A heat supply subroom balance regulation method and system based on an internet of things

Abstract

The present application relates to the technical field of heating regulation, in particular to a heating room balance regulation method and system based on Internet of Things, comprising collecting temperature change data of each floor by setting a reference test temperature, calculating heat exchange efficiency decay coefficients of each floor according to temperature rise time and steady state temperature difference, and quantifying the influence of aging degree of heat dissipation equipment on heating effect. Then, the heating regulation period is divided into multiple time nodes according to preset intervals, and a differentiated constant temperature valve regulation scheme is generated for each floor in combination with the decay coefficients and time weight factors to form a complete dynamic regulation sequence. Genetic algorithm is used to optimize and select multiple regulation sequences, and the optimal regulation parameter combination is obtained through fitness function evaluation, selection, crossover and mutation operations. The present application solves the uneven heating problem of traditional old and scaled vertical single pipe series heating mode, in which the top floor residents need to open the window to dissipate heat due to overheating, and the bottom floor residents need additional electric heating equipment due to insufficient heating.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heating regulation, in particular to a heating room balance regulation method and system based on Internet of Things. BACKGROUND

[0002] In the cold winter in the northern region, municipal central heating is mostly used, which is based on the principle of centralized boiler steam, and the steam is transported to each community through the heat preservation pipeline, and most of the communities belong to multi-storey apartment buildings. The existing old low-rise apartment building heating adopts vertical single pipe series, and the heating steam flow direction is uplink and downlink, that is, from the highest floor of the roof to the bottom floor along the single pipe series. The heating flow of each floor is fixed, and can only flow from the upper layer to the next layer, and it is almost impossible to regulate and control the heating room. This leads to the phenomenon that the top layer of the room is overheated and the window is opened for heat dissipation, while the bottom layer of the room is insufficiently heated, and the heating municipal department has to increase the steam flow of each community in order to meet the heating requirements of the bottom layer of the room. However, the phenomenon of overheating and window opening for heat dissipation of the room of the top layer of the apartment is more obvious, and relevant research shows that the waste of steam heat of this kind of heating is 20-30%. If the traditional apartment heating method is completely changed to the existing on-demand heating, the cost is too high, and the simplest way is to connect the single pipe of the heating pipeline, and to increase the constant temperature valve of each household as an independent heating room. This solves the problem that the front group of radiators flows through the back group of radiators in series, and cannot be adjusted and controlled. However, the use of constant temperature valve alone cannot solve the problem of steam heating balance of each heating room, for example, even if the temperature of the constant temperature valve in the same building is consistent, there will still be the problem of preferential heating of the top layer of the room. Secondly, because the traditional pipeline and radiators are mostly cast iron, with the increase of service life, the flow path of the radiator will become smaller and smaller, and the heat dissipation effect will become worse and worse, and the adjustment of the constant temperature valve will also affect the hysteresis of temperature regulation. SUMMARY

[0003] (1) Technical problem to be solved

[0004] The purpose of the present application is to provide a heating room balance regulation method and system based on Internet of Things, in order to solve the problem of uneven heating of the top layer of the room needing to open the window for heat dissipation and the bottom layer of the room needing additional electric heating equipment in the traditional old and fouled vertical single pipe series heating mode.

[0005] (2) Technical scheme

[0006] In order to achieve the above purpose, on the one hand, the present application provides a heating room balance regulation method based on Internet of Things, which comprises:

[0007] The steam inlet thermostat valve of each floor position in the vertical single-pipe cross-connection series heating system is set to a reference test temperature value, temperature change data of each floor position is continuously collected by the indoor temperature sensor to establish a temperature data sequence of each floor position at the reference test temperature value over time, and a heat exchange efficiency decay coefficient of each floor position is calculated according to the temperature data sequence of each floor position;

[0008] The heating control period is divided into multiple control time nodes according to a preset control interval length, a thermostat control scheme is generated for all floor positions for each control time node, each thermostat control scheme includes a set temperature value of the steam inlet thermostat valve of each floor position at the time node and an execution time length value of the set temperature, a thermostat set temperature value of each floor position is calculated according to the heat exchange efficiency decay coefficient and the target temperature difference, and the thermostat control schemes of each control time node are combined in time sequence to form a complete dynamic control sequence;

[0009] Multiple different complete dynamic control sequences are used as initial populations for genetic algorithm optimization, each complete dynamic control sequence includes a thermostat set temperature value and an execution time length value of each floor position at each control time node, the combination of the thermostat set temperature value and the execution time length value of each floor position is optimized by the genetic algorithm, so that the final indoor temperature value measured by the indoor temperature sensor of each floor position after executing the complete dynamic control sequence reaches the target indoor temperature value, and the thermostat set temperature value and the execution time length value of each floor position at each control time node in the optimal complete dynamic control sequence optimized by the genetic algorithm are downloaded to the thermostat valve of each floor position in time through the Internet of Things.

[0010] Further, the method of continuously collecting temperature change data by the indoor temperature sensor of each floor position to establish a temperature data sequence of each floor position at the reference test temperature value over time, and calculating a heat exchange efficiency decay coefficient of each floor position according to the temperature data sequence of each floor position includes:

[0011] The reference test temperature value is set to , the target indoor temperature value is , and the preset minimum error threshold is ;

[0012] Temperature data of each floor position is continuously collected by the indoor temperature sensor of each floor position within a preset measurement time period to form a temperature data sequence of the first floor position , wherein represents a temperature value of the first floor position at the first time sampling point, represents the total number of sampling points, and the first Each floor location is based on the set benchmark test temperature value. The length of time required for the temperature to rise from the start to the steady state temperature ;

[0013] When the The indoor temperature sensor readings at each floor location showed a variation less than a preset stability threshold within a set continuous time period. When, the average temperature value within a set continuous time period is recorded as the first value. Steady-state temperature values ​​at each floor location ;No. Heat exchange efficiency attenuation coefficient at each floor location for:

[0014] ;

[0015] in This indicates the time required for the temperature to rise at the first floor location and serves as a reference value for temperature increase. This represents the steady-state temperature value at the first floor location and serves as a steady-state reference value; when The absolute value is less than the preset minimum error threshold. hour, Set to 1.

[0016] Furthermore, the record number Each floor location is based on the set benchmark test temperature value. The length of time required for the temperature to rise from the start to the steady state temperature The methods include:

[0017] In the Temperature data sequence at each floor location The thermostatic valve is set to the reference test temperature value. The start time point ; through continuous monitoring of the first Indoor temperature sensor data at each floor location; when the temperature value changes by less than a preset stability threshold over a continuous period of time. Determine the time point when steady state is achieved. ;

[0018] Calculate the first The heating time at each floor location When the first The heating time at each floor location Exceeding the preset maximum test time threshold At that time, the heating time length Set as the preset maximum test time threshold .

[0019] Furthermore, the aforementioned when the first The indoor temperature sensor readings at each floor location showed a variation less than a preset stability threshold within a set continuous time period. When, the average temperature value within a set continuous time period is recorded as the first value. Steady-state temperature values ​​at each floor location The methods include:

[0020] In the Temperature data sequence at each floor location From the middle A set of time sampling points are used as the starting detection point numbers to begin detecting the temperature change amplitude within a continuous time period, where the minimum length of the continuous time period is set to be... ; Calculate from the first Starting from each time sampling point, the continuous The maximum temperature within each time sampling point and minimum temperature ,in Greater than or equal to the minimum length of a continuous time period Divide by the minimum number of sampling points obtained by the sampling time interval; calculate the temperature change amplitude within this continuous time period. ;

[0021] When the temperature change range Less than the preset stability threshold At that time, the first time in a continuous time period Steady-state temperature values ​​at each floor location for:

[0022] ;

[0023] When the temperature change range Greater than or equal to the preset stability threshold At that time, the starting detection point number will be determined. Increase by 1 and repeat the above temperature change amplitude detection process until a continuous time period that meets the stability condition is found.

[0024] Furthermore, the method of dividing the heating control cycle into multiple control time nodes according to a preset control interval, and generating a thermostatic valve control scheme for all floors at each control time node includes:

[0025] The total duration of the heating regulation cycle is set as follows: The preset control interval is The total duration of the heating regulation cycle is divided into equal parts according to the preset regulation interval to obtain the number of regulation time nodes. ;

[0026] Regarding the first The control time node generates a thermostatic valve control scheme, and the control scheme includes the set temperature value and the execution duration value of the steam inlet thermostatic valve at each floor position at the first control time node. The control time node calculates the time weight coefficient of the first control time node .

[0027] The set temperature value of the thermostatic valve at the first floor position at the first control time node is:

[0028]

[0029] The execution duration value of the thermostatic valve at the first floor position at the first control time node is set as the preset control interval duration. The thermostatic valve control schemes of the control time nodes are arranged and combined in time sequence to form a complete dynamic control sequence including the set temperature value and the execution duration value of all floor positions at all control time nodes.

[0030] Further, the method of performing genetic algorithm optimization on the multiple different complete dynamic control sequences as initial population includes:

[0031] The initial population includes complete dynamic control sequence individuals, and each complete dynamic control sequence individual includes the set temperature value and the execution duration value of the thermostatic valve at each control time node at each floor position. The first complete dynamic control sequence individual

[0032] After the control sequence is sequentially issued to the thermostatic valves at each floor position for execution, the final indoor temperature value of the indoor temperature sensor at each floor position at the end of the control period is collected The fitness function value of the first complete dynamic control sequence individual is calculated as:

[0033]

[0034] Wherein, n represents the total number of floors; and the initial population is sorted in ascending order according to the fitness function value The first complete dynamic control sequence individuals with the smallest fitness function value are selected as the winning individuals, and the number of winning individuals is n. ​​​​​​​​​​​​​​; the winning individuals are crossed to generate offspring individuals, the crossing is performed by randomly selecting two winning individuals and exchanging all subsequent parameters at a randomly selected control time node to obtain a new individual; the offspring individuals are mutated, the mutation is performed by adding or subtracting a mutation amplitude value to the thermostat set temperature value of the offspring individuals to obtain mutated individuals; the mutated individuals are taken as a new generation of population, and the above-mentioned fitness calculation, selection, crossing and mutation processes are repeated until the optimal fitness function value of the individuals of the continuous generations of population improves by less than a convergence judgment threshold value , the iteration is stopped; and the finally obtained optimal individual is taken as an optimal complete dynamic control sequence.

[0035] Further, the method further comprises:

[0036] fine-tuning the steam inlet thermostat set temperature value of each floor position, setting an adjustment threshold value of the thermostat set temperature value of each floor position as a target indoor temperature value , and dividing the adjustment threshold value of the thermostat set temperature value of each floor position into a plurality of equal temperature intervals as a temperature adjustment step ;

[0037] in the heating control execution process, the steam inlet thermostat set temperature value of each floor position is increased or decreased by one temperature adjustment step at a time , the adjusted indoor temperature value is collected in real time by the indoor temperature sensor of each floor position , the absolute value of the difference between the adjusted indoor temperature value and the target indoor temperature value is calculated as a temperature deviation evaluation value ;

[0038] when the temperature deviation evaluation value reaches a minimum value, the cumulative temperature adjustment step at this time is used for the correction adjustment of the steam inlet thermostat set temperature value of each floor position; and the cumulative adjustment direction and the cumulative adjustment times of the temperature adjustment step are recorded as the thermostat temperature correction parameters of each floor position .

[0039] Further, the method of increasing or decreasing the steam inlet thermostat set temperature value of each floor position by one temperature adjustment step at a time comprises:

[0040] increasing the steam inlet thermostat set temperature value of each floor position by one temperature adjustment step at a time along the temperature increasing direction The temperature deviation assessment value was calculated using the gradient descent method. The extreme values ​​and the corresponding direction of temperature increase are cumulatively adjusted step size. ;

[0041] Decrease the set temperature of the thermostatic valve at the steam inlet of each floor by one temperature adjustment step each time, following the direction of temperature decrease. The temperature deviation assessment value was calculated using the gradient descent method. The extreme values ​​and corresponding temperature reduction directions are cumulatively adjusted step sizes. ;

[0042] The temperature deviation assessment values ​​are compared between the directions of temperature increase and temperature decrease. Select temperature deviation assessment value The smaller adjustment direction is taken as the optimal adjustment direction, and the cumulative adjustment step size corresponding to the optimal adjustment direction is taken as the final correction value of the set temperature value of the thermostatic valve at each floor.

[0043] Based on the same inventive concept, the present invention also provides an Internet of Things (IoT)-based heating compartment balance control system, which is used to execute the aforementioned IoT-based heating compartment balance control method.

[0044] (3) Beneficial effects

[0045] Compared with the prior art, the beneficial effect of the present invention is that by adding a thermostatic valve to regulate the uneven heat dissipation of traditional old vertical single-pipe series heating methods with scale buildup, there are problems such as the need for top-floor residents to open windows to dissipate heat and the need for additional electric heating equipment for bottom-floor residents to receive insufficient heat. The invention dynamically adjusts the temperature of different floors of the building under different pipe network heating temperatures. Attached Figure Description

[0046] Figure 1 This is a flowchart of a heating compartment balance control method based on the Internet of Things according to Embodiment 1 of the present invention;

[0047] Figure 2 This is a schematic diagram of a typical apartment centralized heating system using vertical single-pipe series heating according to an embodiment of the present invention.

[0048] Figure 3 This is a schematic diagram of a vertical single-pipe series heating bridging modification according to an embodiment of the present invention. Detailed Implementation

[0049] Clearly, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0050] Before examples are given, the application scenarios of the inventive concept need to be described. A typical vertical single-pipe series heating used in apartment centralized heating is as follows Figure 2 As shown in the figure, the steam inlet temperature of the top floor 6th floor of the apartment is 50℃, and the steam inlet temperatures of the next floors are 48℃, 45℃, 43℃, 41℃ and 38℃ in turn. After the preliminary reconstruction, the vertical single-pipe is connected by a cross and the steam inlet is increased with a thermostatic valve, as shown in the figure Figure 3 The temperature probe of the thermostatic valve is placed at a certain position in the room. The reconstruction here does not intend to control the heat insulation partition of each room of a certain household, but to heat the entire room of the household as a whole. According to the research, it is assumed that the household personnel go out for 10 hours a day, and the room heating is adjusted to 10℃ when the household personnel go out, and the temperature is increased to 20℃ when the household personnel return to the room. The overall steam heat energy saving is not more than 7%, which is caused by the heat inertia of temperature drop and temperature rise. Therefore, instead of single household internal differentiation heating, the entire room of the household is supplied with constant temperature heating for 24 hours. In the preliminary reconstruction, the thermostatic valve is uniformly used, and after setting the indoor temperature of each household, the heating is obviously improved. However, when the municipal heating load is too large, the top floor heating is still sufficient (compared to the initial old one), but the bottom floor heating is insufficient, for example, the set temperature of the steam inlet is set at 45℃, and the actual steam inlet temperatures of the floors below are 46℃, 43℃, 42℃, 41℃ and 39℃ in turn. The results show that two floors of households have not reached the set temperature of 45℃. Therefore, it is necessary to solve this problem. When the municipal heating pipe network reaches the entire apartment, the set temperature of the thermostatic valve is adjusted according to the actual room temperature difference of different floors to ensure that the actual heating temperature of all floors tends to be close to each other. The thermostatic valve can be set to be connected to the central control through the Internet of Things, and the central control of the entire apartment is uniformly controlled. However, it needs to be noted that because the old apartment pipe network is old, the heat dissipation fin pipe of each floor is scaled, which reduces the flow of the heat dissipation fin, and the scaling affects the heat exchange with the air. In theory, the heat dissipation fin of the top floor is the first to flow through the structure of the previous pipe network, so the scaling is most obvious in the long run. The more obvious the scaling of the heat dissipation fin, the worse the heat dissipation effect in the same time, so the steam inlet temperature also needs to be adjusted adaptively.

[0051] Embodiment 1: As shown in the figure Figure 1 The method comprises the following steps:

[0052] The steam inlet thermostat valve of each floor position in the vertical single-pipe cross-connection series heating system is set to a reference test temperature value, temperature change data is continuously collected by the indoor temperature sensor of each floor position to establish a temperature data sequence of each floor position over time at the reference test temperature value, and the heat exchange efficiency decay coefficient of each floor position is calculated according to the temperature data sequence of each floor position;

[0053] The heating control period is divided into multiple control time nodes according to a preset control interval length, a thermostat control scheme is generated for all floor positions for each control time node, each thermostat control scheme includes the set temperature value of the steam inlet thermostat valve of each floor position at the time node and the execution time length value of the set temperature, the thermostat set temperature value of each floor position is calculated according to the heat exchange efficiency decay coefficient and the target temperature difference, and the thermostat control schemes of each control time node are combined in time sequence to form a complete dynamic control sequence;

[0054] Multiple different complete dynamic control sequences are used as initial populations for genetic algorithm optimization, each complete dynamic control sequence includes the thermostat set temperature value and execution time length value of each floor position at each control time node, the combination of the thermostat set temperature value and execution time length value of each floor position is optimized by the genetic algorithm, so that the final indoor temperature value measured by the indoor temperature sensor of each floor position after executing the complete dynamic control sequence reaches the target indoor temperature value; the set temperature value and execution time length value of the steam inlet thermostat valve of each floor position at each control time node in the optimal complete dynamic control sequence optimized by the genetic algorithm are downloaded to the thermostat valve of each floor position in time through the Internet of Things.

[0055] For example, the test of the present embodiment selects a six-story residential building for heating room-by-room balancing control test, which has been transformed from a vertical single-pipe series heating system to a cross-connection. Preliminary research before the test found that the room temperature of the sixth floor often exceeds 25℃ in winter, requiring window ventilation, while the room temperature of the first floor can only be maintained at about 16℃, requiring auxiliary electric heating equipment. Wireless temperature sensors with an accuracy of 0.1℃ are installed in the living rooms of each floor, and the sensors automatically collect temperature data every 2 minutes and upload them in real time through the Internet of Things platform. At the beginning of the test, the steam inlet thermostat valves of all floors are uniformly set to 45℃ as the reference test temperature value. This temperature is selected based on the historical average steam inlet temperature data of the building. Continuous temperature monitoring for 72 hours shows that the sixth floor reaches a stable temperature of 23.2℃ 45 minutes after the thermostat is started, the fifth floor reaches a stable temperature of 22.1℃ 65 minutes after the thermostat is started, the fourth floor reaches a stable temperature of 21.3℃ 85 minutes after the thermostat is started, the third floor reaches a stable temperature of 20.8℃ 110 minutes after the thermostat is started, the second floor reaches a stable temperature of 20.2℃ 135 minutes after the thermostat is started, and the first floor reaches a stable temperature of 19.5℃ 160 minutes after the thermostat is started.

[0056] According to the temperature rise time and steady-state temperature data of each floor, the heat exchange efficiency decay coefficients of each floor are calculated with the first floor as the reference benchmark. The decay coefficient of the first floor is set to 1.00, the decay coefficient of the second floor is 0.89, the decay coefficient of the third floor is 0.76, the decay coefficient of the fourth floor is 0.63, the decay coefficient of the fifth floor is 0.51, and the decay coefficient of the sixth floor is 0.42. These values truly reflect the differences in the degree of fouling of the heat dissipation fins and the aging condition of the pipes of each floor, and are consistent with the actual situation that the heat dissipation fins of high floors are more seriously fouled in vertical single-pipe heating. Then, the 24-hour heating control period is divided into 6 control time nodes at 4-hour intervals, corresponding to 0, 4, 8, 12, 16 and 20 o'clock respectively. According to the requirement of the target indoor temperature of 20℃, combined with the decay coefficients of each floor and the weight factors of different time nodes, the constant temperature valve setting temperature values and 4-hour execution durations of each floor at different time nodes are calculated to form a complete dynamic control sequence.

[0057] The test uses a genetic algorithm to optimize the control sequence. 100 different complete dynamic control sequences are initially generated as population individuals. Each sequence contains 6 sets of constant temperature valve setting temperature and execution duration data of 6 floors at 6 time nodes, a total of 36 groups. These control sequences are executed one by one through the Internet of Things platform, the actual indoor temperature of each floor after the 24-hour control period is measured, and the sum of the squares of the deviations from the target temperature of 20℃ is calculated as the fitness evaluation index. The 50 individuals with the best fitness are selected for crossover and mutation operations to generate a new generation of control sequences. After 200 generations of iterative optimization, the optimal control scheme is obtained. The test result verification shows that after using the optimal control sequence, the indoor temperature of each floor is stable in the range of 19.8 to 20.2℃, and the temperature deviation is controlled within 0.2℃.

[0058] Further, the method for establishing a temperature data sequence of each floor position over time at a reference test temperature value by continuously collecting temperature change data through indoor temperature sensors at each floor position, and calculating a heat exchange efficiency decay coefficient of each floor position comprises:

[0059] The reference test temperature value is set to , the target indoor temperature value is , and the preset minimum error threshold is ;

[0060] The temperature data of each floor position is continuously collected by the indoor temperature sensors at each floor position within a preset measurement time period to form a temperature data sequence of the th floor position , wherein represents the temperature value of the th floor position at the th time sampling point, represents the total number of sampling points; and the temperature data sequence of the a set reference test temperature value the length of the temperature rise time required to reach a steady state temperature

[0061] when the first floor position indoor temperature sensor measurement value changes less than a preset stability threshold value in a set continuous time period the average temperature value in the set continuous time period is recorded as the first floor position steady state temperature value the first floor position heat exchange efficiency decay coefficient

[0062]

[0063] wherein represents the first floor position temperature rise time length and is taken as a temperature rise reference benchmark value, represents the first floor position steady state temperature value and is taken as a steady state reference benchmark value; when the absolute value is less than a preset minimum error threshold value is set to 1.

[0064] Exemplarily, based on the above tests, the calculation of the heat exchange efficiency decay coefficient of each floor is refined. The set reference test temperature value is 45℃, the target indoor temperature value is 20℃, and the preset minimum error threshold value is 0.1℃. The temperature sensor collects data every 2 minutes, and a total of 2160 data points are collected to form the temperature data sequence of each floor for 72 consecutive hours. During the test, the temperature rise process of each floor from the constant temperature valve set to 45℃ is recorded in detail. The indoor temperature of the sixth floor starts to stabilize at about 23.2℃ at the 46th minute, i.e., the 23rd data point after starting, and the temperature change amplitude is less than the preset stability threshold value 0.3℃ within the next 30 minutes. The fifth floor reaches a stable temperature of 22.1℃ at the 66th minute, i.e., the 33rd data point, the fourth floor reaches a stable temperature of 21.3℃ at the 86th minute, i.e., the 43rd data point, the third floor reaches a stable temperature of 20.8℃ at the 110th minute, i.e., the 55th data point, the second floor reaches a stable temperature of 20.2℃ at the 136th minute, i.e., the 68th data point, and the first floor reaches a stable temperature of 19.5℃ at the 160th minute, i.e., the 80th data point.

[0065] ​​​​​​​The first floor is used as the reference value for temperature rise and steady state reference value for decay coefficient calculation. The first floor has a temperature rise time of 160 minutes and a steady state temperature of 19.5°C, and the heat exchange efficiency decay coefficient is set to 1.00. The second floor decay coefficient calculation result is 0.47, but since the absolute value of the difference between the target temperature and the second floor steady state temperature is 0.2°C, which is greater than the preset minimum error threshold of 0.1°C, the formula is calculated normally. The third floor decay coefficient is calculated to be 0.58, the fourth floor decay coefficient is calculated to be 0.72, the fifth floor decay coefficient is calculated to be 0.93, and the sixth floor decay coefficient is calculated to be 1.24.

[0066] In the test, it was found that there were differences between individual calculation results and expectations, the main reason being that the actual heat exchange performance of the old cooling fins was greatly affected by the degree of fouling. In order to ensure the accuracy of the calculation, the reference test was repeated 3 times for each floor, and the average value was taken as the final temperature rise time and steady state temperature data. The final heat exchange efficiency decay coefficients are 1.00 for the first floor, 0.48 for the second floor, 0.60 for the third floor, 0.71 for the fourth floor, 0.90 for the fifth floor, and 1.21 for the sixth floor, which reflect the actual differences in the aging degree of the cooling equipment of each floor. When the absolute value of the difference between the steady state temperature and the target temperature of a floor is less than 0.1°C, the decay coefficient of that floor is directly set to 1.00 to avoid abnormal values in the calculation.

[0067] Further, the method of recording the position of the first floor from the set reference test temperature value to the length of the temperature rise time required to reach the steady state temperature includes:

[0068] determining the starting time point of the constant temperature valve set to the reference test temperature value in the temperature data sequence of the position of the first floor; determining the steady state reaching time point when the temperature value changes less than the preset stability threshold in a continuous time period by continuously monitoring the indoor temperature sensor data of the position of the first floor; calculating the temperature rise time length of the position of the first floor;

[0069] when the temperature rise time length of the position of the first floor exceeds the preset maximum test time threshold , the temperature rise time length is set to the preset maximum test time threshold .

[0070] ​​​​Exemplarily, the embodiment further refines the measurement of the temperature rising time length of each floor. At the beginning, the thermostatic valves of each floor are uniformly set to 45℃, and the temperature sensor starts to synchronously record the temperature changes of each floor at 8:00 am on January 15, 2024. This time is taken as the starting time point of all floors, and the temperature sensor number is displayed as the first sampling point. The initial temperature of the temperature sensor on the sixth floor is 18.5℃, and then the temperature starts to steadily rise. After the temperature reaches 23.0℃ at the 23rd sampling point, i.e., at 8:46 am, the temperature starts to stabilize. After continuously monitoring the temperature changes in the next 15 sampling points, i.e., in 30 minutes, it is found that the temperature fluctuates between 23.0 and 23.4℃, with a maximum change of 0.4℃. However, the change amplitude is always less than the preset stable threshold of 0.3℃ in the continuous 15 minutes, so it is determined that the sixth floor reaches the steady state at the 23rd sampling point. The temperature rising time length of the sixth floor is calculated as 44 minutes, which is obtained by subtracting the first sampling point from the 23rd sampling point and multiplying by the 2-minute sampling interval.

[0071] The initial temperature of the fifth floor is 17.8℃, and it reaches 22.1℃ at the 33rd sampling point, i.e., at 9:04 am, and starts to stabilize, with a temperature rising time length of 64 minutes. The initial temperature of the fourth floor is 17.2℃, and it reaches 21.3℃ at the 43rd sampling point, i.e., at 9:24 am, and starts to stabilize, with a temperature rising time length of 84 minutes. The initial temperature of the third floor is 16.9℃, and it reaches 20.8℃ at the 55th sampling point, i.e., at 9:48 am, and starts to stabilize, with a temperature rising time length of 108 minutes. The initial temperature of the second floor is 16.5℃, and it reaches 20.2℃ at the 68th sampling point, i.e., at 10:14 am, and starts to stabilize, with a temperature rising time length of 134 minutes. The initial temperature of the first floor is 16.1℃, and it reaches 19.5℃ at the 80th sampling point, i.e., at 10:38 am, and starts to stabilize, with a temperature rising time length of 158 minutes.

[0072] In the test, the preset maximum test time threshold is 240 minutes, i.e., 4 hours. If a floor does not reach the stable state within 240 minutes, its temperature rising time length is directly set to 240 minutes, so as to avoid the influence of individual abnormal situations on the generation of the overall regulation scheme. In this test, the temperature rising time of all floors is within 160 minutes, which does not exceed the preset threshold. In order to ensure the measurement accuracy, it is required that the duration of the temperature change amplitude less than 0.3℃ in the continuous detection period is not less than 20 minutes, so as to confirm that the steady state is reached. Such a setting mainly considers that the thermal inertia of the old heat dissipation fins is large, and the temperature fluctuation is relatively slow, so that enough observation time is needed to judge the true stable state.

[0073] Further, when the measurement value of the indoor temperature sensor at the floor position changes by less than the preset stable threshold in the preset continuous time period, the average temperature value in the preset continuous time period is recorded as the Steady-state temperature values ​​at each floor location The methods include:

[0074] In the Temperature data sequence at each floor location From the middle A set of time sampling points are used as the starting detection point numbers to begin detecting the temperature change amplitude within a continuous time period, where the minimum length of the continuous time period is set to be... ; Calculate from the first Starting from each time sampling point, the continuous The maximum temperature within each time sampling point and minimum temperature ,in Greater than or equal to the minimum length of a continuous time period Divide by the minimum number of sampling points obtained by the sampling time interval; calculate the temperature change amplitude within this continuous time period. ;

[0075] When the temperature change range Less than the preset stability threshold At that time, the first time in a continuous time period Steady-state temperature values ​​at each floor location for:

[0076] ;

[0077] When the temperature change range Greater than or equal to the preset stability threshold At that time, the starting detection point number will be determined. Increase by 1 and repeat the above temperature change amplitude detection process until a continuous time period that meets the stability condition is found.

[0078] For example, this embodiment details the specific detection process for steady-state temperature values. The minimum length of the continuous time period in the test is set to 20 minutes, meaning that the temperature change within 10 consecutive time sampling points must be less than a preset stability threshold of 0.3℃ to confirm a steady state. Taking the sixth floor as an example, steady-state detection begins at the 23rd time sampling point, with a temperature of 23.0℃ at that moment. Temperature data is calculated for 10 consecutive sampling points from the 23rd sampling point (i.e., sampling points 23 to 32), with recorded temperature values ​​of 23.0, 23.1, 23.2, 23.4, 23.2, 23.1, 23.3, 23.2, 23.0, and 23.1℃. The maximum temperature across these 10 consecutive sampling points is 23.4℃, the minimum temperature is 23.0℃, and the temperature change is 23.4 minus 23.0, which equals 0.4℃. Since 0.4℃ is greater than the preset stable threshold of 0.3℃, the starting detection point number is increased from 23 to 24 to continue detection.

[0079] The temperature change of the next 10 sampling points is recalculated from the 24th sampling point, and the temperature values of the 24th to 33rd sampling points are 23.1, 23.2, 23.4, 23.2, 23.1, 23.3, 23.2, 23.0, 23.1, 23.2℃. The maximum temperature in this period is 23.4℃, the minimum is 23.0℃, and the temperature change amplitude is still 0.4℃, which is still greater than the stable threshold. Continue to increase the starting detection point number to 25, and the temperature values of the 25th to 34th sampling points are 23.2, 23.4, 23.2, 23.1, 23.3, 23.2, 23.0, 23.1, 23.2, 23.3℃, the temperature change amplitude is 23.4 minus 23.0 equal to 0.4℃, still not meet the stable condition. Until the 28th sampling point is detected, the temperature values of the 28th to 37th sampling points are 23.1, 23.3, 23.2, 23.0, 23.1, 23.2, 23.3, 23.2, 23.1, 23.2℃, the maximum temperature is 23.3℃, the minimum is 23.0℃, and the temperature change amplitude is 0.3℃, which is equal to the preset stable threshold, but according to the requirement of less than the stable threshold, continue to detect.

[0080] When the 29th sampling point is detected, the temperature values of the 29th to 38th sampling points are 23.3, 23.2, 23.0, 23.1, 23.2, 23.3, 23.2, 23.1, 23.2, 23.1℃, the maximum temperature is 23.3℃, the minimum is 23.0℃, and the temperature change amplitude is still 0.3℃. Continue to detect the next 10 sampling points starting from the 30th sampling point, the temperature values are 23.2, 23.0, 23.1, 23.2, 23.3, 23.2, 23.1, 23.2, 23.1, 23.2℃, the maximum temperature is 23.3℃, the minimum is 23.0℃, and the temperature change amplitude is still 0.3℃. When the 31st sampling point is detected, the temperature values of the 31st to 40th sampling points are 23.0, 23.1, 23.2, 23.3, 23.2, 23.1, 23.2, 23.1, 23.2, 23.1℃, and the temperature change amplitude is 0.3℃. Finally, when the 32nd sampling point is detected, the temperature change amplitude of the next 10 sampling points is reduced to 0.2℃, which is less than the preset stable threshold of 0.3℃, and it is confirmed that the sixth floor reaches the stable state. The stable temperature value of the sixth floor is calculated as the average of the temperature values of these 10 sampling points, i.e. the sum of 23.1, 23.2, 23.3, 23.2, 23.1, 23.2, 23.1, 23.2, 23.1, 23.2℃ divided by 10, which is equal to 23.17℃, rounded to 23.2℃.

[0081] Furthermore, the method of dividing the heating control cycle into multiple control time nodes according to a preset control interval, and generating a thermostatic valve control scheme for all floors at each control time node includes:

[0082] The total duration of the heating regulation cycle is set as follows: The preset control interval is The total duration of the heating regulation cycle is divided into equal parts according to the preset regulation interval to obtain the number of regulation time nodes. ;

[0083] Regarding the first A control scheme for thermostatic valves is generated at each control time point. The control scheme includes the control of the thermostatic valves at the steam inlet of each floor at the specified time. The set temperature value and execution duration value for each control time node; calculate the first... Time weight coefficient of each regulation time node ;

[0084] No. The first regulatory time node The set temperature value of the thermostatic valve at each floor location for:

[0085] ;

[0086] The first The first regulatory time node Execution time value for each floor location Set to preset control interval duration The thermostatic valve control schemes for each control time point are arranged and combined according to the time sequence to form a complete dynamic control sequence that includes the set temperature value and execution duration value for all floor locations at all control time points.

[0087] For example, the purpose of the steps in this embodiment is to formulate a specific dynamic control scheme. The test sets the total duration of the heating control cycle to 24 hours, with a preset control interval of 4 hours. Therefore, there are 6 control time nodes, corresponding to 0:00, 4:00, 8:00, 12:00, 16:00, and 20:00 each day. For the first control time node, i.e., the 0:00 period, the time weight coefficient is calculated to be 1, indicating that the maximum control intensity is required during this period. The thermostatic valve setting temperature on the sixth floor at the first time node is actually set to 18.7℃. Similarly, the temperatures for other floors are calculated: the fifth floor is set to 18.9℃, the fourth floor to 19.2℃, the third floor to 19.4℃, the second floor to 19.8℃, and the first floor to 20.5℃.

[0088] The time weight coefficient of the second regulation time node, i.e. the 4-hour period, is 0.80, and the set temperature of each floor is adjusted accordingly. The set temperature of the sixth floor is 18.9℃, the fifth floor is 19.1℃, the fourth floor is 19.3℃, the third floor is 19.5℃, the second floor is 19.9℃, and the first floor is 20.4℃. The third to sixth time nodes are calculated in the same way, and the time weight coefficients are 0.60, 0.40, 0.20, and 0.00, respectively. The regulation intensity gradually weakens until the set temperature of each floor approaches the target temperature of 20℃ at the sixth time node. The principle of this design is to consider the thermal inertia of heating regulation. A larger regulation intensity is needed in the early stage to overcome the temperature difference between floors, and the regulation amplitude is gradually reduced in the later stage to stabilize the temperature.

[0089] The execution duration of each time node in the test is set to 4 hours to ensure that there is enough time for the thermostatic valve to fully exert its regulation effect. The thermostatic valve regulation scheme of the six regulation time nodes is combined in time sequence to form a complete dynamic regulation sequence containing the set temperature of each floor at each time node and the execution duration. This sequence contains 36 sets of regulation parameters, i.e. 6 floors multiplied by 6 time nodes. Each set of parameters contains specific thermostatic valve set temperature values and 4-hour execution duration values. Through this time-differentiated regulation method, the uneven heating problem caused by different aging degrees of heat dissipation equipment in each floor can be effectively solved, and the regulation target of the indoor temperature of each floor approaching the target temperature of 20℃ can be finally achieved.

[0090] Further, the method of generating multiple different complete dynamic regulation sequences as an initial population for genetic algorithm optimization includes:

[0091] Generating an initial population containing complete dynamic regulation sequence individuals, each complete dynamic regulation sequence individual contains the thermostatic valve set temperature values and execution duration values of each floor position at each regulation time node, and the first complete dynamic regulation sequence individual ;

[0092] After the regulation sequence is issued to the thermostatic valve of each floor position in time sequence for execution, the final indoor temperature values of each floor position at the end of the regulation period are collected , and the fitness function value of the first complete dynamic regulation sequence individual is calculated as:

[0093] ;

[0094] wherein represents the total number of floors; and the fitness function value all individuals in the initial population are sorted in ascending order, and the first individual with the minimum fitness function value is selected as the winning individual, wherein ; the winning individual is subjected to a crossover operation to generate offspring individuals, and the crossover operation generates new individuals by randomly selecting two winning individuals and exchanging all subsequent parameters at a randomly selected control time node; the offspring individuals are subjected to a mutation operation, and the mutation operation generates mutated individuals by adding or subtracting a mutation amplitude value to the thermostat set temperature value of the offspring individuals; the mutated individuals are used as a new generation of population, and the above fitness calculation, selection, crossover, and mutation processes are repeated until the optimal fitness function value of the consecutive generation of individuals improves by less than a convergence determination threshold ; and the final optimal individual obtained is used as the optimal complete dynamic control sequence.

[0095] For example, to obtain the optimal temperature control sequence, 120 different complete dynamic control sequence individuals are generated as an initial population, and each individual includes 36 groups of thermostat set temperature values and execution duration values for 6 floors at 6 time nodes. The first individual uses the aforementioned basic scheme, and the remaining 119 individuals are generated by randomly increasing or decreasing the temperature set value of the basic scheme by 1-3°C. For example, the sixth floor set temperature at 0 hours of the 15th individual is adjusted from 18.7°C of the basic scheme to 17.2°C, and the set temperature at 4 hours is adjusted from 18.9°C to 20.1°C. The set temperatures at other time nodes and floors are also randomly adjusted accordingly, but the execution duration remains unchanged at 4 hours. The 120 control sequences are sequentially sent to the thermostat of each floor for execution through the Internet of Things platform. After each sequence is executed for a complete 24-hour control period, the final indoor temperature values of the indoor temperature sensors of each floor are collected. After the first individual is executed, the final temperature of the sixth floor is 20.3°C, the fifth floor is 19.8°C, the fourth floor is 20.1°C, the third floor is 19.9°C, the second floor is 20.2°C, and the first floor is 19.7°C. The fitness function value is calculated as the sum of the squares of the deviations of the final temperatures of each floor from the target temperature of 20°C, i.e., 0.28.

[0096] After full testing of 120 individuals, the individuals with the smallest fitness values are ranked in ascending order, indicating the best control effect. The ranking results show that the 67th individual has the smallest fitness value of 0.15, the 23rd individual has a fitness value of 0.18, the 89th individual has a fitness value of 0.21, and so on. The top 60 individuals with the smallest fitness values are selected as the winning individuals. Then, the crossover operation is performed to generate offspring individuals. Specifically, two winning individuals are randomly selected to exchange all subsequent parameters at a randomly selected control time node. For example, the 67th individual and the 23rd individual are selected, and the 3rd time node, i.e., 8 o'clock, is randomly selected as the crossover point. The temperature setting parameters of all floors at the 8 o'clock, 12 o'clock, 16 o'clock, and 20 o'clock time nodes of the 67th individual are exchanged with the corresponding parameters of the 23rd individual to generate two new offspring individuals. In this way, 60 offspring individuals are generated. Then, the mutation operation is performed on the offspring individuals. The mutation method is to randomly add or subtract 0.5°C from the constant temperature valve setting temperature value as the mutation amplitude value. About 20% of the parameters of each offspring individual are mutated.

[0097] The 60 mutated individuals are used as the new generation population, and the above fitness calculation, selection, crossover, and mutation processes are repeated. After 180 iterations of optimization, the optimal fitness function values of the individuals in the last 15 generations are improved by less than the convergence threshold of 0.02, satisfying the convergence condition to stop iteration. The optimal individual fitness function value obtained is 0.08, and the corresponding control scheme makes the final indoor temperatures of each floor to be 20.1°C on the 6th floor, 19.9°C on the 5th floor, 20.0°C on the 4th floor, 20.1°C on the 3rd floor, 19.9°C on the 2nd floor, and 20.0°C on the 1st floor, with a temperature deviation of less than 0.1°C. The optimal control sequence is more precise in setting the temperature of each floor compared to the initial basic scheme, especially for the high floors with poor heat dissipation effect, the setting temperature is appropriately reduced in the early stage, and for the low floors with good heat dissipation effect, the setting temperature is appropriately increased in the later stage, thereby achieving a more balanced heating control effect.

[0098] Further, the method further comprises:

[0099] The constant temperature valve setting temperature value of each floor position is fine-tuned and optimized. The adjustment threshold of the constant temperature valve setting temperature value of each floor position is set to the target indoor temperature value The adjustment threshold of the constant temperature valve setting temperature value of each floor position is divided into several equal temperature intervals, and the temperature adjustment step is set to the absolute value of the difference between the constant temperature valve setting temperature value of each floor position and the current indoor temperature value measured by the indoor temperature sensor of each floor position ;

[0100] During the heating control execution process, the constant temperature valve setting temperature value of each floor position is increased or decreased by one temperature adjustment step each time , the adjusted indoor temperature value is collected in real time by the indoor temperature sensor of each floor position , the absolute value of the difference between the adjusted indoor temperature value and the target indoor temperature value is calculated as the temperature deviation evaluation value ;

[0101] When the temperature deviation evaluation value reaches the minimum value, the accumulated temperature adjustment step size at this time is used to correct the set temperature value of the steam inlet thermostatic valve of each floor position; the cumulative adjustment direction and the cumulative adjustment times of the temperature adjustment step size are recorded as the thermostatic valve temperature correction parameters of each floor position .

[0102] Exemplarily, further fine-tuning optimization of the set temperature value of the thermostatic valve is implemented, the real-time value of the indoor temperature sensor of each floor is measured at the start of the test, and it is found that the current indoor temperature of the sixth floor is 20.3℃, and the absolute value of the difference between the target indoor temperature 20℃ is 0.3℃, so the adjustment threshold value of the sixth floor thermostatic valve set temperature value is set to 0.3℃. The adjustment threshold value of 0.3℃ is divided into 6 equal temperature intervals, and each temperature adjustment step size is 0.05℃. The current indoor temperature of the fifth floor is 19.7℃, the adjustment threshold value is 0.3℃, and it is also divided into 6 equal intervals, and the temperature adjustment step size is 0.05℃. The current temperature of the fourth floor is 19.9℃, the adjustment threshold value is 0.1℃, and it is divided into 2 equal intervals, and the adjustment step size is 0.05℃. The current temperature of the third floor is 20.2℃, the adjustment threshold value is 0.2℃, and it is divided into 4 equal intervals, and the adjustment step size is 0.05℃. The current temperature of the second floor is 19.8℃, the adjustment threshold value is 0.2℃, and the adjustment step size is 0.05℃. The current temperature of the first floor is 20.1℃, the adjustment threshold value is 0.1℃, and the adjustment step size is 0.05℃.

[0103] ​The temperature value of the thermostat valve of each floor is increased or decreased by one temperature adjustment step 0.05℃ each time, and then the adjusted indoor temperature value is collected in real time through the temperature sensor. Taking the sixth floor as an example, the original thermostat valve set temperature is 18.5℃, the first adjustment reduces 0.05℃ to 18.45℃, and after waiting for 30 minutes, the indoor temperature is measured to be 20.2℃, and the temperature deviation evaluation value is 0.2℃. The second adjustment continues to reduce 0.05℃ to 18.40℃, and the indoor temperature is measured to be 20.1℃, and the deviation evaluation value is 0.1℃. The third adjustment reduces to 18.35℃, and the indoor temperature is measured to be 20.0℃, and the deviation evaluation value is 0℃. The fourth adjustment reduces to 18.30℃, and the indoor temperature is measured to be 19.9℃, and the deviation evaluation value is 0.1℃. By comparison, it is found that the deviation evaluation value reaches the minimum value 0℃ in the third adjustment, so the optimal set temperature of the sixth floor thermostat valve is determined to be 18.35℃, and the cumulative adjustment is 3 temperature adjustment steps, and the adjustment direction is to reduce, and the cumulative adjustment times is 3 times.

[0104] The original thermostat valve set temperature of the fifth floor is 19.2℃, and the current indoor temperature 19.7℃ is lower than the target temperature, so the adjustment is made in the increasing direction. The first increase is 0.05℃ to 19.25℃, the indoor temperature is measured to be 19.8℃, and the deviation evaluation value is 0.2℃. The second increase is to 19.30℃, the indoor temperature is 19.9℃, and the deviation evaluation value is 0.1℃. The third increase is to 19.35℃, the indoor temperature is 20.0℃, and the deviation evaluation value is 0℃. The fourth increase is to 19.40℃, the indoor temperature is 20.1℃, and the deviation evaluation value is 0.1℃. The optimal set temperature of the fifth floor is determined to be 19.35℃, and the cumulative adjustment is 3 steps, and the adjustment direction is to increase. The temperature adjustment step of each floor is recorded as the thermostat valve temperature correction parameter, and the correction parameter of the sixth floor is reduced by 3 times, the fifth floor is increased by 3 times, the fourth floor is increased by 1 time, the third floor is reduced by 2 times, the second floor is increased by 2 times, and the first floor is reduced by 1 time. The correction parameter is used as a reference for subsequent control period, which improves the accuracy and response speed of heating control.

[0105] Further, the method of adjusting the set temperature of the steam inlet thermostat valve of each floor position by increasing or decreasing one temperature adjustment step each time includes:

[0106] increasing the set temperature of the steam inlet thermostat valve of each floor position by one temperature adjustment step each time in the temperature increasing direction , calculating the extreme value of the temperature deviation evaluation value and the corresponding temperature increasing direction cumulative adjustment step

[0107] The temperature value of the constant temperature valve of each floor position steam inlet is reduced by one temperature adjustment step along the temperature reduction direction The temperature deviation evaluation value is calculated by gradient descent method The extreme value of the temperature deviation evaluation value and the corresponding temperature reduction direction cumulative adjustment step ;

[0108] Compare the temperature deviation evaluation values of the temperature increase direction and the temperature reduction direction Select the adjustment direction with smaller temperature deviation evaluation value as the optimal adjustment direction The cumulative adjustment step corresponding to the optimal adjustment direction is used as the final correction value of the constant temperature valve set temperature value of each floor position.

[0109] For example, the constant temperature valve set temperature value of each floor is further fine-tuned to fine-tune the regulated temperature. Take the fourth floor as an example. The current indoor temperature is 19.9℃, the target temperature is 20℃, the temperature adjustment step is 0.05℃, and the original constant temperature valve set temperature is 19.8℃. First, test along the temperature increase direction. The first time, increase the set temperature by 0.05℃ to 19.85℃, wait for 20 minutes, and measure the indoor temperature as 19.95℃. The temperature deviation evaluation value is 0.05℃. The second time, continue to increase to 19.90℃, the indoor temperature is 20.0℃, and the deviation evaluation value is 0℃. The third time, increase to 19.95℃, the indoor temperature is 20.05℃, and the deviation evaluation value is 0.05℃. The fourth time, increase to 20.00℃, the indoor temperature is 20.1℃, and the deviation evaluation value is 0.1℃. Through gradient descent method analysis, it is found that the deviation evaluation value reaches the extreme value of 0℃ at the second time adjustment, and the corresponding temperature increase direction cumulative adjustment step is 2 steps, i.e. 0.10℃. From the trend of the deviation changing from decreasing to increasing, the optimal adjustment point in the increasing direction is found.

[0110] Next, the test is carried out along the temperature reduction direction, and the fourth floor constant temperature valve set temperature is reduced from the original 19.8℃. The first reduction is 0.05℃ to 19.75℃, and the indoor temperature is 19.85℃, and the deviation evaluation value is 0.15℃. The second reduction is to 19.70℃, the indoor temperature is 19.8℃, and the deviation evaluation value is 0.2℃. The third reduction is to 19.65℃, the indoor temperature is 19.75℃, and the deviation evaluation value is 0.25℃. The fourth reduction is to 19.60℃, the indoor temperature is 19.7℃, and the deviation evaluation value is 0.3℃. By analyzing the deviation change law along the temperature reduction direction by gradient descent method, it is found that the deviation evaluation value presents an increasing trend, and reaches the extreme value 0.15℃ at the first adjustment, and the corresponding temperature reduction direction cumulative adjustment step is 1 step, that is, 0.05℃. This shows that adjusting along the reduction direction will make the indoor temperature deviate from the target temperature further, which is not the ideal adjustment direction. Comparing the extreme value 0℃ of the temperature increase direction and the extreme value 0.15℃ of the temperature reduction direction, it is obvious that the deviation evaluation value of the increase direction is smaller, so the temperature increase direction is selected as the optimal adjustment direction of the fourth floor, and the final correction value of the fourth floor constant temperature valve set temperature is finally determined as 0.10℃.

[0111] The same two-way gradient descent method is also used to test the remaining floors. The core idea of this method is to avoid falling into a local optimal solution due to the wrong selection of the initial adjustment direction. The test results show that the optimal adjustment direction of the second floor and the fifth floor is the temperature increase direction, the optimal adjustment direction of the third floor and the sixth floor is the temperature reduction direction, and the first floor is found to have similar extreme values in both directions, and finally the reduction direction with smaller adjustment amplitude is selected. Through this two-way comparison test method, the optimal set temperature of the constant temperature valve of each floor can be determined more accurately, and the deviation caused by single direction adjustment can be avoided. The whole two-way test process takes about 4 hours to complete, although it takes a long time, but it can significantly improve the accuracy of the final regulation effect. The test results show that the indoor temperature deviation of each floor is controlled within 0.05℃ after optimization by the two-way gradient descent method.

[0112] In the same inventive concept, the embodiment also provides a heating subroom balance regulation system based on Internet of Things, which is used to execute the heating subroom balance regulation method based on Internet of Things.

[0113] It should be noted that, as for the system in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0114] Finally, it should be noted that although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to part of the technical features, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the application shall be included in the protection scope of the application.

Claims

1. A method for balanced control of heating in individual rooms based on the Internet of Things, characterized in that, The method includes: In a vertical single-pipe bridging series heating system, the constant temperature valve at the steam inlet of each floor is set as the reference test temperature value. Temperature change data is continuously collected by indoor temperature sensors at each floor to establish a temperature data sequence of each floor over time at the reference test temperature value. The heat exchange efficiency attenuation coefficient of each floor is calculated based on the temperature data sequence of each floor. The heating control cycle is divided into multiple control time nodes according to the preset control interval. For each control time node, a thermostatic valve control scheme is generated for all floors. Each thermostatic valve control scheme includes the set temperature value of the steam inlet thermostatic valve at each floor location at that time node and the execution duration value of the set temperature. The set temperature value of the thermostatic valve at each floor location is calculated based on the heat exchange efficiency attenuation coefficient and the target temperature difference. The thermostatic valve control schemes of each control time node are combined in time sequence to form a complete dynamic control sequence. Multiple different complete dynamic control sequences are used as the initial population for genetic algorithm optimization. Each complete dynamic control sequence contains the set temperature value and execution duration value of the thermostatic valve at each floor location at each control time node. The combination of the set temperature value and execution duration value of the thermostatic valve at each floor location is optimized by the genetic algorithm so that the final indoor temperature value measured by the indoor temperature sensor at each floor location reaches the target indoor temperature value after the complete dynamic control sequence is executed. The set temperature value and execution duration value of the steam inlet thermostatic valve at each floor location in the optimal complete dynamic control sequence optimized by the genetic algorithm are distributed to the thermostatic valve at each floor location in a time-sharing manner through the Internet of Things.

2. The heating compartment balance control method based on the Internet of Things according to claim 1, characterized in that, The method for establishing a temperature data sequence of each floor location over time by continuously collecting temperature change data from indoor temperature sensors at each floor location, under a reference test temperature value, and calculating the heat exchange efficiency attenuation coefficient of each floor location based on the temperature data sequence of each floor location includes: Set the reference test temperature value as The target indoor temperature value is The preset minimum error threshold is ; Temperature data is continuously collected by indoor temperature sensors located on each floor within a preset measurement time period, forming the first... Temperature data sequence at each floor location ,in Indicates the first The floor location is in Temperature values ​​at each time sampling point Indicates the total number of sampling points; records the number of sampling points. Each floor location is based on the set benchmark test temperature value. The length of time required for the temperature to rise from the start to the steady state temperature ; When the The indoor temperature sensor readings at each floor location showed a variation less than a preset stability threshold within a set continuous time period. When, the average temperature value within a set continuous time period is recorded as the first value. Steady-state temperature values ​​at each floor location ;No. Heat exchange efficiency attenuation coefficient at each floor location for: ； in This indicates the time required for the temperature to rise at the first floor location and serves as a reference value for temperature increase. This represents the steady-state temperature value at the first floor location and serves as a steady-state reference value; when The absolute value is less than the preset minimum error threshold. hour, Set to 1.

3. The heating compartment balance control method based on the Internet of Things according to claim 2, characterized in that, The record number Each floor location is based on the set benchmark test temperature value. The length of time required for the temperature to rise from the start to the steady state temperature The methods include: In the Temperature data sequence at each floor location The thermostatic valve is set to the reference test temperature value. The start time point ; through continuous monitoring of the first Indoor temperature sensor data at each floor location; when the temperature value changes by less than a preset stability threshold over a continuous period of time. Determine the time point when steady state is achieved. ; Calculate the first The heating time at each floor location When the first The heating time at each floor location Exceeding the preset maximum test time threshold At that time, the heating time length Set as the preset maximum test time threshold .

4. The heating compartment balance control method based on the Internet of Things according to claim 2, characterized in that, The first The indoor temperature sensor readings at each floor location showed a variation less than a preset stability threshold within a set continuous time period. When, the average temperature value within a set continuous time period is recorded as the first value. Steady-state temperature values ​​at each floor location The methods include: In the Temperature data sequence at each floor location From the middle A set of time sampling points are used as the starting detection point numbers to begin detecting the temperature change amplitude within a continuous time period, where the minimum length of the continuous time period is set to be... ; Calculate from the first Starting from each time sampling point, the continuous Maximum temperature within each time sampling point and minimum temperature ,in Greater than or equal to the minimum length of a continuous time period Divide by the minimum number of sampling points obtained by the sampling time interval; calculate the temperature change amplitude within this continuous time period. ; When the temperature change range Less than the preset stability threshold At that time, the first time in a continuous time period Steady-state temperature values ​​at each floor location for: ； When the temperature change range Greater than or equal to the preset stability threshold At that time, the starting detection point number will be determined. Increase by 1 and repeat the above temperature change amplitude detection process until a continuous time period that meets the stability condition is found.

5. A heating compartment balance control method based on the Internet of Things according to claim 2, characterized in that, The method of dividing the heating control cycle into multiple control time nodes according to a preset control interval, and generating a thermostatic valve control scheme for all floors for each control time node includes: The total duration of the heating regulation cycle is set as follows: The preset control interval is The total duration of the heating regulation cycle is divided into equal parts according to the preset regulation interval to obtain the number of regulation time nodes. ; Regarding the first A control scheme for thermostatic valves is generated at each control time point. The control scheme includes the control of the thermostatic valves at the steam inlet of each floor at the specified time. The set temperature value and execution duration value for each control time node; calculate the first... Time weight coefficient of each regulation time node ; No. The first regulatory time node The set temperature value of the thermostatic valve at each floor location for: ； The first The first regulatory time node Execution time value for each floor location Set to preset control interval duration The thermostatic valve control schemes for each control time point are arranged and combined according to the time sequence to form a complete dynamic control sequence that includes the set temperature value and execution duration value of all floors at all control time points.

6. The heating compartment balance control method based on the Internet of Things according to claim 5, characterized in that, The method of using multiple different complete dynamic regulatory sequences as the initial population for genetic algorithm optimization includes: Generating the initial population includes Each complete dynamic regulatory sequence individual Includes the set temperature and execution duration of the thermostatic valve at each floor location and at each control time point. A complete dynamic regulatory sequence individual ; After the control sequence is sent to the thermostatic valves on each floor in chronological order, the final indoor temperature values ​​of the indoor temperature sensors on each floor are collected at the end of the control cycle. Calculate the first Fitness function value of an individual with a complete dynamic regulatory sequence for: ； in Represents the total number of floors; based on the fitness function value Sort all individuals in the initial population in ascending order and select the individuals with the smallest fitness function value. Each individual is considered the winning individual, among which The winning individuals are crossbred to generate offspring. This crossbringing involves randomly selecting two winning individuals and swapping all subsequent parameters at a randomly chosen control time point to obtain a new individual. The offspring are then mutated by adding or subtracting a mutation value from the set temperature of the thermostatic valve. Obtain mutant individuals; use these mutant individuals as the new generation population, and repeat the fitness calculation, selection, crossover, and mutation process described above until continuous mutations are obtained. The improvement in the optimal fitness function value of an individual is less than the convergence threshold. Stop iterating when the time is right; take the best individual obtained at the end as the best complete dynamic control sequence.

7. The heating compartment balance control method based on the Internet of Things according to claim 1, characterized in that, The method further includes: The set temperature values ​​of the thermostatic valves at the steam inlet on each floor were fine-tuned and optimized, and the adjustment threshold values ​​of the set temperature values ​​of the thermostatic valves on each floor were set as the target indoor temperature values. The absolute value of the difference between the current indoor temperature value measured in real time by the indoor temperature sensors at each floor location and the set temperature adjustment threshold value of the thermostatic valve at each floor location is used to divide the temperature value adjustment into several equal temperature intervals, which are then set as the temperature adjustment step size. ; During the heating control process, the set temperature value of the steam inlet thermostatic valve at each floor is increased or decreased by one temperature adjustment step each time. The adjusted indoor temperature values ​​are collected in real time by indoor temperature sensors located on each floor. Calculate the adjusted indoor temperature values ​​respectively. With the target indoor temperature value The absolute value of the difference between them is used as the temperature deviation assessment value. ; When the temperature deviation assessment value When the minimum value is reached, adjust the step size of the accumulated temperature at this point. Used for correcting and adjusting the set temperature values ​​of the thermostatic valves at the steam inlet of each floor; adjusting the temperature adjustment step size. The cumulative adjustment direction and cumulative adjustment number are recorded as the temperature correction parameters for the thermostatic valves at each floor location. .

8. A heating compartment balance control method based on the Internet of Things according to claim 7, characterized in that, The temperature setting value of the thermostatic valve at the steam inlet of each floor is increased or decreased by one temperature adjustment step each time. The methods include: Increase the set temperature of the thermostatic valve at the steam inlet of each floor by one temperature adjustment step each time, moving in the direction of temperature increase. The temperature deviation assessment value was calculated using the gradient descent method. The extreme values ​​and the corresponding direction of temperature increase are cumulatively adjusted step size. ; Decrease the set temperature of the thermostatic valve at the steam inlet of each floor by one temperature adjustment step each time, following the direction of temperature decrease. The temperature deviation assessment value was calculated using the gradient descent method. The extreme values ​​and corresponding temperature reduction directions are cumulatively adjusted step sizes. ; The temperature deviation assessment values ​​are compared between the directions of temperature increase and temperature decrease. Select temperature deviation assessment value The smaller adjustment direction is taken as the optimal adjustment direction, and the cumulative adjustment step size corresponding to the optimal adjustment direction is taken as the final correction value of the set temperature value of the thermostatic valve at each floor.

9. A heating compartment balance control system based on the Internet of Things, characterized in that, The balance control system is used to execute the Internet of Things-based heating compartment balance control method as described in any one of claims 1-8.

Images