Trough type solar heat collection temperature rise optimization method and device based on data analysis

Abstract

The invention provides a trough type solar heat collection temperature rise optimization method and device based on data analysis, and relates to the technical field of trough type solar energy, and the method comprises the steps: inputting a plurality of predicted radiation intensity intervals into a heat supply twin network for simulation, and carrying out the mapping compensation of the predicted heat supply amount of a plurality of intervals based on a plurality of predicted temperature intervals, obtaining a predicted heat supply amount; respectively inputting the plurality of predicted temperature intervals into a heat supply heat prediction unit, a life heat prediction unit, a greenhouse heat prediction unit and a cultivation heat prediction unit, and outputting predicted heat; and if the predicted heat consumption is greater than the predicted heat supply, performing auxiliary heating compensation according to the heat deviation. By means of the method, the technical problems that in an existing method, due to the fact that the fineness and accuracy of solar heating and heating demand analysis are low, the auxiliary heat supply amount cannot be accurately set, and the stability and reliability of regional heating are poor can be solved, the target of accurately setting the auxiliary heat supply amount can be achieved, and safe and stable regional heating is ensured.

Description

Technical Field

[0001] The present application relates to the field of trough solar energy technology, and in particular to a trough solar energy heat collection and heating optimization method and device based on data analysis. Background Art

[0002] With the development of rural areas in my country, more and more rural areas have begun to adopt solar central heating systems to meet the heating needs in winter. Among them, trough solar systems have gradually become one of the main choices for heating in rural areas due to their high efficiency and environmental protection. They can be used for heating houses, heating greenhouses, heating livestock houses, etc.

[0003] Greenhouses are important agricultural production facilities in rural areas. When the temperature is low in winter, the growth of crops in the greenhouse is limited. The use of trough solar heat collection and heating system can effectively increase the temperature in the greenhouse and ensure the normal growth of crops. In addition, the system can also be combined with a heat storage device to meet the heating needs at night or in rainy weather.

[0004] At present, in the existing rural centralized heating method using trough solar energy, due to the low precision and accuracy of solar heating and heating demand analysis, the accuracy and comprehensiveness of heating forecasts and heating demand forecasts are poor, and the auxiliary heating amount cannot be accurately set, resulting in technical problems such as poor stability and reliability of regional heating. Summary of the invention

[0005] The purpose of this application is to provide a trough solar thermal collection and heating optimization method and device based on data analysis, so as to solve the technical problems in the existing rural centralized heating method using trough solar energy, that is, due to the low precision and accuracy of solar heating and heating demand analysis, the accuracy and comprehensiveness of heating forecasts and heating demand forecasts are poor, the auxiliary heating amount cannot be accurately set, and the stability and reliability of regional heating are poor.

[0006] In view of the above problems, the present application provides a trough solar thermal collection and heating optimization method and device based on data analysis.

[0007] In the first aspect, the present application provides a trough solar thermal collection and heating optimization method based on data analysis, which is implemented by a trough solar thermal collection and heating optimization device based on data analysis, including: reading the basic operating data of the target heating system, performing twin simulation based on the basic operating data, and constructing a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operating data include mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds, and water storage tank capacity; obtaining environmental prediction values ​​within a predetermined future window, wherein the environmental prediction values ​​include predicted radiation intensity, predicted sunshine duration, and predicted temperature; clustering the predicted radiation intensity and the predicted temperature to determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals; and clustering the multiple predicted radiation intensities. The heating twin network is simulated by inputting a plurality of intervals, and a plurality of interval predicted heating supplies are outputted; the plurality of interval predicted heating supplies are mapped and compensated based on the plurality of predicted temperature intervals, and the predicted heating supply is obtained by summing up; taking the regional structure characteristics as constraints, a plurality of sample heat usage data sets are retrieved and obtained, and a heat usage predictor is obtained by training based on the plurality of sample heat usage data sets, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit and a breeding heat usage prediction unit; the plurality of predicted temperature intervals are respectively inputted into the heating heat usage prediction unit, the living heat usage prediction unit, the greenhouse heat usage prediction unit and the breeding heat usage prediction unit, and the predicted heat usage is output by summing up; if the predicted heat usage is greater than the predicted heating supply, auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation.

[0008] In the second aspect, the present application also provides a trough solar thermal collection and heating optimization device based on data analysis, which is used to execute a trough solar thermal collection and heating optimization method based on data analysis as described in the first aspect, including: a heating twin network construction module, which is used to read the basic operating data of the target heating system, perform twin simulation according to the basic operating data, and construct a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operating data include mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds and water storage tank capacity; an environmental prediction value acquisition module, which is used to obtain the environmental prediction value within a predetermined future window, wherein the environmental prediction value includes predicted radiation intensity, predicted sunshine duration and predicted temperature; a cluster analysis module, which is used to cluster the predicted radiation intensity and the predicted temperature, and determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals; a predicted heating amount acquisition module, which is used to obtain the environmental prediction value within a predetermined future window, wherein the environmental prediction value includes predicted radiation intensity, predicted sunshine duration and predicted temperature; The multiple predicted radiation intensity intervals are input into the heating twin network for simulation, and multiple interval predicted heating supplies are output. The multiple interval predicted heating supplies are mapped and compensated based on the multiple predicted temperature intervals, and the predicted heating supply is obtained by summing up; a heat usage predictor construction module is used to retrieve and obtain multiple sample heat usage data sets based on regional structural characteristics as constraints, and a heat usage predictor is obtained by training based on the multiple sample heat usage data sets, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit and a breeding heat usage prediction unit; a predicted heat usage output module is used to input the multiple predicted temperature intervals into the heating heat usage prediction unit, the living heat usage prediction unit, the greenhouse heat usage prediction unit and the breeding heat usage prediction unit respectively, and sum up and output the predicted heat usage; an auxiliary heating compensation module is used to perform auxiliary heating compensation on the target area within a predetermined future window according to the heat deviation if the predicted heat usage is greater than the predicted heating supply.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The target heating system is simulated and modeled to build a heating twin network. The environmental prediction values ​​within the predetermined future window are obtained, including the predicted radiation intensity, predicted sunshine duration and predicted temperature. The predicted radiation intensity and predicted temperature are then clustered to determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals. The multiple predicted radiation intensity intervals are then input into the heating twin network for simulation, and multiple interval predicted heat supplies are output. The multiple interval predicted heat supplies are mapped and compensated according to the multiple predicted temperature intervals, and the predicted heat supply is summed up to obtain the predicted heat supply. On the other hand, a heat consumption predictor is built based on a neural network, including a heating heat consumption predictor. The heat measurement unit, the domestic heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit are input into the multiple predicted temperature ranges respectively, and the predicted heat consumption is outputted by summing up; finally, if the predicted heat consumption is greater than the predicted heat supply, the heat deviation is calculated and determined, and auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation; the precision, accuracy and scientificity of heating forecast and heating demand forecast can be improved, thereby achieving the technical goal of accurately setting auxiliary heating supply and ensuring safe and stable heating in rural areas.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 A schematic diagram of a process flow of a trough solar thermal collection and heating optimization method based on data analysis for this application;

[0014] Figure 2 A schematic diagram of a process for determining multiple predicted radiation intensity intervals in a trough solar thermal collection and heating optimization method based on data analysis in this application;

[0015] Figure 3 This is a structural schematic diagram of a trough-type solar thermal collection and heating optimization device based on data analysis in this application.

[0016] Description of reference numerals:

[0017] A heating twin network construction module 11, an environmental prediction value acquisition module 12, a cluster analysis module 13, a predicted heating amount acquisition module 14, a heat predictor construction module 15, a predicted heat output module 16, and an auxiliary heating compensation module 17. DETAILED DESCRIPTION

[0018] This application provides a trough solar thermal heating optimization method and device based on data analysis, which solves the technical problem that the existing rural centralized heating method using trough solar energy has low precision and accuracy in solar heating and heating demand analysis, resulting in poor accuracy and comprehensiveness of heating forecasts and heating demand forecasts, and inability to accurately set auxiliary heating, resulting in poor stability and reliability of regional heating. The precision, accuracy and scientificity of heating forecasts and heating demand forecasts can be improved, and the technical goal of accurately setting auxiliary heating can be achieved, ensuring safe and stable heating in rural areas.

[0019] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0020] For example, please refer to the attached Figure 1 The present application provides a trough solar thermal collection and heating optimization method based on data analysis, which is applied to a trough solar thermal collection and heating optimization device based on data analysis, and specifically includes the following steps:

[0021] Step 1: Read the basic operating data of the target heating system, perform twin simulation based on the basic operating data, and build a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operating data includes mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds, and water storage tank capacity.

[0022] Specifically, first, the basic operating data of the target heating system is read, wherein the target heating system is a trough solar heating system, which is used to provide centralized heating in rural areas. The basic operating data include the mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds and the capacity of the hot water storage tank. The mirror field area of ​​the trough solar system is an important factor affecting the output power of the system; the concentration efficiency refers to the proportion of solar radiation focused on the collector; the conversion efficiency refers to the efficiency of the collector in converting solar radiation energy into thermal energy; pipeline heat loss refers to the heat loss caused by temperature differences during pipeline transmission; the inlet and outlet temperature thresholds refer to the temperature range of the collector inlet and outlet, which is used to ensure the normal operation of the system; the capacity of the hot water storage tank determines the system's ability to store and release heat.

[0023] Digital twin is a digital technology that simulates and predicts the behavior of a physical system by creating a virtual copy of it. It has many advantages such as authenticity and interoperability. Based on digital twin technology, twin simulation is performed according to the basic operating data to build a heating twin network. By building a heating twin network, the operation of the trough solar heating system can be simulated. The model can dynamically adjust parameters according to real-time data to reflect the actual operating status of the system, providing support for the prediction of solar heating capacity.

[0024] Step 2: Obtain environmental prediction values ​​within a predetermined future window, wherein the environmental prediction values ​​include predicted radiation intensity, predicted sunshine duration, and predicted temperature.

[0025] Specifically, a predetermined future window is configured, which is an expected time period and can be set according to actual conditions, such as the next week; then the environmental prediction value within the predetermined future window is obtained, which can be obtained through weather forecasts or climate models, wherein the environmental prediction value includes the predicted radiation intensity, predicted sunshine duration, and predicted temperature. By obtaining the environmental prediction value within the predetermined future window, a basis is provided for subsequent solar heating prediction and heating demand prediction.

[0026] Step three: clustering the predicted radiation intensity and the predicted temperature to determine a plurality of predicted radiation intensity intervals and a plurality of predicted temperature intervals.

[0027] Specifically, the predicted radiation intensity is a predicted radiation intensity sequence, including predicted radiation intensities at multiple time nodes; the predicted temperature is a predicted temperature sequence, including predicted temperatures at multiple time nodes; then the predicted radiation intensity and the predicted temperature are clustered, that is, the time nodes whose radiation intensity deviations and temperature deviations at adjacent time nodes are within a predetermined threshold are clustered into one category; multiple predicted radiation intensity intervals and multiple predicted temperature intervals are obtained, each interval contains multiple time nodes, represented by a time degree, the longer the time degree, the longer the time interval representing the interval; within the same interval, the predicted radiation intensity or predicted temperature can be regarded as the same.

[0028] Step 4: Input the multiple predicted radiation intensity intervals into the heating twin network for simulation, output multiple interval predicted heating amounts, map and compensate the multiple interval predicted heating amounts based on the multiple predicted temperature intervals, and sum them up to obtain the predicted heating amount.

[0029] Specifically, the middle value of the multiple predicted radiation intensity intervals is selected as the predicted radiation intensity mean value, multiple predicted radiation intensity means are obtained, and multiple time degrees of the multiple predicted radiation intensity intervals are obtained; then the multiple predicted radiation intensity means and the multiple time degrees are mapped and combined, and the multiple combination results are input into the heating twin network for simulation, and multiple interval predicted heating amounts are output. On the other hand, based on the multiple predicted temperature intervals, the heat loss analysis of the hot water storage tank is performed to determine multiple heat loss coefficients; further, the multiple interval predicted heating amounts are mapped and compensated according to the multiple heat loss coefficients to obtain multiple initial predicted heating amounts; then, the multiple initial predicted heating amounts are summed, and the summation result is used as the predicted heating amount.

[0030] Step 5: Taking the regional structure characteristics as constraints, retrieve and obtain multiple sample heat usage data sets, and train a heat usage predictor based on the multiple sample heat usage data sets, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit, and a breeding heat usage prediction unit.

[0031] Specifically, regional structural characteristics are obtained, and the regional structural characteristics include housing characteristics, living characteristics, greenhouse characteristics and breeding characteristics; further, with the regional structural characteristics as constraints, multiple sample heat usage data sets are obtained based on big data retrieval, including heating heat usage data set, living heat usage data set, greenhouse heat usage data set and breeding heat usage data set; then, the heating heat usage data set, the living heat usage data set, the greenhouse heat usage data set and the breeding heat usage data set are used as training data, and the BP neural network is supervised and trained respectively to obtain heating heat usage prediction unit, living heat usage prediction unit, greenhouse heat usage prediction unit and breeding heat usage prediction unit; based on the heating heat usage prediction unit, the living heat usage prediction unit, the greenhouse heat usage prediction unit and the breeding heat usage prediction unit, a heat usage predictor is built in parallel.

[0032] By constructing heating heat prediction unit, living heat prediction unit, greenhouse heat prediction unit and breeding heat prediction unit based on BP neural network, the sophistication and intelligence of rural regional heating forecast analysis can be improved, thereby improving the accuracy, comprehensiveness and efficiency of regional heating demand forecast.

[0033] Step six: input the multiple predicted temperature intervals into the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit respectively, and sum and output the predicted heat.

[0034] Specifically, the middle value of the multiple predicted temperature intervals is selected as the predicted temperature mean to obtain multiple predicted temperature means; multiple time degrees of the multiple predicted temperature intervals are obtained, and then the multiple predicted temperature means and the multiple time degrees are mapped and combined, and the multiple mapping combination results are respectively input into the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit to obtain the heating predicted heat, the living predicted heat, the greenhouse predicted heat and the breeding predicted heat; the heating predicted heat, the living predicted heat, the greenhouse predicted heat and the breeding predicted heat are further summed to obtain the predicted heat.

[0035] Step 7: If the predicted heat consumption is greater than the predicted heat supply, auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation.

[0036] Specifically, when the predicted heat demand is greater than the predicted heat supply, in order to ensure that the thermal comfort and heating needs of the target area are met, the predicted heat deviation needs to be compensated. First, the predicted heat demand is subtracted from the predicted heat supply to obtain the heat deviation; further, auxiliary heating compensation is performed on the target area within the predetermined future window based on the heat deviation. For example, according to the characteristics and needs of the target area, a suitable auxiliary heating system is selected, such as a gas boiler, electric heater, etc., and the operation scheduling of the auxiliary heating system is optimized based on real-time data and predicted data to ensure that additional heat is provided when needed.

[0037] The trough solar thermal collection and heating optimization method based on data analysis is applied to a trough solar thermal collection and heating optimization device based on data analysis, which can solve the technical problems that in the existing rural centralized heating method using trough solar energy, the precision and accuracy of the analysis of solar heating and heating demand are low, resulting in poor accuracy and comprehensiveness of heating prediction and heating demand prediction, and the inability to accurately set the auxiliary heating amount, resulting in poor stability and reliability of regional heating. By simulating and modeling the target heating system, a heating twin network is constructed; the environmental prediction values ​​within a predetermined future window are obtained, including predicted radiation intensity, predicted sunshine duration and predicted temperature, and then the predicted radiation intensity and predicted temperature are clustered to determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals; then the multiple predicted radiation intensity intervals are input into the heating twin network for simulation, and multiple interval predicted heating supplies are output, and the multiple interval predicted heating supplies are mapped and compensated according to the multiple predicted temperature intervals, and the predicted heating supply is obtained by summing up; on the other hand, a heat use predictor is constructed based on a neural network, including a heating heat use predictor. The heat measurement unit, the domestic heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit are input into the multiple predicted temperature ranges respectively, and the predicted heat consumption is outputted by summing up; finally, if the predicted heat consumption is greater than the predicted heat supply, the heat deviation is calculated and determined, and auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation; the precision, accuracy and scientificity of heating forecast and heating demand forecast can be improved, thereby achieving the technical goal of accurately setting auxiliary heating supply and ensuring safe and stable heating in rural areas.

[0038] Further, multiple predicted radiation intensity intervals are determined, such as Figure 2 As shown, step three of this application includes:

[0039] Obtain a predicted radiation intensity sequence within a predetermined future window, wherein the predicted radiation intensity sequence includes predicted radiation intensities at multiple time nodes; select a first predicted radiation intensity of a first node and a second predicted radiation intensity of a second node, wherein the first node is the first node of the predicted radiation intensity sequence, and the second node is the adjacent next node of the first node; perform deviation calculation on the first predicted radiation intensity and the second predicted radiation intensity to determine a second intensity deviation, and if the second intensity deviation is less than a predetermined intensity deviation threshold, merge the second node into the first node to construct a first interval; continue to select a third predicted radiation intensity of a third node, perform deviation calculation on the first predicted radiation intensity and the third predicted radiation intensity to determine a third intensity deviation, and if the third intensity deviation is less than a predetermined intensity deviation threshold, merge the third node into the first interval; if the third intensity deviation is greater than or equal to the predetermined intensity deviation threshold, set a segmentation line at the third node, and perform iterative analysis with the third node as the starting point to obtain multiple segmentation lines; cluster multiple time nodes according to the multiple segmentation lines to determine multiple predicted radiation intensity intervals, wherein the predicted radiation intensity intervals have time degree identifiers.

[0040] Specifically, first, obtain the predicted radiation intensity sequence within the predetermined future window, wherein the predicted radiation intensity sequence includes the predicted radiation intensity at multiple time nodes; then select the first node of the predicted radiation intensity sequence as the first node, select the next node adjacent to the first node in the predicted radiation intensity sequence as the second node, and obtain the first predicted radiation intensity of the first node and the second predicted radiation intensity of the second node. Then, perform deviation calculation on the first predicted radiation intensity and the second predicted radiation intensity to determine the second intensity deviation, wherein the second intensity deviation is the absolute value of the difference between the first predicted radiation intensity and the second predicted radiation intensity; configure a predetermined intensity deviation threshold, which can be set according to actual needs. If the second intensity deviation is less than the predetermined intensity deviation threshold, the second node is merged into the first node to construct the first interval.

[0041] Continue to select the next node adjacent to the second node in the predicted radiation intensity sequence as the third node, and obtain the third predicted radiation intensity of the third node; then calculate the deviation between the first predicted radiation intensity and the third predicted radiation intensity to determine the third intensity deviation; if the third intensity deviation is less than the predetermined intensity deviation threshold, the third node is incorporated into the first interval; if the third intensity deviation is greater than or equal to the predetermined intensity deviation threshold, a dividing line is set at the third node, and iterative analysis is performed with the third node as the starting point until all time nodes of the predicted radiation intensity sequence are traversed and multiple dividing lines are obtained. Finally, multiple time nodes are clustered according to the multiple dividing lines to determine multiple predicted radiation intensity intervals, wherein the predicted radiation intensity intervals have time degree identifiers.

[0042] The method of clustering the predicted temperatures to obtain multiple predicted temperature intervals is the same as the method of obtaining multiple predicted radiation intensity intervals, which is not described in detail in this application, and the predicted temperature intervals have time degree identifiers.

[0043] Further, based on the multiple predicted temperature intervals, the multiple interval predicted heating amounts are mapped and compensated, and step 4 of the present application includes:

[0044] Acquire the attribute characteristics of a heat storage tank, wherein the attribute characteristics of the heat storage tank include the material, size and structure of the water tank; retrieve and acquire a water tank sample data set with the capacity, material, size and structure of the heat storage tank as constraints, wherein the water tank sample data set includes a sample external temperature set and a sample heat loss coefficient set; construct a heat loss analysis channel based on the sample external temperature set and the sample heat loss coefficient set based on a random decision forest; select the middle value of the temperature of the multiple predicted temperature intervals as the predicted temperature mean to obtain multiple predicted temperature means; analyze the multiple predicted temperature means through the heat loss analysis channel to determine multiple heat loss coefficients; time-align the multiple heat loss coefficients with the multiple interval predicted heat supplies to determine multiple aligned heat loss coefficients, wherein the aligned heat loss coefficients correspond to the interval predicted heat supplies one by one, and if the interval predicted heat supplies include multiple mapped heat loss coefficients, the aligned heat loss coefficient is the mean of the multiple mapped heat loss coefficients; map and compensate the multiple interval predicted heat supplies according to the multiple aligned heat loss coefficients to obtain multiple interval predicted heat supplies.

[0045] Specifically, first, the attribute characteristics of the hot water storage tank are obtained, wherein the attribute characteristics of the hot water storage tank include the material, size and structure of the water tank; then, the capacity, material, size and structure of the hot water storage tank are used as retrieval constraints, and homologous information retrieval is performed based on big data technology to obtain a water tank sample data set, wherein the water tank sample data set includes a sample external temperature set and a sample heat loss coefficient set, and the sample external temperature and the sample heat loss coefficient correspond one to one, and the heat loss coefficient represents the degree of heat loss of the hot water storage tank at different external temperatures. The larger the heat loss coefficient, the higher the degree of heat loss.

[0046] Random decision forest is an integrated learning method, which improves the prediction accuracy and robustness of the model by constructing multiple decision trees and combining their prediction results; then, a heat loss analysis channel is constructed based on the random decision forest, and the heat loss analysis channel is trained using the sample external temperature set and the sample heat loss coefficient set. During the training process, the performance of the model can be evaluated by cross-validation and other techniques; according to the results of the model evaluation, the model parameters, such as the number of trees, the maximum depth, etc., are adjusted to obtain a heat loss analysis channel that meets the expected requirements.

[0047] Then, the middle temperature value of the multiple predicted temperature intervals is selected as the predicted temperature mean to obtain multiple predicted temperature means; and the multiple predicted temperature means are analyzed through the heat loss analysis channel to determine multiple heat loss coefficients; then the multiple heat loss coefficients are time-aligned with the multiple interval predicted heat supplies, and multiple aligned heat loss coefficients are determined based on the time alignment results, wherein the aligned heat loss coefficients correspond one-to-one to the interval predicted heat supply, and if the interval predicted heat supply contains multiple mapped heat loss coefficients, the aligned heat loss coefficient is the mean of the multiple mapped heat loss coefficients.

[0048] Finally, the multiple interval predicted heating supplies are mapped and compensated according to the multiple aligned heat loss coefficients, that is, the influence of the heat loss coefficient on the performance of the hot water storage tank is considered, and the predicted heating supply is compensated and adjusted. The purpose of the compensation adjustment is to ensure that the predicted heating supply more accurately reflects the actual performance of the hot water storage tank, and to obtain multiple interval predicted heating supplies.

[0049] Further, to obtain the predicted heating amount, step 4 of this application includes:

[0050] Query the historical heating records of the target heating system to obtain multiple historical heating prediction values ​​and multiple historical actual heating values; calculate multiple prediction errors based on the multiple historical heating prediction values ​​and multiple historical actual heating values, wherein the prediction error is the absolute value of the difference between the historical heating prediction value and the historical actual heating value; calculate the mean of the multiple prediction errors, perform difference operations on the multiple prediction errors according to the prediction error mean, and determine multiple error deviations; select the prediction errors corresponding to the error deviations less than a predetermined error threshold, obtain multiple standard prediction errors, and calculate the standard prediction error mean; sum the multiple interval predicted heating values ​​to obtain an initial predicted heating value, attenuate the initial predicted heating value according to the standard prediction error mean, and output the predicted heating value.

[0051] Specifically, query the historical heating records of the target heating system to obtain multiple historical heating prediction values ​​and multiple historical actual heating values; then calculate the prediction error based on the multiple historical heating prediction values ​​and multiple historical actual heating values ​​to obtain multiple prediction errors, wherein the prediction error is the absolute value of the difference between the historical heating prediction value and the historical actual heating value; further calculate the mean of the multiple prediction errors, perform difference calculation on the multiple prediction errors according to the prediction error mean, and determine multiple error deviations. Then, judge the multiple error deviations according to the predetermined error threshold, select the prediction error corresponding to the error deviation less than the predetermined error threshold, obtain multiple standard prediction errors, and calculate the standard prediction error mean; through the above operation, the interference of individual abnormal data can be reduced, and the accuracy and reliability of the standard prediction error mean setting can be improved.

[0052] Then, the predicted heat supply of the multiple intervals is summed to obtain the initial predicted heat supply; finally, the initial predicted heat supply is attenuated according to the standard prediction error mean to obtain the predicted heat supply, that is, the initial predicted heat supply is subtracted from the error ratio, for example, if the standard prediction error mean is 2% and the initial predicted heat supply is 1000, then the predicted heat supply is 1000×(100%-2%), that is, 980. By attenuating the initial predicted heat supply according to the standard prediction error mean, the accuracy and rationality of the predicted heat supply can be further improved to ensure the stability of regional heating.

[0053] Further, to construct a heat usage predictor, step five of this application includes:

[0054] Acquire regional structural features, wherein the regional structural features include housing features, living features, greenhouse features, and breeding features; take the housing features, living features, greenhouse features, and breeding features as constraints, perform positive sample retrieval based on big data, and determine a heating heat data set, a living heat data set, a greenhouse heat data set, and a breeding heat data set; perform supervised training and cross-validation on a BP neural network based on the heating heat data set, the living heat data set, the greenhouse heat data set, and the breeding heat data set, respectively, to obtain a heating heat prediction unit, a living heat prediction unit, a greenhouse heat prediction unit, and a breeding heat prediction unit; construct the heat prediction unit based on the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit, and the breeding heat prediction unit.

[0055] Specifically, the regional structural features are obtained, wherein the regional structural features include housing features, living features, greenhouse features and breeding features, such as housing type, housing area, housing structure, greenhouse area, greenhouse type, etc. Then, the housing features, living features, greenhouse features and breeding features are used as constraints, and positive sample retrieval is performed based on big data. Positive sample retrieval is a method for retrieving samples with certain specific attributes or characteristics in a specific application scenario, that is, the homologous data is retrieved to obtain the heating heat data set, the living heat data set, the greenhouse heat data set and the breeding heat data set. Then, the BP neural network is supervised and cross-validated based on the heating heat data set, the living heat data set, the greenhouse heat data set and the breeding heat data set, respectively, to obtain the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit. Finally, the heat prediction unit is constructed based on the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit.

[0056] Furthermore, the present application also includes the following steps:

[0057] Acquire house structure characteristics, family population, crop type characteristics and livestock species characteristics, wherein the house structure characteristics at least include house type and house area, the crop type characteristics at least include crop type and growth cycle, and the livestock species characteristics at least include animal type and growth stage; subdivide the heating heat data set, the domestic heat data set, the greenhouse heat data set and the livestock heat data set according to the house structure characteristics, the family population, the crop type characteristics and the livestock species characteristics, and determine multiple heating heat data subsets, multiple domestic heat data subsets, multiple greenhouse heat data subsets and multiple livestock heat data subsets; use the multiple heating heat data subsets, multiple domestic heat data subsets, multiple greenhouse heat data subsets and multiple livestock heat data subsets to perform supervised training and cross-validation on the BP neural network to generate the heating heat prediction unit, the domestic heat prediction unit, the greenhouse heat prediction unit and the livestock heat prediction unit.

[0058] Specifically, the house structure characteristics, family population size, crop type characteristics and breeding type characteristics are obtained, among which the house structure characteristics at least include the house type and house area, the crop type characteristics at least include the crop type and growth cycle, and the breeding type characteristics at least include the animal type and growth stage. For example, the house type is used to distinguish different types of residences, such as bungalows, buildings, villas, etc.; the family population size is used to record the number of people in each family, including adults and children; the crop type is used to record the types of crops planted, such as vegetables, fruits, etc.; the growth cycle is used to record the growth cycle of crops, such as sowing period, growing period, harvesting period, etc.; the animal type is used to record the types of animals raised, such as poultry, livestock, etc.; the growth stage is used to record the growth stage of animals, such as cub period, growth period, maturity period, etc.

[0059] Then, the heating heat data set, the living heat data set, the greenhouse heat data set, and the breeding heat data set are subdivided according to the house structure characteristics, the number of family members, the crop type characteristics, and the breeding type characteristics. For example, the heating heat data with the same house type and the house area within the same threshold are grouped into one category, and multiple heating heat data subsets, multiple living heat data subsets, multiple greenhouse heat data subsets, and multiple breeding heat data subsets are determined. Finally, the BP neural network is supervised and cross-validated using the multiple heating heat data subsets, multiple living heat data subsets, multiple greenhouse heat data subsets, and multiple breeding heat data subsets to generate the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit, and the breeding heat prediction unit.

[0060] Furthermore, the present application also includes the following steps:

[0061] Acquire a first house structure feature, wherein the first house structure feature includes a first house type and a first house area threshold; acquire a first heating heat data subset of the first house structure feature, wherein the first heating heat data subset includes a first sample temperature set and a first sample heat consumption set; divide the first heating heat data subset into N equal parts, and select with replacement N times to obtain a first training set, iterate and select N times to generate N training sets; take the first sample temperature as input, take the first sample heat consumption as supervision, use N training sets to perform supervised training and cross-validation on the BP neural network, obtain N first heating heat prediction branches that meet convergence, and construct a first heating heat prediction unit based on the N first heating heat prediction branches, wherein the output of the first heating heat prediction unit is the mode of the output results of the N first heating heat prediction branches; construct the heating heat prediction unit based on a decision tree according to a mapping relationship between the first house structure feature and the first heating heat prediction unit.

[0062] Specifically, this application introduces a method for constructing a heating heat prediction unit, wherein the construction methods of a living heat prediction unit, a greenhouse heat prediction unit and a breeding heat prediction unit are consistent with the construction method of the heating heat prediction unit and are not expanded in this application.

[0063] First, obtain the first house structure feature, wherein the first house structure feature includes the first house type and the first house area threshold; then obtain the first heating heat data subset of the first house structure feature, wherein the first heating heat data subset includes the first sample temperature set and the first sample heat set; further divide the first heating heat data subset into N parts, and select N times with replacement from the N data sets to obtain the first training set, and iterate and select N times to generate N training sets. Then, take the first sample temperature as input, take the first sample heat as supervision, and use the N training sets to supervise the BP neural network. The supervised training can be performed by the back propagation algorithm. After meeting the predetermined number of training times; exchange the N training sets, and after the exchange, verify the BP neural network with the N training sets to obtain N first heating heat prediction branches that meet the expected convergence conditions; finally, construct the first heating heat prediction unit based on the N first heating heat prediction branches, wherein the output of the first heating heat prediction unit is the mode of the output results of the N first heating heat prediction branches. By constructing the first heating heat prediction unit based on N first heating heat prediction branches, the prediction error of a single model can be reduced and the accuracy and reliability of the overall prediction can be improved.

[0064] Using the same method of constructing the first heating heat prediction unit, multiple heating heat prediction units of multiple house structure characteristics are obtained in turn; further based on the decision tree principle, the heating heat prediction unit is constructed with the house structure characteristics as the child node and the heating heat prediction unit as the leaf node of the child node.

[0065] Further, auxiliary heating compensation is performed, and step seven of this application includes:

[0066] With the purpose of meeting the expected heat storage performance, the soil is screened and pretreated to determine the suitable heat storage soil; based on the expected heat storage demand, a greenhouse heat release unit is built according to the predetermined heat release equipment and the suitable heat storage soil; using the target heating system, the greenhouse heat release unit stores heat during the day, and slowly releases heat from the greenhouse through the greenhouse heat release unit at night.

[0067] Specifically, first of all, the soil is screened and pretreated for the purpose of meeting the expected heat storage performance. For example, the appropriate soil type, such as sandy loam, clay, etc., is selected to ensure that the soil has good heat storage performance. At the same time, the soil is treated, such as removing impurities, adjusting the pH value, etc., to improve the heat storage effect of the soil. Then, based on the expected heat storage demand, the heat storage capacity, temperature range and thermal conductivity of the soil are considered to select the appropriate size and shape. On the other hand, suitable heat release equipment is selected, such as floor heating pipes, radiators, etc., to achieve heat release to the greenhouse. At the same time, the heat release equipment should have good thermal conductivity and corrosion resistance to adapt to the soil environment. The heat release equipment is further buried in the soil heat storage body to ensure that the equipment is in full contact with the soil to achieve effective heat transfer and build a greenhouse heat release unit.

[0068] Finally, the target heating system is used to store heat in the greenhouse heat release unit during the day, that is, the target heating system collects solar radiation energy during the day and converts it into thermal energy, which is transferred to the greenhouse heat release unit through the heat exchange system. The greenhouse heat release unit stores the collected heat in the soil. As a huge heat capacity, the soil can absorb and store a large amount of heat. At night or when solar radiation is insufficient, the soil heat storage body begins to slowly release the stored heat, that is, the heat release unit (such as floor heating pipes or radiators) absorbs heat from the soil and transfers it to the inside of the greenhouse.

[0069] Through the above steps, the heat can be stored in the greenhouse heat release unit during the day, and the heat can be slowly released from the greenhouse at night. This method can effectively reduce energy waste, improve the nighttime heating effect of the greenhouse, and promote the growth and yield of crops. At the same time, it can be adjusted and optimized according to actual needs and heating effects to meet the heating needs in different scenarios.

[0070] In summary, the trough solar thermal collection and heating optimization method based on data analysis provided in this application has the following technical effects:

[0071] 1. Construct a heating twin network by simulating and modeling the target heating system; obtain the environmental prediction values ​​within the predetermined future window, including the predicted radiation intensity, predicted sunshine duration and predicted temperature, and then cluster the predicted radiation intensity and predicted temperature to determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals; then input the multiple predicted radiation intensity intervals into the heating twin network for simulation, output multiple interval predicted heating values, and map and compensate the multiple interval predicted heating values ​​according to the multiple predicted temperature intervals, and sum them up to obtain the predicted heating value; on the other hand, construct a heat consumption predictor based on a neural network, including heating heat consumption prediction unit, domestic heat prediction unit, greenhouse heat prediction unit and aquaculture heat prediction unit; then input multiple predicted temperature ranges into the heating heat prediction unit, domestic heat prediction unit, greenhouse heat prediction unit and aquaculture heat prediction unit respectively, and sum up and output the predicted heat consumption; finally, if the predicted heat consumption is greater than the predicted heating supply, the heat deviation is calculated and determined, and auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation; the precision, accuracy and scientificity of heating forecasts and heating demand forecasts can be improved, thereby achieving the technical goal of accurately setting auxiliary heating supply, and ensuring safe and stable heating in rural areas.

[0072] 2. By constructing heating heat prediction unit, living heat prediction unit, greenhouse heat prediction unit and breeding heat prediction unit based on BP neural network, the precision and intelligence of rural regional heating forecast analysis can be improved, thereby improving the accuracy, comprehensiveness and efficiency of regional heating demand forecast.

[0073] 3. By constructing a greenhouse heat release unit, it is possible to store heat in the greenhouse heat release unit during the day and slowly release heat from the greenhouse through the greenhouse heat release unit at night, thereby effectively reducing energy waste, improving the nighttime heating effect of the greenhouse, promoting crop growth and yield, and at the same time improving the flexibility and reliability of greenhouse heating at night.

[0074] Embodiment 2, based on the same inventive concept as the trough type solar energy heat collection and heating optimization method based on data analysis in the above embodiment, this application also provides a trough type solar energy heat collection and heating optimization device based on data analysis, please refer to the attached Figure 3 ,include:

[0075] The heating twin network construction module 11 is used to read the basic operating data of the target heating system, perform twin simulation based on the basic operating data, and construct a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operating data include mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds, and water storage tank capacity; the environmental prediction value acquisition module 12 is used to obtain the environmental prediction value within a predetermined future window, wherein the environmental prediction value includes predicted radiation intensity, predicted sunshine duration, and predicted temperature; the clustering analysis module 13 is used to cluster the predicted radiation intensity and the predicted temperature to determine multiple predicted radiation intensity intervals and multiple predicted temperature intervals; the predicted heat supply acquisition module 14 is used to input the multiple predicted radiation intensity intervals into the heating twin network for simulation, and output multiple interval predicted Measure the heat supply, map and compensate the predicted heat supply of the multiple intervals based on the multiple predicted temperature intervals, and sum them up to obtain the predicted heat supply; a heat usage predictor construction module 15, used to retrieve and obtain multiple sample heat usage data sets based on regional structural characteristics as constraints, and obtain a heat usage predictor based on the multiple sample heat usage data sets for training, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit and a breeding heat usage prediction unit; a predicted heat usage output module 16, used to input the multiple predicted temperature intervals into the heating heat usage prediction unit, the living heat usage prediction unit, the greenhouse heat usage prediction unit and the breeding heat usage prediction unit respectively, and sum up and output the predicted heat usage; an auxiliary heating compensation module 17, used to perform auxiliary heating compensation on the target area within a predetermined future window according to the heat deviation if the predicted heat usage is greater than the predicted heat supply.

[0076] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0077] Obtain a predicted radiation intensity sequence within a predetermined future window, wherein the predicted radiation intensity sequence includes predicted radiation intensities at multiple time nodes; select a first predicted radiation intensity of a first node and a second predicted radiation intensity of a second node, wherein the first node is the first node of the predicted radiation intensity sequence, and the second node is the adjacent next node of the first node; perform deviation calculation on the first predicted radiation intensity and the second predicted radiation intensity to determine a second intensity deviation, and if the second intensity deviation is less than a predetermined intensity deviation threshold, merge the second node into the first node to construct a first interval; continue to select a third predicted radiation intensity of a third node, perform deviation calculation on the first predicted radiation intensity and the third predicted radiation intensity to determine a third intensity deviation, and if the third intensity deviation is less than a predetermined intensity deviation threshold, merge the third node into the first interval; if the third intensity deviation is greater than or equal to the predetermined intensity deviation threshold, set a segmentation line at the third node, and perform iterative analysis with the third node as the starting point to obtain multiple segmentation lines; cluster multiple time nodes according to the multiple segmentation lines to determine multiple predicted radiation intensity intervals, wherein the predicted radiation intensity intervals have time degree identifiers.

[0078] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0079] Acquire the attribute characteristics of a heat storage tank, wherein the attribute characteristics of the heat storage tank include the material, size and structure of the water tank; retrieve and acquire a water tank sample data set with the capacity, material, size and structure of the heat storage tank as constraints, wherein the water tank sample data set includes a sample external temperature set and a sample heat loss coefficient set; construct a heat loss analysis channel based on the sample external temperature set and the sample heat loss coefficient set based on a random decision forest; select the middle value of the temperature of the multiple predicted temperature intervals as the predicted temperature mean to obtain multiple predicted temperature means; analyze the multiple predicted temperature means through the heat loss analysis channel to determine multiple heat loss coefficients; time-align the multiple heat loss coefficients with the multiple interval predicted heat supplies to determine multiple aligned heat loss coefficients, wherein the aligned heat loss coefficients correspond to the interval predicted heat supplies one by one, and if the interval predicted heat supplies include multiple mapped heat loss coefficients, the aligned heat loss coefficient is the mean of the multiple mapped heat loss coefficients; map and compensate the multiple interval predicted heat supplies according to the multiple aligned heat loss coefficients to obtain multiple interval predicted heat supplies.

[0080] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0081] Query the historical heating records of the target heating system to obtain multiple historical heating prediction values ​​and multiple historical actual heating values; calculate multiple prediction errors based on the multiple historical heating prediction values ​​and multiple historical actual heating values, wherein the prediction error is the absolute value of the difference between the historical heating prediction value and the historical actual heating value; calculate the mean of the multiple prediction errors, perform difference operations on the multiple prediction errors according to the prediction error mean, and determine multiple error deviations; select the prediction errors corresponding to the error deviations less than a predetermined error threshold, obtain multiple standard prediction errors, and calculate the standard prediction error mean; sum the multiple interval predicted heating values ​​to obtain an initial predicted heating value, attenuate the initial predicted heating value according to the standard prediction error mean, and output the predicted heating value.

[0082] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0083] Acquire regional structural features, wherein the regional structural features include housing features, living features, greenhouse features, and breeding features; take the housing features, living features, greenhouse features, and breeding features as constraints, perform positive sample retrieval based on big data, and determine a heating heat data set, a living heat data set, a greenhouse heat data set, and a breeding heat data set; perform supervised training and cross-validation on a BP neural network based on the heating heat data set, the living heat data set, the greenhouse heat data set, and the breeding heat data set, respectively, to obtain a heating heat prediction unit, a living heat prediction unit, a greenhouse heat prediction unit, and a breeding heat prediction unit; construct the heat prediction unit based on the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit, and the breeding heat prediction unit.

[0084] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0085] Acquire house structure characteristics, family population, crop type characteristics and livestock species characteristics, wherein the house structure characteristics at least include house type and house area, the crop type characteristics at least include crop type and growth cycle, and the livestock species characteristics at least include animal type and growth stage; subdivide the heating heat data set, the domestic heat data set, the greenhouse heat data set and the livestock heat data set according to the house structure characteristics, the family population, the crop type characteristics and the livestock species characteristics, and determine multiple heating heat data subsets, multiple domestic heat data subsets, multiple greenhouse heat data subsets and multiple livestock heat data subsets; use the multiple heating heat data subsets, multiple domestic heat data subsets, multiple greenhouse heat data subsets and multiple livestock heat data subsets to perform supervised training and cross-validation on the BP neural network to generate the heating heat prediction unit, the domestic heat prediction unit, the greenhouse heat prediction unit and the livestock heat prediction unit.

[0086] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0087] Acquire a first house structure feature, wherein the first house structure feature includes a first house type and a first house area threshold; acquire a first heating heat data subset of the first house structure feature, wherein the first heating heat data subset includes a first sample temperature set and a first sample heat consumption set; divide the first heating heat data subset into N equal parts, and select with replacement N times to obtain a first training set, iterate and select N times to generate N training sets; take the first sample temperature as input, take the first sample heat consumption as supervision, use N training sets to perform supervised training and cross-validation on the BP neural network, obtain N first heating heat prediction branches that meet convergence, and construct a first heating heat prediction unit based on the N first heating heat prediction branches, wherein the output of the first heating heat prediction unit is the mode of the output results of the N first heating heat prediction branches; construct the heating heat prediction unit based on a decision tree according to a mapping relationship between the first house structure feature and the first heating heat prediction unit.

[0088] Furthermore, the trough type solar thermal collection and heating optimization device based on data analysis is also used for:

[0089] With the purpose of meeting the expected heat storage performance, the soil is screened and pretreated to determine the suitable heat storage soil; based on the expected heat storage demand, a greenhouse heat release unit is built according to the predetermined heat release equipment and the suitable heat storage soil; using the target heating system, the greenhouse heat release unit stores heat during the day, and slowly releases heat from the greenhouse through the greenhouse heat release unit at night.

[0090] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The trough solar thermal collection and heating optimization method based on data analysis and the specific examples in the aforementioned embodiment 1 are also applicable to the trough solar thermal collection and heating optimization device based on data analysis in this embodiment. Through the aforementioned detailed description of the trough solar thermal collection and heating optimization method based on data analysis, those skilled in the art can clearly know the trough solar thermal collection and heating optimization device based on data analysis in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method description.

[0091] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0092] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for optimizing the heating of trough solar thermal collectors based on data analysis, characterized in that: include: Read the basic operation data of the target heating system, perform twin simulation based on the basic operation data, and build a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operation data includes the mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds, and heat storage tank capacity; Obtaining environmental prediction values ​​within a predetermined future window, wherein the environmental prediction values ​​include predicted radiation intensity, predicted sunshine duration, and predicted temperature; Clustering the predicted radiation intensity and the predicted temperature to determine a plurality of predicted radiation intensity intervals and a plurality of predicted temperature intervals; Input the multiple predicted radiation intensity intervals into the heating twin network for simulation, output multiple interval predicted heating supplies, perform mapping compensation on the multiple interval predicted heating supplies based on the multiple predicted temperature intervals, and sum them up to obtain the predicted heating supply; Taking the regional structure features as constraints, multiple sample heat usage data sets are retrieved and obtained, and a heat usage predictor is trained based on the multiple sample heat usage data sets, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit, and a breeding heat usage prediction unit; The plurality of predicted temperature intervals are respectively input into the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit, and the predicted heat is summed and output; If the predicted heat consumption is greater than the predicted heat supply, auxiliary heating compensation is performed on the target area within the predetermined future window according to the heat deviation.

2. The trough solar thermal collection and heating optimization method based on data analysis according to claim 1 is characterized in that: Determine multiple predicted radiation intensity intervals, including: Obtaining a predicted radiation intensity sequence within a predetermined future window, wherein the predicted radiation intensity sequence includes predicted radiation intensities at multiple time nodes; Selecting a first predicted radiation intensity of a first node and a second predicted radiation intensity of a second node, wherein the first node is a leading node in a predicted radiation intensity sequence, and the second node is an adjacent next node of the first node; Performing deviation calculation on the first predicted radiation intensity and the second predicted radiation intensity to determine a second intensity deviation, and if the second intensity deviation is less than a predetermined intensity deviation threshold, merging the second node into the first node to construct a first interval; Continue to select the third predicted radiation intensity of the third node, perform deviation calculation on the first predicted radiation intensity and the third predicted radiation intensity, determine a third intensity deviation, and if the third intensity deviation is less than a predetermined intensity deviation threshold, merge the third node into the first interval; If the third intensity deviation is greater than or equal to a predetermined intensity deviation threshold, a segmentation line is set at the third node, and an iterative analysis is performed with the third node as a starting point to obtain multiple segmentation lines; The multiple time nodes are clustered and divided according to the multiple segmentation lines to determine a plurality of predicted radiation intensity intervals, wherein the predicted radiation intensity intervals have time degree identifiers.

3. The trough solar thermal collection and heating optimization method based on data analysis according to claim 1 is characterized in that: Mapping and compensating the multiple predicted temperature intervals for the multiple predicted heating supplies includes: Acquire the attribute characteristics of the hot water storage tank, wherein the attribute characteristics of the hot water storage tank include the material, size and structure of the tank; Taking the capacity, material, size and structure of the water storage tank as constraints, a water tank sample data set is retrieved and obtained, wherein the water tank sample data set includes a sample external temperature set and a sample heat loss coefficient set; Based on the random decision forest, a heat loss analysis channel is constructed according to the sample external temperature set and the sample heat loss coefficient set; Selecting the middle values ​​of the temperatures in the plurality of predicted temperature intervals as the predicted temperature averages to obtain a plurality of predicted temperature averages; Analyzing the plurality of predicted temperature means through the heat loss analysis channel to determine a plurality of heat loss coefficients; Time-align the multiple heat loss coefficients with the multiple interval predicted heat supplies to determine multiple aligned heat loss coefficients, wherein the aligned heat loss coefficients correspond to the interval predicted heat supplies one-to-one, and if the interval predicted heat supplies include multiple mapped heat loss coefficients, the aligned heat loss coefficient is the average of the multiple mapped heat loss coefficients; The multiple interval predicted heating supplies are mapped and compensated according to the multiple aligned heat loss coefficients to obtain multiple interval predicted heating supplies.

4. The trough solar thermal collection and heating optimization method based on data analysis according to claim 3 is characterized in that: Get forecasted heating supply, including: Query the historical heating records of the target heating system to obtain multiple historical heating forecast values ​​and multiple historical actual heating values; Calculating a plurality of prediction errors based on the plurality of historical heat supply prediction values ​​and the plurality of historical actual heat supply values, wherein the prediction error is an absolute value of a difference between a historical heat supply prediction value and a historical actual heat supply value; Calculating a mean of the multiple prediction errors, performing a difference operation on the multiple prediction errors according to the prediction error mean, and determining a plurality of error deviations; Select the prediction error corresponding to the error deviation less than the predetermined error threshold, obtain multiple standard prediction errors, and calculate the mean of the standard prediction errors; The predicted heat supplies of the multiple intervals are summed to obtain an initial predicted heat supply, the initial predicted heat supply is attenuated according to the standard prediction error mean, and the predicted heat supply is output.

5. The trough solar thermal collection and heating optimization method based on data analysis according to claim 1 is characterized in that: Build a heat predictor, including: Acquire regional structural characteristics, where the regional structural characteristics include housing characteristics, living characteristics, greenhouse characteristics, and breeding characteristics; Taking the house characteristics, life characteristics, greenhouse characteristics and breeding characteristics as constraints, positive sample retrieval is performed based on big data to determine a heating heat data set, a life heat data set, a greenhouse heat data set and a breeding heat data set; Based on the heating heat data set, the living heat data set, the greenhouse heat data set and the aquaculture heat data set, supervised training and cross-validation are performed on the BP neural network to obtain a heating heat prediction unit, a living heat prediction unit, a greenhouse heat prediction unit and aquaculture heat prediction unit; The heat usage predictor is constructed based on the heating heat usage prediction unit, the living heat usage prediction unit, the greenhouse heat usage prediction unit and the breeding heat usage prediction unit.

6. The method for optimizing the temperature rise of trough solar thermal collection based on data analysis according to claim 5, characterized in that: Obtain the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit, including: Acquire house structure characteristics, household population, crop type characteristics, and livestock type characteristics, wherein house structure characteristics at least include house type and house area, crop type characteristics at least include crop type and growth cycle, and livestock type characteristics at least include animal type and growth stage; According to the house structure characteristics, the number of family members, the crop type characteristics and the breeding type characteristics, the heating heat data set, the living heat data set, the greenhouse heat data set and the breeding heat data set are respectively subdivided to determine a plurality of heating heat data subsets, a plurality of living heat data subsets, a plurality of greenhouse heat data subsets and a plurality of breeding heat data subsets; The BP neural network is supervisedly trained and cross-validated using the multiple heating heat data subsets, multiple daily heat data subsets, multiple greenhouse heat data subsets and multiple aquaculture heat data subsets to generate the heating heat prediction unit, daily heat prediction unit, greenhouse heat prediction unit and aquaculture heat prediction unit.

7. The trough solar thermal collection and heating optimization method based on data analysis according to claim 6 is characterized in that: Construct a heating heat prediction unit, including: Acquire a first house structure feature, wherein the first house structure feature includes a first house type and a first house area threshold; Acquire a first heating heat usage data subset of the first house structure feature, wherein the first heating heat usage data subset includes a first sample temperature set and a first sample heat usage set; Divide the first heating heat data subset into N equal parts, select it N times with replacement to obtain a first training set, and iterate the selection N times to generate N training sets; Taking the first sample temperature as input and the first sample heat consumption as supervision, the BP neural network is supervised and cross-validated using N training sets to obtain N first heating heat consumption prediction branches that meet convergence, and a first heating heat consumption prediction unit is constructed based on the N first heating heat consumption prediction branches, wherein the output of the first heating heat consumption prediction unit is the mode of the output results of the N first heating heat consumption prediction branches; Based on the decision tree, the heating heat prediction unit is constructed according to the mapping relationship between the first house structure feature and the first heating heat prediction unit.

8. The method for optimizing the temperature rise of trough solar thermal collection based on data analysis according to claim 1, characterized in that: Perform auxiliary heating compensation, including: To meet the expected heat storage performance, the soil is screened and pre-treated to determine the soil suitable for heat storage; Based on the expected heat storage demand, a greenhouse heat release unit is built according to the predetermined heat release equipment and the adapted heat storage soil; By utilizing the target heating system, the heat of the greenhouse heat release unit is stored during the day, and the heat of the greenhouse is slowly released through the greenhouse heat release unit at night.

9. A trough type solar thermal collection and heating optimization device based on data analysis, characterized in that: The steps for implementing the trough solar thermal collection and heating optimization method based on data analysis as described in any one of claims 1 to 8 include: A heating twin network construction module is used to read the basic operating data of the target heating system, perform twin simulation based on the basic operating data, and construct a heating twin network, wherein the target heating system is a trough solar heating system, and the basic operating data include mirror field area, concentration efficiency, conversion efficiency, pipeline heat loss, inlet and outlet temperature thresholds, and heat storage tank capacity; An environmental prediction value acquisition module is used to acquire environmental prediction values ​​within a predetermined future window, wherein the environmental prediction values ​​include predicted radiation intensity, predicted sunshine duration, and predicted temperature; A cluster analysis module, used for clustering the predicted radiation intensity and the predicted temperature to determine a plurality of predicted radiation intensity intervals and a plurality of predicted temperature intervals; A predicted heat supply obtaining module is used to input the multiple predicted radiation intensity intervals into the heat supply twin network for simulation, output multiple interval predicted heat supplies, perform mapping compensation on the multiple interval predicted heat supplies based on the multiple predicted temperature intervals, and obtain the predicted heat supply by summing them up; A heat usage predictor construction module is used to retrieve and obtain multiple sample heat usage data sets based on regional structural features as constraints, and to train a heat usage predictor based on the multiple sample heat usage data sets, wherein the heat usage predictor includes a heating heat usage prediction unit, a living heat usage prediction unit, a greenhouse heat usage prediction unit, and a breeding heat usage prediction unit; A predicted heat output module, used to input the multiple predicted temperature intervals into the heating heat prediction unit, the living heat prediction unit, the greenhouse heat prediction unit and the breeding heat prediction unit respectively, and sum and output the predicted heat; The auxiliary heating compensation module is used to perform auxiliary heating compensation on the target area within the predetermined future window according to the heat deviation if the predicted heat consumption is greater than the predicted heat supply.

Images