New energy storage analysis method and system based on artificial intelligence

Abstract

The invention relates to a new energy storage analysis method and system based on artificial intelligence, and the method comprises the steps: obtaining new energy storage information data and power grid power generation data, carrying out the processing of the energy storage information data, inputting the data into a trained energy storage state model, obtaining the total energy of stored energy, predicting the new energy power generation based on the power grid power generation data, and obtaining the new energy power generation. And finally, inputting the obtained total energy of the stored energy and the energy storage power generation parameters into the trained energy storage analysis model to obtain an energy storage equipment optimization scheduling strategy comprising the optimal charging and discharging time and the charging and discharging power set value. According to the method, energy data resources can be effectively integrated, the utilization efficiency of new energy is improved, energy waste is reduced, the stability and reliability of power grid operation are enhanced, and comprehensive analysis, accurate regulation and control and optimal management of the new energy storage system are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy storage, and particularly relates to a new energy storage analysis method and system based on artificial intelligence. Background Art

[0002] With the development of energy storage technology, new energy storage analysis technology based on artificial intelligence has emerged. As the global demand for sustainable energy surges, the proportion of new energy in the energy system continues to climb. Although new energy sources such as solar energy and wind energy have significant advantages of being clean and renewable, their inherent intermittency and instability pose great challenges to the stable operation of the power grid and the efficient utilization of energy. At the same time, traditional energy storage control technologies are unable to handle the massive and complex data of new energy power grids and energy storage devices, making it difficult to achieve comprehensive analysis, precise control, and optimized management of new energy storage systems, resulting in low utilization efficiency of energy storage devices, energy supply-demand imbalance, and frequent waste phenomena. Summary of the Invention

[0003] Based on this, in view of the above technical problems, it is necessary to provide a new energy storage analysis method and system based on artificial intelligence that can achieve comprehensive analysis, precise control, and optimized management of new energy storage systems.

[0004] In a first aspect, the present application provides a new energy storage analysis method based on artificial intelligence, including:

[0005] Obtain new energy storage information data and power grid power generation data; the energy storage information data includes at least one of power generation amount, energy storage capacity, equipment status data, and energy storage scheduling data; the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data.

[0006] Input the processed energy storage information data into a trained energy storage state model to obtain the total energy storage; predict new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy storage system.

[0007] Input the obtained total energy storage and energy storage power generation parameters into a trained energy storage analysis model to obtain an optimized scheduling strategy for the energy storage device; the optimized scheduling strategy includes the best charge and discharge time and the set value of charge and discharge power.

[0008] In one embodiment, inputting the processed energy storage information data into a trained energy storage state model to obtain the total energy storage includes:

[0009] Construct a data set based on the energy storage information data to obtain a processed energy storage data set.

[0010] Based on historical energy storage information data, a machine learning classifier is used for model training to obtain an energy storage state classifier; the machine learning classifier is at least one of a support vector machine, a decision tree, a neural network, and a random forest.

[0011] The random gradient descent method is used to train each energy storage state classifier to obtain a pre-trained energy storage state model.

[0012] The area under the curve value of each pre-trained energy storage state model is calculated to obtain a trained energy storage state model; the trained energy storage state model is an energy storage state model whose performance meets the preset conditions.

[0013] The data in the processed energy storage data set is input into the trained energy storage state model to obtain the total energy storage.

[0014] In one embodiment, the trained energy storage state model uses the following formula to obtain the total energy storage:

[0015]

[0016] Among them, Z(t) represents the total energy storage of the system at time t, including the weighted sum of the energies of all energy storage devices, R i (t) represents the energy of the i-th energy storage device at time t, n represents the total number of energy storage devices, P i represents the maximum capacity of the i-th energy storage device, θ i represents the efficiency adjustment parameter related to the i-th energy storage device, μ i represents the periodic change amplitude of the efficiency of the i-th energy storage device, ω i is the frequency of change.

[0017] In one embodiment, based on the power grid generation data, the new energy generation is predicted to obtain the energy storage power generation parameters of the new energy storage system, including:

[0018] Each data in the power grid generation data is preprocessed and a data set is constructed to obtain a processed power generation data set.

[0019] Based on the power generation data set, the synthetic minority over-sampling technique is used to balance the data to obtain a balanced data set.

[0020] The trained deep learning model is used to extract features from the balanced data set to obtain corresponding feature information; the feature information includes the operating characteristics of new energy generation equipment, the state characteristics of energy storage devices, and the grid load demand characteristics.

[0021] The least absolute shrinkage and selection operator algorithm is used to assist in verifying the feature information to obtain updated feature information.

[0022] Based on the updated feature information, use a formula to predict new energy power generation and obtain the energy storage power generation parameters of the new energy storage system; the energy storage power generation parameters include charge-discharge power, remaining capacity, and life attenuation parameters.

[0023]

[0024] Among them, M(t) represents the energy storage power generation parameters predicted at time t, and t 0 represents the initial time point, V(x) represents the operating characteristic value of the power generation equipment in the feature information at time x, N(V, W, x) represents a function that combines new energy grid data and the state characteristics of distributed energy storage devices, r represents a normalization factor, exp represents an exponential function, ρ and η represent the average and standard deviation of the grid load demand characteristic values in the feature information, represents the Fourier transform of the distributed energy storage device data, and ω is the frequency variable.

[0025] In one embodiment, input the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain the optimized scheduling strategy for the energy storage device, including:

[0026] Based on the grid power generation data, use a formula to calculate the impact of geographical environment factors on new energy output and obtain the renewable energy amount of the new energy storage.

[0027] According to the fuzzy comprehensive evaluation method, label the obtained renewable energy amount, total energy storage, and energy storage power generation parameters to obtain the corresponding characteristic category parameters.

[0028] Use a formula to construct the energy storage analysis model and continuously adjust the weights and biases of the model through the backpropagation algorithm to obtain the trained energy storage analysis model.

[0029] The formula for constructing the energy storage analysis model is as follows:

[0030]

[0031] Among them, Q i (t + 1) represents the energy state of the i-th energy storage device at time t + 1, and Q i (t) represents the energy state of the i-th energy storage device at time t, Δt represents the time step, Z i (t) represents the total system energy storage of the i-th energy storage device at time t, M i (t) represents the energy storage power generation parameters of the i-th energy storage device predicted at time t, Y i (t) represents the renewable energy amount of the i-th energy storage device at time t, ω 1 、ω 2 and ω 3respectively represent the corresponding weights, γ represents the amplitude of demand change, and ω represents the frequency of demand change.

[0032] Input the obtained feature category parameters into the trained energy storage analysis model to obtain the model output value Q i (t + 1).

[0033] Calculate the model output value Q i (t + 1) for each energy storage device, and compare it with the current energy state Q i (t) to obtain the predicted energy state of each energy storage device at the next time node.

[0034] Determine the charge and discharge time of the energy storage device according to the predicted energy state to obtain the optimized scheduling strategy of the energy storage device.

[0035] In one embodiment, based on the power generation data of the power grid, use the formula to calculate the influence of geographical environment factors on new energy output, and obtain the renewable energy quantity of new energy storage, including:

[0036] Use the following formula to calculate the renewable energy quantity:

[0037]

[0038] H i (t, D(t), K(t), ΔT) = σ * exp(-ε * ΔT) + γ * D(t) - ν * log(1 - K(t))

[0039] Among them, Y(t) represents the renewable energy quantity at time t, that is, the new energy output after environmental factor adjustment, v is the number of new energy types, δ represents the parameter affecting the adjustment time, and H i (t, D(t), K(t), ΔT) represents the environmental factor adjustment coefficient of the i-th new energy, D(t) represents the weather change intensity, K(t) represents the influence of geographical location change, ΔT represents the difference from the ideal temperature, and σ, ε, γ, and ν represent the adjustment coefficients, and the value range is [0, 1].

[0040] In one embodiment, after obtaining the optimized scheduling strategy of the energy storage device, it further includes:

[0041] Simulate and implement the energy storage control plan for the optimized scheduling strategy and evaluate the effect to calculate the optimized evaluation value.

[0042]

[0043] Among them, C(i) represents the optimized evaluation value corresponding to the i-th optimized scheduling strategy, and α i represents the weight factor of the i-th energy storage device, and M i(t) represents the energy storage power generation parameter of the i-th energy storage device at time t, Y i (t) represents the renewable energy quantity of the i-th energy storage device at time t, τ represents the adjustment parameter for balancing the influence of historical data, C(t) represents the regulation data at time t, Z(t) represents the total system energy storage of the energy storage device at time t, and Y(t) represents the renewable energy quantity of the energy storage device at time t.

[0044] Compare the obtained optimized evaluation value C(i) with the set threshold ξ to obtain a comparison result.

[0045] Among them, if C(i) > ξ, the comparison result indicates that the optimized scheduling strategy can be implemented, then implement the optimized scheduling strategy and monitor and record the new energy power grid and energy storage devices in real time to obtain a regulation optimization report.

[0046] If C(i) ≤ ξ, the comparison result indicates that the optimized scheduling strategy cannot be implemented, then re-iterate the new energy power generation and formulate the evaluation effect of the optimized scheduling strategy until ξ and generate an iteration record to obtain a regulation optimization report.

[0047] In a second aspect, the present application also provides a new energy energy storage analysis system based on artificial intelligence. The system includes:

[0048] A data acquisition module for acquiring new energy energy storage information data and power grid power generation data; the energy storage information data includes at least one of power generation quantity, energy storage capacity, device status data, and energy storage scheduling data; the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data.

[0049] A data processing module for inputting the processed energy storage information data into a trained energy storage state model to obtain the total energy storage; and also for predicting new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameter of the new energy energy storage system.

[0050] An energy storage analysis module for inputting the obtained total energy storage and energy storage power generation parameter into a trained energy storage analysis model to obtain an optimized scheduling strategy for the energy storage device; the optimized scheduling strategy includes the best charge and discharge time and the set value of charge and discharge power.

[0051] In a third aspect, the present application also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method as described above.

[0052] In a fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method as described above.

[0053] The above-mentioned new energy energy storage analysis method and system based on artificial intelligence obtain new energy energy storage information data including at least one of power generation amount, energy storage capacity, equipment status data, and energy storage scheduling data, as well as power grid power generation data covering at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data. After processing the energy storage information data, it is input into the trained energy storage status model to obtain the total energy storage, and based on the power grid power generation data, new energy power generation is predicted to obtain the energy storage power generation parameters of the new energy energy storage system. Finally, the obtained total energy storage and energy storage power generation parameters are input into the trained energy storage analysis model to obtain an optimized scheduling strategy for the energy storage device including the best charge and discharge time and the set value of the charge and discharge power. The above steps effectively integrate energy data resources, not only improving the utilization efficiency of new energy, reducing energy waste, but also enhancing the stability and reliability of the power grid operation, realizing the comprehensive analysis, precise control, and optimized management of the new energy energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0055] Figure 1 It is a flowchart of a new energy energy storage analysis method based on artificial intelligence provided by an embodiment of the present invention;

[0056] Figure 2 It is a flowchart of inputting the processed energy storage information data into the trained energy storage status model to obtain the total energy storage;

[0057] Figure 3 It is a flowchart of predicting new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy energy storage system;

[0058] Figure 4 It is a flowchart of inputting the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain an optimized scheduling strategy for the energy storage device;

[0059] Figure 5 It is a structural block diagram of a new energy energy storage analysis system based on artificial intelligence provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] First, the implementation environment of the embodiment of the present application is described. Exemplarily, the implementation environment includes data acquisition hardware, data processing equipment and data storage equipment.

[0062] In the new energy storage analysis method and system based on artificial intelligence, the data acquisition hardware uses wired or wireless communication to send various data of the new energy storage system collected in real time to the data processing equipment. After receiving the data, the data processing equipment performs pre-processing such as cleaning and feature extraction, and then runs the artificial intelligence algorithm for in-depth analysis. The processing equipment transmits the analysis results and some key original data to the data storage device through the connection methods such as the internal local area network, storage area network or network attached storage.

[0063] Data acquisition hardware is responsible for collecting various key data during the operation of the new energy storage system. Among them, sensors are an important part of data acquisition hardware. Temperature sensors capture temperature changes of batteries and energy storage devices in real time; voltage and current sensors accurately measure the charge and discharge status of batteries. In addition, smart meters can accurately measure the input and output of electrical energy and record electricity consumption at different times. Data acquisition hardware devices are distributed at various key nodes of the energy storage system, and the collected raw data is stably and accurately transmitted to subsequent data processing links through wired or wireless communication methods.

[0064] Data processing equipment receives a large amount of raw data transmitted from data acquisition hardware, and performs in-depth processing and analysis on the data. Common data processing equipment includes servers, GPU clusters, etc. First, the raw data is cleaned and preprocessed to remove noise and outliers to improve data quality; then feature extraction is performed to extract representative and analytically valuable features from massive data; finally, artificial intelligence algorithms are used to model and predict the data to explore the laws and potential information behind the data.

[0065] Data storage devices are responsible for long-term and secure storage of processed data and some original data. Common data storage devices include disk arrays, tape libraries, and cloud storage. Disk arrays can quickly respond to the storage needs of data processing equipment; tape libraries are suitable for long-term data archiving; cloud storage has flexible scalability and convenient access methods. Data storage devices are used for system performance evaluation, fault diagnosis, and algorithm optimization.

[0066] In combination with the above implementation environment, the application scenario of the embodiment of the present application is explained.

[0067] A new energy energy storage analysis method and system based on artificial intelligence provided by an embodiment of the present application first performs data collection, uses various sensors and smart meters to collect multi-source data such as battery status, environmental parameters, and power flow in the new energy energy storage system, and then transmits the data to a data processing device. The data processing device performs preprocessing operations such as cleaning and noise reduction on the original data, then uses feature engineering to extract key features, and subsequently uses regression algorithms of machine learning to predict the energy storage capacity and classification algorithms to identify the fault types. The analysis results obtained are not only used to adjust the charge and discharge strategies of the energy storage system in real time and optimize energy management, but also stored in a data storage device for subsequent review, model iteration optimization, and long-term system performance evaluation, so as to achieve the efficient, stable, and safe operation of the new energy energy storage system. Exemplarily, a new energy energy storage analysis method and system based on artificial intelligence provided by an embodiment of the present application can be applied to at least one of the following scenarios including but not limited to the following scenarios.

[0068] First, the new energy energy storage analysis method and system based on artificial intelligence are applied to the real-time status monitoring and fault warning of the energy storage system. For example, when the temperature of the battery suddenly rises and the voltage shows abnormal fluctuations, the data processing device can issue an alarm in a timely manner. The data storage device is responsible for storing the collected original data and the processed analysis results for a long time, providing detailed data support for subsequent fault diagnosis and analysis.

[0069] Second, the new energy energy storage analysis method and system based on artificial intelligence are applied to the energy optimization management of the new energy energy storage system. For example, through a reinforcement learning algorithm, according to the real-time situation of new energy power generation, the load demand of the power grid, and the remaining capacity of the energy storage system, the charge and discharge strategies of the energy storage system are dynamically adjusted. At the same time, a prediction model is used to predict the power of new energy power generation and the load of the power grid in the short term and long term, and a more scientific and reasonable energy management plan is formulated in advance. The data storage device stores a large amount of historical power generation data, load data, and energy storage system operation data.

[0070] In one embodiment, as Figure 1 shown, the present application provides a new energy energy storage analysis method based on artificial intelligence, which may include the following steps:

[0071] Step S101, obtain new energy energy storage information data and power grid power generation data; the energy storage information data includes at least one of power generation amount, energy storage capacity, device status data, and energy storage scheduling data; the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data.

[0072] Specifically, the energy storage information data covers multiple key aspects. The power generation reflects the actual output of new energy converted into electrical energy. The energy storage capacity embodies the ability of the energy storage device to store electrical energy. The device status data shows the current operating health status of the energy storage device. The energy storage scheduling data records the previous energy storage allocation strategies and implementation situations. Similarly, the grid power generation data is also crucial. The real-time power generation data tracks the current power generation power of the grid in real time. The power demand data clarifies the electricity consumption load at different times. The actual energy storage data reflects the real-time status of the energy storage devices in the grid. The environmental data includes natural factors such as light intensity, wind speed, and temperature that affect new energy power generation. The geographical location data involves the specific location information of the power generation and energy storage devices.

[0073] Step S102, input the processed energy storage information data into the trained energy storage status model to obtain the total energy storage; predict the new energy power generation based on the grid power generation data to obtain the energy storage power generation parameters of the new energy storage system.

[0074] Preprocess the energy storage information data, including data cleaning to remove outliers and incorrect data, data standardization to unify the data scale, and data feature extraction to mine key information. Then input the processed energy storage information data into the energy storage status model that has been trained in advance using a large amount of historical data and advanced algorithms. This model accurately calculates and outputs the total energy storage by comprehensively considering key factors such as power generation, energy storage capacity, device status data, and energy storage scheduling data in the input data. At the same time, based on the grid power generation data, use technologies such as time series analysis and machine learning prediction algorithms to accurately predict the future trend of new energy power generation. Further derive the energy storage power generation parameters of the new energy storage system, including key parameters such as the predicted power generation values at different future times and the range of charge and discharge power changes of the energy storage device.

[0075] Step S103, input the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain the optimized scheduling strategy for the energy storage device; the optimized scheduling strategy includes the best charge and discharge times and the set values of charge and discharge power.

[0076] After obtaining the total energy storage and energy storage power generation parameters, the data is input into the trained energy storage analysis model. This model deeply analyzes the total energy storage and energy storage power generation parameters, and through comprehensive consideration of various factors such as the real-time operation status of the power grid, price fluctuations in the power market, and users' electricity consumption habits, accurately plans the optimized scheduling strategy for energy storage devices. Among them, the best charge and discharge time ensures that the energy storage device stores electrical energy in a timely manner when new energy generation is excessive, and accurately releases electrical energy during peak power demand or insufficient new energy generation, realizing the efficient utilization of energy and the balance between supply and demand; while the set value of the charge and discharge power is scientifically and reasonably set based on factors such as the performance characteristics of the energy storage device, the carrying capacity of the power grid, and energy costs, to avoid problems such as energy waste, increased equipment loss, or inability to meet electricity demand due to too large or too small charge and discharge power, ultimately realizing the optimized scheduling of energy storage devices and improving the operation efficiency and economic benefits of the entire new energy storage system.

[0077] The above new energy storage analysis method based on artificial intelligence obtains new energy storage information data including at least one of power generation, energy storage capacity, device status data, and energy storage scheduling data, as well as power grid power generation data covering at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data. After processing the energy storage information data, it is input into the trained energy storage status model to obtain the total energy storage, and based on the power grid power generation data, new energy generation is predicted to obtain the energy storage power generation parameters of the new energy storage system. Finally, the obtained total energy storage and energy storage power generation parameters are input into the trained energy storage analysis model to obtain the optimized scheduling strategy for energy storage devices including the best charge and discharge time and the set value of the charge and discharge power. The above steps effectively integrate energy data resources, not only improving the utilization efficiency of new energy, reducing energy waste, but also enhancing the stability and reliability of power grid operation, realizing the comprehensive analysis, precise control, and optimized management of the new energy storage system.

[0078] In one embodiment, as Figure 2 shown, inputting the processed energy storage information data into the trained energy storage status model to obtain the total energy storage may include the following steps:

[0079] Step S201, constructing a data set based on the energy storage information data to obtain the processed energy storage data set.

[0080] Step S202, using a machine learning classifier to train the model based on historical energy storage information data to obtain an energy storage status classifier; the machine learning classifier is at least one of a support vector machine, a decision tree, a neural network, and a random forest.

[0081] Step S203, using the stochastic gradient descent method to train each energy storage status classifier to obtain a pre-trained energy storage status model.

[0082] Step S204: Calculate the area under the curve value for each pre-trained energy storage state model to obtain a trained energy storage state model; the trained energy storage state model is an energy storage state model whose performance meets the preset conditions.

[0083] Step S205: Input the data in the processed energy storage data set into the trained energy storage state model to obtain the total energy storage.

[0084] Specifically, construct and process a data set based on the energy storage information data to obtain a processed energy storage data set. Then, use at least one machine learning classifier among support vector machine, decision tree, neural network, and random forest with historical energy storage information data to carry out model training, so as to obtain an energy storage state classifier. Subsequently, use the stochastic gradient descent method to train each energy storage state classifier to obtain a pre-trained energy storage state model. After that, calculate the area under the curve value for each pre-trained energy storage state model, screen out the models whose performance meets the preset conditions, and use them as the trained energy storage state models. Finally, input the data in the processed energy storage data set into the trained energy storage state model to obtain the total energy storage.

[0085] This embodiment realizes the accurate quantification of the energy state of the energy storage system, provides a key basis for the reasonable scheduling and efficient management of new energy energy storage, helps to improve energy utilization efficiency, ensure the stable operation of the power grid, and reduce the loss of energy storage equipment.

[0086] In one of the embodiments, the trained energy storage state model can use the following formula to obtain the total energy storage:

[0087]

[0088] Among them, Z(t) represents the total energy storage of the system at time t, including the weighted sum of the energies of all energy storage devices, R i (t) represents the energy of the i-th energy storage device at time t, n represents the total number of energy storage devices, P i represents the maximum capacity of the i-th energy storage device, θ i represents the efficiency adjustment parameter related to the i-th energy storage device, μ i represents the periodic change amplitude of the efficiency of the i-th energy storage device, ω i is the frequency of change.

[0089] This embodiment comprehensively considers multiple key factors and can accurately calculate the total energy storage. By considering the current energy, maximum capacity, and efficiency adjustment parameters of each energy storage device, etc., it ensures that the calculation results are more in line with the actual energy storage situation; incorporating the periodic change amplitude and frequency of device efficiency into the calculation enables the calculation results to dynamically reflect the impact of the efficiency fluctuations of energy storage devices at different times on the total energy, providing an accurate basis for the scientific management and efficient scheduling of energy storage systems, helping to optimize energy distribution, improve energy utilization efficiency, and ensure the stability of energy supply.

[0090] In one embodiment, as Figure 3 shown, predicting new energy power generation based on grid power generation data to obtain the energy storage power generation parameters of a new energy storage system may include the following steps:

[0091] Step S301, preprocess each data in the grid power generation data and construct a data set to obtain a processed power generation data set.

[0092] Step S302, balance the data based on the power generation data set using the Synthetic Minority Over-sampling Technique (SMOTE) to obtain a balanced data set.

[0093] Step S303, use the trained deep learning model to extract features from the balanced data set to obtain the corresponding feature information; the feature information includes the operating characteristics of new energy power generation equipment, the state characteristics of energy storage devices, and the grid load demand characteristics.

[0094] The operating characteristics of new energy power generation equipment include the real-time power generation power of the power generation equipment, the fluctuation of power generation efficiency, and the cumulative operating duration of the equipment; the state characteristics of energy storage devices include the remaining power of the energy storage device, the changes in charge and discharge current and voltage, and the change in the internal resistance of the battery; the grid load demand characteristics include the peak and valley values of power consumption at different times, and the specific time periods of peak and off-peak electricity consumption.

[0095] Step S304, use the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm to assist in verifying the feature information to obtain updated feature information.

[0096] Step S305, predict new energy power generation based on the updated feature information using a formula to obtain the energy storage power generation parameters of the new energy storage system; the energy storage power generation parameters include charge and discharge power, remaining capacity, and life attenuation parameters.

[0097]

[0098] Among them, M(t) represents the energy storage power generation parameters predicted at time t, t 0Let \(t_0\) represent the initial time point, \(V(x)\) represent the operating characteristic value of the power generation equipment in the characteristic information at time \(x\), \(N(V, W, x)\) represent the function of the composite new energy grid data and the state characteristics of the distributed energy storage equipment, \(r\) represent the normalization factor, exp represent the exponential function, and \(\rho\) and \(\eta\) represent the mean and standard deviation of the grid load demand characteristic values in the characteristic information. Represents the Fourier transform of the distributed energy storage equipment data, where \(\omega\) is the frequency variable.

[0099] Preprocess the grid power generation data including real-time power generation data, power demand data, etc., construct a data set through data cleaning, integration and transformation, so as to obtain a processed power generation data set. Then use the Synthetic Minority Over-sampling Technique (SMOTE) to process the power generation data set to balance the data distribution and obtain a balanced data set. Then use the trained deep learning model to extract feature information from the balanced data set, obtaining corresponding feature information including the operating characteristics of new energy power generation equipment, the state characteristics of energy storage equipment, and the grid load demand characteristics. Use the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm to assist in verifying the feature information, further screening and optimizing it to obtain updated feature information. Finally, based on the updated feature information, use the formula to predict new energy power generation, and obtain the energy storage power generation parameters of the new energy storage system, such as charge and discharge power, remaining capacity, and life decay parameters, etc.

[0100] In this embodiment, by preprocessing the grid power generation data and constructing a data set, data noise and inconsistencies are removed, and then the Synthetic Minority Over-sampling Technique (SMOTE) is used to balance the data to obtain a balanced data set. Input it into the trained deep learning model to extract multi-dimensional feature information, comprehensively reflecting the operating conditions of the new energy power generation system. The Least Absolute Shrinkage and Selection Operator (LASSO) algorithm further screens and verifies the features to ensure the effectiveness and importance of the features, obtaining more representative updated feature information. Finally, based on the updated information, use the formula to predict new energy power generation, accurately obtaining the energy storage power generation parameters, which can help the operation and maintenance personnel understand the charge and discharge power, remaining capacity, and life decay of the energy storage equipment in advance, contribute to reasonably arranging the charge and discharge plan of the energy storage equipment, improve the utilization efficiency of new energy, ensure the stable operation of the grid, reduce the loss of energy storage equipment, extend its service life, and promote the efficient and sustainable development of the new energy storage system.

[0101] In one of the embodiments, as Figure 4 shown, input the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain the optimized scheduling strategy of the energy storage equipment, which may include the following steps:

[0102] Step S401, calculate the influence of geographical environment factors on new energy output based on the grid power generation data using the formula to obtain the renewable energy amount of the new energy storage.

[0103] Step S402: Perform data annotation on the obtained renewable energy quantity, total energy storage, and energy storage power generation parameters according to the fuzzy comprehensive evaluation method to obtain corresponding characteristic category parameters.

[0104] Step S403: Construct the energy storage analysis model using the formula and continuously adjust the weights and biases of the model through the backpropagation algorithm to obtain a trained energy storage analysis model.

[0105] The formula for constructing the energy storage analysis model is expressed as follows:

[0106]

[0107] Where, Q i (t + 1) represents the energy state of the i-th energy storage device at time t + 1, Q i (t) represents the energy state of the i-th energy storage device at time t, Δt represents the time step, Z i (t) represents the total system energy storage of the i-th energy storage device at time t, M i (t) represents the energy storage power generation parameter of the i-th energy storage device predicted at time t, Y i (t) represents the renewable energy quantity of the i-th energy storage device at time t, ω 1 、ω 2 and ω 3 respectively represent the corresponding weights, γ represents the amplitude of demand change, and ω represents the frequency of demand change.

[0108] Step S404: Input the obtained characteristic category parameters into the trained energy storage analysis model to obtain the model output value Q i (t + 1).

[0109] Step S405: Calculate the model output value Q i (t + 1) for each energy storage device and compare it with the current energy state Q i (t) to obtain the predicted energy state of each energy storage device at the next time node.

[0110] Step S406: Determine the charging and discharging time of the energy storage device according to the predicted energy state to obtain an optimized scheduling strategy for the energy storage device.

[0111] First, based on the power generation data of the power grid, use a specific formula to calculate the impact of geographical environment factors on new energy output, and then obtain the renewable energy quantity of new energy storage; then use the fuzzy comprehensive evaluation method to label the obtained renewable energy quantity, total energy storage, and energy storage power generation parameters to obtain the corresponding characteristic category parameters; then construct an energy storage analysis model according to the formula, and continuously adjust the weights and biases of the model through the backpropagation algorithm to obtain a trained energy storage analysis model; then input the obtained characteristic category parameters into the trained energy storage analysis model to obtain the model output value; then calculate the model output value for each energy storage device and compare it with the current energy state to obtain the predicted energy state of each energy storage device at the next time node; finally, determine the charging and discharging time of the energy storage device according to the predicted energy state, and finally obtain the optimized scheduling strategy of the energy storage device.

[0112] In this embodiment, the charging and discharging time is determined based on the predicted energy state, and the obtained optimized scheduling strategy can scientifically and reasonably arrange the operation of the energy storage device, greatly improve the energy utilization efficiency, effectively reduce the energy loss and cost, enhance the stability and reliability of the power grid operation, and strongly promote the development of the new energy storage system in the direction of high efficiency and sustainability.

[0113] In one of the embodiments, based on the power generation data of the power grid, using a formula to calculate the impact of geographical environment factors on new energy output, and obtaining the renewable energy quantity of new energy storage may include the following steps:

[0114] Use the following formula to calculate the renewable energy quantity:

[0115]

[0116] H i (t,D(t),K(t),ΔT)=σ*exp(-ε*ΔT)+γ*D(t)-ν*log(1-K(t))

[0117] Among them, Y(t) represents the renewable energy quantity at time t, that is, the new energy output after environmental factor adjustment, n is the number of new energy types, δ represents the parameter that adjusts the time impact, H i (t,D(t),K(t),ΔT) represents the environmental factor adjustment coefficient of the i-th new energy, D(t) represents the weather change intensity, K(t) represents the impact of geographical location change, ΔT represents the difference from the ideal temperature, and σ, ε, γ, and ν represent the adjustment coefficients, and the value range is [0,1].

[0118] This embodiment comprehensively considers the impacts of various geographical environment factors such as the intensity of weather changes, geographical location changes, and temperature differences on the output of different types of new energy, quantifies the impacts through precise mathematical formulas, and thus more accurately obtains the renewable energy quantity. It helps to have a clearer understanding of the energy reserve situation of new energy storage, provides a scientific basis for subsequent energy storage scheduling, energy planning, etc., can effectively improve the utilization efficiency of new energy, promote the reasonable allocation and efficient utilization of energy, and ensure the stability and sustainability of energy supply.

[0119] In one of the embodiments, after obtaining the optimized scheduling strategy for the energy storage device, the following steps may further be included:

[0120] Step S501, simulate and implement the energy storage regulation plan for the optimized scheduling strategy and evaluate the effect, and calculate to obtain an optimized evaluation value.

[0121]

[0122] Among them, C(i) represents the optimized evaluation value corresponding to the i-th optimized scheduling strategy, α i represents the weight factor of the i-th energy storage device, M i (t) represents the energy storage power generation parameter of the i-th energy storage device predicted at time t, Y i (t) represents the renewable energy quantity of the i-th energy storage device at time t, τ represents the adjustment parameter for balancing the influence of historical data, C(t) represents the regulation data at time t, Z(t) represents the total system energy storage of the energy storage device at time t, and Y(t) represents the renewable energy quantity of the energy storage device at time t.

[0123] Step S502, compare the obtained optimized evaluation value C(i) with the set threshold ξ to obtain a comparison result.

[0124] Among them, if C(i)>ξ, the comparison result indicates that the optimized scheduling strategy is feasible, then implement the optimized scheduling strategy and monitor and record the new energy grid and energy storage device in real time to obtain a regulation optimization report.

[0125] If C(i)≤ξ, the comparison result indicates that the optimized scheduling strategy is not feasible, then re-iterate the new energy power generation and formulate and evaluate the effect of the optimized scheduling strategy until ξ and generate an iteration record to obtain a regulation optimization report.

[0126] First, simulate and implement the energy storage regulation plan for the optimized scheduling strategy and evaluate the effect, calculate the optimized evaluation value through the formula. Then compare the obtained optimized evaluation value with the set threshold to obtain a comparison result, and at the same time generate an iteration record, and further obtain a regulation optimization report.

[0127] By simulating the implementation and evaluation of the optimized scheduling strategy, the feasibility and effectiveness of the strategy can be tested in advance, avoiding the risks and losses brought by blind implementation. Calculate the optimized evaluation value and compare it with the set threshold to provide a quantitative basis for decision-making, ensuring that the finally implemented scheduling strategy can achieve the expected effect. When the strategy is infeasible, re-iterate and optimize to ensure the scientific nature and adaptability of the scheduling strategy, which helps to improve the operation efficiency and stability of the new energy energy storage system, realize the reasonable allocation and efficient utilization of energy. At the same time, the generated regulation optimization report and iteration record can provide valuable references for subsequent energy management and decision-making, promoting the sustainable development of the new energy industry.

[0128] In one embodiment, as Figure 5 shown, the present application also provides an artificial intelligence-based new energy energy storage analysis system, which includes:

[0129] A data acquisition module 601, configured to obtain new energy energy storage information data and grid power generation data; the energy storage information data includes at least one of power generation amount, energy storage capacity, device status data, and energy storage scheduling data; the grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data.

[0130] A data processing module 602, configured to input the processed energy storage information data into the trained energy storage state model to obtain the total energy storage; and is also configured to predict new energy power generation based on the grid power generation data to obtain the energy storage power generation parameters of the new energy energy storage system.

[0131] An energy storage analysis module 603, configured to input the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain an optimized scheduling strategy for the energy storage device; the optimized scheduling strategy includes the best charge and discharge time and the set value of the charge and discharge power.

[0132] The above-mentioned new energy energy storage analysis system based on artificial intelligence, in which the data acquisition module is responsible for obtaining new energy energy storage information data and power grid power generation data. The energy storage information data covers at least one of power generation, energy storage capacity, equipment status data, and energy storage scheduling data, while the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data, and geographical location data. The acquired data is transmitted to the data processing module, which inputs the processed energy storage information data into the trained energy storage status model to obtain the total energy storage, and at the same time predicts new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy energy storage system. Finally, the energy storage analysis module inputs the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model, thereby obtaining an optimized scheduling strategy for the energy storage device that includes the optimal charge and discharge time and the set value of the charge and discharge power. The above steps effectively integrate energy data resources, not only improving the utilization efficiency of new energy, reducing energy waste, but also enhancing the stability and reliability of power grid operation, and realizing comprehensive analysis, precise control, and optimized management of the new energy energy storage system.

[0133] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0134] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of a new energy energy storage analysis method and system based on artificial intelligence as described above.

[0135] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the above method embodiments.

[0136] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the descriptions of the method embodiments. The device embodiments described above are only illustrative. The components described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present disclosure solution. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0137] The above embodiments only represent several implementation manners of the embodiments of the present application. The descriptions are relatively specific and detailed, but should not be construed as a limitation on the patent scope of the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. A new energy storage analysis method based on artificial intelligence, characterized in that: The method comprises: Acquire new energy storage information data and power grid power generation data; the energy storage information data includes at least one of power generation, energy storage capacity, equipment status data and energy storage scheduling data; the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data and geographic location data; Input the processed energy storage information data into the trained energy storage state model to obtain the total energy storage; predict the new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy storage system; The obtained total energy storage and energy storage power generation parameters are input into a trained energy storage analysis model to obtain an optimal scheduling strategy for the energy storage device; the optimal scheduling strategy includes an optimal charging and discharging time and a charging and discharging power setting value.

2. The method according to claim 1, characterized in that The step of inputting the processed energy storage information data into the trained energy storage state model to obtain the total energy storage energy includes: Building a data set based on the energy storage information data to obtain a processed energy storage data set; Based on the historical energy storage information data, a machine learning classifier is used to perform model training to obtain an energy storage state classifier; the machine learning classifier is at least one of a support vector machine, a decision tree, a neural network, and a random forest; Using a stochastic gradient descent method to train each of the energy storage state classifiers to obtain a pre-trained energy storage state model; Calculating the area under the curve of each pre-trained energy storage state model to obtain a trained energy storage state model; the trained energy storage state model is an energy storage state model whose performance meets preset conditions; The processed data in the energy storage data set is input into the trained energy storage state model to obtain the total energy storage energy.

3. The method according to claim 2, characterized in that The trained energy storage state model uses the following formula to obtain the total energy storage: Where Z(t) represents the total energy storage of the system at time t, including the weighted sum of the energy of all energy storage devices, R i (t) represents the energy of the i-th energy storage device at time t, n represents the total number of energy storage devices, P i represents the maximum capacity of the i-th energy storage device, θ i represents the efficiency adjustment parameter related to the i-th energy storage device, μ i represents the periodic variation of the efficiency of the i-th energy storage device, ω i The frequency of change.

4. The method according to claim 1, characterized in that: The method of predicting the new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy storage system includes: Preprocessing each data in the power grid power generation data and constructing a data set to obtain a processed power generation data set; Based on the power generation data set, balancing the data using a synthetic minority oversampling technique to obtain a balanced data set; Using the trained deep learning model to extract features from the balance data set, and obtain corresponding feature information; the feature information includes operating features of new energy power generation equipment, state features of energy storage equipment, and grid load demand features; Using the least absolute shrinkage and selection operator algorithm to assist in verifying the feature information, and obtain updated feature information; Based on the updated characteristic information, a formula is used to predict new energy power generation, and energy storage power generation parameters of the new energy storage system are obtained; the energy storage power generation parameters include charge and discharge power, remaining capacity and life attenuation parameters; Wherein, M(t) represents the predicted energy storage power generation parameters at time t, t0 represents the initial time point, V(x) represents the operating characteristic value of the power generation equipment in the characteristic information at time x, N(V,W,x) represents the function of the composite new energy grid data and the state characteristics of the distributed energy storage equipment, r represents the normalization factor, exp represents the exponential function, ρ and η represent the mean and standard deviation of the grid load demand characteristic value in the characteristic information, It represents the Fourier change of distributed energy storage device data, and ω is the frequency variable.

5. The method according to claim 1, characterized in that: The obtained total energy storage and energy storage power generation parameters are input into a trained energy storage analysis model to obtain an optimal scheduling strategy for energy storage equipment, including: Based on the power generation data of the power grid, the influence of geographical environmental factors on the output of new energy is calculated using a formula to obtain the amount of renewable energy stored in the new energy source; The obtained renewable energy amount, total energy storage and energy storage power generation parameters are labeled according to the fuzzy comprehensive evaluation method to obtain corresponding characteristic category parameters; The energy storage analysis model is constructed using the formula and the weight and bias of the model are continuously adjusted through the back propagation algorithm to obtain the trained energy storage analysis model; The formula for constructing the energy storage analysis model is expressed as follows: Among them, Q i (t+1) represents the energy state of the i-th energy storage device at time t+1, Q i (t) represents the energy state of the i-th energy storage device at time t, Δt represents the time step, and Z i (t) represents the total system energy storage energy of the i-th energy storage device at time t, M i (t) represents the predicted energy storage power generation parameters of the i-th energy storage device at time t, Y i (t) represents the amount of renewable energy of the i-th energy storage device at time t, ω1, ω2 and ω3 represent the corresponding weights, γ represents the amplitude of demand change, and ω represents the frequency of demand change; The obtained characteristic category parameters are input into the trained energy storage analysis model to obtain the model output value Q i (t+1); Calculate the model output value Q for each energy storage device i (t+1), and the current energy state Q i (t) compare and obtain the predicted energy state of each energy storage device at the next time node; The charging and discharging time of the energy storage device is determined according to the predicted energy state, and an optimized scheduling strategy for the energy storage device is obtained.

6. The method according to claim 5, characterized in that The method of calculating the influence of geographical environmental factors on the output of new energy by using a formula based on the power grid power generation data to obtain the amount of renewable energy stored in the new energy source includes: The amount of renewable energy is calculated using the following formula: H i (t,D(t),K(t),ΔT)=σ*exp(-ε*ΔT)+γ*D(t)-ν*log(1-K(t)) Where Y(t) represents the amount of renewable energy at time t, that is, the output of new energy after adjusting for environmental factors, n is the number of new energy types, δ represents the parameter affecting the adjustment time, and H i (t, D(t), K(t), ΔT) represents the environmental factor adjustment coefficient of the i-th new energy source, D(t) represents the intensity of weather change, K(t) represents the impact of geographical location change, ΔT represents the difference from the ideal temperature, σ, ε, γ and ν represent adjustment coefficients, and the value range is [0,1].

7. The method according to claim 1, characterized in that After obtaining the optimal scheduling strategy for the energy storage device, the method further includes: Simulate the implementation of the energy storage control scheme for the optimization scheduling strategy and evaluate the effect, and calculate the optimization evaluation value; Among them, C(i) represents the optimization evaluation value corresponding to the i-th optimization scheduling strategy, α i represents the weight factor of the i-th energy storage device, M i (t) represents the predicted energy storage power generation parameters of the i-th energy storage device at time t, Y i (t) represents the renewable energy of the i-th energy storage device at time t, τ represents the adjustment parameter affected by the balance historical data, C(t) represents the control data at time t, Z(t) represents the total system energy storage energy of the energy storage device at time t, and Y(t) represents the renewable energy of the energy storage device at time t; Compare the obtained optimized evaluation value C(i) with the set threshold ξ to obtain a comparison result; Wherein, if C(i)>ξ, the comparison result indicates that the optimization scheduling strategy is feasible, then the optimization scheduling strategy is implemented and the new energy grid and energy storage equipment are monitored and recorded in real time to obtain a control optimization report; If C(i)≤ξ, the comparison result indicates that the optimization scheduling strategy is not feasible, then the new energy power generation is re-iterated and the optimization scheduling strategy is formulated to evaluate the effect until ξ and an iteration record is generated to obtain the regulation optimization report.

8. A new energy storage analysis system based on artificial intelligence, characterized in that: The system comprises: A data acquisition module, used to obtain new energy storage information data and power grid power generation data; the energy storage information data includes at least one of power generation, energy storage capacity, equipment status data and energy storage scheduling data; the power grid power generation data includes at least one of real-time power generation data, power demand data, actual energy storage data, environmental data and geographic location data; A data processing module is used to input the processed energy storage information data into the trained energy storage state model to obtain the total energy storage; it is also used to predict the new energy power generation based on the power grid power generation data to obtain the energy storage power generation parameters of the new energy storage system; The energy storage analysis module is used to input the obtained total energy storage and energy storage power generation parameters into the trained energy storage analysis model to obtain an optimal scheduling strategy for the energy storage equipment; the optimal scheduling strategy includes an optimal charging and discharging time and a charging and discharging power setting value.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images