A method for energy-saving optimization of central heating system

Abstract

The present application relates to the field of heating system control technology, and specifically to an energy-saving optimization method for a centralized heating system, the method comprising: a heating data collection step, obtaining the heat load data of each heating end user under each heating station in the centralized heating system in a continuous time; a heating regulation feature extraction step, obtaining the heat load fluctuation value of each heating end user according to the fluctuation change difference of the heat load data of each heating end user in a local time; an energy-saving optimization control step, according to the overall distribution of the heat load fluctuation values ​​of all heating end users under each heating station, obtaining the heat load change amount to calculate the initial adjustment rate of each heating station; comprehensively combining the heat load fluctuation values ​​and heat load changes of all heating end users under each heating station to obtain an initial correction value to control the initial adjustment rate of each heating station and obtain the optimal adjustment rate. The present application improves the accuracy of the opening control of the electric proportional control valves of different heating stations.

Description

Technical Field

[0001] The present application relates to the technical field of heating system control, and in particular to an energy-saving optimization method for a centralized heating system. Background Art

[0002] Central heating is a heating method in which a heat source generates the heat energy required by heat users, and then supplies heat energy to cities or larger areas through a heat pipe network. The central heating system consists of three main parts: heat source, heat network, and heat users, as well as a primary network and a secondary network. The end users of the central heating system have the ability to adjust the temperature independently to change the heat load, so that the heat load data of the heat station to which the end users belong can change significantly in a short period of time. In the actual operation of the central heating system, there is a difference between the heat supply of the heat source to the heat station and its heat load, resulting in thermal imbalance and heat energy loss.

[0003] At present, when optimizing algorithms are often used to control the opening of electric proportional control valves in different thermal power stations, the initial position vector and velocity vector are randomly selected, ignoring the changing characteristics of the thermal load of the thermal power station. As a result, the obtained opening adjustment rate of the electric proportional control valve falls into the local optimum, causing inaccurate heating control of the heat source to the thermal power stations belonging to different end users. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides an energy-saving optimization method for a centralized heating system to solve the existing problems.

[0005] The energy-saving optimization method of a centralized heating system of the present application adopts the following technical solution:

[0006] The present application provides an energy-saving optimization method for a centralized heating system, the method comprising the following steps:

[0007] The heating data collection step is to obtain the heat load data of each heating end user under each heating station in the centralized heating system over a continuous period of time;

[0008] The step of extracting the characteristics of heat supply regulation is to obtain the heat load dispersion of each heat supply end user according to the fluctuation change of heat load data of each heat supply end user in the local time; and to obtain the heat load fluctuation value of each heat supply end user according to the fluctuation difference and heat load dispersion of heat load data of different heat supply end users under the same heat power station;

[0009] The energy-saving optimization control step is to predict the heat load data of each heating end user according to the overall distribution of the heat load fluctuation value of all heating end users under each heating station, and obtain the heat load change by combining the difference between the predicted results of all heating end users and the heat load data at the current moment, and calculate the initial adjustment rate of the opening of the electric proportional control valve of each heating station; the initial correction value is obtained by combining the heat load fluctuation value and the heat load change of all heating end users under each heating station; the initial adjustment rate of each heating station is controlled according to the initial correction value, and the optimal adjustment rate of the opening of the electric proportional control valve of each heating station is obtained.

[0010] Furthermore, the heat load dispersion of each heating end user is obtained according to the fluctuation of the heat load data of each heating end user in the local time, including:

[0011] For each heat supply terminal user, the ratio of the standard deviation of each heat load data and the previous preset number of heat load data to the mean value is used as the short-term fluctuation degree of each heat load data;

[0012] The absolute value of the difference between each heat load data and the previous heat load data is recorded as the heat load differential value of each heat load data;

[0013] The short-term fluctuation degree of all heat load data and the heat load differential value are comprehensively considered to calculate the heat load dispersion of the heating end users.

[0014] Further, the short-term fluctuation degree of all heat load data and the heat load differential value are integrated to calculate the heat load dispersion of the heat supply end user, including: recording the heat load dispersion of the heat supply end user as , ; Wherein, L represents the number of heat load data; Indicates the short-term fluctuation degree of the i-th heat load data; Indicates the heat load differential value of the i-th heat load data.

[0015] Furthermore, the heat load fluctuation value of each heating end user is obtained according to the fluctuation difference of heat load data of different heating end users under the same heating station and the heat load dispersion, including:

[0016] Obtain a sequence of normalized values ​​of all heat load data of each heat supply terminal user in chronological order, which is recorded as a heat load periodic fluctuation sequence;

[0017] According to the fluctuation difference of heat load period fluctuation sequence of different heating end users under the same heating station, the same type of users of each heating end user are obtained;

[0018] The heat load fluctuation value is calculated based on the difference between the heat load period fluctuation sequence of each heating end user and all the users of the same type and the heat load dispersion.

[0019] Furthermore, the method for obtaining users of the same type as each heating terminal user includes:

[0020] For any heating end user under the same heating station, calculate the difference between the heat load period fluctuation sequence of any heating end user and all other heating end users; perform threshold analysis on all the differences to obtain a difference threshold; and regard the heating end users whose difference threshold is less than the difference threshold as the same type of users as any heating end user.

[0021] Furthermore, the heat load fluctuation value is calculated according to the difference between the heat load period fluctuation sequence of each heat supply end user and all the users of the same type and the heat load dispersion, including:

[0022]

[0023] in, It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station in the central heating system; It represents the heat load dispersion of the nth heating terminal user under the mth heating station; It represents the number of users of the same type as the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the rth user of the same type as the nth heating terminal user under the mth heating station; Represents the heat load cycle fluctuation sequence and The DTW distance between them; A preset constant greater than 0.

[0024] Furthermore, the method for obtaining the initial adjustment rate of the opening of the electric proportional control valve of each thermal power station includes:

[0025] For each heating station, the minimum value of the heat load fluctuation value of all heating end users under the heating station is obtained; the ratio of the heat load fluctuation value of each heating end user to the minimum value is calculated; the integer value of the product of the preset initial sliding window length and the ratio is used as the sliding window length of each heating end user;

[0026] According to the overall proportion of heat load fluctuation values ​​of each heating end user, the polynomial fitting number of each heating end user is obtained;

[0027] Substituting the sliding window length and polynomial fitting times of each heating end user into the polynomial fitting algorithm, fitting all heat load data of each heating end user to obtain a polynomial fitting curve; the horizontal axis of the polynomial fitting curve is time, and the vertical axis is heat load data;

[0028] The heat load data corresponding to the next moment of the current moment on the polynomial fitting curve is obtained and recorded as the heat load prediction data; the absolute value of the difference between the heat load data of the heating end user and the heat load prediction data is recorded as the heat load change of the heating end user; the mean of the heat load change of all heating end users under the thermal power station is calculated, and the arc tangent normalized value of the mean is recorded as the initial adjustment rate of the thermal power station to the opening of the electric proportional control valve.

[0029] Furthermore, the method for obtaining the polynomial fitting times of each heating terminal user includes: recording the polynomial fitting times of the nth heating terminal user under the mth heating station as , ;in, Indicates the preset initial polynomial fitting times; and They represent the minimum and maximum values ​​of heat load fluctuation of all heating end users under the mth heating station respectively; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; is the floor function.

[0030] Furthermore, the method for obtaining the initial correction value includes:

[0031]

[0032] Wherein, C represents the preset correction limit value; norm() represents the normalization function; It represents the number of heating end users under the mth heating station in the centralized heating system; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; It represents the change in heat load of the nth heating terminal user under the mth heating station; Represents the initial correction value of the mth heating station in the central heating system.

[0033] Furthermore, the method for obtaining the optimal adjustment rate of the opening of the electric proportional control valve of each thermal power station includes:

[0034] A first preset number of values ​​are randomly selected within an interval with the initial adjustment rate of the opening of the electric proportional control valve of each thermal power station as the center and the initial correction value as the radius as the update adjustment rate of each thermal power station;

[0035] A vector composed of any update adjustment rate of all thermal power stations in a fixed order is recorded as an initial adjustment vector; an optimization algorithm is used for all initial adjustment vectors to obtain an optimal adjustment vector; wherein each element in the optimal adjustment vector is the optimal adjustment rate of each thermal power station to the opening of the electric proportional control valve.

[0036] This application has at least the following beneficial effects:

[0037] The present application analyzes the influence of fluctuations in heat load data on the accuracy of heating control, calculates the heat load dispersion of the heat load data sequence according to the short-term fluctuation degree of the heat load data and the heat load differential value, and calculates the short-term fluctuation degree by establishing a short-term fluctuation window, thereby enhancing the ability to extract information on heat load data changes, and thus improving the accuracy of subsequent heating system control; further calculates the heat load fluctuation value according to the heat load dispersion, and classifies different end users under the same thermal power station so that different heat load change information complements each other, reduces the deviation in the judgment of the heat load change trend of the end users, and improves the reliability of subsequent heating system control; fits the heat load trend data according to the heat load fluctuation value of the end user, calculates the initial adjustment rate and the initial correction value, and adopts the particle swarm algorithm to obtain the optimal adjustment rate of the opening of the electric proportional control valve of different thermal power stations, completes the energy-saving optimization control of the centralized heating system, and improves the accuracy of the opening control of the electric proportional control valve of different thermal power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 A flowchart of the steps of an energy-saving optimization method for a central heating system provided in this application;

[0040] Figure 2 A comparison chart of heat load data of different end users on the same day provided for this application;

[0041] Figure 3 It is a schematic diagram of the fitting curve of the original polynomial fitting;

[0042] Figure 4 This is the effect diagram of the polynomial fitting curve of this application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0045] The specific scheme of the energy-saving optimization method for a centralized heating system provided by the present application is described in detail below with reference to the accompanying drawings.

[0046] An embodiment of the present application provides an energy-saving optimization method for a centralized heating system. Specifically, the following energy-saving optimization method for a centralized heating system is provided. Figure 1 , the method comprises the following steps:

[0047] S1, obtaining the heat load data of each heating end user under each heating station in the centralized heating system over a continuous period of time.

[0048] This embodiment obtains the heat load data of the end user by measuring with an ultrasonic heat meter installed at each end user of each heating station in the heating system. The frequency of the ultrasonic heat meter measuring the heat load data of the end user is once an hour. During the data collection process, failures in measurement and transmission equipment will cause abnormalities in the collected heat load data. In order to improve the accuracy of subsequent control of the centralized heating system, the standard deviation of all heat load data is calculated, and the abnormal heat load data is eliminated according to the Raida criterion, and the heat load data at the eliminated position is supplemented by linear interpolation to complete the preprocessing of the heat load data.

[0049] S2, according to the fluctuation of heat load data of each heating end user in the local time, obtain the heat load dispersion of each heating end user; according to the fluctuation difference and heat load dispersion of heat load data of different heating end users under the same thermal power station, obtain the heat load fluctuation value of each heating end user.

[0050] The central heating system is a complex multivariable nonlinear system. The change of the heat load of the end user is affected by many factors and has strong nonlinear characteristics, which makes the heat load distribution of the whole system change dynamically, resulting in thermal imbalance of the heating system. Accurate heating system control can reduce the difference between the heat supply and the heat load, improve the utilization rate of heat energy, and reduce the heat energy loss of the central heating system. Figure 2As shown in the figure, the heat load data comparison of different end users on the same day. Compared with the traditional heating system, the central heating system has the ability for end users to adjust the temperature and change the heat load. The heat load data of the heat station to which the end user belongs can change greatly in a short period of time, so it is necessary to adjust the control of the central heating system in a timely manner.

[0051] Specifically, the ratio of the standard deviation of each heat load data and the preset number of heat load data before it to the mean is used as the short-term fluctuation degree of each heat load data. In this embodiment, each heat load data is first taken as the end data, and then U-1 heat load data are taken forward to form a short-term fluctuation window of the heat load data. The heat load data with less than U data at the front end are supplemented by linear interpolation. In this embodiment, the value of U is 5. Due to the different scales of different end users, the measurement scales of their heat loads vary greatly. Therefore, the ratio of the standard deviation of the short-term fluctuation window to the mean is recorded as the short-term fluctuation degree of the data at the end of the window, indicating the short-term relative fluctuation degree before the heat load data, so as to obtain the same type of end users in the future and obtain more heat load change information in the control of the heating system.

[0052] Affected by factors such as indoor and outdoor temperature and wind speed, the heat load data of end users has a similar distribution trend over a long period of time, which can be used as an important basis for the subsequent calculation of the change trend of the heat load of end users. First, the continuously collected heat load data is processed in segments, and the heat load data obtained by each L measurement is analyzed. In this embodiment, the value of L is 24. The absolute value of the difference between each heat load data and its previous heat load data is recorded as the heat load differential value of each heat load data. The heat load dispersion of each heating end user is calculated by combining the short-term fluctuation degree of all heat load data and the heat load differential value. In this embodiment, the calculation formula for heat load dispersion is:

[0053]

[0054] Wherein, L represents the number of heat load data; Indicates the short-term fluctuation degree of the i-th heat load data; represents the heat load difference value of the ith heat load data; It represents the heat load dispersion of the heating end user, reflecting the degree of dispersion of the heat load data distribution of the end user in the previous L hours.

[0055] The larger the heat load differential value in the heat load data sequence, the greater the difference between adjacent heat load data, the greater the discrete degree of heat load data distribution, and the greater the heat load discreteness of the heat load data sequence; at the same time, the greater the short-term fluctuation degree of the heat load data, the greater the short-term fluctuation degree of the heat load data before this data, and the greater the calculated heat load discreteness, indicating that the overall distribution of the heat load data sequence is relatively discrete. By establishing a short-term fluctuation window to calculate the short-term fluctuation degree, the ability to extract heat load data change information is enhanced, thereby improving the accuracy of subsequent heating system control.

[0056] In real life, the heat load data of some end users under the same heat station have similar change trends. The reliability of calculating the heat load change trend only from the heat load data of a single end user is low. The heat load of places such as schools and office buildings is large during working hours on weekdays, while the heat load of residential buildings is small during the same period. In addition, due to the similar construction structure of some buildings, their heat loads are similarly affected by the environment. Therefore, the heat load data of some end users under the same heat station can be supplemented to each other, reducing the deviation in the judgment of the heat load change trend of the heating end users and improving the reliability of the control of the heating system.

[0057] Due to the different functions of different buildings, the heat load changes of different heating end users have a long cycle. Based on the above analysis, in this embodiment, a sequence consisting of the normalized values ​​of all heat load data of each heating end user in chronological order is obtained, which is recorded as a heat load cycle fluctuation sequence; based on the fluctuation difference of the heat load cycle fluctuation sequence of different heating end users under the same thermal power station, the same type of users of each heating end user are obtained.

[0058] Specifically, for any heating end user under the same heating station, calculate the difference between the heat load periodic fluctuation sequence of any heating end user and all other heating end users; perform threshold analysis on all the differences to obtain a difference threshold; and regard the heating end users whose difference threshold is less than the difference threshold as the same type of users as any heating end user. The difference between two sequences indicates the degree of difference in the change trends of the two sequences. In this embodiment, the DTW distance is used to calculate the difference between the heat load periodic fluctuation sequences. The maximum inter-class variance algorithm is used for threshold analysis. Users of the same type indicate that they have similar heat load change trends over a longer period of time.

[0059] The heat load data of the heating end users in the same time period of a day can be used as a reference for calculating the change trend of their current heat load data. The heat load fluctuation value is calculated based on the heat load dispersion and the difference between the heat load cycle fluctuation sequences of the same type of users:

[0060]

[0061] in, It represents the heat load dispersion of the nth heating end user under the mth heating station in the central heating system; It represents the number of users of the same type as the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the rth user of the same type as the nth heating terminal user under the mth heating station; Represents the heat load cycle fluctuation sequence and The DTW distance between them; is a preset constant greater than 0, the purpose of which is to prevent the denominator from being zero. In this embodiment, the value is 1; It represents the heat load fluctuation value of the nth end user under the mth heating station in the centralized heating system, reflecting the degree of fluctuation of the heat load of the end user in the future.

[0062] The greater the degree of dispersion of the heat load data distribution of the end user before, the greater the dispersion of its heat load, the greater the probability of a large degree of fluctuation of its heat load data in the future, and the greater the calculated heat load fluctuation value. In addition, the smaller the DTW distance between the heat load period fluctuation sequence of the end user and its same type of users, the closer the trend of the heat load changes of the two, the higher the reference degree for supplementing the heat load data change information of the end user, and the greater the subsequent adjustment of the control of the heating system. By classifying different end users under the same thermal power station, different heat load change information can complement each other, reduce the deviation in the judgment of the heat load change trend of the end user, and improve the reliability of the subsequent control of the heating system.

[0063] S3, based on the overall distribution of heat load fluctuation values ​​of all heating end users under each heating station, predict the heat load data of each heating end user, combine the difference between the predicted results of all heating end users and the heat load data at the current moment to obtain the heat load change, and calculate the initial adjustment rate of the opening of the electric proportional control valve of each heating station; comprehensively consider the heat load fluctuation values ​​and heat load changes of all heating end users under each heating station to obtain the initial correction value; control the initial adjustment rate of each heating station according to the initial correction value, and obtain the optimal adjustment rate of the opening of the electric proportional control valve of each heating station.

[0064] Since the change of heat load data of heating end users has a certain trend in a short period of time, more accurate heat load data of end users can be obtained by adjusting the sliding window length and polynomial fitting times of the moving polynomial least squares method. According to the heat load fluctuation value of the end user, the sliding window length and polynomial fitting times are calculated:

[0065]

[0066]

[0067] Where K represents the preset initial sliding window length, represents the preset initial polynomial fitting times. In this embodiment, K and The values ​​are 5 and 2 respectively; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; and They represent the minimum and maximum values ​​of heat load fluctuation of all heating end users under the mth heating station respectively; Indicates the sliding window length of the nth heating end user under the mth heating station; It represents the number of polynomial fittings for the nth heating terminal user under the mth heating station; is the floor function.

[0068] Substitute the sliding window length and polynomial fitting times of each heating end user into the polynomial fitting algorithm, fit all the heat load data of each heating end user, and obtain a polynomial fitting curve; the horizontal axis of the polynomial fitting curve is time, and the vertical axis is heat load data; obtain the heat load data corresponding to the next moment of the current moment on the polynomial fitting curve, and record it as heat load prediction data. It represents the next heat load data of the end user under the current heat load change trend.

[0069] When the fluctuation degree of the heat load data of the end user is greater, the heat load fluctuation value is greater. At this time, a larger sliding window length and polynomial fitting times are set to improve the accuracy of data fitting; on the contrary, the smaller the heat load fluctuation value is, the smaller the sliding window length and polynomial fitting times are set to avoid overfitting of the heat load data, improve the accuracy of the calculation of the heat load trend data, and avoid aggravating the thermal imbalance and heat energy loss in the control of the heating system.

[0070] like Figure 3 The figure shows the fitting curve of the original polynomial fitting. Figure 4 This is the polynomial fitting curve of the present application, the horizontal axis is the sequence number of the moment, and the vertical axis is the load value of the heat load data; by comparison, the polynomial fitting curve obtained by the sliding window length and the number of polynomial fitting times set in the present application can better reflect the changes in the heat load data.

[0071] The degree of heat load fluctuation of end users under different thermal power stations is inconsistent, the heat load changes of each thermal power station are different, and the optimal adjustment rate of the heat source to the opening of the electric proportional control valve of each thermal power station is different.

[0072] The absolute value of the difference between the heat load data of the heating end user and the heat load forecast data is recorded as the heat load change of the heating end user. Further, the mean of the heat load change of all heating end users under the heating station is calculated, and the arc tangent normalization is performed on it, and the normalization result is recorded as the initial adjustment rate of the heat source to the opening of the electric proportional control valve of the heating station, wherein the arc tangent normalization is a well-known technology and will not be repeated here.

[0073] In order to improve the accuracy of the calculation of the opening adjustment rate of the electric proportional control valve of different thermal power stations for heat sources, the initial correction value of the initial opening adjustment rate is calculated according to the heat load fluctuation value:

[0074]

[0075] Wherein, C represents the correction limit value, which is 0.1 in this embodiment and is used to limit the value range of the initial correction value; norm() represents the Z-Score standard normalization function; It represents the number of heating end users under the mth heating station in the centralized heating system; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; It represents the change in heat load of the nth heating terminal user under the mth heating station; It represents the initial correction value of the mth heating station in the central heating system, and is used to correct its initial adjustment rate so as to more accurately control the opening of its electric proportional control valve.

[0076] The greater the absolute value of the heat load fluctuation value and the heat load change of the end-user under the thermal power station, the greater the degree of change of the heat load of the thermal power station. A larger correction range is set for the adjustment rate of the opening of its electric proportional control valve to avoid falling into the local optimum in subsequent iterations, so as to reduce the difference between the heat load of the thermal power station and the heat supply of the heat source to it, and improve the energy-saving effect of heating.

[0077] A first preset number of values ​​are randomly selected within an interval with the initial correction value as the radius and the initial adjustment rate of the opening of the electric proportional control valve of each thermal power station as the center, as the update adjustment rate of each thermal power station; in this embodiment, the value of the first preset number is 30.

[0078] A vector composed of any update adjustment rate of all thermal power stations in a fixed order is recorded as the initial adjustment vector; an optimization algorithm is used for all initial adjustment vectors to obtain the optimal adjustment vector; wherein each element in the optimal adjustment vector is the optimal adjustment rate of each thermal power station to the opening of the electric proportional control valve. The initial adjustment vector contains an update adjustment rate for each thermal power station, and the number is the number of permutations and combinations of all update adjustment rates of all thermal power stations. It should be noted that the fixed order indicates that the update adjustment rate at each position in different initial adjustment vectors corresponds to the same thermal power station.

[0079] Specifically, this embodiment adopts the particle swarm optimization algorithm, and the initial correction values ​​of each thermal power station are combined into an initial correction vector as the velocity vector of the particle swarm algorithm position vector. In addition, through the centralized heating system model established by the Flowmaster software, the heating supply of the thermal power station to each heating terminal user under the current heat source opening adjustment rate of the thermal power station is obtained, and the absolute value of the difference between the heating supply and the heat load trend data is calculated, which is recorded as the heat load imbalance, and the sum of the heat load imbalances of all terminal users of the centralized heating system is recorded as the fitness function of the position vector. The maximum number of iterations of the particle swarm algorithm is set to 20, and the optimal adjustment vector is output, and its various components are recorded as the optimal adjustment rate of the opening of the corresponding thermal power station electric proportional control valve, so as to complete the energy-saving optimization control of the centralized heating system.

[0080] Through the above description of the implementation method in combination with the accompanying drawings, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0081] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. A method for energy saving optimization of a central heating system, characterized in that: The method comprises the following steps: The heating data collection step is to obtain the heat load data of each heating end user under each heating station in the centralized heating system over a continuous period of time; The step of extracting the characteristics of heat supply regulation is to obtain the heat load dispersion of each heat supply end user according to the fluctuation change of heat load data of each heat supply end user in the local time; and to obtain the heat load fluctuation value of each heat supply end user according to the fluctuation difference and heat load dispersion of heat load data of different heat supply end users under the same heat power station; The energy-saving optimization control step is to predict the heat load data of each heating end user according to the overall distribution of the heat load fluctuation values ​​of all heating end users under each heating station, and obtain the heat load change by combining the difference between the predicted results of all heating end users and the heat load data at the current moment, and calculate the initial adjustment rate of each heating station for the opening of the electric proportional control valve; the heat load fluctuation values ​​and heat load changes of all heating end users under each heating station are combined to obtain the initial correction value; the initial adjustment rate of each heating station is controlled according to the initial correction value to obtain the optimal adjustment rate of each heating station for the opening of the electric proportional control valve; The heat load fluctuation value of each heating end user is obtained according to the fluctuation difference of heat load data of different heating end users under the same heating station and the heat load dispersion, including: Obtain a sequence of normalized values ​​of all heat load data of each heat supply terminal user in chronological order, which is recorded as a heat load periodic fluctuation sequence; According to the fluctuation difference of heat load period fluctuation sequence of different heating end users under the same heating station, the same type of users of each heating end user are obtained; The heat load fluctuation value is calculated based on the difference between the heat load period fluctuation sequence of each heat supply end user and all the users of the same type and the heat load dispersion; The method for obtaining the initial adjustment rate of the opening of the electric proportional control valve of each thermal power station includes: For each heating station, the minimum value of the heat load fluctuation value of all heating end users under the heating station is obtained; the ratio of the heat load fluctuation value of each heating end user to the minimum value is calculated; the integer value of the product of the preset initial sliding window length and the ratio is used as the sliding window length of each heating end user; According to the overall proportion of heat load fluctuation values ​​of each heating end user, the polynomial fitting number of each heating end user is obtained; Substituting the sliding window length and polynomial fitting times of each heating end user into the polynomial fitting algorithm, fitting all heat load data of each heating end user to obtain a polynomial fitting curve; the horizontal axis of the polynomial fitting curve is time, and the vertical axis is heat load data; Obtain the heat load data corresponding to the next moment of the current moment on the polynomial fitting curve, and record it as the heat load prediction data; record the absolute value of the difference between the heat load data of the heating end user and the heat load prediction data as the heat load change of the heating end user; calculate the mean of the heat load change of all heating end users under the heating station, and record the arc tangent normalized value of the mean as the initial adjustment rate of the heating station to the opening of the electric proportional control valve; The method for obtaining the optimal adjustment rate of the opening of the electric proportional control valve of each thermal power station includes: A first preset number of values ​​are randomly selected within an interval with the initial adjustment rate of the opening of the electric proportional control valve of each thermal power station as the center and the initial correction value as the radius as the update adjustment rate of each thermal power station; A vector composed of any update adjustment rate of all thermal power stations in a fixed order is recorded as an initial adjustment vector; an optimization algorithm is used for all initial adjustment vectors to obtain an optimal adjustment vector; wherein each element in the optimal adjustment vector is the optimal adjustment rate of each thermal power station to the opening of the electric proportional control valve.

2. The energy-saving optimization method for a central heating system according to claim 1, characterized in that: The heat load dispersion of each heating end user is obtained according to the fluctuation of the heat load data of each heating end user in the local time, including: For each heat supply terminal user, the ratio of the standard deviation of each heat load data and the previous preset number of heat load data to the mean value is used as the short-term fluctuation degree of each heat load data; The absolute value of the difference between each heat load data and the previous heat load data is recorded as the heat load differential value of each heat load data; The short-term fluctuation degree of all heat load data and the heat load differential value are comprehensively considered to calculate the heat load dispersion of the heating end users.

3. The energy-saving optimization method for a centralized heating system according to claim 2, characterized in that: The heat load dispersion of the heating end user is calculated by integrating the short-term fluctuation degree of all heat load data and the heat load differential value, including: recording the heat load dispersion of the heating end user as , ; Wherein, L represents the number of heat load data; Indicates the short-term fluctuation degree of the i-th heat load data; Indicates the heat load differential value of the i-th heat load data.

4. The energy-saving optimization method for a centralized heating system according to claim 1, characterized in that: The method for obtaining users of the same type as each heating terminal user comprises: For any heating end user under the same heating station, calculate the difference between the heat load period fluctuation sequence of any heating end user and all other heating end users; perform threshold analysis on all the differences to obtain a difference threshold; and regard the heating end users whose difference threshold is less than the difference threshold as the same type of users as any heating end user.

5. The energy-saving optimization method for a centralized heating system according to claim 1, characterized in that: The heat load fluctuation value is calculated according to the difference between the heat load period fluctuation sequence of each heat supply end user and all the users of the same type and the heat load dispersion, including: in, It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station in the central heating system; It represents the heat load dispersion of the nth heating terminal user under the mth heating station; It represents the number of users of the same type as the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the nth heating terminal user under the mth heating station; It represents the heat load period fluctuation sequence of the rth user of the same type as the nth heating terminal user under the mth heating station; Represents the heat load cycle fluctuation sequence and The DTW distance between them; A preset constant greater than 0.

6. The energy-saving optimization method for a centralized heating system according to claim 1, characterized in that: The method for obtaining the polynomial fitting times of each heating terminal user comprises: recording the polynomial fitting times of the nth heating terminal user under the mth heating station as , ;in, Indicates the preset initial polynomial fitting times; and They represent the minimum and maximum values ​​of heat load fluctuation of all heating end users under the mth heating station respectively; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; is the floor function.

7. The energy-saving optimization method for a centralized heating system according to claim 1, characterized in that: The method for obtaining the initial correction value includes: Wherein, C represents the preset correction limit value; norm() represents the normalization function; It represents the number of heating end users under the mth heating station in the centralized heating system; It represents the heat load fluctuation value of the nth heating terminal user under the mth heating station; It represents the change in heat load of the nth heating terminal user under the mth heating station; Represents the initial correction value of the mth heating station in the central heating system.

Images