A management method and system for energy storage system based on environmental prediction

Abstract

The present invention discloses a management method and system for an energy storage system based on environmental prediction, which belongs to the field of energy storage management technology, including: obtaining serial environmental data and performing environmental prediction results; obtaining energy demand data, and obtaining energy demand prediction results based on the energy demand data and environmental prediction results; then predicting the amount of storage energy according to the environmental prediction results to obtain the amount of storage energy prediction results; and finally scheduling and managing the energy storage system and the power supply system according to the energy demand prediction results and the amount of storage energy prediction results. The present invention predicts the environment in a future period of time, and then predicts the energy demand and storage energy in combination with the environmental prediction results, so as to timely understand the energy demand and storage energy in the future time period, plan energy scheduling in advance, improve energy utilization efficiency, enhance the stability and reliability of the power grid, and promote the large-scale application of renewable energy.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage management, and more specifically to a management method and system for an energy storage system based on environmental prediction. Background Art

[0002] As the quantity and quality of global energy demand continue to increase, energy shortages and environmental pollution problems are becoming increasingly serious, and the problem of low intelligence in centralized power generation systems continues to emerge, which has promoted the development of renewable energy. In order to solve the problem of renewable energy access to the grid, researchers combined renewable energy with energy storage devices, loads, energy conversion devices, monitoring and protection devices, and thus proposed the concept of microgrid systems. As an important part of the microgrid system, the energy storage system plays an important role in promoting the full absorption of renewable energy, enhancing the system inertia of the microgrid, improving the power quality of the microgrid, and improving the economic benefits of the microgrid operation.

[0003] Traditional energy storage system management strategies generally control access to the public power grid based on energy usage priorities and the remaining energy of each power generation module. However, the generation of new energy sources such as wind power and photovoltaics is easily restricted by weather and natural conditions, and is volatile and uncertain. It is impossible to scientifically and accurately formulate power generation plans for wind farms and photovoltaic power stations, resulting in unstable power supply, low energy utilization, and low degree of automation.

[0004] Therefore, traditional dispatching technology can no longer cope with the challenges brought by large-scale renewable energy grid connection to the operation of energy storage systems. Providing a new management method and system for energy storage systems is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a management method and system for an energy storage system based on environmental prediction, which predicts energy demand in combination with environmental data, thereby improving the prediction of energy reserves and improving energy utilization efficiency.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] In one aspect, the present invention provides a method for managing an energy storage system based on environmental prediction, comprising the following steps:

[0008] Acquire sequence environmental data, the sequence environmental data including environmental data at each moment within a preset time period sorted in chronological order, and input the sequence environmental data into a trained environmental prediction network to output an environmental prediction result;

[0009] Acquire energy demand data, the energy demand data comprising energy demand data at each moment in a preset time period sorted in chronological order, and acquire an energy demand forecast result based on the energy demand data and the environmental forecast result;

[0010] Predicting the energy storage capacity according to the environmental prediction result to obtain an energy storage capacity prediction result;

[0011] The energy storage system and the power supply system are dispatched and managed according to the energy demand forecast results and the energy storage forecast results.

[0012] Preferably, the step of inputting the sequence environment data into a trained environment prediction network to output an environment prediction result specifically includes:

[0013] Inputting the sequence environment data into an initial feature extraction layer, obtaining initial sequence environment features output by the initial feature extraction layer, wherein the initial sequence environment features include feature vectors corresponding to the environment data at each moment;

[0014] Based on the initial sequence environment feature, N split sequence environment features are generated, wherein N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector in the n-1th split sequence environment feature;

[0015] The N split sequence environmental features are respectively input into each self-attention unit to obtain the output of each self-attention unit, each self-attention unit includes at least one self-attention layer, the outputs of each self-attention unit are connected and input into the prediction unit to obtain the environmental prediction result output by the prediction unit.

[0016] Preferably, obtaining the energy demand forecast result based on the energy demand data and the environmental forecast result specifically includes:

[0017] Inputting the energy demand data into a pre-trained environmental impact model to obtain environmental impact factor data of energy demand development;

[0018] The current energy demand data, the environmental prediction results and the environmental influencing factor data of energy demand development are input into a pre-trained energy demand prediction model to obtain the energy demand prediction results.

[0019] Preferably, predicting the energy storage capacity according to the environmental prediction result to obtain the energy storage capacity prediction result specifically includes:

[0020] Construct a mapping table of environmental data and energy production;

[0021] Determine a mapping coefficient of the environment prediction result based on the mapping table;

[0022] The energy storage energy prediction result is determined based on the mapping coefficient and the environmental prediction result.

[0023] Preferably, the energy storage system and the power supply system are dispatched and managed according to the energy demand forecast result and the energy storage forecast result, specifically including:

[0024] Dividing the energy demand forecast result and the energy storage forecast result into a plurality of energy demand forecast sub-results and a plurality of energy storage forecast sub-results according to regions;

[0025] Determine whether the energy storage prediction sub-result of any power supply terminal is greater than the energy demand prediction sub-result;

[0026] When the energy storage prediction sub-result of the power supply end is greater than the energy demand prediction sub-result, using the power supply end to supply power;

[0027] When the energy storage prediction sub-result of the power supply end is not greater than the energy demand prediction sub-result, determining whether there is an energy storage prediction sub-result greater than the energy demand prediction sub-result in the remaining power supply ends;

[0028] When it exists, the power supply terminal whose energy storage prediction sub-result is greater than the energy demand prediction sub-result is used for power supply;

[0029] When it is not present or when the power supply end is not supplying power, energy is stored at the power supply end.

[0030] In another aspect, the present invention provides a management system for an energy storage system based on environmental prediction, comprising:

[0031] Data collection module, used to collect serial environmental data and energy demand data;

[0032] An environment prediction module, used to predict the environment at a future moment according to the sequence environment data to obtain an environment prediction result;

[0033] An energy demand prediction module, used for obtaining an energy demand prediction result based on the energy demand data and the environmental prediction result;

[0034] An energy storage prediction module is used to predict the energy storage according to the environmental prediction result to obtain an energy storage prediction result;

[0035] The scheduling module performs scheduling management on the energy storage system and the power supply system according to the energy demand prediction results and the energy storage prediction results.

[0036] Preferably, the data acquisition module includes a sequence environment data acquisition unit and an energy demand acquisition unit.

[0037] Preferably, the environment prediction module includes:

[0038] An initial feature extraction unit, used to obtain initial sequence environment features of the sequence environment data; the initial sequence environment features include feature vectors corresponding to the environment data at each moment;

[0039] A splitting unit, used to generate N split sequence environment features based on the initial sequence environment feature, wherein N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector in the n-1th split sequence environment feature;

[0040] A self-attention unit, used to improve the hierarchical information of the environmental features of the split sequence;

[0041] The first prediction unit is used to obtain the environment prediction result according to the output of the self-attention unit.

[0042] Preferably, the energy demand prediction module includes:

[0043] An influencing factor extraction unit, used for obtaining environmental influencing factor data of energy demand development according to the energy demand data;

[0044] The second prediction unit is used to obtain the energy demand prediction result according to the current energy demand data, the environmental prediction result and the environmental influencing factor data of the energy demand development.

[0045] Preferably, the energy storage prediction module includes:

[0046] A mapping table construction unit, used for constructing a mapping table between environmental data and energy output;

[0047] A mapping coefficient determination unit, configured to determine a mapping coefficient of the environment prediction result based on the mapping table;

[0048] A third prediction unit is used to determine the energy storage energy prediction result based on the mapping coefficient and the environmental prediction result.

[0049] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a management method and system for an energy storage system based on environmental prediction. By predicting the environment in a future period of time, and then predicting energy demand and storage capacity based on the environmental prediction results, it is possible to timely understand the energy demand and storage capacity in the future time period, plan energy scheduling in advance, improve energy utilization efficiency, enhance the stability and reliability of the power grid, and promote the large-scale application of renewable energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0051] Figure 1 It is a schematic diagram of the process of the management method of the present invention;

[0052] Figure 2 It is a structural schematic diagram of the management system of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] The embodiment of the present invention discloses a management method and system for an energy storage system based on environmental prediction. By predicting the environment in a future period of time, and then predicting energy demand and storage capacity in combination with the environmental prediction results, it is possible to timely understand the energy demand and storage capacity in the future period of time, plan energy scheduling in advance, improve energy utilization efficiency, enhance the stability and reliability of the power grid, and promote the large-scale application of renewable energy.

[0055] The embodiment of the present invention discloses a management method of an energy storage system based on environmental prediction, such as Figure 1 As shown, the following steps are included:

[0056] S1: Acquire sequence environmental data, which includes environmental data at each moment in a preset period sorted in chronological order, and input the sequence environmental data into a trained environmental prediction network to output an environmental prediction result;

[0057] Environmental data includes various parameters such as air pressure, temperature, humidity, wind direction, wind speed, rainfall, radiation, sunshine hours, etc., which can be obtained by setting various sensors. The temperature is the temperature at a height of 2 meters above the ground, the humidity is the relative humidity at a height of 2 meters above the ground, and the wind speed is the wind speed at a height of 10 meters above the ground. The operation data of each moment in a period of time are arranged in chronological order to obtain the input sequence data, and the interval between two adjacent moments is fixed.

[0058] S2: Obtain energy demand data, the energy demand data including energy demand data at each moment in a preset time period sorted in chronological order, and obtain energy demand prediction results based on the energy demand data and the environmental prediction results.

[0059] S3: Predict the energy storage capacity according to the environmental prediction result to obtain the energy storage capacity prediction result.

[0060] S4: Dispatch and manage the energy storage system and power supply system based on the energy demand forecast results and the energy storage forecast results.

[0061] Preferably, the sequence environment data is input into the trained environment prediction network to output the environment prediction result, specifically including:

[0062] The sequence environment data is input into the initial feature extraction layer to obtain the initial sequence environment features output by the initial feature extraction layer. The initial sequence environment features include feature vectors corresponding to the environment data at each moment.

[0063] In this embodiment, the input sequence environmental data is processed by a neural network model to output an environmental prediction result. The neural network model includes an initial feature extraction layer. The input sequence environmental data is first input into the initial feature extraction layer. The initial feature extraction layer can adopt the structure of an existing feature extraction network. After the initial feature extraction layer, the environmental data at each moment in the input sequence data is processed into a feature vector to constitute the initial sequence environmental feature.

[0064] Based on the initial sequence environment feature, N split sequence environment features are generated, where N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector of the n-1th split sequence environment feature;

[0065] The N split sequence environmental features are respectively input into each self-attention unit to obtain the output of each self-attention unit. Each self-attention unit includes at least one self-attention layer. The outputs of each self-attention unit are connected and input into the prediction unit to obtain the environmental prediction result output by the prediction unit.

[0066] In this embodiment, a prediction output of a long sequence is generated, and the accuracy of the prediction output of the long sequence also requires the input of the long sequence to provide sufficient information. In the processing of the long sequence, detailed information is likely to be lost, and in particular, the closer the detailed information is to the prediction time period, the more helpful it will be for the prediction result. In this embodiment, in order to retain the detailed information, the initial sequence environment feature is split to obtain the split sequence environment feature, which specifically includes: selecting the feature vector of the last 1 / M in the n-1 split sequence environment feature to obtain the nth split sequence environment feature.

[0067] M is a positive integer greater than 1. Taking M as 2 as an example, for the complete initial sequence environment feature, multiple split sequence environment features are generated. The first split sequence environment feature is the same as the initial sequence environment feature, the second split sequence environment feature includes the feature vector of the last 1 / 2 of the initial sequence environment feature, and the third split sequence environment feature includes the feature vector of the last 1 / 4 of the initial sequence environment feature.

[0068] In this embodiment, the prediction unit generates a prediction result sequence in one step to reduce the complexity of prediction time. In the neural network model of this embodiment, a corresponding self-attention unit is set for each split sequence environmental feature, and each self-attention unit includes at least one self-attention layer, wherein, since the amount of data of the first split sequence environmental feature is larger, the number of self-attention layers included in the self-attention unit corresponding to the first split sequence environmental feature is higher than the number of self-attention layers included in the self-attention units corresponding to other split sequence environmental features, which is used to extract higher-level information.

[0069] Preferably, obtaining energy demand forecast results based on energy demand data and environmental forecast results specifically includes:

[0070] Input the energy demand data into the pre-trained environmental impact model to obtain the environmental impact factor data of energy demand development;

[0071] The current energy demand data, environmental forecast results and data on environmental factors affecting energy demand development are input into a pre-trained energy demand forecast model to obtain energy demand forecast results.

[0072] Preferably, the energy storage capacity is predicted according to the environmental prediction result to obtain the energy storage capacity prediction result, which specifically includes:

[0073] Construct a mapping table of environmental data and energy production;

[0074] Determine a mapping coefficient of the environmental prediction result based on the mapping table;

[0075] The energy storage energy prediction result is determined based on the mapping coefficients and the environmental prediction result.

[0076] Preferably, the energy storage system and the power supply system are dispatched and managed according to the energy demand forecast results and the energy storage forecast results, specifically including:

[0077] Dividing the energy demand forecast results and the energy storage forecast results into a plurality of energy demand forecast sub-results and a plurality of energy storage forecast sub-results according to regions;

[0078] Determine whether the energy storage prediction sub-result of any power supply terminal is greater than the energy demand prediction sub-result;

[0079] When the energy storage prediction sub-result of the power supply end is greater than the energy demand prediction sub-result, the power supply end is used for power supply;

[0080] When the energy storage prediction sub-result of the power supply end is not greater than the energy demand prediction sub-result, determining whether there is an energy storage prediction sub-result greater than the energy demand prediction sub-result in the remaining power supply ends;

[0081] When it exists, the power supply terminal whose energy storage prediction sub-result is greater than the energy demand prediction sub-result is used for power supply;

[0082] When it is not present or when the power supply end is not supplying power, energy is stored at the power supply end.

[0083] On the other hand, the present invention provides a management system for an energy storage system based on environmental prediction, such as Figure 2 As shown, including:

[0084] Data collection module, used to collect serial environmental data and energy demand data;

[0085] An environment prediction module is used to predict the environment at a future moment based on the sequence environment data to obtain an environment prediction result;

[0086] An energy demand prediction module, used for obtaining energy demand prediction results based on energy demand data and environmental prediction results;

[0087] The energy storage prediction module is used to predict the energy storage according to the environmental prediction result to obtain the energy storage prediction result;

[0088] The scheduling module schedules and manages the energy storage system and power supply system according to the energy demand forecast results and the energy storage forecast results.

[0089] Preferably, the data acquisition module includes a sequence environment data acquisition unit and an energy demand acquisition unit.

[0090] Preferably, the environment prediction module includes:

[0091] An initial feature extraction unit is used to obtain initial sequence environment features of the sequence environment data; the initial sequence environment features include feature vectors corresponding to the environment data at each moment;

[0092] A splitting unit is used to generate N split sequence environment features based on the initial sequence environment feature, wherein N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector of the n-1th split sequence environment feature;

[0093] Self-attention unit, used to improve the hierarchical information of the context features of the split sequence;

[0094] The first prediction unit is used to obtain the environment prediction result according to the output of the self-attention unit.

[0095] Preferably, the energy demand prediction module includes:

[0096] An influencing factor extraction unit, used for obtaining environmental influencing factor data of energy demand development according to energy demand data;

[0097] The second prediction unit is used to obtain energy demand prediction results based on current energy demand data, environmental prediction results, and environmental influencing factor data of energy demand development.

[0098] Preferably, the energy storage prediction module includes:

[0099] A mapping table construction unit, used for constructing a mapping table between environmental data and energy output;

[0100] A mapping coefficient determination unit, used to determine a mapping coefficient of an environment prediction result based on a mapping table;

[0101] The third prediction unit is used to determine the energy storage prediction result based on the mapping coefficient and the environmental prediction result.

[0102] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0103] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A management method for an energy storage system based on environmental prediction, characterized in that: The following steps are involved: Acquire sequence environmental data, the sequence environmental data including environmental data at each moment within a preset time period sorted in chronological order, and input the sequence environmental data into a trained environmental prediction network to output an environmental prediction result; Acquire energy demand data, the energy demand data comprising energy demand data at each moment in a preset time period sorted in chronological order, and acquire an energy demand forecast result based on the energy demand data and the environmental forecast result; Predicting the energy storage capacity according to the environmental prediction result to obtain an energy storage capacity prediction result; Dispatching and managing the energy storage system and the power supply system according to the energy demand forecast results and the energy storage forecast results; The step of inputting the sequence environment data into the trained environment prediction network to output the environment prediction result specifically includes: Inputting the sequence environment data into an initial feature extraction layer, obtaining initial sequence environment features output by the initial feature extraction layer, wherein the initial sequence environment features include feature vectors corresponding to the environment data at each moment; Based on the initial sequence environment feature, N split sequence environment features are generated, wherein N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector in the n-1th split sequence environment feature; Input the N split sequence environmental features into each self-attention unit respectively, obtain the output of each self-attention unit, each self-attention unit includes at least one self-attention layer, connect the outputs of each self-attention unit and input them into the prediction unit, and obtain the environmental prediction result output by the prediction unit; Acquiring an energy demand prediction result based on the energy demand data and the environmental prediction result specifically includes: Inputting the energy demand data into a pre-trained environmental impact model to obtain environmental impact factor data of energy demand development; The current energy demand data, the environmental prediction results and the environmental influencing factor data of energy demand development are input into a pre-trained energy demand prediction model to obtain the energy demand prediction results.

2. A method for managing an energy storage system based on environmental prediction according to claim 1, characterized in that: Predicting the energy storage capacity according to the environmental prediction result to obtain the energy storage capacity prediction result specifically includes: Construct a mapping table of environmental data and energy production; Determine a mapping coefficient of the environment prediction result based on the mapping table; The energy storage energy prediction result is determined based on the mapping coefficient and the environmental prediction result.

3. The method for managing an energy storage system based on environmental prediction according to claim 1, characterized in that: The energy storage system and the power supply system are dispatched and managed according to the energy demand forecast result and the energy storage forecast result, specifically including: Dividing the energy demand forecast result and the energy storage forecast result into a plurality of energy demand forecast sub-results and a plurality of energy storage forecast sub-results according to regions; Determine whether the energy storage prediction sub-result of any power supply terminal is greater than the energy demand prediction sub-result; When the energy storage prediction sub-result of the power supply end is greater than the energy demand prediction sub-result, using the power supply end to supply power; When the energy storage prediction sub-result of the power supply end is not greater than the energy demand prediction sub-result, determining whether there is an energy storage prediction sub-result greater than the energy demand prediction sub-result in the remaining power supply ends; When it exists, the power supply terminal whose energy storage prediction sub-result is greater than the energy demand prediction sub-result is used for power supply; When it is not present or when the power supply end is not supplying power, energy is stored at the power supply end.

4. A management system for an energy storage system based on environmental prediction, characterized in that: include: Data collection module, used to collect serial environmental data and energy demand data; An environment prediction module, used to predict the environment at a future moment according to the sequence environment data to obtain an environment prediction result; An energy demand prediction module, used for obtaining an energy demand prediction result based on the energy demand data and the environmental prediction result; An energy storage prediction module is used to predict the energy storage according to the environmental prediction result to obtain an energy storage prediction result; A scheduling module, which performs scheduling management on the energy storage system and the power supply system according to the energy demand prediction result and the energy storage prediction result; The environment prediction module comprises: An initial feature extraction unit, used to obtain initial sequence environment features of the sequence environment data; the initial sequence environment features include feature vectors corresponding to the environment data at each moment; A splitting unit, used to generate N split sequence environment features based on the initial sequence environment feature, wherein N is a positive integer greater than 1, the first split sequence environment feature is the same as the initial sequence environment feature, and the nth split sequence environment feature is a partial feature vector in the n-1th split sequence environment feature; A self-attention unit, used to improve the hierarchical information of the environmental features of the split sequence; A first prediction unit, configured to obtain the environment prediction result according to the output of the self-attention unit; The energy demand prediction module includes: An influencing factor extraction unit, used for obtaining environmental influencing factor data of energy demand development according to the energy demand data; The second prediction unit is used to obtain the energy demand prediction result according to the current energy demand data, the environmental prediction result and the environmental influencing factor data of the energy demand development.

5. The management system of an energy storage system based on environmental prediction according to claim 4, characterized in that: The data acquisition module includes a sequence environment data acquisition unit and an energy demand acquisition unit.

6. The management system of an energy storage system based on environmental prediction according to claim 4, characterized in that: The energy storage prediction module includes: A mapping table construction unit, used for constructing a mapping table between environmental data and energy output; A mapping coefficient determination unit, configured to determine a mapping coefficient of the environment prediction result based on the mapping table; A third prediction unit is used to determine the energy storage energy prediction result based on the mapping coefficient and the environmental prediction result.

Images