Energy storage control method and system considering electricity price and user side load information

Abstract

The invention discloses an energy storage control method and system considering electricity price and user side load information, and relates to the technical field of power grid control, and the method comprises the steps: obtaining load data and electricity price information, and building an LSTM load prediction model and a typical day database based on the load data; performing prediction by using the LSTM load prediction model to obtain load power of a prediction day; performing similarity matching based on the typical day database and the prediction day to obtain similar day data; calculating the energy storage expected charging capacity of the similar day by using the data of the similar day and the electricity price information; calculating predicted day energy storage expected charging capacity by using the similar day data, the energy storage expected charging capacity of the similar day and the load power of the prediction day; and calculating energy storage charging and discharging time and power based on the load power of the prediction day, the energy storage expected charging capacity of the prediction day and an optimization model. By constructing the optimization model, automatic planning of energy storage charging and discharging power and time is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid control, and in particular to an energy storage control method and system that considers electricity price and user-side load information. Background Art

[0002] Electrochemical energy storage technology is increasingly used in power systems, especially in the industrial and commercial fields, where the installed capacity of energy storage is growing year by year. The main purpose of industrial and commercial energy storage is to use the peak-valley electricity price difference of the power grid to achieve a return on investment. The core load comes from the power demand on the user side. However, the existing energy storage control methods still have many shortcomings in practical applications. First, traditional energy storage control strategies are often too simplistic, only considering the fixed peak-valley electricity price difference and ignoring the dynamic characteristics of electricity prices. This method cannot adapt to the complex and changing power market environment, resulting in the energy storage system being unable to fully exert its economic benefits. Secondly, the existing technology does not consider the characteristics of the user-side load. The considerations are not comprehensive enough. Most control methods only rely on historical load data for simple load forecasting and fail to make full use of advanced artificial intelligence algorithms to improve the prediction accuracy. This may lead to insufficient energy storage capacity or over-discharge during peak electricity price periods, affecting the normal electricity demand of users. In addition, existing energy storage control strategies often lack consideration of the life of energy storage equipment. Excessively frequent charging and discharging operations will accelerate the degradation of the battery and reduce the overall performance and service life of the energy storage system. The existence of these problems has seriously restricted the economic benefits and operating efficiency of industrial and commercial energy storage systems. A more intelligent, comprehensive and adaptable energy storage control method is urgently needed to solve these challenges.

[0003] However, the common solutions currently available have many shortcomings, including: existing technologies are insufficient in dealing with the differences between typical daily load patterns and actual operating days. Simply relying on a single typical daily pattern for decision-making cannot adapt to the random fluctuations of daily loads, reducing the adaptability and accuracy of the control strategy; existing technologies are often too simplistic in the construction of optimization models and fail to fully consider multi-dimensional constraints and objective functions, resulting in a large deviation between the optimization results and the actual operating conditions, making it difficult to achieve truly optimal control; finally, existing energy storage control methods are also insufficient in dealing with the balance between long-term and short-term optimization goals. Excessive pursuit of short-term economic benefits may lead to a decline in the long-term stability and reliability of the energy storage system, while over-emphasizing long-term goals may prevent the system from fully utilizing current market opportunities. Summary of the invention

[0004] The present invention is proposed in view of the fact that when controlling the charging and discharging of energy storage equipment, the existing technology relies on a single typical day mode for decision-making, cannot adapt to the random fluctuations of daily loads, reduces the adaptability and accuracy of the control strategy, and has low credibility of the expected charging capacity.

[0005] Therefore, the problem to be solved by the present invention is how to provide a method for energy storage control that takes into account electricity prices and user-side load information, and a method for determining the charging capacity, charging and discharging time, and charging and discharging power of energy storage equipment by obtaining electricity price information and user-side load demand conditions and using a prediction model, a similarity comparison model, and an optimization model.

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

[0007] In a first aspect, an embodiment of the present invention provides an energy storage control method that takes into account electricity prices and user-side load information, which includes acquiring load data and electricity price information, and constructing an LSTM load forecasting model and a typical day database based on the load data; using the LSTM load forecasting model to perform forecasting to obtain the load power of the forecast day; performing similarity matching based on the typical day database and the forecast day to obtain similar day data; using the similar day data and the electricity price information to calculate the expected energy storage charging capacity of the similar day; using the similar day data, the expected energy storage charging capacity of the similar day and the load power of the forecast day to calculate the expected energy storage charging capacity of the forecast day; and calculating the energy storage charging and discharging time and power based on the load power of the forecast day, the expected energy storage charging capacity of the forecast day and the optimization model.

[0008] As a preferred solution of the energy storage control method considering electricity price and user-side load information described in the present invention, wherein: the load data includes the historical load power information on the user side in units of days and the real-time load power information on the user side; the electricity price information includes the daily hourly electricity price information on the user side; the construction of the LSTM load forecasting model includes the following steps: using the historical load power information on the jth day as the input of the LSTM load forecasting model; using the historical load power information on the j+1th day as the output of the LSTM load forecasting model; training the model to construct the LSTM load forecasting model.

[0009] As a preferred solution of the energy storage control method considering electricity price and user-side load information of the present invention, the similarity matching includes the following steps: using the Pearson similar day evaluation algorithm and the typical day database to perform similarity calculation on the predicted day to obtain similar day data with similar characteristics to the predicted day, and the similarity calculation formula is as follows:

[0010]

[0011] in, is the similarity coefficient; m is the total number of samples; x i is the forecast data at the i-th moment; is the mean of the corresponding predicted data; y i is the historical data at the i-th moment; is the mean of the corresponding historical data; the similarity between the predicted day and the typical day is judged. If the similarity between the predicted day and the typical day is greater than c1, the corresponding predicted day is judged to be a similar day, and the load power of the similar day is recorded. and similarity coefficient Otherwise, it is judged that the corresponding predicted day is not a similar day. The specific formula is as follows:

[0012]

[0013] in, is the load power on similar days; is the historical load power information of day j; is the similarity coefficient; c1 is the similarity threshold.

[0014] As a preferred solution of the energy storage control method considering electricity price and user side load information of the present invention, the calculation of the expected charging capacity of energy storage on similar days includes the following steps: according to the electricity price information V of the predicted day i x+1 Record the time period [t1, t2] when the electricity price is at its peak; based on the load power of the similar day Calculate the user demand capacity E under similar peak hours of electricity prices n ; Based on the user demand capacity E during the peak period of similar electricity prices n Calculate and obtain the expected charging capacity of energy storage on the nth similar day The specific formula for the expected charging capacity of energy storage on similar days is as follows:

[0015]

[0016]

[0017] in, is the expected charging capacity of energy storage on the nth similar day; E c is the rated capacity of energy storage; E n is the user demand capacity during the peak period of similar electricity prices; E s is the maximum charging capacity of the energy storage when the electricity price is at the valley value; t1 is the start time of the peak electricity price; t2 is the end time of the peak electricity price; is the load power of the jth similar day when the electricity price is at the peak; t3 is the start time of the electricity price valley; t4 is the end time of the electricity price valley; PD e is the rated power of the energy storage.

[0018] As a preferred solution of the energy storage control method considering electricity price and user side load information of the present invention, the calculation of the predicted daily energy storage expected charging capacity includes the following steps: based on the similarity coefficient Expected charging capacity of energy storage on similar days and the predicted user side load power P on the predicted day x+1 s x+1 Calculate and obtain the predicted daily user demand capacity E in the peak electricity price period [t1, t2] xy ; Use the weighted average method to predict the daily user demand capacity E xy , the similarity coefficient Expected charging capacity of energy storage on similar days as described above Calculate and get the expected daily energy storage charging capacity E cbest The calculation formula for the predicted daily energy storage expected charging capacity is as follows:

[0019]

[0020]

[0021] Δδ=0.05×E c

[0022] Among them, E cbest E is the expected charging capacity of the predicted daily energy storage; xy To predict daily user demand capacity; and is the similarity coefficient; γ1, γ2 and γ n is the similarity discrimination coefficient; and is the expected charging capacity of energy storage on similar days; Δδ is the reserved capacity; n is the number of typical days in the database; E c is the rated capacity of energy storage.

[0023] As a preferred solution of the energy storage control method considering electricity price and user-side load information of the present invention, wherein: the energy storage charging and discharging time and power include energy storage charging time and power and energy storage discharging time and power; the energy storage charging time and power refers to the predicted user-side load power P based on the predicted day x+1 s x+1 、The predicted daily energy storage expected charging capacity E cbest and the electricity price information V on the forecast day i x+1 , with the highest daily return as the goal, the energy storage charging time and power are calculated; the optimization model is calculated using the particle swarm optimization algorithm, and the calculation formula for the optimization target is as follows:

[0024] minf c =p c +e c -s c

[0025]

[0026] Among them, f c is the daily income; p c is the daily maintenance cost; c Charging cost for energy storage; c is the energy storage discharge profit; η1 is the energy storage charging efficiency; V i x+1 is the electricity price information at time i on the forecast day; E di The energy storage charging capacity within a sampling interval; PD e is the rated power of energy storage; V min is the lowest predicted electricity price; i is the sampling time; k i is the charging capacity coefficient at time i; ΔV1 is the set charging electricity price threshold.

[0027] As a preferred solution of the energy storage control method considering electricity price and user-side load information of the present invention, the calculation formula of the energy storage discharge time and power is as follows:

[0028]

[0029] in, is the predicted discharge power of daily energy storage at time t; P now Real-time load power information on the user side; V max V is the maximum value of the predicted electricity price; i x+1 is the electricity price information at time i on the predicted day; ΔV2 is the set discharge electricity price threshold.

[0030] In the second aspect, in order to further solve the safety problems existing in power grid control, the present invention provides an energy storage control system that takes into account electricity prices and user-side load information, which includes: a data processing module, which is used to obtain historical load power information, real-time load power information on the user side and electricity price information, and use the historical load power information to build an LSTM load forecasting model and a typical day database; a load forecasting module, which is used to use the LSTM load forecasting model to predict and obtain the load power of the predicted day; a similarity matching module, which is used to use the typical day database to perform similarity matching on the predicted day to obtain the load power and similarity coefficient of the similar day; a similar capacity module, which is used to use the load power of the similar day and the electricity price information of the predicted day to calculate and obtain the expected energy storage charging capacity of the similar day; a prediction capacity module, which is used to use the similarity coefficient of the similar day, the expected energy storage charging capacity of the similar day and the load power of the predicted day to calculate and obtain the expected energy storage charging capacity of the predicted day; an energy storage calculation module, which is used to calculate based on the load power of the predicted day, the expected energy storage charging capacity of the predicted day and the optimization model to obtain the energy storage charging time and power and the energy storage discharge time and power.

[0031] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of a method for energy storage control that considers electricity prices and user-side load information as described in the first aspect of the present invention is implemented.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of a method for energy storage control that considers electricity prices and user-side load information as described in the first aspect of the present invention is implemented.

[0033] The beneficial effects of the present invention are as follows: the present invention obtains user-side load-related information by utilizing a prediction algorithm and a similarity matching algorithm, obtains the expected charging capacity on the prediction day by weighted average of similarity weights under the premise of considering electricity price information, and obtains the expected charging capacity with greater credibility, thereby ensuring subsequent planning and preventing overcharging of energy storage; by constructing an optimization model that considers electricity price and load information and setting different charging electricity price thresholds, the energy storage peak-valley arbitrage economic maximization and the automatic planning of charging power and time are achieved; by predicting the real-time load on the day and setting different discharge electricity price thresholds, the automatic planning of energy storage discharge power and time is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0035] Figure 1 This is an overall flow chart of the energy storage control method in Example 1 that takes into account electricity prices and user-side load information. DETAILED DESCRIPTION

[0036] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0037] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0038] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0039] Example 1

[0040] Reference Figure 1 , which is the first embodiment of the present invention, provides an energy storage control method considering electricity price and user-side load information, including the following steps:

[0041] S1: Obtain load data and electricity price information, and build an LSTM load forecasting model and a typical day database based on the load data.

[0042] Specifically, Figure 1 The figure shows the overall process of the energy storage control method of the present invention. By constructing an LSTM load forecasting model and a typical day database and using a similar day algorithm to compare similar days, the corresponding similar days are obtained, and then calculations are performed based on all similar days to confirm the expected energy storage charging capacity of the day corresponding to different similar days, and the similar weighted average algorithm is used to obtain the expected charging capacity of the predicted day. After that, the energy storage planning part is performed, and the maximum daily profit is taken as the optimization goal. The PSO algorithm is used to calculate based on the electricity price information of the predicted day to obtain the energy storage charging power and time and the energy storage discharging power and time.

[0043] Preferably, the load data includes the user-side historical load power information in units of days and the user-side real-time load power information, wherein is the historical load power information of the jth day, the power information sampling interval is 1h, a total of 24 power information, the user side real-time load power information P now The power information sampling interval is 1h, with a total of 24 power information.

[0044] Preferably, the electricity price information includes the user side daily hourly electricity price information V i x .

[0045] Specifically, building the LSTM load forecasting model includes the following steps: As the input of LSTM load forecasting model.

[0046] The historical load power information of day j+1 As the output of the LSTM load forecasting model.

[0047] Train the model and build an LSTM load forecasting model.

[0048] Specifically, the calculation formula of the LSTM load forecasting model is as follows:

[0049]

[0050] in, is the historical load power information of day j+1; H t (·) is the LSTM load forecasting model; is the historical load power information of the jth day.

[0051] Furthermore, constructing a typical day database includes the following steps: using the historical load power information of the jth day Construct a typical day database. The specific formula of the typical day database is as follows:

[0052]

[0053] Among them, O j It is a typical day database; is the historical load power information of the jth day.

[0054] S2: Use the LSTM load forecasting model to predict and obtain the load power on the forecast day.

[0055] Preferably, using the LSTM load forecasting model for forecasting means taking the actual load power of the current day as the LSTM load forecasting model H t(·) is input to obtain the predicted user-side load power for the next day. The specific formula is as follows:

[0056]

[0057] in, is the predicted user-side load power on the prediction day x+1, with a prediction interval of 1h; H t (·) is the LSTM load forecasting model; is the actual load power of the current day x, and the sampling interval is 1h.

[0058] Preferably, the LSTM deep learning model is combined with the traditional typical day database method, instead of simply using them in parallel, the advantages of the two methods are cleverly complemented through the similar day algorithm. The LSTM deep learning model is used to provide modeling capabilities for complex nonlinear relationships, and the typical day database is used to provide interpretability support based on historical data, thereby significantly improving the accuracy and robustness of load forecasting.

[0059] S3: Perform similarity matching based on the typical day database and the predicted day to obtain similar day data.

[0060] Preferably, similarity matching includes the following steps: using the Pearson similar day evaluation algorithm and the typical day database to perform similarity calculation on the predicted day to obtain similar day data with similar characteristics to the predicted day, and the similarity calculation formula is as follows:

[0061]

[0062] in, is the similarity coefficient; m is the total number of samples; x i is the forecast data at the i-th moment; is the mean of the corresponding predicted data; i is the historical data at the i-th moment; is the mean of the corresponding historical data.

[0063] The similarity between the predicted day and the typical day is judged. If the similarity between the predicted day and the typical day is greater than c1, the corresponding predicted day is judged to be a similar day, and the load power of the similar day is recorded. and similarity coefficient Otherwise, it is judged that the corresponding predicted day is not a similar day. The specific formula is as follows:

[0064]

[0065] in, is the load power on similar days; is the historical load power information of day j; is the similarity coefficient; c1 is the similarity threshold.

[0066] Specifically, the similarity between the predicted day and the typical day is determined by calculating the similarity of the predicted day using a typical day database, obtaining the similarities corresponding to different typical days and performing statistics, setting an initial threshold and continuously adjusting it, and then obtaining the best threshold for the similarity between the predicted day and the typical day.

[0067] Preferably, by introducing the Pearson correlation coefficient and weighted average method in the similar day analysis, rather than simply applying the existing technology, it achieves accurate screening and rational use of similar days by setting a similarity threshold and introducing a similarity discrimination coefficient, effectively balancing the reference value of historical data and the accuracy of current predictions.

[0068] S4: Calculate the expected charging capacity of energy storage on similar days using similar day data and electricity price information.

[0069] Preferably, calculating the expected charging capacity of energy storage on similar days includes the following steps: i x+1 Record the time period [t1, t2] when the predicted electricity price is at its peak.

[0070] Based on the load power of similar days Calculate the user demand capacity E under similar peak hours of electricity prices n .

[0071] According to the user demand capacity E under similar electricity price peak time period n Calculate and obtain the expected charging capacity of energy storage on the nth similar day

[0072] Specifically, the specific formula for the expected charging capacity of energy storage on similar days is as follows:

[0073]

[0074]

[0075] in, is the expected charging capacity of energy storage on the nth similar day; E c is the rated capacity of energy storage; E n is the user demand capacity during the peak period of similar electricity prices; E s is the maximum charging capacity of the energy storage when the electricity price is at the valley value; t1 is the start time of the peak electricity price; t2 is the end time of the peak electricity price. i x+1 When it is greater than the peak electricity price threshold ΔW1, it is the peak electricity price period; is the load power of the jth similar day when the electricity price is predicted to be at the peak value; t3 is the starting time of the electricity price valley; t4 is the end time of the electricity price valley, where the electricity price information V i x+1 When the power price is less than the valley power price threshold ΔW2, it is the power price valley time period; PD e is the rated power of the energy storage.

[0076] S5: Calculate the expected charging capacity of the energy storage on the forecast day using the data of similar days, the expected charging capacity of the energy storage on similar days, and the load power on the forecast day.

[0077] Preferably, calculating the predicted daily energy storage expected charging capacity includes the following steps: based on the similarity coefficient Expected charging capacity of energy storage on similar days and the predicted user side load power P on the prediction day x+1 s x+1 Calculate and obtain the predicted daily user demand capacity E in the peak electricity price period [t1, t2] xy .

[0078] The weighted average method is used to predict the daily user demand capacity E xy , similarity coefficient and the expected charging capacity of energy storage on similar days Calculate and get the predicted daily energy storage expected charging capacity E cbest .

[0079] Specifically, the calculation formula for predicting the expected daily energy storage charging capacity is as follows:

[0080]

[0081] Δδ=0.05×E c

[0082] Among them, E cbest E is the expected charging capacity of the predicted daily energy storage; xy To predict daily user demand capacity; and is the similarity coefficient; γ1, γ2 and γ n is the similarity discriminant coefficient, where if the similarity coefficient is greater than or equal to 0.8, the similarity discriminant coefficient is 1, otherwise the similarity discriminant coefficient is 0; and is the expected charging capacity of energy storage on similar days; Δδ is the reserved capacity; n is the number of typical days in the database; E c is the rated capacity of energy storage.

[0083] Specifically, similar days are screened by setting an initial similarity discrimination coefficient threshold, corresponding similar day data is obtained and calculated, and the initial similarity discrimination coefficient threshold is continuously updated for the purpose of most accurately predicting the expected daily energy storage charging capacity, and finally the best similarity discrimination coefficient threshold is obtained.

[0084] Preferably, by comprehensively considering the information of multiple similar days and introducing the concept of reserved capacity, not only the accuracy of the prediction is improved, but also the flexibility and reliability of the system are enhanced. Compared with the traditional single prediction model, the effect is better, reflecting the in-depth understanding and innovative application of the characteristics of the energy storage system.

[0085] S6: Calculate the energy storage charging and discharging time and power based on the load power on the forecast day, the expected charging capacity of the energy storage on the forecast day and the optimization model.

[0086] Preferably, the energy storage charging and discharging time and power include energy storage charging time and power and energy storage discharging time and power.

[0087] Preferably, the energy storage charging time and power refer to the predicted user-side load power P based on the predicted day x+1. s x+1 、Forecast daily energy storage expected charging capacity E cbest and the electricity price information V on the forecast day i x+1 , with the goal of maximizing the daily profit, the energy storage charging time and power are calculated.

[0088] Specifically, the optimization model is calculated using the particle swarm optimization algorithm, and the calculation formula of the optimization target is as follows:

[0089] minf c =p c +e c -s c

[0090] p c =E c ×C d

[0091]

[0092] s c =η2×V max ×E cbest

[0093] Among them, f c is the daily income; p c is the daily maintenance cost; c Charging cost for energy storage; c E is the profit of energy storage discharge; c is the rated capacity of energy storage; Cd is the daily maintenance cost of 1MWh energy storage; η1 is the energy storage charging efficiency; V i x+1 is the electricity price information at time i on the forecast day; E di The energy storage charging capacity within a sampling interval; PD e is the rated power of energy storage; V max is the predicted maximum value of the electricity price, where the predicted maximum value of the electricity price V max =1.167 yuan / kWh; V min is the predicted minimum value of the electricity price, where the predicted minimum value of the electricity price V min =0.246 yuan / kWh; i is the sampling time, where sampling time i=1,2,3,...,24; k i is the charging capacity coefficient at time i; ΔV1 is the set charging electricity price threshold, where the set charging electricity price threshold ΔV1 = 0.2 yuan / kWh. If the current electricity price minus the lowest electricity price of the day is greater than the set charging electricity price threshold ΔV1, the energy storage charging power is set to 0; η2 is the energy storage discharge efficiency, where the energy storage discharge efficiency η2 = 0.88.

[0094] Furthermore, based on the calculation results of the particle swarm optimization algorithm, the charging power at time i on the predicted day is obtained. The calculation formula of energy storage charging time and power is as follows:

[0095]

[0096] in, k is the charging power predicted for the daily energy storage at time t; i is the charging capacity coefficient at time i; PD e is the rated power of the energy storage.

[0097] Furthermore, the calculation formula of energy storage discharge time and power is as follows:

[0098]

[0099] in, is the predicted discharge power of daily energy storage at time t; P now Real-time load power information on the user side; V max V is the maximum value of the predicted electricity price; i x+1 is the electricity price information at time i on the predicted day; ΔV2 is the set discharge electricity price threshold, where the set discharge electricity price threshold ΔV2 = 0.3 yuan / kWh. If the highest electricity price of the day minus the current electricity price is greater than the set discharge electricity price threshold ΔV2, the energy storage discharge power is set to 0.

[0100] Preferably, the optimization of multiple indicators is achieved by taking the highest intraday profit as the optimization goal and considering the limitations of energy storage efficiency and charging and discharging power. The LSTM deep learning model, Pearson correlation coefficient and particle swarm optimization algorithm PSO are combined, and the prediction results of the LSTM deep learning model are used to provide basic data for the particle swarm optimization algorithm PSO. Similar day analysis is used to supplement and correct the prediction results of the LSTM deep learning model. The most suitable algorithm is selected according to the characteristics of different problems, and an organic connection is established between different algorithms to realize the charging and discharging control of energy storage equipment.

[0101] This embodiment also provides an energy storage control system that takes into account electricity prices and user-side load information, including: a data processing module, which is used to obtain historical load power information, real-time user-side load power information and electricity price information, and use the historical load power information to build an LSTM load forecasting model and a typical day database; a load forecasting module, which is used to use the LSTM load forecasting model to predict and obtain the load power of the predicted day; a similarity matching module, which is used to use the typical day database to perform similarity matching on the predicted day to obtain the load power and similarity coefficient of the similar day; a similar capacity module, which is used to use the load power of the similar day and the electricity price information of the predicted day to calculate and obtain the expected energy storage charging capacity of the similar day; a prediction capacity module, which is used to use the similarity coefficient of the similar day, the expected energy storage charging capacity of the similar day and the load power of the predicted day to calculate and obtain the expected energy storage charging capacity of the predicted day; an energy storage calculation module, which is used to calculate based on the load power of the predicted day, the expected energy storage charging capacity of the predicted day and the optimization model to obtain the energy storage charging time and power and the energy storage discharging time and power.

[0102] This embodiment also provides a computer device, which is applicable to a method for energy storage control that takes electricity prices and user-side load information into consideration, and includes: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method for energy storage control that takes electricity prices and user-side load information into consideration as proposed in the above embodiment.

[0103] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0104] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a method for energy storage control that takes into account electricity prices and user-side load information as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0105] In summary, the present invention obtains user-side load-related information by utilizing a prediction algorithm and a similarity matching algorithm, and obtains the expected charging capacity of the predicted day by weighted average of similarity weights under the premise of considering electricity price information, so that the expected charging capacity is more credible, ensuring subsequent planning and preventing overcharging of energy storage; by constructing an optimization model that considers electricity price and load information and setting different charging electricity price thresholds, the energy storage peak-valley arbitrage economy is maximized and the charging power and time are automatically planned; by predicting the real-time load of the day and setting different discharge electricity price thresholds, the energy storage discharge power and time are automatically planned.

[0106] Example 2

[0107] Referring to Table 1 and Table 2, a second embodiment of the present invention is shown. This embodiment is different from the first embodiment in that, in order to verify its beneficial effects, operating data and related instructions of the present invention in an actual environment are provided.

[0108] This example obtains the historical load power information of the user side in a certain area and the electricity price information at different times on the user side, and then constructs an LSTM load forecasting model and a typical day database. Table 1 shows the electricity price table for each time period of the day. The load power of the forecast day is obtained by using the LSTM load forecasting model for prediction.

[0109] Table 1 Electricity prices for different periods of the day

[0110] time state Electricity Prices 00:00-10:00 flat 0.682 yuan / kWh 10:00-11:00 valley 0.343 yuan / kWh 11:00-14:00 deep valley 0.246 yuan / kWh 14:00-15:00 valley 0.343 yuan / kWh 15:00-17:00 flat 0.682 yuan / kWh 17:00-20:00 peak 1.167 yuan / kWh 20:00-22:00 peak 1.022 yuan / kWh 22:00-24:00 flat 0.682 yuan / kWh

[0111] Based on the obtained predicted day, the similarity of the characteristics of the predicted day is calculated by using the typical day database to obtain the similarity corresponding to different typical days. The specific data is shown in Table 2. The similarity data is compared with the threshold c1=0.75 of the similarity between the predicted day and the typical day. If the similarity exceeds the threshold, the corresponding typical day is judged to be a similar day, and the corresponding load power and similarity coefficient are recorded.

[0112] Table 2 Similarity data between predicted day and typical day

[0113] sequence name Similarity Typical day 1 0.3145 Typical day 2 0.1124 Typical day 3 0.8204 Typical day 4 0.6172 Typical day 5 0.4586 Typical day 6 0.5698 Typical day 7 0.8712 … …

[0114] According to the data in Table 1 and Table 2, the predicted peak electricity price period is 17:00-22:00, and the corresponding peak electricity price threshold ΔW1=1.00 yuan / kWh. The valley time period is 10:00-15:00, and the corresponding valley electricity price threshold ΔW2=0.40 yuan / kWh. Then, according to the data of the peak time period and the valley time period, the predicted maximum electricity price V is obtained. max = 1.167 yuan / kWh, the lowest predicted electricity price V min =0.246 yuan / kWh, and then judge according to the set charging electricity price threshold ΔV1=0.2 yuan / kWh and the set discharge electricity price threshold ΔV2=0.3 yuan / kWh. If the current electricity price minus the lowest electricity price of the day is greater than the set charging electricity price threshold ΔV1, the energy storage charging power is set to 0. If the highest electricity price of the day minus the current electricity price is greater than the set discharge electricity price threshold ΔV2, the energy storage discharge power is set to 0, and then the maximum intraday profit is used as the optimization goal for calculation to obtain the energy storage charging time and power and the energy storage discharge time and power. Therefore, based on the above content, it can be seen that the present invention effectively improves the accuracy of the expected charging capacity, prevents overcharging of energy storage, and realizes the maximization of energy storage peak-valley arbitrage economy and the automatic planning of energy storage charging and discharging power and time.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An energy storage control method considering electricity price and user-side load information, characterized in that: include: Obtain load data and electricity price information, and build an LSTM load forecasting model and a typical day database based on the load data; The LSTM load forecasting model is used to perform forecasting to obtain the load power on the forecasted day; Perform similarity matching based on the typical day database and the predicted day to obtain similar day data; Calculating the expected charging capacity of energy storage on similar days using the similar day data and the electricity price information; Calculating the expected charging capacity of energy storage on the forecast day by using the similar day data, the expected charging capacity of energy storage on the similar day and the load power on the forecast day; The energy storage charging and discharging time and power are calculated based on the load power on the forecast day, the expected charging capacity of the energy storage on the forecast day and the optimization model.

2. The energy storage control method considering electricity price and user-side load information as claimed in claim 1, characterized in that: The load data includes historical load power information on the user side in units of days and real-time load power information on the user side; The electricity price information includes user-side daily hourly electricity price information; The construction of the LSTM load forecasting model includes the following steps: Taking the historical load power information of the jth day as the input of the LSTM load forecasting model; Taking the historical load power information of the j+1th day as the output of the LSTM load forecasting model; Train the model and build an LSTM load forecasting model.

3. The energy storage control method considering electricity price and user-side load information as claimed in claim 2, characterized in that: The similarity matching comprises the following steps: The similarity of the predicted day is calculated using the Pearson similar day evaluation algorithm and the typical day database to obtain similar day data with similar characteristics to the predicted day. The similarity calculation formula is as follows: in, is the similarity coefficient; m is the total number of samples; x i is the forecast data at the i-th moment; is the mean of the corresponding predicted data; y i is the historical data at the i-th moment; is the mean of the corresponding historical data; The similarity between the predicted day and the typical day is judged. If the similarity between the predicted day and the typical day is greater than c1, the corresponding predicted day is judged to be a similar day, and the load power of the similar day is recorded. and similarity coefficient Otherwise, it is judged that the corresponding predicted day is not a similar day. The specific formula is as follows: in, is the load power on similar days; is the historical load power information of day j; is the similarity coefficient; c1 is the similarity threshold.

4. The energy storage control method considering electricity price and user-side load information as claimed in claim 3, characterized in that: The calculation of the expected charging capacity of energy storage on similar days includes the following steps: Based on the electricity price information on the forecast day Record the time period [t1, t2] when the predicted electricity price is at its peak; According to the load power of the similar day Calculate the user demand capacity E under similar peak hours of electricity prices n ; According to the user demand capacity E under similar electricity price peak time period n Calculate and obtain the expected charging capacity of energy storage on the nth similar day The specific formula for the expected charging capacity of energy storage on similar days is as follows: in, is the expected charging capacity of energy storage on the nth similar day; E c is the rated capacity of energy storage; E n is the user demand capacity during the peak period of similar electricity prices; E s is the maximum charging capacity of the energy storage when the electricity price is at the valley value; t1 is the start time of the peak electricity price; t2 is the end time of the peak electricity price; is the load power of the jth similar day when the electricity price is at the peak; t3 is the start time of the electricity price valley; t4 is the end time of the electricity price valley; PD e is the rated power of the energy storage.

5. The energy storage control method considering electricity price and user-side load information as claimed in claim 4, characterized in that: The calculation and prediction of the expected daily energy storage charging capacity comprises the following steps: Based on the similarity coefficient Expected charging capacity of energy storage on similar days and the predicted user side load power on the predicted day x+1 Calculate and obtain the predicted daily user demand capacity E in the peak electricity price period [t1, t2] xy ; The predicted daily user demand capacity E is calculated by using the weighted average method. xy , the similarity coefficient Expected charging capacity of energy storage on similar days as described above Calculate and get the expected daily energy storage charging capacity E cbest ; The calculation formula for the predicted daily energy storage expected charging capacity is as follows: Δδ=0.05×E c Among them, E cbest E is the expected charging capacity of the predicted daily energy storage; xy To predict daily user demand capacity; and is the similarity coefficient; γ1, γ2 and γ n is the similarity discrimination coefficient; and is the expected charging capacity of energy storage on similar days; Δδ is the reserved capacity; n is the number of typical days in the database; E c is the rated capacity of energy storage.

6. The energy storage control method considering electricity price and user-side load information as claimed in claim 5, characterized in that: The energy storage charging and discharging time and power include energy storage charging time and power and energy storage discharging time and power; The energy storage charging time and power refer to the predicted user-side load power based on the predicted day x+1. The predicted daily energy storage expected charging capacity E cbest and the electricity price information for the forecast day The calculation is based on the goal of maximizing the daily profit to obtain the energy storage charging time and power; The optimization model is calculated using the particle swarm optimization algorithm, and the calculation formula of the optimization target is as follows: minf c =p c +e c -s c Among them, f c is the daily income; p c is the daily maintenance cost; c Charging cost for energy storage; c is the energy storage discharge profit; η1 is the energy storage charging efficiency; is the electricity price information at time i on the forecast day; E di The energy storage charging capacity within a sampling interval; PD e is the rated power of energy storage; V min is the lowest predicted electricity price; i is the sampling time; k i is the charging capacity coefficient at time i; ΔV1 is the set charging electricity price threshold.

7. The energy storage control method considering electricity price and user-side load information as claimed in claim 6, characterized in that: The calculation formula of the energy storage discharge time and power is as follows: in, is the predicted discharge power of daily energy storage at time t; P now Real-time load power information on the user side; V max V is the maximum value of the predicted electricity price; i x+1 is the electricity price information at time i on the predicted day; ΔV2 is the set discharge electricity price threshold.

8. An energy storage control system considering electricity price and user-side load information, based on an energy storage control method considering electricity price and user-side load information according to any one of claims 1 to 7, characterized in that: include, The data processing module is used to obtain historical load power information, user-side real-time load power information and electricity price information, and use the historical load power information to build an LSTM load forecasting model and a typical day database; The load forecasting module is used to use the LSTM load forecasting model to make forecasts and obtain the load power on the forecast day; A similarity matching module is used to perform similarity matching on the predicted day using the typical day database to obtain the load power and similarity coefficient of the similar day; The similar capacity module is used to calculate the expected charging capacity of energy storage on similar days by using the load power on similar days and the electricity price information on the forecast day; The prediction capacity module is used to calculate the expected charging capacity of energy storage on the prediction day by using the similarity coefficient of similar days, the expected charging capacity of energy storage on similar days and the load power on the prediction day; The energy storage calculation module is used to calculate based on the load power of the forecast day, the expected charging capacity of the energy storage on the forecast day and the optimization model to obtain the energy storage charging time and power and the energy storage discharging time and power.

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 energy storage control method considering electricity price and user-side load information as described in 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 an energy storage control method considering electricity prices and user-side load information as described in any one of claims 1 to 7 are implemented.

Images