A control method for a photovoltaic direct-driven air source heat pump unit

Abstract

The present invention relates to the technical field of heat pump unit control, and in particular to a control method for a photovoltaic direct-drive air-source heat pump unit. The method comprises the following steps: collecting historical environmental data of the heat pump unit at a sampling frequency, performing trend identification based on the historical environmental data, obtaining a predicted value of the environmental data at the current moment, and then obtaining a trend value of the power generation at the current moment; obtaining an initial trend identification accuracy and confidence level at the next moment based on the actual power generation value and the trend value at the current moment, and obtaining a target trend identification accuracy at the next moment based on the accuracy and confidence level; sequentially adjusting the initial trend identification window, and finally controlling the unit using an MPC algorithm based on the adjusted trend identification window. The present invention calculates the trend identification accuracy and confidence level based on the difference between the trend identification result and the actual data, and adjusts the trend identification window based on the confidence level and accuracy level to obtain an accurate trend identification result.

Description

Technical Field

[0001] The present invention relates to the technical field of heat pump unit control, and in particular to a control method for a photovoltaic direct-driven air-source heat pump unit. Background Art

[0002] Against the backdrop of the global energy structure accelerating its transition to a low-carbon one, severe challenges are being posed to traditional energy utilization models. As an innovative system for the coordinated utilization of "light-heat-electricity", photovoltaic direct-drive air-source heat pump units, with their unique advantages, have become the core technology path for achieving low-carbon transformation in areas such as building heating and industrial heating. This system uses photovoltaic modules to directly convert solar energy into electricity to drive the heat pump, effectively avoiding the multi-stage conversion losses of traditional grid power supply. In theory, it can achieve near-zero-carbon heating, combining environmental value with economic potential. However, with the increasing complexity of the system's operating scenarios, how to achieve precise control to improve operational efficiency has become a key issue restricting its large-scale promotion.

[0003] Currently, utilizing data trend identification technology to optimize model predictions is a key area for improving the control accuracy of photovoltaic heat pump units. Existing technologies often employ MPC models for predictive control of unit data. In this process, trend identification methods with fixed window sizes exhibit significant flaws. On the one hand, fixed windows struggle to adapt to the complex and changing trend characteristics of data sequences, easily missing subtle changes in areas with less obvious trends, resulting in the loss of critical information and an inability to fully capture data dynamics. On the other hand, in areas with significant fluctuations, fixed windows can overemphasize local trend characteristics, leading to biased and misinterpreted trend judgments. This can reduce the accuracy of the MPC prediction model and make it difficult for the system to respond promptly and effectively to complex operating conditions such as sudden changes in light intensity and drastic fluctuations in ambient temperature. This, in turn, can impact the economic efficiency, reliability, and full realization of the low-carbon benefits of unit operation. Summary of the Invention

[0004] In order to solve the technical problem that the trend identification method using a fixed window size affects the accuracy of the MPC prediction model, the present invention provides a control method for a photovoltaic direct-driven air source heat pump unit; the method comprises the steps of: collecting the historical actual value of the heat pump unit's power generation according to a preset sampling frequency, and historical environmental data including multiple environmental parameters such as ambient temperature and wind speed; setting an initial trend identification window, performing trend identification based on the historical environmental data, and obtaining the predicted value of the environmental data at the current moment; calculating the power generation trend value at the current moment through a photovoltaic power generation model based on the predicted value of the environmental data at the current moment; obtaining the initial trend identification accuracy and confidence of the power generation at the next moment based on the actual power generation value and the power generation trend value at the current moment, and obtaining the target trend identification accuracy of the power generation at the next moment based on the initial trend identification accuracy and confidence; adjusting the initial trend identification window according to the target trend identification accuracy to obtain a target trend identification window; and controlling the heat pump unit using an MPC algorithm according to the target trend identification window.

[0005] The present invention combines the dynamic adjustment of the trend identification window with the MPC algorithm to achieve precise control of the photovoltaic direct-drive heat pump unit. Specifically, after obtaining the power generation trend value based on the sampled data, the initial trend identification accuracy and confidence are calculated by the difference between the actual value and the trend value, and then the target trend identification accuracy is obtained, thereby dynamically optimizing the trend identification window size. This mechanism can adapt to fluctuations in environmental data, expand the window to capture stable trends when the data changes drastically, and reduce the window to improve computing efficiency when the data is stable, avoiding the limitations of fixed windows. Combined with the rolling optimization characteristics of the MPC algorithm, it can respond to changes in photovoltaic power and environmental parameters in real time, accurately adjust the heat pump operation strategy, improve the synergistic efficiency of "light-heat-electricity", reduce equipment loss and energy consumption caused by prediction deviations, enhance the system's adaptability to complex working conditions, and achieve efficient and low-carbon operation.

[0006] Preferably, the step of obtaining the initial trend recognition accuracy and confidence of the power generation at the next moment based on the actual power generation value and the power generation trend value at the current moment includes: obtaining the current moment and the previous moment; The actual value of the generated power at a moment is recorded as the generated power neighborhood data; based on the relative deviation between the actual value of the generated power and the generated power trend value of the generated power neighborhood data, and the degree of fluctuation of the actual value of the generated power in the generated power neighborhood data, the initial trend recognition accuracy of the generated power at the next moment is obtained; based on the actual value of the generated power at the next moment and the generated power neighborhood data, the confidence level of the initial trend recognition accuracy of the next moment is obtained.

[0007] This method quantifies initial accuracy based on the difference between actual power generation and trend values, and corrects for interference errors by combining the fluctuation of neighboring data. This effectively filters the impact of short-term outliers on trend identification. Furthermore, it dynamically assesses the stability of data sequences by deriving confidence from the correlation of actual neighboring values. This mechanism enables the system to automatically downweight unreliable trend identification results when the environment suddenly changes, thus avoiding misjudgments. It also enhances the effectiveness of trend predictions during periods of data stability, helping the MPC algorithm to more accurately adjust heat pump operating parameters.

[0008] Preferably, the acquisition of the initial trend recognition accuracy at the next moment includes: recording the next moment as the first At a certain moment, the initial trend recognition accuracy at the next moment satisfies the relationship: ;in, It is The accuracy of identifying the initial trend of power generation at a given moment, is the natural exponential function, It is The index of the power generation trend value at the moment is the same as the The difference between the power generation neighborhood data indexes, It is The power generation at the moment The relative deviation between the actual value of the power generation and the trend value of the power generation of the power neighborhood data is It is the influencing factor of the fluctuation degree of the actual value of the power generation in the power generation neighborhood data. is the length of the preset trend identification window, The power generation at the moment The power generation neighborhood data is the data in its preset trend identification window.

[0009] The present invention uses the natural exponential function to associate the neighborhood data index difference with the actual trend difference, thereby mining the local fluctuation characteristics of the power sequence and having strong noise resistance. It also incorporates the factors affecting the fluctuation of the actual value of the generated power to adapt to environmental changes and enhance robustness. It also combines the preset window length and integrates the local and global trends to provide a reliable basis for the MPC algorithm to control the unit, thereby improving the system energy efficiency and stability.

[0010] Preferably, the The power generation at the moment The relative deviation between the actual power generation value and the power generation trend value of the power generation neighborhood data satisfies the relationship: ;in, It is The power generation trend value at the moment The power generation trend value of the power generation neighborhood data, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is an absolute value.

[0011] Preferably, the influencing factor of the fluctuation degree of the actual value of the generated power in the generated power neighborhood data satisfies the relationship: ;in, It is The first environmental parameter The standard deviation of the power generation neighborhood data of the actual power generation value, It is The first environmental parameter The average value of the power generation neighborhood data of the actual power generation value, is a small value used to prevent the denominator from being zero. is the total number of environmental parameters in each environmental data.

[0012] Preferably, the confidence level of the initial trend recognition accuracy at the next moment satisfies the relationship: ;in, It is The confidence level of the accuracy of the initial trend identification at the moment, is the length of the preset trend identification window, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is The actual value of generated power at a moment, is the range of the actual value of the generated power at multiple moments, is a natural exponential function.

[0013] The present invention builds a model based on the deviation between the actual power value of the neighborhood and the actual value at the current moment, combined with the natural exponential function. Associated neighborhood data to determine the actual value of power generation Normalized deviation accurately quantifies the confidence level of initial trend identification accuracy. It not only adapts to data fluctuations but also strengthens the impact of deviation on confidence through exponential calculations. This provides a reliable confidence assessment for subsequent trend identification window adjustments and unit control optimization, improving the system's adaptability and control accuracy to complex operating conditions.

[0014] Preferably, the target trend recognition accuracy is the product of the initial trend recognition accuracy of the power generation at the next moment and its confidence.

[0015] Preferably, the setting of the initial trend identification window, performing trend identification based on the historical environmental data, and obtaining the environmental data prediction value at the current moment include: obtaining multiple environmental parameters in the historical environmental data, forming time series data for each environmental parameter, and setting the initial trend identification window for each data point in the time series data corresponding to each environmental parameter. The data points are recorded as environmental parameter neighborhood data; curve fitting is performed according to the environmental parameters at each moment in the environmental parameter neighborhood data to obtain the fitting slope at the current moment, and the environmental parameters at the current moment are predicted according to the fitting slope to obtain the environmental data prediction value at the current moment.

[0016] Preferably, the predicted value of the environmental data at the current moment is based on the photovoltaic power generation power model to calculate the power generation power trend value at the current moment, including: constructing a photovoltaic power generation power prediction model based on the heat pump COP function and the environmental data at the current moment to obtain the power generation power trend value at the current moment.

[0017] Preferably, the target trend identification window satisfies the relationship: ;in, is the length of the target trend identification window, is the length of the preset trend identification window, It is The target trend recognition accuracy of power generation at each moment, is the maximum function, yes decimal, used to limit The minimum value of .

[0018] The present invention is based on the preset window length and dynamically adjusts the window according to the target trend recognition accuracy. Limits prevent excessive window shrinkage, ensure data reliability, and allow trend identification windows to adapt to changes in working conditions.

[0019] Beneficial effects of the present invention: The present invention realizes precise control of photovoltaic direct-drive heat pump units through a multi-dimensional strategy. The initial trend identification accuracy and confidence are calculated by combining the difference between the actual value of generated power and the trend value of generated power, and then the target accuracy is derived to dynamically optimize the trend identification window, thereby obtaining more accurate trend identification results. The window can adapt to fluctuations in environmental data, expand when the data changes drastically to capture stable trends, and shrink when it is stable to improve calculation efficiency, breaking through the limitations of fixed windows. With the help of the natural exponential function to associate neighborhood data, the local fluctuation characteristics of the generated power sequence are mined to enhance noise resistance; the influencing factors of the fluctuation of the actual value of generated power are incorporated to adapt to environmental changes and enhance robustness. At the same time, based on the deviation between the neighborhood and the actual value of generated power at the current moment, a confidence model is constructed using the natural exponential function and range normalization to accurately quantify the credibility of the initial trend identification accuracy, providing a reliable basis for trend identification window adjustment and unit control optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a control method for a photovoltaic direct-drive air-source heat pump unit provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The embodiment of the present invention provides a control method for a photovoltaic direct-driven air source heat pump unit, such as Figure 1 As shown, the method includes steps S100 to S600:

[0022] Step S100: collecting historical actual power generation values ​​of the heat pump unit and historical environmental data including multiple environmental parameters such as ambient temperature and wind speed according to a preset sampling frequency.

[0023] It should be noted that during the operation of a PV-powered air-source heat pump system, the power generation efficiency of PV modules is affected by a variety of natural environmental factors, such as solar irradiance, ambient temperature, humidity, and wind speed. Solar irradiance determines the photovoltaic conversion efficiency of PV modules. In actual operation, its value fluctuates dramatically from early morning to midday and has a significant positive correlation with power generation efficiency. Ambient temperature, on the other hand, is negatively correlated with power generation efficiency. As temperature increases, the power generation efficiency of silicon-based PV modules generally decreases. Humidity has a dual impact on power generation efficiency. On the one hand, high humidity can cause condensation on the module surface, forming water stains that block light. On the other hand, humidity accelerates dust absorption on the module surface. When dust accumulation reduces light transmittance, power generation efficiency is significantly suppressed. Although wind speed does not directly participate in photovoltaic conversion, it can reduce module temperature through convection heat dissipation, thereby improving power generation efficiency. The real-time changes in these environmental data jointly influence the power generation efficiency of PV modules. Therefore, collecting and analyzing these data is key to optimizing the operation and control of PV-powered air-source heat pump systems.

[0024] Specifically, the sampling frequency is set to 10 minutes, and sampling points are set up. Temperature sensors, humidity sensors, irradiance sensors, wind speed sensors, and other equipment are installed at the sampling points to collect environmental data such as ambient temperature, humidity, solar radiation intensity, and wind speed. The current sampling moment is recorded as the current moment, and the moments before the current moment are recorded as historical moments. After collecting data in this way, the ambient temperature, humidity, solar radiation intensity, and wind speed data of several historical moments will be obtained, which is the historical environmental data.

[0025] In addition, it is necessary to set up sensors built into the PV combiner box or inverter to collect the actual power generation data of the unit. The sampling frequency of this data should be consistent with the environmental data, and this data should be recorded as the actual power generation value. Similarly, the actual power generation value at several historical moments can be obtained.

[0026] At this point, several historical actual values ​​of generated power and historical environmental data including multiple environmental parameters such as ambient temperature and wind speed are obtained. In addition, the actual value of generated power and environmental data at the current moment can also be obtained.

[0027] Step S200: Set an initial trend identification window, perform trend identification based on historical environmental data, and obtain a predicted value of the environmental data at the current moment.

[0028] It should be noted that in order to obtain the trend value of the generated power, it is necessary to first obtain the predicted values ​​of the environmental data, and then calculate the generated power trend value using the generated power prediction model based on the predicted values ​​of each environmental data.

[0029] For the predicted value of each environmental parameter, curve fitting or exponential smoothing can be used. Taking curve fitting as an example, curve fitting can effectively explore the potential nonlinear change rules in the time series. Compared with simple mean, difference and other methods, it can more accurately characterize the complex fluctuation trend of environmental variables. Trend prediction based on the fitting slope can dynamically reflect the rate and direction of change of data. When environmental conditions suddenly change, such as a sudden increase in wind speed or a sharp drop in temperature, the real-time update of the slope can quickly capture trend changes, allowing the prediction model to adapt to new working conditions in a timely manner, and improving the response speed and prediction accuracy to dynamic changes in the environment. Therefore, the present invention performs trend identification through curve fitting. According to trend prediction, more accurate prediction values ​​of various environmental parameters can be obtained. Combining trend prediction with photovoltaic power generation can obtain more accurate and long-term photovoltaic power generation prediction results.

[0030] Specifically, first, the length of the initial trend identification window is set to . Secondly, each environmental parameter of the historical environmental data obtained through the above steps constitutes a time series data sequence of the corresponding environmental parameter, which is arranged in order by timestamp. Then, the initial trend recognition window of each data point in the time series data corresponding to each environmental parameter corresponding to each historical sampling moment is obtained. Each data point is recorded as its environmental data neighborhood data. Next, a curve is fitted based on the environmental parameters at each moment in the environmental parameter neighborhood data to obtain the fitting slope at the current moment. Data prediction is then performed based on the fitting slope to obtain the trend value of the predicted value of each environmental parameter in the environmental data at the current moment. Curve fitting and exponential smoothing are existing technologies and will not be discussed in detail here. The length of the initial trend identification window can be set as required.

[0031] For example, change the length of the initial trend identification window to Set to 3, take the ambient temperature as an example to illustrate the acquisition of neighborhood data. Assume that the time series data of the ambient temperature prediction value obtained according to the above steps is , taking the third sampling time point as an example, its environmental data neighborhood data is .

[0032] At this point, the current environmental data forecast value is obtained

[0033] Step S300: Based on the environmental data prediction value at the current moment, the photovoltaic power generation power trend value at the current moment is calculated using the photovoltaic power generation power model.

[0034] It should be noted that the MPC algorithm is a control strategy that predicts the future behavior of the system based on current and historical data by establishing a mathematical model of the system. At each sampling moment, MPC calculates the control sequence for a period of time in the future according to the prediction model to minimize the objective function, and applies the first control quantity to the system. This process is repeated at the next moment to achieve rolling optimization. The COP function is a key indicator for measuring the energy conversion efficiency of a heat pump system. The higher the COP value, the more heat or energy the heat pump system can output while consuming the same amount of electrical energy, and the higher the energy efficiency. The COP function is usually related to factors such as ambient temperature and heat load demand. In the present invention, the current photovoltaic power generation efficiency data is predicted by the MPC prediction algorithm, and the difference in subsequent data is obtained based on the predicted value.

[0035] Specifically, a photovoltaic power generation prediction model is constructed based on the heat pump COP function and the predicted values ​​of historical environmental data and current environmental data to predict photovoltaic power generation. The predicted value of photovoltaic power generation at the current sampling moment is obtained and recorded as the power generation power trend value at the current moment.

[0036] At this point, the current power generation trend value is obtained.

[0037] Step S400: Based on the actual value of the generated power and the generated power trend value at the current moment, obtain the initial trend recognition accuracy and confidence of the generated power at the next moment, and obtain the target trend recognition accuracy of the generated power at the next moment based on the initial trend recognition accuracy and confidence.

[0038] It should be noted that, according to the above steps, a power generation trend value is obtained through curve fitting and trend prediction methods. This power generation trend value reflects the predicted power generation trend based on historical data and environmental variables, providing an important reference for system operation control and optimization. However, the accuracy of the power generation trend value is directly related to the effectiveness of subsequent control strategies, so its accuracy needs to be compared with the actual power generation.

[0039] Specifically, get the current time and its previous The actual value of the generated power at the moment is recorded as the generated power neighborhood data; the present invention evaluates the accuracy of the generated power trend value at the next moment based on the gap between the generated power trend value at the current moment and the actual value of the generated power. Since the data needs to be further adjusted later, the accuracy of the generated power trend value at the next moment obtained in this step is recorded as the initial trend recognition accuracy at the next moment. The next moment is recorded as the first At a certain moment, the initial trend recognition accuracy at the next moment satisfies the expression:

[0040] ;

[0041] in, It is The accuracy of identifying the initial trend of power generation at a given moment, is the natural exponential function, It is The index of the power generation trend value at the moment is the same as the The difference between the power generation neighborhood data indexes, It is The power generation at the moment The relative deviation between the actual value of the power generation and the trend value of the power generation of the power neighborhood data is It is the influencing factor of the fluctuation degree of the actual value of the power generation in the power generation neighborhood data. is the length of the preset trend identification window, The power generation at the moment The power generation neighborhood data is the data in its preset trend identification window.

[0042] In this formula It is The power generation at the moment The relative deviation between the actual power value of the power generation neighborhood data and the power generation trend value. The larger the relative deviation, the less accurate the trend identification result. The difference satisfies the relationship:

[0043] ;

[0044] in, It is The power generation at the moment The relative deviation between the actual value of the power generation and the trend value of the power generation of the power neighborhood data is It is The power generation trend value at the moment The power generation trend value of the power generation neighborhood data, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is an absolute value.

[0045] In this formula This is the influencing factor of the fluctuation degree of the actual value of the generated power. The larger the value, the greater the fluctuation degree of the data, and the greater its impact on trend identification, and the accuracy of trend identification will also be reduced accordingly. The influencing factor of the fluctuation degree of the actual value of the generated power satisfies the relationship:

[0046] ;

[0047] in, It is the influencing factor of the fluctuation degree of the actual value of the power generation in the power generation neighborhood data. It is The first environmental parameter The standard deviation of the power generation neighborhood data of the actual power generation value, It is The first environmental parameter The average value of the power generation neighborhood data of the actual power generation value, is a small value used to prevent the denominator from being zero. is the total number of environmental parameters in each environmental data is a natural exponential function.

[0048] The above describes how to calculate the accuracy of initial trend identification at the next moment, and the following describes how to calculate the confidence level of the accuracy of initial trend identification at the next moment.

[0049] It should be noted that for the initial trend recognition accuracy of any data at the next moment, if the trend change of the data itself is not obvious, the accuracy calculation of the trend recognition also needs to be revised.

[0050] Specifically, the confidence level of the initial trend recognition accuracy at the next moment satisfies the relationship:

[0051] ;

[0052] in, It is The confidence level of the accuracy of the initial trend identification at the moment, is the length of the preset trend identification window, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is The actual value of generated power at a moment, is the range of the actual value of the generated power at multiple moments, is a natural exponential function.

[0053] In this formula Part of it is the gap between the actual value of the power generation data and its neighboring data. The gap between the neighboring power generation data reflects its volatility, and the confidence of the trend identification accuracy is calculated. If the volatility of the power generation data is greater, it means that its trend is not obvious, and the confidence of its trend identification result is lower.

[0054] So far, the initial trend recognition accuracy and confidence level at the next moment have been obtained based on the above operations. Next, we will explain how to obtain the target trend recognition accuracy at the next moment based on these two data. The target trend recognition accuracy at the next moment satisfies the relationship:

[0055] ;

[0056] in, It is The target trend recognition accuracy of power generation at each moment, It is The accuracy of identifying the initial trend of power generation at each moment; It is The confidence level of the accuracy of the initial trend identification of the power generation at a certain moment.

[0057] In this formula, confidence reflects the degree of confidence in the accuracy of the initial trend identification. By multiplying the two, the final accuracy can be dynamically adjusted based on factors such as data stability and model credibility. For example, when environmental data fluctuates significantly or the model is in the parameter calibration phase, a lower confidence level will reduce the target accuracy, avoiding decision-making errors caused by over-reliance on unreliable predictions.

[0058] At this point, the target trend recognition accuracy at the next moment is obtained.

[0059] Step S500: Adjust the initial trend recognition window according to the target trend recognition accuracy to obtain the target trend recognition window.

[0060] It should be noted that the greater the trend recognition accuracy of the data, the more accurate the trend recognition. Trend recognition can be performed through a smaller window to reduce the amount of data calculation. For areas with lower trend recognition accuracy, trend recognition should be performed based on a larger window size to obtain more accurate trend recognition results.

[0061] Specifically, the ratio of the preset trend recognition window size and the adjusted trend recognition accuracy is used to obtain the adaptive trend recognition window size of each environmental parameter at the next moment, which is recorded as the target trend recognition window. The target trend recognition window satisfies the relationship:

[0062] ;

[0063] in, is the length of the target trend identification window, is the length of the preset trend identification window, It is The target trend recognition accuracy at each moment, is the maximum function, yes A decimal number used to limit The minimum value of .

[0064] In this formula, by The minimum value is limited to a decimal between 0 and 1, such as 0.5, to ensure that the trend identification window will not be too large and distorted. In actual use, it can be set according to the actual implementation situation. .

[0065] At this point, the target trend recognition window for each environmental parameter at the next moment is obtained.

[0066] Step S600: Control the heat pump unit using the MPC algorithm according to the target trend identification window.

[0067] The target trend identification window obtained through the above steps is no longer a fixed-size window, which solves the technical problem that the fixed-window trend identification method used in the prior art affects the accuracy of the MPC prediction model. The remaining control operations after obtaining the target trend identification window are the same as those in the prior art MPC algorithm and will not be detailed here.

[0068] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A control method for a photovoltaic direct-driven air source heat pump unit, characterized in that: Including steps: Collect the actual historical power generation value of the heat pump unit and historical environmental data including ambient temperature, wind speed and other environmental parameters according to the preset sampling frequency; Setting an initial trend identification window, performing trend identification based on the historical environmental data, and obtaining a predicted value of the environmental data at the current moment; Calculating the power generation trend value at the current moment using a photovoltaic power generation model based on the environmental data prediction value at the current moment; Get the current time and its previous The actual value of the generated power at a certain moment is recorded as the generated power neighborhood data; According to the relative deviation between the actual value of the generated power and the trend value of the generated power in the generated power neighborhood data, and the degree of fluctuation of the actual value of the generated power in the generated power neighborhood data, the initial trend recognition accuracy of the generated power at the next moment is obtained; the next moment is recorded as the first At a certain moment, the initial trend recognition accuracy at the next moment satisfies the relationship: ; in, It is The accuracy of identifying the initial trend of power generation at a given moment, is the natural exponential function, It is The index of the power generation trend value at the moment is the same as the The difference between the power generation neighborhood data indexes, It is The power generation at the moment The relative deviation between the actual value of the power generation and the trend value of the power generation of the power neighborhood data is It is the influencing factor of the fluctuation degree of the actual value of the power generation in the power generation neighborhood data. is the length of the preset trend identification window, The power generation at the moment The data of each power generation neighborhood is the data in its preset trend identification window; Obtaining a confidence level of the initial trend recognition accuracy at the next moment based on the actual value of the generated power at the next moment and the generated power neighborhood data, and obtaining a target trend recognition accuracy of the generated power at the next moment based on the initial trend recognition accuracy and its confidence level; Adjusting the initial trend recognition window according to the target trend recognition accuracy to obtain a target trend recognition window; According to the target trend identification window, the heat pump unit is controlled using the MPC algorithm.

2. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The said The power generation at the moment The relative deviation between the actual power generation value and the power generation trend value of the power generation neighborhood data satisfies the relationship: ; in, It is The power generation trend value at the moment The power generation trend value of the power generation neighborhood data, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is an absolute value.

3. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The influencing factor of the fluctuation degree of the actual value of the power generation in the power generation neighborhood data satisfies the relationship: ; in, It is The first environmental parameter The standard deviation of the power generation neighborhood data of the actual power generation value, It is The first environmental parameter The average value of the power generation neighborhood data of the actual power generation value, is a small value used to prevent the denominator from being zero. is the total number of environmental parameters in each environmental data.

4. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The confidence level of the initial trend recognition accuracy at the next moment satisfies the relationship: ； in, It is The confidence level of the accuracy of the initial trend identification at the moment, is the length of the preset trend identification window, It is The power generation trend value at the moment The actual value of the power generation of the power generation neighborhood data, It is The actual value of generated power at a moment, is the range of the actual value of the generated power at multiple moments, is a natural exponential function.

5. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The target trend recognition accuracy is the product of the initial trend recognition accuracy of the power generation at the next moment and its confidence.

6. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The step of setting an initial trend identification window, performing trend identification based on the historical environmental data, and obtaining a predicted value of the environmental data at the current moment includes: Acquire multiple environmental parameters in historical environmental data to form time series data for each environmental parameter, and convert the initial trend recognition window of each data point in the time series data corresponding to each environmental parameter into the time series data. The data points are recorded as the neighborhood data of environmental parameters; Perform curve fitting based on the environmental parameters at each moment in the environmental parameter neighborhood data to obtain the fitting slope at the current moment, predict the environmental parameters at the current moment based on the fitting slope, and obtain the environmental data prediction value at the current moment.

7. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The step of calculating the power generation trend value at the current moment by using a photovoltaic power generation model based on the environmental data prediction value at the current moment includes: According to the heat pump COP function, combined with the current environmental data, a photovoltaic power generation prediction model is constructed to obtain the current power generation trend value.

8. The control method of the photovoltaic direct-driven air source heat pump unit according to claim 1, characterized in that: The target trend identification window satisfies the relationship: ; in, is the length of the target trend identification window, is the length of the preset trend identification window, It is The target trend recognition accuracy of power generation at each moment, is the maximum function, yes decimal, used to limit The minimum value of .

Images