Air conditioner management method and system with electricity consumption metering and grading limit control functions

Abstract

The invention discloses an air conditioner management method and system with electricity consumption metering and graded limit control, and relates to the technical field of energy consumption management and control. The method comprises the steps that air conditioner operation parameters are collected in a preset period, the total actual electricity consumption is calculated based on current and voltage, then an initial water supply temperature set value is determined by combining outdoor temperature, and a correction value is obtained through deviation calculation of indoor temperature and the initial set value and integrated into a final-state water supply temperature set value; and inputting the total actual power consumption, historical data and outdoor temperature into a preset model to obtain a total power consumption predicted value, adopting a corresponding power limit control strategy according to the interval, and finally adjusting the operation state of the air conditioner in combination with a final state water supply temperature set value and the limit control strategy. According to the method, the temperature control precision is guaranteed by dynamically correcting the water supply temperature, reasonable regulation and control of energy consumption are achieved by means of electricity consumption prediction and graded limit control, and the management efficiency of the air conditioner is improved while the energy-saving target is achieved.

Description

Technical Field

[0001] This invention belongs to the field of energy consumption management technology, specifically relating to an air conditioning management method and system with electricity metering and hierarchical limiting control. Background Technology

[0002] With the increasing demand for building energy conservation and the ever-growing requirements for indoor environmental comfort, the refined management of air conditioning systems has become a core element in reducing overall building energy consumption and ensuring user experience. However, current air conditioning management faces numerous complex challenges. For example, electricity metering relies heavily on overall loop statistics, making it difficult to accurately calculate energy consumption by combining real-time operating parameters. Water supply temperature settings lack dynamic correction mechanisms, failing to adapt to changes in indoor and outdoor temperatures. Furthermore, electricity consumption trend prediction is susceptible to interference from multiple factors, leading to insufficient accuracy. Additionally, prediction model parameter optimization is inefficient and prone to getting trapped in local optima. These issues not only result in significant energy waste and low operating efficiency in air conditioning systems but may also cause large fluctuations in indoor temperature and decreased user comfort. Therefore, constructing an air conditioning management method that integrates accurate electricity metering, dynamic parameter correction, reliable trend prediction, and tiered control, and leveraging advanced technologies to achieve real-time regulation of air conditioning operation, has become an urgent need to promote building energy conservation and improve the operational stability of air conditioning systems.

[0003] Patent CN117606114A discloses an indoor air conditioning management system and an air conditioning billing and energy-saving management method. The system includes: a management remote control for setting air conditioning usage parameters, temperature control parameters, time management parameters, and clock parameters, and sending these parameters to indoor controllers; and several indoor controllers. Each indoor controller synchronizes its parameters upon receiving them from the management remote control, measures air conditioning usage in real time, and controls the air conditioner to lock or operate at a low load when the remaining usage is measured to be zero. This system constructs a low-cost, network-free, and platform-free air conditioning billing and energy-saving management system. It not only solves air conditioning billing but also achieves three-pronged energy-saving management: temperature management, time management, and quantitative management. Furthermore, it can be installed by electricians and plumbers, and all parameter adjustments are condensed into a simple management remote control, requiring no professional personnel for debugging, resulting in low implementation and maintenance costs. However, this invention does not consider determining the initial water supply temperature setpoint based on the outdoor temperature and correcting it by combining the indoor temperature deviation to obtain the final water supply temperature setpoint. It also does not obtain the predicted electricity consumption value through multi-scale time series decomposition, temperature lag effect analysis and prediction model of optimized parameters. Therefore, it cannot implement graded limiting control based on the prediction, resulting in parameter mismatch with demand and energy consumption redundancy, making it difficult to achieve the energy saving target and the air conditioning management efficiency is low. Summary of the Invention

[0004] The purpose of this invention is to solve the problem of balancing energy-saving goals and air conditioning management efficiency, and to propose an air conditioning management method and system with electricity metering and hierarchical limiting control.

[0005] In a first aspect of this invention, an air conditioning management method with electricity metering and tiered limiting control is first proposed, the method comprising: Obtain the operating parameter values ​​of the air conditioner within a preset period; the operating parameter values ​​include current parameter values, voltage parameter values, indoor temperature value, outdoor temperature value, and temperature setpoint value; Within a preset period, the actual power consumption of the air conditioner is calculated based on the current parameter value and the voltage parameter value to obtain the total actual power consumption. The initial water supply temperature setting value is determined based on the outdoor temperature value. The deviation between the indoor temperature value and the initial water supply temperature setting value is calculated to obtain the water supply temperature correction value; The initial water supply temperature setpoint and the water supply temperature correction value are integrated to obtain the final water supply temperature setpoint; The total actual electricity consumption, historical total actual electricity consumption, and outdoor temperature value are input into a preset electricity consumption trend prediction model to obtain the total electricity consumption prediction value. Based on the electricity consumption range where the total electricity consumption forecast value is located, corresponding electricity consumption control strategies shall be adopted; The operating status of the air conditioner is adjusted according to the final water supply temperature setpoint and the power limiting strategy.

[0006] Optional, via The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This refers to the outdoor temperature.

[0007] Optional, via The deviation between the indoor temperature and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction coefficient.

[0008] Optionally, the total actual electricity consumption and the historical total actual electricity consumption can be decomposed into multi-scale time series to obtain electricity consumption components containing different frequency characteristics; A thermodynamic cumulative effect analysis was performed on the outdoor temperature values ​​to obtain a temperature hysteresis characteristic sequence; The coupled features are obtained by nonlinearly coupling the power consumption component and the temperature hysteresis feature sequence through a feature cross-network based on an attention gating mechanism. A preliminary total electricity consumption prediction sequence is obtained by performing time-series progressive prediction on the coupling characteristics using a preset electricity consumption trend prediction model. The total electricity consumption forecast value is obtained by synthesizing the preliminary total electricity consumption forecast value sequence through reverse decomposition and reconstruction.

[0009] Optionally, step 1: Obtain the parameter combination, use the parameter combination as the initial solution, and initialize the parameters of the algorithm; the parameters include population size and maximum number of iterations; Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameter combinations for a support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the set of initial solutions; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

[0010] In a second aspect of this invention, an air conditioning management system with electricity metering and tiered limiting control is proposed, comprising: The parameter acquisition module is used to acquire the operating parameter values ​​of the air conditioner within a preset period; the operating parameter values ​​include current parameter values, voltage parameter values, indoor temperature values, outdoor temperature values, and temperature setpoint values; The total actual power consumption calculation module is used to calculate the total actual power consumption of the air conditioner based on the current parameter value and the voltage parameter value within a preset period. An initial value generation module is used to determine the initial water supply temperature setpoint based on the outdoor temperature value. The correction value generation module is used to calculate the deviation between the indoor temperature value and the initial water supply temperature setting value to obtain the water supply temperature correction value. The final state value generation module is used to integrate the initial water supply temperature setpoint and the water supply temperature correction value to obtain the final state water supply temperature setpoint. The total electricity consumption forecast generation module is used to input the total actual electricity consumption, historical total actual electricity consumption and outdoor temperature value into a preset electricity consumption trend prediction model to obtain the total electricity consumption forecast value; The strategy generation module is used to adopt corresponding power consumption control strategies based on the power consumption range in which the total power consumption forecast value is located. The adjustment module is used to adjust the operating status of the air conditioner according to the final water supply temperature setpoint and the power limiting strategy.

[0011] Optionally, the initial value generation module includes: pass The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This refers to the outdoor temperature.

[0012] Optionally, the correction value generation module includes: pass The deviation between the indoor temperature and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction coefficient.

[0013] Optionally, the total electricity consumption forecast generation module includes: The decomposition module is used to perform multi-scale time-series decomposition on the total actual electricity consumption and the historical total actual electricity consumption to obtain electricity consumption components containing different frequency characteristics. The analysis module is used to perform thermodynamic cumulative effect analysis on the outdoor temperature value to obtain a temperature hysteresis characteristic sequence; The coupling feature generation module is used to obtain coupling features by nonlinearly coupling the power consumption component and the temperature hysteresis feature sequence through a feature cross-network based on an attention gating mechanism; The preliminary sequence generation module is used to obtain a preliminary total electricity consumption prediction value sequence by performing time-series progressive prediction on the coupling features through a preset electricity consumption trend prediction model. The total electricity consumption forecast generation module is used to synthesize the preliminary total electricity consumption forecast sequence through reverse decomposition and reconstruction to obtain the total electricity consumption forecast value.

[0014] Optionally, the preset electricity consumption trend prediction model is obtained by optimizing the parameters of the support vector machine model using an optimization algorithm. Specifically, the steps for optimizing the support vector machine parameters include: Step 1: Obtain the parameter combination and use it as the initial solution to initialize the algorithm's parameters; the parameters include population size and maximum number of iterations; Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameter combinations for a support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the set of initial solutions; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

[0015] The beneficial effects of this invention are as follows: This invention proposes an air conditioning management method with electricity metering and tiered limiting control. By collecting air conditioning operating parameters within a preset period, the total actual electricity consumption is first calculated based on current and voltage. Then, an initial water supply temperature setpoint is determined by combining the outdoor temperature. A correction value is calculated based on the deviation between the indoor temperature and the initial setpoint, and this is integrated into the final water supply temperature setpoint. Simultaneously, the total actual electricity consumption, historical data, and outdoor temperature are input into a preset model to obtain a total electricity consumption prediction value. Based on the predicted range, corresponding electricity limiting strategies are adopted. Finally, the air conditioning operating status is adjusted by combining the final water supply temperature setpoint and the limiting strategies. This method ensures temperature control accuracy through dynamic correction of the water supply temperature and achieves reasonable energy consumption control through electricity consumption prediction and tiered limiting control, thereby improving air conditioning management efficiency while achieving energy-saving goals. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 A flowchart illustrating an air conditioning management method with electricity metering and tiered limiting control, provided as an embodiment of the present invention; Figure 2 This is a framework diagram of an air conditioning management system with electricity metering and hierarchical limiting control provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention provides an air conditioning management method with electricity metering and tiered limiting. See also... Figure 1 , Figure 1 A flowchart illustrating an air conditioning management method with electricity metering and tiered limiting provided in an embodiment of the present invention. The method includes the following steps: S101, Obtain the operating parameter values ​​of the air conditioner within a preset period; S102, within a preset period, calculates the actual power consumption of the air conditioner based on the current parameter value and voltage parameter value to obtain the total actual power consumption; S103, determine the initial water supply temperature setting value based on the outdoor temperature value; S104, calculates the deviation between the indoor temperature value and the initial water supply temperature setting value to obtain the water supply temperature correction value; S105, integrate the initial water supply temperature setpoint and the water supply temperature correction value to obtain the final water supply temperature setpoint; S106, input the total actual electricity consumption, historical total actual electricity consumption and outdoor temperature value into the preset electricity consumption trend prediction model to obtain the total electricity consumption prediction value; S107, adopt corresponding power consumption control strategies based on the power consumption range where the total power consumption forecast value is located; S108, adjusts the operating status of the air conditioner according to the final water supply temperature setpoint and power limiting strategy; The operating parameters include current, voltage, indoor temperature, outdoor temperature, and temperature setting.

[0021] This invention provides an air conditioning management method with electricity metering and tiered limiting control. The method collects air conditioning operating parameters within a preset period, first calculating the total actual electricity consumption based on current and voltage, then determining the initial water supply temperature setpoint based on the outdoor temperature. A correction value is calculated based on the deviation between the indoor temperature and the initial setpoint, and integrated into the final water supply temperature setpoint. Simultaneously, the total actual electricity consumption, historical data, and outdoor temperature are input into a preset model to obtain a total predicted electricity consumption value. Based on the predicted range, corresponding electricity limiting strategies are implemented. Finally, the air conditioning operating status is adjusted by combining the final water supply temperature setpoint and the limiting strategies. This method ensures temperature control accuracy through dynamic correction of the water supply temperature and achieves reasonable energy consumption control through electricity prediction and tiered limiting control, improving air conditioning management efficiency while achieving energy-saving goals.

[0022] In one implementation, within a preset period, current and voltage parameters are collected in real time using current and voltage sensors integrated within the air conditioner. These sensors need to be calibrated to ensure data accuracy. Simultaneously, an indoor temperature sensor collects the real-time indoor temperature, and an outdoor temperature sensor collects the real-time outdoor temperature. Furthermore, user-set or system-default temperature settings are retrieved from the air conditioner control panel or control system backend as the target temperature reference. All collected parameter values ​​are synchronously stored with timestamps to form a continuous parameter sequence, providing complete data support for subsequent steps.

[0023] In one implementation, within the aforementioned preset period, based on the collected current and voltage parameter values, and according to the electrical formula for calculating power (power = current × electricity), the instantaneous power of the air conditioner at different times is first calculated. Then, the instantaneous power is integrated within the preset period or the average power over the time period is multiplied by the time to accumulate the total actual power consumption of the air conditioner within that period, thereby accurately reflecting the actual energy consumption of the air conditioner during that period.

[0024] In one implementation, after obtaining the initial water supply temperature setpoint and the water supply temperature correction value, a fusion calculation needs to be performed according to a preset integration rule. The integration rule must consider both the benchmark nature of the initial setpoint and the adjustability of the correction value. A weighted summation formula can be used: Final water supply temperature setpoint = Initial water supply temperature setpoint × a + Water supply temperature correction value × b, where a and b are weighting coefficients and a + b = 1, with a greater than b to ensure the dominant position of the initial setpoint. If the correction value is positive, the water supply temperature is increased based on the initial setpoint; if it is negative, it is decreased. Simultaneously, upper and lower limit constraints need to be set during the integration process to prevent the final setpoint from exceeding the equipment's operating limits, ultimately obtaining a final water supply temperature setpoint that considers both the outdoor temperature benchmark and indoor temperature feedback.

[0025] In one implementation, multiple power consumption zones are pre-defined based on the air conditioning system's power capacity, grid power supply limitations, and energy-saving targets. Zone 1 is low load: total predicted power consumption ≤ 50% of rated capacity; Zone 2 is normal load: 50% of rated capacity < total predicted power consumption ≤ 80% of rated capacity; Zone 3 is warning load: 80% of rated capacity < total predicted power consumption ≤ 90% of rated capacity; Zone 4 is overload: total predicted power consumption > 90% of rated capacity. Each zone corresponds to a specific power consumption control strategy: Zone 1 requires no control and maintains normal operation; Zone 2 optimizes operating parameters appropriately, such as by fine-tuning the fan speed; Zone 3 initiates active control by reducing the compressor's operating frequency; Zone 4 adopts mandatory control through phased shutdowns. After obtaining the total predicted power consumption, the zone to which it belongs is determined, and the corresponding strategy is invoked to ensure that power consumption is controlled within a reasonable range and to avoid exceeding limits.

[0026] In one implementation, the final-state water supply temperature setpoint is the core control objective, while the power consumption control strategy serves as a constraint, enabling coordinated adjustments to the air conditioning operation. First, based on the final-state water supply temperature setpoint, the target flow rate of the air conditioning water system and the target output of the compressor are determined. Then, the target parameters are corrected in conjunction with the power consumption control strategy: if the system is in a warning or over-limit range, the compressor operating frequency is reduced, the fan speed is lowered, or the number of operating water pumps is adjusted without significantly affecting the final-state water supply temperature; if the system is in a normal or low-load range, priority is given to ensuring the final-state water supply temperature meets the target, maintaining efficient equipment operation. During the adjustment process, real-time monitoring of indoor temperature and total power consumption is required for dynamic feedback and correction, ensuring that while meeting indoor ambient temperature requirements, strict power consumption control is implemented to achieve the dual objectives of temperature control and energy saving.

[0027] In one embodiment, via The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This refers to the outdoor temperature.

[0028] In one implementation, the process employs a linear algorithm that directly adjusts the control temperature based on the outdoor temperature, exhibiting high sensitivity and rapid response to changes in external heat load. However, this process is a feedforward control process, making it highly sensitive to uncertainties. When the indoor load changes abruptly, or the system encounters other unforeseen disturbances, its control performance may be affected. To compensate for these shortcomings, a feedback control mechanism based on indoor air temperature can be introduced. By correcting residual errors, the overall robustness and control accuracy of the control system can be further improved.

[0029] In one embodiment, via The deviation between the indoor temperature and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction coefficient.

[0030] In one implementation, and The value is set by technical personnel.

[0031] In one implementation, through By calculating the deviation between indoor temperature and the initial temperature setpoint to obtain the water supply temperature correction value, the impact of the difference between indoor temperature and the initial water supply temperature setpoint on the water supply temperature can be accurately and efficiently quantified, using an incremental correction coefficient. and offset correction coefficient It can flexibly adapt to the correlation between temperature deviation and water supply temperature correction in different scenarios, thereby quickly and accurately obtaining water supply temperature correction values ​​that meet actual needs. This provides a reliable basis for the precise control of subsequent water supply temperature, helps improve the accuracy and intelligence of water supply temperature control, and ensures that the indoor temperature is maintained within a comfortable and stable range.

[0032] In one embodiment, multi-scale time-series decomposition of total actual electricity consumption and historical total actual electricity consumption is performed to obtain electricity consumption components containing different frequency characteristics. Thermodynamic cumulative effect analysis of outdoor temperature values ​​yields temperature hysteresis characteristic sequences; Coupled features are obtained by nonlinearly coupling the electricity consumption component and temperature lag feature sequence through a feature cross-network based on an attention gating mechanism. A preliminary total electricity consumption forecast sequence is obtained by performing time-series progressive prediction of coupling characteristics using a preset electricity consumption trend prediction model. The total electricity consumption forecast value is obtained by synthesizing the preliminary total electricity consumption forecast value sequence through reverse decomposition and reconstruction.

[0033] In one implementation, when performing multi-scale time series decomposition on the total actual electricity consumption and the historical total actual electricity consumption, appropriate time series decomposition algorithms such as wavelet transform and empirical mode decomposition are used to expand the electricity consumption data on different time scales, decomposing the originally complex electricity consumption time series into multiple subsequences. Each subsequence represents the electricity consumption change characteristics in different frequency ranges. For example, some subsequences reflect long-term trend changes, while others reflect short-term fluctuations, thereby obtaining electricity consumption components containing different frequency characteristics.

[0034] In one implementation, the thermodynamic cumulative effect analysis of outdoor temperature values ​​is based on thermodynamic principles. It considers the lag in the impact of temperature changes on electricity consumption. By analyzing the temperature change process over time, the cumulative degree of temperature impact on subsequent electricity consumption at different times is calculated. This leads to the construction of a temperature lag characteristic sequence that reflects the law of temperature lag effect. This sequence can reflect the effect of temperature changes accumulated over a period of time on current and future electricity consumption.

[0035] In one implementation, the preliminary total electricity consumption forecast sequence is synthesized by reverse decomposition and reconstruction. This involves reversing the previous multi-scale time series decomposition process, recombining the components of different frequency characteristics in the preliminary total electricity consumption forecast sequence according to certain rules and methods, restoring and integrating the decomposed subsequences into a complete time series, and finally obtaining the total electricity consumption forecast that reflects the overall electricity consumption situation.

[0036] In one embodiment, step 1: obtain the parameter combination, use the parameter combination as the initial solution, and initialize the parameters of the algorithm; the parameters include the population size and the maximum number of iterations; Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameters of the support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the initial solution set; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

[0037] In one implementation, the parameter combination consists of a penalty factor and a kernel parameter. The penalty factor is used to control the model's tolerance to error; if it is too large, it will lead to overfitting, and if it is too small, it will lead to underfitting. The kernel parameter is used to determine the mapping capability of the kernel function; if it is too large, the mapping will be too strong, and if it is too small, the mapping will be insufficient.

[0038] In one implementation, the formula for calculating the dynamic switching probability p is as follows: ,in For the maximum switching probability, The minimum switching probability is given by M, the maximum number of iterations, and t, the current iteration number. In the early stages of iteration, when t is relatively small, p is close to... The algorithm is more likely to enter the global exploration, thus quickly covering the parameter space. In the later stages of iteration, when t approaches M, p approaches M. The algorithm is more likely to delve into local optimization, thus enabling a fine search near the optimal range; the formula for calculating the adaptive weight w. In the early stages of iteration: when w is relatively large, close to 1.5, the weight has a weak impact on position updates, thus preserving the randomness of global exploration; when w is relatively small, close to 0.5, the weight has a strong impact on position updates, guiding individuals to fine-tune towards the current optimal solution, thereby reducing invalid searches.

[0039] In one implementation, the algorithm's pheromone concentration The calculation formula is: Where c is the perception modality constant and a is the perception index, both of which need to be set according to the experience of technical personnel and are used to adjust the correlation strength between pheromone concentration and fitness.

[0040] In one implementation, when updating the position of the optimal initial solution, a random long step size perturbation is introduced to enhance population diversity, allowing the algorithm to potentially escape local optima in the later stages of iteration. Levy flight is a random walking model characterized by short-step dense movement and long-step random jumps. The formula for calculating its step size 's' is... ,in, The Levy exponent is usually taken as 1.5, and u and v follow a normal distribution. ,v~N(0,1)), Depend on Decide, During the position update and integration, Levy flight is combined with the current optimal position, and the optimal individual position is perturbed in the later stage of the iteration.

[0041] Based on the same inventive concept, this invention also provides an air conditioning management system with electricity metering and tiered limiting control. See also Figure 2 , Figure 2A framework diagram of an air conditioning management system with electricity metering and tiered limiting control provided in an embodiment of the present invention includes: The parameter acquisition module is used to acquire the operating parameter values ​​of the air conditioner within a preset period; the operating parameter values ​​include current parameter values, voltage parameter values, indoor temperature values, outdoor temperature values, and temperature setpoint values; The total actual power consumption calculation module is used to calculate the total actual power consumption of the air conditioner based on the current parameter value and voltage parameter value within a preset period. The initial value generation module is used to determine the initial water supply temperature setpoint based on the outdoor temperature value. The correction value generation module is used to calculate the deviation between the indoor temperature value and the initial water supply temperature setpoint to obtain the water supply temperature correction value. The final state value generation module is used to integrate the initial water supply temperature setpoint and the water supply temperature correction value to obtain the final state water supply temperature setpoint. The total electricity consumption forecast generation module is used to input the total actual electricity consumption, historical total actual electricity consumption and outdoor temperature value into the preset electricity consumption trend prediction model to obtain the total electricity consumption forecast value; The strategy generation module is used to adopt corresponding power consumption control strategies based on the power consumption range where the total power consumption forecast value is located. The adjustment module is used to adjust the operating status of the air conditioner according to the final water supply temperature setpoint and the power limiting strategy.

[0042] An air conditioning management system with electricity metering and tiered limiting control, provided by an embodiment of the present invention, collects air conditioning operating parameters within a preset period. First, it calculates the total actual electricity consumption based on current and voltage. Then, it determines the initial water supply temperature setpoint by combining the outdoor temperature. A correction value is calculated based on the deviation between the indoor temperature and the initial setpoint, and integrated into the final water supply temperature setpoint. Simultaneously, the total actual electricity consumption, historical data, and outdoor temperature are input into a preset model to obtain a total predicted electricity consumption value. Based on the predicted range, corresponding electricity limiting strategies are adopted. Finally, the air conditioning operating status is adjusted by combining the final water supply temperature setpoint and the limiting strategies. This method ensures temperature control accuracy through dynamic correction of the water supply temperature and achieves reasonable energy consumption control through electricity prediction and tiered limiting control, thereby improving air conditioning management efficiency while achieving energy-saving goals.

[0043] In one embodiment, the initial value generation module includes: pass The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This refers to the outdoor temperature.

[0044] In one embodiment, the correction value generation module includes: pass The deviation between the indoor temperature and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction coefficient.

[0045] In one embodiment, the total electricity consumption forecast generation module includes: The decomposition module is used to perform multi-scale time-series decomposition of the total actual electricity consumption and the historical total actual electricity consumption to obtain electricity consumption components containing different frequency characteristics. The analysis module is used to perform thermodynamic cumulative effect analysis on outdoor temperature values ​​to obtain temperature hysteresis characteristic sequences; The coupling feature generation module is used to obtain coupled features by nonlinearly coupling the power consumption component and temperature lag feature sequence through a feature cross-network based on an attention gating mechanism. The preliminary sequence generation module is used to obtain a preliminary total electricity consumption prediction value sequence by performing time-series progressive prediction of coupling features through a preset electricity consumption trend prediction model. The total electricity consumption forecast generation module is used to synthesize the total electricity consumption forecast value by reversing the decomposition and reconstruction of the preliminary total electricity consumption forecast value sequence.

[0046] In one embodiment, the preset electricity consumption trend prediction model is obtained by optimizing the parameters of the support vector machine model using an optimization algorithm. Specifically, the steps for optimizing the support vector machine parameters include: Step 1: Obtain the parameter combination and use it as the initial solution to initialize the algorithm's parameters; the parameters include population size and maximum number of iterations. Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameters of the support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the initial solution set; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

[0047] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An air conditioning management method with electricity metering and tiered limiting control, characterized in that, The method includes: Obtain the operating parameter values ​​of the air conditioner within a preset period; the operating parameter values ​​include current parameter values, voltage parameter values, indoor temperature value, outdoor temperature value, and temperature setpoint value; Within a preset period, the actual power consumption of the air conditioner is calculated based on the current parameter value and the voltage parameter value to obtain the total actual power consumption. The initial water supply temperature setting value is determined based on the outdoor temperature value. The deviation between the indoor temperature value and the initial water supply temperature setting value is calculated to obtain the water supply temperature correction value; The initial water supply temperature setpoint and the water supply temperature correction value are integrated to obtain the final water supply temperature setpoint; The total actual electricity consumption, historical total actual electricity consumption, and outdoor temperature value are input into a preset electricity consumption trend prediction model to obtain the total electricity consumption prediction value. Based on the electricity consumption range where the total electricity consumption forecast value is located, corresponding electricity consumption control strategies shall be adopted; The operating status of the air conditioner is adjusted according to the final water supply temperature setpoint and the power limiting strategy.

2. The air conditioning management method with electricity metering and tiered limiting control according to claim 1, characterized in that, Determining the initial water supply temperature setpoint based on the outdoor temperature includes: pass The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This represents the outdoor temperature value.

3. The air conditioning management method with electricity metering and tiered limiting control according to claim 1, characterized in that, The deviation calculation between the indoor temperature and the initial water supply temperature setpoint yields the water supply temperature correction value, including: pass The deviation between the indoor temperature and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction factor.

4. The air conditioning management method with electricity metering and tiered limiting control according to claim 1, characterized in that, The total actual electricity consumption, historical total actual electricity consumption, and outdoor temperature value are input into a preset electricity consumption trend prediction model to obtain the total electricity consumption prediction value, which includes: Multi-scale time-series decomposition of the total actual electricity consumption and the historical total actual electricity consumption yields electricity consumption components containing different frequency characteristics; A thermodynamic cumulative effect analysis was performed on the outdoor temperature values ​​to obtain a temperature hysteresis characteristic sequence; The coupled features are obtained by nonlinearly coupling the power consumption component and the temperature hysteresis feature sequence through a feature cross-network based on an attention gating mechanism. A preliminary total electricity consumption prediction sequence is obtained by performing time-series progressive prediction on the coupling characteristics using a preset electricity consumption trend prediction model. The total electricity consumption forecast value is obtained by synthesizing the preliminary total electricity consumption forecast value sequence through reverse decomposition and reconstruction.

5. An air conditioning management method with electricity metering and tiered limiting control according to claim 4, characterized in that, The preset electricity consumption trend prediction model is obtained by optimizing the parameters of the support vector machine model using an optimization algorithm. Specifically, the steps for optimizing the support vector machine parameters include: Step 1: Obtain the parameter combination and use it as the initial solution to initialize the algorithm's parameters; the parameters include population size and maximum number of iterations; Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameter combinations for a support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the set of initial solutions; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

6. An air conditioning management system with electricity metering and tiered limiting control, characterized in that, The system includes: The parameter acquisition module is used to acquire the operating parameter values ​​of the air conditioner within a preset period; the operating parameter values ​​include current parameter values, voltage parameter values, indoor temperature values, outdoor temperature values, and temperature setpoint values; The total actual power consumption calculation module is used to calculate the total actual power consumption of the air conditioner based on the current parameter value and the voltage parameter value within a preset period. An initial value generation module is used to determine the initial water supply temperature setpoint based on the outdoor temperature value. The correction value generation module is used to calculate the deviation between the indoor temperature value and the initial water supply temperature setting value to obtain the water supply temperature correction value. The final state value generation module is used to integrate the initial water supply temperature setpoint and the water supply temperature correction value to obtain the final state water supply temperature setpoint. The total electricity consumption forecast generation module is used to input the total actual electricity consumption, historical total actual electricity consumption and outdoor temperature value into a preset electricity consumption trend prediction model to obtain the total electricity consumption forecast value; The strategy generation module is used to adopt corresponding power consumption control strategies based on the power consumption range in which the total power consumption forecast value is located. The adjustment module is used to adjust the operating status of the air conditioner according to the final water supply temperature setpoint and the power limiting strategy.

7. An air conditioning management system with electricity metering and tiered limiting control as described in claim 6, characterized in that, The initial value generation module includes: pass The initial water supply temperature setpoint was calculated; where, This is the initial water supply temperature setpoint. This is the water supply temperature setting corresponding to the upper limit of the outdoor temperature. This is the water supply temperature setpoint corresponding to the lower limit of the outdoor temperature. The upper limit of outdoor temperature, The lower limit of outdoor temperature, This refers to the outdoor temperature.

8. An air conditioning management system with electricity metering and tiered limiting control as described in claim 6, characterized in that, The correction value generation module includes: pass The deviation between the indoor temperature value and the initial temperature setpoint is calculated to obtain the water supply temperature correction value; where T2 is the water supply temperature correction value and T1 is the initial water supply temperature setpoint. Indoor temperature, Incremental correction coefficient, This is the offset correction factor.

9. An air conditioning management system with electricity metering and tiered limiting control according to claim 6, characterized in that, The total electricity consumption forecast generation module includes: The decomposition module is used to perform multi-scale time-series decomposition on the total actual electricity consumption and the historical total actual electricity consumption to obtain electricity consumption components containing different frequency characteristics. The analysis module is used to perform thermodynamic cumulative effect analysis on the outdoor temperature value to obtain a temperature hysteresis characteristic sequence; The coupling feature generation module is used to obtain coupling features by nonlinearly coupling the power consumption component and the temperature hysteresis feature sequence through a feature cross-network based on an attention gating mechanism; The preliminary sequence generation module is used to obtain a preliminary total electricity consumption prediction value sequence by performing time-series progressive prediction on the coupling features through a preset electricity consumption trend prediction model. The total electricity consumption forecast generation module is used to synthesize the preliminary total electricity consumption forecast sequence through reverse decomposition and reconstruction to obtain the total electricity consumption forecast value.

10. An air conditioning management system with electricity metering and tiered limiting control according to claim 9, characterized in that, The preset electricity consumption trend prediction model is obtained by optimizing the parameters of the support vector machine model using an optimization algorithm. Specifically, the steps for optimizing the support vector machine parameters include: Step 1: Obtain the parameter combination and use it as the initial solution to initialize the algorithm's parameters; the parameters include population size and maximum number of iterations; Step 2: Randomly generate a number of initial solutions corresponding to the population size to obtain an initial solution set; each initial solution in the initial solution set corresponds to a set of parameter combinations for a support vector machine; Step 3: Calculate the fitness value of each initial solution in the initial solution set, and select the initial solution with the highest fitness as the initial optimal solution; Step 4: Calculate the dynamic switching probability and adaptive weights based on the number of iterations; Step 5: Generate random numbers for the target initial solution. If the random numbers are less than the dynamic switching probability, then proceed... Update the initial solution position; if the random number is greater than or equal to the dynamic switching probability, then... Update the initial solution position; the target initial solution is any initial solution in the set of initial solutions; where, Let be the updated position of the initial solution for the i-th objective in generation t. Let be the position of the i-th objective before the initial solution of the t-th generation is updated. Let be the position of the j-th initial solution randomly selected from the population in generation t (j and i are different individuals), and r be a random number. The position of the k-th initial solution randomly selected from the population in generation t (k is different from i and j) The pheromone concentration is the value of the algorithm, and w is the adaptive weight. Step 6: Through The position is updated based on the optimal initial solution; where, Let be the position after updating the optimal initial solution in generation t. This represents the position before updating the optimal initial solution in generation t. Levy index It is the element-wise product; Step 7: Determine if the maximum number of iterations has been reached. If not, return to step 3. If it has been reached, use the parameter combination corresponding to the optimal initial solution as the optimal parameters of the support vector machine.

Images