Photovoltaic air conditioner energy consumption optimization method

Abstract

The invention discloses a photovoltaic air conditioner energy consumption optimization method. Based on a system architecture composed of a photovoltaic array, an energy storage unit, an air conditioning unit, a data acquisition module, a central control module and a mains supply complementary module, the method is optimized through the following steps that firstly, the data acquisition module acquires multi-dimensional data according to a fixed frequency and preprocesses the multi-dimensional data to ensure accuracy; secondly, an LSTM algorithm is adopted to predict photovoltaic output in the next 24 hours, and a GBDT algorithm is adopted to predict the air conditioner load in the same period; thirdly, an intelligent scheduling strategy of photovoltaic priority, photovoltaic-energy storage cooperation or commercial power supplementation is executed according to a prediction result; meanwhile, air conditioner operation parameters are dynamically adjusted in combination with a human body comfort model; and finally, the prediction model and the energy storage control logic are adaptively corrected through real-time state feedback. According to the method, the photovoltaic energy utilization rate can be increased by 15%-20%, the commercial power dependence degree can be reduced by 30%-40%, the user comfort is considered while stable operation of the air conditioner is guaranteed, and remarkable energy-saving benefits and practicability are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of photovoltaic application and air conditioner energy consumption control technology, in particular to a photovoltaic air conditioner energy consumption optimization method based on combination of photovoltaic power supply, energy storage regulation and intelligent scheduling. BACKGROUND

[0002] With the global energy crisis and the improvement of environmental protection awareness, photovoltaic as a clean energy is widely used. The photovoltaic air conditioner system can effectively reduce the consumption of traditional power supply by using solar energy to power the air conditioner. However, the existing photovoltaic air conditioner system has many shortcomings in energy consumption optimization. Firstly, the photovoltaic output is strongly intermittent and unstable due to the influence of natural factors such as day and night, season, weather, etc. When the photovoltaic power generation is insufficient, it needs to rely on the power supply, and when the power generation is excessive, it cannot be effectively stored, resulting in energy waste. Secondly, the air conditioner running energy consumption is strongly related to the indoor and outdoor environment and user usage habits. The existing system mostly uses fixed operation mode and does not dynamically adjust the air conditioner running parameters according to the photovoltaic output, so it cannot realize the dynamic matching of "photovoltaic output-air conditioner energy consumption". Thirdly, although some systems are equipped with energy storage units, they lack prediction-based charging and discharging scheduling strategies, and the utilization rate of energy storage devices is low, which further limits the energy consumption optimization effect. Therefore, it is urgent to develop a comprehensive energy consumption optimization method that can realize photovoltaic output prediction, air conditioner running intelligent regulation and efficient scheduling of energy storage, so as to improve the energy utilization efficiency of photovoltaic air conditioner system and reduce the dependence on power supply. SUMMARY

[0003] The purpose of the present application is to overcome the defects of low energy utilization efficiency, poor matching degree of photovoltaic and air conditioner load, and unreasonable energy storage scheduling of the existing photovoltaic air conditioner system, and to provide a photovoltaic air conditioner energy consumption optimization method. Through multi-dimensional data acquisition, intelligent prediction and dynamic scheduling, the maximum utilization of photovoltaic energy, the precise regulation and control of air conditioner energy consumption and the efficient allocation of energy storage resources are realized, so as to finally reduce the overall energy consumption and power consumption of the system.

[0004] To achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] A photovoltaic air conditioner energy consumption optimization method applied to a photovoltaic air conditioner system comprising a photovoltaic array, an energy storage unit, an air conditioner unit, a data acquisition module, a central control module and a power supply complementary module, the method comprising the following steps:

[0006] (I) System architecture building

[0007] 1. Photovoltaic array: adopt polycrystalline silicon photovoltaic components, form an array through series and parallel connection, and convert solar energy into direct current; configure a maximum power point tracking (MPPT) controller to track the maximum power output of the photovoltaic array in real time and improve the photovoltaic conversion efficiency.

[0008] 2. Energy storage unit: composed of lithium iron phosphate battery pack, charge and discharge controller, and battery management system (BMS); BMS monitors battery voltage, current, temperature, and remaining capacity (SOC) in real time to ensure safe and stable operation of the battery.

[0009] 3. Air conditioning unit: uses variable frequency air conditioner, equipped with temperature sensor, humidity sensor, and power sensor; can dynamically adjust compressor speed and fan power according to indoor environmental parameters and photovoltaic output.

[0010] 4. Data acquisition module: real-time acquisition of photovoltaic array output power, energy storage unit SOC, indoor and outdoor temperature, humidity, air conditioning operating power, and power supply parameters, and data transmission to the central control module.

[0011] 5. Central control module: embedded microprocessor as the core, integrated data processing, prediction algorithm, scheduling strategy, and communication module; through the communication module to realize information interaction with each component, based on the collected data to execute prediction and scheduling instructions.

[0012] 6. City power complementary module: contains a bidirectional inverter, when photovoltaic output is insufficient and energy storage unit SOC is below the threshold, access to city power to supply air conditioning; when photovoltaic output is excessive, through the inverter to feed the excess power into the grid (if it has grid-connected conditions).

[0013] (II) Energy consumption optimization process

[0014] 1. Multi-dimensional data acquisition and preprocessing

[0015] The data acquisition module collects photovoltaic array output power, energy storage unit SOC, voltage, temperature, indoor and outdoor temperature, humidity, air conditioning real-time power, and city power price (in time-of-use pricing scenarios) at a frequency of 1 minute / second.

[0016] The central control module preprocesses the collected data, removes noise interference through Kalman filtering algorithm, and uses linear interpolation method to complete the missing data, ensuring the accuracy and integrity of the data.

[0017] 2. Photovoltaic output and air conditioning load prediction

[0018] Photovoltaic output prediction: the central control module based on historical photovoltaic output data, real-time weather data (light intensity, cloud coverage, wind speed, etc.), and 24-hour weather forecast, uses long short-term memory network (LSTM) algorithm to build a prediction model, outputs the photovoltaic output prediction value every 15 minutes in the next 24 hours.

[0019] Air conditioning load prediction: Combined with historical air conditioning operation data, indoor and outdoor temperature difference, user set temperature and work and rest habits, the gradient boosting decision tree (GBDT) algorithm is used to construct an air conditioning load prediction model, and the air conditioning load prediction value every 15 minutes in the next 24 hours is output.

[0020] 3. Intelligent scheduling strategy execution based on prediction

[0021] Photovoltaic priority utilization mode: When the photovoltaic output prediction value ≥ the air conditioning load prediction value, the central control module controls the photovoltaic array to directly power the air conditioning unit, and the excess photovoltaic power is charged to the energy storage unit through the charge and discharge controller until the SOC of the energy storage unit reaches 100%; if there is still surplus power and the grid-connected condition is met, the power is fed into the grid through the power supply complementary module.

[0022] Photovoltaic- energy storage collaborative mode: When the photovoltaic output prediction value < the air conditioning load prediction value, but the photovoltaic output prediction value + the energy storage unit dischargeable capacity ≥ the air conditioning load prediction value, the central control module controls the photovoltaic array and the energy storage unit to jointly power the air conditioning; at the same time, the discharge power is dynamically adjusted according to the SOC of the energy storage unit to ensure that the SOC is not lower than 20% (protection threshold).

[0023] Power supply complementary mode: When the photovoltaic output prediction value + the energy storage unit dischargeable capacity < the air conditioning load prediction value, the central control module controls the photovoltaic and energy storage to supply power preferentially, and the insufficient part is supplemented by the power supply complementary module; in the time-of-use electricity price scenario, the power supply complementary module is preferentially used to supplement the energy storage unit with power at the low electricity price period to reduce the electricity cost.

[0024] 4. Dynamic adjustment of air conditioning operation parameters

[0025] The central control module dynamically adjusts the air conditioning operation parameters according to the real-time photovoltaic output, energy storage SOC and indoor environmental parameters: when the photovoltaic output is sufficient, the air conditioning set temperature is adjusted to the lower limit of the user comfort interval (in summer) or the upper limit (in winter), and the fan speed is increased to improve the refrigeration / heating efficiency; when the photovoltaic output is insufficient, the compressor speed and fan power are reduced to reduce the air conditioning energy consumption under the premise of ensuring the basic comfort of the user (temperature fluctuation not exceeding ±1℃).

[0026] The human comfort model is introduced, and the temperature, humidity and wind speed parameters are combined to calculate the human comfort index (PMV) to ensure that the air conditioning adjustment meets the energy consumption optimization demand and does not affect the user experience.

[0027] 5. System operation state feedback and adaptive adjustment

[0028] The central control module compares the deviation between the actual value and the predicted value of the photovoltaic output and the actual value and the predicted value of the air conditioning load in real time, and when the deviation exceeds 10%, the prediction model parameters are automatically corrected to improve the prediction accuracy.

[0029] The battery management system monitors the SOC of the energy storage unit in real time. When the SOC is lower than 15%, the BMS triggers a low power alarm, and the central control module preferentially limits the unnecessary energy consumption of the air conditioner. When the SOC is higher than 90%, the charging is stopped to avoid damage to the battery caused by overcharging.

[0030] Compared with the prior art, the present application has the following beneficial effects:

[0031] 1. Improve the utilization rate of photovoltaic energy: through MPPT control and photovoltaic output prediction, realize the maximum power output of photovoltaic array and preferential utilization, the utilization rate of photovoltaic energy is improved by 15%-20% compared with traditional system.

[0032] 2. Reduce the consumption of commercial power: through photovoltaic-energy storage collaborative scheduling and dynamic adjustment of air conditioner parameters, the dependence of the system on commercial power is reduced by 30%-40%, significantly reducing the operation cost.

[0033] 3. Ensure the stability of the system: based on real-time data feedback and adaptive correction of the prediction model, effectively respond to photovoltaic output fluctuation, ensure the stable operation of the air conditioner; the addition of BMS improves the safety and service life of the energy storage unit.

[0034] 4. Consider user comfort: adjust the air conditioner parameters combined with the human comfort model, optimize energy consumption while ensuring user experience, and have strong practicality. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a system module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0036] Figure 2 It is a photovoltaic array module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0037] Figure 3 It is a storage unit module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0038] Figure 4 It is a data acquisition module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0039] Figure 5 It is a central control module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0040] Figure 6 It is an air conditioner unit module architecture diagram of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0041] Figure 7 It is a total flow chart of the core link of the photovoltaic air conditioner energy consumption optimization method of the present application;

[0042] Figure 8 Data layer flowchart for the photovoltaic air conditioner energy consumption optimization method of the present application;

[0043] Figure 9 Prediction layer flowchart for the photovoltaic air conditioner energy consumption optimization method of the present application;

[0044] Figure 10 Scheduling layer flowchart for the photovoltaic air conditioner energy consumption optimization method of the present application;

[0045] Figure 11 Execution layer flowchart for the photovoltaic air conditioner energy consumption optimization method of the present application;

[0046] Figure 12 Feedback layer flowchart for the photovoltaic air conditioner energy consumption optimization method of the present application. DETAILED DESCRIPTION

[0047] In order to make the technical means, creative features, purposes and effects achieved by the present application easy to understand, the present application will be further described below in conjunction with specific embodiments.

[0048] I. System construction

[0049] 1. The photovoltaic array is selected from 20 pieces of 250W polycrystalline silicon photovoltaic components, which are connected in series to form 4 strings, each string has 5 pieces, and 1 piece of 5kW MPPT controller is configured.

[0050] 2. The energy storage unit adopts 10kWh lithium iron phosphate battery pack, equipped with 1 piece of 3kW bidirectional charge and discharge controller and BMS, and the SOC protection threshold is set to 15%-90%.

[0051] 3. The air conditioning unit is selected from 1.5 variable frequency air conditioners, equipped with DS18B20 temperature sensor, SHT30 humidity sensor and SCT013 power sensor.

[0052] 4. The central control module adopts STM32F407 microprocessor, integrates 4G communication module, communicates with each component through MQTT protocol, and adopts Kalman filter algorithm for data preprocessing, and builds LSTM and GBDT models based on TensorFlow framework.

[0053] 5. The utility complementary module is selected from 3kW bidirectional inverter, which supports grid-connected and off-grid switching.

[0054] II. Optimization process execution

[0055] 1. The data acquisition module acquires data every 1 minute, including photovoltaic output power, energy storage SOC, indoor and outdoor temperature, humidity, air conditioning power and utility time-of-use electricity price (peak segment 08:00-22:00, valley segment 22:00-Next day 08:00).

[0056] 2. Central control module predicts photovoltaic output of next day and air conditioning load of next day through LSTM model and GBDT model respectively based on historical data and weather forecast of next day at 20:00 every day.

[0057] 3. Dispatching strategy execution:

[0058] On sunny day in summer from 10:00 to 14:00, photovoltaic output is sufficient (predicted value 3kW ≥ air conditioning load predicted value 1.2kW), photovoltaic directly supplies power for air conditioner, and the excess power charges the energy storage until the SOC reaches 90%, and the remaining power is fed into the power grid.

[0059] On cloudy day from 9:00 to 10:00, photovoltaic output predicted value 0.8kW < air conditioning load predicted value 1.2kW, but 0.8kW + energy storage dischargeable amount 0.5kW ≥ 1.2kW, photovoltaic and energy storage jointly supply power, and the energy storage discharge power is controlled at 0.4kW, maintaining SOC ≥ 20%.

[0060] At night from 23:00 to 6:00 of next day, there is no photovoltaic output, and the energy storage SOC is lower than 20%, the city power is connected to supply power for air conditioner (at this time, it is the valley section of electricity price), and the energy storage is charged to SOC 50%.

[0061] 4. Air conditioning parameter adjustment: when photovoltaic is sufficient, the set temperature is 25℃ in summer and the fan speed is high; when photovoltaic is insufficient, the set temperature is adjusted to 26℃ and the fan speed is medium, and the PMV index is maintained between -0.5 and 0.5 (comfort interval).

[0062] 5. When the actual value of photovoltaic output deviates from the predicted value by more than 10%, the central control module automatically updates the training samples of LSTM model and corrects the model parameters; when the battery temperature monitored by BMS exceeds 45℃, the heat dissipation device is triggered, and the charging and discharging power is reduced.

[0063] III. Implementation effect

[0064] After 3 months of actual operation test, the photovoltaic energy utilization rate of the photovoltaic air conditioning system reaches 82%, which is increased by 18% compared with the traditional photovoltaic air conditioning system; the power consumption is reduced by 35% compared with the traditional photovoltaic air conditioning, and about 80 yuan of electricity fee is saved per month; the air conditioner runs stably, the indoor temperature fluctuation is ≤0.8℃, and the user comfort is good; the energy storage unit runs safely, there is no overcharge and overdischarge phenomenon, and the battery capacity attenuation rate is lower than 2% / year.

[0065] The above shows and describes the basic principles and main features of the present application and the advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A photovoltaic air conditioner energy consumption optimization method applied to a photovoltaic air conditioner system, characterized in that, The photovoltaic air conditioning system comprises a photovoltaic array, an energy storage unit, an air conditioning unit, a data acquisition module, a central control module and a mains power complementary module; the method comprises the following steps: Step S1: data acquisition and preprocessing: the data acquisition module acquires the photovoltaic array output power, the energy storage unit SOC, the indoor and outdoor temperature, humidity, air conditioning operating power and mains power parameters in real time, and transmits them to the central control module after preprocessing; Step S2: prediction model construction and execution: the central control module predicts the photovoltaic output by using the LSTM algorithm and predicts the air conditioning load by using the GBDT algorithm based on historical data and real-time weather information; Step S3: intelligent scheduling strategy execution: according to the photovoltaic output and air conditioning load prediction values, the photovoltaic priority utilization, photovoltaic-energy storage coordination or mains power complementary mode is executed to realize energy distribution; Step S4: dynamic adjustment of air conditioning parameters: the air conditioning set temperature, compressor speed and fan power are dynamically adjusted in combination with the real-time energy state and the human comfort model; Step S5: state feedback and adaptive adjustment: the prediction model parameters are corrected by comparing the deviation between the prediction value and the actual value; the energy storage state is monitored by the BMS to ensure safe operation of the system.

2. The method of claim 1, wherein, The photovoltaic array is configured with an MPPT controller to track the maximum power output in real time; the energy storage unit comprises a lithium iron phosphate battery pack, a charge-discharge controller and a BMS, the BMS monitors the battery voltage, current, temperature and SOC, and sets the SOC protection threshold to 15%-90%.

3. The method of claim 1, wherein, In step S1, the Kalman filtering algorithm is used for noise removal and the linear interpolation method is used for missing data completion; the data acquisition frequency is 1 minute / time.

4. The method of claim 1, wherein, In step S3, in the time-of-use electricity price scenario, the mains power is used to supplement the energy storage unit during the low electricity price period.

5. The method of claim 1, wherein, In step S4, the human comfort model is constructed based on the PMV index, and the PMV index is controlled to be between-0.5 and 0.5.

Images