Quantum-based photovoltaic ultra-short-term power prediction method

Abstract

The invention provides a photovoltaic ultra-short-term power prediction method based on quantum. The method comprises the steps of performing quantization processing on historical actual power, historical weather information, historical sky imaging information and weather forecast information of a photovoltaic power station to be predicted to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function and a weather forecast wave function; constructing a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function; constructing an interaction function between information at different moments; and constructing a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function so as to carry out photovoltaic ultra-short-term power prediction. Therefore, the influence of various factors on the photovoltaic power can be accurately captured based on the quantum information processing technology, so that the accuracy and reliability of photovoltaic power prediction are effectively improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of photovoltaic power generation, and in particular to a quantum-based photovoltaic ultra-short-term power prediction method. Background Art

[0002] As the proportion of renewable energy in the global energy structure continues to rise, photovoltaic power stations, as an important source of clean energy, are crucial to the accurate prediction of power output for grid dispatching, power market transactions and energy management systems. Photovoltaic power is affected by many factors, including solar radiation intensity, weather conditions, temperature, cloud cover, etc. The rapid changes of these factors have brought great challenges to ultra-short-term (generally referring to the next few minutes to hours) power prediction.

[0003] The related art cannot accurately capture the complex and nonlinear dynamic changes of various influencing factors when predicting photovoltaic power, resulting in low accuracy of power prediction results. Summary of the invention

[0004] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the purpose of the present disclosure is to propose a quantum-based photovoltaic ultra-short-term power prediction method, device, computer equipment and storage medium, thereby being able to accurately capture the impact of various factors on photovoltaic power based on quantum information processing technology, thereby effectively improving the accuracy and reliability of photovoltaic power prediction.

[0006] To achieve the above-mentioned purpose, the quantum-based photovoltaic ultra-short-term power prediction method proposed in the first aspect of the present disclosure includes:

[0007] Obtain the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted;

[0008] Quantizing the historical actual power, the historical weather information, the historical sky imaging information, and the weather forecast information respectively to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function, and a weather forecast wave function;

[0009] Constructing a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, wherein the time correlation function is used to indicate the correlation relationship between the historical weather information and the historical sky imaging information and the weather forecast information;

[0010] Constructing an interaction function between information at different times, wherein the interaction function is used to indicate a dependency relationship between information at different times;

[0011] Based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, a power prediction model is constructed, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

[0012] To achieve the above-mentioned purpose, the quantum-based photovoltaic ultra-short-term power prediction device proposed in the second embodiment of the present disclosure includes:

[0013] An acquisition module is used to acquire the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted;

[0014] A processing module, used to perform quantization processing on the historical actual power, the historical weather information, the historical sky imaging information and the weather forecast information respectively, so as to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function and a weather forecast wave function;

[0015] A first construction module is used to construct a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, wherein the time correlation function is used to indicate the correlation relationship between the historical weather information and the historical sky imaging information and the weather forecast information;

[0016] A second building module is used to build an interaction function between information at different moments, wherein the interaction function is used to indicate a dependency relationship between information at different moments;

[0017] The third building module is used to build a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

[0018] The computer device proposed in the third aspect embodiment of the present disclosure includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the quantum-based photovoltaic ultra-short-term power prediction method proposed in the first aspect embodiment of the present disclosure is implemented.

[0019] The fourth aspect embodiment of the present disclosure proposes a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the quantum-based photovoltaic ultra-short-term power prediction method proposed in the first aspect embodiment of the present disclosure.

[0020] The fifth aspect of the present disclosure provides a computer program product. When the instructions in the computer program product are executed by a processor, the quantum-based photovoltaic ultra-short-term power prediction method proposed in the first aspect of the present disclosure is performed.

[0021] The quantum-based photovoltaic ultra-short-term power prediction method, device, computer equipment and storage medium provided by the present disclosure obtain the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted; respectively quantize the historical actual power, historical weather information, historical sky imaging information and weather forecast information to generate photovoltaic power station power wave function, historical weather information wave function, historical sky imaging wave function and weather forecast wave function; construct a time correlation function according to the historical weather information wave function, historical sky imaging wave function and weather forecast wave function, wherein the time correlation function is used to indicate the correlation between the historical weather information and the historical sky imaging information and the weather forecast information; construct an interaction function between information at different times, wherein the interaction function is used to indicate the dependency between information at different times; construct a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used to perform photovoltaic ultra-short-term power prediction. Thus, it is possible to accurately capture the influence of various factors on photovoltaic power based on quantum information processing technology, thereby effectively improving the accuracy and reliability of photovoltaic power prediction.

[0022] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0024] Figure 1 It is a flow chart of a method for photovoltaic ultra-short-term power prediction based on quantum proposed in one embodiment of the present disclosure;

[0025] Figure 2 is a flow chart of a quantum-based photovoltaic ultra-short-term power prediction method proposed in another embodiment of the present disclosure;

[0026] Figure 3 It is a calculation logic diagram of the quantum-based photovoltaic ultra-short-term power prediction method proposed in the present disclosure;

[0027] Figure 4 It is a schematic diagram of the steps of the quantum-based photovoltaic ultra-short-term power prediction method proposed in the present disclosure;

[0028] Figure 5 It is a schematic diagram of the structure of a quantum-based photovoltaic ultra-short-term power prediction device proposed in one embodiment of the present disclosure;

[0029] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0030] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present disclosure, and are not to be construed as limitations of the present disclosure. On the contrary, the embodiments of the present disclosure include all changes, modifications, and equivalents that fall within the spirit and connotation of the appended claims.

[0031] Figure 1 It is a flow chart of a quantum-based photovoltaic ultra-short-term power prediction method proposed in an embodiment of the present disclosure.

[0032] Among them, it should be noted that the executor of the quantum-based photovoltaic ultra-short-term power prediction method of this embodiment is a quantum-based photovoltaic ultra-short-term power prediction device, which can be implemented by software and / or hardware. The device can be configured in a computer device, and the computer device can include but is not limited to a terminal, a server, etc. For example, the terminal can be a mobile phone, a handheld computer, etc.

[0033] like Figure 1 As shown, the quantum-based photovoltaic ultra-short-term power prediction method includes:

[0034] S101: Acquire historical actual power, historical weather information, historical sky imaging information, and weather forecast information of the photovoltaic power station to be predicted.

[0035] Among them, historical actual power refers to the actual power generation data of the photovoltaic power station in a certain period of time in the past. These data directly reflect the performance of the photovoltaic system under different environmental conditions.

[0036] The historical weather information refers to weather data in the area where the photovoltaic power station is located that may affect the working performance of the photovoltaic panels within a certain period of time in the past, such as temperature, humidity, etc., without limitation.

[0037] Among them, historical sky imaging information refers to the sky imaging information in the area where the photovoltaic power station is located in a certain period of time in the past. This historical sky imaging information can more intuitively evaluate the cloud cover and light transmittance, and provide more intuitive spatial feature information for prediction.

[0038] Among them, weather forecast information may refer to future weather forecast data calculated based on meteorological models, which helps to estimate the impact of upcoming weather changes on photovoltaic power output.

[0039] In the disclosed embodiment, when the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted are obtained, multi-dimensional data support can be provided for the photovoltaic ultra-short-term power prediction, thereby effectively improving the reliability of the photovoltaic ultra-short-term power prediction process.

[0040] S102: quantizing the historical actual power, historical weather information, historical sky imaging information and weather forecast information respectively to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function and a weather forecast wave function.

[0041] In the disclosed embodiments, historical actual power, historical weather information, historical sky imaging information, and weather forecast information can be used as a space-time modality, respectively. "Space-time modality" is an abstract conceptual model that attempts to integrate the characteristics of the above-mentioned types of data in time and space dimensions. On the time axis, the changing trend of the historical data sequence over time is considered; on the space axis, attention is paid to the relationship between different types of data (such as weather variables, image features, etc.) and their comprehensive performance at specific time and space points. From the perspective of "entangled superposition", it means that this information not only exists independently, but also in certain states it is intertwined in a complex and interdependent way to form a unique prediction model.

[0042] Among them, the photovoltaic power station power wave function is a quantum representation of the spatiotemporal mode of photovoltaic power station power. The historical weather information wave function is a quantum representation of the spatiotemporal mode of historical weather information. The historical sky imaging wave function is a quantum representation of the spatiotemporal mode of historical sky imaging. The weather forecast wave function is a quantum representation of the spatiotemporal mode of weather forecast.

[0043] In the embodiments of the present disclosure, when the historical actual power, historical weather information, historical sky imaging information and weather forecast information are quantized to generate the photovoltaic power station power wave function, the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, the historical actual power, the historical weather information, the historical sky imaging information and the weather forecast information can be preprocessed, including removing outliers, filling missing data, etc.; and then the preprocessed data are standardized to a suitable range, usually between 0 and 1, for subsequent quantization processing; then the standardized data is quantized, for example, using a quantizer or segmenting the data to achieve this, and the quantizer can convert continuous power data into discrete values ​​according to a pre-set quantization level; finally, the quantized data is converted into a wave function form. In quantum mechanics, the wave function describes the state of the system.

[0044] Optionally, in some embodiments, the historical actual power includes a power value P(t) corresponding to time t, and the power wave function of the photovoltaic power station is generated based on the following method:

[0045] Based on the preset spatial dimension N and the power value P(t) at the moment, a photovoltaic power station power wave function |ψ(t)> corresponding to the moment t is generated; wherein the photovoltaic power station power wave function |ψ(t)> is expressed by the following formula:

[0046]

[0047] Among them, A j and θ j denote the complex amplitude and real phase of the photovoltaic power station power wave function |ψ(t)>, θ j ∈[-π,π], the normalization condition of the photovoltaic power station power wave function |ψ(t)> is

[0048] Therefore, the power information of the PV power station can be encoded by a complex vector (i.e., the power wave function of the PV power station), which not only captures the change of power over time, but also contains richer information structure through the amplitude and phase of the complex number.

[0049] The preset spatial dimension N may refer to a Hilbert space dimension pre-set for historical actual power, historical weather information, historical sky imaging information, and weather forecast information.

[0050] In the embodiment of the present disclosure, an N-dimensional complex vector |ψ(t)> can be used to represent the power information of the photovoltaic power station at time t. The power value at each time t is P(t).

[0051] It can be understood that in the embodiments of the present disclosure, A jand θ j The value of can be trained in the model and is applicable to any time t.

[0052] Optionally, in some embodiments, the historical weather information includes the actual weather information corresponding to time t The actual weather information includes the temperature, humidity, pressure and light radiation value corresponding to time t. Recorded as The historical weather information wave function is generated based on the following method:

[0053] Based on the preset spatial dimension N and actual weather information Generate the historical weather information wave function |φ corresponding to time t R (t)>; where the historical weather information wave function |φ R (t)> is represented by the following formula:

[0054]

[0055] Among them, B j Represents the historical weather information wave function |φ R (t)>, Represents the historical weather information wave function |φ R (t)> is the jth real coefficient vector with dimension 4, and the normalization condition of the historical weather information wave function is

[0056] Therefore, the resulting historical weather information wave function realizes the spatiotemporal modal quantum representation of historical weather information. The above formula extends the concept of quantum state to the representation of multi-dimensional meteorological data. The wave function structure is similar to the previous historical power representation of photovoltaic power plants, but its construction is more complex because it needs to capture weather information in four dimensions: temperature, humidity, pressure, and light irradiation. This structure allows the phase change of the wave function to be associated with all meteorological parameters, thereby delicately encoding the multi-dimensional characteristics of the weather state.

[0057] It can be understood that in the embodiments of the present disclosure, B, and It can be trained in the model and is applicable to any time t. j and Essentially, a set of weights is defined to measure the importance of each meteorological parameter in forming a representation of the weather state. Through training, these weights can be adaptively adjusted to best match the patterns in historical data. The use of normalization conditions ensures the probabilistic interpretation of the wave function, that is, the sum of the probabilities of all weather states is 1, which is the basis for building an effective physical model. The determination of parameters relies on the training process of historical weather data, the purpose of which is to learn the inherent laws of weather changes and the complex interactions between different meteorological parameters.

[0058] Optionally, in some embodiments, the historical sky imaging information includes the sky imaging at the time corresponding to the time t, and the historical sky imaging wave function is generated based on the following method:

[0059] Extract N eigenvalues ​​from the sky image at the moment based on the preset spatial dimension N Based on the N eigenvalues, the historical sky imaging wave function |φ corresponding to time t is generated I (t)>; where the historical sky imaging wave function |φ I (t)> is represented by the following formula:

[0060]

[0061] Among them, C j represents the historical sky imaging wave function |φ I (t)>, the complex amplitude, ω j represents the historical sky imaging wave function |φ I (t)>, the normalization condition of the historical sky imaging wave function is

[0062] In this way, the spatiotemporal modal quantum representation of historical sky imaging can be achieved. The above formula converts the characteristics of sky images into the form of quantum states, providing a novel perspective for the analysis and prediction of sky images.

[0063] In the present disclosure, C j and ω j It can be trained in the model and is applicable to any time t.

[0064] In the disclosed embodiment, the features extracted from sky imaging may include cloud coverage, color distribution, texture features, etc., which together constitute the key information of the sky image. The wave function structure can capture the dynamic characteristics of the sky image over time, and the change in phase reflects the evolution of the feature intensity or state. The normalization condition is the same as the quantum representation of weather information mentioned above, ensuring that the wave function can be interpreted as a probability distribution, and the sum of the probability of occurrence of each feature is 1, which is crucial for subsequent probabilistic analysis. Parameter C j and ω jThe determination of also relies on the training of historical sky image datasets. Learning these parameters can effectively encode the temporal and spatial variation patterns of sky images. By learning the patterns of historical sky images, quantum representation can be used to predict future sky conditions, such as cloud changes, sunset / sunrise times, etc., which is of great value to aviation, photography, outdoor activities and other fields. Combining the quantum representation of sky imaging with traditional meteorological data can provide a more comprehensive perspective for weather forecasts and improve the accuracy of warnings for extreme weather events. Quantum representation methods can provide deeper data analysis capabilities. The spatiotemporal modal quantum representation of historical sky imaging theoretically expands the boundaries of image processing and pattern recognition.

[0065] Optionally, in some embodiments, the weather forecast information includes the predicted weather information corresponding to time t The predicted weather information includes the temperature, humidity, pressure and light radiation value corresponding to time t. Recorded as The weather forecast wave function is generated based on the following method:

[0066] Based on the preset spatial dimension N and predicted weather information Generate the weather forecast wave function |φ corresponding to time t P (t)>; where the weather forecast wave function |φ P (t)> is represented by the following formula:

[0067]

[0068] Among them, D j represents the weather forecast wave function |φ P (t)>, represents the weather forecast wave function |φ P (t)> is the jth real coefficient vector with dimension 4, and the normalization condition of the weather forecast wave function is

[0069] This provides an innovative framework for understanding and predicting weather systems. The weather forecast information at each moment is encoded as a four-dimensional vector, representing four key meteorological parameters: temperature, humidity, pressure, and light irradiance. This representation directly maps continuous variables in the physical world, facilitating the construction of subsequent quantum states.

[0070] Understandably, D j and It can be trained in the model and is applicable to any time t.

[0071] In the disclosed embodiment, the phase of the weather forecast wave function captures the complex interactions and dynamics between different meteorological parameters, and the normalization condition maintains the normalization of the wave function to ensure that it can be interpreted as a probability distribution, corresponding to the probability of a specific weather "pattern" occurring, and the sum of all pattern probabilities is 1, which is the basis of physical interpretability and statistical inference. The determination of parameters relies on the learning process of historical weather data, aiming to automatically extract the characteristics and correlations of weather patterns from a large number of observations. Weather models based on quantum representation can more accurately simulate and predict weather changes, especially for extreme weather events, because they can better capture the nonlinear relationships between complex meteorological parameters.

[0072] S103: constructing a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, wherein the time correlation function is used to indicate the correlation relationship between the historical weather information and the historical sky imaging information and the weather forecast information.

[0073] In the disclosed embodiments, when constructing a time correlation function based on the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function may be input into a pre-trained machine learning model to obtain the corresponding time correlation function. Alternatively, a relationship analysis may be performed on the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function based on a third-party device to obtain the time correlation function.

[0074] In the disclosed embodiment, when a time correlation function is constructed based on the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, the obtained time correlation function can accurately indicate the correlation between weather information at different times, thereby providing reliable reference information for the subsequent construction of a power prediction model.

[0075] S104: constructing an interaction function between information at different moments, wherein the interaction function is used to indicate a dependency relationship between information at different moments.

[0076] It is understandable that the longer the time interval, the weaker the correlation between the two information. Therefore, in the embodiment of the present disclosure, an interaction function between information at different times can be constructed, thereby providing a reliable judgment basis for analyzing the interaction strength between information at different times in the subsequent construction of a power prediction model.

[0077] Optionally, in some embodiments, the interaction function is represented by the following formula:

[0078]

[0079] Among them, time t′ is earlier than time t, Λ is the preset time cutoff factor, V0 is the interaction strength, and μ is the attenuation factor of the interaction between different time information.

[0080] Therefore, the dependency between information at different times can be formally expressed through the Gaussian attenuation function, effectively capturing the phenomenon that information correlation weakens with increasing time distance.

[0081] The specific value of the preset time cutoff factor can be set manually, such as one week, three days, etc., and there is no restriction on this.

[0082] Among them, the attenuation factor μ can be obtained during the model training process.

[0083] From the above expression of the interaction function, it can be seen that the smaller the two time intervals are, the stronger the interaction between the information is. When the time interval is greater than Λ, it is considered that there is no interaction between the information.

[0084] In the disclosed embodiment, the parameter interaction strength V0 determines the baseline strength of information interaction, reflecting how strong the direct correlation between information is when there is no time gap. The decay factor μ controls the speed at which information decays over time. A larger value indicates that information correlation decays faster over time, reflecting the system's higher dependence on recent information. The time truncation factor defines a time window within which historical information is considered to contribute to the forecast. Beyond this time window, the information is considered no longer relevant. This parameter can be set based on the memory effect of the actual weather system to help the model focus on the most relevant historical data.

[0085] S105: constructing a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

[0086] That is to say, after obtaining the weather forecast wave function, time correlation function, interaction function and photovoltaic power station power wave function in the embodiment of the present disclosure, a power prediction model can be constructed based on the weather forecast wave function, time correlation function, interaction function and photovoltaic power station power wave function, thereby achieving accurate prediction of photovoltaic ultra-short-term power.

[0087] In the disclosed embodiment, the power prediction model can be dynamically changing, that is, the corresponding weather forecast wave function, time correlation function, interaction function and photovoltaic power station power wave function can be obtained for any time t when power prediction is required, thereby ensuring the applicability of the constructed power prediction model.

[0088] In this embodiment, by obtaining the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted; quantizing the historical actual power, historical weather information, historical sky imaging information and weather forecast information respectively to generate photovoltaic power station power wave function, historical weather information wave function, historical sky imaging wave function and weather forecast wave function; constructing a time correlation function according to the historical weather information wave function, historical sky imaging wave function and weather forecast wave function, wherein the time correlation function is used to indicate the correlation between the historical weather information and the historical sky imaging information and the weather forecast information; constructing an interaction function between information at different times, wherein the interaction function is used to indicate the dependency between information at different times; constructing a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used to perform photovoltaic ultra-short-term power prediction. Thus, it is possible to accurately capture the influence of various factors on photovoltaic power based on quantum information processing technology, thereby effectively improving the accuracy and reliability of photovoltaic power prediction.

[0089] Figure 2 It is a flow chart of a quantum-based photovoltaic ultra-short-term power prediction method proposed in another embodiment of the present disclosure.

[0090] like Figure 2 As shown, the quantum-based photovoltaic ultra-short-term power prediction method includes:

[0091] S201: Acquire historical actual power, historical weather information, historical sky imaging information, and weather forecast information of the photovoltaic power station to be predicted.

[0092] S202: quantizing the historical actual power, historical weather information, historical sky imaging information and weather forecast information respectively to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function and a weather forecast wave function.

[0093] The description of S201 and S202 can be specifically referred to the above embodiment, which will not be repeated here.

[0094] S203: Calculate the outer product of the historical weather information wave function and its Dirac conjugate to obtain the historical weather information density matrix.

[0095] S204: Calculate the outer product of the historical sky imaging wave function and its Dirac conjugate to obtain the historical sky imaging density matrix.

[0096] That is, in the embodiment of the present disclosure, the historical weather information density matrix ρ can be generated for each time t R (t) and the historical sky imaging density matrix ρ I (t).

[0097] Historical weather information density matrix ρ R (t) and the historical sky imaging density matrix ρ I (t) are respectively the wave functions of historical weather information |φ R (t)> and the historical space imaging wave function |φ I (t)> is constructed, and the formula is as follows:

[0098] ρ R / I (t)=|φ R / I (t)><φ R / I (t)|

[0099] Among them, <φ R / I (t)| is the wave function |φ R / I (t)> Take the vector transpose and then take the complex conjugate result, and the ρ obtained from the above formula R / I (t) is a matrix of N×N dimensions.

[0100] According to the above formula, the historical weather information density matrix and the historical sky imaging density matrix can be constructed by their corresponding wave functions. Specifically, both density matrices are obtained by the outer product of the wave function with its own Dirac conjugate (i.e., complex conjugate transpose), which reflects the purity and information content of the quantum state. The density matrix provides a complete description of the state of the quantum system, not only for pure states (as directly constructed by wave functions here), but also for mixed states. Each density matrix captures all the statistical characteristics of weather information or sky imaging at the corresponding historical moment. In particular, for pure states, the elements on the diagonal of the density matrix represent the probabilities of each eigenstate, and the off-diagonal elements are associated with the coherence between different eigenstates. The density matrix provides a compact representation of historical weather or sky image information that evolves over time, which is conducive to tracking and analyzing state changes. In this way, encoding weather forecasts and sky imaging information into density matrices provides a quantum mechanical perspective and tools for processing such complex systems, which helps to deeply understand the dynamics of atmospheric systems and promote the development of more efficient data analysis and prediction technologies.

[0101] S205: Constructing a historical weather information mixed density matrix based on a preset first proportionality coefficient, a historical weather information density matrix, and a historical sky imaging density matrix.

[0102] That is, in the embodiment of the present disclosure, for each time t, a historical weather information mixed density matrix ρ can be generated. sky (t).

[0103] Since historical weather information includes historical temperature, irradiance and other information and historical sky imaging, the mixed density matrix of historical weather information is constructed with the following formula:

[0104] ρ sky (t) = αρ R (t)+(1-α)ρ I (t)

[0105] The preset first proportionality coefficient α can be obtained by training in the model.

[0106] According to the above formula, the historical weather information mixed density matrix is ​​constructed by combining the historical weather information density matrix and the historical sky imaging density matrix, which reflects a method of integrating two different types of data. The proportional coefficient α is determined during the model training process and is used to balance the relative importance of weather information (such as temperature, irradiance, etc.) and sky imaging information. The density matrix of historical weather information contains comprehensive information on meteorological parameters (such as temperature, humidity, wind speed, irradiance, etc.) at each time. The density matrix of historical sky imaging reflects the visual characteristics of the sky at the same time point, such as cloud cover, color distribution, etc.

[0107] α is a weighting factor between 0 and 1, which determines the contribution ratio of weather information and sky imaging information in the mixed density matrix. When α is close to 1, the mixed matrix focuses more on weather information; when it is close to 0, it relies more on sky image information. The mixed density matrix provides a unified framework for simultaneously considering and analyzing the physical properties and visual appearance of weather conditions, which is very useful for application scenarios that require comprehensive prediction or analysis of multiple factors, such as advanced meteorological models, environmental monitoring, solar resource assessment, etc. By adjusting α, the two types of data can be customized to be fused according to actual needs or model performance optimization, thereby improving the accuracy and practicality of the model.

[0108] S206: Constructing a first correlation function based on the historical weather information mixed density matrix, the weather forecast wave function and the historical weather information wave function.

[0109] The first correlation function may be used to indicate the correlation between historical weather information and weather forecast.

[0110] S207: Construct a second correlation function based on the historical weather information mixed density matrix, the weather forecast wave function and the historical sky imaging wave function.

[0111] The second correlation function can be used to indicate the correlation between historical sky imaging and weather forecasts.

[0112] That is to say, in the embodiments of the present disclosure, historical weather information and weather forecast information may be fused to generate a time correlation function.

[0113] The implementation process of weather forecasting is to predict the weather information at future time (t) based on historical weather information (t′ < t). This process reflects the entanglement between information at different times. For the weather system, the time correlation function is used to characterize the information correlation between different times, and the formula is as follows:

[0114]

[0115]

[0116] where Tr represents taking the trace of a matrix, and the first correlation function G R (t, t′) (the second correlation function G I (t, t′)) reveals the correlation between historical weather information (historical sky imaging) and weather forecasting.

[0117] S208: Construct a time correlation function based on a preset second proportionality coefficient, the first correlation function, and the second correlation function.

[0118] In the embodiments of the present disclosure, the total correlation function can be constructed as follows:

[0119] G(t, t′) = βG R (t, t′) + (1 - β)G I (t, t′)

[0120] where the preset second proportionality coefficient β can be obtained through training in the model.

[0121] The above formula quantifies the connection between historical weather information and future weather forecasting by introducing the time correlation function. This step utilizes an important concept in quantum theory. Although in fact the weather system does not follow the laws of quantum mechanics, this abstract method can inspire new analysis means. The historical weather information correlation function G R (t, t′) measures the degree of correlation between the historical weather information at time t′ and the weather forecasting information at time t under the actual weather information environment at historical time t′ (characterized by the historical weather information mixed density matrix ρ sky (t′)). The historical sky imaging correlation function G I (t, t′) measures the degree of correlation between the historical sky imaging at time t′ and the weather forecasting information at time t under the actual weather information environment at historical time t′, capturing how the visual features of the sky affect the prediction of future weather.

[0122] By combining the two and introducing a new weight coefficient β, the total correlation function can obtain a total correlation measure that integrates the impact of historical weather information and historical sky imaging information on weather forecasts. β once again reflects the optimal choice of the importance of different information sources during model training. Such a correlation function can not only help us better understand the spatiotemporal continuity of weather systems, but also optimize weather forecast models in practice. By learning and adjusting the β parameter, the model can learn the most effective information fusion strategy to improve forecast accuracy. In particular, when faced with complex and changing climate conditions and nonlinear dynamic systems, this approach provides a new way to quantify and exploit the interactions between data of different dimensions, quantifying information transfer across time scales through the temporal correlation function.

[0123] That is to say, in the embodiment of the present disclosure, after generating the photovoltaic power station power wave function, the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, the outer product of the historical weather information wave function and its Dirac conjugate can be calculated to obtain the historical weather information density matrix; the outer product of the historical sky imaging wave function and its Dirac conjugate can be calculated to obtain the historical sky imaging density matrix; based on the preset first proportionality coefficient, the historical weather information density matrix and the historical sky imaging density matrix, a historical weather information mixed density matrix is ​​constructed; based on the historical weather information mixed density matrix, the weather forecast wave function and the historical weather information wave function, a first correlation function is constructed; based on the historical weather information mixed density matrix, the weather forecast wave function and the historical weather information wave function, a second correlation function is constructed; based on the preset second proportionality coefficient, the first correlation function and the second correlation function, a time correlation function is constructed. Thus, it can be ensured that the obtained time correlation function accurately indicates the total correlation measure of the impact of the historical weather information and the historical sky imaging information on the weather forecast, thereby effectively improving the practicality and reliability of the time correlation function.

[0124] S209: constructing an interaction function between information at different moments, wherein the interaction function is used to indicate a dependency relationship between information at different moments.

[0125] S210: constructing a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

[0126] The description of S209 and S210 can be specifically referred to the above embodiment, which will not be repeated here.

[0127] In this embodiment, the outer product of the historical weather information wave function and its Dirac conjugate is calculated to obtain the historical weather information density matrix; the outer product of the historical sky imaging wave function and its Dirac conjugate is calculated to obtain the historical sky imaging density matrix; based on the preset first proportionality coefficient, the historical weather information density matrix and the historical sky imaging density matrix, a historical weather information mixed density matrix is ​​constructed; based on the historical weather information mixed density matrix, the weather forecast wave function and the historical weather information wave function, a first correlation function is constructed; based on the historical weather information mixed density matrix, the weather forecast wave function and the historical sky imaging wave function, a second correlation function is constructed; based on the preset second proportionality coefficient, the first correlation function and the second correlation function, a time correlation function is constructed. Thus, it can be ensured that the obtained time correlation function accurately indicates the total correlation measure of the impact of the historical weather information and the historical sky imaging information on the weather forecast, thereby effectively improving the practicality and reliability of the time correlation function.

[0128] Optionally, in some embodiments, the power prediction model is based on a Lippman-Schwinger type equation to represent the change of the photovoltaic power station power wave function |ψ(t)> over time. The Lippman-Schwinger type equation is represented by the following formula:

[0129]

[0130] Among them, |φ P (t)> is the weather forecast wave function, Λ is the preset time cutoff factor, time t′ is earlier than time t, G(t, t′) is the time correlation function, and V(t, t′) is the interaction function.

[0131] Therefore, the power prediction model can incorporate the dynamic impact of historical information on the current power forecast while considering weather forecast information, and enhance the prediction accuracy through the interaction between historical information and current forecast.

[0132] It can be seen from the above formula that when there is no historical information, the power can only be predicted by relying on weather forecasts; when there is historical information, the historical information will have an important impact on the predicted power.

[0133] Lippmann-Schwinger type equations have been widely used in quantum scattering theory. Here, they are creatively used to describe and fuse information from multiple spatiotemporal modalities, especially in the context of photovoltaic power prediction. This equation not only takes into account the immediate weather forecast information, but also incorporates the dynamic impact of historical information on the current power forecast, and enhances the prediction accuracy through the interaction between historical information and the current forecast. The equation describes the evolution of the power wave function over time, which contains two parts: one is the part that directly depends on the weather forecast information, representing the direct prediction based on the future weather forecast; the other is the correction term of the historical information on the current prediction, which is reflected in the form of an integral. This term acts on the historical power wave function through the time correlation function and the interaction function.

[0134] The fusion of historical information shows how the power wave function at a historical moment affects the power forecast at the current moment through the time correlation function and the interaction function. The time correlation function quantifies the correlation between historical weather information and sky imaging information and the current forecast, while the interaction function controls the intensity and attenuation of this influence. The lower limit of the integral sets a truncation factor, which means that only historical information within the recent time range is considered, which reflects the limited memory effect of the system on historical data, and information beyond this time window is considered invalid. By fusing instant weather forecast information and historical data, especially by considering the dynamic impact of historical information on the current state, the equation can theoretically provide a more robust and accurate ultra-short-term photovoltaic power forecast. The addition of historical information helps the model to use historical patterns and trends for correction when faced with sudden weather changes or inaccurate forecasts, thereby improving the robustness of the forecast.

[0135] Optionally, in some embodiments, the photovoltaic power station power wave function ψ(t)> may be substituted into a preset formula to obtain a photovoltaic power station power prediction value, wherein the preset formula is expressed by the following formula:

[0136]

[0137] Among them, ψ j (t) is the jth element of the photovoltaic power station power wave function |ψ(t)>, abs represents the absolute value of the calculation result, N is the preset spatial dimension, A j and θ j They represent the complex amplitude and real phase of the power wave function |ψ(t)> of the photovoltaic power station respectively.

[0138] In this way, the quantum wave function generated by the prediction process can be converted back into the actual power prediction value, ensuring that the prediction result is closely consistent with the actual power generation performance of the photovoltaic power station, thereby improving the practicality and reliability of the prediction.

[0139] The above formula expresses several steps. First, the contribution of each component is calculated, reflecting the deviation of each wave function component from its ideal state and its importance. Then, the contributions of all components are summed (or equivalently converted to a continuous multiplication form) and multiplied by i for normalization and adjustment. Finally, the absolute value of the result is taken to ensure that the actual power obtained is a positive real number.

[0140] Optionally, in some embodiments, the loss function of the power prediction model is expressed by the following formula:

[0141]

[0142] Where MSE stands for mean square error, n is the number of samples, and y k represents the true value of the kth sample, Represents the predicted value of the kth sample.

[0143] Therefore, the optimization effect of the power prediction model can be effectively improved based on the loss function, ensuring the accuracy and reliability of the model prediction results.

[0144] That is, in the embodiments of the present disclosure, the mean squared error (MSE) loss function and the Adam algorithm can be used to optimize hyperparameters, and the best model after the hyperparameter optimization is used for prediction.

[0145] Based on the above embodiments, the present invention provides a quantum-based photovoltaic ultra-short-term power prediction method, which combines the concept of quantum physics with traditional energy prediction technology, and opens up a new path in the field of photovoltaic power prediction. The spatiotemporal modal construction forms a comprehensive "spatial modality" by integrating the historical photovoltaic power data of the time series with multi-dimensional spatial data (meteorological conditions, sky imaging, numerical weather forecasts, etc.). This integration goes beyond the limitations of a single data source in traditional methods and captures the complex factors that affect photovoltaic output more comprehensively. The present invention is inspired by quantum theory. Although it does not directly rely on quantum hardware, the method cleverly draws on the core concepts of quantum theory, such as wave functions, probability amplitudes, and correlation functions. These concepts are used to design new data analysis algorithms to help models better understand and simulate nonlinear relationships and dynamic interactions between data, similar to entanglement and superposition phenomena in quantum systems. The present invention uses high-dimensional vectors to express spatiotemporal characteristics, which not only enhances the model's ability to handle complex data structures, but also promotes the effective fusion of multi-source data. This method improves the prediction accuracy of the model in the face of high nonlinearity and uncertainty. The present invention is a product of the cross-integration of multiple disciplines such as physics, statistics, and computer science. It not only relies on mature machine learning and deep learning technologies, but also incorporates the thinking mode of quantum theory, showing the strong potential of interdisciplinary research. Especially for photovoltaic power prediction in the next few hours, this method shows the potential to significantly improve the prediction accuracy through the above innovative strategies. This is crucial for optimizing grid scheduling, enhancing the utilization of renewable energy, and improving the overall flexibility and reliability of the energy system.

[0146] Based on the above embodiments, Figure 3 and Figure 4 As shown, Figure 3 is a calculation logic diagram of the quantum-based photovoltaic ultra-short-term power prediction method proposed in the present disclosure, Figure 4 This is a schematic diagram of the steps of the quantum-based photovoltaic ultra-short-term power prediction method proposed in the present disclosure. In combination with the above embodiments, the technical key point of the present invention is that it innovatively combines the concept of quantum theory with the needs of photovoltaic ultra-short-term power prediction, which is specifically reflected in the following aspects:

[0147] 1. Quantum representation construction: Convert the data in the traditional prediction model into quantum state representation, including the quantization of power spatiotemporal modalities, historical weather information, sky imaging data, and weather forecast information. This method uses complex wave functions to carry rich information about power changes, environmental impacts, and weather parameters, providing a higher-dimensional and deeper way of expressing data.

[0148] 2. Spatiotemporal modal analysis: By integrating information in both time and space, a prediction framework is constructed that can capture the complex interactions between historical trends, environmental changes, and future forecasts. This approach goes beyond the linear or simple nonlinear processing of traditional prediction models and can better deal with the nonlinearity, complexity, and uncertainty issues in photovoltaic power prediction.

[0149] 3. Information fusion technology: The density matrix is ​​used to integrate information from different sources, and the contribution of historical data and visual information is balanced through the trained proportional coefficient, forming a unified prediction platform that includes both historical statistical characteristics and real-time visual analysis. This helps to improve the model's responsiveness to real-time changes and prediction accuracy.

[0150] 4. Dynamic adjustment of the prediction model: The Lippman-Schwinger equation is used as the core prediction model. The model can dynamically integrate the real-time weather forecast with historical data, adapt to the rapid changes in weather in the short term, and significantly improve the robustness and accuracy of the prediction model.

[0151] 5. Decoding process: Convert the quantum wave function generated by the prediction process back to the actual power prediction value to ensure that the prediction result is closely aligned with the actual power generation performance of the photovoltaic power station, thereby improving the practicality and reliability of the prediction.

[0152] In summary, the key technical point of the present invention is that it innovatively uses the analogy of quantum theory for data characterization and analysis, combines spatiotemporal modal analysis with information fusion technology, and achieves a significant improvement in the accuracy and efficiency of ultra-short-term power prediction of photovoltaic power stations.

[0153] Points to be protected: 1. Construction and analysis methods of space-time modalities: Innovative ways to integrate historical power data, historical weather information, historical sky imaging information and numerical weather forecast information into space-time modalities, especially how these modalities are related to each other in time series and spatial characteristics, and how to understand and optimize the processing of these data through the concept of "entanglement superposition" in quantum theory.

[0154] 2. Innovative application of time correlation function: Using time correlation function to characterize the relationship between different modes and the intrinsic correlation of time series data is a unique data fusion method that aims to more accurately capture and predict photovoltaic power changes.

[0155] 3. Algorithm innovation steps: Based on the analogy of quantum theory, new data analysis algorithms or frameworks may be developed, including but not limited to using probability amplitude and interference principles to deal with uncertainty in predictions, and developing high-dimensional data representation methods to more efficiently characterize and process spatiotemporal features.

[0156] 4. Improved accuracy of ultra-short-term predictions: The invention aims to significantly improve the accuracy of ultra-short-term power predictions of photovoltaic power stations through the above-mentioned innovations, especially when dealing with rapidly changing weather conditions. This is a direct benefit brought about by its technical improvement and an important protection point.

[0157] These protection points together constitute the originality and practicality of the invention.

[0158] The prediction method provided by the present invention elevates the ultra-short-term power prediction of photovoltaic power stations to a more theoretical level. It mainly includes the following aspects:

[0159] 1. Quantum theory simulation of space-time modes: Historical power data, historical weather information, historical sky imaging information and numerical weather forecasts are regarded as four independent "space-time modes". This approach innovatively introduces concepts from the field of physics into data analysis. Under the framework of quantum theory, each mode can be compared to a quantum system, and its state can be described by a wave function. The evolution of the wave function over time reflects the law of data change over time. Although actual operations do not involve real quantum physics experiments, this idea uses a dynamic and probabilistic perspective to understand and model the time series characteristics of data.

[0160] 2. Application of time correlation function: In classical statistical physics, time correlation function is often used to describe the correlation properties of the system state evolving over time. In the present invention, the time correlation function is cleverly used to quantify and describe the "entanglement" relationship between different modes, that is, the mutual dependence and information transfer between them. The establishment of this correlation helps to capture the deep connections hidden behind complex data, especially those reflected in the synergistic effects of data changes on different time scales.

[0161] 3. Reasonable response to internal correlation: In this way, the model not only considers the changing trend within a single mode, but also effectively integrates the interaction between the modes, so that the prediction model can more comprehensively and accurately reflect the complex impact of weather conditions and sky conditions on photovoltaic power. This cross-modal "entanglement" description theoretically enhances the model's generalization ability and ability to capture nonlinear relationships, which is particularly important for uncertainty management in ultra-short-term forecasts.

[0162] In summary, this paper proposes a novel data processing and analysis framework by drawing on the thinking mode of quantum theory, aiming to improve the accuracy and reliability of photovoltaic power plant power forecasting by highly integrating multi-source data, especially in the face of rapidly changing weather conditions. This innovative method provides a new perspective combining theory and practice in the field of energy forecasting.

[0163] The beneficial effects brought by the present invention are mainly in the following aspects: first, the prediction accuracy is significantly improved. By integrating historical data with real-time weather forecast information and innovatively using the concept of "entanglement superposition" of quantum theory to process and analyze data, this method can more accurately predict the power output of photovoltaic power stations in the next few hours under complex and changeable natural environments. This high-precision prediction is of great value for power dispatching, grid balance and energy trading. Secondly, it can optimize energy management and dispatching. Accurate ultra-short-term power forecasting helps photovoltaic power station operators to more effectively plan the generation and distribution of electric energy, reduce energy waste or insufficient supply caused by prediction errors, optimize the use of energy storage systems, and improve the flexibility and efficiency of the entire energy system. In addition, the technical solution of the invention can enhance the stability and reliability of the power grid. For power grid operators, accurate power forecasting can predict the fluctuation of photovoltaic power in advance, so as to better adjust the ratio between traditional energy generation and new energy generation, ensure the stable operation of the power grid, reduce the risk of power outages, and improve the quality of power supply. In addition, it can promote the integration of renewable energy. With the increase in the proportion of renewable energy such as photovoltaics, the power grid needs more accurate forecasting tools to deal with the intermittent and uncertain nature of renewable energy generation. This invention improves the predictability of photovoltaic power, helps access and integrate large-scale renewable energy, and promotes the clean transformation of the energy structure. In summary, this invention not only promotes the advancement of photovoltaic power prediction methods at the technical level, but also brings significant economic, environmental and social benefits to the energy industry in practical applications.

[0164] Figure 5 It is a schematic diagram of the structure of a quantum-based photovoltaic ultra-short-term power prediction device proposed in one embodiment of the present disclosure.

[0165] like Figure 5 As shown, the quantum-based photovoltaic ultra-short-term power prediction device 50 comprises:

[0166] An acquisition module 501 is used to acquire the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted;

[0167] The processing module 502 is used to quantize the historical actual power, historical weather information, historical sky imaging information and weather forecast information to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function and a weather forecast wave function;

[0168] A first construction module 503 is used to construct a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, wherein the time correlation function is used to indicate the correlation relationship between the historical weather information and the historical sky imaging information and the weather forecast information;

[0169] The second construction module 504 is used to construct an interaction function between information at different moments, wherein the interaction function is used to indicate a dependency relationship between information at different moments;

[0170] The third construction module 505 is used to construct a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

[0171] It should be noted that the aforementioned explanation of the quantum-based photovoltaic ultra-short-term power prediction method is also applicable to the quantum-based photovoltaic ultra-short-term power prediction device of this embodiment, and will not be repeated here.

[0172] In this embodiment, by obtaining the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted; quantizing the historical actual power, historical weather information, historical sky imaging information and weather forecast information respectively to generate photovoltaic power station power wave function, historical weather information wave function, historical sky imaging wave function and weather forecast wave function; constructing a time correlation function according to the historical weather information wave function, historical sky imaging wave function and weather forecast wave function, wherein the time correlation function is used to indicate the correlation between the historical weather information and the historical sky imaging information and the weather forecast information; constructing an interaction function between information at different times, wherein the interaction function is used to indicate the dependency between information at different times; constructing a power prediction model based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, wherein the power prediction model is used to perform photovoltaic ultra-short-term power prediction. Thus, it is possible to accurately capture the influence of various factors on photovoltaic power based on quantum information processing technology, thereby effectively improving the accuracy and reliability of photovoltaic power prediction.

[0173] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Figure 6 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0174] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0175] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnection (PCI) bus.

[0176] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0177] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 Not shown, often called a "hard drive").

[0178] although Figure 6 Not shown in the figure, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a compact disc read only memory (hereinafter referred to as: CD-ROM), a digital versatile disc read only memory (hereinafter referred to as: DVD-ROM) or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.

[0179] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described in the present disclosure.

[0180] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a human body to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via an input / output (I / O) interface 22. In addition, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer device 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0181] The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28, such as implementing the quantum-based photovoltaic ultra-short-term power prediction method mentioned in the above embodiment.

[0182] In order to implement the above embodiments, the present disclosure also proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the quantum-based photovoltaic ultra-short-term power prediction method proposed in the above embodiments of the present disclosure is implemented.

[0183] In order to implement the above embodiments, the present disclosure further proposes a computer program product. When the instruction processor in the computer program product is executed, the quantum-based photovoltaic ultra-short-term power prediction method proposed in the above embodiments of the present disclosure is executed.

[0184] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0185] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

[0186] It should be noted that, in the description of the present disclosure, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present disclosure, unless otherwise specified, the meaning of "plurality" is two or more.

[0187] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.

[0188] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0189] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0190] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0191] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0192] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0193] Although the embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A quantum-based photovoltaic ultra-short-term power prediction method, characterized in that: include: Obtain the historical actual power, historical weather information, historical sky imaging information and weather forecast information of the photovoltaic power station to be predicted; Quantizing the historical actual power, the historical weather information, the historical sky imaging information, and the weather forecast information respectively to generate a photovoltaic power station power wave function, a historical weather information wave function, a historical sky imaging wave function, and a weather forecast wave function; Constructing a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function, wherein the time correlation function is used to indicate the correlation relationship between the historical weather information and the historical sky imaging information and the weather forecast information; Constructing an interaction function between information at different times, wherein the interaction function is used to indicate a dependency relationship between information at different times; Based on the weather forecast wave function, the time correlation function, the interaction function and the photovoltaic power station power wave function, a power prediction model is constructed, wherein the power prediction model is used for photovoltaic ultra-short-term power prediction.

2. The method according to claim 1, characterized in that The historical actual power includes the power value P(t) corresponding to the time t, and the photovoltaic power station power wave function is generated based on the following method: Based on the preset spatial dimension N and the power value P(t) at the time, a photovoltaic power station power wave function |ψ(t)> corresponding to the time t is generated; wherein the photovoltaic power station power wave function |ψ(t)> is expressed by the following formula: Among them, A j and θ j denote the complex amplitude and real phase of the photovoltaic power station power wave function |ψ(t)>, θ j ∈[-π,π], the normalization condition of the photovoltaic power station power wave function |ψ(t)> is 3. The method according to claim 1, characterized in that The historical weather information includes the actual weather information corresponding to time t The actual weather information includes the temperature, humidity, pressure and light radiation value corresponding to the time t. Recorded as The historical weather information wave function is generated based on the following method: Based on the preset spatial dimension N and the actual weather information Generate the historical weather information wave function |φ corresponding to time t R (t)>; wherein the historical weather information wave function |φ R (t)> is represented by the following formula: Among them, B j Represents the historical weather information wave function |φ R (t)>, Represents the historical weather information wave function |φ R (t)> is the jth real coefficient vector of dimension 4, and the normalization condition of the historical weather information wave function is 4. The method according to claim 1, characterized in that The historical sky imaging information includes the moment sky imaging corresponding to the moment t, and the historical sky imaging wave function is generated based on the following method: Extract N eigenvalues ​​from the sky image at the time based on the preset spatial dimension N And based on the N eigenvalues, a historical sky imaging wave function |φ corresponding to time t is generated I (t)>; wherein the historical sky imaging wave function |φ I (t)> is represented by the following formula: Among them, C j represents the historical sky imaging wave function |φ I (t)>, the complex amplitude, ω j represents the historical sky imaging wave function |φ I (t)>, the normalization condition of the historical sky imaging wave function is 5. The method according to claim 1, characterized in that The weather forecast information includes the predicted weather information corresponding to time t The predicted weather information includes the temperature, humidity, pressure and light radiation value corresponding to the time t. Recorded as The weather forecast wave function is generated based on the following method: Based on the preset spatial dimension N and the predicted weather information Generate the weather forecast wave function |φ corresponding to time t P (t)>; wherein the weather forecast wave function |φ P (t)> is represented by the following formula: Wherein, Dj represents the weather forecast wave function |φ P The complex amplitude of (t)>, Denotes the weather forecast wave function |φ P (t)> is the jth real coefficient vector of dimension 4, and the normalization condition of the weather forecast wave function is 6. The method according to claim 1, characterized in that The step of constructing a time correlation function according to the historical weather information wave function, the historical sky imaging wave function and the weather forecast wave function comprises: Calculating the outer product of the historical weather information wave function and its Dirac conjugate to obtain a historical weather information density matrix; Calculating the outer product of the historical sky imaging wave function and its Dirac conjugate to obtain a historical sky imaging density matrix; Constructing a historical weather information mixed density matrix based on a preset first proportionality coefficient, the historical weather information density matrix, and the historical sky imaging density matrix; Constructing a first correlation function based on the historical weather information mixed density matrix, the weather forecast wave function and the historical weather information wave function; Constructing a second correlation function based on the historical weather information mixed density matrix, the weather forecast wave function and the historical sky imaging wave function; The time correlation function is constructed based on a preset second proportionality coefficient, the first correlation function and the second correlation function.

7. The method according to claim 1, characterized in that The interaction function is represented by the following formula: Among them, time t′ is earlier than time t, Λ is the preset time cutoff factor, V0 is the interaction strength, and μ is the attenuation factor of the interaction between different time information.

8. The method according to claim 1, characterized in that The power prediction model is based on the Lippman-Schwinger type equation to represent the change of the photovoltaic power station power wave function |ψ(t)> over time. The Lippman-Schwinger type equation is represented by the following formula: Among them, |φ P (t)> is the weather forecast wave function, Λ is the preset time cutoff factor, time t′ is earlier than time t, G(t, t′) is the time correlation function, and V(t, t′) is the interaction function.

9. The method according to claim 8, characterized in that The method further comprises: Substitute the photovoltaic power station power wave function |ψ(t)> into a preset formula to obtain a photovoltaic power station power prediction value, wherein the preset formula is expressed by the following formula: Among them, ψ j (t) is the jth element of the photovoltaic power station power wave function |ψ(t)>, abs represents the absolute value of the calculation result, N is the preset spatial dimension, A j and θ j represent the complex amplitude and real phase of the photovoltaic power station power wave function |ψ(t)> respectively.

10. The method according to claim 1, characterized in that The loss function of the power prediction model is expressed as follows: Where MSE stands for mean square error, n is the number of samples, and y k represents the true value of the kth sample, Represents the predicted value of the kth sample.

Images