An intelligent control method for the secondary network of a heating system based on demand-side regulation

Abstract

An intelligent control method for the secondary network of a heating system based on demand-side regulation, which includes: Step 1, setting the building regulation period of the heating system; Step 2, temperature acquisition and building regulation, performing temperature acquisition, building user classification and building regulation, and the building user classification is divided into super-cooled buildings, over-heated buildings and suitable-temperature buildings through a clustering algorithm; Step 3, heat source regulation; according to the different divisions of building users in super-cooled buildings and suitable-temperature buildings at the moment of the building regulation period, different heat source regulation strategies are executed, and so on, forming a cyclic control mode with nested building regulation periods and heat source control periods, and completing the demand-side regulation of the secondary network of the heating system. The method of the present invention makes up for the lack of demand-side regulation of the heating system, has accurate control and good applicability. Under this regulation method, the heating system greatly reduces the excessive heating of the system, improves the comfort level of users at the same time, and improves the overall operation efficiency of the heating system.

Description

Technical Field

[0001] The present invention relates to the field of centralized heating systems, and in particular to an intelligent control method for a secondary network of a heating system based on demand-side regulation. Background Art

[0002] Currently, energy-saving regulation is a major challenge facing heat exchanger station operations in my country. Research on steady-state operation regulation strategies is relatively mature. Using various machine learning algorithms, operating parameters can be adjusted based on load demand to achieve on-demand heating. Compared to conventional heat load maps, this approach can improve the control accuracy of individual heat exchanger stations or heat sources by reducing the genericity of strategies. However, balancing the circulating flow in pipe networks is often difficult. For a long time, heat supply companies have relied on broad and crude control methods such as quality and quantity regulation. This makes it difficult to achieve precise and accurate control. Traditional heating systems utilize various control methods, including quality regulation, quantity regulation, phased quality regulation, and intermittent regulation. Traditional systems employ source-side regulation. The control strategy involves adjusting the entire heating system based on the outdoor temperature and the overall return water temperature at the heat source. However, hydraulic imbalances are common in these situations. Many heat users close to the heat source open windows for ventilation, while users farther away complain, resulting in significant system overheating losses. Therefore, currently proposed smart heating systems are gradually promoting the implementation of demand-side control to reduce unnecessary system overheating losses. However, the implementation of demand-side control relies not only on user-specific control and adjustment of demand, but also includes timely supply control at the heat source. However, the proposed demand-side control does not consider supply control at the heat source, and therefore, actual heating system operation and control lacks demand-side control methods. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, this invention provides a method for intelligently controlling the secondary network of a heating system based on demand-side regulation. This method addresses the lack of demand-side regulation in heating systems, offers accurate control and excellent applicability. This method significantly reduces overheating in the heating system while improving user comfort.

[0004] A method for intelligently controlling a secondary network of a heating system based on demand-side regulation includes:

[0005] Step 1: Set the building control cycle of the heating system to T B ;

[0006] Step 2: Temperature collection and building control, every time mT BDuring a period, temperature acquisition, building user classification, and building regulation are performed. The building user classification is divided into supercooled buildings, overheated buildings, and suitable-temperature buildings through a clustering algorithm;

[0007] The (m + 1)T B During the period, mT is executed B The temperature range of the suitable-temperature building at the moment, and the building users are reclassified. And whenever (m + 1)T B At the moment, heat source regulation is performed under the conditions satisfying Step 3, where m is an odd number and m≥1;

[0008] Step 3: Heat source regulation;

[0009] (1) When there is a heating building that is classified as a supercooled building at both the mT B moment and the (m + 1)T B moment, at this time, increase the heat supply of the heat source and increase the frequency of the heat source water pump;

[0010] (2) When there is a building marked as a supercooled building at the mT B moment and marked as a suitable-temperature building at the (m + 1)T B moment, and the opening degree of the marked valve is not 100% for the heating building, at this time, reduce the heat supply of the heat source and reduce the frequency of the heat source water pump;

[0011] (3) When there are no supercooled buildings at both the mT B moment and the (m + 1)T B moment, and there is no suitable-temperature building with a valve opening of 100% at the (m + 1)T B moment, at this time, reduce the heat supply of the heat source and reduce the frequency of the heat source water pump;

[0012] And so on, forming a cyclic control mode with the building regulation period nested with the heat source control period to complete the demand-side regulation of the secondary network of the heating system.

[0013] The beneficial effects of the present invention compared with the prior art are:

[0014] Different from the traditional centralized heat source end regulation of the heating system, the regulation strategy of the present invention provides an overall heating system regulation strategy based on the demand side. The system regulation not only eliminates the random characteristics during the autonomous regulation process of heating building users through the clustering method, but also considers the time-delay factor of the temperature change of the building system users through the cyclic control mode with the building regulation period nested with the heat source control period. The overall regulation method has a strong control purpose and good applicability, providing a practical regulation method for the demand-side regulation of the heating system. Under this regulation method, the heating system greatly reduces the excessive heating of the system, improves the comfort level of users, and improves the overall operation efficiency of the heating system.

[0015] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and embodiments: Description of the Drawings

[0016] Figure 1 It is a cyclic control mode diagram of the nested building control cycle and heat source control cycle of the present invention;

[0017] Figure 2 It is a schematic diagram of the temperature distribution of building users in the first building control cycle in the embodiment;

[0018] Figure 3 It is a schematic diagram of the distribution of building user categories in the first building control cycle in the embodiment;

[0019] Figure 4 For Figure 1 The first heat source control cycle diagram;

[0020] Figure 5 For Figure 1 The heat source control cycle diagram after the first heat source control cycle; Specific Embodiments

[0021] The embodiments of the technical solution of the present invention will be described in detail below in conjunction with the accompanying drawings. Unless otherwise specified, the technical terms or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art to which the present invention belongs.

[0022] The intelligent control method for the secondary network in this embodiment effectively utilizes the operation monitoring data of the heating system through a clustering algorithm to achieve demand-side regulation of the secondary network of the heating system, improve the overall energy efficiency of the heating system, and reduce unnecessary carbon emissions of the system. Different from the traditional heat source-side control, the demand-side regulation includes two levels of operation control. The bottom layer is the demand side (referring to the user side in the heating system) to control according to its own needs, and the top layer is the heat source side to supply heat according to the needs of the demand side. In the overall control process of the system, the control end only reaches the building level.

[0023] In the disordered and random regulation state of demand-side regulation, a large-scale oscillation effect will be formed in the whole control process, and all control devices are in a constantly changing state. Therefore, the demand-side regulation process of the system is a periodic regulation. During the operation of the system, the heating system data acquisition center collects the real-time indoor temperature of each building user and the opening degree of the building entrance valve.

[0024] First Embodiment

[0025] Example 1. Refer to Figure 1 、 Figure 4 And Figure 5As shown in the figure, an intelligent control method for the secondary network of a heating system based on demand-side regulation includes:

[0026] Step 1: Set the building regulation period of the heating system as T B ;

[0027] Step 2: Temperature acquisition and building regulation. Set the initial moment as moment 0. Whenever it is within the mT B period, perform temperature acquisition, building user classification, and building regulation. The building user classification is divided into subcooled buildings, overheated buildings, and suitable-temperature buildings through a clustering algorithm. The temperature acquisition is to collect the temperature data of the building users in the heating system during the stable period of the building;

[0028] During the (m + 1)T B period, execute the temperature range of the suitable-temperature buildings at the moment of mT B , reclassify the building users, and whenever at the moment of (m + 1)T B , perform heat source regulation under the conditions of Step 3, where m is an odd number greater than or equal to 1;

[0029] Step 3: Heat source regulation;

[0030] (1) When there are heating buildings that are classified as subcooled buildings at both the moment of the building regulation period mT B and the moment of (m + 1)T B , at this time, increase the heat supply of the heat source and increase the frequency of the heat source water pump;

[0031] (2) When there is a building marked as a subcooled building at the moment of mT B and a building marked as a suitable-temperature building at the moment of (m + 1)T B , and the opening degree of the marked valve is not 100% for the heating building, at this time, reduce the heat supply of the heat source and reduce the frequency of the heat source water pump;

[0032] (3) When there are no subcooled buildings at both the moment of mT B and the moment of (m + 1)T B , and there are no suitable-temperature buildings with a valve opening degree of 100% at the moment of (m + 1)T B , at this time, reduce the heat supply of the heat source and reduce the frequency of the heat source water pump; and so on, forming a cyclic control mode with the building regulation period nested with the heat source control period, and completing the demand-side regulation of the secondary network of the heating system. Under this regulation method, the heat source regulation period is twice the building regulation period. The heating system can greatly reduce the excessive heating of the system, improve the comfort level of users, and improve the overall operation efficiency of the heating system. This method makes up for the lack of demand-side regulation of the heating system, has clear regulation logic, accurate control, and has a wide range of potential applications. Among them Figure 4 and Figure 5 are Figure 1The up-down decomposition diagram.

[0033] Example 2: Using the same clustering algorithm as in Example 1, the building control and heat source control in the first building control cycle and the second building control cycle are described in detail:

[0034] During the operation of the entire heating system, in a certain building, the time period 1 hour after the regulation of the building thermal inlet is called the building stable period. The initial time is set as time 0, and at this time, all building users in the system are in the designed indoor temperature state. The building control cycle is T B , and the heat source control cycle T S = 2T B . After one building control cycle, the time is T B . At this time, according to the building temperature data collected by the system data center, the temperature data of the building users in the heating system during the building stable period [1, T B is collected, and the clustering algorithm is used to classify the building users into sub-cooled buildings, over-heated buildings and suitable-temperature buildings, and the temperature range of the suitable-temperature buildings is given. A sub-cooled building refers to a type of user whose building temperature is much lower than the target temperature after the heating system clusters the building user data. An over-heated user refers to a type of user whose building temperature is much higher than the target temperature after the heating system clusters the building user data. A suitable-temperature building refers to a type of user whose building temperature is close to the target temperature after the heating system clusters the building user data. It is determined according to the analysis after clustering. After clustering the buildings, at this moment, the sub-cooled buildings and over-heated buildings are adjusted by the regulating device at the building thermal inlet, and the suitable-temperature buildings are not regulated at all. After another building control cycle, the time is 2T B , according to the temperature range of the suitable-temperature buildings given by clustering at time T B in the first building control cycle, the clustering algorithm is used to classify the building users again. The building types are still sub-cooled buildings, over-heated buildings and suitable-temperature buildings. Those higher than the upper limit of the temperature range of the suitable-temperature buildings are over-heated buildings, and those lower than the lower limit of the temperature range of the suitable-temperature buildings are sub-cooled buildings;

[0035] (1). When there are heating users whose buildings are classified as sub-cooled buildings at both time T B and 2T B , at this time, increase the heat supply of the heat source and increase the frequency of the heat source water pump;

[0036] (2). There is a heating building marked as a sub-cooled building at time T B and marked as a suitable-temperature building at 2T B , and the valve opening of the marked building is not 100%, at this time, reduce the heat supply of the heat source and reduce the frequency of the heat source water pump;

[0037] (3) When T B and at the moment of 2T B there are no subcooled buildings, and at the moment of 2T B there is no suitable-temperature building with a building valve opening of 100%. At this time, reduce the heat supply of the heat source and lower the frequency of the heat source water pump.

[0038] And so on, in the (m + 1)T B period, execute the temperature range of the suitable-temperature building at the moment of mT B , reclassify the building users, and whenever at the moment of (m + 1)T B , execute the heat source regulation under the conditions of Step 3, where m is an odd number greater than or equal to 1; form a loop control mode with nested building regulation cycles and heat source control cycles to complete the demand-side regulation of the secondary network of the heating system.

[0039] Example 3: Using the same steps as in Example 1, classify the building users by the K-means clustering algorithm. There are n data points in a dataset. Randomly select k points from these n points as the initial clustering centers, and classify the objects with the closest Euclidean distance to the initial clustering centers; adopt an iterative method. During the iterative process, calculate and continuously update the values of the clustering centers of each class successively until reaching a stable state to obtain the best clustering effect, determine the clustering centers (both actual points and virtual points can be used as clustering centers), and make the sum of the Euclidean distances between each data in the dataset and its nearest clustering center the smallest. For the convenience of calculation, take the sum of the squares of the distances as the objective function, and its mathematical formula is expressed as:

[0040]

[0041] In the formula: v i represents the data of the i-th dimension of the actual point;

[0042] c j represents the j-th clustering point;

[0043] W n represents the sum of the Euclidean distances.

[0044] Furthermore, in the regulation process, the basic steps of classifying the building users by the K-means algorithm are as follows:

[0045] Input: A dataset of the room temperatures of building users in a heating system with n data;

[0046] Output: 3 clustering centers;

[0047] The first step: Arbitrarily select 3 data in the space of the temperature dataset as the initial clustering centers;

[0048] Step 2: Calculate the distances between the cluster centers and all data points, and assign the minimum distance among them.

[0049] Step 3: Calculate the mean value of all objects in the same cluster, and update and replace this result with the cluster center, representing the total number of data in the j-th class.

[0050] Step 4: Stop the calculation until all cluster centers no longer iterate and tend to be stable, that is, the iterative clustering function converges. Otherwise, return to Step 2 to continue the iterative calculation.

[0051] The control device for the building heat inlet regulates the opening of the building user valves for the subcooled building user group to increase, and the opening of the building user valves for the superheated building user group to decrease. Optionally, the building control cycle is 1.5 h.

[0052] Second Embodiment

[0053] Example 4: Based on the control schemes and steps of Examples 1 - 3, this example is carried out under the condition that a secondary network heating system includes 1 heat source and 100 heating buildings. At the initial state of the heating system at time 0, after one building control cycle, the time is T B At time, during the building stable period, the opening degrees of all building valves in the heating system and the average temperature of the buildings are shown in Table 1, and the user temperature distribution is as Figure 2 shown;

[0054] Table 1 Data related to building users in the heating system after the first building control cycle

[0055]

[0056]

[0057] At T B At time, through the K-means clustering analysis algorithm, after analysis, the clustering division results of all buildings in the system are as Figure 3 shown (from top to bottom in the figure are superheated buildings, suitable-temperature buildings, and subcooled buildings), and the building users are divided into superheated buildings, subcooled buildings, and suitable-temperature buildings. After clustering, the temperature range of the suitable-temperature buildings is [18.84 °C, 21.22 °C]. At the first building control cycle T B At time, the building users divided into superheated buildings and subcooled buildings are each subject to their respective control adjustments, where the superheated buildings reduce the valve opening degree, and the subcooled buildings increase the valve opening degree. For the suitable-temperature building users, the building valves do not operate.

[0058] After the building control ends, after the second building control cycle, the system time is then 2T B . According to T BClassify the system users again according to the temperature range of the suitable-temperature users at a certain moment. The average indoor temperature during the stable period of the building users, the valve opening of the building, and the classification of the building users at this moment are shown in Table 2.

[0059] Table 2 Information of system building users after the second regulation cycle

[0060]

[0061]

[0062]

[0063] It can be seen that after experiencing the second building regulation cycle, there are cases of building users that satisfy T B At the moment, it is marked as an over-cooled building, and at 2T B At the moment, it is marked as a heat supply building with suitable temperature, and the valve opening of the marked building is not 100%. The information of the heat supply buildings that meet this condition is shown in Table 3.

[0064] Table 3 Information of building users that meet the heat source regulation conditions after the second regulation cycle

[0065]

[0066]

[0067] Because, at 2T B At the moment, there is a heat supply building type in Table 3, so it is selected to reduce the heat output of the heat source, that is, to reduce the frequency of the heat source water pump.

[0068] The present invention has been disclosed above with preferred embodiments. However, it is not intended to limit the present invention. Any person skilled in the art, without departing from the scope of the technical solution of the present invention, can make some changes or modifications to the above-disclosed structure and technical content as equivalent embodiments of equivalent changes, and all still fall within the scope of the technical solution of the present invention.

Claims

1. An intelligent control method for the secondary network of a heating system based on demand-side regulation, characterized in that: Including: Step 1. Set the building control period of the heating system to be ; Step 2: Temperature acquisition and building control. Whenever in the cycle, temperature acquisition, building user classification, and building control are performed. The building user classification is divided into subcooled buildings, overheated buildings, and comfortable-temperature buildings through a clustering algorithm; The temperature collection is to collect the temperature data of building users in the heating system during the building's stable period. The building regulation is the heating of overcooled and overheated buildings, which is regulated by the regulating device at the building's thermal inlet. No regulation is carried out for buildings at the appropriate temperature. The stable period of the building during the building regulation cycle is ; During the cycle, perform the suitable temperature range for the building at the moment, reclassify the building users, and whenever is an odd number greater than or equal to 1; Step 3: Heat source regulation; (1) When there is a building regulation cycle moments and moments are both classified as heating buildings with supercooled buildings. At this time, increase the heat supply of the heat source and increase the frequency of the heat source water pump; (2)When there is a moment marked as a supercooled building, a moment marked as a thermally comfortable building, and the opening degree of the marked valve is not 100% for the heating building, at this time, reduce the heat supply of the heat source and lower the frequency of the heat source water pump; (3) When the moment and there is no subcooled building at the moment, and there is no suitable temperature building with a valve opening of 100% at the moment. At this time, reduce the heat supply of the heat source and lower the frequency of the heat source water pump; And so on, a cyclic control mode with nested building regulation cycles and heat source regulation cycles is formed to complete the demand-side regulation of the secondary network of the heating system. The heat source regulation cycle is twice the building regulation cycle.

2. The intelligent control method for the secondary network of a heating system based on demand-side regulation according to claim 1, characterized in that: Use the K-means algorithm to classify building users in the building user classification; In the formula: represents the data of the i-th dimension of the actual point; Denote the j-th clustering point; Indicates the sum of Euclidean distances.

3. The intelligent control method for the secondary network of a heating system based on demand-side regulation according to claim 1 or 2, characterized in that: The steps of the clustering algorithm are as follows: Input: A dataset of the room temperatures of building users in a heating system with n data; Output: 3 cluster centers; First step: Arbitrarily select 3 data in the space of the temperature dataset as the initial cluster centers; Second step: Calculate the distances between the cluster centers and all data points, and allocate them by taking the minimum of these distances; Third step: Calculate the mean of all objects in the same cluster, and update and replace this result with the cluster center, representing the total number of data in the j-th class; Fourth step: Stop the calculation until all cluster centers no longer iterate and tend to be stable, that is, the iterative clustering function converges, otherwise return to the second step to continue the iterative calculation.

4. The intelligent control method for the secondary network of a heating system based on demand-side regulation according to claim 1, characterized in that: The building users are classified into sub-cooled buildings, overheated buildings, and suitable-temperature buildings. Sub-cooled buildings are those users whose building temperatures are lower than the target temperature after the heating system clusters the building user data. Overheated users are those users whose building temperatures are much higher than the target temperature after the heating system clusters the building user data. Suitable-temperature buildings are those users whose building temperatures are close to the target temperature after the heating system clusters the building user data.

5. The intelligent control method for the secondary network of a heating system based on demand-side regulation according to claim 1, characterized in that: The regulation device at the building heat inlet regulates by opening the building user valves for the sub-cooled building user group and closing the building user valves for the overheated building user group.

6. The intelligent control method for the secondary network of a heating system based on demand-side regulation according to claim 5, characterized in that: The temperature range of a comfort building is .

Images