Energy management method, device and equipment based on air conditioning system and medium

Abstract

The invention relates to an energy management method, device and equipment based on an air conditioning system and a medium. The method comprises the steps that in a preset period, a target cold storage capacity calculation model matched with the current working condition is selected from a plurality of candidate cold storage capacity calculation models according to the pre-obtained photovoltaic power generation power, the air conditioner system load and the change trend of the environment temperature; the environment temperature and the electric energy are input into the target cold storage capacity calculation model, and the cold storage capacity, output by the target cold storage capacity calculation model, of a phase change cold storage material in the air conditioner system is obtained; based on the cold storage capacity, the photovoltaic power generation power and pre-obtained planned electricity consumption probability distribution, analysis processing is conducted, and an energy management strategy of the air conditioning system is obtained; the energy management strategy is used for guiding the air conditioning system to adopt commercial power supply or photovoltaic power supply, whether to utilize the phase change cold storage material for cold storage or not and whether to start a compressor to assist in executing the cold storage process or not. By adopting the method, the power consumption cost can be reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air conditioner control, in particular to an energy management method, device, equipment and medium based on an air conditioner system. BACKGROUND

[0002] With the increasing prominence of energy issues, how to effectively utilize renewable energy and reduce energy consumption has become a research hotspot. In air conditioning systems, traditional operation modes often rely on grid power supply, resulting in high energy costs and increased grid burden during peak electricity consumption periods. The development of photovoltaic technology provides a new energy solution for air conditioning systems, while phase change cold storage technology can store cold at low electricity prices and release it during peak electricity consumption periods, thereby reducing operating costs.

[0003] Current photovoltaic and phase change cold storage air conditioning systems cannot precisely regulate electricity consumption strategies, resulting in low energy utilization efficiency and thus failing to maximize cost reduction. Therefore, there is an urgent need for an air conditioning electricity control strategy that can reduce electricity costs to solve the above problems. SUMMARY

[0004] Therefore, it is necessary to provide an energy management method, device, equipment and medium based on an air conditioner system that can reduce electricity costs to solve the above problems.

[0005] In a first aspect, the present application provides an energy management method based on an air conditioner system, which comprises:

[0006] In a predetermined period, a target cold storage amount calculation model matching the current working condition is selected from a plurality of candidate cold storage amount calculation models according to the pre-acquired variation trends of photovoltaic power generation, air conditioning system load and environmental temperature;

[0007] The environmental temperature and the electrical energy are input into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model;

[0008] Based on the cold storage amount, the photovoltaic power generation and the pre-acquired planning electricity consumption probability distribution, an energy management strategy of the air conditioning system is obtained; the energy management strategy is used to guide the air conditioning system to adopt grid power supply or photovoltaic power supply, whether to utilize phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0009] In a second aspect, the present application further provides an energy management device based on an air conditioner system, which comprises:

[0010] The matching module is configured to select, in a preset period, a target cold storage amount calculation model matching a current working condition from a plurality of candidate cold storage amount calculation models according to a variation trend of the photovoltaic power, the air conditioning system load and the ambient temperature obtained in advance;

[0011] The output module is configured to input the ambient temperature and the electric energy into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model;

[0012] The analysis module is configured to perform analysis and processing based on the cold storage amount, the photovoltaic power and the planned power consumption probability distribution obtained in advance to obtain an energy management strategy of the air conditioning system; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to utilize the phase change cold storage material to perform cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0013] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0014] In a preset period, a target cold storage amount calculation model matching a current working condition is selected from a plurality of candidate cold storage amount calculation models according to a variation trend of the photovoltaic power, the air conditioning system load and the ambient temperature obtained in advance;

[0015] The ambient temperature and the electric energy are input into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model;

[0016] Analysis and processing are performed based on the cold storage amount, the photovoltaic power and the planned power consumption probability distribution obtained in advance to obtain an energy management strategy of the air conditioning system; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to utilize the phase change cold storage material to perform cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0017] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0018] In a preset period, a target cold storage amount calculation model matching a current working condition is selected from a plurality of candidate cold storage amount calculation models according to a variation trend of the photovoltaic power, the air conditioning system load and the ambient temperature obtained in advance;

[0019] The ambient temperature and the electric energy are input into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model;

[0020] The energy management strategy of the air conditioning system is obtained based on the cold storage capacity, the photovoltaic power generation power and the pre-acquired planning power consumption probability distribution; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use the phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0021] In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the following steps:

[0022] In a preset period, a target cold storage capacity calculation model matching a current working condition is selected from a plurality of candidate cold storage capacity calculation models according to pre-acquired variation trends of photovoltaic power generation power, air conditioning system load and environmental temperature;

[0023] The environmental temperature and the electric energy are input into the target cold storage capacity calculation model to obtain the cold storage capacity of the phase change cold storage material in the air conditioning system output by the target cold storage capacity calculation model;

[0024] The energy management strategy of the air conditioning system is obtained based on the cold storage capacity, the photovoltaic power generation power and the pre-acquired planning power consumption probability distribution; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use the phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0025] The above-mentioned energy management method, device, equipment and medium based on the air conditioning system select a target cold storage capacity calculation model matching a current working condition from a plurality of candidate cold storage capacity calculation models according to pre-acquired variation trends of photovoltaic power generation power, air conditioning system load and environmental temperature in a preset period; input the environmental temperature and the electric energy into the target cold storage capacity calculation model to obtain the cold storage capacity of the phase change cold storage material in the air conditioning system output by the target cold storage capacity calculation model; obtain the energy management strategy of the air conditioning system based on the cold storage capacity, the photovoltaic power generation power and the pre-acquired planning power consumption probability distribution; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use the phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process; the method obtains accurate cold storage capacity data by dynamically matching the target cold storage capacity calculation model, generates the energy management strategy in combination with the photovoltaic power generation power and the planning power consumption probability distribution, preferentially uses photovoltaic power supply, stores cold in a low-valley electricity price period, and reduces the high-cost operation of the compressor by releasing cold through the phase change cold storage material, which realizes the synergy of renewable energy consumption, low-price electricity storage and high-price electricity peak avoidance, thereby reducing the electricity cost of the air conditioning system. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 It is an internal structure diagram of a computer device in one embodiment;

[0027] Figure 2 Flowchart of the energy management method based on air conditioning system in one embodiment;

[0028] Figure 3 Flowchart of the energy management method based on air conditioning system in another embodiment;

[0029] Figure 4 Flowchart of the energy management method based on air conditioning system in another embodiment;

[0030] Figure 5 Flowchart of the energy management method based on air conditioning system in another embodiment;

[0031] Figure 6 Flowchart of the energy management method based on air conditioning system in another embodiment;

[0032] Figure 7 Flowchart of the energy management method based on air conditioning system in another embodiment;

[0033] Figure 8 Structure block diagram of the energy management device based on air conditioning system in one embodiment. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0035] In one embodiment, a computer device, which can be a server, has an internal structure diagram as shown in Figure 1 The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store relevant data in the process of energy management based on air conditioning system. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement an energy management method based on air conditioning system.

[0036] Those skilled in the art can understand that Figure 1 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0037] In one embodiment, as Figure 2 Indicated, an energy management method based on an air conditioning system is provided. The method is applied to a computer device as Figure 1 for example, and includes the following steps:

[0038] S201, in a preset period, according to the variation trend of the pre-acquired photovoltaic power generation, air conditioning system load and environmental temperature, select a target cold storage amount calculation model matching the current working condition from a plurality of candidate cold storage amount calculation models.

[0039] The photovoltaic power generation is the power generation capacity of the photovoltaic power generation system per unit time, which is used to evaluate the available amount of renewable energy.

[0040] The air conditioning system load is the refrigeration / heating amount required by the air conditioning system to meet the indoor thermal environment demand, which is affected by factors such as indoor temperature setting, outdoor environmental temperature, building envelope thermal resistance, and personnel activity.

[0041] The environmental temperature refers to the external environmental temperature when the air conditioning system is running. The environmental temperature will affect the cold storage efficiency of the phase change cold storage material, the photovoltaic power generation efficiency and the air conditioning system energy consumption.

[0042] The candidate cold storage amount calculation model is a set of mathematical models for calculating the cold storage amount of the phase change cold storage material, which contains physical relationships under various working conditions, including charging-cold storage power and cold storage model (cold storage process), cold release-cold model (cold release process), combined charging and discharging model (simultaneous charging and discharging), and non-charging and non-discharging environmental temperature-cold model (natural decay).

[0043] In the embodiment of the present application, in a pre-set time period, the computer device first acquires the trend of key data related to energy management. The computer device uses meteorological information and historical power generation data to analyze the variation of photovoltaic power generation in the future period. For example, on a sunny day with sufficient sunlight, the photovoltaic power generation usually reaches a peak from 10 o'clock in the morning to 3 o'clock in the afternoon; while in rainy weather, the photovoltaic power generation may be at a low level all day.

[0044] Meanwhile, the computer device combines the building internal environment data and external climate conditions to predict the load change of the air conditioning system, and obtains the air conditioning system load. For example, the air conditioning refrigeration load is high when people concentrate in the office during the daytime on weekdays, and the load is significantly reduced when there is no one at night. The change trend of the environmental temperature is monitored in real time through the outdoor sensor, and is comprehensively judged in combination with the short-term weather forecast, for example, the outdoor temperature may rise to 35℃ in the afternoon in summer, and drop to 28℃ at night.

[0045] The computer device selects the target cold storage amount calculation model that best fits the current working condition from multiple candidate cold storage amount calculation models according to the photovoltaic power generation power, the air conditioning system load, and the change trend of the environmental temperature. For example, the computer device selects the target cold storage amount calculation model that best fits the current working condition from the candidate cold storage amount calculation models according to the current air conditioning system running state (such as whether the phase change cold storage material is being charged, and whether the current environmental temperature is in the working interval of the phase change cold storage material). For example, when the computer device detects that the night electricity price is in the valley and the photovoltaic system stops generating electricity, the charging-cold storage electricity cold storage model is selected.

[0046] S202, input the environmental temperature and the electric energy into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model.

[0047] The phase change cold storage material is a medium (such as paraffin, salt hydrate) that stores and releases cold energy by using the material phase change process (such as solid-liquid phase change), and is used in the air conditioning system to balance the supply and demand side cold energy and reduce the compressor energy consumption in the peak electricity period.

[0048] In the embodiments of the present application, the computer device inputs the real-time acquired environmental temperature data and the electric energy input into the phase change cold storage material into the selected target cold storage amount calculation model. The environmental temperature is, for example, high temperature of 30℃ outdoors in summer or low temperature of 10℃ at night in winter, and the electric energy includes the charging amount of the photovoltaic residual electricity or the valley electricity of the power grid.

[0049] The target cold storage amount calculation model is constructed based on the physical properties of the phase change cold storage material (such as melting point, heat storage capacity) and a large amount of historical test data, and can simulate the cold storage process under different temperature and power inputs. Taking the charging process as an example, when the ambient temperature is high and the computer equipment is charging the phase change cold storage material, the target cold storage amount calculation model will comprehensively consider the factors such as the efficiency of converting electric energy into cold energy and the loss of the ambient temperature on the cold storage effect. For example, charging in a 30°C environment, due to the fact that the ambient temperature is close to the melting point of the phase change material, part of the electric energy may be lost due to heat dissipation, and the target cold storage amount calculation model needs to calculate the actual stored cold energy. The result output by the target cold storage amount calculation model is not only the specific cold storage amount value, but also provides a confidence range in combination with the data fluctuation. For example, the calculated cold storage amount is 50 kWh, and the confidence range is between 48-52 kWh, which provides a reference for subsequent strategy formulation.

[0050] In step 203, based on the cold storage amount, the photovoltaic power generation power, and the pre-acquired planning power consumption probability distribution, an energy management strategy of the air conditioning system is obtained through analysis and processing; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0051] The planning power consumption probability distribution is a future power consumption probability model calculated by a machine learning algorithm based on user historical power consumption behavior data, reflecting the possibility of user use of air conditioning and power consumption demand in different time periods.

[0052] The energy management strategy is an optimized scheduling scheme formulated by comprehensively utilizing the cold storage amount, the photovoltaic power generation power, and the planning power consumption probability distribution, including power supply source selection (mains / photovoltaic), phase change material cold storage / cooling decision, and compressor auxiliary enabling condition, and the target is to reduce operating cost and improve energy efficiency.

[0053] The compressor auxiliary cold storage is a process of enabling a traditional compressor refrigeration system to supplement cold energy when the phase change material cold storage is insufficient or rapid cold storage is needed.

[0054] In the embodiments of the present application, the computer equipment generates an energy management strategy based on the calculated phase change cold storage material cold storage amount, the photovoltaic power generation power prediction value, and the planning power consumption probability distribution. The planning power consumption probability distribution is obtained by analyzing user historical behavior, for example, the probability of the user turning on the air conditioner at 12:00-14:00 on weekdays is 85%, and the corresponding power consumption prediction value is 10 kWh; the probability of turning on at the same time on weekends is only 30%, and the power consumption prediction value is 3 kWh.

[0055] In some embodiments, photovoltaic power is prioritized to reduce grid electricity cost, phase change material is charged during off-peak hours to reduce energy storage cost, and the stored cold is released during peak hours to reduce compressor operation energy consumption. The specific strategy includes:

[0056] (1) Power supply selection: the computer device determines whether the current photovoltaic power generation meets the air conditioning load. If the photovoltaic power generation is 15 kW and the air conditioning load is 12 kW, photovoltaic power supply is preferred; if the load is 18 kW, the insufficient part is supplemented by the grid.

[0057] (2) Cold storage control: the computer device determines whether to charge or release cold with the current electricity. For example, during the night off-peak hours (such as 0:00-8:00), if there is no photovoltaic output and the air conditioning load is low, the charging program is started; during the day peak hours (such as 17:00-22:00), the stored cold is released to meet the air conditioning demand.

[0058] (3) Compressor assistance: when the phase change cold storage material is insufficient (such as less than 20% of the maximum cold storage capacity) and the air conditioning load is high, the compressor auxiliary refrigeration is automatically started to ensure the stability of indoor temperature.

[0059] The energy management strategy generated by the computer device is executed by the intelligent control system. In the power supply link, the intelligent switch device switches the power supply source in real time. For example, when the photovoltaic system generates electricity, the power is preferentially delivered to the air conditioning system, and the remaining electricity is used to charge the phase change cold storage material; if the photovoltaic power is insufficient or there is no electricity generation at night, the grid power supply is automatically switched. In the cold storage link, the computer device controls the amount of electricity input to the phase change cold storage material by adjusting the power of the charging device to avoid overcharging or insufficient cold storage; in the cold release link, the stored cold is released to the air conditioning cycle through the heat exchange system to reduce the working time of the compressor.

[0060] In the above energy management method based on the air conditioning system, in a preset period, a target cold storage amount calculation model matching a current working condition is selected from a plurality of candidate cold storage amount calculation models according to a previously acquired variation trend of photovoltaic power generation, air conditioning system load and environmental temperature; the environmental temperature and the electric energy are input into the target cold storage amount calculation model to obtain a cold storage amount of a phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model; and the cold storage amount, the photovoltaic power generation and a previously acquired planning power consumption probability distribution are analyzed and processed to obtain an energy management strategy of the air conditioning system; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to utilize the phase change cold storage material to store cold, and whether to enable the compressor to assist in performing the cold storage process; the method obtains accurate cold storage amount data by dynamically matching the target cold storage amount calculation model, generates the energy management strategy in combination with the photovoltaic power generation and the planning power consumption probability distribution, preferentially utilizes photovoltaic power supply and valley electricity price period cold storage, and reduces high-cost operation of the compressor through the phase change cold storage material to release cold, which realizes the synergy of renewable energy consumption, low-price electricity energy storage and high-price electricity peak avoidance, thereby reducing the air conditioning system power consumption cost.

[0061] In one embodiment, as shown in FIG. 3, Figure 3 The above "analyzing and processing based on the cold storage amount, the photovoltaic power generation and the previously acquired planning power consumption probability distribution to obtain the energy management strategy of the air conditioning system" includes:

[0062] Step 301, determining an optimization model according to the planning power consumption probability distribution, the photovoltaic power generation and the cold storage amount.

[0063] The optimization model is a mathematical model constructed based on the planning power consumption probability distribution, the photovoltaic power generation prediction value and the cold storage amount of the phase change cold storage material, which balances the operation cost, energy efficiency and system reliability and the like by integrating the uncertainty of power consumption demand, the volatility of renewable energy and the energy storage state and the like multidimensional data, and then solves the optimal energy management strategy. The optimization model takes the planning power consumption probability distribution, the photovoltaic power generation prediction and the cold storage amount as input parameters, takes the minimization of operation cost or carbon emission as an objective function, and sets power balance, device capacity, upper and lower limits of cold storage amount and the like constraint conditions to generate an energy scheduling scheme.

[0064] In the embodiments of the present application, the computer device constructs an optimization model based on the planned electricity consumption probability distribution, the photovoltaic power prediction value and the phase change cold storage material storage capacity. The planned electricity consumption probability distribution is generated by analyzing historical electricity consumption behavior (such as air conditioner usage frequency and load size at different time periods), for example, it is statistically concluded that the probability of air conditioner load reaching peak value from 9:00 am to 5:00 pm on weekdays is 85%. The photovoltaic power prediction combines meteorological data and photovoltaic system historical power generation data, for example, it is predicted that the photovoltaic power will be stable at 18-22 kW from 10:00 to 14:00 tomorrow. The cold storage capacity data comes from the phase change cold storage model calculation, for example, the current phase change material stores 60% of the maximum capacity (30 kWh) of cold. The optimization model takes the minimization of operation cost as the goal, sets the power balance, the upper and lower limits of the cold storage capacity and other constraint conditions, and forms a mathematical optimization framework.

[0065] In step 302, the electricity price information corresponding to the preset period is obtained; wherein the electricity price information includes the unit price of each time period and the fluctuation range of the time-of-use electricity price.

[0066] The electricity price information refers to the electricity price data corresponding to the preset period, specifically including the unit price of each time period and the fluctuation range of the time-of-use electricity price. For example, the time-of-use electricity price may divide a day into peak period (such as 17:00-22:00, electricity price 1.2 yuan / kWh), flat period (such as 11:00-17:00, electricity price 0.8 yuan / kWh) and valley period (such as 0:00-8:00, electricity price 0.3 yuan / kWh), and at the same time, the electricity price of each period may have a fluctuation range of ±10% to cope with the real-time price changes of the electricity market. The role of the electricity price information is to guide the system to preferentially use the power grid for energy storage or charging during the electricity price valley period, to reduce the use of grid electricity during the electricity price peak period, and to meet the load demand through photovoltaic power or energy storage release instead, so as to reduce the overall operation cost.

[0067] In the embodiments of the present application, the computer device obtains the electricity price information in the preset period, including the unit price and the fluctuation range of each time period. The time-of-use electricity price is usually divided into valley period (such as 0:00-8:00, unit price 0.35 yuan / kWh), flat period (such as 8:00-17:00, unit price 0.65 yuan / kWh) and peak period (such as 17:00-22:00, unit price 1.05 yuan / kWh), and the fluctuation range is set according to the electricity market rules (such as ±10%).

[0068] In step 303, the risk control parameters in the air conditioning system running process are obtained, including the photovoltaic output uncertainty tolerance threshold, the load fluctuation range and the indoor temperature control stability index.

[0069] The risk control parameters are a set of indexes for quantifying and controlling the uncertainty risk in the energy management process, and are intended to ensure that the air conditioning system can still operate stably and meet the indoor environmental requirements when facing uncertain factors such as photovoltaic output fluctuation and load change. Specifically, the risk control parameters include:

[0070] In the embodiment of the present application, the computer device obtains the risk control parameters of the air conditioning system operation.

[0071] The photovoltaic output uncertainty tolerance threshold is the maximum deviation range (such as ±15%) between the predicted value and the actual value of the photovoltaic power generation, which is used to avoid power shortage or excess caused by prediction error and ensure that the system reserves sufficient adjustment capacity.

[0072] The load fluctuation range is the maximum amplitude (such as ±20%) of the actual power demand deviating from the predicted value. By setting this parameter, the air conditioning system can reserve redundant capacity in energy scheduling to prevent insufficient cooling caused by sudden increase in load.

[0073] The indoor temperature control stability index is the allowable indoor temperature fluctuation range (such as ±1℃), which is used to constrain the adjustment accuracy of the air conditioning system to ensure that the indoor thermal environment always remains within the user's comfortable range during energy scheduling, and to avoid large temperature fluctuations caused by frequent adjustment of equipment operating state.

[0074] In step 304, the electricity price information and the risk control parameters are substituted into the optimization model for solving processing to obtain an optimal solution, and the optimal solution is determined as the energy management strategy.

[0075] In the embodiment of the present application, the computer device substitutes the electricity price information and the risk control parameters into the optimization model and solves the optimal solution by algorithm. For example, the phase change material is charged with low-price electricity of 0.35 yuan / kWh during the low valley period, the cold storage capacity is preferentially released (avoiding the use of high-price electricity of 1.05 yuan / kWh) during the peak period, and the power supply proportion is allocated in combination with the photovoltaic power prediction value (such as 15kW). The solving result forms the energy management strategy, which specifically includes: the power supply proportion of photovoltaic, grid, and cold storage in each period (such as 70% of photovoltaic power supply and 30% of cold storage release during the peak period), the charging and discharging period and power of the phase change material (such as charging at a power of 6kW for 4 hours during the low valley period), and the compressor start-stop condition (such as starting when the cold storage capacity is less than 20% and the load exceeds the prediction by 15%).

[0076] In some embodiments, after the energy management strategy is generated, the computer device simulates scenarios such as photovoltaic output fluctuation ± 15%, load fluctuation ± 20%, etc., to verify whether the indoor temperature is controlled within a range of ± 1°C and whether there is a gap in power supply. In a preset period (such as every 15 minutes), the computer device obtains actual data in real time, and if photovoltaic power drops or load surges exceed the risk threshold, the model is immediately re-optimized and the strategy is updated to ensure that energy scheduling always adapts to real-time changes in operating conditions.

[0077] In the above embodiments, by integrating the planned electricity consumption probability distribution, photovoltaic power generation, cold storage capacity, electricity price information, and risk control parameters into the optimization model for solving, an energy management strategy that takes into account the cost difference of time-of-use electricity prices, photovoltaic output fluctuations, and load uncertainty can be dynamically generated. This strategy achieves cold storage during off-peak electricity price periods, releases cold during peak periods to replace high-priced grid electricity, and reduces electricity purchase costs through photovoltaic power priority. At the same time, while ensuring stable indoor temperature, the strategy reduces the dependence of the air conditioning system on high-priced grid electricity, thereby effectively reducing overall electricity costs.

[0078] In one embodiment, as shown in FIG. 4, Figure 4 The above "determining an optimization model according to the planned electricity consumption probability distribution, photovoltaic power generation, and cold storage capacity" includes:

[0079] Step 401, based on the planned electricity consumption probability distribution, normalizing the photovoltaic power generation and the cold storage capacity, and constructing a decision variable matrix according to the results of the normalization.

[0080] The decision variable matrix is a two-dimensional data structure composed of normalized photovoltaic power generation and cold storage capacity according to the planned electricity consumption probability distribution. The rows of the matrix represent different time periods within a preset period (such as every 15 minutes as a period, a total of 96 periods per day), and the columns represent different decision variables (such as photovoltaic power supply proportion, grid power supply proportion, and phase change material charging cold power normalized value). For example, the decision variables for a certain period may be: photovoltaic power normalized value 0.8 (corresponding to actual power 20kW), cold storage capacity normalized value 0.7 (corresponding to 35kWh), and charging cold power normalized value 0.3 (corresponding to charging power 15kW). The decision variable matrix provides structured input data for the optimization model, facilitating matrix operations or algorithm iterations to solve the optimal decision combination.

[0081] In the embodiments of the present application, the computer device first normalizes the photovoltaic power generation and the cold storage capacity based on the planned electricity consumption probability distribution. The planned electricity consumption probability distribution is used to analyze user historical electricity consumption data (such as the probability density of air conditioning load at different times) to provide statistical characteristics (such as mean, variance) of electricity demand at each time period.

[0082] wherein, the photovoltaic power normalization is to scale the predicted power value to the interval [0, 1] according to its maximum value and minimum value, for example, if the photovoltaic power prediction of a certain period is 10-30kW, then the actual value 20kW is normalized to 0.5. The cold storage capacity normalization is to proportionally convert the real-time cold storage capacity of the phase change material according to its maximum cold storage capacity, for example, if the maximum cold storage capacity is 50kWh, then the current cold storage capacity 30kWh is normalized to 0.6. Through normalization, the interference of different physical dimensions (power, energy) on model calculation is eliminated, and the data is comparable.

[0083] In step 402, an optimization model with the minimum air conditioning system operation cost as the objective function is constructed based on the decision variable matrix.

[0084] wherein, the objective function is a mathematical expression with the minimum air conditioning system operation cost as the core, including the cost of purchasing electricity from the grid, the cost of equipment loss, etc., as shown in formula (1).

[0085] (1)

[0086] wherein, t represents the time variable, representing a single period within the preset period. For example, if 15 minutes is taken as a period, one day can be divided into 96 periods (t=1, 2,..., 96). N represents the total number of time variables, i.e. the complete time range from the first period to the Nth period. For example, if the operation cost of the next 24 hours is calculated, and each period is 1 hour, then N=24; if the period is 15 minutes, then N=96.

[0087] In the embodiment of the present application, the computer device constructs a decision variable matrix according to the normalized photovoltaic power and cold storage capacity. The rows of the decision variable matrix correspond to each period within the preset period, and the columns correspond to different decision variables (such as photovoltaic power supply ratio, grid power supply ratio, cold storage capacity state, and cold charging power coefficient). For example, the normalized photovoltaic power of a certain period is 0.8, and the cold storage capacity is 0.7. The decision variables may include photovoltaic power supply ratio 0.7, grid power supply ratio 0.3, and cold charging power coefficient 0.2 (indicating that the maximum charging power is 20% for cold storage). The decision variable matrix integrates multi-dimensional decision information in a structured form, providing a unified input framework for subsequent model construction. The computer device constructs an optimization model based on the decision variable matrix with the minimum air conditioning system operation cost as the objective.

[0088] In the above embodiments, by normalizing photovoltaic power generation and cold storage capacity based on the planned electricity consumption probability distribution, and constructing a decision variable matrix and an optimization model, the key data of different dimensions and different magnitudes can be converted into a unified scale, so that the optimization model can more accurately capture the correlation and influence between factors, and then systematically optimize the power supply mode, cold storage strategy and equipment operation arrangement of the air conditioning system, realize the coordinated scheduling of efficient use of renewable energy, energy storage during low valley electricity price period and energy release during peak period, and effectively reduce the overall electricity cost and operation energy consumption.

[0089] In one embodiment, as shown in Figure 5 The above "constructing an optimization model with the minimum running cost of the air conditioning system as the objective function based on the decision variable matrix" includes:

[0090] Step 501, determining an optimization variable set based on the decision variable matrix, the optimization variable set at least including the cold storage power proportion of the phase change cold storage material, the utilization proportion of the photovoltaic power generation, the power supply mode switching time point of the air conditioner, the start-stop state of the compressor and the running time length.

[0091] The optimization variable set is an adjustable parameter extracted from the decision variable matrix, which is used to find the optimal solution in the optimization model. Specifically, it includes:

[0092] (1) The cold storage power proportion of the phase change cold storage material: the ratio of the cold storage power to the maximum allowed cold storage power (such as 0.5 indicating charging at 50% of the maximum power), which is used to control the energy storage speed and cost.

[0093] (2) The utilization proportion of the photovoltaic power generation: the distribution proportion of photovoltaic power in air conditioning load (such as 0.7 indicating that 70% of the load is supplied by photovoltaic power), which reflects the utilization degree of renewable energy.

[0094] (3) The power supply mode switching time point of the air conditioner: the specific time point of switching the power supply mode between the grid, photovoltaic and cold storage (such as 17:00 switching from photovoltaic power supply to cold storage release and grid supplement), which is used to avoid peak electricity price or optimize energy scheduling.

[0095] (4) The start-stop state and running time length of the compressor: the on-off state (such as "start-stop" or "running") and continuous running time (such as running for 30 minutes) of the compressor in each period, which affects the refrigeration efficiency and energy consumption cost.

[0096] In the embodiment of the present application, the computer device selects core adjustable parameters affecting energy scheduling from the decision variable matrix to form an optimization variable set. For example, the proportion of the charging power of the phase change cold storage material is extracted, which represents the relationship between the charging power and the maximum allowed power, such as 0.6 representing charging at 60% of the maximum power; the utilization proportion of photovoltaic power is extracted, such as 0.8 representing that 80% of the photovoltaic power is used to meet the air conditioning load. In addition, the power supply mode switching time point is determined, such as setting the time point of switching from photovoltaic power supply to the combination of municipal power and cold storage cold release as 18:00; the start-stop state and the running time of the compressor are recorded, such as the compressor running for 30 minutes in a certain period. These variables directly affect the energy distribution and cost of the air conditioning system.

[0097] In step 502, a target function is constructed based on the optimization variable set, and the target function is used to represent the running cost of the air conditioning system in a preset period. The running cost includes the power cost generated by municipal power supply, the energy consumption cost generated by compressor operation, and the cost generated by the use of electric energy in the cold storage process.

[0098] The target function is a mathematical expression for quantifying the running cost of the air conditioning system, and the minimization of the total cost in the preset period is the optimization direction. The components include:

[0099] (1) Municipal power supply power cost: the sum of the product of the power consumption of each period using municipal power and the corresponding electricity price (for example, the unit price of municipal power in the peak period is 1.2 yuan / kWh, and the cost of using 5 kWh is 6 yuan).

[0100] (2) Compressor operation energy consumption cost: the product of the compressor operation power, the time length, and the device loss coefficient (for example, the operation power is 10 kW, the time length is 2 hours, and the loss coefficient is 0.1 yuan / kWh, and the cost is 2 yuan).

[0101] (3) Cold storage process electric energy use cost: the product of the consumed electric quantity in the cold storage process (such as charging by using off-peak municipal power) and the corresponding electricity price (such as the off-peak electricity price of 0.3 yuan / kWh, and the cost of charging 10 kWh is 3 yuan). By weighted sum of these cost items, the target function provides a clear optimization target for the optimization model.

[0102] In the embodiment of the present application, based on the optimization variable set, the computer device constructs a target function for quantifying the running cost.

[0103] First, the municipal power supply cost is calculated: the unit price of municipal power in the off-peak period is 0.3 yuan / kWh, and if 5 kWh of municipal power is used in a certain period, the cost is 1.5 yuan; the unit price in the peak period is 1.2 yuan / kWh, and if 3 kWh is used, the cost is 3.6 yuan, and the total municipal power cost is obtained by accumulating the cost of each period.

[0104] Secondly, calculate the compressor energy consumption cost: the compressor operating power is 10 kW, if a certain period of time runs for 1 hour, the equipment loss coefficient is 0.1 yuan / kWh, then the loss cost of this period is 1 yuan, and the loss of all running periods is accumulated. Finally, calculate the cost of the cold storage process: use the city power to charge in the valley period, the charging power is 5 kW (corresponding to the cold storage power ratio 0.5, the maximum power 10 kW), 10 kWh is consumed in 2 hours of charging, and the cost is 3 yuan according to the low valley electricity price 0.3 yuan / kWh.

[0105] Add the above three costs to form the objective function, and minimize the total operation cost as the optimization direction.

[0106] Step 503, combine the objective function with the operation constraints to build an optimization model with the objective function of minimizing the operation cost of the air conditioning system, and the constraints include at least that the photovoltaic power generation power does not exceed the maximum output capacity, the phase change cold storage material storage capacity is within the rated range, the power load meets the planned power demand, the switching of the air conditioning system operating state meets the minimum time interval requirement, and the indoor temperature is maintained within the preset temperature interval.

[0107] Among them, the operation constraint is a mandatory rule to ensure the safe and stable operation of the air conditioning system, which is used to limit the value range of the optimization variable. Specifically, it includes:

[0108] (1) Photovoltaic power generation power limit: the actual power corresponding to the photovoltaic utilization ratio should not exceed the maximum output capacity of the photovoltaic system (for example, the maximum output of photovoltaic is 20 kW, and the actual power corresponding to the utilization ratio of 0.8 is 16 kW≤20 kW).

[0109] (2) Phase change cold storage material storage capacity boundary: the storage capacity needs to be kept within the rated range (such as 20%-100% capacity), to avoid equipment damage due to overcharging (more than 100%) or affecting the cooling capacity due to over-discharging (less than 20%).

[0110] (3) Power load matching: the total power supply (photovoltaic+city power+cold storage cooling release) in each period needs to meet the planned power demand, such as a predicted load of 15 kW, and the power supply needs to be ≥15 kW×(1-load fluctuation tolerance).

[0111] (4) Running state switching limit: the switching of the air conditioning system between different power supply modes or equipment start-stop states needs to meet the minimum time interval (such as the compressor start-stop interval≥10 minutes), to prevent frequent switching from causing equipment loss.

[0112] (5) Indoor temperature control: through adjustment means such as cold storage cooling release and compressor operation, ensure that the indoor temperature is maintained within the preset interval (such as 24℃±1℃).

[0113] In the embodiments of the present application, the computer device converts physical limitations and comfort requirements into specific constraints. For example, the photovoltaic power constraint requires that the actual power corresponding to the utilization ratio of photovoltaic power generation does not exceed the maximum output capacity. If the maximum output of photovoltaic power is 20 kW, and the utilization ratio is 0.9, the actual power is 18 kW, which meets the limitation of not exceeding 20 kW. The cold storage capacity needs to be maintained within the rated range, for example, the maximum cold storage capacity of the phase change material is 50 kWh, and the current cold storage capacity needs to be between 10 kWh (20%) and 50 kWh (100%). The load matching constraint requires that the total power supply meets the planning demand, for example, the predicted load is 15 kW, and the fluctuation is allowed to be 10%, so the power supply needs to be at least 13.5 kW. The state switching constraint stipulates that the compressor needs to be started and stopped at least 10 minutes apart to avoid frequent operation, for example, after the compressor is stopped in a certain period, it cannot be started immediately in the next period. The temperature control constraint requires that the indoor temperature be maintained at 24±1℃, which is achieved by adjusting the cold storage release rate and the compressor operation power, for example, when the temperature approaches 25℃, the compressor operation time is increased or the cold storage release amount is increased.

[0114] The computer device combines the objective function and the constraint conditions to form an optimization model. Taking linear programming as an example, the objective function is the linear accumulation of the electricity cost, compressor loss, and cold storage charging cost in each period, and the constraint conditions are linear inequalities such as power, cold storage capacity, and load. Through the solver, the optimal variable combination is obtained. Exemplarily, in the low valley period (0:00-8:00), the cold storage power ratio is set to 1 (full power charging), and the photovoltaic utilization ratio is 1 (full use of photovoltaic power); in the peak period (17:00-22:00), the cold storage power ratio is 0 (stop charging and preferentially release cold), and the photovoltaic utilization ratio is 0.7 (the remaining 30% load is supplemented by the grid), and the compressor is started when the cold storage capacity is insufficient, and the single operation lasts at least 15 minutes. The strategy generated by the model ensures that the running cost is the lowest under the premise of meeting all the constraints, such as reducing the cold storage cost through low valley charging, reducing the use of grid power in the peak period to avoid high electricity prices, while ensuring stable indoor temperature.

[0115] In the above embodiments, by defining core optimization variables such as the cold storage power ratio of the phase change cold storage material and the utilization ratio of photovoltaic power generation, the air conditioning system operation cost is refined into grid electricity cost, compressor energy consumption, and cold storage electricity cost, and then an optimization model is constructed by combining multi-dimensional constraints such as photovoltaic power limitation, cold storage capacity boundary, and load demand matching, which can systematically plan the air conditioning system operation strategy from the whole element and whole process level.

[0116] In one embodiment, as shown in Figure 6 The above "substituting the electricity price information and the risk control parameter into the optimization model for solving and processing to obtain an optimal solution" includes:

[0117] Step 601, input the electricity price information and the risk control parameter into the optimization model, combine the optimization variable set and the target function.

[0118] In the embodiment of the present application, the computer device first inputs the acquired electricity price information and risk control parameter into the constructed optimization model. At the same time, the computer device integrates the optimization variable set and the target function oriented to minimize the operation cost into the optimization model, so that the optimization model has complete input elements and optimization direction.

[0119] Step 602, under the constraint of the target of minimizing the operation cost, a plurality of optimization variable sets are solved, and the optimal solution satisfying all constraint conditions is determined according to the plurality of optimization variable sets.

[0120] In the embodiment of the present application, under the constraint of the target of minimizing the operation cost, the computer device uses a mathematical solving algorithm to solve the optimization model, and generates a plurality of optimization variable sets.

[0121] For example, scheme one: in the valley period, the phase change cold storage material cold storage power ratio is 0.8, and the photovoltaic power utilization ratio is 0; scheme two: in the same period, the phase change cold storage material cold storage power ratio is 0.7, and the photovoltaic power utilization ratio is 0.1. These different variable combinations form a plurality of candidate solutions, and each solution corresponds to the value condition of each key parameter under an energy management strategy.

[0122] In some embodiments, the computer device verifies the plurality of optimization variable sets solved one by one to determine whether they satisfy all operation constraint conditions. For example, whether the photovoltaic power utilization ratio exceeds the maximum output capacity, whether the phase change cold storage material cold storage capacity is within the rated range, whether the electricity load can meet the planned electricity demand, and the like. For example, in a certain optimization variable set, the photovoltaic power utilization ratio reaches 1.2, which exceeds the maximum output capacity, so the solution is excluded; in another set, the indoor temperature exceeds the range of 24℃±1℃, which also does not meet the requirements. After screening, the optimal solution that satisfies all constraint conditions and can minimize the operation cost is determined from the plurality of candidate solutions, and the optimal solution is the final energy management strategy parameter combination.

[0123] In the above embodiment, the electricity price information and the risk control parameter are integrated into the optimization model, and the optimization variable set and the target function are solved, which can find the optimal operation scheme on the basis of time-of-use electricity price difference and system operation risk. This can not only utilize the low valley electricity price period to reduce cold storage cost and reduce the use of high peak period high price grid electricity, but also can guarantee the stable operation of the system through the risk control parameter when the photovoltaic output fluctuates and the load changes, avoid sacrificing indoor temperature comfort or equipment operation reliability due to excessive pursuit of cost reduction, and finally realize the balance between cost control and risk prevention of the air conditioning system, and effectively reduce the comprehensive electricity cost.

[0124] In one embodiment, the method further comprises:

[0125] inputting the environmental data into a pre-constructed photovoltaic power generation amount prediction model to obtain photovoltaic power output by the photovoltaic power generation amount prediction model.

[0126] In the embodiments of the present application, the computer device first collects environmental data affecting photovoltaic power generation, including real-time and historical meteorological information such as solar irradiance, temperature, humidity, wind speed, cloud conditions, and geographic location information of the photovoltaic power station, including longitude, latitude, altitude, and device parameters such as photovoltaic panel type, installation angle, area, and conversion efficiency. For example, the computer device collects hourly solar irradiance data, temperature changes, and specific information such as a photovoltaic panel installation angle of 30 degrees and a conversion efficiency of 18% for a certain location on a given day. After the collection is complete, the computer device cleans the raw data to remove obvious error values, such as negative solar irradiance values, and fills in missing data. If temperature data is missing for a certain period, the average value of the previous and subsequent periods is used to supplement it, thereby ensuring the accuracy and completeness of the data.

[0127] The computer device extracts key features from the processed environmental data. The horizontal solar irradiance is converted to the actual effective solar irradiance of the inclined surface of the photovoltaic panel, because the installation angle of the photovoltaic panel affects the actual amount of light received; the actual operating temperature is calculated based on the temperature and temperature characteristics of the photovoltaic panel, and the increase in temperature will reduce the power generation efficiency. Subsequently, the computer device organizes these features in chronological order into the input format required by the model, forming multi-dimensional data such as hourly solar irradiance, temperature, humidity, etc., and performs standardization processing to adjust the data range, facilitating model recognition and calculation.

[0128] The computer device calls a pre-trained photovoltaic power generation prediction model, which can be a machine learning model trained based on historical data or a model constructed based on physical principles. Taking a physical model as an example, the photovoltaic power generation prediction model will integrate parameters such as solar irradiance, photovoltaic panel area, conversion efficiency, and temperature to calculate power generation at different times. If it is a machine learning model, it automatically finds the relationship between environmental features and power generation by analyzing the rules in historical data. For example, if the solar irradiance at 10 am is 800 W / m² and the temperature is 28°C, the model can calculate the photovoltaic power at that time to be 20 kW.

[0129] The computer device verifies the power generation prediction value output by the photovoltaic power generation prediction model, compares the prediction result with the historical actual power generation data, evaluates the prediction error, and usually requires that the error be controlled within 5%. If the error is too large, the computer device will automatically trigger the model optimization process and update or retrain the model parameters. After verification, the computer device outputs the hourly photovoltaic power generation prediction result for a future period of time, generates a power generation power curve such as tomorrow 9:00-17:00, shows the expected power generation capacity of each period, and provides a basis for formulating an energy management strategy for the air conditioning system.

[0130] In the above embodiment, the environmental data is input into the photovoltaic power generation prediction model to obtain the photovoltaic power generation power. Based on real-time meteorological conditions and equipment parameters, the power generation capacity of the photovoltaic system can be predicted. By predicting the trend of photovoltaic power generation power in advance, the air conditioning system can optimize the energy storage and power supply strategy, for example, preferentially using clean energy when photovoltaic power is sufficient to reduce power consumption; in the power generation valley period, pre-cooling is performed to avoid high-priced power purchase, thereby effectively reducing the overall power consumption cost and improving the utilization efficiency of renewable energy.

[0131] In one embodiment, the method further comprises:

[0132] The historical air conditioner usage data is input into a pre-constructed behavior prediction model to obtain a probability distribution of planned power consumption output by the behavior prediction model.

[0133] In the embodiments of the present application, the computer device first comprehensively collects historical air conditioner usage data of users or in a region. The historical air conditioner usage data can cover multiple dimensions, including time information, such as date, hour, day of the week, and season; environmental parameters, such as indoor and outdoor temperature and humidity; equipment parameters, such as air conditioner power, set temperature, and operating mode; and user behavior data, such as on-off time, temperature adjustment frequency, and actual power consumption.

[0134] For example, the computer device collects that the air conditioner in an office is turned on at 9:00 am and turned off at 6:00 pm every weekday in the past year, the set temperature is 26℃ in summer and 22℃ in winter, and the power consumption in the corresponding period is recorded. The computer device integrates data from multiple channels such as smart meters and air conditioner control logs to ensure that the data covers a complete time period and avoids missing or missing data.

[0135] After obtaining the raw data, the computer device performs cleaning operations on it. Obvious outliers, such as unreasonable data where the power consumption of a certain hour is suddenly several times higher than usual, are removed. For missing values, a reasonable method is used to fill them in. If the set temperature for a certain period is not recorded, the average of the set temperatures for the previous and subsequent periods is used to fill it in. After cleaning, the computer device further extracts key features. In terms of time, the date is converted into a seasonal attribute, distinguishing between weekdays and weekends, and dividing into specific hour segments (such as the morning rush hour, lunch break, and night). In terms of environmental features, the indoor-outdoor temperature difference is calculated, and the humidity trend is analyzed. In terms of user behavior features, the daily air conditioner on-time, temperature adjustment frequency, and operating mode switching frequency are counted. In addition, external information such as holidays and weather warnings is used to generate associated features. For example, the data "2024-08-10 15:00" is labeled as "summer, weekend, afternoon period", and is associated with the information that the outdoor temperature is 36°C and the indoor set temperature is 25°C on that day.

[0136] The computer device uses the cleaned and feature-extracted data to train the pre-constructed behavior prediction model. These models can be random forest, recurrent neural network, etc. During training, historical data is used as input, and the probability of electricity consumption in the future period is used as output target, for example, the probability of electricity consumption between 4-5 degrees at 17:00-18:00 tomorrow is 70%. To ensure the accuracy and generalization ability of the model, the computer device uses cross-validation to adjust the model parameters to prevent overfitting. In a specific embodiment, the model is trained using data from the past 300 days, with the last 30 days reserved as a test set. The performance of the behavior prediction model is evaluated by calculating the error (such as root mean square error less than 0.5 degrees) between the predicted probability and the actual electricity consumption.

[0137] The computer device inputs real-time data, including the current time, real-time indoor and outdoor temperature, and user's latest air conditioner settings, into the trained behavior prediction model. The model calculates the probability distribution of planned electricity consumption in future periods based on the rules learned during historical data training and the real-time input feature information. For example, input "current time is 2025-05-30 10:00, outdoor temperature is 29°C, user sets cooling to 24°C", the model calculates the probability distribution of electricity consumption from 13:00 to 15:00 today: 3-4 degrees (25%), 4-5 degrees (55%), 5-6 degrees (20%).

[0138] In some embodiments, the computer device does not directly output the probability distribution result calculated by the behavior prediction model as the final output, but calibrates the result in combination with real-time dynamic factors. If the weather forecast shows that there will be a sudden temperature rise or other sudden weather changes in the next period, or it is monitored that the user has temporarily adjusted the air conditioning setting, the probability of the high electricity consumption interval in the corresponding period will be adjusted accordingly. After calibration, the computer device outputs the electricity consumption probability distribution result of each period in a clear and intuitive way, which can be in the form of a table listing the probability of each electricity consumption interval in each period, or in the form of a chart, such as a bar chart showing the probability values of each electricity consumption interval, providing a reliable basis for the energy management system to make a scientific scheduling strategy or for the user to reasonably optimize the electricity consumption plan.

[0139] In the above embodiments, the historical air conditioning usage data is input into the behavior prediction model to obtain the planned electricity consumption probability distribution, which can deeply mine the user's electricity consumption behavior rules, combine environmental factors and device parameters, and accurately predict the electricity demand fluctuation in different periods. Based on the probability distribution, the air conditioning system can optimize the cold storage and power supply strategy in advance, release the cold storage capacity or preferentially use photovoltaic power to meet the load demand during the electricity consumption peak period, avoid the consumption of a large amount of high-priced commercial power, and dynamically adjust the cold storage scale during the low valley period to balance the electricity cost and device energy consumption, thereby realizing fine control and effective reduction of the whole cycle electricity cost.

[0140] In one embodiment, as shown in FIG. 7, the method further includes: Figure 7

[0141] Step 701, real-time acquisition of environmental temperature disturbance and photovoltaic output deviation.

[0142] The environmental temperature disturbance refers to unexpected deviation or fluctuation between the actual environmental temperature and the predicted value or normal range. For example, if the outdoor temperature is predicted to be 28°C according to the weather forecast, but the actual temperature rises to 33°C due to sudden weather changes, or drops to 22°C due to cloud cover, such unexpected temperature changes constitute environmental temperature disturbance.

[0143] The photovoltaic output deviation is the difference between the actual power generation of the photovoltaic system and the predicted value. For example, the predicted photovoltaic output is 15kW per hour, but due to factors such as sudden decrease in light intensity, equipment failure or abnormal environmental temperature, the actual output is only 12kW, and the difference between the two is 3kW, which is the photovoltaic output deviation. This deviation reflects the uncertainty of renewable energy generation.

[0144] ​In the embodiments of the present application, the computer device collects environmental temperature and actual power generation data of the photovoltaic system in real time through the outdoor temperature sensor and the photovoltaic power station power monitoring device. For example, the outdoor temperature is obtained every 15 minutes, and compared with the predicted temperature of the period to calculate the temperature disturbance value; at the same time, the real-time power of the photovoltaic inverter is read, and the output deviation value is obtained by comparing the predicted power.

[0145] The computer device compares the collected temperature disturbance and photovoltaic output deviation with the pre-set threshold. The pre-set conditions may include that the temperature disturbance does not exceed ±2℃, the photovoltaic output deviation does not exceed ±10% of the predicted value, and the like. If the actual temperature disturbance reaches +3℃, or the photovoltaic output is only 85% of the predicted value, which exceeds the pre-set threshold, the computer device determines that the current environment is abnormal, and triggers the rolling horizon control process.

[0146] Step 702, in the case that the environmental temperature disturbance and / or the photovoltaic output deviation does not meet the pre-set conditions, the model parameters of each candidate cold storage capacity calculation model are updated by using the rolling horizon control method.

[0147] The pre-set conditions are pre-set standards or thresholds for judging whether the environmental temperature disturbance and the photovoltaic output deviation are within an acceptable range. For example, it is set that “the environmental temperature disturbance should not exceed ±3℃” or “the photovoltaic output deviation should not exceed ±15% of the predicted value”. When the actual disturbance or deviation exceeds these thresholds, it indicates that the current air conditioning system operating environment has deviated significantly from the expectation, and the control mechanism needs to be triggered to optimize and adjust the model.

[0148] The rolling horizon control is a dynamic optimization control method. In each control period, the candidate cold storage capacity calculation model is updated based on the latest real-time data (such as the current environmental temperature and the actual photovoltaic output), and the optimization problem is solved again to generate a control strategy that adapts to the current state. For example, the rolling optimization is performed every 15 minutes, and the latest 15 minutes of data is used to correct the cold storage capacity calculation model parameters for the next 1 hour, so as to ensure that the strategy always fits the actual operation.

[0149] The model parameters are adjustable variables in the cold storage capacity calculation model for characterizing the physical properties of the system, and the values thereof directly affect the reliability of the model output results. For example, the candidate cold storage capacity calculation model includes parameters such as the heat loss coefficient of the cold storage device (for calculating the cold loss in the cold storage process), the environmental temperature sensitivity factor (reflecting the influence of temperature change on the cold storage rate), and the coupling coefficient of photovoltaic output and cold storage power (for coordinating the relationship between the excess electricity of photovoltaic and the cold storage demand).

[0150] In the embodiments of the present application, when an anomaly is detected, the computer device starts the rolling horizon control, sets the current optimization period and divides it into multiple sub-periods. Based on the latest temperature, photovoltaic output data and historical operation records, the current state of the air conditioning system is updated, including the real-time value of the cold storage capacity and the prediction of the air conditioning load, providing an initial basis for subsequent model parameter optimization.

[0151] For each candidate cold storage capacity calculation model, the computer device optimizes its parameters using the rolling horizon control algorithm. First, input real-time temperature disturbance, photovoltaic deviation and other data into the model, then minimize the prediction error of the cold storage capacity as the target, combine with the optimization targets such as operation cost, cold storage efficiency, etc., to build the target function of parameter adjustment. At the same time, considering the physical constraints such as the upper and lower limits of the cold storage capacity, the power limit of the equipment, etc., the model parameters such as heat loss coefficient, temperature influence factor on cold storage rate, etc. are iteratively adjusted through optimization algorithms such as gradient descent, so that the candidate cold storage capacity calculation model is more suitable for the cold storage characteristics under the current abnormal environment.

[0152] After parameter updating, the computer device verifies the prediction accuracy of each candidate cold storage capacity calculation model, compares the cold storage capacity output by the candidate cold storage capacity calculation model with the actual measured value (such as the real-time cold storage capacity calculated by the temperature of the cold storage device). If the prediction error of a certain candidate cold storage capacity calculation model is significantly reduced (such as a 15% reduction in mean square error), it will be set as the target cold storage capacity calculation model for subsequent cold storage capacity calculation; if none of the models meet the accuracy requirements, the computer device will call the backup model or start the model training process to ensure the reliability of the cold storage strategy.

[0153] In the above embodiments, the environmental temperature disturbance and photovoltaic output deviation are obtained in real time, and the parameters of the candidate cold storage capacity calculation model are updated based on the rolling horizon control, which can make the model quickly adapt to environmental changes and power fluctuations. By continuously correcting the model parameters, the real-time cold storage capacity of the phase change cold storage material can be accurately calculated, ensuring that the cold storage strategy can be dynamically optimized when the temperature changes suddenly or the photovoltaic output is unstable, such as increasing the cold storage capacity in time when the photovoltaic output is excessive, to avoid insufficient or excessive cold storage due to model lag, thereby effectively reducing the additional electricity cost of the air conditioning system due to energy scheduling mismatch, and improving the operation economy and reliability.

[0154] In one embodiment, the method further comprises:

[0155] Step 1. In a predetermined period, select a target cold storage capacity calculation model that matches the current working condition from a plurality of candidate cold storage capacity calculation models according to the variation trends of the pre-acquired photovoltaic power generation, air conditioning system load and environmental temperature.

[0156] Step 2. Input the environmental temperature and electrical energy into the target cold storage capacity calculation model to obtain the cold storage capacity of the phase change cold storage material in the air conditioning system output by the target cold storage capacity calculation model.

[0157] Step 3, input the environmental data into the pre-constructed photovoltaic power generation prediction model to obtain the photovoltaic power output by the photovoltaic power generation prediction model.

[0158] Step 4, input the historical air conditioner usage data into the pre-constructed behavior prediction model to obtain the planning electricity consumption probability distribution output by the behavior prediction model.

[0159] Step 5, based on the planning electricity consumption probability distribution, normalize the photovoltaic power and the cold storage capacity, and construct a decision variable matrix according to the results of the normalization processing.

[0160] Step 6, determine an optimization variable set based on the decision variable matrix, the optimization variable set at least including the cold storage power ratio of the phase change cold storage material, the utilization ratio of the photovoltaic power, the power supply mode switching time point of the air conditioner, the start-stop state of the compressor, and the running time length.

[0161] Step 7, construct a target function based on the optimization variable set, the target function being used to represent the operation cost of the air conditioning system within a preset period, the operation cost including the power cost generated by the mains power supply, the energy consumption cost generated by the compressor operation, and the cost generated by the use of electric energy in the cold storage process.

[0162] Step 8, combine the target function with the operation constraint conditions to construct an optimization model with the minimization of the operation cost of the air conditioning system as the target function, the constraint conditions at least including that the photovoltaic power does not exceed the maximum output capacity, the cold storage capacity of the phase change cold storage material is within the rated range, the electricity load meets the planning electricity demand, the switching of the air conditioning system operation state meets the minimum time interval requirement, and the indoor temperature is maintained within the preset temperature interval.

[0163] Step 9, obtain the electricity price information corresponding to the preset period; wherein the electricity price information includes the unit price of each time period and the time-of-use electricity price fluctuation range.

[0164] Step 10, obtain the risk control parameters in the operation process of the air conditioning system, the risk control parameters including the photovoltaic output uncertainty tolerance threshold, the load fluctuation range, and the indoor temperature control stability index.

[0165] Step 11, input the electricity price information and the risk control parameters into the optimization model, combine the optimization variable set and the target function.

[0166] Step 12, under the constraint of the minimization of the operation cost, solve to obtain a plurality of optimization variable sets, and determine the optimal solution that meets all the constraint conditions according to the plurality of optimization variable sets, and determine the optimal solution as the energy management strategy; the energy management strategy is used to guide the air conditioning system to adopt the mains power supply or the photovoltaic power supply, whether to utilize the phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0167] Step 13, real-time acquisition of environmental temperature disturbance and photovoltaic output deviation;

[0168] Step 14, in the case that the environmental temperature disturbance and / or the photovoltaic output deviation does not meet the preset condition, updating the model parameters of each candidate cold storage amount calculation model by using a rolling horizon control mode.

[0169] It should be understood that, although each step in the flowchart involved in each of the above embodiments is shown in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each of the above embodiments can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be alternately executed with at least part of other steps or steps or stages in other steps.

[0170] Based on the same inventive concept, the embodiments of the present application also provide an air conditioning system-based energy management device for implementing the above-mentioned air conditioning system-based energy management method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more air conditioning system-based energy management device embodiments provided below can refer to the limitations of the air conditioning system-based energy management method described above, which will not be repeated here.

[0171] In one embodiment, as shown in Figure 8 An air conditioning system-based energy management device is provided, comprising: a matching module 801, an output module 802 and an analysis module 803, wherein:

[0172] The matching module 801 is configured to select, in a preset period, a target cold storage amount calculation model matching a current working condition from a plurality of candidate cold storage amount calculation models according to the variation trends of the pre-acquired photovoltaic power generation, air conditioning system load and environmental temperature.

[0173] The output module 802 is configured to input the environmental temperature and the electric energy into the target cold storage amount calculation model to obtain the cold storage amount of the phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model.

[0174] The analysis module 803 is configured to perform analysis and processing based on the cold storage amount, the photovoltaic power generation power, and the pre-acquired planning electricity consumption probability distribution, to obtain an energy management strategy of the air conditioning system. The energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use the phase change cold storage material for cold storage, and whether to enable the compressor to assist in performing the cold storage process.

[0175] In one embodiment, the analysis module 803 is specifically configured to determine an optimization model according to the planning electricity consumption probability distribution, the photovoltaic power generation power, and the cold storage amount; acquire electricity price information corresponding to a preset period; the electricity price information includes a mains unit price of each time period and a time-of-use electricity price fluctuation range; acquire a risk control parameter in an air conditioning system running process, the risk control parameter includes a photovoltaic output uncertainty tolerance threshold, a load fluctuation range, and an indoor temperature control stability index; substitute the electricity price information and the risk control parameter into the optimization model for solving processing to obtain an optimal solution, and determine the optimal solution as the energy management strategy.

[0176] In one embodiment, the analysis module 803 is specifically configured to perform normalization processing on the photovoltaic power generation power and the cold storage amount based on the planning electricity consumption probability distribution, and construct a decision variable matrix according to a result of the normalization processing.

[0177] In one embodiment, the analysis module 803 is specifically configured to determine an optimization variable set based on the decision variable matrix, the optimization variable set at least including a cold storage power proportion of the phase change cold storage material, a utilization proportion of the photovoltaic power generation power, a power supply mode switching time point of the air conditioner, a start-stop state of the compressor, and a running time length; construct a target function based on the optimization variable set, the target function being used to represent a running cost of the air conditioning system in the preset period, the running cost including a power cost generated by mains power supply, an energy consumption cost generated by compressor running, and a cost generated by use of electric energy in the cold storage process; combine the target function and a running constraint condition to construct an optimization model with the running cost minimization of the air conditioning system as the target function, the constraint condition at least including that the photovoltaic power generation power does not exceed the maximum output capacity, the cold storage amount of the phase change cold storage material is within a rated range, the electricity consumption load meets the planning electricity consumption demand, the switching of the air conditioning system running state meets the minimum time interval requirement, and the indoor temperature is kept within a preset temperature interval.

[0178] In one embodiment, the analysis module 803 is specifically configured to input the electricity price information and the risk control parameter into the optimization model, and combine the optimization variable set and the target function; under the constraint of the running cost minimization target, a plurality of optimization variable sets are obtained by solving, and an optimal solution meeting all constraint conditions is determined according to the plurality of optimization variable sets.

[0179] In an embodiment, the apparatus further includes:

[0180] The first input module is configured to input the environmental data into the pre-constructed photovoltaic power generation amount prediction model to obtain the photovoltaic power generation power output by the photovoltaic power generation amount prediction model.

[0181] In an embodiment, the apparatus further includes:

[0182] The second input module is configured to input the historical air conditioner usage data into the pre-constructed behavior prediction model to obtain the planning electricity consumption probability distribution output by the behavior prediction model.

[0183] In an embodiment, the apparatus further includes:

[0184] The obtaining module is configured to obtain the environmental temperature disturbance and the photovoltaic output deviation.

[0185] The updating module is configured to update the model parameters of each candidate cold storage amount calculation model by using a rolling horizon control method in a case where the environmental temperature disturbance and / or the photovoltaic output deviation does not meet a preset condition.

[0186] The above various modules in the air conditioner system-based energy management apparatus can be realized by software, hardware, or a combination thereof, in whole or in part. The above various modules can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform the operations corresponding to the above various modules.

[0187] In an embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above various method embodiments when executing the computer program.

[0188] In an embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program implements the steps in the above various method embodiments when executed by a processor.

[0189] In an embodiment, a computer program product is provided, including a computer program, and the computer program implements the steps in the above various method embodiments when executed by a processor.

[0190] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use, and processing of related data need to comply with relevant laws, regulations, and standards of relevant countries and regions.

[0191] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0192] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0193] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. An energy management method based on an air conditioning system, characterized by, The method comprises: In a preset period, a target cold storage amount calculation model matching a current working condition is selected from a plurality of candidate cold storage amount calculation models according to a previously acquired variation trend of photovoltaic power generation, air conditioning system load and environmental temperature; The environmental temperature and electric energy are input into the target cold storage amount calculation model to obtain a cold storage amount of a phase change cold storage material in the air conditioning system output by the target cold storage amount calculation model; Based on the cold storage amount, the photovoltaic power generation and a previously acquired planning power consumption probability distribution, an energy management strategy of the air conditioning system is obtained through analysis and processing; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to utilize the phase change cold storage material for cold storage, and whether to enable a compressor to assist in performing a cold storage process.

2. The method of claim 1, wherein, The energy management strategy of the air conditioning system is obtained through analysis and processing based on the cold storage amount, the photovoltaic power generation and the planning power consumption probability distribution, and comprises: An optimization model is determined according to the planning power consumption probability distribution, the photovoltaic power generation and the cold storage amount; Electricity price information corresponding to the preset period is acquired; wherein the electricity price information comprises a mains unit price and a time-of-use electricity price fluctuation range of each time period; A risk control parameter in the running process of the air conditioning system is acquired, and the risk control parameter comprises a photovoltaic output uncertainty tolerance threshold, a load fluctuation range and an indoor temperature control stability index; The electricity price information and the risk control parameter are substituted into the optimization model for solving processing to obtain an optimal solution, and the optimal solution is determined as the energy management strategy.

3. The method of claim 2, wherein, The optimization model is determined according to the planning power consumption probability distribution, the photovoltaic power generation and the cold storage amount, and comprises: Based on the planning power consumption probability distribution, the photovoltaic power generation and the cold storage amount are normalized, and a decision variable matrix is constructed according to the results of the normalization processing; An optimization model with the running cost minimization of the air conditioning system as an objective function is constructed based on the decision variable matrix.

4. The method of claim 3, wherein, The optimization model with the running cost minimization of the air conditioning system as an objective function is constructed based on the decision variable matrix, and comprises: An optimization variable set is determined based on the decision variable matrix, and the optimization variable set at least comprises a cold storage power proportion of the phase change cold storage material, a utilization proportion of the photovoltaic power generation, an air conditioner power supply mode switching time point, a start-stop state and a running time length of the compressor; A target function is constructed based on the optimization variable set, and the target function is used to represent a running cost of the air conditioning system in the preset period, and the running cost comprises an electricity fee generated by mains power supply, an energy consumption cost generated by compressor running and a cost generated by electric energy use in the cold storage process; The target function is combined with operation constraints to build an optimization model with the minimum operation cost of the air conditioning system as the target function, and the constraints at least include that the photovoltaic power generation power does not exceed the maximum output capacity, the cold storage capacity of the phase change cold storage material is within a rated range, the power load meets the planned power consumption demand, the switching of the air conditioning system operation state meets the minimum time interval requirement, and the indoor temperature is kept within a preset temperature interval.

5. The method of claim 4, wherein, The solving processing of the optimization model by substituting the electricity price information and the risk control parameter into the optimization model includes: The electricity price information and the risk control parameter are input into the optimization model, and the optimization variable set and the target function are combined; Under the constraint of the minimum operation cost target, a plurality of optimization variable sets are solved, and the optimal solution meeting all the constraints is determined according to the plurality of optimization variable sets.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: The environmental data is input into a pre-constructed photovoltaic power generation prediction model to obtain the photovoltaic power generation power output by the photovoltaic power generation prediction model.

7. The method according to any one of claims 1 to 5, characterized in that, The method further includes: The historical air conditioning use data is input into a pre-constructed behavior prediction model to obtain the planned power consumption probability distribution output by the behavior prediction model.

8. The method according to any one of claims 1 to 5, characterized in that, The method further includes: The environmental temperature disturbance and the photovoltaic output deviation are acquired in real time; In the case that the environmental temperature disturbance and / or the photovoltaic output deviation does not meet the preset condition, the model parameters of each candidate cold storage capacity calculation model are updated by using a rolling horizon control method.

9. An energy management device based on an air conditioning system, characterized by, The device includes: A matching module is configured to select, in a preset period, a target cold storage capacity calculation model matching a current working condition from a plurality of candidate cold storage capacity calculation models according to pre-acquired variation trends of photovoltaic power generation power, air conditioning system load and environmental temperature; An output module is configured to input the environmental temperature and electrical energy into the target cold storage capacity calculation model to obtain the cold storage capacity of the phase change cold storage material in the air conditioning system output by the target cold storage capacity calculation model; An analysis module is configured to perform analysis processing based on the cold storage capacity, the photovoltaic power generation power and the pre-acquired planned power consumption probability distribution to obtain an energy management strategy of the air conditioning system; the energy management strategy is used to guide the air conditioning system to adopt mains power supply or photovoltaic power supply, whether to use the phase change cold storage material for cold storage, and whether to enable a compressor to assist in performing a cold storage process. 10.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-9. The processor executes the computer program to realize the steps of the method in any one of claims 1 to 8.

11. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 8.

12. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 8.

Images