Ground source heat pump intelligent control method and system for improving comprehensive energy efficiency based on Internet of Things

Abstract

The invention relates to the technical field of neural network algorithm control, in particular to a ground source heat pump intelligent control method and system for improving comprehensive energy efficiency based on the Internet of Things, and the method comprises the following steps: matching temperature and flow trends to generate correction, adjusting power priority embedding logic to generate collaborative priority, extracting pressure difference and flow slope to judge an unbalance trend, and controlling a ground source heat pump according to the unbalance trend. And generating an energy efficiency optimization atlas in combination with feedback. The temperature and flow deviation trend is dynamically matched, the control response opportunity is corrected in combination with the soil temperature gradient change, the heat source switching accuracy and timeliness are improved, cooperative execution of all control elements is ensured through time sequence offset calculation and power priority adjustment, and the heat source switching efficiency is improved. The branch differential pressure fluctuation and flow slope joint analysis enhances the identification capability for the unbalance trend, the flow matching degree and the differential pressure direction are linked to adjust the valve response, the adjustment precision and the execution efficiency are improved, the compensation and energy efficiency fluctuation association is constructed through the feedback spectrum and the flow balance degree, and efficient cooperative control and energy-saving optimization in the multi-source state are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of neural network algorithm control, and particularly to a ground source heat pump intelligent control method and system for improving comprehensive energy efficiency based on the Internet of Things. Background Art

[0002] The technical field of neural network algorithm control includes a method system for dynamically optimizing complex systems using data-driven models. Its core is to collect environmental and device state parameters through a sensor network, realize real-time data processing by combining edge computing and cloud collaboration architecture, and generate control instructions using deep learning algorithms. This technical field involves multi-source heterogeneous data fusion technologies, including dynamic monitoring of soil thermal conductivity, analysis of the operating efficiency of heat pump units, prediction of user load demand, and at the same time, it is necessary to solve the synchronization and stability problems of control strategies at different time scales. Traditional methods rely on fixed thresholds or linear adjustment mechanisms and are difficult to cope with the interference caused by non-linear heat exchange processes and climate fluctuations.

[0003] Among them, the ground source heat pump intelligent control method and system for improving comprehensive energy efficiency based on the Internet of Things refer to constructing a data acquisition network by deploying temperature sensors, flow meters, and power monitoring devices, obtaining parameters such as the soil layer temperature gradient, the temperature difference between the inlet and outlet of the circulating water, and the compressor speed at a second-level frequency, establishing a mapping relationship model between the heat pump output and the energy efficiency ratio using a backpropagation neural network, combining edge nodes to perform short-term load prediction and parameter pre-adjustment, and the cloud platform completing medium- and long-term soil heat balance analysis and algorithm iteration, and finally generating coordinated optimization instructions for the compressor frequency, valve opening, and water pump flow rate according to the dynamic weight distribution mechanism. The technical matters of this method cover sensor data correction based on Kalman filtering, construction of a neural network training set under thermodynamic constraints, and division of control instruction priorities at the edge-cloud two-level.

[0004] Traditional neural network control methods collect environmental and equipment status parameters at a fixed frequency, making it difficult to dynamically reflect the mutual evolution relationship between various trend items in the system, resulting in a response delay between the control rhythm and load changes. In terms of the scheduling strategy, it relies on static instruction - issuing logic and fixed - threshold determination methods, unable to achieve feed - forward adjustment of the timing offset of control actions, easily causing action conflicts or instruction repetitions, and increasing the ineffective energy consumption of equipment operation. In data fusion, it only processes the single - dimensional information of each monitoring item, fails to conduct logical linkage judgment on continuous evolution processes such as differential pressure fluctuations and flow slope, resulting in a lag in the identification of local abnormal trends, and further affecting the accuracy of the control link. During the response adjustment process, it is unable to precisely quantify the flow matching degree, resulting in insufficient pertinence of valve adjustment actions, the adjustment results deviating from the target parameters, and increasing the amplitude of energy consumption fluctuations. In the evaluation of execution results, no collaborative matching mechanism is established between feedback data and energy - efficiency response, lacking the ability of multi - source attribution for the energy - saving effect of the system, resulting in a lack of data support for optimizing control strategies. A control system operating in this way is prone to problems such as insufficient response, poor control synchronization, and unstable energy - saving effect when facing dynamic loads, non - linear heat transfer, and complex climate disturbances. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art, and to propose a ground - source heat pump intelligent control method and system for improving comprehensive energy efficiency based on the Internet of Things.

[0006] To achieve the above - mentioned purpose, the present invention adopts the following technical solutions: A ground - source heat pump intelligent control method for improving comprehensive energy efficiency based on the Internet of Things, including the following steps: S1: Obtain the temperature trend at the outlet of the ground - source side pipeline and the deviation trend of the circulating water flow rate on the user side, dynamically match the directions of the two, and combine the cumulative change of the soil temperature gradient to generate a correction amount for the heat - source switching delay; S2: According to the correction amount for the heat - source switching delay, calculate the direction of the time - sequence offset between the heat - source switching instruction and the valve action, adjust the priority of the power signal, and embed the adjustment logic within the control period before the valve action to generate a time - sequence coordination priority; S3: According to the time - sequence coordination priority, obtain the fluctuation range of the deviation of the branch differential pressure from the design value and the slope of the flow rate difference between adjacent branches, and determine whether the differential pressure fluctuation exceeds the boundary and whether the flow slope continuously increases to generate a trigger flag for the branch imbalance trend; S4: Based on the trigger flag for the branch imbalance trend, combine the matching degree between the current branch flow rate and the design flow rate and the differential pressure fluctuation direction of the adjacent branch to adjust the direction and amplitude of the valve opening to generate a branch flow compensation instruction; S5: Based on the feedback of the execution of the branch flow compensation instruction, aggregate the fluctuation spectrum of the return water temperature on the ground - source side and the balance degree of the branch flow distribution to determine the collaborative relationship between branch compensation and energy - efficiency fluctuation, and generate a multi - source energy - efficiency optimization map.

[0007] As a further solution of the present invention, the heat source switching delay correction amount includes the temperature trend at the outlet of the ground source side pipeline, the deviation trend of the circulating water flow rate on the user side, and the cumulative change of the soil temperature gradient. The time sequence coordination priority includes the heat source switching instruction, the time sequence offset direction of the valve action, and the power signal priority adjustment logic. The branch imbalance trend trigger flag includes the deviation of the branch pressure difference from the designed value fluctuation range, the slope of the flow rate difference between adjacent branches, the pressure difference fluctuation boundary state, and the flow rate slope continuity state. The branch flow compensation instruction includes the matching degree between the current branch flow rate and the designed flow rate, the pressure difference fluctuation direction between adjacent branches, the valve opening direction, and the valve amplitude adjustment parameter. The multi-source energy efficiency optimization map includes the ground source side return water temperature fluctuation spectrum, the branch flow distribution balance degree, and the coordination relationship between branch compensation and energy efficiency fluctuation.

[0008] As a further solution of the present invention, the specific steps of S1 are as follows: S101: Obtain the temperature at the outlet of the ground source side pipeline and the circulating water flow rate data on the user side collected by the Internet of Things sensing nodes, determine the trend based on the temperature change direction and the flow rate deviation direction within the same time period, match and identify the two types of trend directions, select the time period with the same trend direction, and generate the numerical quantity of the trend direction matching section; S102: According to the time period corresponding to the numerical quantity of the trend direction matching section, extract the change data of the soil temperature within the time period, identify the gradient difference between consecutive time points, and accumulate the overall temperature change intensity of the section in the order of time series to generate the total temperature gradient cumulative change; S103: Combine the total temperature gradient cumulative change and the numerical quantity of the trend direction matching section, judge the deviation of the intensity relationship between the two data, compare it with the heat source switching sensitivity reference value, and generate the heat source switching delay correction amount according to the degree of intensity difference.

[0009] As a further solution of the present invention, the specific steps of S2 are as follows: S201: Based on the heat source switching delay correction amount and the heat source switching instruction trigger time, valve action response time, and actuator power adjustment time collected within the current control cycle, identify the time offset between the heat source switching instruction and the valve action, combine the response offset threshold to judge the offset direction, and generate the offset direction determination result; S202: According to the offset direction determination result, combine the power signal adjustment amplitude, switching node number, and valve response period position, set the response order of the power signal within the control cycle, and mark the priority value of each signal to generate the signal priority distribution quantity; S203: Adjust the signals with priorities lower than the response threshold to the control cycle position before valve actuation according to the signal priority distribution quantity and the valve response period position, and re-encode the signal sequence to generate a timing collaborative priority.

[0010] As a further solution of the present invention, the specific steps of S3 are as follows: S301: Based on the timing collaborative priority, combine the Internet of Things sensing network to obtain the real-time data of the branch pressure difference and the designed pressure difference reference value, detect the pressure difference offset situation within the detection time period, extract the pressure difference difference boundary interval of each branch within a continuous time period, and generate a pressure difference offset fluctuation interval; S302: According to the pressure difference offset fluctuation interval, call the flow rate data of adjacent branches within the corresponding time period, extract the difference change trend, determine the difference change rate sequence within a continuous time period, and generate an adjacent branch flow rate slope sequence; S303: Based on the pressure difference offset fluctuation interval and the adjacent branch flow rate slope sequence, judge whether the pressure difference offset exceeds the fluctuation boundary, and check whether the slope sequence shows a continuous increasing trend to obtain a branch imbalance trend trigger flag.

[0011] As a further solution of the present invention, the specific calculation formula based on the pressure difference offset fluctuation interval and the adjacent branch flow rate slope sequence is: ; Wherein, represents the third standardized amplitude of the pressure difference offset, represents the cube of the flow rate value of the i-th measurement, represents the standard deviation of the flow rate value, v represents the number of measurements, i represents the sequence with an index from 1 to v, and the summation symbol The associated letter i indicates the operation on all measured values.

[0012] As a further solution of the present invention, the specific steps of S4 are as follows: S401: Based on the branch imbalance trend trigger flag, call the current branch flow rate and the designed flow rate, extract the corresponding data of the branch, establish a corresponding relationship according to the number, identify the difference direction between the current value and the designed value, and form a flow rate offset direction and amplitude set in combination with the difference magnitude to generate a flow rate offset degree sequence; S402: According to the flow rate offset degree sequence, call the adjacent branch pressure difference information, form a branch group according to the branch number, judge the fluctuation direction of the pressure difference between adjacent moments, compare it with the offset direction of the corresponding branch, and extract the direction consistency feature in combination with the pressure difference change and the offset amplitude to generate a pressure difference guiding coefficient sequence; S403: Extract the current valve opening value of the corresponding branch according to the differential pressure guiding coefficient sequence, determine whether the opening adjustment direction is consistent with the guiding direction, adjust the opening amplitude according to the guiding coefficient, set it as the corresponding mapping value when the directions are consistent, and set it as the reverse correction value when the directions are inconsistent, and generate a branch flow compensation instruction.

[0013] As a further solution of the present invention, the specific calculation formula for adjusting the opening amplitude according to the guiding coefficient is: ; Wherein, represents the corrected opening adjustment amplitude of the j-th valve in the i-th branch, represents the guiding weight coefficient of the j-th valve in the i-th branch under the k-th measurement, represents the pressure difference value of the j-th valve in the i-th branch under the k-th measurement, represents the target deviation penalty coefficient of the j-th valve in the i-th branch, represents the expected reference opening value of the j-th valve in the i-th branch, represents the opening value of the j-th valve in the i-th branch at the previous moment, represents the opening adjustment sensitivity coefficient corresponding to the j-th valve in the i-th branch, represents the flow interference factor of the j-th valve in the i-th branch, represents the minimum dynamic suppression coefficient of the j-th valve in the i-th branch, which is used to characterize the correction stability of the valve when the disturbance is extremely small, and n represents the total number of measurement times.

[0014] As a further solution of the present invention, the specific steps of S5 are as follows: S501: Based on the branch flow compensation instruction, call the time series data of the ground source side return water temperature and the instantaneous flow rate, complete synchronous matching, screen continuous data, extract the corresponding relationship between the flow rate change and the temperature offset, and generate a branch offset correlation eigenvalue; S502: According to the branch offset correlation eigenvalue, extract the flow rate change and the temperature fluctuation amplitude of the branch, construct a frequency distribution diagram, count the fluctuation density of the frequency interval, and analyze the difference in combination with the curve trend, and generate the degree of change of the branch fluctuation spectrum; S503: Call the degree of change of the branch fluctuation spectrum and the branch flow rate change, compare the change relationship between the flow rate distribution and the spectrum fluctuation, and screen the corresponding branches in combination with the change trend of the energy efficiency utilization rate to generate a multi-source energy efficiency optimization map.

[0015] The ground source heat pump intelligent control system for improving the comprehensive energy efficiency based on the Internet of Things includes: The temperature trend matching module obtains the deviation value between the outlet temperature of the ground source side pipeline and the user side circulating water flow rate, detects the trend directions of the two, calculates the cumulative value of the change in the soil temperature gradient, and generates a heat source switching delay correction amount according to the ratio of the direction differences among the three; The valve control priority scheduling module calls the heat source switching delay correction amount, obtains the valve action response period and the power signal response period, compares the difference in the overlapping interval of their time axes, and generates a timing coordination priority; The branch imbalance identification module calls the timing coordination priority, obtains the branch pressure difference monitoring value, the designed pressure difference value, and the real-time flow rate of the adjacent branch, calculates the amplitude of the difference between the pressure difference and the designed value, determines whether the upper and lower limits of the pressure difference are exceeded, screens the sequence in which the slope of the flow rate difference in the adjacent branches continuously increases, and generates a branch imbalance trend trigger flag based on the overlapping area between the two; The flow compensation control module calls the branch imbalance trend trigger flag, obtains the branch flow rate and the designed value, and the change direction of the pressure difference of the adjacent branch, calculates the degree of consistency between the flow deviation direction and the pressure difference fluctuation direction, screens the branches that need to be adjusted, and adjusts the valve opening direction and amplitude, and generates a branch flow compensation instruction; The energy efficiency map construction module calls the branch flow compensation instruction, obtains the branch flow rate distribution data and the ground source side return water temperature spectrum, calculates the balance index and the fluctuation range, and generates a multi-source energy efficiency optimization map.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by dynamically matching the temperature and flow deviation trends, and combining the change of the soil temperature gradient to correct the control response timing, the accuracy and timeliness of heat source switching are improved. The timing offset calculation and power priority adjustment ensure the coordinated execution of each control element. The joint analysis of the branch pressure difference fluctuation and the flow rate slope enhances the ability to identify the imbalance trend. The flow matching degree and the pressure difference direction are linked to adjust the valve response, improving the adjustment accuracy and execution efficiency. By constructing the compensation and energy efficiency fluctuation correlation through the feedback spectrum and flow balance degree, efficient coordinated control and energy-saving optimization in the multi-source state are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is the step flow schematic diagram of the present invention; Figure 2 is the system module diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0020] Please refer to Figure 1 , a ground source heat pump intelligent control method for improving comprehensive energy efficiency based on the Internet of Things, comprising the following steps: S1: Collect the temperature change trend at the outlet of the ground source side pipeline and the deviation trend of the circulating water flow on the user side through the Internet of Things sensing nodes, dynamically match the temperature trend direction and the flow deviation direction, and combine the cumulative change amount of the soil temperature gradient to generate a heat source switching delay correction amount; S2: Based on the heat source switching delay correction amount, calculate the timing offset direction of the heat source switching instruction and the valve action, adjust the power signal priority through the Internet of Things control node, and embed the control cycle before the valve action to generate a timing coordination priority; S3: Based on the timing coordination priority, obtain the fluctuation range of the branch pressure difference deviating from the design value and the slope of the flow difference between adjacent branches through the Internet of Things sensing network, determine whether the pressure difference fluctuation breaks through the boundary and whether the flow slope continuously increases, and generate a branch imbalance trend trigger flag; S4: Based on the branch imbalance trend trigger flag, combine the matching degree between the current branch flow and the design flow and the pressure difference fluctuation direction of adjacent branches, adjust the valve opening direction and amplitude, and generate a branch flow compensation instruction; S5: Based on the execution feedback of the branch flow compensation instruction, aggregate the ground source side return water temperature fluctuation spectrum and the branch flow distribution balance degree through the Internet of Things platform, determine the coordination relationship between branch compensation and energy efficiency fluctuation, and generate a multi-source energy efficiency optimization map.

[0021] The heat source switching delay correction amount includes the temperature trend at the outlet of the ground source side pipeline, the deviation trend of the circulating water flow on the user side, and the cumulative change of the soil temperature gradient. The time sequence coordination priority includes the heat source switching instruction, the deviation direction of the valve action time sequence, and the power signal priority adjustment logic. The branch imbalance trend trigger flag includes the deviation of the branch pressure difference from the designed value fluctuation range, the slope of the flow difference between adjacent branches, the boundary state of the pressure difference fluctuation, and the continuity state of the flow slope. The branch flow compensation instruction includes the matching degree between the current branch flow and the designed flow, the fluctuation direction of the pressure difference between adjacent branches, the valve opening direction, and the valve amplitude adjustment parameter. The multi-source energy efficiency optimization map includes the fluctuation spectrum of the return water temperature on the ground source side, the balance degree of the branch flow distribution, and the collaborative relationship between the branch compensation and the energy efficiency fluctuation.

[0022] The specific steps of S1 are as follows: S101: Obtain the temperature at the outlet of the ground source side pipeline and the circulating water flow data on the user side collected by the Internet of Things sensing nodes, perform trend determination based on the temperature change direction and the flow deviation direction within the same time period, and match and identify the two types of trend directions. Select the time period with the same trend direction and generate the numerical quantity of the trend direction matching section; To obtain the temperature at the outlet of the ground source side pipeline and the circulating water flow data, it is necessary to deploy Internet of Things sensing nodes with remote transmission functions. On the ground source side, temperature collection can use high-precision temperature sensors installed at the pipeline outlet position, record the value every few seconds and upload it to the server through the transmission module. For the flow collection on the user side, electromagnetic or ultrasonic flow meters can be used. These devices are connected to the data collection terminal through industrial communication protocols to achieve real-time recording and synchronous upload of the circulating water flow. A time-based comparison table is established uniformly in the data platform to correspond the temperature and flow data according to time nodes. Define the temperature change direction between adjacent moments as rising or falling, and the flow change direction as increasing or decreasing. Then, determine the trend matching based on whether the two are in the same direction. In the matching determination, if the temperature rises and the flow increases simultaneously, or the temperature drops and the flow decreases simultaneously, it is considered that the trend directions are the same, otherwise it is considered inconsistent. Record the matching situation item by item and accumulate the total number to construct the number of sections with trend direction matching. Combining the actual scenario, if the temperature rises successively and the flow also increases, this stage should be recorded as a group of trend matching sections. If this cooperation relationship appears in multiple consecutive time periods, the total amount is the numerical quantity of the trend direction matching section, which is used to reflect the consistency of the data trend.

[0023] S102: According to the time period corresponding to the numerical quantity of the trend direction matching section, extract the change data of the soil temperature within the time period, identify the gradient difference between consecutive time points, and accumulate the overall temperature change intensity of the section in the order of time series to generate the total amount of the temperature gradient cumulative change; After obtaining the number of sections with matching trend directions, it is necessary to extract the soil temperature data within this time period based on its time range. Usually, soil temperature is collected using buried sensors and recorded at a frequency of once per minute. After extracting the temperature sequence within the time range corresponding to the section matching the trend, calculate the temperature difference between adjacent moments in chronological order, construct a gradient change list reflecting the degree of temperature fluctuation, accumulate the temperature difference values between each time point item by item and summarize them in absolute value form to obtain the overall change intensity of the temperature within this time period. If the continuous change amplitude is too small, a minimum change threshold needs to be set as the filtering criterion to ignore small changes and reduce data noise interference. In actual operation, a temperature change within 0.05 degrees can be considered an invalid change, so as to eliminate insignificant changes. If there is ten-minute data in a matching section and the temperature continuously fluctuates between 0.1 and 0.3 degrees, the cumulative value of the temperature gradient in this stage can be used as a basic index for evaluating the soil response intensity, and finally form the total change value of the temperature gradient corresponding to the trend matching stage, laying a data basis for subsequent deviation judgment.

[0024] S103: Combine the total cumulative change of the temperature gradient with the numerical quantity of the section with matching trend directions, judge the deviation of the intensity relationship between the two data, and compare it with the reference value of the heat source switching sensitivity, and generate a heat source switching delay correction amount according to the degree of intensity difference; After completing the calculation of the numerical value of the trend matching section and its corresponding total temperature gradient change, deviation judgment can be further carried out through the ratio of the two. This ratio reflects the temperature change intensity brought by each trend matching section and provides a reference for evaluating the heat source response characteristics. According to the level of this ratio, it can be compared with the set reference standard. If the ratio is higher than the set standard, it means that the current soil temperature response speed is faster. At this time, it is necessary to delay the heat source switching time to maintain system stability. The determination of the delay correction amount can be set according to the product of an adjustment coefficient and the ratio deviation. This adjustment coefficient should be determined by referring to historical operation data, and its value represents the time correction degree corresponding to the unit change intensity. For example, for every 0.1-degree increase in intensity, the delay can be about 0.5 to 1 minute. According to this setting, the actual correction value can be gradually formed. At the same time, when making intensity judgments, it is necessary to clarify the division of the response interval. For example, when the ratio is greater than 0.8, it is judged as a strong response, between 0.4 and 0.8 is a medium response, and less than 0.4 is considered a weak response. Such interval standards should be obtained from the data analysis results during the system debugging stage and adjusted in combination with different seasons and load conditions, and finally form the heat source switching delay adjustment time to guide the update of the control logic.

[0025] The specific steps of S2 are as follows: S201: Based on the heat source switching delay correction amount, the triggering time of the heat source switching instruction collected within the current control period, the valve action response time, and the actuator power adjustment time, identify the time offset between the heat source switching instruction and the valve action, combine with the response offset threshold to judge the offset direction, and generate an offset direction determination result; First, obtain the triggering time of the heat source switching instruction in the current control period. For example, if the heat source switching instruction is issued at the 20th second, this time is the analysis starting point. At the same time, record the starting time of the valve response corresponding to this instruction. If the valve starts to act at the 27th second, it can be inferred that the response delay is 7 seconds. Also record the time when the actuator power starts to be adjusted. For example, the power adjustment starts at the 3rd second after the instruction is issued, that is, the 23rd second. At this time, the delay amount is the difference between the valve action time and the instruction triggering time. By comparing this time difference with the pre-set response offset threshold, it can be judged whether there is an abnormal response. For example, when the set response offset threshold is 5 seconds, if the actual time difference is greater than 5 seconds, it means that the system has a delayed response. Otherwise, it can be considered that the response is advanced or normal. Such response offset judgment needs to be combined with the specific system working conditions. For example, in the heat source switching scenario from a natural gas boiler to an electric heating device, if the electric heating device supplies energy in advance and the hot water valve closes late, it will lead to mismatched heat flow. Therefore, by calculating the time difference between the instruction and the valve response, and further combining with the offset judgment threshold, it can be clearly judged whether the response is offset and the offset direction, and finally generate a judgment result of the offset direction.

[0026] S202: According to the offset direction determination result, combine with the power signal adjustment amplitude, the switching node number, and the valve response period position, set the response order of the power signal within the control period, and mark the priority values of each signal to generate a signal priority distribution quantity; After the response offset direction has been identified, the response order of the power adjustment signal within the current control cycle is set according to the offset direction. First, an evaluation is made based on the adjustment amplitude of each group of power signals. For example, the adjustment amplitude of the electric heater is 5 kW, the corresponding value for the hot water valve is 2 kW, and the natural gas auxiliary device is 3 kW. The adjustment capabilities of these three groups of signals are sorted from large to small, and combined with the switching node numbers to which they belong. For example, the electric heater is node 2, the valve is node 3, and the natural gas device is node 1. Then, combined with the expected time period position of the valve response, such as the expected response time of the valve being from the 25th second to the 30th second, the signals with high adjustment amplitude, concentrated response time, and belonging to key components are set as signals with higher priorities. For example, the electric heater is set as priority 1, the valve is set as priority 2, and the natural gas device is set as priority 3. After marking this priority order, a signal priority distribution list is formed, which provides an order reference for subsequent signal scheduling. Among them, the lower the priority value, the earlier the response and the higher the priority for processing, while the higher the priority value, the response can be appropriately postponed, thus forming a complete signal priority order to guide the response arrangement of power signals in the control system.

[0027] S203: According to the signal priority distribution quantity and the valve response time period position, adjust the signals with priorities lower than the response threshold to the control cycle position before the valve action, and re-encode the signal sequence to generate a time-sequence collaborative priority; Based on the existing signal priority distribution, according to the position of the valve response time period, adjust the order of the power signals with priorities lower than the response threshold. For example, the key response time period for the valve action is set from the 25th second to the 30th second, and the priority threshold is set to 2. Then the signal with priority 3 belongs to the object that needs to be adjusted. At this time, adjust the control time of this signal to one control cycle before the valve action, for example, advance it to the 24th second, to avoid signal response interfering with the valve operation sequence. After completing the time adjustment, re-arrange all the signals according to the current response time and assign an encoding order. For example, the response times are from early to late as the 24th second, the 26th second, and the 28th second, then the corresponding power signal order is the signals with adjustment amplitudes of 3 kW, 2 kW, and 5 kW. Subsequently, according to the new time-sequence position and combined with the priority value, re-sort to form the final time-sequence collaborative priority sequence. This sequence not only reflects the sequence of responses but also synchronously retains the priority information of the signals within the control cycle, which is used to adjust the execution order in the control logic to ensure the logical consistency and response rhythm coordination among the signals within the control cycle.

[0028] The specific steps of S3 are as follows: S301: Based on the time-sequence collaborative priority, combined with the Internet of Things sensing network to obtain the real-time data of the branch pressure difference and the designed pressure difference reference value, detect the pressure difference offset situation during the detection time period, extract the pressure difference difference boundary interval of each branch within a continuous time period, and generate a pressure difference offset fluctuation interval; First, based on the distribution of each branch in the physical structure and the control logic set by the system, a unified time alignment model is constructed. This model can synchronize data at different time periods through the timestamp information collected by sensors, ensuring the consistency of data within the same time window. Within each branch, the collected differential pressure data is divided into sliding analysis windows according to a set time length, for example, set to 10 minutes. All data during this period can be extracted as a sequence for volatility analysis. Subsequently, based on the system-designed differential pressure reference value corresponding to each data point during this period, a point-by-point difference comparison is performed to obtain a continuous record of the offset situation. Taking a certain branch as an example, the differential pressure recorded within 10 minutes is five different values, and the corresponding reference differential pressure is a set value. The difference between each actual differential pressure and the reference differential pressure is calculated to form a set of offset data for this period. In this set, the maximum offset value and the minimum offset value are extracted and defined as the fluctuation boundaries of this branch during the current period. To further eliminate interference factors, the original data needs to be preprocessed. By using methods such as filtering and moving average, after removing noise points and abnormal jumps, the differential pressure offset fluctuation range of this branch is finally determined and recorded as the basic parameter for subsequent linkage analysis.

[0029] S302: According to the differential pressure offset fluctuation range, call the flow data of adjacent branches during the corresponding time period, extract the difference change trend, determine the difference change rate sequence within a continuous time period, and generate the flow slope sequence of adjacent branches; Based on the obtained differential pressure fluctuation range of a certain branch, the flow monitoring data of multiple adjacent branches during the same time period can be further called. The flow data of each branch is also divided into sequences according to the time window, and the flow change amplitude during the current time period is calculated by using the point-by-point difference method. In the obtained flow change sequence, the increase and decrease situations between each time point are identified, the change rate is recorded, and it is organized into a sequence of flow slopes. For example, the flow of a certain adjacent branch gradually increases at five consecutive time points. By taking the difference between the values of two adjacent time points, the flow change speed can be obtained. After standardizing this sequence, its change trend can be analyzed. If it shows a stable upward trend as a whole, this trend is recorded as a typical feature. In the case of multiple branches, the flow change trends of each branch are extracted separately, and whether there are trend features that synchronize with the differential pressure offset of the target branch is evaluated. When identifying abnormal flow change amplitudes or sudden increases in rates, the association information between this period and the offset range is recorded to provide data support for subsequent trend judgment and imbalance identification.

[0030] S303: Based on the differential pressure offset fluctuation range and the flow slope sequence of adjacent branches, judge whether the differential pressure offset exceeds the fluctuation boundary, and check whether the slope sequence shows a continuous increasing trend to obtain the trigger flag for the branch imbalance trend; The specific calculation formula based on the pressure difference offset fluctuation range and the adjacent branch flow rate slope sequence is as follows: ; Wherein, represents the three - time standardized amplitude of the pressure difference offset, represents the cube of the flow rate value measured at the i - th time, represents the standard deviation of the flow rate value, v represents the number of measurements, i represents the sequence with indices from 1 to v, and the summation symbol with the associated letter i indicating the operation on all measured values; Detailed explanation of the formula and the derivation process of the formula calculation: In the formula, to calculate the three - time standardized amplitude of the pressure difference offset, considering the actual monitored values of the flow rate data, and cubing them to amplify the difference between data, and then normalizing by taking the cube root and the standard deviation of the flow rate. The calculation example with actual data is as follows: Suppose there is a set of measured flow rate data , with the unit of .

[0031] First, calculate the cube of each flow rate value to get ; Find the average of these cube values, ; Take the cube root, ; Calculation of the standard deviation of the flow rate value: First, obtain the average flow rate ; The sum of the squares of the differences between each flow rate value and the average value: ; Standard deviation ; The final three - time standardized amplitude of the pressure difference offset ; This result shows that the calculated three - time standardized amplitude of the pressure difference offset is 5.17, indicating that the pressure difference offset in the current flow rate data is relatively significant. This value is a key indicator for further judging whether the pressure difference offset exceeds the fluctuation boundary, and also provides a quantitative basis for examining whether the slope sequence shows a continuous increasing trend. The monitoring of specific flow rate values and the calculation of the standard deviation reflect the volatility and dispersion degree of the flow rate data, which are the core steps in the analysis of the pressure difference offset.

[0032] The specific steps of S4 are as follows: S401: Based on the branch imbalance trend trigger flag, call the current branch flow rate and the designed flow rate, extract the corresponding data of the branch, establish the corresponding relationship according to the number, identify the difference direction between the current value and the designed value, and combine the difference magnitude to form a set of flow rate offset directions and amplitudes, and generate a flow rate offset degree sequence; The branch imbalance trend trigger flag is used to identify the differences between each branch and the design parameters in the current state. It is necessary to separately extract the current flow values from multiple branches. For example, in a pipeline system with branch numbers A-1, B-1, and C-1, the current instantaneous flows are 1.25 liters per second, 0.95 liters per second, and 1.5 liters per second respectively. In the initial design state, the corresponding flow values of these branches are 1.00, 1.20, and 1.40 liters per second. After comparing the current flows with the design values one by one, the flow differences are calculated, specifically, A-1 deviates by 0.25 liters per second, B-1 deviates by -0.25 liters per second, and C-1 deviates by 0.10 liters per second. Combining with the branch numbers, a flow number mapping structure is established, and the current deviation trend of the branch is identified through the positive and negative directions of the difference. For example, A-1 is floating up, B-1 is floating down, and C-1 is also floating up. The deviation direction and the corresponding flow difference are combined into a data set, and further arranged in the order of the numbers to form a unified flow deviation sequence, which is used to reflect the current deviation situation and its amplitude of all branches. For example, in the formed sequence, the value corresponding to A-1 is positive, indicating that the flow exceeds the design value, B-1 is negative, indicating that it is lower than the design state, and C-1 is positive. Such a structure is widely used in the commissioning of air-conditioning hydraulic systems. For example, during the commissioning stage of the central air-conditioning system in a certain shopping mall, the branches with uneven water flow supply can be located through this deviation sequence, and it provides a basis for the subsequent analysis and adjustment of the differential pressure direction.

[0033] S402: According to the flow deviation degree sequence, call the differential pressure information of adjacent branches, form branch groups based on the branch numbers, judge the fluctuation direction of the differential pressure between adjacent moments, compare it with the deviation direction of the corresponding branch, and extract the direction consistency feature by combining the differential pressure change and the deviation amplitude to generate a differential pressure guiding coefficient sequence; According to the flow deviation degree sequence, paired structures need to be identified from the branches. For example, A-1 and B-1, B-1 and C-1 can be classified into two branch pairs. The pressure differences of each branch pair at two adjacent time points are collected. For example, the pressure difference between A-1 and B-1 is 30 kPa at the first time point and rises to 32 kPa at the second time point, indicating that the direction of the pressure difference of this pair is upward. The pressure difference between B-1 and C-1 drops from 28 kPa to 25 kPa, showing a downward direction. The direction of the pressure difference is compared with the flow deviation direction in the previous stage. If the pressure difference and the deviation are in the same direction, they are considered consistent; if they are in the opposite direction, they are marked as inconsistent. For example, if the flow of A-1 rises and the pressure difference rises, it is in the same direction as B-1. If the flow of C-1 rises while the pressure difference drops, it is considered to be in the opposite direction. The direction consistency of each pair of branches needs to be evaluated in combination with the deviation amplitude. For example, the deviation difference between A-1 and B-1 is 0.25 liters per second and the directions are consistent, so it can be assigned a value of 0.25 as a weight reference item. To avoid the interference of extreme values, a normalization factor can be used to adjust it to make it fall into a unified scale range, such as uniformly scaling it between 0 and 1. Finally, a sequence of pressure difference guiding coefficients is formed in the order of numbers to identify whether the flow deviation is affected by the pressure difference. In the adjustment link of the building heating and cooling system, through this sequence, it can be clearly pointed out whether the adjustment response conforms to the pressure difference change characteristics, which is the intermediate basis for dynamic adjustment.

[0034] S403: According to the pressure difference guiding coefficient sequence, extract the current valve opening value of the corresponding branch, determine whether the opening adjustment direction is consistent with the guiding direction, adjust the opening amplitude according to the guiding coefficient, set it as the corresponding mapping value when the directions are consistent, and set it as the reverse correction value when the directions are inconsistent, and generate a branch flow compensation instruction; The specific calculation formula for adjusting the opening amplitude according to the guiding coefficient is: ; Among them, represents the corrected opening adjustment amplitude of the j-th valve of the i-th branch, represents the guiding weight coefficient of the j-th valve of the i-th branch under the k-th measurement, represents the pressure difference value of the j-th valve of the i-th branch under the k-th measurement, represents the target deviation penalty coefficient of the j-th valve of the i-th branch, represents the expected reference opening value of the j-th valve of the i-th branch, represents the opening value of the j-th valve of the i-th branch at the previous moment, represents the opening adjustment sensitivity coefficient corresponding to the j-th valve of the i-th branch, represents the flow interference factor of the j-th valve of the i-th branch, represents the minimum dynamic suppression coefficient of the j-th valve of the i-th branch, which is used to characterize the correction stability of the valve when the disturbance is extremely small, and n represents the total number of measurement times; Parameter settings and calculation examples: Obtained through historical data analysis and real-time performance monitoring. For example, under specific working conditions, the weight may be determined by factors such as flow velocity and flow rate, and the set value is 0.95.

[0035] Obtained by real-time monitoring through a differential pressure sensor, assuming the specific value is 10 Pascals.

[0036] Set based on the analysis of the difference between the historical operation data of the device and the predetermined target opening. For example, it is set to 0.03, and this value is adjusted according to the stability of the device operation.

[0037] Set by the control system according to the optimized operation parameters. For example, it is set to 45 degrees.

[0038] Recorded by the real-time monitoring system. For example, the previous moment record is 44 degrees.

[0039] Reflects the response speed of the valve to the adjustment command, obtained through the valve characteristic table provided by the manufacturer. For example, it is set to 0.1.

[0040] Measures the influence degree of other factors on the valve performance, obtained based on fluid dynamics simulation, and the set value is 0.2.

[0041] Represents the correction stability of the valve under extremely small disturbances, set according to the dynamic test results of the valve. For example, 0.05.

[0042] Formula calculation example Assume there is 1 measurement : ; ; ; ; ; The result shows that under the current differential pressure and historical opening of the valve, the opening of the valve needs to be adjusted significantly, increasing by approximately 117.5 units to achieve the desired flow control effect. Further organizing this result can be directly used as the input of the control system for actual adjustment of the valve to ensure that the flow or pressure of the system reaches the predetermined operation parameters.

[0043] The specific steps of S5 are: S501: Based on the branch flow compensation instruction, call the time series data of the return water temperature and instantaneous flow on the ground source side, complete synchronous matching, screen continuous data, extract the corresponding relationship between flow change and temperature offset, and generate the branch offset correlation eigenvalue; Based on the branch flow compensation instruction, first identify the independent control units of each branch in the heating and cooling system. Through the system monitoring module, call the historical record data of the return water temperature and instantaneous flow on the ground source side. The data acquisition frequency can be set to once per minute and continue for a period of not less than seven days to form a time series. Synchronize the obtained temperature and flow data according to a unified time reference. The synchronization method generally adopts the timestamp pairing method to ensure that the temperature and flow at each moment correspond one by one. For missing data points, linear interpolation or spline interpolation can be used for repair. Subsequently, in the paired data, eliminate the data segments with a continuous duration of less than one hour and only retain the records with strong continuity. Gradually traverse the valid data through a sliding time window, and record the flow difference and temperature difference between the current time point and the previous time point at each step. These differences constitute the basic data points representing the relationship between flow change and temperature change. In a specific example, if the flow of a certain branch rises from 2.5 liters per second to 3.0 liters per second within a certain time period, and the return water temperature drops from 39.0 degrees Celsius to 38.2 degrees Celsius in the same time period, a corresponding set of data points can be identified, indicating that the temperature decreases during the flow increase of this branch. Traverse all data segments in turn, extract similar data points to construct a data set, and then statistically analyze all pairs of flow and temperature change points, use the correlation evaluation method to evaluate the relationship strength between the two, and record the representative strength value as the offset correlation eigenvalue of this branch. Finally, form the offset feature set of each branch in the entire system.

[0044] S502: According to the branch offset correlation eigenvalue, extract the flow change and temperature fluctuation amplitude of the branch, construct a frequency distribution diagram, count the fluctuation density in the frequency interval, and analyze the difference in combination with the curve trend to generate the degree of change of the branch fluctuation spectrum; After obtaining the offset correlation eigenvalues of each branch, further segmentally extract the flow rate change amplitude and temperature fluctuation degree of each branch. First, statistically calculate the maximum value, minimum value, and change range of the flow rate change within each branch, and at the same time obtain the amplitude range of the temperature fluctuation. Then divide these change values into several continuous intervals, and record the actual occurrence times within each interval to form a frequency distribution. For each flow rate change section, correspondingly count the frequency quantity of the temperature fluctuation within this section to construct a frequency distribution diagram. By using the horizontal axis to represent the flow rate change amplitude and the vertical axis to represent the occurrence frequency of the temperature change at this amplitude, the data density situation can be presented in the form of a bar chart or a heat map. For example, the temperature fluctuation occurred 20 times within the flow rate change range of 0.3 to 0.4 for a certain branch, and reached 30 times in the section of 0.4 to 0.5. Such a frequency distribution helps to depict the density of fluctuations at different change amplitudes. Further, by statistically calculating the fluctuation density of each section as the evaluation basis, average the fluctuation densities of all sections of each branch to obtain the overall fluctuation intensity, and then calculate the fluctuation center position to reflect the gravity trend of the fluctuations, providing auxiliary information for judging the concentrated direction of the branch operation changes. In addition, the smoothness degree of the change trend can be calculated in combination with the trend of the temperature fluctuation frequency curve, and the differential method is used to evaluate the amplitude change degree of the frequency change in each section. A higher change amount represents obvious fluctuations in the curve, and vice versa represents a relatively stable curve. Each branch outputs two characteristic values, namely the density and the fluctuation smoothness degree, as the basic data for the fluctuation spectrum change degree.

[0045] S503: Invoke the branch fluctuation spectrum change degree and the branch flow rate change, compare the change relationship between the flow rate distribution and the spectrum fluctuation, and combine the change trend of the energy efficiency utilization rate to screen the corresponding branches to generate a multi-source energy efficiency optimization map; After obtaining the fluctuation spectrum change degree and the flow rate change amplitude of each branch, it is necessary to make a horizontal comparison between the two to discover the trend matching relationship between them. Whether there is a trend of synchronous growth or alternating change can be analyzed through the comparison graph. In addition, introduce the change value of the energy efficiency utilization rate of each branch as the third index, which can be obtained by calculating the difference between the energy output and the energy consumption ratio of the branch within different time periods. Then, in combination with the change amplitudes of the flow rate and the spectrum, analyze the correlation degree between various parameters. By setting the minimum reference standard for the energy efficiency change, only retain those branches whose change amplitudes reach or exceed this standard during the evaluation period. If at the same time these branches show high-amplitude fluctuations in the fluctuation spectrum, it indicates that the branch may have strong optimization potential. Record the branches that meet the conditions as the target optimization objects, and construct a three-dimensional data map, with the flow rate change amplitude as the horizontal axis, the fluctuation spectrum degree as the vertical axis, and the change amplitude of the energy efficiency utilization rate as the vertical axis. All the selected branches correspond to a point position in the map. Identify the overall optimization direction and the priority order of resource allocation through the distribution of these points, and assist in the dynamic adjustment task execution and key path identification in the multi-branch energy efficiency management.

[0046] Please refer to Figure 2 , a ground-source heat pump intelligent control system for improving comprehensive energy efficiency based on the Internet of Things, including: The temperature trend matching module obtains the deviation value between the outlet temperature of the ground-source side pipeline and the circulating water flow of the user side, detects the trend directions of the two, calculates the cumulative value of the change in the soil temperature gradient, and generates a heat source switching delay correction amount according to the ratio of the direction differences among the three; The valve control priority scheduling module calls the heat source switching delay correction amount, obtains the valve action response period and the power signal response period, compares the difference in the overlapping interval of the two time axes, and generates a timing coordination priority; The branch imbalance identification module calls the timing coordination priority, obtains the branch pressure difference monitoring value, the designed pressure difference value and the real-time flow of adjacent branches, calculates the amplitude of the difference between the pressure difference and the designed value, determines whether the upper and lower limits of the pressure difference are exceeded, screens the sequence of continuously increasing slopes of the flow difference in adjacent branches, and generates a branch imbalance trend trigger flag according to the overlapping area of the two; The flow compensation control module calls the branch imbalance trend trigger flag, obtains the branch flow and the designed value, and the change direction of the pressure difference of adjacent branches, calculates the degree of consistency between the flow deviation direction and the pressure difference fluctuation direction, screens the branches to be adjusted and adjusts the valve opening direction and amplitude, and generates a branch flow compensation instruction; The energy efficiency map construction module calls the branch flow compensation instruction, obtains the branch flow distribution data and the ground-source side return water temperature spectrum, calculates the balance index and the fluctuation range, and generates a multi-source energy efficiency optimization map.

[0047] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any person skilled in the relevant art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent control method for ground source heat pump based on the Internet of Things to improve comprehensive energy efficiency, characterized in that: The following steps are involved: S1: Obtain the outlet temperature trend of the ground source side pipeline and the deviation trend of the circulating water flow rate on the user side, dynamically match the directions of the two, and generate the heat source switching delay correction value in combination with the cumulative change of the soil temperature gradient; S2: Calculate the offset direction of the heat source switching instruction and the valve action timing according to the heat source switching delay correction amount, adjust the power signal priority, embed the adjustment logic in the control cycle before the valve action, and generate the timing coordination priority; S3: According to the timing coordination priority, obtain the fluctuation range of the branch pressure difference deviation from the design value and the slope of the flow difference of adjacent branches, determine whether the pressure difference fluctuation exceeds the boundary and whether the flow slope increases continuously, and generate a branch imbalance trend trigger mark; S4: Based on the branch imbalance trend trigger mark, combined with the matching degree between the current branch flow and the design flow and the pressure difference fluctuation direction of the adjacent branches, the valve opening direction and amplitude are adjusted to generate a branch flow compensation instruction; S5: Based on the feedback of the branch flow compensation instruction execution, the return water temperature fluctuation spectrum on the ground source side and the branch flow distribution balance are aggregated to determine the synergistic relationship between branch compensation and energy efficiency fluctuation, and generate a multi-source energy efficiency optimization map.

2. According to claim 1, a method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things is characterized in that: The heat source switching delay correction includes the outlet temperature trend of the ground source side pipeline, the deviation trend of the user side circulating water flow, and the cumulative change of the soil temperature gradient. The timing coordination priority includes the heat source switching instruction, the valve action timing offset direction, and the power signal priority adjustment logic. The branch imbalance trend trigger mark includes the branch pressure difference deviation from the design value fluctuation range, the adjacent branch flow difference slope, the pressure difference fluctuation boundary state, and the flow slope continuity state. The branch flow compensation instruction includes the current branch flow and the design flow matching degree, the adjacent branch pressure difference fluctuation direction, the valve opening direction, and the valve amplitude adjustment parameter. The multi-source energy efficiency optimization map includes the ground source side return water temperature fluctuation spectrum, the branch flow distribution balance, and the branch compensation and energy efficiency fluctuation coordination relationship.

3. According to claim 1, a method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things is characterized in that: The specific steps of S1 are: S101: Obtain the outlet temperature of the ground source side pipeline and the circulating water flow data on the user side collected by the IoT sensor node, perform trend determination based on the temperature change direction and flow deviation direction in the same time period, match and identify the two types of trend directions, select the time period with the same trend direction, and generate the numerical value of the trend direction matching section; S102: According to the trend direction, the time period corresponding to the segment numerical value is matched, the soil temperature change data within the time period is extracted, the gradient difference between consecutive time points is identified, and the temperature change intensity of the entire segment is accumulated according to the order of the time series to generate the total amount of temperature gradient cumulative change; S103: Based on the total accumulated change of the temperature gradient and the numerical value of the trend direction matching section, a deviation judgment is made on the intensity relationship of the two data, and the difference is compared with the heat source switching sensitivity reference value, and a heat source switching delay correction amount is generated according to the degree of intensity difference.

4. According to claim 3, a method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things is characterized in that: The specific steps of S2 are: S201: Based on the heat source switching delay correction amount and the heat source switching instruction trigger time, valve action response time, and actuator power adjustment time collected in the current control cycle, identify the time offset between the heat source switching instruction and the valve action, determine the offset direction in combination with the response offset threshold, and generate an offset direction determination result; S202: according to the offset direction determination result, combined with the power signal adjustment amplitude, the switching node number, and the valve response period position, the response order of the power signal within the control cycle is set, and the priority value of each signal is marked to generate a signal priority distribution amount; S203: According to the signal priority distribution and the valve response period position, adjust the signal with a priority lower than the response threshold to the control period position before the valve is actuated, and re-encode the signal sequence to generate a timing coordination priority.

5. According to claim 4, a method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things is characterized in that: The specific steps of S3 are: S301: Based on the timing coordination priority, the real-time data of branch pressure difference and the design pressure difference reference value are obtained in combination with the IoT sensor network, the pressure difference deviation in the time period is detected, the pressure difference difference boundary interval of each branch in the continuous time period is extracted, and the pressure difference deviation fluctuation interval is generated; S302: according to the pressure difference deviation fluctuation interval, calling the flow data of the adjacent branches in the corresponding time period, extracting the difference change trend, determining the difference change rate sequence in the continuous time period, and generating the adjacent branch flow slope sequence; S303: Based on the pressure difference deviation fluctuation interval and the adjacent branch flow slope sequence, determine whether the pressure difference deviation exceeds the fluctuation boundary, and check whether the slope sequence shows a continuous increasing trend, and obtain a branch imbalance trend trigger mark.

6. The intelligent control method for ground source heat pump based on Internet of Things to improve comprehensive energy efficiency according to claim 5 is characterized in that: The specific calculation formula based on the pressure difference deviation fluctuation interval and the adjacent branch flow slope sequence is: ； in, represents the cubic normalized magnitude of the differential pressure excursion, represents the cube of the flow value measured at the i-th time, represents the standard deviation of the flow value, v represents the number of measurements, i represents the sequence indexed from 1 to v, and the summation symbol The associated letter i indicates that the calculation is performed on all measured values.

7. According to claim 5, a method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things is characterized in that: The specific steps of S4 are: S401: Based on the branch imbalance trend trigger flag, call the current branch flow and the design flow, extract the branch corresponding data, establish a corresponding relationship according to the number, identify the difference direction between the current value and the design value, and form a flow deviation direction and amplitude set based on the difference size, and generate a flow deviation degree sequence; S402: according to the flow deviation sequence, call the pressure difference information of adjacent branches, form a branch group according to the branch number pair, determine the fluctuation direction of the pressure difference between adjacent moments, compare it with the deviation direction of the corresponding branch, and extract the direction consistency feature in combination with the pressure difference change and the deviation amplitude, and generate a pressure difference guidance coefficient sequence; S403: According to the pressure difference guide coefficient sequence, the current valve opening value of the corresponding branch is extracted, and it is determined whether the opening adjustment direction is consistent with the guide direction. The opening amplitude is adjusted according to the guide coefficient. When the directions are consistent, it is set to the corresponding mapping value. When the directions are inconsistent, it is set to the reverse correction value, and a branch flow compensation instruction is generated.

8. The intelligent control method for ground source heat pump for improving comprehensive energy efficiency based on Internet of Things according to claim 7 is characterized in that: The specific calculation formula for adjusting the opening amplitude according to the guide coefficient is: ； in, represents the corrected opening adjustment amplitude of the jth valve of the i-th branch, represents the guidance weight coefficient of the jth valve of the i-th branch under the k-th measurement, represents the pressure difference value of the jth valve of the i-th branch under the k-th measurement, represents the target deviation penalty coefficient of the jth valve of the i-th branch, represents the expected reference opening value of the jth valve of the i-th branch, represents the opening value of the jth valve of the i-th branch at the last moment, represents the opening adjustment sensitivity coefficient corresponding to the jth valve of the i-th branch, represents the flow disturbance factor of the jth valve in the i-th branch, Represents the minimum dynamic suppression coefficient of the jth valve in the ith branch, which is used to characterize the corrected stability of the valve when the disturbance is extremely small. n represents the total number of measurements.

9. The intelligent control method for ground source heat pump for improving comprehensive energy efficiency based on Internet of Things according to claim 7 is characterized in that: The specific steps of S5 are: S501: Based on the branch flow compensation instruction, call the time series data of the return water temperature and the instantaneous flow at the ground source side, complete the synchronous matching, filter the continuous data, extract the corresponding relationship between the flow change and the temperature offset, and generate the branch offset associated characteristic value; S502: extracting the flow change and temperature fluctuation amplitude of the branch according to the branch offset associated characteristic value, constructing a frequency distribution diagram, statistically analyzing the fluctuation density of the frequency interval, and combining the curve trend analysis to generate the branch fluctuation spectrum change degree; S503: Call the branch fluctuation spectrum change degree and branch flow change, compare the change relationship between flow distribution and spectrum fluctuation, select the corresponding branch in combination with the change trend of energy efficiency utilization, and generate a multi-source energy efficiency optimization map.

10. An intelligent control system for ground source heat pumps based on the Internet of Things to improve comprehensive energy efficiency, characterized in that: According to any one of claims 1 to 9, the method for intelligently controlling a ground source heat pump for improving comprehensive energy efficiency based on the Internet of Things comprises: The temperature trend matching module obtains the deviation value between the outlet temperature of the ground source side pipeline and the circulating water flow rate on the user side, detects the trend direction of the two, calculates the cumulative value of the soil temperature gradient change, and generates the heat source switching delay correction value based on the ratio of the three direction differences; The valve control priority scheduling module calls the heat source switching delay correction amount, obtains the valve action response cycle and the power signal response cycle, compares the difference between the overlapping intervals of the two time axes, and generates a timing coordination priority; The branch imbalance identification module calls the timing coordination priority, obtains the branch pressure difference monitoring value, the design pressure difference value and the real-time flow of the adjacent branches, calculates the difference between the pressure difference and the design value, determines whether the pressure difference exceeds the upper and lower limits, and screens the continuous increasing sequence of the flow difference slope in the adjacent branches, and generates a branch imbalance trend trigger mark based on the overlap area of ​​the two; The flow compensation control module calls the branch imbalance trend trigger flag, obtains the branch flow and the design value, the direction of change of the pressure difference of the adjacent branches, calculates the consistency degree between the flow deviation direction and the pressure difference fluctuation direction, selects the branch to be adjusted and adjusts the valve opening direction and amplitude, and generates a branch flow compensation instruction; The energy efficiency map construction module calls the branch flow compensation instruction, obtains the branch flow distribution data and the return water temperature spectrum on the ground source side, calculates the balance index and the fluctuation range, and generates a multi-source energy efficiency optimization map.

Images