0126] The system execution result is fed back to the effectiveness evaluation module, which is evaluated by the energy efficiency evaluation model and engine. The evaluation index dimension includes energy efficiency, energy efficiency and unit heat value power consumption, which respectively judges from the aspects of energy utilization efficiency, scheduling accuracy and system economy. Further, through the multi-strategy cluster of energy efficiency optimal strategy, carbon emission minimum strategy, performance optimal strategy, cost minimum strategy and comprehensive strategy, the current strategy result is compared and analyzed and dynamically optimized, and the evaluation result is returned to the intelligent scheduling module for strategy optimization and algorithm optimization, so as to realize the continuous iteration and optimization of the scheduling result.

[0127] It should be noted that although the steps of the method in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired result. In addition or alternatively, some steps can be omitted, a plurality of steps can be combined into one step, and / or one step can be divided into a plurality of steps, etc.

[0128] In addition, in the present example embodiment, a multi-level geothermal well cooperative scheduling system based on multi-time scale prediction is also provided. Referring to Figure 4 As shown in the figure, the multi-level geothermal well cooperative scheduling system based on multi-time scale prediction 400 can include an output model construction module 410, a prediction model construction module 420, a load prediction module 430 and an intelligent scheduling module 440. Among them:

[0129] The output model construction module 410 can be configured to construct a well group output model based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, and the well group output model can represent the unit heat output capacity, heat storage capacity and temperature attenuation characteristics of each well group;

[0130] The prediction model construction module 420 can be configured to construct a multi-scale load prediction model based on historical load data and historical meteorological data.

[0131] The load prediction module 430 can be configured to determine a load prediction result of a target period by using the multi-scale load prediction model, and the load prediction result includes day-ahead load, day-time load and real-time load.

[0132] The intelligent scheduling module 440 can be configured to jointly calculate the well group output model, the load prediction result and a multi-objective optimization function to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period.

[0133] The specific details of the modules of the multi-level geothermal well cooperative scheduling system based on multi-time scale prediction have been described in detail in the corresponding multi-level geothermal well cooperative scheduling method based on multi-time scale prediction, and thus will not be described here.

[0134] It should be noted that although several modules or units of the multi-level geothermal well cooperative scheduling system based on multi-time scale prediction are mentioned in the above detailed description, such division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units.

[0135] In addition, in the exemplary embodiments of the present disclosure, an electronic device capable of implementing the above-mentioned multi-level geothermal well cooperative scheduling method based on multi-time scale prediction is also provided.

[0136] Those skilled in the art can understand that each aspect of the present disclosure can be implemented as a system, a method or a program product. Therefore, each aspect of the present disclosure can be embodied as a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module" or "system" here.

[0137] The electronic device 500 according to this embodiment of the present disclosure will be described below with reference to Figure 5 Figure 5 The electronic device 500 shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present disclosure. ​

[0138] As Figure 5 shown, the electronic device 500 is in the form of a general-purpose computing device. The components of electronic device 500 can include, but are not limited to, the at least one processing unit 510, the at least one storage unit 520, a bus 530 that connects the various system components, including the storage unit 520 and the processing unit 510, a display unit 540.

[0139] The storage unit stores program code that can be executed by the processing unit 510 such that the processing unit 510 performs the steps described in the above "Exemplary Methods" section according to various exemplary embodiments of the present disclosure. The storage unit 520 can include a readable medium in the form of volatile storage such as random access memory (RAM) 521 and / or cache 522, and further can include a read only memory (ROM) 523.

[0140] The storage unit 520 can also include program / utility 524 having a set of at least one program modules 525, including but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which can include implementation of a network environment, alone or in combination.

[0141] The bus 530 can represent one or more of several types of bus structures, including a storage bus or bus controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of a variety of bus architectures.

[0142] The electronic device 500 can also communicate with one or more external devices 570 such as a keyboard or pointing device, a Bluetooth device, etc.; other devices that enable a user to interact with the electronic device 500; and / or one or more devices that enable the electronic device 500 to communicate with one or more other computing devices. Such communication can be facilitated by an input / output (I / O) interface 550. Still yet, the electronic device 500 can communicate with one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or the Internet, through a network adapter 560. As depicted, the network adapter 560 communicates with the other components of the electronic device 500 through the bus 530. It should be appreciated that although the network adapter 560 is depicted as a single component, the network adapter 560 can comprise two or more components that work together to facilitate communications between the electronic device 500 and one or more other computing devices. It should also be appreciated that not all of the components shown in FIG. 5 can be required, that implementations of an electronic device 500 can include other components that are not explicitly shown, and that one or more components can be consolidated. For example, the storage unit 520 and these storage unit-related components can be consolidated into a single component (e.g., component 520 can be a single storage unit that includes one or more of the above-described storage unit-related components).

[0143] From the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware.

[0144] In the example embodiments of the present disclosure, a computer readable storage medium having stored thereon a program product capable of implementing the method described above is also provided. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program codes for causing an end device to perform the steps described in the “Example Method” section above according to various example embodiments of the present disclosure when the program product is run on the end device.

[0145] Reference Figure 6 As shown, a program product 600 for implementing the multi-time scale prediction based multi-level geothermal well coordinated scheduling method described above according to embodiments of the present disclosure is described, which can adopt a portable compact disc read-only memory (CD-ROM) and include program codes, and can be run on an end device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program, which can be used or combined with an instruction execution system, device or apparatus.

[0146] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0147] The computer readable signal medium can include a data signal propagated in a baseband or as a carrier wave in a propagated data signal, in which readable program codes are borne. Such a propagated data signal can take on many forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination thereof. The readable signal medium can also be any readable medium that is not a readable storage medium, which can send, propagate or transmit programs for use by or in connection with an instruction execution system, device or apparatus.

[0148] Further, the above-described diagrams are merely schematic illustrations of processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended for a limiting purpose. It is readily understood that the processes illustrated in the above-described diagrams do not indicate or limit the time sequence of the processes. In addition, it is readily understood that the processes can be executed, for example, synchronously or asynchronously in a plurality of modules.

[0149] It is to be understood that the present disclosure is not limited to the precise construction described above and illustrated in the accompanying drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A multi-level geothermal well coordinated scheduling method based on multi-time scale prediction, characterized in that, The method comprises the following steps: constructing a well group output model based on the operation parameters of the shallow geothermal well group and the deep geothermal well group, the well group output model being capable of representing the unit heat output capacity, heat storage capacity and temperature decay characteristics of each well group; constructing a multi-scale load prediction model based on historical load data and historical meteorological data; determining a load prediction result of a target period by using the multi-scale load prediction model, the load prediction result including day-ahead load, daytime load and real-time load; performing joint calculation by using the well group output model, the load prediction result and a multi-objective optimization function to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period; wherein the step of constructing a well group output model based on the operation parameters of the shallow geothermal well group and the deep geothermal well group comprises the following steps: obtaining operation parameters of the shallow geothermal well group and the deep geothermal well group, the operation parameters including temperature field data, well output parameters, soil thermal conductivity, auxiliary equipment performance parameters and shallow geothermal well recharge parameters; constructing a temperature response function for representing the temperature variation trend of each geothermal well during operation based on the temperature field data; constructing an output function for representing the heat production capacity of each geothermal well per unit time based on the well output parameters; constructing a geotemperature decay function for representing the decay of the formation temperature of each geothermal well with time based on the soil thermal conductivity; constructing an auxiliary energy supply function for representing the output characteristics of the auxiliary equipment per unit time based on the auxiliary equipment performance parameters; constructing a shallow heat storage function for representing the heat storage capacity of the shallow geothermal well during a non-heating period based on the shallow geothermal well recharge parameters; and constructing the well group output model based on the temperature response function, the output function, the geotemperature decay function, the auxiliary energy supply function and the shallow heat storage function; the step of constructing a multi-scale load prediction model based on historical load data and historical meteorological data comprises the following steps: extracting structural index features and time series variation features from the historical load data and the historical meteorological data; constructing a day-ahead prediction sub-model for outputting a first time granularity load based on the structural index features; constructing a daytime prediction sub-model for outputting a second time granularity load and a real-time prediction sub-model for outputting a third time granularity load based on the time series variation features; wherein the time interval of the first time granularity is longer than that of the second time granularity, and the time interval of the second time granularity is longer than that of the third time granularity; the step of performing joint calculation by using the well group output model, the load prediction result and a multi-objective optimization function to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period comprises the following steps: constructing the multi-objective optimization function based on formation heat balance, operation cost, carbon emissions and energy efficiency; obtaining a plurality of sets of candidate scheduling parameters in the current scheduling period based on the well group output model and the load prediction result; and obtaining optimal scheduling parameters satisfying the multi-objective optimization function from the plurality of sets of candidate scheduling parameters by using a target genetic algorithm.

2. The multi-time scale prediction based multi-level geothermal well coordinated scheduling method according to claim 1, characterized in that, The feature extraction is performed on the historical load data and the historical meteorological data to obtain structural index features and time sequence change features, including: Based on the historical load data and the historical meteorological data, a plurality of types of structural data are extracted, wherein the plurality of types of structural data include any one or more of a date type, a month, a power supply area, a historical peak load and an average load index; The plurality of types of structural data are discretized to obtain the structural index features; The time sequence change features are constructed by using a time sequence coupling relationship of the historical load data and the historical meteorological data.

3. The multi-time scale prediction based multi-level geothermal well coordinated scheduling method according to claim 2, characterized in that, The time sequence change features are constructed by using a time sequence coupling relationship of the historical load data and the historical meteorological data, including: Based on the historical load data, a load change rate sequence is constructed, and based on the historical meteorological data, a temperature change sequence and a humidity change sequence are constructed; The load change rate sequence and the temperature change sequence are cross-operated to obtain a load-temperature cross sequence; The load change rate sequence and the humidity change sequence are spliced to obtain a load-humidity combined sequence; Based on the load change rate sequence, the temperature change sequence, the humidity change sequence, the load-temperature cross sequence and the load-humidity combined sequence, the time sequence change features are constructed.

4. The multi-time scale prediction based multi-level geothermal well coordinated scheduling method according to claim 1, characterized in that, The multi-objective optimization function is constructed based on the formation heat balance, the operation cost, the carbon emission and the power efficiency, including: According to the thermal conductivity coefficient of the geological formation corresponding to the geothermal well group, the temperature difference between the recharge fluid and the water outlet fluid, and the temperature gradient between the wellhead and the well bottom, a temperature change rate function reflecting the influence of the heat extraction process on the formation temperature is constructed; According to the temperature change rate function, a heat extraction upper limit value is determined, which makes the temperature change rate equal to a preset rock formation temperature change rate threshold value; The multi-objective optimization function is constructed based on the heat extraction upper limit value, the operation cost function, the carbon emission function and the power efficiency function.

5. The multi-time scale prediction based multi-level geothermal well coordinated scheduling method according to claim 1, characterized in that, The target genetic algorithm is a non-dominated sorting multi-objective genetic algorithm, and the optimal scheduling parameter satisfying the multi-objective optimization function is obtained from the plurality of sets of candidate scheduling parameters by using the target genetic algorithm, including: Based on the candidate scheduling parameter set, an initial scheduling population is constructed, and each scheduling parameter individual in the initial scheduling population includes output configuration variables of the shallow geothermal well group and the deep geothermal well group; The multi-objective fitness of the scheduling parameter individuals in the initial scheduling population is evaluated by using the multi-objective optimization function, and based on the evaluation result, the non-dominated levels are divided according to the non-dominated sorting principle, and the corresponding crowding degree indexes are calculated in each level; Based on the non-dominated levels and the crowding degree indexes, part of the scheduling parameter individuals are selected as parent individuals, and cross operation and mutation operation are performed to generate a next generation scheduling parameter population; The multi-objective fitness evaluation and non-dominated sorting operation are repeatedly performed on the next generation scheduling parameter population, and when a preset iteration termination condition is met, the optimal scheduling parameter is selected from all the current scheduling parameter individuals.

6. The multi-time scale prediction based multi-level geothermal well coordinated scheduling method according to claim 1, characterized in that, Further comprising: Collecting operation parameters of the shallow geothermal well group, the deep geothermal well group, the short-time heat storage device and the air source heat pump, and determining the operation state of each heat source unit based on the operation parameters; In response to the operation state of any heat source unit being faulty, calling a backup output configuration strategy corresponding to the fault type, and adjusting the output proportion of each heat source unit in the scheduling parameter in combination with the current load prediction result and the backup output capacity; During the non-heating period, using the deep geothermal well group and / or building heat to replenish heat for the shallow geothermal well group; Obtaining temperature distribution data of the shallow geothermal well, and updating the shallow geothermal well recharge parameters in the well group output model according to the temperature distribution data.

7. A multi-level geothermal well coordinated scheduling system based on multi-time scale prediction, characterized in that, The system for implementing the multi-time scale prediction based multi-level geothermal well cooperative scheduling method of any one of claims 1 to 6, comprising: An output model construction module configured to construct a well group output model based on operation parameters of the shallow geothermal well group and the deep geothermal well group, the well group output model being capable of representing unit heat output capacity, heat storage capacity and temperature decay characteristics of each well group; A prediction model construction module configured to construct a multi-scale load prediction model based on historical load data and historical meteorological data; A load prediction module configured to determine a load prediction result of a target period by using the multi-scale load prediction model, the load prediction result including day-ahead load, day-time load and real-time load; An intelligent scheduling module configured to perform joint calculation by using the well group output model, the load prediction result and a multi-objective optimization function, to obtain optimal scheduling parameters of the shallow geothermal well group and the deep geothermal well group in a current scheduling period.

Images