A control method for cooperative operation of a water chiller and a full variable frequency pump tower

Abstract

The application discloses a control method for cooperative operation of a water chilling unit and a full variable frequency pump tower, comprising: collecting personnel trajectory data, extracting trajectory space-time features, calculating a total space density, mapping the total space density value to a color space, thereby generating a static heat map, arranging the static heat map in time sequence to generate a space-time evolution heat map, splicing the trajectory space-time features into a feature vector, and predicting a predicted load impact by a prediction model; collecting temperature data and equipment heat generation, and calculating an original refrigeration capacity; correcting the original refrigeration capacity according to the predicted impact curve to obtain a target refrigeration capacity; adjusting the fan speed of the full variable frequency pump tower according to the target refrigeration capacity; and accurately predicting the load impact and calculating the target refrigeration capacity, so that the refrigeration system operation parameters can be adjusted more accurately according to the personnel trajectory and the equipment heat generation, overcooling or insufficient cooling can be avoided, unnecessary energy consumption can be reduced, and the cooperative control of the water chilling unit and the fan speed of the full variable frequency pump tower can be realized.

Description

Technical Field

[0001] This invention relates to the field of air conditioning and industrial refrigeration control technology, and in particular to a control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower. Background Technology

[0002] In modern architecture and industry, air conditioning and industrial refrigeration systems are key facilities for maintaining the stability of indoor environments and processes. Their operating efficiency and energy consumption directly affect overall performance and cost-effectiveness. Especially in scenarios requiring centralized cooling, such as data centers, large commercial buildings, and precision manufacturing, as well as in industries with high industrial cooling demands, such as chemical, pharmaceutical, and food processing, cold source stations (usually composed of multiple chillers, chilled water pumps, cooling water pumps, and cooling towers) have become the main energy-consuming units, often accounting for a significant proportion of the total energy consumption of a building or factory.

[0003] Traditional air conditioning and industrial refrigeration system designs generally employ fixed water supply temperatures and static control strategies. This approach may be able to maintain operation under stable and predictable loads. However, in practical applications, the dynamic changes in the distribution of people within buildings, fluctuations in equipment heat generation, and frequent load adjustments during industrial production all contribute to the high dynamism and uncertainty of refrigeration system loads. Particularly in large commercial buildings, office buildings, complexes, and industrial plants, frequent personnel movement and complex and variable equipment usage lead to significant load fluctuations in air conditioning and industrial refrigeration systems. Traditional control methods struggle to achieve precise temperature regulation and efficient energy utilization. Traditional temperature feedback control suffers from significant lag, failing to respond promptly to sudden increases or decreases in localized cooling loads caused by short-term gatherings or movement of people, or sudden load changes during industrial production. This lag not only results in inaccurate temperature control of the indoor environment or processes, affecting personnel comfort or product quality, but also easily leads to substantial energy waste. For example, in densely populated or equipment-rich areas, a sudden increase in cooling load may cause traditional systems to be unable to increase cooling capacity in time, resulting in excessively high temperatures; conversely, when the load decreases, excessive cooling may waste energy.

[0004] Meanwhile, traditional systems use a fixed water supply temperature, making it difficult to meet the differentiated needs of different areas or process sections. Different floors, functional areas, or process sections have different cooling load requirements. A fixed water supply temperature often leads to insufficient cooling in some areas and excessive cooling in others, causing hydraulic imbalance in the pipe network and energy waste. Summary of the Invention

[0005] This application provides a control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower. By accurately predicting load impacts and calculating the target cooling capacity, the operating parameters of the refrigeration system can be adjusted more accurately according to personnel trajectories and equipment heating conditions, avoiding over-cooling or under-cooling, adjusting the speed of water pumps and fans, reducing unnecessary energy consumption, achieving energy-saving operation, and monitoring and predicting personnel trajectories and load changes in real time to adjust the operation of the refrigeration system in advance, enabling the system to better cope with load fluctuations, improving system stability and reliability, achieving coordinated control of the chiller unit and the fully variable frequency pump tower fan speed, and reducing energy consumption at the same time.

[0006] This application provides a control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower, including:

[0007] S101: Collect personnel trajectory data, extract trajectory spatiotemporal features based on the personnel trajectory data, calculate the total spatial density based on the trajectory spatiotemporal features, map the total spatial density value to the color space to generate a static heat map, arrange the static heat map in chronological order to generate a spatiotemporal evolution heat map, use a temporal convolutional network as a prediction model, concatenate the trajectory spatiotemporal features into a feature vector, input the feature vector into the prediction model, and the prediction model outputs the predicted load impact;

[0008] The load impact includes the impact intensity and the impact curve. The impact intensity is the instantaneous cooling load value at a certain moment, reflecting the immediate load demand of the air conditioning system on the trajectory of people. The impact curve is the sequence of changes in the load impact intensity over time in the future.

[0009] The spatiotemporal characteristics of the trajectory include movement speed, dominant direction, personnel density, and density change.

[0010] S102 collects temperature data and equipment heat output, calculates the original cooling capacity based on the temperature data and equipment heat output, and corrects the original cooling capacity according to the predicted impact curve to obtain the target cooling capacity;

[0011] S103, the chiller unit is adjusted based on the predicted load impact, and the fan speed of the full variable frequency pump tower is adjusted according to the target cooling capacity. The predicted load impact and the target cooling capacity are used to achieve coordinated control of the chiller unit and the full variable frequency pump tower.

[0012] The fully variable frequency pump tower includes a chilled water pump, a cooling water pump, and a cooling tower;

[0013] The method for adjusting the speed of the fully variable frequency pump tower fan based on the target cooling capacity is as follows: calculate the chilled water flow rate based on the target cooling capacity, and adjust the speed of the chilled water pump based on the chilled water flow rate; calculate the cooling water flow rate based on the target cooling capacity, and adjust the speed of the cooling water pump and cooling tower fan based on the cooling water flow rate.

[0014] Preferably, the steps for adjusting the chiller unit based on predicted load shocks are as follows: setting the initial outlet water temperature and initial operating capacity of the chiller unit as control variables, setting temperature change gradient thresholds and capacity change gradient thresholds; if the shock curve in the predicted load shock shows an upward trend, the outlet water temperature of the chiller unit is reduced by calculating the difference between the initial outlet water temperature and the temperature change gradient threshold; simultaneously, the initial operating capacity is increased by adding the capacity change gradient threshold; if the upward trend of the shock curve slows down, the adjustment is stopped; if the shock curve in the predicted load shock shows a downward trend, the initial water temperature and initial operating capacity are adjusted in the opposite direction.

[0015] Preferably, the collaborative control method further includes:

[0016] S201: Collect elevator operation data and personnel distribution data. Based on the elevator operation data and personnel distribution data, use machine learning algorithms to construct a transportation distribution model. Based on the transportation distribution model, predict the vertical personnel distribution. Combine the vertical personnel distribution with the personnel trajectory data from step S101 to generate a three-dimensional personnel distribution heat map.

[0017] S202 divides the building into vertical units by floor, calculates the floor load value based on the elevator load-person distribution coupling coefficient in the load distribution model, and allocates chilled water based on the floor load value;

[0018] S203 monitors the actual load demand of each floor in real time based on the chilled water distribution results. Based on the real-time monitored load demand, it sends control rules to the chiller units and the full variable frequency pump tower, and performs coordinated control of the chiller units and the full variable frequency pump tower through the control rules.

[0019] Preferably, the method for generating a three-dimensional heat map of personnel distribution is as follows: based on a three-dimensional model of the building, different colors are used to represent the density of personnel distribution according to the distribution density of personnel on each floor and in each area.

[0020] Preferably, the steps for calculating the floor load value are as follows: the floor load value is equal to the sum of the personnel load and the heat transfer load, wherein the personnel load is equal to the product of the personnel load coefficient and the net inflow of the floor, and the heat transfer load is equal to the product of the heat transfer coefficient, the contact area, and the temperature of the upper and lower floors.

[0021] Preferably, the steps for sending the control rules are as follows: when the return water temperature of a certain floor is detected to be greater than the set value, it is determined that the cooling supply in that area is insufficient. While maintaining the current distribution ratio of the regulating valve opening, the system simultaneously sends an instruction to the chiller to increase the cooling capacity. At the same time, based on the feedback signal of system pressure drop or insufficient flow, it sends an acceleration instruction to the variable frequency pump tower to increase the pump speed. Conversely, if the return water temperature of a certain floor is continuously lower than the set value, it is determined that the cooling capacity is excessive. The system immediately sends an instruction to the chiller to reduce the cooling capacity and controls the variable frequency pump tower to decelerate to reduce the pump speed.

[0022] Preferably, the collaborative control method further includes:

[0023] S301 collects data from end devices, uses sensors to monitor the operation data of end devices, calculates the demand urgency index based on the operation data of end devices, and identifies the user with the highest demand urgency index as the most unfavorable user.

[0024] S302 obtains the minimum allowable water supply temperature based on the real-time operating conditions of the most unfavorable user, calculates the optimal water supply temperature based on the minimum allowable water supply temperature, and simultaneously performs coordinated control and adjustment of the water pump frequency and the chiller unit outlet water temperature based on the optimal water supply temperature.

[0025] Preferably, based on the optimal supply water temperature and the predicted load intensity, the specific heat capacity of water, the density of water, and the temperature difference are multiplied, where the temperature difference is the temperature difference between the optimal supply water temperature and the return water temperature. The total water flow required by the system is equal to the ratio of the predicted load intensity to the product of the specific heat capacity of water, the density of water, and the temperature difference. The total water flow required by the system in the pipe network is recalculated. Based on the known water flow of the original system, the water flow required by the new system, and the original pump frequency, and according to the proportional relationship between pump flow and frequency, the new pump frequency is calculated by multiplying the original pump frequency by the ratio of the new total water flow required by the system to the original system water flow. The system sends the calculated new pump frequency command to the variable frequency pump, causing the variable frequency pump to operate at the new frequency, thereby adjusting the pump's water flow. The system sends the optimal supply water temperature to the chiller unit. The chiller unit obtains the target temperature to be adjusted and, based on the received optimal supply water temperature command, adjusts its chilled water outlet temperature to the target temperature to achieve optimized operation of the entire air conditioning system.

[0026] One or more technical solutions provided in this application have at least the following technical effects or advantages: by accurately predicting load impacts and calculating target cooling capacity, the operating parameters of the refrigeration system can be adjusted more accurately according to personnel trajectories and equipment heating conditions, avoiding over-cooling or under-cooling, adjusting the speed of water pumps and fans, reducing unnecessary energy consumption, achieving energy-saving operation, real-time monitoring and prediction of personnel trajectories and load changes, adjusting the operation of the refrigeration system in advance, enabling the system to better cope with load fluctuations, improving system stability and reliability, achieving coordinated control of the speed of chiller units and fully variable frequency pump tower fans, while reducing energy consumption;

[0027] The optimization object is extended from two-dimensional plane to three-dimensional space, which solves the dynamic balance problem of vertical load unique to high-rise buildings. The load of each floor is accurately calculated through the floor load coupling equation, realizing precise layered control of vertical space, avoiding energy waste caused by uniform temperature setting of the whole building, thereby improving the overall energy efficiency of the system, and building a complete intelligent control system from plane to three-dimensional and from short-term to long-term, so as to realize the efficient operation of air conditioning system in high-rise buildings and improve energy utilization efficiency.

[0028] By determining the optimal water supply temperature, we can avoid the waste of energy in transmission and distribution caused by forced low-temperature operation to meet the needs of a few high-demand terminals. At the same time, we can take into account the differentiated needs of different regions, solve the problem of hydraulic imbalance in the pipeline network, and achieve optimized operation of the entire air conditioning system. The optimal water supply temperature adjusts the pump frequency and the chiller outlet water temperature to minimize the total power consumption of the pump and improve the system operating efficiency. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating a control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower according to the present invention.

[0030] Figure 2 This is a schematic diagram of the process for calculating floor load values ​​according to the present invention;

[0031] Figure 3 This is a schematic diagram of the process for obtaining the optimal water supply temperature according to the present invention. Detailed Implementation

[0032] To facilitate understanding of the present invention, a more complete description of this application will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the present invention.

[0033] It should be noted that the terms "vertical," "horizontal," "up," "down," "left," "right," and similar expressions used in this article are for illustrative purposes only and do not represent the only possible implementation.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0035] Example 1: Traditional temperature feedback control is lagging and cannot cope with rapidly changing personnel distribution. Short-term gatherings and movements of people can cause a sudden surge or sharp decrease in the local space's cooling load, resulting in lagging temperature control, decreased comfort, and energy waste.

[0036] Figure 1 This is a flowchart illustrating a control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower according to an embodiment of the present invention, including:

[0037] S101: Collect personnel trajectory data, extract spatiotemporal features of the trajectory data, construct a spatiotemporal evolution heat map based on the trajectory spatiotemporal features, construct a prediction model based on the trajectory spatiotemporal features, and predict load impact based on the prediction model; wherein, load impact includes impact intensity and impact curve;

[0038] Furthermore, sensors are deployed at entrances and passageways to capture data on rapid personnel movement. At intersections, the sensors are set to a high sampling frequency, while in open areas, the sampling frequency is set to a low frequency. The sensors acquire the position coordinates of personnel movement and the corresponding timestamp for each position. The collected data is stored, and median filtering is used to eliminate sensor data. The collected position data is then arranged according to timestamps to form personnel trajectory data. The stationary and movement states of the personnel trajectory are identified. The method for identifying stationary states is as follows: a minimum distance threshold is set experimentally. Several consecutive points in the personnel trajectory data are collected as sampling points. The movement distance of these consecutive sampling points is calculated using the Euclidean distance formula. The calculated movement distance is compared with the preset minimum distance threshold. If the calculated movement distance is less than or equal to the preset minimum distance threshold, the time period consisting of these sampling points is marked as a stationary state, and the coordinates of the stationary center are recorded. The station center coordinates are the average of several sampling points. The steps for identifying movement states are as follows: Based on the recorded station center coordinates, the trajectory segment between adjacent station center coordinates is defined as the movement state; the start and end timestamps and coordinate sequences are recorded; a movement distance threshold is set; the movement distance threshold is used to filter out minor fluctuations in the movement state; the obtained movement state is compared with the preset movement distance threshold, and moving supervisors whose movement distance is less than or equal to the threshold are removed; the spatiotemporal features of the trajectory are extracted based on the personnel trajectory data. These features include movement speed, dominant direction, personnel density, and density change. Movement speed includes instantaneous speed and average speed. Based on the personnel trajectory data, the instantaneous speed between adjacent movement data is calculated, and the average speed is calculated based on the movement speed. The calculation of the instantaneous speed and average speed mentioned above are existing technologies and will not be elaborated upon in this embodiment. The dominant direction is calculated using vector synthesis based on the personnel trajectory data, using the following formula: ,

[0039] Where θ is the dominant direction, M is the number of moving segments (i.e., the number of line segments into which the trajectory is divided), and i is the index of the moving segment. For the first The vertical coordinates of each point For the first The vertical coordinates of each point

[0040] This represents the total vertical displacement, used to accumulate the vertical displacement components of all moving segments, reflecting the overall northward or southward offset.

[0041] For the first The horizontal coordinates of the points For the first The horizontal coordinates of the points This represents the total horizontal displacement, used to accumulate the horizontal displacement components of all moving segments, reflecting the overall eastward or westward offset.

[0042] The monitoring area is divided into a uniform grid, and the population density in each grid is calculated using the following formula: ,

[0043] in, At time t, the grid The population density within the grid is given by t, where t is the current time, i and j are the two-dimensional spatial indices of the grid (the grid in the i-th row and j-th column), N is the total number of people, and k is the individual index, representing the k-th person. This represents the planar coordinate position of the k-th person at time t. Let be a grid region in two-dimensional space, uniquely identified by the index (i,j). I[·] is an indicator function that takes the value 1 when the condition is true and 0 otherwise.

[0044] The formula for calculating density change based on population density is: ,

[0045] in, For time intervals, For grid Rate of change in internal population density For time (i.e., at the next moment) the grid The number of people inside.

[0046] Based on the above spatiotemporal characteristics of the trajectory, the total density of the space is calculated using a Gaussian kernel function, with the following formula: ,

[0047] in, Let be the total density at (x,y) in space, and h be the bandwidth of the Gaussian kernel, used to control the smoothness of the density length. , Let be the coordinates of the i-th trajectory point. , Let be the coordinates of any point in the space where the density is to be calculated. Based on the total density calculated above, the calculated total density value is mapped to a color space, with different density values ​​corresponding to different color intensities, thus generating a static heat map. The generated static heat maps are then sorted according to time steps to generate a dynamic heat map, used to capture rapidly changing crowd flow. Load impact is predicted based on trajectory spatiotemporal characteristics. The load impact refers to the instantaneous or short-term fluctuation of the cooling load of the air conditioning system caused by the characteristics of people's movement trajectory (such as changes in speed, direction, and density). Its essence is the coupling effect of people's behavior and the thermal dynamics of the building environment, manifested as impact intensity and impact curve. Impact intensity is the instantaneous cooling load value at a certain moment, reflecting the immediate load demand of the air conditioning system on people's trajectory. Impact curve is the sequence of changes in load impact intensity over time in the future. The movement speed, dominant direction, people density, and density change obtained above are concatenated into a feature vector, with the formula: ,

[0048] in, Average moving speed, The dominant direction, i.e., the direction of movement, is represented by the polar coordinate angle of the line connecting the trajectory points. The rate of change in population density The current temperature of the area is collected by a temperature and humidity sensor. The air supply temperature is collected by the fan coil unit sensor. H is the trajectory-load impact transfer function. A temporal convolutional network (TCN) is used as the prediction model. The collected trajectory spatiotemporal features are divided into training set, validation set and test set. Based on the real-time collected trajectory data, the trajectory spatiotemporal features are recalculated. The updated trajectory spatiotemporal features are concatenated into a feature vector and input into the trained prediction model. The prediction model outputs the dynamic load impact (impact intensity and impact curve) for the future time period based on the built-in method.

[0049] S102 collects temperature data and equipment heat output, calculates the original cooling capacity based on the temperature data, personnel density data, and equipment heat output, and corrects the original cooling capacity based on the predicted load impact to obtain the target cooling capacity;

[0050] Specifically, temperature sensors are installed indoors, outdoors, and at the air conditioning vents to collect temperature data. The heat generation of the equipment is calculated by monitoring the real-time power consumption of devices such as lighting and computers. Based on the total spatial density calculated in step S101, the original cooling capacity is calculated using the following formula: ,

[0051] in, This is the original cooling capacity. The target temperature setpoint for the refrigeration system. The actual temperature of the current environment. This is the temperature difference coefficient, representing the amount of cooling required to lower the temperature by 1°C. Population density, which is the number of people per unit area. The heat dissipation coefficient represents the amount of heat dissipated by each person. Total heat generation refers to the heat generated by all equipment operating within the area. This is the equipment heating adjustment coefficient, used to correct the contribution of equipment heating to the cooling capacity. It is used to correct the original cooling capacity based on the load impact curve output by the prediction model.

[0052] The formula is: ,in, For the target cooling capacity, 原始 This is the original cooling capacity. This is the load impact curve.

[0053] S103, the chiller unit is adjusted based on the predicted load impact, and the speed of the fully variable frequency pump tower fan is adjusted according to the target cooling capacity. The predicted load impact and the target cooling capacity are used to achieve coordinated control of the chiller unit and the fully variable frequency pump tower. The method for adjusting the speed of the fully variable frequency pump tower fan (chilled water pump, cooling water pump, and cooling tower fan according to the target cooling capacity) is as follows: the chilled water flow rate is calculated based on the target cooling capacity, and the speed of the chilled water pump is adjusted based on the chilled water flow rate; the cooling water flow rate is calculated based on the target cooling capacity, and the speed of the cooling water pump and cooling tower fan is adjusted based on the cooling water flow rate.

[0054] Furthermore, adjustments are made to the chiller unit based on predicted load shocks. The chiller unit's outlet water temperature and operating capacity are set as control variables, along with temperature and capacity change gradient thresholds. If the shock curve in the predicted load shock shows an upward trend, the chiller unit's outlet water temperature is reduced. The difference between the initial outlet water temperature and the temperature change gradient threshold is calculated, lowering the initial outlet water temperature to enhance cooling capacity. Simultaneously, the initial operating capacity is increased by adding the capacity change gradient threshold, thereby increasing the main unit's output power. If the upward trend of the shock curve slows down, the adjustment stops. If the shock curve in the predicted load shock shows a downward trend, the outlet water temperature and operating capacity are adjusted in the opposite direction to avoid excessive cooling and energy waste. The fan speed of the fully variable frequency pump tower (including chilled water pumps, cooling water pumps, and a cooling tower) is adjusted according to the target cooling capacity. The required chilled water flow rate is calculated based on the target cooling capacity and the chiller unit's outlet water temperature.

[0055] The formula is: ,in, The chilled water flow rate represents the amount of water that the chilled water circulation system needs to deliver at time t to meet the cooling demand of the terminal equipment (air conditioning terminals). The target cooling capacity represents the amount of cooling the system needs to provide to the terminal devices at time t, and is determined by the population density and indoor temperature. Specific heat capacity of water, representing the amount of heat required to raise the temperature of a unit mass of water by 1°C. ρ is the density of water, representing the mass of water per unit volume. The chilled water supply and return temperature difference refers to the temperature difference between the chilled water flowing out of the chiller unit (supply water) and returning to the chiller unit (return water). The chilled water pump speed is adjusted using a frequency converter based on the chilled water flow rate.

[0056] The formula is: ,

[0057] in, This refers to the real-time speed of the chilled water pump, indicating the pump speed adjusted by the frequency converter at time t. The rated speed of the chilled water pump indicates the design speed of the pump under rated operating conditions. This refers to the chilled water flow rate. The rated flow rate of the chilled water pump indicates the flow rate corresponding to the pump's rated speed; the cooling water flow rate is calculated based on the target cooling capacity.

[0058] The formula is: ,in, The cooling water flow rate represents the amount of water that the cooling water circulation system needs to deliver at time t to remove heat from the chiller unit's condenser. This represents the total heat dissipation of the chiller unit, indicating the total heat that the chiller unit's condenser needs to release, including the target cooling capacity and the power consumption of the main unit. = + , The power consumption of the main unit represents the electrical energy consumed by the chiller unit's compressor, fan, and other equipment during operation. The cooling water supply and return temperature difference refers to the temperature difference between the cooling water flowing out of the chiller unit (supply water) and the temperature returning to the cooling tower (return water). Similarly, the cooling water flow rate is used to adjust the cooling water pump speed.

[0059] At the same time, through the function Adjust the cooling tower fan speed, among which, This refers to the cooling tower fan speed. The setting is the outlet temperature of the cooling water. If the outlet temperature of the cooling water is higher than the set value, increase the fan speed to enhance heat dissipation; if the temperature difference between the supply and return cooling water is too large, reduce the fan speed or check if the water flow is insufficient.

[0060] The technical solutions described in the above embodiments of this application have at least the following technical effects or advantages: By accurately predicting load impacts and calculating target cooling capacity, the operating parameters of the refrigeration system can be adjusted more accurately according to personnel trajectories and equipment heating conditions, avoiding over-cooling or under-cooling, adjusting the speed of water pumps and fans, reducing unnecessary energy consumption, achieving energy-saving operation, real-time monitoring and prediction of personnel trajectories and load changes, and advance adjustment of the refrigeration system operation, enabling the system to better cope with load fluctuations, improving system stability and reliability, achieving coordinated control of the speed of chiller units and fully variable frequency pump tower fans, and reducing energy consumption.

[0061] Example 2: Example 1 only considered the planar distribution of people, which could not predict the instantaneous load impact generated when an elevator concentrates a large number of people to a specific floor in a short period of time, resulting in lag in the cooling response of the vertical space and energy waste. This example establishes a model of elevator carrying capacity and personnel distribution, extending the two-dimensional planar prediction to a three-dimensional spatiotemporal collaborative prediction system of horizontal movement + vertical transportation, thereby achieving accurate prediction and control of the building's vertical space load, such as... Figure 2 As shown.

[0062] S201: Collect elevator operation data and personnel distribution data, construct a transportation distribution model based on the elevator operation data and personnel distribution data, predict the vertical personnel distribution based on the transportation distribution model, and combine the vertical personnel distribution with the personnel trajectory data in step S101 to generate a three-dimensional personnel distribution heat map.

[0063] Specifically, a data acquisition module is installed in the elevator control system to collect elevator operation data, including elevator load change data, maximum load data, number of stops, and total number of runs. The number of stops refers to the number of times the elevator stops within a certain period, and the total number of runs is the total frequency of elevator operations from the start of operation to the current moment. Infrared sensors PTC-08 are installed at floor entrances and elevator lobbies to collect personnel distribution data. These sensors detect personnel entry and exit by sensing infrared radiation emitted by the human body. Based on the personnel entry and exit information collected by the infrared sensors, the net inflow of personnel on each floor is calculated after data processing; this is the difference between the number of people entering and leaving a certain floor within a certain period. Simultaneously, temperature sensors are installed on each floor to collect temperature data for each floor of the building. The collected elevator operation data and personnel distribution data are preprocessed to remove outliers and noise, and the cleaned data is normalized. A transport distribution model is constructed using machine learning algorithms (neural networks) with historical elevator operation data and personnel distribution data as the training set.

[0064] The formula for calculating the elevator transport-personnel distribution coupling coefficient in the transport distribution model is as follows: ,

[0065] in, This is the elevator load-person distribution coupling coefficient, used to measure the degree of coupling between elevator load conditions and person distribution. Its value reflects the degree of matching between elevator operating status and personnel flow demand. Load change refers to the change in the elevator's load over a specific period of time. The maximum load capacity of the elevator is the maximum weight that the elevator can safely bear as specified in its design; it is a fixed parameter. For the number of stops, The total number of elevator runs. The weighting is based on peak hours, which typically refer to periods of high pedestrian traffic and frequent elevator use. The weighting is based on normal time periods, which are periods with relatively low pedestrian traffic and lower elevator usage frequency compared to peak periods. , , The parameter weights obtained through machine learning training are used to adjust the influence of load change rate, stop frequency ratio, and peak running time ratio on the elevator carrying-personnel distribution coupling coefficient. The trained model is fully validated using an independent test set. The data in the test set is input into the model, and the error between the model's prediction and the actual value is calculated. Based on the validation results, the model is optimized in a targeted manner.

[0066] Vertical passenger distribution is predicted based on a passenger load distribution model. Specifically, this involves considering the elevator's direction of travel (up or down) and the floors it stops at, as well as the current passenger load on each floor.

[0067] The net inflow of people to each floor is calculated using a function, which is: ,

[0068] in, This refers to the net inflow of people to each floor. The elevator carrying capacity-personnel distribution coupling coefficient. The threshold is used to distinguish between low and high coupling. , The slope coefficient, This indicates the marginal impact of the elevator carrying capacity-personnel distribution coupling coefficient on personnel flow when the elevator's carrying capacity is not fully utilized. This indicates the marginal inhibition effect of the elevator load-person distribution coupling coefficient on personnel flow when the elevator is close to saturation. , For the intercept term, This represents the baseline passenger flow when there is no impact from elevator transportation. The baseline passenger flow under elevator saturation conditions is used. For example, when the elevator stops at a floor and goes up, the model may calculate how many people will enter the elevator based on the current passenger load and the remaining load of the elevator, thus predicting the net passenger outflow for that floor. Conversely, when the elevator goes down and stops, the net passenger inflow for that floor is predicted. The starting position and direction of movement of people on each floor are determined based on the personnel trajectory data in step S101. Based on the net passenger inflow for each floor, the distribution of people in different areas of each floor is further refined. For example, if the net passenger inflow for a floor is positive and the personnel trajectory data shows that people are moving from the elevator lobby to the office area, it can be inferred that the personnel density in the office area will increase. The personnel distribution information of different areas of each floor is mapped to the three-dimensional space of the building. The specific method is as follows: based on the three-dimensional model of the building, different colors are used to represent the density of personnel distribution on each floor and in each area, generating a three-dimensional personnel distribution heat map.

[0069] S202 divides the building into vertical units by floor, calculates the floor load value based on the elevator load-person distribution coupling coefficient in the load distribution model, and allocates chilled water based on the floor load value;

[0070] Furthermore, the building is divided into vertical units by floor, each unit including floor area, contact area between upper and lower floors, and floor function type. Information on wall materials, floor height, and distance between adjacent floors is collected. The thermal conductivity of the building materials is divided by the thickness of the floor slab between floors to obtain the inter-floor heat transfer coefficient. A floor occupancy load coefficient is set according to the floor function type: 150 for office floors, 250 for commercial floors, and 80 for residential floors. The floor load value is calculated based on the elevator load-occupancy distribution coupling coefficient. The floor load value equals the occupancy load plus the heat transfer load. The occupancy load equals the occupancy load coefficient multiplied by the floor net inflow, and the heat transfer load equals the heat transfer coefficient multiplied by the contact area multiplied by the temperature of the upper and lower floors. Chilled water is allocated based on the obtained floor load values, establishing a linear relationship between valve opening and floor load. An initial valve opening is set as the baseline state under no-load conditions.

[0071] The formula for calculating the valve opening based on the floor load value is as follows: Where θ is the valve opening degree. The initial valve opening is given by k, which is the adjustment coefficient, indicating that the valve opening increases by 1% for every 10% increase in the floor load relative to the rated load. This represents the floor load value. Using the rated load as a baseline value, and based on the valve opening formula described above, the valve opening is updated periodically based on the floor load value. If the floor load value increases, the valve opening increases proportionally; if the floor load value decreases, the valve opening decreases. For high-load or high-priority floors, the valve opening is additionally increased using a weighted coefficient.

[0072] The formula is: ,in, The adjusted valve opening represents the final opening value of the chilled water distribution valve on the i-th floor after priority weighting, while θ represents the valve opening before adjustment, reflecting the initial impact of the current floor load on the valve opening. This is a priority weighting coefficient used to control the increase in the opening size of high-load floors relative to other floors. This represents the floor load value. The system measures the total real-time load for all floors. It verifies the chilled water distribution effect by measuring the return water temperature at the terminal equipment. Temperature sensors are installed at the air conditioning terminals on each floor to monitor the return water temperature in real time. An allowable deviation range is set. If the return water temperature continuously deviates from the set value, valve fine-tuning is triggered. The valve fine-tuning method is as follows: when the return water temperature is greater than the set allowable deviation range, it indicates insufficient cooling, so the valve opening is increased; when the return water temperature is less than the set allowable deviation range, it indicates excessive cooling, so the valve opening is decreased.

[0073] S203 monitors the actual load demand of each floor in real time based on the chilled water distribution results. Based on the real-time monitored load demand, it sends control rules to the chiller units and the full variable frequency pump tower, and performs coordinated control of the chiller units and the full variable frequency pump tower through the control rules.

[0074] Specifically, after optimizing chilled water distribution and dynamically adjusting the opening of vertical riser regulating valves, the system collects chilled water return temperature, pipeline flow rate, and system pressure data for each floor based on temperature, flow rate, and pressure sensors. Combined with valve opening results, the actual cooling load demand for each floor is obtained. Based on the real-time monitored load demand, control rules are sent to the chiller units and the variable frequency pump tower. The steps for sending control rules are as follows: when the return water temperature of a certain floor is continuously higher than the set value (e.g., exceeding the set temperature + 1℃), it is determined that the cooling capacity supply in that area is insufficient. At this time, the central control system, while maintaining the current regulating valve opening distribution ratio, simultaneously sends a command to the chiller units to increase cooling capacity (e.g., increase compressor frequency or refrigerant flow). Simultaneously, based on feedback signals of system pressure drop or insufficient flow, a command is sent to the variable frequency pump tower. The system accelerates the pump speed to ensure that chilled water is distributed to each floor as needed, thereby meeting the cooling capacity shortage by increasing both cooling capacity output and circulation flow. Conversely, if the return water temperature on a floor remains below the set value (e.g., below the set temperature -1°C), it is determined that there is excess cooling capacity. The system immediately sends a command to the chiller to reduce the cooling capacity and controls the variable frequency pump tower to decelerate, thereby reducing the pump speed, cooling capacity output, and circulation flow, and avoiding energy waste. Throughout the entire control process, the central control system dynamically corrects the control commands by comparing the return water temperature, flow rate, and pressure parameters with the set thresholds in real time. This ensures that the cooling capacity output of the chiller and the circulation flow of the variable frequency pump tower are always precisely matched with the actual load demand of each floor, ultimately achieving efficient and stable operation of the air conditioning system under different operating conditions.

[0075] The technical solutions in the above embodiments of this application have at least the following technical effects or advantages: the optimization object is extended from a two-dimensional plane to a three-dimensional space, which solves the dynamic balance problem of vertical load unique to high-rise buildings. The load of each floor is accurately calculated through the floor load coupling equation, realizing precise layered control of the vertical space, avoiding energy waste caused by uniform temperature setting of the whole building, thereby improving the overall energy efficiency of the system, constructing a complete intelligent control system from plane to three-dimensional and from short-term to long-term, realizing the efficient operation of the air conditioning system of high-rise buildings, and improving energy utilization efficiency.

[0076] Example 3: The three-dimensional spatiotemporal collaborative prediction system in Examples 1 and 2 above achieves accurate prediction and control of the building's vertical spatial load. However, in actual operation, we found that using a fixed water supply temperature still makes it difficult to take into account the differentiated needs of different areas, leading to hydraulic imbalance in the pipe network and energy waste. This example is based on the optimal water supply temperature calculation method for identifying the most unfavorable user. By dynamically adjusting the water supply temperature and the operating parameters of the water pump and chiller unit, the entire air conditioning system is further optimized. Figure 3 As shown.

[0077] S301 collects data from end devices, uses sensors to monitor the operating data of end devices, and identifies the most disadvantaged user based on the operating data of end devices;

[0078] Specifically, data is collected from various terminal devices (fan coil units and air conditioning units) within the building. This data includes the location information of the terminal devices and the heat transfer characteristic parameters of the heat exchangers. The location data refers to their specific location within the building, such as whether they are located on the east or west side of the floor, near the stairwell or elevator shaft, etc. The inlet and outlet fluid temperatures and flow rates of the heat exchangers under specific operating conditions are measured using the heat balance method. The structural parameters of the heat exchangers, such as heat exchange area, heat exchange tube material, and tube diameter, are recorded. The heat transfer capacity and heat transfer coefficient of the heat exchangers are calculated using heat transfer formulas. Sensors are used to monitor the operating data of the terminal devices, including personnel distribution at different times, valve opening data, flow rate data, temperature difference data, and heat exchange efficiency data. Based on the collected terminal equipment data and operational data, a pipeline network model (EPANET-Thermal) is constructed based on fluid mechanics and thermodynamics principles. This model divides the pipeline network system into multiple nodes and pipe segments. It describes the hydraulic and thermal characteristics of the network system by establishing node flow balance equations, pipe segment pressure drop equations, and heat balance equations. The collected pipeline network and terminal equipment data are input into the model, and the model is validated using monitored actual operational data. If significant errors exist, the model is adjusted and optimized. The collected data and the predicted load impact curve from step S101 are standardized, and the demand urgency index is calculated using the following formula: + d+ ,

[0079] in, This is the demand urgency index, reflecting the degree of urgency of demand at each terminal. A higher demand urgency index indicates a more urgent need for that terminal device at the current moment, and a potentially less than ideal operating condition. The adjusted valve opening is given by d, where d represents the geographical and hydraulic distance. This is the impact strength value. , , The weighting coefficients for valve opening, geographical and hydraulic distance, and impact intensity are respectively used to calculate the demand urgency index. An urgency threshold is then set. If the demand urgency index exceeds the preset threshold, the terminal device is identified as a potentially most unfavorable user. The urgency threshold is determined based on actual conditions and historical data. By statistically analyzing the demand urgency indices of a large number of terminal devices, the index range of devices with high demand urgency can be identified. Based on this, an urgency threshold is set. All potentially most unfavorable users are then sorted from highest to lowest demand urgency index, and the one or several terminals with the highest index are selected as the most unfavorable users at the current moment.

[0080] S302 obtains the minimum allowable water supply temperature based on the real-time operating conditions of the most unfavorable user, calculates the optimal water supply temperature based on the minimum allowable water supply temperature, and simultaneously controls and adjusts the water pump frequency and the chiller outlet water temperature based on the optimal water supply temperature.

[0081] Furthermore, based on the most unfavorable user identified in step S301, real-time flow rate, supply and return water temperature difference, and heat exchange efficiency data for the most unfavorable user are obtained from the system. Real-time flow rate reflects the water demand of the most unfavorable user at a given moment; the supply and return water temperature difference reflects the temperature change of the water before and after passing through the heat exchange equipment of the most unfavorable user; and the heat exchange efficiency reflects the ability of the heat exchange equipment to transfer heat from the water to cold. The minimum allowable supply water temperature is calculated using the heat balance equation. According to the principle of heat balance, the specific heat capacity of water is known to be a fixed value, and the mass flow rate is related to the flow rate obtained from the system. Based on the cold load required by the most unfavorable user, combined with the heat balance relationship (i.e., the heat released by the water equals the heat required by the most unfavorable user), the minimum allowable supply water temperature is calculated. The required cooling capacity (the heat released by the water is the product of the water's specific heat capacity, mass flow rate, and the supply and return water temperature difference) is calculated. Given the required cooling capacity, flow rate, and supply and return water temperature difference, the minimum allowable supply water temperature is calculated. Using this minimum allowable supply water temperature as a starting point, the initial supply water temperature is set equal to the minimum allowable supply water temperature. The pump power consumption is iteratively calculated, increasing the supply water temperature setpoint by 0.5℃ each time. For each assumed higher supply water temperature, simulation calculations are performed using a pipe network model. Under the premise that all terminals (especially the most unfavorable users) can still obtain the required cooling capacity, the required pump head and flow rate are determined. Under similar operating conditions, the pump power consumption is positively correlated with the cube of the flow rate and the pump head. Based on this relationship, and combining the head and flow rate obtained from each simulation calculation, the total power consumption of the water pump at different supply temperatures is calculated. Among all the calculated total power consumption of the water pump corresponding to different supply temperatures, the supply temperature that minimizes the total power consumption of the water pump is found; this temperature is the optimal supply temperature. Based on the optimal supply temperature and the predicted load intensity, the specific heat capacity of water, the density of water, and the temperature difference are multiplied. The temperature difference is the temperature difference between the optimal supply temperature and the return water temperature. The total water flow rate required by the system is equal to the ratio of the predicted load intensity to the product of the specific heat capacity of water, the density of water, and the temperature difference. The total water flow rate required by the system in the pipe network is recalculated. Based on the known water flow rate of the original system and the new system... The system calculates the new pump frequency based on the relationship between pump flow rate and rotational speed (frequency), i.e., the ratio of flow rates at different times is equal to the ratio of corresponding rotational speeds (frequency). The calculation method is to multiply the original pump frequency by the ratio of the new total system flow rate to the original system flow rate. The system sends the calculated new pump frequency command to the variable frequency pump, causing the variable frequency pump to operate at the new frequency, thereby adjusting the pump's water flow rate. The system also sends the optimal supply water temperature to the chiller unit. The chiller unit obtains the target temperature that needs to be adjusted and adjusts its chilled water outlet temperature to the target temperature according to the received optimal supply water temperature command, thereby achieving optimized operation of the entire air conditioning system.

[0082] The technical solutions in the above embodiments of this application have at least the following technical effects or advantages: by obtaining the optimal water supply temperature, the energy waste caused by forced low-temperature operation to meet the needs of a few high-demand terminals is avoided. At the same time, the differentiated needs of different regions are taken into account, the hydraulic imbalance of the pipeline network is solved, and the optimized operation of the entire air conditioning system is achieved. The optimal water supply temperature adjusts the pump frequency and the chiller outlet water temperature, so as to minimize the total power consumption of the pump and improve the system operating efficiency.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A control method for the coordinated operation of a chiller unit and a fully variable frequency pump tower, characterized in that, include: S101: Collect personnel trajectory data, extract trajectory spatiotemporal features based on the personnel trajectory data, calculate the total spatial density based on the trajectory spatiotemporal features, map the total spatial density value to the color space to generate a static heat map, arrange the static heat map in chronological order to generate a spatiotemporal evolution heat map, use a temporal convolutional network as a prediction model, concatenate the trajectory spatiotemporal features into a feature vector, input the feature vector into the prediction model, and the prediction model outputs the predicted load impact; The load impact includes the impact intensity and the impact curve. The impact intensity is the instantaneous cooling load value at a certain moment, reflecting the immediate load demand of the air conditioning system on the trajectory of people. The impact curve is the sequence of changes in the load impact intensity over time in the future. The spatiotemporal characteristics of the trajectory include movement speed, dominant direction, personnel density, and density change. S102 collects temperature data and equipment heat output, calculates the original cooling capacity based on the temperature data and equipment heat output, and corrects the original cooling capacity according to the predicted impact curve to obtain the target cooling capacity; S103, the chiller unit is adjusted based on the predicted load impact, and the fan speed of the full variable frequency pump tower is adjusted according to the target cooling capacity. The predicted load impact and the target cooling capacity are used to achieve coordinated control of the chiller unit and the full variable frequency pump tower. The fully variable frequency pump tower includes a chilled water pump, a cooling water pump, and a cooling tower; The method for adjusting the speed of the fully variable frequency pump tower fan according to the target cooling capacity is as follows: calculate the chilled water flow rate according to the target cooling capacity, and adjust the speed of the chilled water pump based on the chilled water flow rate; calculate the cooling water flow rate according to the target cooling capacity, and adjust the speed of the cooling water pump and cooling tower fan according to the cooling water flow rate. Collaborative control methods also include: S201: Collect elevator operation data and personnel distribution data. Based on the elevator operation data and personnel distribution data, use machine learning algorithms to construct a transportation distribution model. Based on the transportation distribution model, predict the vertical personnel distribution. Combine the vertical personnel distribution with the personnel trajectory data from step S101 to generate a three-dimensional personnel distribution heat map. S202, the building is divided into vertical units by floor. The floor load value is calculated based on the elevator load-personnel distribution coupling coefficient in the load distribution model. Chilled water is then allocated based on the floor load value. The steps for calculating the floor load value are as follows: the floor load value equals the sum of the personnel load and the heat transfer load. The personnel load equals the product of the personnel load coefficient and the net inflow of the floor. The heat transfer load equals the product of the heat transfer coefficient, the contact area, and the temperatures of the upper and lower floors. Chilled water is allocated based on the obtained floor load value, i.e., a linear relationship is established between valve opening and floor load. At regular intervals, the valve opening is updated based on the floor load value. If the floor load value increases, the valve opening increases proportionally; if the floor load value decreases, the valve opening decreases. For high-load or high-priority floors, the valve opening is additionally increased using a weighted coefficient, as shown in the formula: ,in, The adjusted valve opening represents the final opening value of the chilled water distribution valve on the i-th floor after priority weighting, while θ represents the valve opening before adjustment, reflecting the initial impact of the current floor load on the valve opening. This is a priority weighting coefficient used to control the increase in the opening size of high-load floors relative to other floors. This represents the floor load value. The real-time total load for all floors; S203 monitors the actual load demand of each floor in real time based on the chilled water distribution results. Based on the real-time monitored load demand, it sends control rules to the chiller units and the full variable frequency pump tower, and performs coordinated control of the chiller units and the full variable frequency pump tower through the control rules.

2. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 1, characterized in that, The formula for calculating the original cooling capacity is: ,in, This is the original cooling capacity. The target temperature setpoint for the refrigeration system. The actual temperature of the current environment. This is the temperature difference coefficient, representing the amount of cooling required to lower the temperature by 1°C. Population density, which is the number of people per unit area. The heat dissipation coefficient represents the amount of heat dissipated by each person. Total heat generation refers to the heat generated by all equipment operating within the area. This is the equipment heating adjustment coefficient, used to correct the contribution of equipment heating to cooling capacity.

3. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 2, characterized in that, The formula for calculating the target cooling capacity is: ,in, For the target cooling capacity, 原始 This is the original cooling capacity. This is the load impact curve.

4. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 1, characterized in that, The steps for adjusting the chiller unit based on predicted load shocks are as follows: Set the initial outlet water temperature and initial operating capacity of the chiller unit as control variables, set temperature change gradient thresholds and capacity change gradient thresholds. If the shock curve in the predicted load shock shows an upward trend, reduce the outlet water temperature of the chiller unit by calculating the difference between the initial outlet water temperature and the temperature change gradient threshold. At the same time, increase the initial operating capacity by adding the capacity change gradient threshold. If the upward trend of the shock curve slows down, stop the adjustment. If the shock curve in the predicted load shock shows a downward trend, reverse the adjustment of the initial water temperature and initial operating capacity.

5. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 1, characterized in that, The method for generating a 3D heat map of people distribution is as follows: based on the 3D model of the building, different colors are used to represent the density of people distribution on each floor and in each area.

6. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 1, characterized in that, The steps for sending control rules are as follows: When the return water temperature of a certain floor is detected to be greater than the set value, it is determined that the cooling supply in that area is insufficient. While maintaining the current distribution ratio of the regulating valve opening, the system simultaneously sends an instruction to the chiller to increase the cooling capacity. At the same time, based on the feedback signal of system pressure drop or insufficient flow, it sends an acceleration instruction to the variable frequency pump tower to increase the pump speed. Conversely, if the return water temperature of a certain floor is continuously lower than the set value, it is determined that the cooling capacity is excessive. The system immediately sends an instruction to the chiller to reduce the cooling capacity and controls the variable frequency pump tower to decelerate to reduce the pump speed.

7. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 1, characterized in that, Collaborative control methods also include: S301 collects data from end devices, uses sensors to monitor the operation data of end devices, calculates the demand urgency index based on the operation data of end devices, and identifies the user with the highest demand urgency index as the most unfavorable user. S302 obtains the minimum allowable water supply temperature based on the real-time operating conditions of the most unfavorable user, calculates the optimal water supply temperature based on the minimum allowable water supply temperature, and simultaneously performs coordinated control and adjustment of the water pump frequency and the chiller outlet water temperature based on the optimal water supply temperature.

8. The control method for coordinated operation of a chiller unit and a fully variable frequency pump tower as described in claim 7, characterized in that, Based on the optimal supply water temperature and the predicted load intensity, the specific heat capacity of water, the density of water, and the temperature difference are multiplied. The temperature difference is the temperature difference between the optimal supply water temperature and the return water temperature. The total water flow required by the system is equal to the ratio of the predicted load intensity to the product of the specific heat capacity of water, the density of water, and the temperature difference. The total water flow required by the system in the pipe network is recalculated. Based on the known water flow of the original system, the water flow required by the new system, and the original pump frequency, and according to the proportional relationship between pump flow and frequency, the new pump frequency is calculated by multiplying the original pump frequency by the ratio of the new total water flow required by the system to the original system water flow. The system sends the calculated new pump frequency command to the variable frequency pump, causing the variable frequency pump to operate at the new frequency, thereby adjusting the pump's water flow. The system sends the optimal supply water temperature to the chiller unit. The chiller unit obtains the target temperature to be adjusted and, based on the received optimal supply water temperature command, adjusts its chilled water outlet temperature to the target temperature to achieve optimized operation of the entire air conditioning system.

Images