Fabricated refrigerating room energy-saving monitoring system based on data analysis

Abstract

The invention discloses an assembly type refrigerating machine room energy-saving monitoring system based on data analysis, and relates to the technical field of energy-saving control, and the system comprises an interval module which is used for dividing the space of a machine room into a plurality of monitoring intervals, and is bound with a structural unit; the excitation module is used for applying power disturbance to the structural unit to obtain structural excitation data; the identification module is used for generating a heat influence coefficient matrix according to the excitation data and the temperature change of the adjacent monitoring interval; and the calculation module is used for calculating a blind area temperature and outputting an estimated temperature and a covariance based on the heat influence coefficient matrix, the structure energy consumption data and the adjacent interval temperature. According to the invention, the space of the machine room is subjected to three-dimensional grid division through the interval module, and each monitoring interval is bound with the function nodes in the structure of the machine room, so that the temperature monitoring grid with the corresponding relation between the space and the structure is formed, the space expression precision of monitoring data is improved, and the monitoring efficiency is improved. And a structural support is provided for positioning and temperature compensation of a blind area position.

Description

Technical Field

[0001] The present invention relates to the field of energy-saving control technology, and in particular to an energy-saving monitoring system for an assembled refrigeration room based on data analysis. Background Art

[0002] In modern building energy management, refrigeration rooms, as energy-intensive areas, require precise temperature monitoring and dynamic regulation based on heat load, which are key to achieving efficient energy conservation. While prefabricated refrigeration rooms offer advantages such as rapid construction and standard modular deployment due to their pre-configured structures and compact spatial layout, they also introduce new monitoring challenges.

[0003] In prefabricated wiring structures, some pipe areas, heat source nodes, or terminal return points are often difficult to directly deploy temperature sensors due to structural obstruction, concentrated wiring, or physical limitations. This creates multiple monitoring blind spots. These blind spots can easily prevent the system from sensing local heat distribution, affecting the accuracy of overall energy consumption regulation.

[0004] After searching, a Chinese patent (publication number: CN116007122A) discloses an efficient refrigeration room energy-saving monitoring system based on data analysis. The patent includes an energy-saving monitoring platform, which is communicatively connected to an energy consumption monitoring module, a production capacity optimization module, an energy consumption optimization module, a parameter adjustment module and a storage module. The energy consumption monitoring module is used to monitor and analyze the energy consumption of the refrigeration room, and the production capacity optimization module is used to perform production capacity optimization analysis on the refrigeration room and obtain the production capacity characteristic parameters of the spatial interval, and send the production capacity characteristic parameters of the spatial interval to the energy-saving monitoring platform.

[0005] Existing solutions for blind-spot temperature estimation primarily rely on linear compensation of adjacent areas or empirical models. These methods are poorly able to integrate the spatial layout, energy consumption, and heat diffusion trends of the assembly structure itself, and struggle to identify external environmental interference or implement dynamic adjustments. Therefore, this paper proposes an energy-saving monitoring system for prefabricated refrigeration rooms based on data analysis. Summary of the Invention

[0006] The purpose of the present invention is to provide an energy-saving monitoring system for an assembled refrigeration room based on data analysis to solve the problems mentioned in the above background technology.

[0007] The present invention can be implemented through the following technical solutions: an energy-saving monitoring system for assembled refrigeration rooms based on data analysis, comprising: an interval module, an excitation module, an identification module, a processing module and an energy-saving control module;

[0008] The interval module is used to divide the computer room space into multiple monitoring intervals and bind them with the structural units;

[0009] The excitation module is used to apply power disturbance to the structural unit to obtain structural excitation data;

[0010] The identification module is used to generate a thermal influence coefficient matrix based on the excitation data and the temperature changes in adjacent monitoring intervals;

[0011] The calculation module is used to calculate the blind zone temperature based on the thermal influence coefficient matrix, structural energy consumption data and adjacent interval temperatures, and output the estimated temperature and covariance;

[0012] The processing module is used to determine the confidence level of the estimated temperature and trigger the update of the thermal impact coefficient matrix when the preset conditions are not met;

[0013] The energy-saving control module is used to adjust the operating parameters of the refrigeration system according to the spatial temperature field and load prediction results to achieve the energy-saving operation goal.

[0014] A further technical improvement of the present invention is that when the blind spot is adjacent to the standard area and is disturbed by the external environment, the calculation module determines whether to correct the estimated temperature of the current blind spot based on the temperature change sequence of the standard area, the external environment change trend and the duration of the interference.

[0015] A further technical improvement of the present invention is that the determination method of the calculation module includes the following steps:

[0016] A1. Edge area identification: Based on the monitoring interval and structure space binding table output by the interval module, the monitoring area that is in direct contact with the external structure boundary is identified as a candidate interference sensitive area;

[0017] A2. Temperature sequence analysis: For the standard area adjacent to the blind zone, extract the temperature change sequence within the current estimation period and construct a temperature rise trend model;

[0018] A3. External environment linkage judgment: Synchronously obtain environmental parameters for the corresponding time period (such as external radiation intensity, external air temperature, etc.) and associate them with temperature changes in the standard area;

[0019] A4. If the temperature rise rate per unit time in the standard area is significantly higher than the historical steady-state average, and there is no corresponding structural power output fluctuation, then mark the standard area as a "potential external interference area" and record the duration of the interference.

[0020] A further technical improvement of the present invention is that after receiving the estimated blind zone temperature and covariance, the processing module enters the following operation:

[0021] B1. Associated blind spot identification:

[0022] b11. Obtain the standard areas marked as "potentially affected by external disturbances";

[0023] b12. Determine whether the adjacent blind spot is the estimated blind spot j. If so, mark the corresponding blind spot as a candidate for correction.

[0024] B2. Modify the condition judgment:

[0025] If the following two conditions are met:

[0026] The temperature of the standard area marked as "potentially affected by external disturbance area" continues to rise for a time greater than the preset temperature rise time threshold;

[0027] Covariance of the estimated temperature in the current blind area ∑ j,j (t) is less than the upper confidence limit;

[0028] Then enter the estimated temperature correction process;

[0029] B3. Calculation of temperature rise in standard area:

[0030] By formula: ΔT std =T std,now -T std,base , calculate the temperature change ΔT in the standard area std ;

[0031] Where, T std,now is the real-time temperature value of the standard area at the current moment, collected by the temperature sensor; T std,base It is the reference steady-state temperature value of the standard area under a stable external environment and undisturbed conditions, usually the historical statistical average of the area or the baseline temperature measured under low-disturbance conditions;

[0032] B4. Blind zone estimated temperature correction:

[0033] Estimated temperature Perform offset adjustment to obtain the corrected estimated temperature

[0034] α∈(0,1) is the disturbance correction coefficient, which can be set according to the duration of the disturbance, for example:

[0035] If the disturbance lasts for more than 15 minutes, set α = 0.8;

[0036] The duration of the disturbance is 5-15 minutes, and α is set to 0.5;

[0037] If the disturbance lasts less than 5 minutes, no correction will be triggered;

[0038] B5. Output and recovery mechanism:

[0039] b51, the corrected estimated temperature Replace the original estimated temperature Input value energy saving control module;

[0040] b52. If the interference state is subsequently eliminated, the original estimated temperature will be restored. And update the blind spot status mark.

[0041] A further technical improvement of the present invention is that when a closed-loop structure is formed by multiple standard areas with temperature rise exceeding the threshold outside the monitoring blind area, the processing module calculates the average temperature rise of the closed-loop area and corrects the current estimated temperature of the blind area based on the temperature rise to improve the accuracy of the temperature estimation. The specific correction method includes the following steps:

[0042] Z1, closed-loop boundary identification;

[0043] The processing module traverses the surrounding area of ​​each blind spot j based on the spatial topological structure and adjacency matrix of the monitoring interval;

[0044] If the blind area j is surrounded by multiple adjacent standard areas, and the temperature rise of the standard areas surrounding the blind area j is greater than a preset temperature rise threshold, the processing module determines that there is a closed high-temperature boundary structure outside the blind area j;

[0045] Z2, statistics of temperature rise in boundary area;

[0046] The processing module extracts the temperature rise of all standard areas that form a closed loop and calculates their average temperature rise

[0047] Z3, blind zone temperature correction;

[0048] Use the formula: Estimate the temperature of the current blind area Perform offset to obtain the estimated temperature after current blind spot correction

[0049] Where β∈(0.5,1.0) is the closed-loop isothermal correction coefficient, which is dynamically adjusted according to the boundary strength and temperature rise consistency;

[0050] Z4, output and status mark;

[0051] The corrected estimated temperature Replace the original estimated temperature Input the energy-saving control module, mark the current blind spot j as "closed-loop correction state", and record the correction value and closed-loop identification source for subsequent trend analysis and model adjustment.

[0052] A further technical improvement of the present invention is that when the processing module does not form a closed loop outside the monitoring blind area, if it is identified that there are multiple arc-shaped distribution structures with similar temperatures in the standard area, the estimated temperature of the blind area is corrected according to the temperature difference between the distribution structures.

[0053] A further technical improvement of the present invention is that the method for correcting the estimated temperature by the processing module comprises:

[0054] Y1. The processing module first determines whether there are multiple standard monitoring areas distributed in different directions outside the blind area, and analyzes the temperature changes in these areas respectively;

[0055] When two or more consecutive standard areas are identified in at least two spatial directions, and their temperatures show an upward trend exceeding the set temperature rise threshold within the current estimation cycle, and these standard areas cannot be spatially enclosed by continuous standard areas, the blind area is considered to be in a non-closed-loop structure condition;

[0056] Y2, the processing module further analyzes the spatial distribution of these temperature-rising areas to determine whether they conform to the thermal diffusion trend;

[0057] Y3. The processing module selects standard monitoring areas with similar temperatures in the candidate arc areas to form an arc distribution structure set. The judgment basis includes:

[0058] y31. The difference in temperature rise in multiple standard areas is less than the preset difference threshold;

[0059] y32. Spatial arrangement forms visible arcs or gradient curves;

[0060] y33. The distance between regions and the temperature change trend meet the expected direction of heat diffusion;

[0061] Y4. The processing module determines the dominant direction of temperature change based on the positional relationship of the blind area relative to each arc structure and the temperature difference along the arc path of the identified multiple arc structures;

[0062] On this basis, according to the relative geometric position of the blind spot in the arc structure, the average or weighted average of the temperature differences in multiple arc directions is used to make a directional correction to the current estimated temperature.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] The present invention divides the computer room space into three-dimensional grids along the x, y, and z axes using "interval modules" and binds each monitoring interval to a functional node within the computer room structure, thereby forming a temperature monitoring grid with a spatial and structural correspondence. This not only improves the spatial expression accuracy of monitoring data, but also provides structural support for the positioning of blind spots and temperature compensation, resolving the ambiguity problem of traditional methods in spatial thermal field modeling.

[0065] Furthermore, by integrating the thermal influence coefficient matrix output by the "identification module," the blind spot estimation model constructed by the "calculation module," and the confidence assessment and correction strategy of the "processing module," the present invention can dynamically estimate temperature data in sensorless areas and perform multiple rounds of corrections based on estimation errors, external environmental interference, or thermal diffusion structural characteristics. In particular, when the blind spot is in an unclosed loop or is subject to multi-directional thermal interference, the estimated amplitude can be automatically constructed based on the arc temperature distribution trend and offset correction can be performed, significantly improving the stability and reliability of the estimation.

[0066] On the other hand, the present invention combines the estimated blind-spot temperature with the measured temperature in the standard area to generate a full-space temperature field. This data is then used as input to drive the "energy-saving control module," which dynamically adjusts parameters such as the chiller's outlet water temperature setting, the main pump / terminal pump frequency, and the water valve and air valve openings. If the blind-spot estimate is unreliable, the system automatically limits the adjustment range or switches to a safe operating mode, ensuring safe system operation while minimizing energy consumption. This closed-loop control mechanism significantly improves the accuracy and responsiveness of energy-saving strategies and is applicable to a wide range of prefabricated building energy systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0068] Figure 1 It is a schematic diagram of the external structure of the present invention. DETAILED DESCRIPTION

[0069] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0070] Example 1

[0071] See also Figure 1 As shown, the present invention provides an energy-saving monitoring system for assembled refrigeration rooms based on data analysis, including: an interval module, an excitation module, an identification module, a processing module and an energy-saving control module;

[0072] The interval module is used to build the basic framework for spatial monitoring of the refrigeration room. It determines the location of each monitoring interval in three-dimensional space and binds it to a specific room structural unit. This supports the association between structural data and spatial temperature data in subsequent load excitation, thermal impact identification, and virtual estimation modules.

[0073] To work, enter:

[0074] BIM model or laser scanning data provides coordinate information of the computer room structure units;

[0075] Identification and location of structural units in the computer room (such as pipelines, heat source equipment, and terminal return water inlets);

[0076] deal with:

[0077] 1. Divide the computer room space into multiple three-dimensional monitoring zones along the x, y, and z directions. These zones include standard areas with sensor acquisition capabilities and blind areas where sensors cannot be deployed due to structural limitations. Temperature sensors are installed in the standard areas to collect basic temperature data in real time. This collected basic temperature data serves as the observational data input for blind area temperature estimation, enabling reconstruction and dynamic updating of the blind area temperature field.

[0078] 2. Based on the distance relationship between the structural coordinates and the center of gravity of the monitoring interval, complete the spatial binding of the monitoring interval and the structural unit;

[0079] 3. Output the mapping relationship table between the interval number and the computer room structure unit number;

[0080] Output:

[0081] 1. Monitoring interval spatial grid S(x,y,z);

[0082] 2. Monitoring interval-structural unit binding table;

[0083] The excitation module is used to actively apply power disturbances to induce changes in the thermal behavior of structural units. This in turn provides a data basis for modeling thermal influence relationships through the temperature responses of adjacent areas, acting as a "data-driven identification stimulator."

[0084] To work, enter:

[0085] 1. Monitoring interval-structural unit binding table, clearly defining the monitoring interval to which the target structural unit belongs;

[0086] 2. Incentive task instructions, including target structural unit ID, incentive power increase, duration, etc.;

[0087] deal with:

[0088] 1. Apply a step load change (±10-30% power) to the target structural unit (such as pumps, chillers, heat sources), and the control time is generally 2-5 minutes;

[0089] 2. Record the structural unit number, excitation amplitude, start time, end time and other information of the excitation process to form an event log;

[0090] Output:

[0091] 1. Stimulus event log (including structural unit ID, Δ power, and timestamp);

[0092] 2. The identification module is used to calculate the heat transfer effect of the structure on the spatial temperature based on the energy input of the excitation source and the temperature changes in the adjacent monitoring intervals. It outputs the thermal influence coefficient matrix and provides quantifiable modeling parameters for blind zone temperature estimation.

[0093] To work, enter:

[0094] 1. Incentive event log;

[0095] 2. Real-time temperature sequence of adjacent monitoring intervals before and after the excitation process;

[0096] 3. Three-dimensional space coordinate matrix;

[0097] deal with:

[0098] 1. Correlate and pair the structural excitation power change ΔP with the adjacent area temperature change ΔT;

[0099] 2. Use the recursive least squares algorithm (RLS) to build an input-output model;

[0100] 3. Identify the thermal influence weights of each structure on multiple intervals and generate the coefficient matrix K;

[0101] Output:

[0102] 1. The thermal influence coefficient matrix K, and Indicates that there are m structural units (such as heat source equipment, pipe sections, etc.) in the corresponding computer room, and n spatial monitoring intervals; and K ij It represents the unit thermal impact capability of equipment i in the computer room on monitoring interval j;

[0103] 2. Confidence index CI of thermal influence coefficient matrix K K*j , such as covariance or confidence intervals;

[0104] The calculation module is used to solve the problem of temperature monitoring in blind areas where sensors cannot be deployed. It combines the structural energy consumption with the temperature of adjacent areas and uses the heat impact matrix to perform real-time temperature estimation of blind areas, which serves as one of the basic input data for energy-saving monitoring.

[0105] To work, enter:

[0106] 1. Thermal influence coefficient matrix K;

[0107] 2. Real-time energy consumption data of structural units (such as power, current, and operating time);

[0108] 3. The temperature of the monitoring area adjacent to the target blind area;

[0109] deal with:

[0110] 1. Construct a state estimation model based on the thermal impact model;

[0111] 2. Use the extended Kalman filter (EKF) to dynamically estimate the blind zone temperature;

[0112] 3. Output estimated temperature and its covariance, which are used to reflect the forecast uncertainty;

[0113] Output:

[0114] 1. Estimate temperature and covariance matrix ∑ j (t);

[0115] The processing module is used to estimate the temperature output by the virtual temperature estimation processor Conduct credibility assessment and, based on the credibility results, decide whether to use the temperature value for energy-saving control strategy generation or initiate auxiliary correction and model modification processes, thereby forming a closed-loop adaptive adjustment capability for the system.

[0116] The processing module is also responsible for generating and issuing energy-saving control instructions, ensuring that the overall operation of the refrigeration room can be optimized even if there are no sensors in some blind areas;

[0117] To work, enter:

[0118] 1. Estimate temperature

[0119] 2. Covariance matrix ∑ j (t), reflecting the uncertainty in the estimated temperature;

[0120] 3. Confidence index CI of thermal influence coefficient matrix K K*j ;

[0121] deal with:

[0122] 1. Confidence calculation: to estimate temperature Error covariance matrix ∑ j The diagonal element ∑ corresponding to the jth monitoring interval in (t) j,j (t) as the basis, calculate its standard deviation

[0123] Make judgments based on the following conditions:

[0124] σ j (t)≤σ max , and CI K*j ≤CI max ; Among them, σ max and CI max are the maximum acceptable uncertainty standards preset for the system;

[0125] If both conditions are met, the confidence level is defined as "credible", otherwise it is "uncredible";

[0126] 2. Confidence response logic:

[0127] When the confidence level is "trusted":

[0128] The estimated temperature As a valid input parameter, it is passed to the energy-saving optimization algorithm module (such as the chiller energy efficiency control and fan-pump coordination strategy module), and the downstream model maintains equal weighting of the estimated temperature and the measured temperature;

[0129] When the confidence level is "untrustworthy":

[0130] The estimated temperature The participation weight of is set to λ(0<λ<1), which means that its influence is reduced in the system;

[0131] And in this embodiment, the system scheduling layer generates a mobile thermal imager inspection task and controls the thermal image acquisition device to go to the corresponding monitoring interval;

[0132] Get thermal image temperature T IR,j (t) and the estimated temperature Perform error evaluation. If the error between the two is greater than the preset error threshold, the thermal image temperature T IR,j (t) replaces the current estimated temperature And trigger an update calculation of the thermal influence coefficient matrix K to correct the thermal influence coefficient matrix K.

[0133] Specifically, the steps of updating and calculating the thermal influence coefficient matrix K include:

[0134] Collect current structural energy consumption data, including the power, operating time and other parameters of the bound structural unit at the current time point;

[0135] Obtain the actual temperature value measured during inspection and use it as the corrected actual temperature;

[0136] Energy consumption data and measured temperature are combined into sample pairs to form the input and output data required for thermal impact modeling;

[0137] Call the recursive least squares method and perform an online iterative update on the thermal influence coefficient matrix based on the current sample;

[0138] Generate a modified thermal influence coefficient matrix to replace the original matrix and continue to participate in subsequent temperature estimation;

[0139] The steps of correcting the thermal influence coefficient matrix K include:

[0140] Determine whether the error between the estimated temperature in the blind area and the temperature actually measured during inspection exceeds the error threshold;

[0141] When the error exceeds the error threshold, it is considered that the original temperature model estimation is inaccurate and the model correction process begins;

[0142] The measured temperature is used to replace the current estimated value and the point is marked as a "calibration data point";

[0143] Based on the calibration data, the corresponding parameters in the thermal influence coefficient matrix are updated to achieve the correction of local parameters;

[0144] After completing the parameter correction, the blind zone temperature is re-estimated and the new result is passed to the energy-saving control system for use;

[0145] The energy-saving control module dynamically adjusts the following parameters based on the full-space temperature field (including the fusion of measured and estimated results), combined with the cooling load prediction model and the energy efficiency curves of each device:

[0146] The outlet water temperature setpoint of the chiller;

[0147] Frequency or flow rate of main pump and terminal water pump;

[0148] The opening degree of the air valve or water valve at the air conditioner terminal;

[0149] The control goal is to keep the temperature of each monitoring area within the comfortable range and minimize overall energy consumption. If the system detects that the blind area temperature estimate is not reliable enough, it will tighten the adjustment range or switch to safe mode operation;

[0150] The cooling load forecasting model refers to a time series forecasting model built based on historical operating data, indoor and outdoor temperature and humidity trends, and personnel activity patterns, and is used to predict changes in cooling demand in a short period of time.

[0151] The energy efficiency curve refers to the relationship function between the unit energy consumption and performance output of equipment such as chillers and water pumps under different load conditions, usually obtained through factory testing or on-site calibration;

[0152] The estimated fusion result refers to the virtual temperature data obtained by the blind zone temperature estimation algorithm, which is weightedly fused with the measured temperature to construct a more complete spatial temperature field;

[0153] The energy-saving control module's control objective is to minimize overall system energy consumption and improve operational efficiency while maintaining the temperature in each monitoring zone within the comfort control range (e.g., 22°C to 26°C). If the system detects insufficient confidence in the blind zone temperature estimate (e.g., the estimated standard deviation exceeds a preset threshold, or the thermal impact model deviates significantly), the control module automatically triggers a degradation strategy, specifically:

[0154] Tighten the adjustment range, that is, reduce the response range of the control variable to avoid over-reliance on the estimated value;

[0155] Or switch to safe mode, that is, use conservative control setting values ​​to ensure stable temperature control in key areas until the estimated results return to a confidence level.

[0156] When the blind area is adjacent to the standard area and is disturbed by the external environment, the calculation module determines whether to correct the estimated temperature of the current blind area based on the temperature change sequence of the standard area, the external environment change trend, and the duration of the interference. The judgment method includes the following steps:

[0157] A1. Edge area identification: Based on the monitoring interval and structure space binding table output by the interval module, the monitoring area that is in direct contact with the external structure boundary is identified as a candidate interference sensitive area;

[0158] A2. Temperature sequence analysis: For the standard area adjacent to the blind zone, extract the temperature change sequence within the current estimation period and construct a temperature rise trend model;

[0159] A3. External environment linkage judgment: Synchronously obtain environmental parameters for the corresponding time period (such as external radiation intensity, external air temperature, etc.) and associate them with temperature changes in the standard area;

[0160] A4. If the temperature rise rate per unit time in the standard area is significantly higher than the historical steady-state average, and there is no corresponding structural power output fluctuation, then mark the standard area as a "potential external interference area" and record the duration of the interference.

[0161] After receiving the estimated blind zone temperature and covariance, the processing module proceeds to the following operations:

[0162] B1. Associated blind spot identification:

[0163] b11. Obtain the standard areas marked as "potentially affected by external disturbances";

[0164] b12. Determine whether the adjacent blind spot is the estimated blind spot j. If so, mark the corresponding blind spot as a candidate for correction.

[0165] B2. Modify the condition judgment:

[0166] If the following two conditions are met:

[0167] The temperature of the standard area marked as "potentially affected by external disturbance area" continues to rise for a time greater than the preset temperature rise time threshold;

[0168] Covariance of the estimated temperature in the current blind area ∑ j,j (t) is less than the upper confidence limit;

[0169] Then enter the estimated temperature correction process;

[0170] B3. Calculation of temperature rise in standard area:

[0171] By formula: ΔT std =T std,now -T std,base , calculate the temperature change ΔT in the standard area std ;

[0172] Where, T std,now is the real-time temperature value of the standard area at the current moment, collected by the temperature sensor; T std,base It is the reference steady-state temperature value of the standard area under a stable external environment and undisturbed conditions, usually the historical statistical average of the area or the baseline temperature measured under low-disturbance conditions;

[0173] B4. Blind zone estimated temperature correction:

[0174] Estimated temperature Perform offset adjustment to obtain the corrected estimated temperature

[0175] α∈(0,1) is the disturbance correction coefficient, which can be set according to the duration of the disturbance, for example:

[0176] If the disturbance lasts for more than 15 minutes, set α = 0.8;

[0177] The duration of the disturbance is 5-15 minutes, and α is set to 0.5;

[0178] If the disturbance lasts less than 5 minutes, no correction will be triggered;

[0179] B5. Output and recovery mechanism:

[0180] b51, the corrected estimated temperature Replace the original estimated temperature Input value energy saving control module;

[0181] b52. If the interference state is subsequently eliminated, the original estimated temperature will be restored. And update the blind spot status mark;

[0182] In this embodiment, the j=5th blind area is adjacent to the standard area std=4;

[0183] The time period is from 10:00 to 11:00 in the morning, which is the period of high solar radiation;

[0184] Structural unit status: The power of the cold water pipe section corresponding to blind zone 5 has no significant fluctuation during this period;

[0185] External environment changes: external radiation intensity from 400W / m 2 Increased to 800W / m 2 The outside air temperature rose by 2.1°C; the sunshine duration exceeded 12 minutes;

[0186] That is, the current temperature T in the standard area std,now =29.6℃; historical steady-state reference temperature (under no external disturbance conditions) T std,base =27.1℃;

[0187] Get the standard area temperature rise: ΔT std =T std,now -T std,base =2.5℃;

[0188] The disturbance duration is 12 minutes, which is between 5 and 15 minutes, so the disturbance correction coefficient α is set to 0.5 by looking up the table;

[0189] The current estimated temperature of the 5th blind zone is:

[0190] According to the modified formula

[0191] Finally, the corrected temperature Used for energy-saving control logic during this period;

[0192] Original estimated temperature No longer used as a control input;

[0193] The system marks the blind spot as "under correction" and will restore to the original model estimated path after the interference is removed.

[0194] Example 2

[0195] The energy-saving monitoring system for assembled refrigeration rooms based on data analysis is characterized by including:

[0196] The interval module is used to divide the computer room space into multiple monitoring intervals and bind them with the structural units;

[0197] An excitation module is used to apply power disturbance to the structural unit and obtain structural excitation data;

[0198] Identification module, used to generate a thermal influence coefficient matrix based on the excitation data and the temperature changes in adjacent monitoring intervals;

[0199] A calculation module is used to calculate the blind zone temperature based on the thermal influence coefficient matrix, structural energy consumption data and adjacent interval temperatures, and output the estimated temperature and covariance;

[0200] a processing module, configured to determine the confidence level of the estimated temperature and trigger an update of the thermal impact coefficient matrix when a preset condition is not met;

[0201] The energy-saving control module is used to adjust the operating parameters of the refrigeration system according to the space temperature field and load prediction results to achieve energy-saving operation goals.

[0202] When a closed-loop structure is formed by multiple standard areas with temperature rise exceeding the threshold outside the monitoring blind area, the processing module calculates the average temperature rise of the closed-loop area and corrects the current estimated temperature of the blind area based on the temperature rise to improve the accuracy of the temperature estimation. The specific correction method includes the following steps:

[0203] Z1, closed-loop boundary identification;

[0204] The processing module traverses the surrounding area of ​​each blind spot j based on the spatial topological structure and adjacency matrix of the monitoring interval;

[0205] If the blind area j is surrounded by multiple adjacent standard areas, and the temperature rise of the standard areas surrounding the blind area j is greater than a preset temperature rise threshold, the processing module determines that there is a closed high-temperature boundary structure outside the blind area j;

[0206] Z2, statistics of temperature rise in boundary area;

[0207] The processing module extracts the temperature rise of all standard areas that form a closed loop and calculates their average temperature rise

[0208] Z3, blind zone temperature correction;

[0209] Use the formula: Estimate the temperature of the current blind area Perform offset to obtain the estimated temperature after current blind spot correction

[0210] Where β∈(0.5,1.0) is the closed-loop isothermal correction coefficient, which is dynamically adjusted according to the boundary strength and temperature rise consistency;

[0211] Z4, output and status mark;

[0212] The corrected estimated temperature Replace the original estimated temperature Input the energy-saving control module and mark the current blind spot j as "closed-loop correction state". The correction value and closed-loop identification source are recorded for subsequent trend analysis and model adjustment.

[0213] In this embodiment, the default initial value of β is set to 0.7. When the temperature rise of all boundary standard areas is highly consistent, that is, the standard deviation is less than 0.2°C, β is adjusted to 0.9;

[0214] If the subsequent boundary closed-loop structure disintegrates (for example, the temperature in a certain standard area drops below the threshold), the correction state is removed and the original estimated temperature is restored.

[0215] Example 3

[0216] The energy-saving monitoring system for assembled refrigeration rooms based on data analysis is characterized by including:

[0217] The interval module is used to divide the computer room space into multiple monitoring intervals and bind them with the structural units;

[0218] An excitation module is used to apply power disturbance to the structural unit and obtain structural excitation data;

[0219] Identification module, used to generate a thermal influence coefficient matrix based on the excitation data and the temperature changes in adjacent monitoring intervals;

[0220] A calculation module is used to calculate the blind zone temperature based on the thermal influence coefficient matrix, structural energy consumption data and adjacent interval temperatures, and output the estimated temperature and covariance;

[0221] a processing module, configured to determine the confidence level of the estimated temperature and trigger an update of the thermal impact coefficient matrix when a preset condition is not met;

[0222] The energy-saving control module is used to adjust the operating parameters of the refrigeration system according to the space temperature field and load prediction results to achieve energy-saving operation goals.

[0223] Compared with Example 2, in Example 3, when the processing module does not form a closed loop outside the monitoring blind area, if multiple arc-shaped distribution structures with similar temperatures are identified in the standard area, the estimated temperature of the blind area is corrected according to the temperature difference between the distribution structures.

[0224] The method for processing the module to correct the estimated temperature includes:

[0225] Y1. Non-closed loop structure determination;

[0226] The processing module first determines whether there are sufficient standard areas outside the current blind spot where the temperature rise is significant but does not form a closed enclosing structure. If there are standard areas with temperature rise in at least two directions (for example, only on the east and south sides) and there are discontinuous boundaries in space, the current structure is considered to be a non-closed loop condition.

[0227] Y2, thermal diffusion structure identification;

[0228] The processing module further analyzes the spatial distribution of these temperature-rising areas to determine whether they conform to typical thermal diffusion trends. If these standard areas are arranged in a certain arc, ellipse, or fan shape in a plane or three-dimensional space, and their temperature values ​​gradually decrease in the spatial direction, it indicates that the structure may have an open thermal diffusion boundary.

[0229] Y3, screening of similar temperature areas;

[0230] The processing module selects standard monitoring areas with similar temperatures from the candidate arc areas to form an arc distribution structure set. The judgment basis includes:

[0231] y31. The difference in temperature rise in multiple standard areas is less than the preset difference threshold;

[0232] y32. Spatial arrangement forms visible arcs or gradient curves;

[0233] y33. The trend of distance and temperature change between regions meets the expected direction of heat diffusion (temperature gradually decreases with distance);

[0234] Y4, blind zone estimated temperature correction;

[0235] The processing module combines the positional relationship of the blind area with respect to each arc structure, and the processing module corrects the current estimated temperature of the blind area according to the identified temperature differences of the multiple arcs;

[0236] The correction direction is based on the arc temperature center trend, and the correction amplitude is determined by the accumulation or weighted average of the temperature difference distribution trend in each direction. The corrected temperature is used as the new estimated temperature to participate in the energy-saving control module;

[0237] In this embodiment, the specific correction method includes:

[0238] The processing module first determines the three-dimensional coordinates of the current blind spot within the overall equipment room monitoring grid and marks the arc-shaped temperature distribution structures adjacent to the blind spot. Within each arc-shaped structure, the "temperature center position" is defined as the standard area with the highest temperature within the structure.

[0239] The processing module calculates the spatial distance between the center of the blind area and the temperature center of each arc structure, and determines whether the blind area is located inside, at the edge, or outside the heat diffusion direction, thereby determining the directionality of the temperature correction.

[0240] For each arc structure identified as valid, extract its representative temperature rise value, that is, the temperature difference between the temperature center of the structure and the peripheral low-temperature boundary standard area, and record it as the local temperature difference of the structure;

[0241] If multiple arc structures are identified (e.g., the south, southwest, and southeast sides all exist), the temperature difference and relative spatial angle in each direction are recorded separately;

[0242] The processing module constructs a weighted distribution model of the temperature difference based on the spatial direction, angle, and distance between the blind spot and each arc structure. The temperature difference in each direction will be weighted based on its spatial correlation with the center of the blind spot:

[0243] Directions closer to the blind spot have higher weights;

[0244] The spatial angle has a higher weight on the structure facing the center of the blind spot;

[0245] Directions with more stable temperature differences have higher weights;

[0246] Finally, the weighted average value is calculated based on the product of these directional weights and the corresponding temperature difference as the correction amplitude;

[0247] Y5, status output and marking;

[0248] This round of correction is marked as "non-closed-loop thermal diffusion correction state", and the temperature difference characteristics used for estimation are retained in the data record for subsequent trend learning and adaptive updating of the estimation model.

[0249] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0250] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis is characterized by: include: The interval module is used to divide the computer room space into multiple monitoring intervals and bind them with the structural units; An excitation module is used to apply power disturbance to the structural unit and obtain structural excitation data; Identification module, used to generate a thermal influence coefficient matrix based on the excitation data and the temperature changes in adjacent monitoring intervals; A calculation module is used to calculate the blind zone temperature based on the thermal influence coefficient matrix, structural energy consumption data and adjacent interval temperatures, and output the estimated temperature and covariance; a processing module, configured to determine the confidence level of the estimated temperature and trigger an update of the thermal impact coefficient matrix when a preset condition is not met; The energy-saving control module is used to adjust the operating parameters of the refrigeration system according to the space temperature field and load prediction results to achieve energy-saving operation goals.

2. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 1 is characterized in that: When the blind area adjacent to the standard area is disturbed by the external environment, the calculation module determines whether to correct the estimated temperature of the current blind area based on the temperature change sequence of the standard area, the external environment change trend and the duration of the interference.

3. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 2 is characterized in that: The determination method of the calculation module comprises the following steps: A1. Identify monitoring areas that are in direct contact with external structural boundaries as candidate interference-sensitive areas; A2. For the standard area adjacent to the blind area, extract the temperature change sequence within the current estimation period and construct a temperature rise trend model; A3. Synchronously obtain environmental parameters for the corresponding time period and correlate them with temperature changes in the standard area; A4. If the temperature rise rate per unit time in the standard area is significantly higher than the historical steady-state average, and there is no corresponding structural power output fluctuation, then mark the standard area as a "potential external interference area" and record the duration of the interference.

4. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 3 is characterized in that: The processing module enters the following operations: B1. Obtain a standard area marked as a "potentially disturbed area" and determine whether its adjacent blind area is an estimated blind area j. If so, mark the corresponding blind area as a candidate for correction. B2. Modify the condition judgment: If the following two conditions are met: The temperature of the standard area marked as "potentially affected by external disturbance area" continues to rise for a time greater than the preset temperature rise time threshold; Covariance ∑ of the estimated temperature in the current blind zone j j,j (t) is less than the upper confidence limit; Then enter the estimated temperature correction process; B3. Calculation of temperature rise in standard area: By formula: ΔT std =T std,now -T std,base , calculate the temperature change ΔT in the standard area std ; Where, T std,now is the real-time temperature value of the standard area at the current moment; T std,base is the reference steady-state temperature value of the standard area; B4. Blind zone estimated temperature correction: Estimated temperature Perform offset adjustment to obtain the corrected estimated temperature α is the disturbance correction coefficient.

5. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 1 is characterized in that: When a closed-loop structure is formed by multiple standard areas with temperature rise exceeding a threshold value outside the monitoring blind area, the processing module calculates the average temperature rise of the closed-loop area and corrects the current estimated temperature of the blind area based on the temperature rise.

6. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 5 is characterized in that: The correction method comprises the following steps: Z1, the processing module traverses the surrounding area of ​​each blind spot j based on the spatial topological structure and adjacency matrix of the monitoring interval; If the blind area j is surrounded by multiple adjacent standard areas, and the temperature rise of the standard areas surrounding the blind area j is greater than a preset temperature rise threshold, the processing module determines that there is a closed high-temperature boundary structure outside the blind area j; Z2, the processing module extracts the temperature rise of all standard areas that form a closed loop and calculates their average temperature rise Z3. Use the formula: Estimate the temperature of the current blind area Perform offset to obtain the estimated temperature after current blind spot correction Where β∈(0.5,1.0) is the closed-loop isothermal correction coefficient.

7. The energy-saving monitoring system for assembled refrigeration rooms based on data analysis according to claim 1 is characterized in that: When the processing module detects that there are multiple arc-shaped distribution structures with similar temperatures in the standard area when no closed loop is formed outside the monitoring blind area, the estimated temperature of the blind area is corrected according to the temperature difference between the distribution structures.

8. The data analysis-based energy-saving monitoring system for assembled refrigeration rooms according to claim 7 is characterized in that: The method for the processing module to correct the estimated temperature includes: Y1. The processing module first determines whether there are multiple standard monitoring areas distributed in different directions outside the blind area, and analyzes the temperature changes in these areas respectively; When two or more consecutive standard areas are identified in at least two spatial directions, and their temperatures show an upward trend exceeding the set temperature rise threshold within the current estimation cycle, and these standard areas cannot be spatially enclosed by continuous standard areas, the blind area is considered to be in a non-closed-loop structure condition; Y2, the processing module further analyzes the spatial distribution of these temperature-rising areas to determine whether they conform to the thermal diffusion trend; Y3. The processing module selects standard monitoring areas with similar temperatures in the candidate arc areas to form an arc distribution structure set. The judgment basis includes: y31. The difference in temperature rise in multiple standard areas is less than the preset difference threshold; y32. Spatial arrangement forms visible arcs or gradient curves; y33. The distance between regions and the temperature change trend meet the expected direction of heat diffusion; Y4. The processing module determines the dominant direction of temperature change based on the positional relationship of the blind area relative to each arc structure and the temperature difference along the arc path of the identified multiple arc structures; On this basis, according to the relative geometric position of the blind spot in the arc structure, the average or weighted average of the temperature differences in multiple arc directions is used to make a directional correction to the current estimated temperature.

Images