Heat supply system secondary network regulation and control method considering heat consumer classification and building thermal inertia
By conducting multi-level classification and thermal load prediction on the thermal user building, combined with building thermal inertia analysis, the precise regulation of the secondary network of the heating system is achieved, and the problems of intricate classification and inaccurate regulation in the existing technology are solved, achieving the effect of energy conservation and emission reduction.
Patent Information
- Application Number
- CN202411605723.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-12
- Publication Date
- 2025-02-25
AI Technical Summary
The prior art is difficult to classify the end-of-level thermal users of the secondary network in a detailed manner, and it has failed to effectively utilize building thermal inertia to achieve accurate on-demand regulation of the secondary network.
The composite heat transfer coefficients of each thermal user building are calculated by fitting, and the first classification is performed using a clustering algorithm, and then the second classification is performed based on the personalized heat coefficient, indoor temperature and return water temperature. Combined with thermal load prediction and building thermal inertia analysis, the front-home valve control strategy is determined to achieve accurate control in time-divided periods.
It improves the accuracy of thermal user classification, enhances the accuracy of thermal load prediction, and realizes accurate on-demand regulation of the secondary network to achieve the purpose of energy conservation and emission reduction.
Smart Images
Figure CN119511982A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of intelligent heating systems, and specifically relates to a secondary network control method for a heating system that takes into account the classification of heat users and thermal inertia of buildings. Background Art
[0002] The primary network of the centralized heating system is responsible for delivering high-temperature hot water or steam from the heat source to the heat exchange station, while the secondary network is responsible for distributing heat from the heat exchange station to each heat user to meet the heating needs of the user end. The heat users at the end of the secondary network include residential buildings, commercial buildings, industrial users, etc. Different types of heat users have different heating time periods and heating needs. Moreover, the same type of heat users may have different heating needs due to factors such as heating behavior, building location, orientation, and structural materials. Therefore, it is necessary to finely classify each heat user.
[0003] The building envelopes of different types of heat users have thermal inertia, which is reflected in the fact that the envelopes have a certain ability to store and release heat and have attenuation and hysteresis characteristics in response to external changes. How to finely classify the heat users at the end of the secondary network and effectively use the thermal inertia of buildings to achieve precise regulation of the secondary network on demand is an urgent problem that needs to be solved.
[0004] Based on the above technical problems, it is necessary to design a new secondary network control method for the heating system that takes into account the classification of heat users and the thermal inertia of buildings. Summary of the invention
[0005] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a secondary network control method for a heating system taking into account the classification of heat users and the thermal inertia of buildings.
[0006] In order to solve the above technical problems, the technical solution of the present invention is:
[0007] The present invention provides a secondary network control method for a heating system taking into account the classification of heat users and the thermal inertia of buildings, which comprises:
[0008] S1. Considering the basic attributes of each heat user building and the different heating state parameters of each heat user, the composite heat transfer coefficient of each heat user building is calculated by fitting, and the clustering algorithm is used to perform the first classification of each heat user type based on the composite heat transfer coefficient, and then the second classification of each heat user type is performed based on the personalized heat coefficient, indoor temperature and supply and return water temperature level of each heat user;
[0009] S2. Select characteristic variables for load prediction for each type of heat user after the second classification, including outdoor environmental parameters, heat user heating status parameters, heat user indoor temperature, heat user personalized heat coefficient and heat load, and calculate the influence weight of each characteristic variable on the heat load and establish a heat load prediction model for each type of heat user in different time periods;
[0010] S3. According to the heat load prediction results of each type of heat user in different time periods, adjust the valves in front of each type of heat user, simulate the heat storage and release process of the building of each type of heat user, and conduct a comprehensive analysis of the thermal inertia of the building of each type of heat user in combination with the heat transfer process of the building of each type of heat user, the thermal inertia index and the attenuation and hysteresis characteristics of the response to external changes;
[0011] S4. Determine the indoor temperature setting value for each type of heat user in different time periods, and combine the heat load prediction results for each type of heat user in different time periods and the comprehensive analysis results of the thermal inertia of the building for each type of heat user to determine the valve control time for each type of heat user and the heat storage and release mechanism of the building in each time period, establish a valve control model for each type of heat user, and output the valve opening control strategy for each type of heat user in different time periods;
[0012] S5. After household regulation is performed according to the pre-household valve regulation strategy for each type of heat user in different time periods, multi-dimensional evaluation indicators are selected to evaluate the effect of household regulation.
[0013] Furthermore, in S1, fitting and calculating the composite heat transfer coefficient of each heat user building includes:
[0014] The heat balance model of each heat user building is established and expressed as:
[0015]
[0016] K is the heat transfer coefficient of the radiator; F is the heat dissipation area of the radiator; K′ is the heat transfer coefficient of the external wall; F′ is the area of the external wall; t g Water supply temperature for heat users; t h is the return water temperature of the heat user; t n is the indoor temperature of the heat user; t w is the outdoor temperature;
[0017] The composite heat transfer coefficient of each heat user building is calculated by fitting, which is expressed as:
[0018]
[0019] Furthermore, in S1, when a clustering algorithm is used to perform a first classification on each heat user type according to the composite heat transfer coefficient, the clustering algorithm used is an agglomerative hierarchical clustering algorithm, which clusters each heat user and divides similar heat users into the same cluster to form a clustering tree, including:
[0020] Obtain the basic attributes of each heat user building and the heating status parameters of the heat user, including the building area, structure type, building properties, heat transfer coefficient of the radiator, heat dissipation area of the radiator, area of the enclosure structure, heat transfer coefficient of the enclosure structure, supply and return water temperature of the heat user, circulation flow, supply and return water pressure, indoor temperature and outdoor temperature;
[0021] The basic attributes of each heat user's building, the heat user's heating state parameters and the composite heat transfer coefficient of each heat user are used as the heat user clustering sample data;
[0022] Choose a similarity measure, including Euclidean distance, Manhattan distance, Chebyshev distance, and cosine similarity;
[0023] Each hot user is initially treated as an independent cluster to construct the initial clusters;
[0024] Using the selected similarity measure method, calculate the similarity between each two clusters;
[0025] Find the two clusters with the smallest similarity among all the current cluster pairs, merge the two clusters into a new cluster, remove the original two clusters from the cluster list, add the merged cluster, and recalculate the similarity between the merged cluster and other clusters;
[0026] The most similar clusters are gradually merged until the predetermined termination condition is reached, and a clustering tree is constructed based on the information of each merged cluster;
[0027] The optimal number of clusters is determined based on the clustering tree, and the characteristic differences between different clusters and the distribution law of the composite heat transfer coefficient of each heat user building are obtained.
[0028] Furthermore, in S1, the second classification of each heat user type is performed according to the personalized heat coefficient, indoor temperature and supply and return water temperature level of each heat user, including:
[0029] According to the heat usage behavior habits, heat usage time distribution, and heat energy consumption of each heat user, the personalized heat usage coefficient of each heat user is fitted, which is expressed as:
[0030] PUI=f(H duration ,T profile ,E history ,C behavior );
[0031] H duration T is the heating time preference of heat users, indicating the heating time in a day or a specific time period; profile is the time distribution characteristics of heat users, indicating the temperature setting mode at different times of the day; E history is the historical energy consumption data, indicating the total amount of heat used in the historical period; Cbehavior is the weighting factor of the behavioral factor, indicating each heat user’s awareness of energy saving and sensitivity to external temperature changes;
[0032] Select an indoor temperature benchmark value, and divide each heat user into room temperature intervals according to the indoor temperature benchmark value, and users in the same room temperature interval are of the same type of heat users;
[0033] According to the historical water supply temperature and return water temperature data of each heat user in each period, the change trend of the supply and return water temperature of each heat user in each period is analyzed, and the average supply and return water temperature of each heat user in each period is selected to divide the supply and return water temperature interval, and the supply and return water temperatures in each period are in the same change trend and the same supply and return water temperature interval are divided into the same type of heat users;
[0034] Heat users with the same level of personalized heat utilization coefficient, the same room temperature range, the same supply and return water temperature change trend, and the same supply and return water temperature range are classified as the same type of heat users, and a second classification of each type of heat user is performed.
[0035] Furthermore, in S2, the influence weight of each characteristic variable on the heat load is calculated and a heat load prediction model for each heat user type and time period is established, including:
[0036] Select outdoor environmental parameters, indoor temperature, heating status parameters, personalized heat consumption coefficient and heat load as the first-level indicator characteristics; select outdoor temperature, outdoor humidity and outdoor wind speed in different time periods at historical moments and future moments as the second-level indicator characteristics of outdoor environmental parameters; select indoor temperature in different time periods at historical moments as the second-level indicator characteristics of indoor temperature; select heat user water supply temperature, return water temperature and circulation flow rate in different time periods at historical moments as the second-level indicator characteristics of heating status parameters; select heat consumption behavior habits, heat consumption time distribution and heat consumption energy consumption of each heat user as the second-level indicator characteristics of personalized heat consumption coefficient; select heat load of heat user building in different time periods at historical moments as the second-level indicator characteristics of heat load;
[0037] Calculate the first weight of each primary indicator feature and the second weight of each secondary indicator feature to obtain the comprehensive impact weight of each secondary indicator feature on the heat load of the heat user;
[0038] The comprehensive impact weights are sorted, the key secondary indicator characteristics and the comprehensive impact weights of the corresponding indicator characteristics are selected as input variables, the heat load is selected as the output variable, and the machine learning algorithm is used to establish a heat load prediction model for each type of heat user in different time periods.
[0039] Furthermore, the first weight calculation adopts the hierarchical analysis method, and the second weight calculation adopts the entropy weight method;
[0040] The comprehensive influence weight of each secondary indicator characteristic on the heat load of heat users is expressed as:
[0041]
[0042] λ j is the comprehensive influence weight of the j-th secondary indicator feature; α i is the first weight of the i-th first-level indicator feature; β j is the second weight of the jth secondary indicator feature; m is the number of primary indicator features; n is the number of secondary indicator features.
[0043] Furthermore, in S3, according to the heat load prediction results of each heat user type in different time periods, the valves in front of each heat user type are adjusted to simulate the heat storage and release process of the building of the heat user type, including:
[0044] According to the heat load forecast results of each type of heat user in different time periods, formulate a preliminary adjustment strategy for the valve opening before each type of heat user. For periods with high heat load, increase the valve opening in advance, and for periods with low heat load, reduce the valve opening.
[0045] For different heat user buildings, the constructed secondary network simulation model of the heating system is used to input the predicted heat load, environmental conditions and the valve opening in front of the house in each period, monitor and analyze the changes in the indoor temperature of the heat user when the valve opening in front of the house increases or decreases, and simulate the heat storage and heat release process of the heat user building in different time periods;
[0046] The heat storage process is expressed as:
[0047] Q storage =C×(T f -T 0 );
[0048] Q storage is the heat storage; C is the total heat capacity of the building; T f The indoor temperature of the building after the valve in front of the house is adjusted up; T 0 Adjust the indoor temperature of the building in front of the house for the valve in front of the house;
[0049] The exothermic process is expressed as:
[0050] Q loss =Q conv +Q rad +Q cond ;
[0051] Q loss is the heat release; Q conv is the heat release due to convection; Q rad It is radiation heat release; Q cond To conduct heat and release heat.
[0052] Furthermore, in S3, the heat transfer process of each heat user type building includes a heat conduction process, a convection heat transfer process and a radiation heat transfer process;
[0053] The thermal inertia index of the building is expressed as:
[0054] D = R * S;
[0055] R is the thermal resistance of the building envelope of the heat user; S is the thermal storage coefficient of the building envelope of the heat user; the attenuation and hysteresis characteristics of the building in response to external changes are expressed as:
[0056]
[0057] ζ=t 0 -t f ;
[0058] f is the attenuation coefficient; T 0 max Maximum indoor temperature of the building; T 0 min is the minimum indoor temperature of the building; T e max is the maximum outdoor temperature; T e min is the minimum outdoor temperature; It is the lag time of the indoor temperature of the building responding to the change of outdoor temperature; is the time when the indoor temperature reaches its peak value; is the time when the outdoor temperature peaks; ζ is the lag time of the building's indoor temperature responding to the change in the valve adjustment in front of the house; t 0 is the time when the indoor temperature of the building changes; t f Adjust the time for the valve in front of the house.
[0059] Furthermore, in S4, a valve control model for each type of heat user is established, including:
[0060] According to the indoor temperature setting values of each heat user type in different time periods, the heat load prediction results of each heat user type in different time periods, and the comprehensive analysis results of the thermal inertia of the building of each heat user type, the predicted change time and amount of indoor temperature and heat load, as well as the lag time and attenuation characteristics of the building response change are obtained, and the control time of the front valve response change of each heat user type and the heat storage and heat release of the building in each time period after the front valve is controlled are determined;
[0061] Based on the heat load prediction results of each type of heat user in different time periods and the heat storage and heat release of the building, the heat supply value of each type of heat user in different time periods is determined, and the minimum control cost is taken as the objective function. The model predictive control method is used to establish the household valve control model of each type of heat user, and the household valve opening control strategy of each type of heat user in different time periods is output.
[0062] Furthermore, in S5, the multi-dimensional evaluation index includes: comfort index, energy efficiency index, stability index and economy index.
[0063] The beneficial effects of the present invention are:
[0064] (1) The present invention considers the basic attributes of each heat user building and the different heating state parameters of each heat user, fits and calculates the composite heat transfer coefficient of each heat user building, and uses a clustering algorithm to perform a first classification of each heat user type based on the composite heat transfer coefficient, and then performs a second classification of each heat user type based on the personalized heat utilization coefficient, indoor temperature and supply and return water temperature level of each heat user; it can comprehensively consider the static data, dynamic data and personalized heat utilization coefficient of each heat user building, perform a two-layer classification of the heat user building type, refine each heat user type, improve the accuracy of heat user classification, and facilitate subsequent on-demand regulation;
[0065] (2) The present invention selects characteristic variables for load prediction for each type of heat user after the second classification, including outdoor environmental parameters, heat user heating state parameters, heat user indoor temperature, heat user personalized heat coefficient and heat load, and calculates the influence weight of each characteristic variable on the heat load and establishes a heat load prediction model for each type of heat user in different time periods; it is possible to deeply understand the influence of different characteristic variables on the heat load, and further improve the accuracy of heat load prediction;
[0066] (3) The present invention adjusts the valves in front of each type of heat user according to the heat load prediction results of each type of heat user in different time periods, simulates the heat storage and release process of the building of each type of heat user, and combines the heat transfer process of the building of each type of heat user, thermal inertia index and the attenuation and hysteresis characteristics of the building in response to external changes to conduct a comprehensive analysis of the thermal inertia of the building of each type of heat user; determines the indoor temperature setting value of each type of heat user in different time periods, and combines the heat load prediction results of each type of heat user in different time periods and the comprehensive analysis results of the thermal inertia of the building of each type of heat user to determine the valve control time in front of each type of heat user and the heat storage and release mechanism of the building in each time period, establishes a valve control model in front of each type of heat user, and outputs the valve opening control strategy in front of each type of heat user in different time periods; can comprehensively consider the thermal inertia of the building of each type of heat user, effectively utilize the heat storage and release mechanism and hysteresis characteristics, perform on-demand precise control of the valves in front of each type of heat user, and achieve the purpose of energy saving and emission reduction;
[0067] (4) The present invention performs household regulation based on the household valve regulation strategy of each heat user type and time period, and then selects multi-dimensional evaluation indicators to evaluate the household regulation effect, which can comprehensively evaluate the regulation effect and clarify the regulation effect.
[0068] Other features and advantages will be described in the following description, and partly become apparent from the description, or understood by practicing the invention. The purpose and other advantages of the invention are realized and obtained by the structures particularly pointed out in the description and the drawings.
[0069] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0070] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0071] Figure 1 A flow chart of a secondary network control method for a heating system considering heat user classification and building thermal inertia according to the present invention;
[0072] Figure 2 It is a schematic diagram of the secondary network control structure of the heating system of the present invention. DETAILED DESCRIPTION
[0073] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0074] Example 1
[0075] like Figure 1 As shown, this embodiment 1 provides a secondary network control method for a heating system taking into account the classification of heat users and thermal inertia of buildings, which includes:
[0076] S1. Considering the basic attributes of each heat user building and the different heating state parameters of each heat user, the composite heat transfer coefficient of each heat user building is calculated by fitting, and the clustering algorithm is used to perform the first classification of each heat user type based on the composite heat transfer coefficient, and then the second classification of each heat user type is performed based on the personalized heat coefficient, indoor temperature and supply and return water temperature level of each heat user;
[0077] S2. Select characteristic variables for load prediction for each type of heat user after the second classification, including outdoor environmental parameters, heat user heating status parameters, heat user indoor temperature, heat user personalized heat coefficient and heat load, and calculate the influence weight of each characteristic variable on the heat load and establish a heat load prediction model for each type of heat user in different time periods;
[0078] S3. According to the heat load prediction results of each type of heat user in different time periods, adjust the valves in front of each type of heat user, simulate the heat storage and release process of the building of each type of heat user, and conduct a comprehensive analysis of the thermal inertia of the building of each type of heat user in combination with the heat transfer process of the building of each type of heat user, the thermal inertia index and the attenuation and hysteresis characteristics of the response to external changes;
[0079] S4. Determine the indoor temperature setting value for each type of heat user in different time periods, and combine the heat load prediction results for each type of heat user in different time periods and the comprehensive analysis results of the thermal inertia of the building for each type of heat user to determine the valve control time for each type of heat user and the heat storage and release mechanism of the building in each time period, establish a valve control model for each type of heat user, and output the valve opening control strategy for each type of heat user in different time periods;
[0080] S5. After household regulation is performed according to the pre-household valve regulation strategy for each type of heat user in different time periods, multi-dimensional evaluation indicators are selected to evaluate the effect of household regulation.
[0081] In this embodiment, in S1, fitting and calculating the composite heat transfer coefficient of each heat user building includes:
[0082] The heat balance model of each heat user building is established and expressed as:
[0083]
[0084] K is the heat transfer coefficient of the radiator; F is the heat dissipation area of the radiator; K′ is the heat transfer coefficient of the external wall; F′ is the area of the external wall; t g Water supply temperature for heat users; t h is the return water temperature of the heat user; t n is the indoor temperature of the heat user; t w is the outdoor temperature;
[0085] The composite heat transfer coefficient of each heat user building is calculated by fitting, which is expressed as:
[0086]
[0087] In this embodiment, in S1, when a clustering algorithm is used to perform a first classification on each heat user type according to the composite heat transfer coefficient, the clustering algorithm used is an agglomerative hierarchical clustering algorithm, which clusters each heat user and divides similar heat users into the same cluster to form a cluster tree, including:
[0088] Obtain the basic attributes of each heat user building and the heating status parameters of the heat user, including the building area, structure type, building properties, heat transfer coefficient of the radiator, heat dissipation area of the radiator, area of the enclosure structure, heat transfer coefficient of the enclosure structure, supply and return water temperature of the heat user, circulation flow, supply and return water pressure, indoor temperature and outdoor temperature;
[0089] The basic attributes of each heat user's building, the heat user's heating state parameters and the composite heat transfer coefficient of each heat user are used as the heat user clustering sample data;
[0090] Choose a similarity measure, including Euclidean distance, Manhattan distance, Chebyshev distance, and cosine similarity;
[0091] Each hot user is initially treated as an independent cluster to construct the initial clusters;
[0092] Using the selected similarity measure method, calculate the similarity between each two clusters;
[0093] Find the two clusters with the smallest similarity among all the current cluster pairs, merge the two clusters into a new cluster, remove the original two clusters from the cluster list, add the merged cluster, and recalculate the similarity between the merged cluster and other clusters;
[0094] The most similar clusters are gradually merged until the predetermined termination condition is reached, and a clustering tree is constructed based on the information of each merged cluster;
[0095] The optimal number of clusters is determined based on the clustering tree, and the characteristic differences between different clusters and the distribution law of the composite heat transfer coefficient of each heat user building are obtained.
[0096] It should be noted that as an unsupervised learning algorithm, agglomerative hierarchical clustering aims to divide the data set into multiple clusters so that the similarity between clusters is high and the similarity between clusters is low. Compared with the K-Means clustering algorithm, which is affected by the initial clustering points, agglomerative hierarchical clustering only needs to specify the clustering end conditions.
[0097] In this embodiment, in S1, the second classification of each heat user type is performed according to the personalized heat coefficient, indoor temperature and supply and return water temperature level of each heat user, including:
[0098] According to the heat usage behavior habits, heat usage time distribution, and heat energy consumption of each heat user, the personalized heat usage coefficient of each heat user is fitted, which is expressed as:
[0099] PUI=f(H duration ,T profile ,E history ,C behavior );
[0100] H duration T is the heating time preference of heat users, indicating the heating time in a day or a specific time period; profile is the time distribution characteristics of heat users, indicating the temperature setting mode at different times of the day; E history is the historical energy consumption data, indicating the total amount of heat used in the historical period; C behavior is the weighting factor of the behavioral factor, indicating each heat user’s awareness of energy saving and sensitivity to external temperature changes;
[0101] Select an indoor temperature benchmark value, and divide each heat user into room temperature intervals according to the indoor temperature benchmark value, and users in the same room temperature interval are of the same type of heat users;
[0102] According to the historical water supply temperature and return water temperature data of each heat user in each period, the change trend of the supply and return water temperature of each heat user in each period is analyzed, and the average supply and return water temperature of each heat user in each period is selected to divide the supply and return water temperature interval, and the supply and return water temperatures in each period are in the same change trend and the same supply and return water temperature interval are divided into the same type of heat users;
[0103] Heat users with the same level of personalized heat utilization coefficient, the same room temperature range, the same supply and return water temperature change trend, and the same supply and return water temperature range are classified as the same type of heat users, and a second classification of each type of heat user is performed.
[0104] It should be noted that heat user buildings are first divided according to their uses, including commercial heat user buildings, civil heat user buildings, corporate heat user buildings and school heat user buildings, etc. The heating time of each type of heat user building is different. For example, commercial heat user buildings use heat during the opening and closing time of shopping malls, and the peak and trough periods of heat use are also different from those of buildings with other uses; different heat users in civil heat user buildings have different work and rest behaviors, work nature and other factors. Some heat users have no one at home during the day on weekdays and have low heat demand during the day; some heat users have elderly and children at home during the day on weekdays and have high heat demand; some heat users like high indoor temperature at home, which can reach 25 degrees; some heat users prefer indoor temperature of 22 degrees; corporate heat user buildings The heat use time period and peak heat use period of a building are related to the company's commuting time and the number of staff in the company; the heat use behavior of school heat user buildings matches the normal teaching time of the school; the heat use of enterprise heat user buildings is related to the office and production process operation of the company; there are also personalized differences in heat use between heat user buildings for each purpose due to the structure, orientation, location, etc. of the building. For example, in the same unit building, the indoor temperature of heat users in the lower, middle and upper floors will be different, and for heat users on the same floor, due to factors such as orientation, structure, radiator deployment, whether they are close to the heating well, there will also be obvious differences in indoor temperature. The supply and return water temperatures of radiators are also different, resulting in substandard heating in some users' homes. Therefore, for heat user buildings with different purposes, detailed division of heat user buildings is carried out according to personalized heat use coefficients, indoor temperature levels, and supply and return water temperature levels, which can better carry out on-demand regulation and meet the heat use needs of various heat user buildings. In addition, the personalized heat use coefficient of each heat user is not limited to heat use behavior habits, heat use time distribution, and heat use energy consumption.
[0105] The secondary network control structure of the heating system is as follows Figure 2 As shown, there are multiple heat user buildings connected behind the heat exchange station, which may be civil heat user buildings, commercial heat user buildings and industrial heat user buildings. Heat users can be classified according to each heat user building, and a valve is installed in front of each heat user to adjust the heat user radiator.
[0106] In this embodiment, in S2, calculating the influence weight of each characteristic variable on the heat load and establishing a heat load prediction model for each heat user type in different time periods include:
[0107] Select outdoor environmental parameters, indoor temperature, heating status parameters, personalized heat consumption coefficient and heat load as the first-level indicator characteristics; select outdoor temperature, outdoor humidity and outdoor wind speed in different time periods at historical moments and future moments as the second-level indicator characteristics of outdoor environmental parameters; select indoor temperature in different time periods at historical moments as the second-level indicator characteristics of indoor temperature; select heat user water supply temperature, return water temperature and circulation flow rate in different time periods at historical moments as the second-level indicator characteristics of heating status parameters; select heat consumption behavior habits, heat consumption time distribution and heat consumption energy consumption of each heat user as the second-level indicator characteristics of personalized heat consumption coefficient; select heat load of heat user building in different time periods at historical moments as the second-level indicator characteristics of heat load;
[0108] Calculate the first weight of each primary indicator feature and the second weight of each secondary indicator feature to obtain the comprehensive impact weight of each secondary indicator feature on the heat load of the heat user;
[0109] The comprehensive impact weights are sorted, the key secondary indicator characteristics and the comprehensive impact weights of the corresponding indicator characteristics are selected as input variables, the heat load is selected as the output variable, and the machine learning algorithm is used to establish a heat load prediction model for each type of heat user in different time periods.
[0110] In this embodiment, the first weight calculation adopts the hierarchical analysis method, and the second weight calculation adopts the entropy weight method;
[0111] The comprehensive influence weight of each secondary indicator characteristic on the heat load of heat users is expressed as:
[0112]
[0113] λ j is the comprehensive influence weight of the j-th secondary indicator feature; α i is the first weight of the i-th first-level indicator feature; β j is the second weight of the jth secondary indicator feature; m is the number of primary indicator features; n is the number of secondary indicator features.
[0114] In this embodiment, in S3, according to the heat load prediction results of each heat user type in different time periods, the valves in front of each heat user type are adjusted to simulate the heat storage and release process of the building of the heat user type, including:
[0115] According to the heat load forecast results of each type of heat user in different time periods, formulate a preliminary adjustment strategy for the valve opening before each type of heat user. For periods with high heat load, increase the valve opening in advance, and for periods with low heat load, reduce the valve opening.
[0116] For different heat user buildings, the constructed secondary network simulation model of the heating system is used to input the predicted heat load, environmental conditions and the valve opening in front of the house in each period, monitor and analyze the changes in the indoor temperature of the heat user when the valve opening in front of the house increases or decreases, and simulate the heat storage and heat release process of the heat user building in different time periods;
[0117] The heat storage process is expressed as:
[0118] Q storage =C×(T f -T 0 );
[0119] Q storage is the heat storage; C is the total heat capacity of the building; T f The indoor temperature of the building after the valve in front of the house is adjusted up; T 0 Adjust the indoor temperature of the building in front of the house for the valve in front of the house;
[0120] The exothermic process is expressed as:
[0121] Q loss =Q conv +Q rad +Q cond ;
[0122] Q loss is the heat release; Q conv is the heat release due to convection; Q rad It is radiation heat release; Q cond To conduct heat and release heat.
[0123] In this embodiment, in S3, the heat transfer process of each heat user type building includes a heat conduction process, a convection heat transfer process and a radiation heat transfer process;
[0124] The thermal inertia index of the building is expressed as:
[0125] D = R * S;
[0126] R is the thermal resistance of the building envelope of the heat user; S is the thermal storage coefficient of the building envelope of the heat user; the attenuation and hysteresis characteristics of the building in response to external changes are expressed as:
[0127]
[0128] ζ=t 0 -t f ;
[0129] f is the attenuation coefficient; T 0 max Maximum indoor temperature of the building; T 0 minis the minimum indoor temperature of the building; T e max is the maximum outdoor temperature; T e min is the minimum outdoor temperature; It is the lag time of the indoor temperature of the building responding to the change of outdoor temperature; is the time when the indoor temperature reaches its peak value; is the time when the outdoor temperature peaks; ζ is the lag time of the building's indoor temperature responding to the change in the valve adjustment in front of the house; t 0 is the time when the indoor temperature of the building changes; t f Adjust the time for the valve in front of the house.
[0130] It should be noted that the heat storage and release process considering valve regulation includes:
[0131] 1) Calculate the adjusted heat flow: Calculate the actual heat flow based on the current valve opening;
[0132] 2) Update indoor temperature prediction: Based on the current indoor temperature, heat flow, and heat release in the previous period, predict the indoor temperature in the next period;
[0133] 3) Calculate heat release: Recalculate convection, radiation, and conduction heat release using the updated indoor temperature and environmental conditions (outdoor temperature, sky temperature, etc.);
[0134] 4) Calculate the heat storage: Calculate the heat storage based on the predicted indoor temperature change and actual heat flow.
[0135] It should be noted that convection heat release: Considering the convection heat transfer between the building surface and the air, it can be expressed as:
[0136] Q conv =h×A×(T surf -T air );
[0137] h is the convective heat transfer coefficient; A is the building surface area; T surf is the surface temperature; T air is the ambient air temperature;
[0138] Radiative heat release: Considering the radiative heat exchange between the building surface and the surrounding environment, it can be expressed as:
[0139]
[0140] ε is the emissivity of the building surface; σ is the Stepan-Boltzmann constant; is the fourth power of the surface temperature; is the fourth power of the average radiation temperature of the surrounding environment;
[0141] Heat conduction and heat release: Considering the heat conduction inside the building walls, roof, ground and other structures, it can be expressed as:
[0142]
[0143] k is the thermal conductivity of the building envelope material; δ is the thickness of the building envelope material; T in , T out are the temperatures of the inner and outer surfaces of the building envelope, respectively.
[0144] In this embodiment, in S4, a valve control model for each type of heat user is established, including:
[0145] According to the indoor temperature setting values of each heat user type in different time periods, the heat load prediction results of each heat user type in different time periods, and the comprehensive analysis results of the thermal inertia of the building of each heat user type, the predicted change time and amount of indoor temperature and heat load, as well as the lag time and attenuation characteristics of the building response change are obtained, and the control time of the front valve response change of each heat user type and the heat storage and heat release of the building in each time period after the front valve is controlled are determined;
[0146] Based on the heat load prediction results of each type of heat user in different time periods and the heat storage and heat release of the building, the heat supply value of each type of heat user in different time periods is determined, and the minimum control cost is taken as the objective function. The model predictive control method is used to establish the household valve control model of each type of heat user, and the household valve opening control strategy of each type of heat user in different time periods is output.
[0147] It should be noted that the model predictive control method is a stochastic model predictive control method combined with affine disturbance, and the state space model is expressed as:
[0148] x(k+1)=Ax(k)+Bu(k)+Gd(k)+Ew(k);
[0149] k is the current moment; x, u, d, and w are state quantities, control input quantities, external input quantities, and independent random disturbance quantities, respectively; state quantities include indoor temperature of heat users, heat load of heat users, heat storage, and heat release; control input quantities include the opening of valves in front of the house; external input quantities include outdoor ambient temperature; random disturbance quantities include heat load prediction error and outdoor ambient temperature prediction error; A, B, G, and E are state matrices, control matrices, input matrices, and disturbance matrices, respectively;
[0150] In addition, considering the influence of random disturbances, which often leads to the inability to fully meet the conventional hard constraints, the description method of chance constraints can be adopted to express the system input and state constraints in probabilistic form.
[0151] In this embodiment, in S5, the multi-dimensional evaluation index includes: comfort index, energy efficiency index, stability index and economy index.
[0152] The specific description of each evaluation indicator is as follows:
[0153] 1) Comfort index
[0154] Indoor temperature satisfaction: Through automatic monitoring, statistics are collected on user satisfaction with indoor temperature, whether it reaches the expected set value, and whether the fluctuation range is within an acceptable range.
[0155] Temperature uniformity: Evaluate the uniformity of temperature distribution within the same heat user or between different heat users to avoid local overheating or overcooling.
[0156] 2) Energy efficiency indicators
[0157] Energy efficiency: Compare the changes in energy consumption per unit area or total energy consumption before and after the implementation of the control strategy to evaluate the energy-saving and emission reduction effects.
[0158] Improvement of heat source efficiency: By analyzing the matching degree between heat source output and user demand, the energy efficiency improvement on the heat source side is evaluated.
[0159] Heat network transmission efficiency: Inspect the water supply temperature, return water temperature and circulation pump energy consumption to evaluate whether the heat network transmission efficiency has improved.
[0160] 3) System stability indicators
[0161] Control response time: Measures the time from when a control instruction is issued to when the system actually responds and reaches the target state.
[0162] Regulation accuracy: Evaluates the deviation between the actual effect of valve regulation and the target set value, reflecting the accuracy of the regulation system.
[0163] System anti-interference ability: The ability of the system to maintain stable operation when analyzing changes in the external environment (such as sudden changes in temperature).
[0164] 4) Economic indicators
[0165] Operating costs: Comprehensively consider the changes in total operating costs caused by the implementation of control strategies, such as electricity costs, maintenance costs, and labor management costs.
[0166] Return on investment: Evaluate the long-term economic benefits of investments in advanced control equipment or technologies.
[0167] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.
[0168] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0169] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.
Claims
1. A secondary network control method for a heating system considering heat user classification and building thermal inertia, characterized in that: It includes: S1. Considering the basic attributes of each heat user building and the different heating state parameters of each heat user, the composite heat transfer coefficient of each heat user building is calculated by fitting, and the clustering algorithm is used to perform the first classification of each heat user type based on the composite heat transfer coefficient, and then the second classification of each heat user type is performed based on the personalized heat coefficient, indoor temperature and supply and return water temperature level of each heat user; S2. Select characteristic variables for load prediction for each type of heat user after the second classification, including outdoor environmental parameters, heat user heating status parameters, heat user indoor temperature, heat user personalized heat coefficient and heat load, and calculate the influence weight of each characteristic variable on the heat load and establish a heat load prediction model for each type of heat user in different time periods; S3. According to the heat load prediction results of each type of heat user in different time periods, adjust the valves in front of each type of heat user, simulate the heat storage and release process of the building of each type of heat user, and conduct a comprehensive analysis of the thermal inertia of the building of each type of heat user in combination with the heat transfer process of the building of each type of heat user, the thermal inertia index and the attenuation and hysteresis characteristics of the response to external changes; S4. Determine the indoor temperature setting value for each type of heat user in different time periods, and combine the heat load prediction results for each type of heat user in different time periods and the comprehensive analysis results of the thermal inertia of the building for each type of heat user to determine the valve control time for each type of heat user and the heat storage and release mechanism of the building in each time period, establish a valve control model for each type of heat user, and output the valve opening control strategy for each type of heat user in different time periods; S5. After household regulation is performed according to the pre-household valve regulation strategy for each type of heat user in different time periods, multi-dimensional evaluation indicators are selected to evaluate the effect of household regulation.
2. The secondary network control method for a heating system according to claim 1, characterized in that: In S1, the composite heat transfer coefficient of each heat user building is fitted and calculated, including: The heat balance model of each heat user building is established and expressed as: K is the heat transfer coefficient of the radiator; F is the heat dissipation area of the radiator; K′ is the heat transfer coefficient of the external wall; F′ is the area of the external wall; t g Water supply temperature for heat users; t h is the return water temperature of the heat user; t n is the indoor temperature of the heat user; t w is the outdoor temperature; The composite heat transfer coefficient of each heat user building is calculated by fitting, which is expressed as:
3. The secondary network control method for a heating system according to claim 1, characterized in that: In S1, when a clustering algorithm is used to perform a first classification of each heat user type according to the composite heat transfer coefficient, the clustering algorithm used is an agglomerative hierarchical clustering algorithm, which clusters each heat user and divides similar heat users into the same cluster to form a cluster tree, including: Obtain the basic attributes of each heat user building and the heating status parameters of the heat user, including the building area, structure type, building properties, heat transfer coefficient of the radiator, heat dissipation area of the radiator, area of the enclosure structure, heat transfer coefficient of the enclosure structure, supply and return water temperature of the heat user, circulation flow, supply and return water pressure, indoor temperature and outdoor temperature; The basic attributes of each heat user's building, the heat user's heating state parameters and the composite heat transfer coefficient of each heat user are used as the heat user clustering sample data; Choose a similarity measure, including Euclidean distance, Manhattan distance, Chebyshev distance, and cosine similarity; Each hot user is initially treated as an independent cluster to construct the initial clusters; Using the selected similarity measure method, calculate the similarity between each two clusters; Find the two clusters with the smallest similarity among all the current cluster pairs, merge the two clusters into a new cluster, remove the original two clusters from the cluster list, add the merged cluster, and recalculate the similarity between the merged cluster and other clusters; The most similar clusters are gradually merged until the predetermined termination condition is reached, and a clustering tree is constructed based on the information of each merged cluster; The optimal number of clusters is determined based on the clustering tree, and the characteristic differences between different clusters and the distribution law of the composite heat transfer coefficient of each heat user building are obtained.
4. The secondary network control method for a heating system according to claim 1, characterized in that: In S1, the second classification of each heat user type is performed according to the individual heat coefficient, indoor temperature and supply and return water temperature level of each heat user, including: According to the heat usage behavior habits, heat usage time distribution, and heat energy consumption of each heat user, the personalized heat usage coefficient of each heat user is fitted, which is expressed as: PUI=f(H duration ,T profile ,E history ,C behavior ); H duration T is the heating time preference of heat users, indicating the heating time in a day or a specific time period; profile is the time distribution characteristics of heat users, indicating the temperature setting mode at different times of the day; E history is the historical energy consumption data, indicating the total amount of heat used in the historical period; C behavior is the weighting factor of the behavioral factor, indicating each heat user’s awareness of energy saving and sensitivity to external temperature changes; Select an indoor temperature benchmark value, and divide each heat user into room temperature intervals according to the indoor temperature benchmark value, and users in the same room temperature interval are of the same type of heat users; According to the historical water supply temperature and return water temperature data of each heat user in each period, the change trend of the supply and return water temperature of each heat user in each period is analyzed, and the average supply and return water temperature of each heat user in each period is selected to divide the supply and return water temperature interval, and the supply and return water temperatures in each period are in the same change trend and the same supply and return water temperature interval are divided into the same type of heat users; Heat users with the same level of personalized heat utilization coefficient, the same room temperature range, the same supply and return water temperature change trend, and the same supply and return water temperature range are classified as the same type of heat users, and a second classification of each type of heat user is performed.
5. The secondary network control method for a heating system according to claim 1, characterized in that: In S2, the influence weight of each characteristic variable on the heat load is calculated and a heat load prediction model for each heat user type and time period is established, including: Select outdoor environmental parameters, indoor temperature, heating status parameters, personalized heat consumption coefficient and heat load as the first-level indicator characteristics; select outdoor temperature, outdoor humidity and outdoor wind speed in different time periods at historical moments and future moments as the second-level indicator characteristics of outdoor environmental parameters; select indoor temperature in different time periods at historical moments as the second-level indicator characteristics of indoor temperature; select heat user water supply temperature, return water temperature and circulation flow rate in different time periods at historical moments as the second-level indicator characteristics of heating status parameters; select heat consumption behavior habits, heat consumption time distribution and heat consumption energy consumption of each heat user as the second-level indicator characteristics of personalized heat consumption coefficient; select heat load of heat user building in different time periods at historical moments as the second-level indicator characteristics of heat load; Calculate the first weight of each primary indicator feature and the second weight of each secondary indicator feature to obtain the comprehensive impact weight of each secondary indicator feature on the heat load of the heat user; The comprehensive impact weights are sorted, the key secondary indicator characteristics and the comprehensive impact weights of the corresponding indicator characteristics are selected as input variables, the heat load is selected as the output variable, and the machine learning algorithm is used to establish a heat load prediction model for each type of heat user in different time periods.
6. The secondary network control method for a heating system according to claim 5, characterized in that: The first weight calculation adopts the hierarchical analysis method, and the second weight calculation adopts the entropy weight method; The comprehensive influence weight of each secondary indicator characteristic on the heat load of heat users is expressed as: λ j is the comprehensive influence weight of the j-th secondary indicator feature; α i is the first weight of the i-th first-level indicator feature; β j is the second weight of the jth secondary indicator feature; m is the number of primary indicator features; n is the number of secondary indicator features.
7. The secondary network control method for a heating system according to claim 1, characterized in that: In S3, according to the heat load prediction results of each heat user type in different time periods, the valves in front of each heat user type are adjusted to simulate the heat storage and release process of the building of the heat user type, including: According to the heat load forecast results of each type of heat user in different time periods, formulate a preliminary adjustment strategy for the valve opening before each type of heat user. For periods with high heat load, increase the valve opening in advance, and for periods with low heat load, reduce the valve opening. For different heat user buildings, the constructed secondary network simulation model of the heating system is used to input the predicted heat load, environmental conditions and the valve opening in front of the house in each period, monitor and analyze the changes in the indoor temperature of the heat user when the valve opening in front of the house increases or decreases, and simulate the heat storage and heat release process of the heat user building in different time periods; The heat storage process is expressed as: Q storage =C×(T f -T0); Q storage is the heat storage; C is the total heat capacity of the building; T f T0 is the indoor temperature of the building after the valve in front of the household is adjusted up; T1 is the indoor temperature of the building before the valve in front of the household is adjusted up; The exothermic process is expressed as: Q loss =Q conv +Q rad +Q cond ; Q loss is the heat release; Q conv is the heat release due to convection; Q rad It is radiation heat release; Q cond To conduct heat and release heat.
8. The secondary network control method for a heating system according to claim 1, characterized in that: In S3, the heat transfer process of each heat user type building includes a heat conduction process, a convection heat transfer process and a radiation heat transfer process; The thermal inertia index of the building is expressed as: D = R * S; R is the thermal resistance of the building envelope of the heat user; S is the thermal storage coefficient of the building envelope of the heat user; The attenuation and hysteresis characteristics of the building body in response to external changes are expressed as: ζ=t0-t f ; f is the attenuation coefficient; Maximum indoor temperature of the building; is the minimum indoor temperature of the building; is the maximum outdoor temperature; is the minimum outdoor temperature; It is the lag time of the indoor temperature of the building responding to the change of outdoor temperature; is the time when the indoor temperature reaches its peak value; is the time when the outdoor temperature peaks; ζ is the lag time of the indoor temperature of the building responding to the change of the valve adjustment in front of the house; t0 is the time when the indoor temperature of the building changes; t f Adjust the time for the valve in front of the house.
9. The method for controlling the secondary network of a heating system according to claim 1, characterized in that: In S4, a valve control model for each type of heat user is established, including: According to the indoor temperature setting values of each heat user type in different time periods, the heat load prediction results of each heat user type in different time periods, and the comprehensive analysis results of the thermal inertia of the building of each heat user type, the predicted change time and amount of indoor temperature and heat load, as well as the lag time and attenuation characteristics of the building response change are obtained, and the control time of the front valve response change of each heat user type and the heat storage and heat release of the building in each time period after the front valve is controlled are determined; Based on the heat load prediction results of each type of heat user in different time periods and the heat storage and heat release of the building, the heat supply value of each type of heat user in different time periods is determined, and the minimum control cost is taken as the objective function. The model predictive control method is used to establish the household valve control model of each type of heat user, and the household valve opening control strategy of each type of heat user in different time periods is output.
10. The secondary network control method for a heating system according to claim 1, characterized in that: In S5, the multi-dimensional evaluation index includes: comfort index, energy efficiency index, stability index and economy index.